Preface

These engineering guidelines have been prepared by the Office of Energy Projects (OEP) to provide
guidance to the technical Staff in the processing of applications for license and in the evaluation of dams
under Part 12 of the Commission's regulations. The Guidelines will also be used to evaluate proposed
modifications or additions to existing projects under the jurisdiction of the Federal Energy Regulatory
Commission (Commission). Staff technical personnel consist of the professional disciplines (e.g.
professional engineers and geologists) that have the responsibility for reviewing studies and evaluating
designs prepared by owners or developers of dams.

The guidelines are intended to provide technical personnel of the Office of Energy Projects, including
the Regional Office and Washington Office personnel, with procedures and criteria for the engineering
review and analysis of projects over which the Commission has jurisdiction. In addition, these
guidelines should be used by staff in the evaluation of consultant or licensee/exemptee conducted
studies. The guidance is intended to cover the majority of studies usually encountered by Staff.
However, special cases may require deviation from, or modification of, the guidelines. When such
cases arise, Staff must determine the applicability of alternate criteria or procedures based upon their
experience and must exercise sound engineering judgment when considering situations not covered by
the guidelines. The alternate procedures, or criteria, used in these situations should be justified and
accompanied by any suggested changes for incorporation in the guidelines. Since every dam site and
hydropower related structure is unique, individual design considerations and construction treatment will
be required. Technical judgment is therefore required in most analytical studies.

These guidelines are not a substitute for good engineering judgment, nor are the procedures
recommended herein to be applied rigidly in place of other analytical solutions to engineering problems
encountered by staff. Staff should keep in mind that the engineering profession is not limited to a
specific solution to each problem, and that the results are the desired end to problem solving.

These guidelines are primarily intended for internal use by OEP staff, but also provide licensees,
exemptees, and applicants with general guidance that should be considered when presenting any studies
presented to the Commission under Parts 4 and 12 of the Regulations (18 CFR, Parts 4 and 12).
When any portions of the Guidelines becomes outdated, obsolete, or needs revision for any reason, it
will be revised and supplemented as necessary. Comments on, or recommended changes, in these
Guidelines should be forwarded to the Director of the Division of Inspections for consideration and
possible inclusion in future updates. New pages will be prepared and issued with instructions for page
replacements.
 CHAPTER I

GENERAL REQUIREMENTS

April 1991
 Chapter I

General Requirements

1-0 Contents

Title Page

1-1 Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

1-1.1 Regulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

1-1.2 Operating Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

1-2 Project Classifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2

1-2.1 Hazard Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2

1-2.2 Downstream Hazard
Potential - Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2

1-3 Study Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3

1-3.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3

1-3.2 Regional Office Inspections
and Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
1-3.3 Washington Office Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
1-3.3.1 License Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
1-3.3.2 Review of Consultants Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
1-3.3.3 Review of Staff Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4

1-4 Deviations from Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4

1-4.1 Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4

1-4.2 Deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4

1-5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5

1-i
 Chapter I

General Requirements

1-1 Purpose and Scope

The purpose of this chapter of the guidelines is to establish the basis for determining the need
for engineering review and studies conducted by Commission Staff during the processing
stage of license applications and the review of reports prepared by licensees, exemptees, or
independent consultants.

The following Federal Power Act regulations, and Division of Dam Safety and Inspections,
Office of Hydropower Licensing Operating Manual, provide requirements and general
guidance concerning: the contents, deposition, and evaluation of applications for licenses or
exemption, and the supervision of existing licenses and exemptions.

1-1.1 Regulations

Application for License for Major Unconstructed Project and Major Modified Project; and
Application for Amendment to License - Subchapter B, Part 4, Subpart E, Section 4.40 and
Subpart H, Sections 4.70 and 4.41.

Application for License for Major Project - Existing Dam - Subchapter B, Part 4, Subpart F,
Sections 4.50 and 4.51.

Application for License for Minor Water Power Projects and Major Water Power Projects
5 Megawatts or Less - Subchapter B, Part 4, Subpart G. Sections 4.60 and 4.61.

Exemptions of Small Hydroelectric Power Projects of 5 Megawatts or Less - Subchapter B,

Subpart K, Sections 4.101 to 4.108.

Amendment of License - Subchapter B, Subpart L, Part 4, Sections 4.200 to 4.202.

1-1.2 Operating Manual

For Inspection of Projects and Supervision of Licenses for Water Power Projects published
in 1986.

1-1
1-2 Project Classification

1-2.1 Hazard Classification

The hazard potential classification of a project determines the level of engineering review
and the criteria that are applicable. Therefore, it is critical to determine the appropriate
hazard potential of a dam, because it sets the stage for the analyses that must be completed
to properly evaluate the structural integrity of any dam.

1-2.2 Downstream Hazard Potential - Definitions

The hazard potential of dams describes the potential for loss of human life or property
damage in the area downstream or upstream of the dam in event of failure or incorrect
operation of a dam. Hazard classification does not indicate the structural integrity of the dam
itself, but rather the effects if a failure should occur. The hazard potential assigned to a dam
is based on consideration of the effects of a failure during both normal and flood flow
conditions.

Dams conforming to criteria for the low hazard potential category generally are located in
rural or agricultural areas where failure may damage farm buildings, limited agricultural
land, or township and country roads. Low hazard potential dams have a small storage
capacity, the release of which would be confined to the river channel in the event of a failure
and therefore would represent no danger to human life.

Significant hazard potential category structures are usually located in predominately rural or
agricultural areas where failure may damage isolated homes, secondary highways or minor
railroads; cause interruption of use or service of relatively important public utilities; or cause
some incremental flooding of structures with possible danger to human life.

Dams in the high hazard potential category are those located where failure may cause serious
damage to homes, agricultural, industrial and commercial facilities, important public utilities,
main highways, or railroads, and there would be danger to human life.

The hazard potential evaluation includes consideration of recreational development and use
and socio-economic matters. Included in the high hazard potential category are dams where
failure would cause serious damage to permanently established or organized recreational
areas or activities. Also included in the high hazard potential category are dams where
failure could result in loss of life of people gathered for an unorganized recreational activity
(such as salmon fishermen and kayakers) where concentrated use of a confined area below
the dam is a common annual occurrence during certain times each year.

1-2
1-3 Study Requirements

1-3.1 General

The following guidance shall establish the basic requirements for reviews and studies
conducted by both the Washington and Regional offices. It is recognized that unique
situations may require deviations from these guidelines, however, they are considered
flexible enough to be followed for most of the basic types of reviews and studies anticipated.
Any engineering study which is conducted, shall be consistent with the applicable sections
of these guidelines.

1-3.2 Regional Office Inspections and Studies

The operating manual, prepared by the Division of Dam Safety and Inspections (D2SI),
establishes minimum requirements for reports and field inspections of hydroelectric projects
conducted pursuant to the Federal Power Act.

1-3.3 Washington Office Studies

1-3.3.1 License Applications

Review for Deficiencies - All license applications shall be reviewed for compliance with the
engineering requirements of FERC regulations. Application deficiencies should be
documented so the applicant can be appropriately and timely notified. A preliminary review
should then be conducted to preliminarily assess economic feasibility and to ensure that the
project's power output can be utilized. These preliminary studies should be conducted prior
to the acceptance of the application. Items which should be examined include: the need for
project power; the existence (or absence) of an agreement or memorandum of understanding
for sale of project power; the impact of changes in fish habitat preservation flow releases
on power generation; and the reasonableness of the project construction cost estimate. This
study should resolve any basic questions concerning the ability of the Applicant to build the
project and/or sell the project power.

Safety and Design Assessment - The safety and design assessment report shall include a
summary of the conclusions and recommendations resulting from the engineering data in the
license applications and technical review and studies based on such data.

1-3
1-3.3.2 Review of Consultants Reports

Review of Board of Consultants Report - All licenses authorizing major construction

require the licensee to employ a board of qualified independent engineering consultants,
approved by the Director, D2SI, to review the design, plans and specifications, and
construction of the project. Also, the board is expected to assess the construction inspection
program, construction procedures and progress, planned instrumentation, the filling
procedures for the reservoir, and plans for surveillance during initial filling of the reservoir.
Staff review of consultants reports should examine all recommendations made by the Board.
Recommendations which are inconsistent with engineering guidelines, or previously stated
staff positions on a particular problem with the project, should be reviewed and the
differences resolved.

Review of Part 12, Subpart D Inspection Reports - Reference is made to the Operating
Manual which establishes the Commission's policy concerning Part 12 (Independent
Consultant 5-year) inspections. Specific guidance is given to Regional Directors concerning
the review of Consultants' Part 12 reports.

1-3.3.3 Review of Staff Studies

Independent analyses conducted by any member of the Staff shall be reviewed by another
staff member for completeness and appropriate application of analytical methods.

1-4 Deviations from the Guidelines

1-4.1 Changes

Guideline criteria and recommendations which are found to be technically incorrect, or

outdated, should be brought to the attention of the Director, D2SI. This shall be done in
writing with the incorrect or outdated passages cited, and shall include the Staff members'
recommendations for correcting the deficiency.

1-4.2 Deviations

Deviations from the guidelines shall be subject to the approval of the Director, D2SI. The
procedures, or criteria, used in lieu of guideline recommendations shall be justified in writing
for inclusion in the guideline files, and shall be accompanied by any suggested changes in
the guidelines that may be necessary to incorporate such procedures or criteria in future
revisions.

1-4
1-5 References

1. Federal Energy Guidelines, Statutes and Regulations, Federal Energy Regulatory

Commission, Subchapter B - Regulations Under the Federal Power Act, Parts 4
through 12.

2. Operating Manual for Inspection of Projects and Supervision of Licensees for

Water Power Projects.

1-5
 CHAPTER II

SELECTING AND ACCOMMODATING INFLOW

DESIGN FLOODS FOR DAMS

October 1993
 Chapter II

Selecting and Accommodating Inflow

Design Floods for Dams

2-0 Contents

Title Page

2-1 Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

2-2 Definition of Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

2-3 Determination of the Inflow Design Flood . . . . . . . . . . . . . . . . . . . . . . . . 2-3

2-3.1 Hazard Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4

2-3.1.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2-3.1.2 Defining the Hazard Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2-3.1.3 Evaluating the Consequences of Dam Failure . . . . . . . . . . . . . . . . 2-5
2-3.1.4 Studies to Define the Consequences of Dam Failure . . . . . . . . . . 2-6
2-3.1.5 Incremental Hazard Evaluation for Inflow
Design Flood Determination . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9
2-3.1.6 Criteria for Selecting the Inflow Design Flood . . . . . . . . . . . . . . . 2-10

2-3.2 Probable Maximum Floods for Dam Safety . . . . . . . . . . . . . . . . . 2-12

2-3.2.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12
2-3.2.2 Probable Maximum Precipitation . . . . . . . . . . . . . . . . . . . . . . . . . 2-12

2-3.3 Floods to Protect Against Loss of Benefits During the Life

of the Project - Applicable Only to Low Hazard Dams . . . . . . . . 2-14

2-i October 1993

 2-0 Contents
(Cont.)

Title Page

2-4 Accommodating Inflow Design Floods . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14

2-4.1 Flood Routing Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14

2-4.1.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
2-4.1.2 Guidelines for Initial Elevation Based on Storage Allocation . . . . 2-14
2-4.1.3 Reservoir Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
2-4.1.4 Reservoir Regulation Requirements . . . . . . . . . . . . . . . . . . . . . . . 2-15
2-4.1.5 Evaluation of Domino-like Failure . . . . . . . . . . . . . . . . . . . . . . . . 2-16

2-4.2 Spillway and Flood Outlet Selection and Design . . . . . . . . . . . . . . 2-17

2-4.2.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-17
2-4.2.2 Gated or Ungated Spillways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-17
2-4.2.3 Design Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18

2-4.3 Freeboard Allowances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-20

2-4.3.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-20
2-4.3.2 Freeboard Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-20

2-5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-23

2-6 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-25

Appendix II-A Dambreak Studies

Appendix II-B Hydrometeorological Report (HMR)
Nos. 51 and 52 vs. HMR No. 33
Appendix II-C Inflow Design Flood Selection Procedures

2-ii October 1993

 Chapter II

Selecting and Accommodating

Inflow Design Floods for Dams

2-1 Purpose and Scope

The purpose of this chapter of the Guidelines is to provide technical guidance for
determining the appropriate Inflow Design Flood (IDF) to be used in the review of spillway
and appurtenant structure designs and to conform to the provisions of the Federal Guidelines
for Dam Safety.

This chapter is not intended to provide a complete manual of all procedures used for
estimating inflow design floods for spillways, because the selection of procedures is
dependent upon available hydrologic data and individual watershed characteristics. All
studies submitted to the Commission should be performed by a competent engineer
experienced in hydrology and hydraulics, and should contain a summary of the design
assumptions, design analyses, and methodologies used to evaluate the inflow design flood.

2-2 Definition of Terms

This section contains definitions of some specialized technical terms used in this chapter:

Flood Routing - A process of determining progressively over time the amplitude of a flood
wave as it moves past a dam and continues downstream to successive points along a river or
stream.

Freeboard - Vertical distance between a specified stillwater reservoir surface elevation and
the top of the dam, without camber.

Hazard - A situation which creates the potential for adverse consequences such as loss of
life, property damage, or other adverse impacts. Impacts in the area downstream of a dam
defined by the flood waters released through spillways and outlet works of the dam or waters
released by partial or complete failure of the dam. There may also be impacts upstream of
the dam due to backwater flooding or landslides around the reservoir perimeter.

Hydrograph - A graphical representation of the streamflow stage or discharge as a function

of time at a particular point on a watercourse.

2-1 October 1993

Inflow Design Flood (IDF) - The floodflow above which the incremental increase in water
surface elevation due to failure of a dam or other water impounding structure is no longer
considered to present an unacceptable threat to downstream life or property. The IDF of a
dam or other water impounding structure flood hydrograph is used in the design of a dam and
its appurtenant works particularly for sizing the spillway and outlet works, and for
determining maximum height of a dam, freeboard, and temporary storage requirements.

Maximum Wind - The most severe wind for generating waves that is reasonably possible
at a particular reservoir. The determination will generally include results of meteorologic
studies which combine wind velocity, duration, direction, and seasonable distribution
characteristics in a realistic manner.

One Percent Chance Flood - A flood that has 1 chance in 100 of being equaled or exceeded
in a specified time period, usually 1 year.

Outlet Works - A dam appurtenance that provides release of water (generally controlled)
from a reservoir.

Probable Maximum Flood (PMF) - The flood that may be expected from the most severe
combination of critical meteorologic and hydrologic conditions that are reasonably possible
in the drainage basin under study. This is the upper limit for determining the IDF.

Probable Maximum Precipitation (PMP) - Theoretically, the greatest depth of precipitation

for a given duration that is physically possible over a given size storm area at a particular
geographical location during a certain time of the year.

Reservoir Regulation Procedure (Rule Curve) - Compilation of operating procedures that

govern reservoir storage and releases.

Spillway - A gated or ungated hydraulic structure used to discharge water from a reservoir.
Definition of specific types of spillways follow:

• Service Spillway. A spillway that is designed to provide continuous or frequent

regulated or unregulated releases from a reservoir without significant damage to either
the dam or its appurtenant structures.

• Auxiliary Spillway. Any secondary spillway which is designed to be operated

very infrequently; possibly, in anticipation of some degree of structural
damage or erosion to the spillway would occur during operation.

2-2 October 1993

• Emergency Spillway. A spillway that is designed to provide additional
protection against overtopping of dams and is intended for use under extreme
flood conditions or mis-operation or malfunction of the service spillway.

• Spillway Capacity - The maximum outflow flood which a dam can safely pass.

Stillwater Level - The elevation that a water surface would assume if all wave action were
absent.

Wave Runup - Vertical height above the stillwater level to which water from a specific wave
will run up the face of a structure or embankment.

Wind Setup - The vertical rise of the stillwater level at the face of a structure or embankment
caused by wind stresses on the surface of the water.

2-3 Determination of the Inflow Design Flood

The Commission's Order No. 122, issued January 21, 1981, states that the adequacy of a
spillway must be evaluated by considering the hazard potential which would result from
failure of the project works during flood flows. (See Section 1-2.2 of Chapter I of these
Guidelines for definition of hazard potential.) If failure of the project works would present
a threat to human life or would cause significant property damage, the project works must
be designed to either withstand overtopping or the loading condition that would occur during
a flood up to the probable maximum flood, or to the point where a failure would no longer
constitute a hazard to downstream life or property. In the alternative, the capacity of the
spillway must be adequate to prevent the reservoir from rising to an elevation that would
endanger the safety of the project works.

The Inflow Design Flood (IDF) is the flood flow above which the incremental increase in
water surface elevation due to failure of a dam or other water impounding structure is no
longer considered to present an unacceptable threat to downstream life and property.

The procedures used to determine whether or not the failure of a project would constitute a
threat to human life or could cause significant property damage vary with the physical
characteristics and location of the project.

Analyses of dam failures are complex with many historical dam failures not completely
understood. The principal uncertainties in determining outflow from a dam failure involve
the mode and degree of failure. These uncertainties can be circumvented in situations where
it can be shown that the complete and sudden removal of the dam would not endanger human

2-3 October 1993

life or cause extensive property damage. Otherwise, reasonable failure postulations and
sensitivity analyses such as those suggested in Appendix II-A should be used. Although a
study using the breach parameters suggested in Appendix II-A of this chapter may indicate
that a hazard does not exist, a hazard could exist for a more extensive mode of failure. If it
is judged that a more extensive mode of failure is possible, then an analyses should be done
to determine whether a need for remedial action is required. The possibility of more
extensive modes of failure should particularly be considered when failure is due to
overtopping.

2-3.1 Hazard Evaluation

A properly designed, constructed, and operated dam can be expected to improve the safety
of downstream developments during floods. However, the impoundment of water by a dam
can create a potential hazard to downstream developments greater than that which would
exist without the dam because of the potential for dam failure. There are several potential
causes of dam failure, including hydrologic, geologic, seismic, and structural. This chapter
of the Guidelines is limited to the selection of the IDF for the hydrologic design of a dam to
reduce the likelihood of failure from a flood occurrence to an acceptable level.

2-3.1.1 General

Once a dam is constructed, the downstream hydrologic regime may change, particularly
during flood events. The change in hydrologic regime could alter land use patterns to
encroach on a flood plain that would otherwise not be developed without the dam.
Consequently, evaluation of the consequences of dam failure must be based on the dam
being in place, and must compare the impacts of with-failure and without-failure
conditions on existing development and known and prospective future development.
Comparisons between existing downstream conditions with and without the dam are not
relevant.

2-3.1.2 Defining the Hazard Potential

The hazard potential of a dam pertains to the potential for loss of human life or property
damage in the area downstream of the dam in the event of failure or incorrect operation of
a dam. Hazard potential does not refer to the structural integrity of the dam itself, but rather
the effects if a failure should occur.

The hazard potential classification assigned to a dam (see Section 1-2.2, Chapter I, of
these Guidelines) should be based on the worst-case failure condition. That is, the
classification is based on failure consequences resulting from the failure condition that

2-4 October 1993

will result in the greatest potential for loss of life and property damage. For example, a
failure during normal operating conditions may result in the released water being confined
to the river channel, indicating a low hazard potential. However, if the dam were to fail
during a floodflow condition, and the result would be a potential loss of life or serious
damage to property, the dam would have high hazard potential classification.

In many cases, the hazard potential classification can be determined by field investigations
and a review of available data, including topographic maps. However, when the hazard
potential classification is not apparent from a field reconnaissance, detailed studies, including
dambreak analyses, are required for various floodflow conditions to evaluate the incremental
effects of a failure of a dam in order to identify the flood level above which the consequences
of failure become acceptable--that is, the floodflow condition above which the additional
incremental increase in elevation due to failure of a dam is no longer considered to present
an unacceptable threat to downstream life and property.

The selection of the appropriate IDF for a dam is related to the hazard classification for the
dam. The IDF for a dam having a low hazard potential is selected primarily to protect
against loss of the dam and its benefits should a failure occur. The IDF for high and
significant hazard potential dams is the maximum flood above which there are no significant
incremental impacts on downstream life and property.

2-3.1.3 Evaluating the Consequences of Dam Failure

The possible consequences resulting from a dam failure include loss of human life;
economic, social, and environmental impacts; damage to national security installations; and
political and legal ramifications. Estimates of the potential for loss of human life and the
economic impacts of damage resulting from dam failure are the usual bases for defining
hazard potential. Social and environmental impacts, damage to national security
installations, and political and legal ramifications are not easily evaluated, and are more
susceptible to subjective or qualitative evaluation. Therefore, these other considerations do
not usually affect decisions on hazard potential. Because their actual impacts cannot be
clearly defined, particularly in economic terms, their consideration as factors for
determining the hazard potential rating must be on a case-by-case basis, as determined by
the Regional Director in consultation with the Director or Deputy Director, D2SI.

The following factors should be evaluated regarding potential for loss of human life when
estimating the potential for fatalities resulting from dam failure:

• The number and location of habitable structures within the potential

area inundated by dam failure. The presence of public facilities within

2-5 October 1993

 the potential area inundated by dam failure that would attract people on
a temporary basis (e.g., improved campgrounds, organized or
unorganized recreation areas, State or national parks, etc.) requires
special consideration.

• Type of flow conditions based on water depths, temperatures and velocities,

rate of rise of the flood wave, duration of floodflow, and special hazardous
conditions such as the presence of surface waves, debris flow or terrain
conditions which may increase potential for loss of lives.

The evaluation of the economic impacts of failure should consider damages to residences;
commercial property; industrial property; public utilities and facilities including transmission
lines and substations; transportation systems; agricultural buildings, lands, and equipment;
dams; and loss of production and other benefits from project operation.

In summary, in most situations the investigation of the impacts of failure on downstream

life and property is sufficient in itself to determine the appropriate hazard potential
rating and to select the appropriate IDF for a project. However, in determining the
appropriate IDF for a project, there could be circumstances beyond loss of life and property
damage, particularly when a failure would have minimal or no impact on downstream life
and property, that would dictate using a more conservative hazard potential rating and IDF.
For example, the reservoir of a dam that would normally be considered to have a low hazard
potential based on insignificant incremental increases (in elevation) due to a failure may be
known to contain extensive toxic sediments. If released, those toxic sediments would be
detrimental to the eco-system. Therefore, a low hazard potential rating would not be
appropriate. Instead, a higher standard should be used for selecting the hazard potential
rating and IDF.

2-3.1.4 Studies to Define the Consequences of Dam Failure

The degree of study required to sufficiently define the impacts of dam failure for selecting
an appropriate IDF will vary with the extent of existing and potential downstream
development, the size of reservoir (depth and storage volume), and type of dam. Evaluation
of the river reach and areas impacted by a dam failure should proceed only until sufficient
information is generated to reach a sound decision or there is a good understanding of the
consequences of failure. In some cases, it may be apparent, from a field inspection or a
review of aerial photographs, Flood Insurance Rating Maps, and recent topographic maps,
that loss of life and extensive economic impacts attributable to dam failure would occur and
be unacceptable. In other cases, detailed studies including dambreak analyses will be
required. It may also be necessary to perform field surveys to determine the basement and

2-6 October 1993

first floor elevations of potentially affected habitable structures (residential, commercial,
etc.).

When conducting dambreak studies, the consequences of the incremental increase due to
failure under both normal (full reservoir with normal streamflow conditions prevailing) and
floodflow conditions up to the point where a dam failure would no longer significantly
increase the threat to life or property should be considered. For each flood condition, water
surface elevations with and without dam failure, flood wave travel times, and rates of rise
should be determined. This evaluation is known as an incremental hazard evaluation. Since
dambreak analyses and flood routing studies do not provide precise results, evaluation of the
consequences of failure should be reasonably conservative.

The upper limit of flood magnitude to be considered in an IDF evaluation is the Probable
Maximum Flood (PMF) (see Chapter VIII of these Guidelines).

The type of dam and the mechanism that could cause failure require careful consideration
if a realistic breach is to be assumed. Special consideration should be given to the following
factors:

• Size and shape of the breach,

• Time of breach formation,
• Hydraulic head, and
• Storage in the reservoir.
• Reservoir inflow

In addition, special cases where a dam failure could cause domino-like failure of downstream
dams resulting in a cumulative flood wave large enough to cause a threat should be
considered.

The area affected by dam failure is the additional area inundated by the incremental increase
in flood levels over that which would occur by natural flooding with the dam in place. The
area affected by a flood wave resulting from a theoretical dam breach is a function of the
height of the flood wave and the length and width of the river at a particular location. An
associated and important factor is the flood wave travel time. These elements are primarily
a function of the rate and extent of dam failure, but also are functions of channel and
floodplain geometry and roughness and channel slope.

The flood wave should be routed downstream to the point where the incremental effect of
a failure will no longer constitute a threat to life or property. When routing a dambreak flood
through the downstream reaches, appropriate local inflows should be considered in the

2-7 October 1993

computations. Downstream concurrent inflows can be determined using one of the following
approaches:

• Concurrent inflows can be based on historical records, if these records indicate that
the tributaries contributing to the reservoir volume are characteristically in flood stage
at the same time that flood inflows to the reservoir occur. Concurrent inflows based
on historical records should be adjusted so they are compatible with the magnitude
of the flood inflow computed for the dam under study.

• Concurrent inflows can be developed from flood studies for downstream reaches
when they are available. However, if these concurrent floods represent inflows to a
downstream reservoir, suitable adjustments must be made to properly distribute flows
among the tributaries.

• Concurrent inflows may be assumed equal to the mean annual flood (approximately
bankfull capacity) for the channel and tributaries downstream from the dam. The
mean annual flood can be determined from flood flow frequency studies. As the
distance downstream from the dam increases, engineering judgment may be required
to adjust the concurrent inflows selected.

In general, the study should be terminated when the potential for loss of life and significant
property damage caused by routing floodflows appears limited. This point could occur
when:

• There are no habitable structures, and anticipated future development in the

floodplain is limited,
• Floodflows are contained within a large downstream reservoir,
• Floodflows are confined within the downstream channel, or
• Floodflows enter a bay or ocean.

The failure of a dam during a particular flood may increase the area flooded and also alter
the flow velocity and depth of flow as well as the rate of rise of flood flows. These changes
in flood flows could also affect the amount of damage. To fully evaluate the hazard created
by a dam, a range of flood magnitudes needs to be examined. Water surface profiles, flood
wave travel times, and rates of rise should be determined for each condition.

The results of the downstream routing should be clearly shown on inundation maps with the
breach wave travel time indicated at critical downstream locations. The inundation maps
should be developed at a scale sufficient to identify downstream habitable structures within
the impacted area. Guidance on inundation map requirement appears in Section 6-2.3 of

2-8 October 1993

Chapter VI of these Guidelines and in the Commission's Revised Emergency Action Plan
Guidelines issued February 22, 1988, located in Appendix VI-C of Chapter VI.

Dambreak studies should be performed in accordance with one or more of the techniques
presented in Appendix II-A and Section 6-2 of Chapter VI of these Guidelines.

The most widely used and recommended method for dambreak analysis is the unsteady
flow and dynamic routing method used in the National Weather Service Dambreak
model. In fact, the Corps of Engineers Hydrologic Engineering Center (HEC) HEC-I
Manual defers to the NWS DAMBRK model when studies require higher levels of accuracy.
The NWS FLOODWAV model, released in 1993, combines the NWS DAMBRK model with
the NWS DWOPER model. FLOODWAV is also recommended as a preferred model for
dambreak analysis.

Most of the methods used for estimating dambreak hydrographs, including the widely used
NWS Dambreak Model, require selecting the size, shape, and time of formation of the dam
breach as input parameters for the computations. Therefore, sensitivity analyses are
considered necessary. Sensitivity analyses, based on varying flood inflow conditions and
breach parameters, should be performed only to the extent necessary to make a decision.

2-3.1.5 Incremental Hazard Evaluation for Inflow Design Flood Determination

The IDF is determined through an iterative process known as an incremental hazard

evaluation. In other words, to evaluate the incremental increase in consequences due to dam
failure, you would begin with the normal inflow condition and the reservoir at normal full
reservoir level with normal streamflow conditions prevailing. That condition should be
routed through the dam and downstream areas, with the assumption that the dam remains in
place. The same flow should then be routed through the dam with the assumption that the
dam fails.

The incremental increase in downstream water surface elevation between the with-failure and
without-failure conditions should then be determined (in other words, how much higher
would the water downstream be if the dam failed than if the dam did not fail?). The amount
of damage that could result should then be identified. If the incremental rise in flood water
downstream indicates an additional threat to downstream life and/or property, assess the need
for remedial action.

If the study under normal flow conditions indicates no adverse consequences, the same
analyses should be done for several larger flood levels to determine the greatest unacceptable
threat to downstream life and/or property. Under each incrementally larger inflow condition

2-9 October 1993

identify the consequences of failure. For each larger assumed flood inflow condition (which
can be percentages of the PMF):

• assume the dam remains in place during the non-failure conditions, and
• assume the dam fails when the peak reservoir elevation is attained for the assumed
inflow condition.

It is not appropriate to assume that a dam fails on the rising limb of the inflow
hydrograph. For example, current methods available cannot accurately determine the extent
of overtopping that an earth dam can withstand or how rapidly the dam will erode and
ultimately breach from overtopping. Therefore, until such methodologies are available and
proven, a conservative approach should be followed that assumes that failure occurs at the
peak of the flood hydrograph. The assumption should also be made that the dam has been
theoretically modified to contain or safely pass all lower inflow floods. This is an
appropriate assumption since this procedure requires that the dambreak analyses start at the
normal operating condition, with incremental increases in the flood inflow condition for each
subsequent failure scenario up to the point where a failure no longer constitutes a threat to
downstream life and property. In summary, before one selects larger floods for analysis, you
should determine that failure at a lower flood constituted a threat to downstream life and
property.

The above procedure should be repeated until the flood inflow condition is identified such
that a failure at that flow or larger flows (up to the PMF) will no longer result in an
additional hazard to downstream life and property. The resultant flood flow is the IDF for
the project. The maximum IDF is always the PMF, but in many cases the IDF will be
substantially less than the PMF.

It is important to investigate the full range of flood flow conditions to verify that a failure
under flood flows larger than the selected IDF up through the PMF will not result in any
additional hazard. In addition, once the design for remedial repairs is selected, the IDF
should be verified for that design.

Appendix II-C provides specific guidance and procedures, including a comprehensive

flowchart, for conducting an incremental hazard evaluation to select the appropriate
IDF for a dam and determine the need for remedial measures.

2-3.1.6 Criteria for Selecting the Inflow Design Flood

The selection of the appropriate IDF for a dam is related to the hazard potential classification
and is the result of the incremental hazard evaluation.

2-10 October 1993

There is not a separate IDF for each different section of a dam. A dam is assigned only one
IDF, and it is determined based on the consequences of failure of the section of the dam that
creates the worst hazard potential downstream. This should not, however, be confused with
the design criteria for different sections of a dam which may be based on the effect of their
failure on downstream areas.

The criteria for selecting an IDF for the design of a dam requires consideration of the
consequences of dam failure under both normal and flood flow conditions.

The PMF should be adopted as the IDF in those situations where consequences attributable
to dam failure for flood conditions less than the PMF are unacceptable. The determination
of unacceptability clearly exists when the area affected is evaluated and indicates there is a
potential for loss of human life and extensive property damage.

A flood less than the PMF may be adopted as the IDF in those situations where the
consequences of dam failure at flood flows larger than the selected IDF are acceptable. In
other words, where detailed studies conclude that the risk is only to the dam owners' facilities
and no increased damage to downstream areas is created by failure, a risk-based approach
is acceptable. Generally, acceptable consequences exist when evaluation of the area affected
indicates:

• There are no permanent human habitations, or known national security installations,

commercial or industrial development, nor are such habitations, or commercial or
industrial developments projected to occur within the potential hazard area in the
foreseeable future.

• There are permanent human habitations within the potential hazard area that would
be affected by failure of the dam, but there would be no significant incremental
increase in the hazard to life or property resulting from the occurrence of a
failure during floods larger than the proposed IDF. For example, if an
impoundment has a small storage volume and failure would not add appreciably to
the volume of the outflow flood hydrograph, it is likely that downstream inundation
would be essentially the same with or without failure of the dam.

The consequences of dam failure may not be acceptable if the hazard potential to these
habitations is increased appreciably by the failure flood wave or level of inundation. When
a dambreak analysis shows downstream incremental effects of approximately two feet
or more, engineering judgment and further analysis will be necessary to finally evaluate
the need for modification to the dam. In general, the consequences of failure are
considered acceptable when the incremental effects of failure on downstream structures

2-11 October 1993

are approximately two feet or less. However, the two-foot increment is not an absolute
decision-making point. Sensitivity analyses and engineering judgment are the tools used
in making final decisions. For example, if it is determined that a trailer sitting on blocks can
be moved and displaced by as little as six inches of water, then the acceptable incremental
impact would be much less than two feet. As a second example, if a sensitivity analysis
demonstrates that the largest breach width recommended by this chapter is the only condition
that results in an incremental rise of two feet, then engineering judgment becomes necessary
to determine whether a smaller breach having acceptable consequences of failure is more
realistic for the given conditions (e.g. flow conditions, characteristics of dam, velocity in
vicinity of structures, location and type of structures).

In addition, selection of the appropriate magnitude of the IDF may include consideration of
whether a dam provides vital community services such as municipal water supply or energy.
Therefore, a higher degree of protection may be required against failure to ensure those
services are continued during and following extreme flood conditions when alternate services
are unavailable. If the economic risk of losing such services is acceptable, the IDF can be
less conservative. However, loss of water supply for domestic purposes may not be an
acceptable public health risk.

2-3.2 Probable Maximum Floods for Dam Safety

The PMF is the upper limit of floods to be considered when selecting the appropriate IDF
for a dam.

2-3.2.1 General

A deterministic approach should be used to determine the PMF. In the deterministic

approach, a flood hydrograph is generated by modeling the physical atmospheric and
drainage basin hydrologic and hydraulic processes. The approach attempts to represent the
most severe combination of meteorologic and hydrologic conditions considered reasonably
possible for a given drainage basin. The PMF represents an estimate of the upper limit of
run-off that is capable of being produced on the watershed. Chapter VIII of these
Guidelines provides criteria for determining the PMF.

2-3.2.2 Probable Maximum Precipitation (PMP)

The concept that the PMP represents an upper limit to the level of precipitation the
atmosphere can produce has been stated in many hydrometeorological documents. The
commonly used approach in deterministic PMP development for non-orographic regions is
to determine the limiting surface dew point temperature (used to obtain the moisture

2-12 October 1993

maximization factor) and collect a "sufficient" sample of extreme storms. The latter is done
through a method known as storm transposition, i.e., the adjustment of moisture observed in
a storm at its actual site of occurrence to the corresponding moisture level at the site from
which the PMP is to be determined. Storm transposition is based on the concept that all
storms within a meteorologically homogeneous region could occur at any other location
within that region with appropriate adjustments for effects of elevation and moisture supply.
The maximized transposed storm values are then enveloped both depth-durationally and
depth-areally to obtain PMP estimates for a specific basin. Several durations of PMP should
be considered to ensure the most appropriate duration is selected.

In orographic regions, where local influences affect the delineation of meteorological

homogeneity, transposition is generally not permitted. Alternative procedures are offered
for these regions that are less reliant on the adequacy of the storm sample. Most of these
procedures involve development of both non-orographic and orographic components
(sometimes an orographic intensification factor is used) of PMP. Orographic and
non-orographic PMP's are then combined to obtain total PMP estimates for an orographic
basin.

To date, no single orographic procedure has been developed that offers universal
applicability. These techniques have been discussed at length in various National Weather
Service (NWS) reports and in the Manual for Estimation of PMP (WMO, 1986). Currently,
PMP estimates are available for the entire conterminous United States, as well as Alaska,
Hawaii, and Puerto Rico.

As our understanding and the availability of data increases, the "particular" PMP estimates
that appear in NWS Hydrometeorological Reports may require adjustment in order to better
define the conceptual PMP for a specific site. Therefore, it is appropriate to refine PMP
estimates with site specific or regional studies performed by a qualified
hydrometeorologist with experience in determining PMP. The results of available
research such as that developed by the Electric Power Research Institute for the Wisconsin
and Michigan areas should be considered in performing site specific studies. Since these
studies can become very time consuming and costly, the benefit of a site specific study must
be carefully considered.

See Appendix IIB for guidelines adopted by FERC staff on the use of Hydrometeorological
Report (HMR) Nos. 51 and 52 vs. HMR No. 33.

2-13 October 1993

2-3.3 Floods to Protect Against Loss of Benefits During the
Life of the Project - Applicable Only to Low Hazard Dams

Dams identified as having a low hazard potential should be designed to at least meet a
minimum standard to protect against the risk of loss of benefits during the life of the project;
to hold 0&M costs to a reasonable level; to maintain public confidence in owners and
agencies responsible for dam safety; and to be in compliance with local, State, or other
regulations applicable to the facility. Flood frequency and risk base analyses may be used
for this analysis. Generally, it would not be an appropriate risk to design a dam having a low
hazard potential for a flood frequency of less than 100 years. (See Section 3-3.3 of Chapter
III of these Guidelines.)

2-4 Accommodating Inflow Design Floods

2-4.1 Flood Routing Guidelines

2-4.1.1 General

Site-specific considerations should be used to establish flood routing criteria for each dam
and reservoir. The criteria for routing the IDF should be consistent with the reservoir
regulation procedure that is to be followed in actual operation. General guidelines to be used
in establishing criteria follow.

2-4.1.2 Guidelines for Initial Elevations

Specific guidance for establishing the initial reservoir elevation during the PMF is provided
in Section 8-3.1 of Chapter VIII of these Guidelines. This criteria should also be applied to
routing the IDF when the IDF is less than the PMF. In general, if there is no allocated or
planned flood control storage (e.g. run-of-river), the flood routing usually begins with the
reservoir at the normal maximum pool elevation. If regulation studies show that pool levels
would be lower than the normal maximum pool elevation during the critical IDF season, then
the results of those specific regulation studies would be analyzed to determine the
appropriate initial pool level for routing the IDF.

2-4.1.3 Reservoir Constraints

Flood routing criteria should recognize constraints that may exist on the maximum desirable
water surface elevation. A limit or maximum water surface reached during a routing of the
IDF can be achieved by providing spillways and outlet works with adequate discharge
capacity. Backwater effects of floodflow into the reservoir must specifically be considered

2-14 October 1993

when constraints on water surface elevation are evaluated. Reservoir constraints may include
the following:

• Topographic limitations on reservoir stage which exceed the economic limits of dike
construction.

• Public works around the reservoir rim which are not to be relocated, such as water
supply facilities and sewage treatment plants.

• Dwellings, factories, and other developments around the reservoir rim which are not
to be relocated.

• If there is a loss of storage capacity caused by sediment accumulation in portions of

the reservoir, then this factor should be accounted for in routing the IDF. Sediment
deposits in reservoir headwater areas may build up a delta which can increase
flooding in that area, as well as reduce flood storage capacity, thereby having an
effect on routings.

• Geologic features that may become unstable when inundated, and result in landslides
which would threaten the safety of the dam, domestic and/or other developments, or
displace needed storage capacity.

• Flood plain management plans and objectives established under Federal or State
regulations and/or authorities.

2-4.1.4 Reservoir Regulation Requirements

Considerations to be evaluated when establishing flood routing criteria for a project include:

(1) regulation requirements to meet project purposes;

(2) the need to impose a maximum regulated release rate to prevent flooding
or erosion of downstream areas and control rate of drawdown;
(3) the need to provide a minimum regulated release capacity to recover flood
control storage for use in regulating subsequent floods; and
(4) the practicability of evacuating the reservoir for emergencies and for
performing inspection, maintenance, and repair.

Spillways, outlet works, and penstocks for powerplants are sized to satisfy project
requirements and must be operated in accordance with specific instructions if these project
works are relied upon to make flood releases, subject to the following limitations:

2-15 October 1993

• Only those release facilities which can be expected to operate reliably under the
assumed flood condition should be assumed to be operational for flood routing.
Reliability depends upon structural competence and availability for use. Availability
and reliability of generating units for flood release during major floods should be
justified. Availability of a source of auxiliary power for gate operation, effects of
reservoir debris on operability and discharge capacity of gates and other facilities,
accessibility of controls, design limits on operating head, reliability of access roads,
and availability of operating personnel at the site during flood events are other factors
to be considered in determining whether to assume release facilities are operational.

• A positive way of making releases to the natural watercourse by use of a bypass or

wasteway must be available if canal outlets are to be considered available for making
flood releases.

• Bypass outlets for generating units may be used if they are or can be isolated from the
turbines by gates or valves.

• In flood routing, assumed releases are generally limited to maximum values

determined from project uses, by availability of outlet works, tailwater conditions
including effects of downstream tributary inflows and wind tides, and downstream
nondamaging discharge capacities until allocated storage elevations are exceeded.
When a reservoir's capacity in regulating flows is exceeded, then other factors,
particularly dam safety, will govern releases.

• During normal flood routing, the rate of outflow from the reservoir should not exceed
the rate of inflow until the outflow begins to exceed the maximum project flood
discharge capacity at normal pool elevation, nor should the maximum rate of increase
of outflow exceed the maximum rate of increase of inflow. This is to prevent outflow
conditions from being more severe than pre-dam conditions. An exception to the
preceding would be the case where streamflow forecasts are available and pre-flood
releases could reduce reservoir levels to provide storage for flood flows.

2-4.1.5 Evaluation of Domino-like Failure

If one or more dams are located downstream of the site under review, the failure wave should
be routed downstream to determine if any of the downstream dams would breach in a
domino-like action. The flood routing of flows entering the most upstream of a series of
such dams may be either dynamic or level pool. The routing through all subsequent
downstream reservoirs should be dynamic. Tailwater elevations should consider the effect
of backwater from downstream constrictions.

2-16 October 1993

2-4.2 Spillway and Flood Outlet Selection and Design

2-4.2.1 General

Spillways and flood outlets should be designed to safely convey major floods to the
watercourse downstream from the dam and to prevent overtopping of the dam. They are
selected for a specific dam and reservoir on the basis of release requirements, topography,
geology, dam safety, and project economics.

2-4.2.2 Gated or Ungated Spillways

An ungated spillway releases water whenever the reservoir elevation exceeds the spillway
crest level. A gated spillway can regulate releases over a broad range of water levels.

Ungated spillways are more reliable than gated spillways. Gated spillways provide greater
operational flexibility and large discharge capacity per unit length. Operation of gated
spillways and/or their regulating procedures should generally ensure that the peak flood
outflow does not exceed the natural downstream flow that would occur without the dam.

The selection of a gated or ungated type of spillway for a specific dam depends upon site
conditions, project purposes, economic factors, costs of operation and maintenance, and
other considerations.

The following paragraphs focus on considerations that influence the choice between gated
and ungated spillways:

(1) Discharge capacity - For a given spillway crest length and maximum allowable water
surface elevation, a gated spillway can be designed to release higher discharges than
an ungated spillway because the crest elevation may be lower than the normal
reservoir storage level. This is a consideration when there are limitations on spillway
crest length or maximum water surface elevation.

(2) Project objectives and flexibility - Gated spillways permit a wide range of releases
and have capability for pre-flood drawdown.

(3) Operation and maintenance - Gated spillways may experience more operational
problems and are more expensive to construct and maintain than ungated spillways.
Constant attendance or several inspections per day by an operator during high water
levels is highly desirable for reservoirs with gated spillways, even when automatic or
remote controls are provided. During periods of major flood inflows where automatic

2-17 October 1993

 or remote controls are not provided, the spillway should be constantly manned. Gated
spillways are more subject to clogging from debris and jamming from ice, whereas,
properly designed ungated spillways are basically free from these problems. Gated
spillways require regular maintenance, and, as a minimum, an annual operation test
for safety purposes. However, ungated spillways can have flashboards, trip gates,
stop log sections, etc. which can have operational problems during floods and may
require constant attendance or several inspections per day during high water levels.

(4) Reliability - The nature of ungated spillways reduces dam failure potential associated
with improper operation and maintenance. Where forecasting capability is unreliable,
or where time from the beginning of runoff to peak inflow is only a few hours,
ungated spillways are more reliable, particularly for high hazard structures.
Consequences of failure of operation equipment or errors in operation are more severe
for gated spillways.

(5) Data and control requirements - Gated spillways require reliable real time hydrologic
and meteorologic data to make proper regulation possible.

(6) Emergency evacuation - Unless ungated spillways have removable sections such as
flashboards, trip gates, or stop log sections, they cannot be used to evacuate a
reservoir during emergencies. The capability of gated spillways to draw down pools
from the top of the gates to the spillway crest can be an advantage when emergency
evacuation to reduce head on the dam is a concern.

(7) Economics and selection - Designs to be evaluated should be technically adequate

alternatives. Economic considerations often indicate whether gated or ungated
spillways are selected. The possibility of selecting a combination of more than one
type of spillway is also a consideration. Final selection of the type of crest control
should be based on a comprehensive analysis of all pertinent factors, including
advantages, disadvantages, limitations, and feasibility of options.

2-4.2.3 Design Considerations

Dams and their appurtenant structures should be designed to give satisfactory performance
and to practically eliminate the probability of failure. These guidelines identify three
specific classifications of spillways (service, auxiliary, and emergency) and outlet works that
are used to pass floodwaters, each serving a particular function. The following paragraphs
discuss functional requirements.

2-18 October 1993

Service spillways should be designed for frequent use and should safely convey releases
from a reservoir to the natural watercourse downstream from the dam. Considerations must
be given to waterway freeboard, length of stilling basins, if needed, and amount of
turbulence and other performance characteristics. It is acceptable for the crest structure,
discharge channel (e.g., chute, conduit, tunnel), and energy dissipator to exhibit marginally
safe performance characteristics for the IDF. However, they should exhibit excellent
performance characteristics for frequent and sustained flows such as up to the 1 percent
chance flood event. Other physical limitations may also exist which have an effect on
spillway sizing.

Auxiliary spillways are usually designed for infrequent use and it is acceptable to sustain
limited damage during passage of the IDF. The design of auxiliary spillways should be
based on economic considerations and be subject to the following requirements:

• The auxiliary spillway should discharge into a watercourse sufficiently separated

from the abutment to preclude abutment damage and should discharge into the main
stream a sufficient distance downstream from the toe of the dam so that flows will not
endanger the dam's structural integrity or usefulness of the service spillway.

• The auxiliary spillway channel should either be founded in competent rock or an

adequate length of protective surfacing should be provided to prevent the spillway
crest control from degrading to the extent that it results in an unacceptable loss of
conservation storage or a large uncontrolled discharge which exceeds peak inflow.

Emergency spillways may be used to obtain a high degree of hydrologic safety with minimal
additional cost. Because of their infrequent use it is acceptable for them to sustain
significant damage when used and they may be designed with lower structural standards than
those used for auxiliary spillways.

An emergency spillway may be advisable to accommodate flows resulting from mis-

operation or malfunction of other spillways and outlet works. Generally, they are sized to
accommodate a flood smaller than the IDF. The crest of an emergency spillway should be
set above the normal maximum water surface (attained when accommodating the IDF) so it
will not overflow as a result of reservoir setup and wave action. The design of an emergency
spillway should be subject to the following limitations:

• The structural integrity of the dam should not be jeopardized by spillway operation.

• Large conservation storage volumes should not be lost as a result of degradation of

the crest during operation.

2-19 October 1993

• The effects of a downstream flood resulting from uncontrolled release of reservoir
storage should not be greater than the flood caused by the IDF without the dam.

Outlet works used in passing floods and evacuating reservoir storage space should be
designed for frequent use and should be highly reliable. Reliability is dependent on
foundation conditions which influence settlement and displacement of waterways, on
structural competence, on susceptibility of the intake and conduit to plugging, on hydraulic
effects of spillway discharge, and on operating reliability.

2-4.3 Freeboard Allowances

2-4.3.1 General

Freeboard provides a margin of safety against overtopping failure of dams. It is generally

not necessary to prevent splashing or occasional overtopping of a dam by waves under
extreme conditions. However, the number and duration of such occurrences should not
threaten the structural integrity of the dam, interfere with project operation, or create hazards
to personnel. Freeboard provided for concrete dams can be less conservative than for
embankment dams because of their resistance to wave damage or erosion. If studies
demonstrate that concrete dams can withstand the PMF while overtopped without significant
erosion of foundation or abutment material, then no freeboard should be required for the
PMF condition. Special consideration may be required in cases where a powerplant is
located near the toe of the dam. The U.S. Bureau of Reclamation has developed guidelines
(Ref. 12) that provide criteria for freeboard computations.

Normal freeboard is defined as the difference in elevation between the top of the dam and
the normal maximum pool elevation. Minimum freeboard is defined as the difference in pool
elevation between the top of the dam and the maximum reservoir water surface that would
result from routing the IDF through the reservoir. Intermediate freeboard is defined as the
difference between intermediate storage level and the top of the dam. Intermediate freeboard
may be applicable when there is exclusive flood control storage.

2-4.3.2 Freeboard Guidelines

Following are guidelines for determining appropriate freeboard allowances:

• Freeboard allowances should be based on site-specific conditions and the type of dam
(concrete or embankment).

2-20 October 1993

• Both normal and minimum freeboard requirements should be evaluated in
determining the elevation of the top of the dam. The resulting higher top of dam
elevation should be adopted for design.

• Freeboard allowances for wind-wave action should be based upon the most reliable
wind data available that are applicable to the site. The significant wave should be the
minimum used in determining wave runup; and the sum of wind setup and wave
runup should be used for determining requirements for this component of freeboard.

• Computations of wind-generated wave height, setup, and runup should incorporate

selection of a reasonable combined occurrence of pool level, wind velocity, wind
direction, and wind durations based on site-specific studies.

• It is highly unlikely that maximum winds will occur when the reservoir water surface
is at its maximum elevation resulting from routing the IDF, because the maximum
level generally persists only for a relatively short period of time (a few hours).
Consequently, winds selected for computing wave heights should be appropriate for
the short period the pool would reside at or near maximum levels.

• Normal pool levels persist for long periods of time. Consequently, maximum winds
should be used to compute wave heights.

• Freeboard allowance for settlement should be applied to account for consolidation of

foundation and embankment materials when uncertainties exists in computational
methods or data used yield unreliable values for camber design. Freeboard allowance
for settlement should not be applied where an accurate determination of settlement
can be made and is included in the camber.

• Freeboard allowance for embankment dams for estimated earthquake-generated

movement, resulting seiches, and permanent embankment displacements or
deformations should be considered if a dam is located in an area with potential for
intense seismic activity.

• Reduction of freeboard allowances on embankment dams may be appropriate for

small fetches, obstructions that impede wave generation, special slope and crest
protection, and other factors.

• Freeboard allowance for wave and volume displacement due to potential landslides
which cannot be economically removed or stabilized should be considered if a
reservoir is located in a topographic setting where the wave or higher water resulting

2-21 October 1993

 from displacement may be destructive to the dam or may cause serious downstream
damage.

• Total freeboard allowances should include only those components of freeboard which
can reasonably occur simultaneously for a particular water surface elevation.
Components of freeboard and combinations of those components which have a
reasonable probability of simultaneous occurrence are listed in the following
paragraphs for estimating minimum, normal, and intermediate freeboards. The top
of the dam should be established to accommodate the most critical combination of
water surface and freeboard components from the following combinations.

For minimum freeboard combinations the following components, when they can reasonably
occur simultaneously, should be added to determine the total minimum freeboard
requirement:

(1) Wind-generated wave runup and setup for a wind appropriate for maximum reservoir
stage for the IDF.

(2) Effects of possible malfunction of spillway and/or outlet works during routing of the
IDF.

(3) Settlement of embankment and foundation not included in crest camber.

(4) Landslide-generated waves and/or displacement of reservoir volume (only cases

where landslides are triggered by the occurrence of higher water elevations and
intense precipitation associated with the occurrence of the IDF).

For normal freeboard combinations, the most critical of the following two combinations of
components should be used for determining normal freeboard requirements:

(1) Combination 1

(a) Wind-generated wave runup and setup for maximum wind, and

(b) Settlement of embankment and foundation not included in camber.

(2) Combination 2

(a) Landslide-generated waves and/or displacement of reservoir

volume;

2-22 October 1993

 (b) Settlement of embankment and foundation not included in camber;
and

(c) Settlement of embankment and foundation or seiches as a result of

the occurrence of the maximum credible earthquake.

For intermediate freeboard combinations, in special cases, a combination of intermediate

winds and water surface between normal and maximum levels should be evaluated to
determine whether this condition is critical. This may apply where there are exclusive flood
control storage allocations.

2-5. References

1. American Nuclear Society (1981). Determining Design Basis Flooding at Power

Reactor Sites. ANSI/ANS-2.8 1981.

2. Committee on Safety Criteria for Dams (1985). Safety of Dams - Flood and
Earthquake Criteria. Prepared under the Auspices of Water Science and Technology
Board, Commission on Engineering and Technical Systems, National Research
Council, Washington, D.C.: National Academy Press, 374 pp.

3. Federal Emergency Management Agency (1984). Federal Guidelines for Selecting

and Accommodating Inflow Design Floods for Dams. Prepared by Working Group
on Inflow Design Flood Subcommittee 1 of Interagency Committee on Dam Safety.

4. Federal Energy Regulatory Commission (1993). Operating Manual for Inspection of

Projects and Supervision of Licenses for Water Power Projects.

5. Hydrology Subcommittee, (1981). Estimating Peak Flow Frequencies for Natural

Ungaged Watersheds: A Proposed Nationwide Test. U.S. Water Resources Council,
346 pp.

6. Hydrology Subcommittee, (1982). Guidelines for Determining Flood Flow

Frequency. Hydrology Subcommittee Bulletin 17B, with editorial corrections.
Interagency Advisory Committee on Water Data, U.S. Geological Survey, 28 pp.

7. Interagency Committee on Water Data (1985 Draft). Feasibility of Assigning a

Probability to the Probable Maximum Flood. Prepared by Working Group of
Hydrology Committee.

2-23 October 1993

8. International Symposium on Flood Frequency and Risk Analyses at Louisiana State
University, Baton Rouge, La. (May 14-17, 1986). Proceedings to be published late
1986.

9. Lane, W. L., (1985). Rare Flood Frequency Estimation - A Case Study of the Pecos
River, (Abstract). EOS Transactions of AGU, 66(18), 267.

10. Myers, V.A., (1969). The Estimation of Extreme Precipitation as the Basis for
Design Floods - Resume of Practice in the United States. Proceedings of the
Leningrad Symposium on Floods and Their Computation, August 1967, Vol. 1,
International Association of Scientific Hydrology, Gentbrugge, Belgium, 84-101.

11. Naef, F., (1981). Can We Model the Rainfall-Runoff Process Today? Hydrological
Sciences Bulletin, 26.

12. U.S. Bureau of Reclamation, (1981). Freeboard Criteria and Guidelines for
Computing Freeboard Allowances for Storage Dams. ACER Technical Memorandum
No. 2.

2-24 October 1993

2.6 APPENDICES

2-25 October 1993

 Appendix II-A

DAMBREAK STUDIES

October 1993
APPENDIX II-A

Dambreak Studies

The evaluation of the downstream consequences in the event of a dam failure is a main
element in determining hazard potential and formulating emergency action plans for
hydroelectric projects. The solution requires knowledge of the lateral and longitudinal
geometry of the stream, its frictional resistance, a discharge-elevation relationship at one
boundary, and the time-varying flow or elevation at the opposite boundary.

The current state-of-the-art is to use transient flow or hydraulic methods to predict dambreak
wave formation and downstream progression. The transient flow methods solve and
therefore account for the essential momentum forces involved in the rapidly changing flow
caused by a dambreak. Another technique, referred to as storage routing or the hydrologic
method, solves one-dimensional equations of steady flow ignoring the pressure and
acceleration contributions to the total momentum force. For the same outflow hydrograph,
the storage routing procedures will always yield lower water surface elevations than
hydraulic or transient flow routing.

When routing a dambreak flood through the downstream reaches appropriate local inflows
should be included in the routing which are consistent with the assumed storm centering.

The mode and degree of dam failure involves considerable uncertainty and cannot be
predicted with acceptable engineering accuracy; therefore, conservative failure postulations
are necessary. Uncertainties can be circumvented in situations where it can be shown that
the complete and sudden removal of a dam (or dams) will not endanger human life or cause
significant property damage.

The following provides references on dambreak analyses and criteria which may prove useful
as indicators of reasonableness of the breach parameters, peak discharge, depth of flow, and
travel time determined by the licensee. In addition, Section 6-2 and Appendix VI-C of
Chapter VI of these Guidelines provides additional criteria on analytical requirements
for dambreak analyses.

I. REFERENCES

Suggested acceptable references regarding dam failure studies include the following:

A. Fread, D. L. "DAMBRK - The NWS Dam-Break Flood Forecasting Model,"

National Weather Service, Silver Spring, Maryland, 1988 Version. This (or

2-A-1 October 1993

 the most recent version) is the preferred method for performing dambreak
studies.

B. Fread, D. L. "NWS FLDWAV Model: The Replacement of DAMBRK for Dam-

Break Flood Prediction", Proceedings, Association of State Dam Safety Officials,
10th Annual Conference, Kansas City, Missouri, September 26-29, 1993. Since this
model combines the NWS DAMBRK model and the NWS DWOPER model, it is
also considered the preferred method.

C. Westmore, Jonathan N. and Fread, Danny L., "The NWS Simplified Dam-
Break Flood Forecasting Model," National Weather Service, Silver Spring,
Maryland, 1981. (Copy previously furnished to each Regional Office with a
detailed example).

D. Fread, D. L., 1977: The development and testing of a dam-break flood

forecasting model, "Proceedings, Dam-Break Flood Modeling Workshop,"
U.S. Water Resources Council, Washington, D.C., 1977, pp. 164-197.

E. Hydrologic Engineering Center, "Flood Hydrograph Package (HEC-1) Users

Manual for Dam Safety Investigations," September, 1990.

F. Gandlach, D. L. and Thomas, W. A., "Guidelines for Calculating and Routing

a Dam-Break Flood," Research Note No. 5, U.S. Army Corps of Engineers,
Hydrologic Engineering Center, 1977.

G. Cecilio, C. B. and Strassburger, A. G., "Downstream Hydrograph from Dam

Failure," Engineering Foundation Conference on Evaluation of Dam Safety,
1976.

H. Soil Conservation Service, "Simplified Dam-Breach Routing Procedure,"

March 1979. (To be used only for flood routing technique, not dambreak
discharge).

I. Chow, V. T., Open Channel Hydraulics, McGraw-Hill Book Company, Inc.,

New York, 1959, Chapter 20.

J. Henderson, F. M., Open Channel Flow, McMillan Company, New York,

1966, Chapters 8 and 9.

2-A-2 October 1993

K. Hydrologic Engineering Center, "Flood Emergency Plans, Guidelines for
Corps Dam," June 1980. (Forwarded to all Regional Engineers by
memorandum dated February 11, 1981).

L. Hydrologic Engineering Center, "UNET, One-Dimensional Unsteady Flow Through

a Full Network of Open Channels", September 1992.

II. CRITERIA

The following criteria may prove useful as an indicator of the reasonableness of a dambreak
study:

A. If the dambreak analysis has been performed by an acceptable method (References A

and B are the preferred methods), then generally only the breach parameters, peak
discharge, and flood wave travel time should be verified as an indicator of the licensee's
correct application of the method selected. Downstream routing parameters (i.e., Manning's
"n") should be reviewed for acceptability and inundation maps should be reviewed for clarity
and completeness of information (i.e., travel times). The following criteria are considered
to be adequate and appropriate for verifying the selected breach parameters and peak
discharge:

1. Breach Parameters - Most serious dam failures result in a situation resembling weir
conditions. Breach width selection is judgmental and should be made based on the channel
or valley width with failure occurring at the deepest section. The bottom of the breach
should generally be assumed to be at the foundation elevation of the dam. Pages 2-A-8
through 2-A-11 of this appendix contain suggested breach parameters and should be used
when verifying the selected breach parameters. For worst case scenarios, the breach width
should be in the upper range while the time of failure should be in the lower range. However
a sensitivity analysis is recommended to determine the reasonableness of the assumptions.

2. Peak Discharge - The peak discharge may be verified by use of equations (11) and (13)
of Reference No. 1. Although the equations assume a rectangular-shaped breach, a
trapezoidal breach may be analyzed by specifying a rectangular breach width that is equal
to the average width of the trapezoidal breach.

Equation 11:

23.4A s
C'
BR

2-A-3 October 1993

 Where: C = constant
As = reservoir surface area, in acres
BR = average breach width, in feet

Equation 13:

C
Qbmax'3.1BR( )3
C
(tf% )
H

Where: Qbmax = maximum breach outflow, in cfs

tf = time of failure, in hours
H = maximum head over the weir, in feet

This equation for Qbmax has been found to give results within +5% of the Qpeak from the full
DAMBRK model.

In a rare case where a dam impounding a small storage volume has a large time of failure,
the equations above will predict a much higher flow than actually occurs.

At a National Weather Service Dam-Break Model Symposium held in Tulsa, Oklahoma, June
27-30, 1983, Dr. Danny Fread presented an update to his simplified method. Equation 13
has been modified as follows to include additional outflow not attributed to breach outflow:

C
Qbmax'Qo%3.1BR( )3
C
(tf% )
H

Where: Qo = Additional (non-breach) outflow (cfs) at time tf (i.e.,

spillway flow and/or crest overflow) (optional data value,
may be set to 0).

2-A-4 October 1993

This equation has also been modified to address instantaneous failure, because in some
situations where a dam fails very rapidly, the negative wave that forms in the reservoir may
significantly affect the outflow from the dam.

3. Flood Wave Travel Time - Reasonableness of the flood wave travel time may be
determined by use of the following "rule-of-thumb" approximation for average wave speed:

(a) Assume an equivalent rectangular channel section for the selected irregular
channel section.

(b) Assume a constant average channel slope.

(c) Compute depth of flow from the following adjusted Manning's equation.

Qn
d'( )0.6
0.5
1.46B(S)

Where: d = depth of flow for assumed rectangular section, ft.

Q = peak discharge, cfs
B = average width (rectangular), ft.
S = average slope, ft./ft.
n = Manning's roughness coefficient

(d) Compute average velocity from Manning's Equation:

1.49(S)0.5(d)0.67
V'
n

Where: V = average velocity, fps

(e) Compute wave speed, C (Kinematic velocity):

5
C' V(0.68)
3

2-A-5 October 1993

 Where: C = wave speed (mph)

Note: 1 fps = 0.68 mph

(f) Determination travel time, TT

X
TT'
C

Where: TT = travel time, hr.

X = distance from dam, mi.

Note: If the slope is flat, the following "rule-of-thumb" provides a

very rough estimate of the wave speed:

C'2(S)0.5

Where: C = wave speed, mph

S = average slope, ft./mi.

In addition, as a "rule-of-thumb", the dynamic routing (NWS) method should be used

whenever severe backwater conditions at downstream areas occur and/or the slope is
less than 20 ft/mi. When these restrictions are not present normal hydrologic routing
(HEC-1) may provide reasonable results. It is recommended that HEC-2 be used to
determine the resulting water surface elevations when HEC-1 is used for the dambreak
study.

The HEC-I Manual (Reference E) states that when "a higher order of accuracy is
needed, then an unsteady flow model, such as the National Weather Service's DAMBRK
should be used." Experience demonstrates that the higher order of accuracy is usually
required. Therefore, the NWS DAMBRK model and the more recent NWS
FLOODWAV model are the preferred methods and recommended for all situations
requiring dambreak studies.

B. If a dambreak analysis has been performed by a method other than one of the
suggested acceptable methods, the selected breach parameters, peak discharge, depth of flow
and travel time of the flood wave shall be verified by one of the two methods:

2-A-6 October 1993

1. Unsteady Flow - Dynamic Routing Method (Recommended)

The NWS "DAMBRK" Model (Reference A) and the NWS "FLOODWAV" Model
(Reference B) are the recommended methods. Each FERC Regional Office has received
the software using the NWS DAMBRK program and should use this program, as necessary,
to verify dambreak studies. As the flood wave travels downstream, the peak discharge and
wave velocity generally, but not always, decrease. This attenuation in the flood wave is
primarily due to energy dissipation when it is near the dam and to valley storage as it
progresses in an unsteady flow downstream. It is important that the NWS model be
calibrated to historical floods, if at all possible.

2. Steady Flow Method (Provides a rough estimate)

If this method is selected, the breach parameters and peak discharge shall be verified as in
part "A" above. The method described below should be utilized only for preliminary
assessments and the obtained values may be far from the actually expected results. Sound
judgement and extensive numerical experience is necessary when evaluating the results.

For a rough estimate of the travel time and flood wave, it is recommended that one of the
following two steady state methods be used for verification of the licensee's values:

a. When steam gage data are available, the depth of flow and travel time can
be estimated as follows (This method will indirectly take valley storage into
consideration):

(1) Identify existing stream gages located downstream of the dam.

(2) Obtain the stage-discharge curve for each gage.

(3) Assuming Qpeak remains constant, extrapolate the curves to the

Qpeak value of the flood wave and determine the corresponding
water surface elevation.

(4) Using the continuity equation to determine the velocity, estimate the
travel time between each cross-section.

b. When stream gage data is not available, the depth of flow and travel time
can be estimated based on the following steady-state method:

2-A-7 October 1993

 (1) Assume the area downstream of the dam is a channel. This will
neglect valley storage.

(2) Identify on topographic maps all abrupt changes in channel width

and/or slope. Using this as a basis, select and plot channel cross-
sections.

(3) Assume Qbmax remains constant throughout the entire stream length
under consideration.

(4) Selecting a fairly rough Manning's n value, determine the depth of

flow by applying Manning's equation to each cross-section.
Assume the energy slope is equal to the slope of the channel.

(5) Using the continuity equation to determine the velocity, estimate the
travel time between each cross-section.

C. The above criteria for breach parameters, peak discharge, depth of flow, and travel time
should provide the necessary "ballpark figures" needed for comparison with licensee's
estimates. When large discrepancies in compared values exist, or questions arise about
assumptions to be made, or it appears that an extensive review will be necessary, the
Regional Director should contact the Washington Office, D2SI for guidance. The
methodology used by the licensee should be a part of the study and should be requested if
not included.

2-A-8 October 1993

 TABLE 1
SUGGESTED BREACH PARAMETERS
(Definition Sketch Shown in Figure 1)

Parameter Value Type of Dam

Average width of Breach (BR) BR = Crest Length Arch
(See Comment No. 1)*
BR = Multiple Slabs Buttress
BR = Width of 1 or more Masonry, Gravity
Monoliths,
Usually BR # 0.5 W

HD # BR # 5HD . . . . . . . . . . Earthen, Rockfill,

(usually between . . . . . . . . . . . . Timber Crib
2HD & 4HD)
BR $ 0.8 x Crest . . . . . . . . . . . Slag, Refuse
Length

Horizontal Component of Side 0 # Z # slope of valley walls . . . Arch

Slope of Breach (Z)
(See Comment No. 2)*
Z = O . . . . . . . . . . . . . . . . . . . Masonry, Gravity
Timber Crib, Buttress
¼ # Z # 1 . . . . . . . . . . . . . . . . . Earthen (Engineered,
Compacted)
1 # Z # 2 . . . . . . . . . . . . . . . . . . Slag, Refuse
(Non-Engineered)

Time to Failure (TFH) TFH # 0.1 . . . . . . . . . . . Arch

(in hours) 0.1 # TFH # 0.3 . . . . . . . . . . . Masonry, Gravity,
(See Comment No. 3)* Buttress
0.1 # TFH # 1.0 . . . . . . . . . . . Earthen (Engineered,
Compacted) Timber Crib
0.1 # TFH # 0.5 . . . . . . . . . . . Earthen (Non Engineered
Poor Construction)
0.1 # TFH # 0.3 . . . . . . . . . . . Slag, Refuse

Definition:
HD - Height of Dam
Z - Horizontal Component of Side Slope of Breach
BR - Average Width of Breach
TFH - Time to Fully Form the Breach
W - Crest Length

Note: See Page 2-A-12 for definition Sketch

*Comments: See Page 2-A-10 - 2-A-11

2-A-9 October 1993

Comments:

1. BR is the average breach width, which is not necessarily the bottom width. BR is the
bottom width for a rectangle, but BR is not the bottom width for a trapezoid.

2. Whether the shape is rectangular, trapezoidal, or triangular is not generally critical

if the average breach width for each shape is the same. What is critical is the
assumed average width of the breach.

3. Time to failure is a function of height of dam and location of breach. Therefore, the
longer the time to failure, the wider the breach should be. Also, the greater the height
of the dam and the storage volume, the greater the time to failure and average breach
width will probably be. Time to failure is the time from the start of the breach
formation until the complete breach is formed. It does not include the time leading
up to the start of the breach formation. For example, the time to erode away the
downstream slope of an earth dam is not included. In this situation, the time to failure
commences after sufficient erosion of the downstream slope has occurred and actual
formation of the breach (the lowering of the crest) has begun.

4. The bottom of the breach should be at the foundation elevation.

5. Breach width assumptions should be based on the type of dam, the height of dam, the
volume of the reservoir, and the type of failure (e.g. piping, sustained overtopping,
etc.). Slab and buttress dams require sensitivity analyses that vary the number of
slabs assumed to fail.

6. For a worst-case scenario, the average breach width should be in the upper portion of
the recommended range, the time to failure should be in the lower portion of the
range, and the Manning's "n" value should be in the upper portion of the
recommended range. In order to fully evaluate the impacts of a failure on
downstream areas, a sensitivity analysis is required to estimate the confidence and
relative differences resulting from varying assumptions.

a. To compare relative differences in peak elevation based on variations in breach

widths, the sensitivity analysis should be based on the following assumptions:

1. Assume a probable (reasonable) maximum breach width, a probable

minimum time to failure, and a probable maximum Manning's "n" value.
Manning's "n" values for sections immediately below the dam and up to
several thousand feet or more downstream of the dam should be assumed
to be larger than the maximum value suggested by field investigations in

2-A-10 October 1993

 order to account for uncertainties of high energy losses, velocities,
turbulence, etc., resulting from the initial failure.

2. Assume a probable minimum breach width, a probable maximum

time to failure, and a probable minimum Manning's "n" value.

Plot the resulting water surface elevation at selected locations downstream from
the dam for each run on the same graph. Compare the differences in elevation
with respect to distance downstream from the dam for the two cases.

b. To compare differences in travel time of the flood wave, the sensitivity analysis
should be based on the following assumptions:

1. Use criteria in a. 1.

2. Assume a probable maximum breach width, a probable minimum time

to failure, and a probable minimum Manning's "n" value.

Plot the results (elevation-distance downstream) of both runs on the same graph
to compare the changes in travel time with respect to distance downstream from
the dam.

c. To compare differences in elevation between natural flood conditions and

natural flood conditions plus dambreak, the sensitivity analysis should be based
on the following assumptions:

1. Route natural flood without dambreak assuming maximum probable

Manning's "n" value.

2. Use criteria in a. 1.

Plot the results (elevation-distance downstream) of both runs on the same graph
to compare the changes in elevation with respect to distance downstream from
the dam.

7. When dams are assumed to fail from overtopping, wider breach widths than those
suggested in Table 1 should be considered if overtopping is sustained for a long
period of time.

2-A-11 October 1993

2-A-12 October 1993
 APPENDIX II-B

HYDROMETEOROLOGICAL REPORT (HMR)

Nos. 51 and 52 vs HMR No. 33

October 1993
APPENDIX IIB

Hydrometeorological Report (HMR) Nos. 51 and 52 vs HMR No. 33

In accordance with Section 12.35(b)(1) of the Commission's Regulations, if structural failure

of project works (water impounding structures) would present a hazard to human life or
cause significant property damage, licensed or exempted project works subject to Part 12 of
the Commission's Regulations must be analyzed to evaluate their capability to withstand the
loading conditions and/or overtopping which may occur from a flood up to the probable
maximum flood (PMF) or the capacity of spillways to prevent the reservoir from rising to
an elevation that would endanger downstream life and property.

As a result of the recent publications of Hydrometeorological Reports Nos. 51 and 52 (HMR

Nos. 51 and 52), the FERC Staff has adopted the following guidelines for evaluating the
spillway adequacy of all licensed and exempted projects located east of the 105th meridian:

(1) For existing structures where a reasonable determination of the Probable

Maximum Precipitation (PMP) has not previously been made using suitable
methods and data such as contained in HMR No. 33 or derived from specific
meteorologic studies, or the PMF has not been properly determined, the ability
of the project structures to withstand the loading or overtopping which may
occur from the PMF must be re-evaluated using HMR Nos. 51 and 52.

(2) For existing structures where a reasonable determination of the PMP has
previously been made, a PMF has been properly determined, and the project
structures can withstand the loading or overtopping imposed by that PMF, a re-
evaluation of the adequacy of the spillway using HMR Nos. 51 and 52 is not
required. Generally, no PMF studies will be repeated solely because of the
publication of HMR Nos. 51 and 52. However, there is no objection to using
the two reports for necessary PMF studies for any water retaining structure,
should you so desire.

(3) For all unconstructed projects and for those projects where any proposed or
required modification will significantly affect the stability of water impounding
project structures, the adequacy of the project spillway must be evaluated
using:

(a) HMR Nos. 51 and 52, or

(b) Specific basin studies where the project lies in the stippled areas on
Figures 18 through 47 of HMR No. 51.

2-B-1 October 1993

 PROCEDURE FOR SELECTING APPROPRIATE INFLOW DESIGN FLOOD (IDF)
AND DETERMINING NEED FOR REMEDIAL ACTION

INTRODUCTION

The purpose of this appendix is to describe the procedures used to select the appropriate
inflow design flood (IDF) for a dam, and to determine the need for remedial action. These
procedures are presented in two flowcharts. The first flowchart describes the steps needed
to determine. . .

• If the probable maximum flood (PMF) was used in the original design of the dam,

• If the PMF or some lesser flood is the appropriate IDF, and,

• Whether remedial action at the dam is needed to enable it to safely accommodate

the appropriate PMF and/or IDF.

In order to determine whether the PMF or some lesser flood is the appropriate IDF, it may
be necessary to conduct an incremental hazard evaluation. This process is presented in the
second flowchart.

Following each flowchart is a breakdown of the procedures. Each block is presented

individually, and includes an explanation of the steps taken.

PROCEDURES FOR DETERMINING THE APPROPRIATE IDF

AND THE NEED FOR REMEDIAL ACTION

Flowchart 1 in Figure 1 presents a logical, step-by-step approach for evaluating the

hydrologic design of an existing dam, and determining the appropriate IDF for the dam and
whether remedial action is needed in order for the dam to safely accommodate the IDF.

Flowchart 1 is on the next page.

2-C-1 October, 1993

 FIGURE 1. FLOWCHART 1 -- PROCEDURES FOR DETERMINING
THE APPROPRIATE INFLOW DESIGN FLOOD (IDF)
AND THE NEED FOR REMEDIAL ACTION

Review the flood

used for the
original design

2 3 4 5
Is the
Was the PMF yes original PMF yes yes STOP
used for the Methodology Is the dam No further
original acceptable safe for the action is
design? (including PMF? required
flood routing)?

no

7 6
Conduct
no or
incremental Is the
hazard evaluation PMF the
to determine
determine thethe not appropriate
appropriate IDF apparent IDF?
(see Flowchart 2)

8 9 10 11
Based on the Conduct a
yes yes
incremental new PMF study STOP
hazard evaluation and flood routing Is the dam
No further
is the PMF the based on current safe for the
action is
appropriate IDF? criteria, as new PMF?
required
necessary

no

13 14 12 Remedial
yes action is
Is the dam STOP required for
safe for the No further the dam to
appropriate action is safely the
IDF? required new PMF.

no

15

Remedial action
is required for the
dam to safely
accommodate the
appropriate IDF

2-C-2
 EXPLANATION OF FLOWCHART 1
An explanation of the IDF flowchart is presented below.

The initial step in selecting the appropriate IDF and

determining the need for dam safety modification is
Block 1 to review the basis for the original hydrologic design
1 of an existing dam. This information will provide
valuable insight regarding whether the flood originally
Review the flood
used for the used for design purposes satisfies current criteria or
original design whether detailed investigations and analyses will be
required to determine the appropriate IDF for the dam.

In those situations where the original design information

has been lost, detailed investigations and analyses will
normally be required.

Block 2
Block
Once you have identified the basis for the original
1 hydrologic design, the next step is to determine if the
flood used for the original design is the probable
maximum flood (PMF). This question is important,
since the upper limit of the IDF is the PMF.
2
yes
Was the PMF
Block If your answer is YES, continue to Block 3.
used by the
original design? 3
If your answer is NO, go to Block 7. In Block 7 you
will perform an incremental hazard evaluation to
no
determine the appropriate IDF.

Block
7

Block 3 To ensure the reliability of the original PMF study

3
or the assumptions made on the various parameters
Is the
affecting the study, it is necessary to determine if
yes original PMF yes
Block Methodology Block
the PMF methodology originally used is still
2 acceptable 4 acceptable under current criteria.
(including
flood routing)?
no If your answer is YES, continue to Block 4.

Block
If your answer is NO, go to Block 6. In Block 6, you
6 will answer the question: Is the PMF the appropriate
IDF?

2-C-3 Continued . . .
 EXPLANATION OF FLOWCHART 1 (Continued)

Block 4
Determine if the dam is safe for the PMF. Your
4 answer to this question will indicate whether remedial
Block yes yes action will be required.
3 Is the dam
safe for the
PMF? If your answer is YES, continue to Block 5.

If your answer is NO, go to Block 6, you will answer

no
Block the question: Is the PMF the appropriate IDF?
6

Block 5
If the PMF is considered to be the appropriate IDF
5 for the dam, no further investigations or remedial
work for hydrologic conditions will be required
yes STOP
Block
4 No further
action is
required

2-C-4 Continue . . .
EXPLANATION OF FLOWCHART 1 (Continued)

Block 6 IF . . .
Block Block
3 4 In Block 3 you determined that the original PMF
methodology is NOT acceptable,
no

6 OR . . .
no or
Block Is the
4 PMF the In Block 4 you determined that the dam is NOT
not appropriate
apparent IDF?
safe for the PMF,

THEN . . .

You need to determine if the PMF is the appropriate

Block
9 IDF.

In some cases, such as when the dam is totally

submerged during the PMF, it may be obvious that
the appropriate IDF is something less than the PMF.
In other cases, it will not be apparent whether the
IDF should be the PMF or something less. In these
two cases, it will be necessary to perform an
incremental hazard evaluation to determine the
appropriate IDF for the dam. Continue to Block 7.

Sometimes, based on the size and volume of the

dam and reservoir, the proximity of the dam to
downstream communities, or even because of
political decisions, it will be obvious that the IDF
should be the PMF. If this is the case, a new PMF
study will be required. Go to Block 9.

2-C-5 Continued
EXPLANATION OF FLOWCHART 1 (Continued)

Block 7
IF . . .
Block
2
In Block 2 you determined that the flood used in the
original design is NOT the PMF,

OR . . .
7
Conduct
no or
incremental
hazard evaluation Block In Block 6 you determined that it is obvious that the
to determine not 6 IDF should be less than the PMF or it is not apparent
determine thethe
appropriate IDF apparent
(see Flowchart 2)
if the IDF should be the PMF or something less,

THEN . . .
Block
9 You need to perform an incremental hazard evaluation
to determine the appropriate IDF. Performing the
incremental hazard evaluation involves:

• Conducting dambreak sensitivity studies,

• Reviewing incremental rises between with-

failure and without-failure conditions for a
range of flood inflows (see Flowchart 2).

• Selecting the appropriate IDF on the basis of

the dambreak studies and incremental impacts
on downstream areas.

A procedural flowchart for performing a hazard

evaluation appears in Flowchart 2 (Figure 2),
followed by an explanation of the procedure.

2-C-6 Continued . . .
 EXPLANATION OF FLOWCHART 1 (Continued)

Block 8 You should use the results of the incremental hazard

Block evaluation and dambreak studies conducted in Block 7
7
to determine if the PMF is the appropriate IDF.

The IDF should be the PMF when the incremental

8
Based on the
yes
consequences of failure are unacceptable, regardless
incremental
Block of how large the assumed flood inflow becomes.
hazard evaluation
is the PMF the 9
appropriate IDF? If your answer is YES, continue to Block 9.

If your answer is NO, go to Block 13. In Block 13

you will answer the question: Is the dam safe for
Block the appropriate IDF?
13

Block 9 IF . . .
Block
6
In Block 6 you determined that the PMF is obviously
the appropriate IDF,
yes

OR . . .
9
Conduct a
yes
Block new PMF study
8 and flood routing Block If, based on the incremental hazard evaluation
based on current 10 conducted in Block 8, the PMF is the appropriate
criteria, as
necessary
IDF,

THEN . . .

You should conduct a new PMF study and flood

routing based on current criteria, unless it was
determined in Block 3 that the original PMF is
acceptable under current criteria.

2-C-7 Continued...
 EXPLANATION OF FLOWCHART 1 (Continued)

Blocks 10, 11 and 12

Once the new PMF is calculated, you should
determine if the dam is safe for the new PMF.
10 11
yes
STOP If the dam is SAFE for the new PMF, no further
Block 9 Is the dam
No further
safe for the
action is
investigations or remedial actions for hydrologic
new PMF? conditions are required.
required

no If the dam is NOT SAFE for the new PMF,

remedial action is required for the dam to safely
12 accommodate the PMF.
Remedial action
is required for the
dam to safely
accommodate the
new PMF.

Blocks 13, 14 and 15

IF….

Block 8 In Block 8 you determined that the PMF is NOT

the appropriate IDF,
no

10 11 THEN…..
yes
Is the dam STOP
No further
You need to determine if the dam is safe for the
safe for the
appropriate action is appropriate IDF.
IDF? required

If the dam is SAFE for the appropriate IDF, no

no
further investigations or remedial action for hydro-
logic conditions are required.
12

Remedial action If the dam is NOT SAFE for the appropriate IDF,
is required for the
dam to safely remedial action is required for the dam to safety
accommodate the accommodate the appropriate IDF.
appropriate IDF.

Depending on the type of remedial action

considered, it may be necessary to reevaluate the
IDF to ensure that the appropriate IDF has been
selected for the design of any modification.

2-C-8 October 1993

INTRODUCTION

As stated prviously, if the PMF was not used for the original design of a dam, or if the PMF
is not the appropriate IDF, an incremental hazard evlauation must be performed to
determine the appropriate IDF.

PROCEDURES FOR CONDUCTING AN INCREMENTAL HAZARD EVALUATION

Flowchart 2 in Figure 2 shows the procedures for performing an incremental hazard

evaluation. This flowchart is an expansion of Block 7 in Flowchart 1, Figure 1.

Flowchart 1 is on the next page.

2-C-9
FIGURE 2. FLOWCHART 2 -- PROCEDURES FOR CONDUCTING
AN INCREMENTAL HAZARD EVALUATION

1 Assume that
the normal
Reservoir level with
normal streamflow
conditions pre-
vailing is the initial
failure condition

2 Conduct
dambreak
Analysis and route
the dambreak flood
downstream to the
point where the
flood no longer
constitutes a threat

Assume a
3 Is the 3
new (larger)
Incremental no Flood inflow
incremental incremental
condition (up to
increase in
hazard evaluation hazard evaluation
the PMF, if
consequences
determine the determine theas the
necessary)
due to failure
appropriate IDF new failureIDF
appropriate
acceptable?
(see Flowchart 2) condition
(see Flowchart 2)

yes

5 Could a
Failure at a larger
flood inflow result yes
in unacceptable
consequences not sure

no

Select
6
appropriate
IDF on the basis
of dambreak
studies and in-
cremental impacts
on downstream
areas

no

Continue with the

procedures in
Flowchart 1.
Beginning with
Block 8.

2-C-10 October, 1993

EXPLANATION OF FLOWCHART 2
An explanation of the Hazard Evaluation Flowchart is presented below.

Block 1
1 Assume that Assume that the normal reservoir level with normal
the normal
Reservoir level with streamflow conditions prevailing is the initial failure
normal streamflow
conditions pre-
condition. Starting at this point will ensure that the
vailing is the initial full range of flood inflow conditions will be
failure condition.
investigated and will include the “sunny day” failure
condition. It will also assist in verifying the initial
hazard rating assigned to the dam. Using the normal
Block maximum water surface level as the initial condition
2 is particularly important if the initial hazard rating
was low.

Block 2

Block Next, conduct dambreak sensitivity studies (of various

1 breach parameters) and route the dambreak flood to the
point downstream where it no longer constitutes a threat
to downstream life and property.
Conduct
2
dambreak
Analysis and route It is important to remember that the incremental increases
the dambreak flood
downstream to the should address the differences between the nonfailure
point where the
flood no longer Block condition with the dam remaining in place and the
constitutes a threat 4 failure condition. Also, the dam should not be assumed
to fail until the peak reservoir water surface elevation is
attained for the assumed flood inflow condition being
Block
3 analyzed. Dams should be assumed to fail as described in
Chapter II of these Engineering Guidelines.

Continued...

2-C-11
 EXPLANATION OF FLOWCHART 2 (Continued)

Block 3
Block
2 Now, determine if the additional increase in
consequences due to failure is acceptable. Answering
this question is critical in the incremental hazard
3 evaluation and doing so involves an estimate of loss
Is the incremental no Block
of life and property with and without dam failure.
increase in 4
consequences due
to failure If the consequences of failure under the assumed
acceptable?. floodflow conditions are NOT ACCEPTABLE, go to
Block 4.
yes

If the consequences of failure ARE ACCEPTABLE,

Block
5
continue to Block 5..

Block 4

IF …
Block
2
In Block 3 it was determined that the consequences
of failure under the assumed floodflow conditions are
Assume a new
4
(larger) flood NOT ACCEPTABLE,
Inflow condition
no (up to the PMF, if
Block
3 necessary) as the THEN …
new failure
condition.
Assume a new (larger flood inflow condition (e.g.,
yes
some percentage of the PMF) and perform a new
Block
5 not sure dambreak analysis (see Block 2). This procedure
should be repeated until an acceptable level of
flooding is identified, or the full PMF has been reached.

Continued...

2-C-12 October, 1993

EXPLANATION OF FLOWCHART 2 (Continued)
IF ...
Block 5
Block In Block 3 you determined that the consequences of failure
3 under the assumed floodlfow conditions are ACCEPTABLE,
yes i.e., failure of the dam under “sunny day” conditoins was
insignificant,
Block
4
5 THEN ...
Could a failure at
yes
a larger flood
inflow result in not sure Determine if failure at a larger flood inflow condition will
unacceptable result in unacceptable consequences. This question is very
consequences? important. For example, situations exist where a failure
during normal water surface conditions results in the flood
no
wave being contained completely within the banks of a river
and obviously would not cause a threat to life and property
Block downstream. However, under some floodflow conditions,
6
the natural river flows may go out-of-bank, and a failure on
top of that flood condition will result in an additional threat
to downstream life and property.

If failure at another flood level will result in UNACCEPTABLE

consequences, or if you are NOT SURE, return to Block 4.
Assume larger flood inflow conditions and perform new
dambreak studies. This procedure should be repeated to
determine the acceptable level of flooding.

If failure at another flood level will NOT result in unacceptable

consequences, continue to Block 6.

Block 6 You should now select the appropriate IDF based on the
Block
5
results of dambreak studies and incremental impacts on
downstream areas.
no

6
Select appropriate
IDF on the basis of
dambreak studies
and incremental
impacts on
downstream areas.

Block
7

2-C-13 Continued...
EXPLANATION OF FLOWCHART 2 (Continued)

Block
6
Block 7
Continue this process with the steps in Flowchart 1,
7 Figure 1, starting with Block 8. In Block 8 you will
Continue with answer the question: Based on the incremental
the procedures hazard evaluation, is the PMF the appropriate IDF?
in Flowchart 1,
beginning with
Block 8.

2-C-14
 CHAPTER III

GRAVITY DAMS

(Revised October 2002)

 Chapter III

Gravity Dams

3-0 Contents

Title Page

3-1 Purpose and Scope 3-1

3-1.1 General 3-1

3-1.2 Review Procedures 3-1

3-2 Forces 3-2

3-2.1 General 3-2

3-2.2 Dead Loads 3-2
3-2.3 External Water Imposed 3-2
3-2.4 Internal Hydrostatic Loads (uplift) 3-3
3-2.5 Earth and Silt Pressures 3-12
3-2.6 Earthquake Forces 3-13
3-2.7 Ice Loading 3-13
3-2.8 Temperature and Aggregate Reactivity 3-14

3-3 Loading Combinations 3-15

3-3.1 General 3-15

3-3.2 Case I - Usual Loading Combination-Normal 3-15
Operating Condition
3-3.3 Case II - Unusual Loading Combination-Flood 3-15
Discharge
3-3.4 Case IIA - Unusual Loading Combination-Ice 3-15
3-3.5 Case III - Extreme Loading Combination- 3-15
Case I + Earthquake

3-i
3-4 Methods of Analysis 3-16

3-4.1 General 3-16

3-4.2 Gravity Method 3-16
3-4.3 Finite Element Methods 3-16
3-4.4 Dynamic Methods 3-20
3-4.5 Cracked Base Analysis 3-23
3-4.6 Review of Computer Analyses 3-24

3-5 Stability Criteria 3-25

3-5.1 General 3-25

3-5.2 Acceptance Criteria 3-25
3-5.3 Safety Factor Evaluation 3-29
3-5.4 Foundation Stability 3-29

3-6 Construction Materials 3-31

3-6.1 General 3-31

3-6.2 Site Investigations 3-32
3-6.3 Concrete Properties 3-32
3-6.4 Foundation Properties 3-34

3-7 References 3-37

Appendices

Appendix 3A Hydrodynamic Forces 3A

Appendix 3B Block Rocking Analysis 3B
Appendix 3C Example Problem 3C
Appendix 3D Example Seismic Analysis 3D

3-ii
 GRAVITY DAMS

3-1 Purpose and Scope

3-1.1 General

The objective of this chapter of the Guidelines is to provide Staff engineers,

licensees, and their consultants with recommended procedures and stability criteria for
use in the stability analysis of concrete gravity structures. Engineering judgement must
be exercised by staff when evaluating procedures or situations not specifically covered
herein. Unique problems or unusual solutions may require deviations from the criteria
and/or procedures outlined in this chapter. In these cases, such deviations must be
evaluated on an individual basis in accordance with Chapter 1, paragraph 1-4 of these
Engineering Guidelines

3-1.2 Review Procedures

Review by the staff of analyses performed by licensees, or their consultants,

should concentrate on the assumptions used in the analysis. The basis for critical
assumptions such as allowable stresses, shear strengths, drain effectiveness, and loading
conditions should be carefully examined. The consultant's reports, exhibits, and
supplemental information must provide justification for these assumptions such as
foundation exploration and testing, concrete testing, instrumentation data, and records
maintained during the actual construction of the project. Also, the staff engineer's
independent knowledge of the dam gained through site inspections or review of
operations inspection report as well as familiarity with previous reports and analyses,
should be used to verify that the exhibits presented are representative of actual
conditions. M ethods of analysis should conform to the conventional procedures used in
the engineering profession.

Conservative assumptions can reduce the amount of exploration and testing

required. For example, if no cohesion or drain effectiveness is assumed in an analysis,
there would be no need to justify those assumptions with exploration and testing. For this
reason, it may sometimes be more beneficial to analyze the dam with conservative
assumptions rather than to try to justify less conservative assumptions. There is however
a minimum knowledge of the foundation that must be obtained. The potential for sliding
on the dam foundation is generally investigated. However, the potential for failure
through a deep surface deep in the foundation should be investigated. Experience has
shown that the greatest danger to dam stability results when critical attributes of the
foundation are not known. For example, in the case of Morris Shephard Dam, 26/ P-
1494, a horizontal seam underlaid the dam, providing a plane of weakness that was not

3-1
considered. This oversight was only discovered after the dam had experienced significant
downstream movement.

3-2 Forces

3-2.1 General

Many of the forces which must be considered in the design of the gravity dam
structure are of such a nature that an exact determination cannot be made. The intensity,
direction and location of these forces must be estimated by the engineer after
consideration of all available facts and, to a certain extent, must be based on judgment
and experience.

3-2.2 Dead Loads

Unless testing indicates otherwise, the unit weight of concrete can be assumed to
be 150 lb/ft3 . In the determination of the dead load, relatively small voids, such as
galleries, normally are not deducted unless the engineer judges that the voids constitute a
significant portion of the dam's volume. The dead loads considered should include
weights of concrete and superimposed backfill, and appurtenances such as gates and
bridges.

3-2.3 External Water Imposed Loads

3-2.3.1 Hydro Static Loads

Although the weight of water varies slightly with temperature, the weight of fresh
water should be taken at 62.4 lb/ft3. A linear distribution of the static water pressure
acting normal to the surface of the dam should be applied.

3-2.3.2 Nappe Forces

The forces acting on an overflow dam or spillway section are complicated by

steady state hydrodynamic effects. Hydrodynamic forces result from water changing
speed and direction as it flows over a spillway. At small discharges, nappe forces may
be neglected in stability analysis; however, when the discharge over an overflow spillway
approaches the design discharge, nappe forces can become significant and should be
taken into account in the analysis of dam stability.

Previous FERC gravity dam guidance dealt with nappe forces by ignoring the
weight of the nappe on top of the structure and by requiring that the tailwater be assumed

3-2
to be 60% of its expected height. This method does not sufficiently account for sub-
atmospheric crest pressures and high bucket pressures, and in some cases it can yield un-
conservative results. While this practice is still acceptable, it may be desirable to
determine forces due to the nappe and tailwater more rigorously. References 3 and 4 can
be used to determine more accurate nappe pressure distribution. Also, Appendix A of
this chapter presents a general method for the determination of nappe pressures.

If the tailwater is greater than the conjugate depth, tailwater will fall back against
the dam, submerging the jet and lessening hydrodynamic effects. However, unless there
is clear evidence that tailwater will be in excess of the conjugate depth, it shall be
assumed that tailwater is blown downstream of the dam by the discharge, and that
tailwater has no effect on the nappe pressures on the dam. Downstream channel
conveyance characteristics are typically not well known for unprecedented discharges.
For this reason, it should not be assumed that tailwater will drown out the hydraulic jump
without sufficient justification.

3-2.4 Internal Hydrostatic Loads (Uplift)

3-2.4.1 General

Any stability analysis of the dam should seek to apply forces that are compatible
with the failure mechanism being assumed. For this reason, it is less important to
determine what the uplift pressures on a dam are at present than it is to determine what
they would be during failure. Uplift should be assumed to exist between the dam and its
foundation, and within the foundation below the contact plane and it should also be
applied within any cracks within the dam. Uplift is an active force which must be
included in the analysis of stability. Uplift shall be assumed to act over 100 percent of
the area of any failure plane whether that plane is within the dam, at the contact with the
foundation or at any plane within the foundation.

3-2.4.2 Horizontal Planes within the Dam

Uplift along failure planes within the body of the dam shall be assumed to vary
from 100% of normal headwater at the upstream face to 100% of tailwater or zero, as the
case may be, at the downstream face. When a vertical drainage system has been
provided within the dam, the drain effectiveness and uplift assumptions should follow the
guidance provided in paragraph 3-2.4.3 below, and should be verified by instrumentation.

3-3
3-2.4.3 Rock Foundations

Uplift distribution at the plane of contact of the dam and its foundation, and within
the foundation depends on depth and spacing of drains, grout curtain, rock permeability,
jointing, faulting, and any other geologic features which may modify the seepage or flow
of water. Effective downstream drainage, whether natural or artificial will usually limit
the uplift pressure at the toe of the dam to tailwater pressure. However, there are
situations where the orientation of joint systems and/or bedding planes transmit high
uplift pressures to areas of the base downstream of the drainage system.

The uplift reduction due to seepage control measures such as drainage and/or
grouting should not be assumed unless the geologic characteristics of the foundation have
been thoroughly investigated and the design of the seepage control measures has been
tailored to correct the specific deficiencies of the site. Uplift reduction can only be
assured by implementation of a comprehensive monitoring and maintenance program.
For this reason structures with closed drainage systems, which do not allow inspection or
maintenance, are considered to be subject to full uplift loading unless a monitoring
system is installed to verify uplift pressures on a periodic basis. Increasing uplift
pressures that can not be corrected because of the closed system may necessitate
structural modifications.

It is helpful to coordinate closely with an experienced engineering geologist to

evaluate the character of the foundation and the effectiveness of a drainage system. Any
drain effectiveness assumptions made should be coupled with a testing and monitoring
program aimed at verifying the assumptions. The system should include instrumentation
to verify continued operation of the drains and to determine the effects of corrosion or
clogging upon the original effectiveness assumption. A maintenance program for the
system should be developed and implemented that is consistent with the nature of the
system. In general, maintenance should include, but not be limited to: periodic testing to
locate clogged and inoperative drains; redrilling or cleaning of drains which have become
clogged; installation of additional drains to achieve design concept; and periodic
monitoring and calibration of pressure gages.

Uplift reduction due to drainage assumes that the drainage system vents the high
pressure area under the dam to tailwater pressure. This intended purpose can be thwarted
however if the drainage system exits into a region of high hydrodynamic pressure as
shown in figure 1. In this case, the drainage system is vented to tailwater under normal
conditions, however, during flood discharges the drain system can become pressurized.

3-4
 Staff review of assumptions concerning
uplift reduction should always be
conservative. Instrumentation data should be
submitted in support of uplift reduction
assumptions, and even when instrumentation
indicates that uplift reduction is occurring, the
reviewer must question whether or not the
headwater, tailwater and foundation stresses
that control the magnitude and distribution of
uplift pressure will remain the same under
more severe conditions.

The following guidance shall be

applied to staff review of the design
assumptions. The uplift criteria cited herein Figure 1
may be relaxed only when sufficient field
measurements of actual uplift pressures justify any proposed deviations.

3-2.4.3.1 Uplift Assumptions

Uplift at the foundation-concrete interface for structures having no foundation

drains or an unverified drainage system should be assumed to vary as a straight line from
100% of the headwater pressure at the upstream face (heel) to 100% of the tailwater
pressure at the downstream face (toe) applied over 100% of the base area. Local
reductions in tailwater elevations produced by hydrodynamic effects described in section
3-2.3.2 shall not be included in uplift computation.

Uplift at the concrete/rock interface for structures having an open verifiable

drainage system should be assumed to vary as a straight line from full headwater pressure
at the heel or theoretical crack tip, to reduced uplift at the drain, and then to full tailwater
pressure at the toe (See figure 2). The drain effectiveness (E) must be verified by
instrumentation and an effective maintenance plan as outlined in paragraph 3-2.4.3 must
be implemented. Note that if heads are measured from any other datum than the dam
base, the dam base elevation must be subtracted from the absolute heads to yield uplift
pressure. It is also assumed that the gallery is free draining.

3-5
 Figure 2

The assumption of full

reservoir uplift in the non-
compressive zone results from the
realization that if the crack width
becomes sufficiently large, the base
will become exposed to the reservoir
and the drains will become
completely in-effective. This Figure 3
assumption is compatible with the
limit state failure mechanism that is considered in an overturning failure. For this reason,

3-6
uplift on any portion of the base or section not in compression should be assumed to be
100% of the assumed upstream head except when the non-compressive foundation
pressure is the result of earthquake forces. If, however, instrumentation can verify use of
less than 100%, then uplift pressure may be reduced accordingly. Uplift distribution for
the case in which the theoretical foundation crack extends beyond the line of drains is
shown in figure 3.

Deviations from the pressure distributions shown in figures 2 and 3 may be

considered provided there is sufficient justification such as instrumentation of foundation
abnormalities.

Typically, measured drain efficiency must be considered valid only for the
reservoir loading at which the measurement was taken. Extrapolation to higher reservoir
levels in the absence of supporting field data is not valid, especially where the applied
forces from the unusual loading condition are significantly different than the usual
loading condition. However, extrapolation of drain efficiencies for higher reservoir levels
may be allowed on a case-by-case basis. Staff engineers should consider the specific
conditions at each project to determine if extrapolation of drain efficiencies is valid.
Factors which should be considered are as follows:

a. The difference in the character of foundation stresses produced. Crack extent

and dimensions are influenced by the stresses imposed on the foundation. If
analysis indicates that the foundation stresses will be significantly different, crack
geometry and therefore drain efficiency may be different.

b. The difference between drain efficiency assumed in the design and the
measured drain efficiency. If there is some margin for error, extrapolation is
easier to justify.

c. Whether or not a theoretical crack propagates beyond the location of the drains
during the loading conditions at which measurements are available. If a crack is
indicated, then the drainage system may be assumed to perform adequately under
cracked base conditions.

d. The degree of understanding of the geology of the foundation of the dam. As

outlined in paragraph 3-5.3, a reduction in the uncertainties associated with the
selection of design parameters can lead to a corresponding reduction in required
factors of safety. This principle can also be applied to the extrapolation of drain
efficiencies. Better definition of the geologic characteristics of the foundation
which affect seepage parameters can also reduce the uncertainties associated with
drain efficiency extrapolation.

3-7
 e. The sensitivity of the stability drain
effectiveness assumptions. If drain
efficiency is required to keep the
theoretical base crack from extending
all the way through the dam,
extrapolation of drain efficiency
assumptions into unprecedented
loading conditions should be viewed
with great skepticism.

When analysis indicates that a

theoretical crack propagates beyond the drains
for an unprecedented load condition such as Figure 4
the PMF, the amount of drain efficiency that
can exist is limited by certain physical constraints. Even if the pressure at a given drain is
zero, the effect of this pressure reduction is very local as can be seen in figure 4.

For cases in which the theoretical base crack extends beyond the drains, the resulting
uplift force should not be assumed to be less than that calculated by the idealization
shown below , where L2AP = 0 and the boundary conditions are those depicted in figure 5.

Figure 5

3-8
3-2.4.3.2 Grouting

Grouting alone should not be considered sufficient justification to assume an uplift

reduction. A grout curtain may retard foundation flows initially, but the degree of uplift
relief may be lessened as the age of the dam increases due to deterioration of the curtain.
A drainage system should be utilized downstream of grout curtains and, a monitoring
system should be employed to determine actual uplift pressures and to detect any
reduction in drain efficiency due to clogging of the drains.

3-2.4.3.3 Aprons

Upstream and downstream aprons have the effect of increasing the seepage path
under the dam. For an upstream apron properly sealed to prevent leakage, the effect is to
reduce the uplift under the dam. The effectiveness of upstream aprons in reducing uplift
is compromised if cracks and
joints in the apron permit leakage.
Conversely, downstream aprons
such as stilling basins have the
effect of increasing uplift under
the dam. (See figure 6) Uplift
reduction should be justified by
instrumentation.

In the case of downstream Figure 6

aprons, it may be assumed that
uplift is limited to that which would float the apron.

3-2.4.3.4 Reservoir Silt

Reservoir silt can reduce uplift under a dam in a manner similar to an upstream
apron. 14/ Uplift reduction should be justified by instrumentation.

Because of potential liquefaction of the silt during a seismic event, uplift

reduction due to silt may be lost in seismic situations. If liquefaction occurs, pore
pressure in the silt will increase. This condition of elevated pore pressure may persist for
some time after the seismic event. For this reason, uplift reduction due to silt may not be
relied upon when considering post earthquake stability.

3-9
3-2.4.3.5 Earthquake

Uplift pressures should be assumed to be those existing under normal conditions

during earthquake loading. However, when performing post earthquake stability analysis,
the effects of silt liquefaction, apron cracking, or potential offsets must be considered.

3-2.4.3.6 Flood Loading

Uplift reductions should not be based on the assumption that the IDF flood event
will be of such short duration and the permeability of the foundation so low that the
elevated headwater and tailwater pressures are not transmitted under the base of the dam.
This less than conservative assumption is invalid because extreme design floods and the
resulting elevated water levels often last many hours, if not days, and because in a
saturated rigid system such as a rock foundation with joints, extremely small volume
changes can transmit large pressure changes. In the absence of corroborative evidence
(e.g., measurements of piezometer levels during prior floods) the uplift should be
assumed to vary directly with changes in headwater and tailwater levels. For more
discussion of flood loading, refer to Chapter 2 of these guidelines.

3-2.4.4 Soil Foundations

Uplift pressures acting upon the base of a gravity structure constructed on a

pervious soil foundation are related to seepage through permeable materials. Water
percolating through pore spaces in the materials is retarded by frictional resistance,
somewhat the same as water flowing through a pipe. The intensity of the uplift can be
controlled by construction of properly placed aprons, cutoffs and other devices. 19/

Base cracking may not affect the uplift distribution under a soil founded dam as
much as under a dam founded on rock. If the soil is relatively pervious, a small crack
between the dam and foundation may cause no effect. For this reason, the standard
cracked base uplift distributions in section 3-2.4.3.1 of this chapter may not be
applicable. One of the following methods should be used to estimate the magnitude of the
uplift pressure:

3-2.4.4.1 Creep Theory

The word "Creep" in this usage refers to a simplified method which can be used
to estimate uplift pressure under a structure. Under creep theory, the uplift pressure is
assumed to be the sum of two components; the seepage potential and the position
potential.

3-10
 The seepage potential is calculated by first determining the creep distance which a
molecule of water would follow as it flows beneath the structure. The creep distance
starts at a point on the ground line directly over the heel, and ends at another point on the
ground line directly above the toe, following the boundary of the sides and bottom of
buried concrete. The seepage and position potentials are then calculated as shown in
figure 7.

The effective uplift pressure at a

point is then calculated by multiplying the
sum of the seepage and position potentials
of the points by the unit weight of water.

In most cases, the vertical and

horizontal permeability of soil are not equal. Figure 7
Typically, the horizontal permeability (kh) is
3 times as great as the vertical permeability (kv) A "weighted creep" recognizes the
differences in vertical and horizontal permeability of most soil foundations by
multiplying the horizontal distances along the creep path by the ratio (kv)/(kh).

The weighted creep distance Lw , should be calculated as shown below:

Where: K h= horizontal permeability

K v= vertical permeability
L h= horizontal length of creep path
L v= vertical length of creep path
Lw = weighted creep distance

3-11
3-2.4.4.2 Flow Net Method

This method is a graphical procedure which involves the construction of flow lines
and lines of equal potential (lines drawn through points of equal total head) in subsurface
flow. Flow lines and equipotential lines are superimposed upon a cross section of the soil
through which the flow is taking place. Reference for the procedure is made to any
standard text book on soil mechanics, or reference 19/.

3-2.4.4.3 Finite Element Method

Two and 3 dimensional finite element ground water modeling can also be used in a
manner similar to the flow net method. Material anisotropy can be factored into these
analyses.

3-2.5 Earth and Silt Pressures

3-2.5.1 Earth Pressures

Earth pressures exerted on dams or other gravity structures by soil backfills should
be calculated as outlined in reference 19. In most cases, at rest earth pressures should be
assumed. The rigidity of the foundation and the character of the backfill, along with the
construction sequence, may affect this assumption. The unit weight of the backfill and
material strength parameters used in the analysis should be supported by site
investigations. If the backfill is submerged, the unit weight of the soil should be reduced
by the unit weight of water to determine the buoyant weight.

Earth backfill on the downstream side of a gravity dam has a beneficial effect on
stability, however, if flood conditions can overtop the dam and lead to erosion of the
backfill, it can not be relied upon for its stabilizing effects.

3-2.5.2 Silt Pressures

The silt elevation should be determined by hydro graphic surveys. Vertical

pressure exerted by saturated silt is determined as if silt were a saturated soil, the
magnitude of pressure varying directly with depth. Horizontal pressure exerted by the silt
load is calculated in the same manner as submerged earth backfill. Silt shall be assumed
to liquefy under seismic loading. Thus, for post earthquake analysis, silt internal shear
strength shall be assumed to be zero unless site investigations demonstrate that
liquefaction is not possible.

3-12
3-2.6 Earthquake Forces

Earthquake loadings should be selected after consideration of the accelerations

which may be expected at each project site as determined by the geology of the site,
proximity to major faults, and earthquake history of the region as indicated by available
seismic records. Seismic risk maps can be used to establish the probability zone for
projects which do not have detailed seismicity studies. A set of seismic risk maps are
available from the United States Geological Survey, (USGS) at:

http://geohazards.cr.usgs.gov/eq/

Other widely accepted seismic risk maps can also be used as a starting point for the
determination of seismic loading.

While a variety of sources can be cited, the determination of the Maximum

Credible Earthquake for a site remains the responsibility of the licensee. General seismic
hazard maps such as that cited above may not sufficiently account for local seismicity.
Site specific seismic studies may be required. See the seismicity chapter of these
guidelines for more information.

Seismic loading need not be considered for structures for which the MCE
produces a peak ground acceleration of less than 0.1g at the site. Procedures for
evaluating the seismic response of the dam are given in Section 3-4.4 of this chapter.

3-2.7 Ice Loading

3-2.7.1 Ice Pressures

Ice pressure is created by thermal expansion of the ice and by wind drag.
Pressures caused by thermal expansion are dependent on the temperature rise of the ice,
the thickness of the ice sheet, the coefficient of expansion, the elastic modulus and the
strength of the ice. Wind drag is dependent on the size and shape of the exposed area, the
roughness of the surface, and the direction and velocity of the wind. Ice loads are usually
transitory. Not all dams will be subject to ice pressure and the engineer should decide
whether an ice load is appropriate after consideration of the above factors. An example
of the conditions conducive to the development of potentially high ice pressure would be
a reservoir with hard rock reservoir walls which totally restrain the ice sheet. In addition,
the site meteorological conditions would have to be such that an extremely rigid ice sheet
develops. For the purpose of the analysis of structures for which an ice load is expected,
it is recommended that a pressure of 5000 pounds per square foot be applied to the
contact surface of the structure in contact with the ice, based upon the expected ice

3-13
thickness. The existence of a formal system for the prevention of ice formation, such as
an air bubble system, may reduce or eliminate ice loadings. Information showing the
design and maintenance of such a system must be provided in support of this assumption.
Ice pressure should be applied at the normal pool elevation. If the dam is topped with
flashboards, the strength of the flashboards may limit the ice load. Further information
concerning ice loadings can be found in reference 21/.

3-2.7.2 Ice /Debris Impact

Some rivers are subject to ice and debris flow. Current bourne ice sheets weighing
several tons, and/or debris can impact dams and cause local damage to piers, gates or
machinery. Several dams have experienced very large reservoir surcharges under
moderate flood events due to plugging of spillway bays by debris or floating ice. When
the ability of a spillway to pass floods is evaluated, the effect of ice and debris should be
considered.

3-2.8 Temperature & Aggregate Reactivity

Volumetric changes caused by thermal expansion and contraction, or by

alkali/aggregate reactivity affect the cross valley stresses in the dam. These stresses are
important when 3 dimensional behavior is being considered. Expansion will cause a dam
to wedge itself into the valley walls more tightly, increasing its stability. Contraction has
the opposite effect. While these effects are acknowledged, the beneficial effect of
expansion is difficult to quantify even with very elaborate finite element models because
it is contingent on the modulus of deformation of the abutments which is highly variable.
For this reason, the beneficial effects of expansion should not be relied upon in three
dimensional stability analysis. If it appears that contraction will cause monolith joints to
open, and thus compromise force transfer from monolith to monolith, this effect should
be considered.

3-3 Loading Combinations

3-3.1 General

The following loading conditions and requirements are suitable in general for
gravity dams of moderate height. Loads which are not indicated, such as wave action, or
any unusual loadings should be considered where applicable.

3-14
3-3.2 Case I Usual Loading Combination - Normal Operating Condition

The reservoir elevation is at the normal power pool, as governed by the crest
elevation of an overflow structure, or the top of the closed spillway gates whichever is
greater. Normal tailwater is used. Horizontal silt pressure should also be considered, if
applicable.

3-3.3 Case II Unusual Loading Combination - Flood Discharge Loading

For high and significant hazard potential projects, the flood condition that results
in reservoir and tailwater elevations which produce the lowest factor of safety should be
used. Flood events up to and including the Inflow Design Flood, if appropriate, should
be considered. For further discussion on the Inflow Design Flood, refer to chapter 2 of
these guidelines.

For dams having a low hazard potential, the project should be stable for floods up
to and including the 100 year flood.

3-3.4 Case II A Unusual Loading Combination - Ice

Case I loading plus ice loading, if applicable.

3-3.5 Case III Extreme Loading Combination - Case 1+Earthquake

In a departure from the way the FERC has previously considered seismic
loading, there is no longer any acceptance criteria for stability under earthquake
loading. Factors of safety under earthquake loading will no longer be evaluated.
Acceptance criteria is based on the dam's stability under post earthquake static loading
considering damage likely to result form the earthquake. The purpose of considering
dynamic loading is to determine the damage that will be caused so that this damage can
be accounted for in the subsequent post earthquake static analysis.

Factors to consider are as follows:

- Loss of cohesive bond in regions of seismically induced tensile stress.

- Degradation of friction angle due to earthquake induced movements or

rocking.

- Increase in silt pressure and uplift due to liquefaction of reservoir silt. (See
section 3.2.5.2)

3-15
Recommended procedures for seismic analysis are presented in section 3-4.4.

3-4 Methods of Analysis

3-4.1 General

Selection of the method of analysis should be governed by the type and

configuration of the structure being considered. The gravity method will generally be
sufficient for the analysis of most structures, however, more sophisticated methods may
be required for structures that are curved in plan, or structures with unusual
configurations.

3-4.2 Gravity Method

The gravity method assumes that the dam is a 2 dimensional rigid block. The
foundation pressure distribution is assumed to be linear. It is usually prudent to perform
gravity analysis before doing more rigorous studies. In most cases, if gravity analysis
indicates that the dam is stable, no further analyses need be done. An example gravity
analysis is presented in Appendix C of this chapter. Stability criteria and required factors
of safety for sliding are presented in Section 3-5.

3-4.3 Finite Element Methods

3-4.3.1 General

In most cases, the gravity analysis method discussed above will be sufficient for
the determination of stability. However, dams with irregular geometries or spillway
sections with long aprons may require more
rigorous analysis. The Finite Element Method
(FEM) permits the engineer to closely model
the actual geometry of the structure and
account for its interaction with the foundation.
For example, consider the dam in figure 8.
Note that the thinning spillway that forms the
toe of the dam is not stiff enough to produce
the foundation stress distribution assumed in
the gravity method. In this case, gravity
analysis alone would have under-predicted
base cracking.

Figure 8
3-16
 Finite element analysis allows not only modeling of the dam, but also the
foundation rock below the dam. One of the most important parameters in dam/foundation
interaction is the ratio of the modulus of deformation of the rock to the modulus of
elasticity of the dam concrete. Figure 9 illustrates the effect that this ratio has on
predicted crack length. As the modular ratio varies, the amount of predicted base
cracking varies also. As can be seen in figure 9, assuming a low deformation modulus
(E r), is not necessarily conservative.

Figure 9

In gravity analysis, the distribution of foundation shear stress is not specifically

addressed. However, it is implicitly assumed that shear stress is distributed uniformly
across the base. This assumption is arbitrary and not very accurate. Finite element
modeling can give some insight into the distribution of base contact stress. As can be
seen in figure 10, shear stress is at a maximum at the tip of the propagating base crack. In
this area, normal stress is zero, thus all shear resistance must come from cohesion. Also,
the peak shear stress is about twice the average shear
stress. An un-zipping failure mode can be seen here,
as local shear strength is exceeded near the crack tip,
the crack propagates causing shear stress to increase in
the area still in contact. This is one reason why this
chapter favors allowing lower factors of safety for no
cohesion analysis. In this example, the dam is being
significantly overtopped.

3-17 Figure 10
3-4.3.2 Two-Dimensional Finite Element Analysis

Two-dimensional finite element analysis is adaptable to gravity dam analysis when

the assumption of plane strain is used. Sections including auxiliary works can be
analyzed to determine their stress distribution. As seen above, 2-dimensional finite
element analysis allows the foundation, with its possible wide variation in material
properties, to be included with the dam in the analysis.

3-4.3.2.1 Uplift Loads for Finite Element Studies

Uplift pressures must be included in finite element studies. Pressures are

calculated using the same procedures as conventional gravity dam analyses as outlined in
Section 3-2.4. Figure 11 shows a very effective means of uplift application. The use of
a thin interface layer of elements (standard Q4 elements) allows the uplift pressure to be
applied to the bottom of the dam and the top of the foundation. The resulting stress
output for these interface elements then includes the effects of uplift. The procedure also
has the benefit of allowing interface elements to be systematically deleted so that a
cracked base analysis may be performed in an iterative manner. As in conventional
gravity analysis, whenever base cracking is indicated by the presence of tensile stress
normal to the foundation in the interface elements, the uplift distribution should be
modified accordingly.

Figure 11

There are many non-linear finite element codes available which allow base cracking and
sliding to be modeled automatically.

3-18
3-4.3.3 Three-Dimensional Finite Element Analysis

Three-dimensional (3-D) FEM should be used when the structure or loading is

such that plane strain conditions may not be assumed, such as when the geometry is such
that the stability of the dam depends upon stress distribution parallel to its axis, as is the
case of a gravity-arch dam which is curved-in
plan, or when a dam is in a narrow valley. Three
dimensional analysis allows the rigorous
determination of what forces will be applied to the
foundation, and where. If 3 dimensional behavior
is to be considered, features that enable horizontal
force transfer such as shear keys or curvature in
plane must be present

3-4.3.4 Analysis of Results of Finite Element Method Studies

It is important to realize that the question before the reviewer is whether or not the
dam will fail under a given loading condition. In the review of finite element analyses, it
is easy to lose sight of the original question in view of the voluminous stress output that
typically results. The reviewer should never forget that stress at a point in the dam may
or may not be informative with respect to whether or not the dam will fail. Unlike the
conventional gravity technique which pre-supposes failure mechanisms, namely sliding
and overturning, the standard linear elastic finite element method does not address failure
mechanisms. It is up to the reviewer to determine the value of the analysis based on how
it addresses the possibility of failure mechanisms.

Whatever distribution of stress that results from an finite element analysis, it

should be verified that global force and moment equilibrium are satisfied. In addition, the
stress states in individual elements must be within the limits of the material strength. For
example, if the analysis indicates tension at the dam/foundation interface, the analysis
should be re-run with tensile elements eliminated from the stiffness matrix.

Excessive shear stress at the interface can also be a problem. For example, figure
10 (3-4.3.1) shows that the peak shear stress on the dam/foundation interface is in
elements with zero normal stress. This means that there is no frictional resistance
available at this location, and that all shear stress must be transferred through cohesive
bond alone. If the reviewer questions the availability of cohesive strength at the
interface, the analysis should be re-run with the shear stiffness of these elements
effectively reduced so that shear stress can be re-distributed. This can be handled
automatically with many finite element programs using gap-friction elements.

3-19
3-4.4 Dynamic Methods

Dynamic analysis refers to analysis of loads whose duration is short with respect
to the first period of vibration of the structure. Such loads include seismic, blast, and
impact. Dynamic methods described in this chapter are appropriate to seismic loading.

Because of the oscillatory nature of earthquakes, and the subsequent structural

responses, conventional moment equilibrium and sliding stability criteria are not
valid when dynamic and pseudo dynamic methods are used. The purpose of these
investigations is not to determine dam stability in a conventional sense, but rather to
determine what damage will be caused during the earthquake, and then to determine if the
dam can continue to resist the applied static loads in a damaged condition with possible
loading changes due to increased uplift or silt liquefaction (See 3-2.4.3.5). It is usually
preferable to use simple dynamic analysis methods such as the pseudo dynamic method
or the response spectrum method (described below), rather than the more rigorous
sophisticated methods.

3-4.4.1 Pseudo Dynamic Method

This procedure was developed by Professor Anil Chopra as a hand calculated

alternative to the more general analytical procedures which require computer programs.
It is a simplified response spectrum analysis which determines the structural response, in
the fundamental mode of vibration, to only the horizontal component of ground motion.
This method can be used to evaluate the compressive and tensile stresses at locations
above the base of the dam. Using this information, degree of damage can be estimated
and factored into a post earthquake stability analysis. References 8 and 13 provide an
explanation of this method, and sample calculations.

3-4.4.2 Modal Dynamic Methods

Dynamic response analysis is typically performed using finite element modal

analysis. The major modes of vibration are calculated, and the response of the structure
to the earthquake is expressed as a combination of individual modal responses. There are
2 acceptable techniques for modal analysis, Response Spectrum Analysis and Time
History Analysis.

3-20
3-4.4.2.1 Response Spectrum Method

In the response spectrum method, the modes of vibration determined from finite
element modeling are amplitude weighted by a response spectrum curve which relates the
maximum acceleration induced in a single
degree of freedom mechanical oscillator to the
oscillator's natural period. A typical response
spectrum curve is shown in figure 12. Because
the timing of the peaks of individual modal
responses is not taken into account, and
because peaks of all modes will not occur
simultaneously, modal responses are not
combined algebraically. Modal responses are
combined using the SRSS (square root of sum
of squares) or the CQC (complete quadratic
combination) methods. For further guidance
on the use of this method, refer to chapter 11 of Figure 12
these guidelines, or reference 28.

3-4.4.2.2 Time History Method

The time history method is a more rigorous solution technique. The response of
each mode of vibration to a specific acceleration record is calculated at each point in time
using the Duhamel integral. All modal responses are then added together algebraically
for each time step throughout the earthquake event. While this method is more precise
than the response spectrum method for a given acceleration record, its results are
contingent upon the particulars of the acceleration record used. For this reason, time
history analysis should consider several accelerograms. For further guidance on the use
of this method, refer to chapter 11 of these guidelines, or reference 28.

3-4.4.3 Direct Solution Methods

The modal superposition methods described above require the assumption of

material linearity. Direct solution techniques solve the differential equations of motion
in small time steps subject to material stress strain relationships which can be arbitrary,
and therefore the development of damage can be accounted for. Their results are also
highly affected by the particular accelerogram used.

3-4.4.4 Block Rocking Analysis

3-21
 When dynamic analysis techniques such as those discussed above indicate that
concrete cracking will occur, a block rocking analysis can be done. This type of analysis
is useful to determine the stability of gravity structures or portions thereof, when it is
determined that cracking will progress to the extent that free blocks will be formed. The
dynamic behavior of free blocks can be determined by summing moments about the pivot
point of rocking. More information on this method can be found in reference 12, or in
Appendix 3B of this chapter.

3-4.4.5 Pseudo Static Method

The Pseudo Static method is not acceptable.

3-4.4.6 Reservoir Added Mass

During seismic excitation the motion of the dam causes a portion of the water in
the reservoir to move also. Acceleration of this added mass of water produces pressures
on the dam that must be taken into account in dynamic analysis. Westergaard derived a
pressure distribution assuming that the dam would move upstream and downstream as a
rigid body, in other words, the base and crest accelerations of the dam are assumed to be
identical. 27/ This pressure distribution is accurate to the extent that the rigid body
motion assumption is valid. The
dam's structural response to the
earthquake will cause additional
pressure. Figure 13 shows the
difference in pressure
distributions resulting from rigid
body motion and modal
vibration.

Figure 13

3-22
 Westergaard's theory is based on expressing the motion of the dam face in terms of
a fourier series. If the acceleration of the upstream face of the dam can be expressed as:

where " is the ground acceleration, then the resulting pressure is given by :

While, Westergaard assumed a rigid body acceleration, the above equations can be
generalized to accommodate any mode shape.

As with the application of finite element techniques for static analysis, the
reviewer must not lose sight of the purpose of the analysis, ie to determine whether or
not a given failure mode is possible. Finite element techniques assume linear stress strain
characteristics in the materials, and almost always ignore the effect of cracking in the
dam. These assumptions can constitute rather gross errors. For this reason when
reviewing the finite element results, the stress output should be viewed qualitatively
rather than quantitatively. Finite element dynamic output can show where the structure is
most highly stressed, but the stress values should not be considered absolute.

3-4.5 Cracked Base Analysis

The dam/foundation interface shall be assumed to crack whenever tensile stress

normal to the interface is indicated. This assumption is independent of the analysis
procedure used. The practical implementation of this requirement is illustrated in the
gravity analysis shown below.

3-4.5.1 Determination of Resultant Location - Static Cases Only

All forces, including uplift are applied

to the structure. Moments are taken about 0,0
which does not necessarily have to be at the
toe of the dam. The line of action of the
resultant is then determined as shown in the
figure 14. The intersection of the resultant
line of action and the sloping failure plane is
the point of action of the resultant on the
structure.

3-23 Figure 14
3.4.5.2 Determination of Theoretical Crack Length

A crack is assumed to develop between the base and foundation if the stress normal to the
base is tensile. Since the gravity analysis technique assumes a linear effective stress
distribution along the dam base, the
length of this crack is uniquely
determined by the location of the
resultant and the assumption of a linear
effective stress distribution. (See figure
15)

3-4.5.3 Cracking Induced by Dynamic

Loading

Dynamic loading is equally

capable of causing base cracking,
Figure 15
however, cracked base analyses are not
typically performed for dynamic loadings because of the computational difficulty
involved. The conventional gravity analysis procedure is not appropriate for dynamic
loading because it ignores the dynamic response of the structural system. Standard
dynamic finite element techniques are not appropriate because they are based on an
assumption of material linearity and structural continuity. What is typically assumed is
that during the earthquake, extensive base cracking does occur. Stability under post
earthquake conditions, which include whatever damage results from the earthquake, must
be verified.

3-4.6 Review of Computer Analyses

The FERC does not endorse specific computer programs. The FERC has on
occasion requested very detailed information about the internal workings of computer
programs. For this reason, those who submit computer analyses should have full
knowledge of not only what the results of the analysis were, but also why. No matter
which program is used, the engineer must stand behind the result.

Output data should be spot checked and compared to hand calculated solutions
wherever possible, to assure that the basic laws of statics have been satisfied, i.e.,
summation of forces and moments equals zero. For additional guidance on the finite
element method, refer to reference 29.

3-24
3-5 Stability Criteria

3-5.1 General

Specific stability criteria for a particular loading combination are dependent upon
the degree of understanding of the foundation structure interaction and site geology, and
to some extent, on the method of analysis.

Assumptions used in the analysis should be based upon construction records and
the performance of the structures under historical loading conditions. In the absence of
available design data and records, site investigations may be required to verify
assumptions.

Safety factors are intended to reflect the degree of uncertainty associated with the
analysis. Uncertainty resides in the knowledge of the loading conditions and the material
parameters that define the dam and the foundation. Uncertainty can also be introduced
by simplifying assumptions made in analyses. When sources of uncertainty are removed,
safety factors can be lowered.

3-5.2 Acceptance Criteria

3-5.2.1 Basic Requirements

The basic requirement for stability of a gravity dam subjected to static loads is that
force and moment equilibrium be maintained without exceeding the limits of concrete,
foundation or concrete/foundation interface strength. This requires that the allowable unit
stresses established for the concrete and foundation materials not be exceeded. The
allowable stresses should be determined by dividing the ultimate strengths of the
materials by the appropriate safety factors in Table 2.

3-5.2.2 Internal Concrete Stresses

In most cases, the stresses in the body of a gravity dam are quite low, however if
situations arise in which stress is a concern, the following guidance in table 1 is
applicable.

3-25
Table 1
Load Shear Stress on Pre-cracked Principal Axis Tension W ithin
Condition Failure Plane1 Intact Concrete 2,4

Worst Static 1.4 Fn 1.7(F'c)2/3

Max. Dynamic N.A. 3 N.A. 3

1
ACI 318 has specified that the ultimate shear strength of concrete along a
pre-existing crack in monolithically cast concrete is 1.4 times the normal
stress on the crack ( Fn) , provided of course that the normal stress is
compressive. (See reference 1)

2
Shear failure of intact concrete is governed by the tensile strength of
concrete normal to the plane of maximum principal axis tension. The limits
shown are taken from reference 16.

3
It is expected that earthquakes will induce stresses that exceed the strength
of materials. For this reason, this guideline does not specify an allowable
stress levels for this load case. Post earthquake analysis should be done
with using procedures outlined for static analysis. If dynamic analysis does
indicate that tensile cracking, shear displacements, or rocking are likely to
occur, post earthquake static allowables should be downgraded accordingly.
For example, if dynamic analysis indicates that a region will crack, the post
earthquake tensile stress allowable for that region would be 0. If dynamic
analysis indicates that shear failure will occur, then residual shear strenghts
should be used.

4
If pre-existing cracks exist, the tensile strength normal to the plane of the
crack is 0. Also, the tensile strength of horizontal lift joints within the dam
may be less than the parent concrete and testing may be required to
establish allowable stresses. This is especially true in RCC dams, which
often have low ensile strength across their lift joints.

The tensile strength of the rock-concrete interface should be assumed to be 0.

Rock foundations may consist of adversely-oriented joints or fractures such that even if
the interface could resist tension, the rock formation immediately below may not be able
to develop any tensile capacity. Therefore, since stability would not be enhanced by an
interface with tensile strength when a joint, seam or fracture in the rock only a few inches
or feet below the interface has zero tensile strength, no tension will be allowed at the
interface.

3-26
3-5.2.3 Sliding Stability Safety Factors

Recommended factors of safety are listed in table 2 and 2A.

TABLE 2

Recommended Minimum Factors of Safety 1/

Dams having a high or significant hazard potential.

Loading Condition 2/ Factor of Safety 3/

Usual 3.0
Unusual 2.0
Post Earthquake 4/ 1.3

Dams having a low hazard potential.

Loading Condition Factor of Safety

Usual 2.0
Unusual 1.25
Post Earthquake Greater than 1.0
Notes:

1/ Safety factors apply to the calculation of stress and the Shear Friction
Factor of Safety within the structure, at the rock/concrete interface and in
the foundation.

2/ Loading conditions as defined in paragraph 3-3.0.

3/ Safety factors should not be calculated for overturning, i.e., M r / M 0.

4/ For clarification of this load condition, see paragraph 3-4.4.

For definitions of "High", "Significant", and "Low" hazard potential dams, see
Chapter 1 of this guideline.

One of the main sources of uncertainty in the analysis of gravity dam stability is
the amount of cohesive bond present at the dam foundation interface. The FERC
recognizes that cohesive bond is present, but it is very difficult to quantify through
borings and testing. It has been the experience of the FERC that borings often fail to

3-27
recover intact interface samples for testing. In addition, strengths of intact samples that
are recovered exhibit extreme variability. For this reason, table 2A below offers
alternative recommended safety factors that can be used if cohesion is not relied upon for
stability.
TABLE 2A

Alternate Recommended M inimum Factors of Safety

for Use in Conjunction with a No Cohesion Assumption

Loading Condition Factor of Safety

Worst Static Case 5/ 1.5

Flood if Flood is PMF 6/ 1.3
Post Earthquake 1.3
Notes:

5/ The worst static case is defined as the static load case with the lowest factor
of safety. It shall be up to the analyst to determine the worst static case and
to demonstrate that it truly is the worst static case.

6/ Because the PMF is by definition the flood that will not be exceeded, a
lower factor of safety may be tolerated. Therefore if the worst static case is
the PMF, a factor of safety of 1.3 is acceptable. If the IDF is not the PMF,
then the safety factor for the worst static case shall control.

3-5.2.4 Cracked Base Criteria

For existing structures, theoretical base cracking will be allowed for all loading
conditions, provided that the crack stabilizes, the resultant of all forces remains within the
base of the dam, and adequate sliding safety factors are obtained. Cohesion may only be
assumed on the uncracked portion of the base. Limitations may be necessary on the
percentage of base cracking allowed if foundation stresses become high with respect to
the strength of the concrete or foundation material.

When remediation is required, the remediation should be designed to attempt to

eliminate theoretical base cracking for static load cases.

3-28
3-5.3 Safety Factor Evaluation

The safety factors determined in accordance with the previous sections shall be
evaluated on a case-by-case basis in order to assess the overall safety of a particular
project. Engineering judgment must be used to evaluate any calculated safety factor
which does not conform to the recommendations of tables 1, 2 or 2A of section 3-5.2. In
applying engineering judgment, consideration must be given to both the adequacy of the
data presented in support of the analyses and the loading case for which the safety factor
does not meet the criteria.

It is preferable to conservatively define strength parameters and loading conditions

than to utilize higher safety factors to accommodate uncertainties in the analysis.
Therefore, if the analyst can demonstrate that there is sufficient conservatism in the
strength parameters and analysis assumptions, lower factors of safety may be considered
adequate on a case by case basis. Any decision to accept safety factors lower than those
shown in Table 2A of this chapter will be based on: (1) the degree of uncertainty in the
data and analyses provided and (2) the nature of the loading condition, i.e. its probability
of exceedance.

In accepting any lower safety factor as outlined herein, the stability analyses must
be supported by a program that includes, but is not limited to, adequate field level
investigations to define material (dam and foundation) strength parameters, installation
and verification of necessary instrumentation to evaluate uplift assumptions and loading
conditions, a detailed survey of the condition of the structure, and proper analysis
procedures. This program should be submitted for approval by the Director, Division of
Dam Safety and Inspections. Flexibility on safety factors beyond that discussed above
will be infrequent and on special case-specific consideration.

3-5.4 Foundation Stability

3-5.4.1 Rock Foundations

The foundation or portions of it must be analyzed for stability whenever the

structural configuration of the rock is such that direct shear failure is possible, or
whenever sliding failure is possible along faults, shears and/or joints. Associated with
stability are problems of local over stressing in the dam due to foundation deficiencies.
The presence of such weak zones can cause problems under either of two conditions: (1)
when differential displacement of rock blocks occurs on either side of weak zones, and
(2) when the width of a weak zone represents an excessive span for the dam to bridge
over.

3-29
 Sliding failure may result when the rock foundation contains discontinuities and/or
horizontal seams close to the surface. Such discontinuities are particularly dangerous
when they contain clay, bentonite, or other similar substances, and when they are
adversely oriented.26/ Appropriate foundation investigation and exploration must be
done to identify potential adverse features.

3-5.4.2 Soil Foundations

Gravity dams constructed on soil foundations are usually relatively small

structures which exert low bearing pressures upon the foundation. Large structures on
soil foundations are usually supported by bearing or friction piles. Piles supported
structures are addressed in Chapter 10 of these guidelines. When the foundation consists
of pervious sands and gravels, such as alluvial deposits, two possible problems exist; one
pertains to the amount of underseepage, and the other is concerned with the forces
exerted by the seepage. Loss of water through underseepage may be of economic
concern for a storage or hydro electric dam but may not adversely affect the safety of the
dam. However, adequate measures must be taken to ensure the safety of the dam against
failure due to piping, regardless of the economic value of the seepage.

The forces exerted by the water as it flows through the foundation can cause an
effective reduction in the weight of the soil at the toe of a dam and result in a lifting of
the soil. If uncontrolled, these seepage forces can cause a progressive erosion of the
foundation, often referred to as "piping" and allow a sudden collapse of the structure.
The design of the erosion, seepage and uplift control measures requires extensive
knowledge of type, stratification, permeability, homogeneity, and other properties of the
foundation materials.

One way to limit this type of material transport is to insure that the weighted creep
ratio is greater than the minimum values shown in Table 3. The weighted creep ratio is
defined as the total weighted creep distance Lw , defined in section 3-2.4.4.1, divided by
the head differential (HW-TW).

3-30
 Table 3 17/
Minimum Weighted Creep Ratios, Cw for Various Soils

Very fine sand or silt 8.5

Fine sand 7.0
Medium sand 6.0
Coarse sand 5.0
Fine gravel 4.0
Medium gravel 3.5
Coarse gravel including cobbles 3.0
Boulders with some cobbles 2.5
and gravel.

Some of the control measures which may be required may include some, all or
various combinations of the following devices:

a. Upstream apron, usually with cut offs at the upstream end.

b. Downstream apron, with scour cut offs at the downstream end, and with or
without filters and drains under the apron.

c. Cutoffs at the upstream or downstream end or at both ends of the overflow

section, with or without filters or drains under the section.

A detailed discussion of these measures and their usages is given in reference 7. For
guidance on the evaluation of concrete dams on earth soil foundations, refer to chapter 10
of these guidelines or reference 17.

3-6 Construction Materials

3-6.1 General

The compressive stresses in a concrete in a gravity dam are usually much lower
than the compressive strength of the concrete. Therefore compressive strength is rarely
an issue. Tensile strength is typically the limiting criteria. It is addressed in section 3-
5.2.2 of this chapter.

3-31
 3-6.2 Site Investigations

During staff review of foundation investigation reports, staff geologists and soils
engineers should be consulted concerning the adequacy of the data submitted with respect
to defining the structural and geological capability of the foundation. Foundation borings
and testing can be helpful in identification of weak zones in the foundation beneath the
dam. In addition, construction photographs of the foundation during construction can
provide valuable information on the characteristics of the dam/foundation interface and
on the orientation of jointing. Specific details concerning geological investigations are
contained in Chapter 5 of these guidelines.

3-6.3 Concrete Properties

3-6.3.1 General

Many factors affect the strength and durability of mass concrete. The concrete
must be of sufficient strength to safely resist the design loads throughout the life of the
structure. Durability of the concrete is required to withstand the effects of weathering
(freeze-thaw), chemical action and erosion.

In recent years, the use of Roller Compacted Concrete (RCC) has become
increasingly popular as a construction material for new gravity dams, to repair existing
concrete structures and to armor existing embankment dams against overtopping events.
While RCC used to construct gravity dams should have the same general properties as
conventional mass concrete, there are some differences which must be considered. For
example, the construction of roller compacted concrete results in horizontal lifts joints at
about 1 foot spacing. The potential for failure along these lift joints due to lower strength
and higher permeability must be considered. More information on RCC dams can be
found in references 2, 11, and 24.

3-6.3.2 Structural Properties

Stresses in a gravity dam are usually low; therefore, concrete of moderate strength
is generally sufficient to withstand design loads. Laboratory tests are often unnecessary
if conservative assumptions on concrete strength result in adequate factors of safety.
Tests can be performed if concrete parameters are in question.

For existing structures, non-destructive acoustic testing techniques have proven

valuable for the qualitative evaluation of concrete strength and continuity. Drilling and
testing can also be performed. Drilling and testing should be used to correlate concrete
strength with acoustic wave velocities.

3-32
 Staff review of these tests should compare the laboratory results to the original
design assumptions, and should examine the testing procedures to determine if the tests
were conducted in conformance with recommended ASTM and ACI procedures as listed
below:

a. Compressive tests: ASTM-C39

b. Tensile tests: ASTM C78

c. Shear tests: RTH 203-80 23/

d. Modulus of elasticity, Static: ASTM C469, Dynamic: ASTM C215

e. Poisson's ratio: ASTM C469

f. Collection of test samples: ASTM C31, C172, and C192

g. Evaluation of test results: ACI 214

Additional guidance concerning the design of mass concrete mixes and the
determination of the cured properties of the concrete are presented in reference 17.

3-6.3.3 Durability

The durability of concrete or RCC is influenced by the physical nature of the

component parts, and although performance is largely influenced by mix proportions and
degree of compaction, the aggregates constitute nearly 85 percent of the constituents in a
mass concrete and good aggregates are essential for durable concrete. The environment
in which the structure will exist must be considered in the mix design and in the
evaluation of the suitability of aggregate sources proposed for use in the mix. Generally,
the environmental considerations which must be examined are: weathering due to
freezing and thawing cycles; chemical attack from reactions between the elements in the
concrete, exposure to acid waters, exposure to sulfates in water and leaching by
mineral-free water; and erosion due to cavitation or the movement of abrasive material in
flowing water.

3-33
3-6.3.4 Dynamic Properties

The compressive and tensile strengths of concrete vary with the speed of testing.
As the rate of loading increases compressive and tensile strengths and modulus of
elasticity also increase, therefore, these properties of concrete under dynamic loadings
such as during an earthquake are greater than under static conditions.

References 16 and 18 provide a detailed discussion of the rates and types of testing
which should be conducted to determine the dynamic properties of concrete for use in
linear finite element analyses. The rates of testing should be coordinated with the
expected stress cycles of the design seismic event.

3-6.4 Foundation Properties

In many instances, a gravity dam is keyed into the foundation so that the
foundation will normally be adequate if it has enough bearing capacity to resist the loads
from the dam. If, however, weak planes or zones of inferior rock are present within the
foundation, the stability of the dam will be governed by the sliding resistance of the
foundation. The foundation investigations should follow the recommendations of
Chapter 5 of these guidelines, and should establish the following strength parameters for
use in stability and stress analyses:

a. Shear strengths along any discontinuities and the intact rock.

b. Bearing capacity (compressive strength).

c. Deformation Modulus of the rock mass.

d. Poisson's ratio of the rock mass.

These parameters are usually established by laboratory tests on samples obtained

at the site. In some instances, in situ testing may be justified. In either instance, it is
important that samples and testing methods be representative of the site conditions. The
results of these tests will, generally, yield ultimate strength or peak values and must,
therefore, be divided by the appropriate factors of safety in order to obtain the allowable
working stresses. Recommended factors of safety are presented in table 2 of section 3-
5.2.

Foundation permeability tests may be helpful in conjunction with the drilling

program, or as a separate study, in order to establish uplift parameters and to design an
appropriate drainage system. Permeability testing programs should be designed to

3-34
establish the permeability of the rock mass and not an isolated sample of the rock
material. The mass permeability will usually be higher, due to jointing and faulting, than
an individual sample.

Prior to the selection of representative foundation properties, all available geologic

and foundation information should be reviewed for descriptions of the type of material
and structural formation on which the dam was constructed. A general description of the
foundation material can be used as a basis for choosing a range of allowable strengths
from published data, if testing data is not available. Staff geologists should be consulted
if the available information refers to material parameters or structural features which are
suspected to be indications of poor foundation conditions. Situations which should alert
the engineer to possible problem areas are listed below:

a. Low RQD ratio (RQD = Rock Quality Designation).

b. Solution features such as caves, sinkholes and fissures.

c. Columnar jointing.

d. Closely spaced or weak horizontal seams or bedding planes.

e. Highly weathered and/or fractured material.

f. Shear zones or faults and adversely oriented joints.

g. Joints or bedding planes described as slickensided, or filled with gouge

materials such as bentonite or other swelling clays.

h. Foliation surfaces.

i. Drill fluid loss.

j. Large water takes during pumping tests.

k. Large grout takes.

l. Rapid penetration rate during drilling.

Compressive - In general, the compressive strength of a rock foundation will be

greater than the compressive strength of the concrete within the dam. Therefore, crushing
(or compressive failure) of the concrete will usually occur prior to compression failure of

3-35
the foundation material. When testing information is not available this can be assumed,
and the allowable compressive strength of the rock may be taken as equal to that of the
concrete. However, if testing data is available, the safety factors from Table 2 should be
applied to the ultimate compressive strength to determine the allowable stress. Where the
foundation rock is nonhomogeneous, tests should be performed on each type of rock in
the foundation.

Tensile - A determination of tensile strength of the rock is seldom required

because unhealed joints, shears, etc., cannot transmit tensile stress within the foundation.
Therefore, the allowable tensile strength for the foundation should be assumed to be zero.

Shear - Resistance to shear within the foundation and between the dam and its
foundation depends upon the zero normal stress shear strength (cohesion) and internal
friction inherent in the foundation materials, and in the bond between concrete and rock
at the contact surface. Ideally, these properties are determined in the laboratory by
triaxial and direct shear tests on samples taken during construction, during a post-
construction drilling program, or in the field through insitu testing. The possible sliding
surface may consist of several different materials, some intact and some fractured. Intact
rock reaches its maximum break bond resistance with less deformation than is necessary
for fractured materials to develop their maximum frictional resistances. Therefore, the
shear resistance developed by each fractured material depends upon the displacement of
the intact rock part of the surface. This raises several issues, including strain
compatibility, point crushing strength, creep, and progressive failure which must be
considered in the selection of reasonable shear strength parameters. The shear resistance
versus normal load relationship for each material along the potential sliding plane should
be determined by testing wherever possible. Staff geotechnical engineers should be
consulted concerning the adequacy of any foundation evaluation program and the
interpretation of test results.

In many cases, photographic records of the foundation before and during

construction are very useful in estimating overall foundation contact shear strength.
Large scale roughness which interrupts shear planes can force a shear through rock or
shear through concrete situation, justifying apparent cohesion, or much higher friction
angles than small sample testing would indicate. The reviewer should be aware however,
that there may be one "weak link" in the foundation. If large scale asperities prohibit
sliding along the interface between concrete and rock, attention should be focused on
other area, such as planar concrete lift joints, or adversely oriented rock joints beneath the
dam.

3-36
3-7 References

1. American Concrete Institute, "Building Code Requirements for Reinforced

Concrete", ACI-318-95

2. American Concrete Institute, "Roller Compacted Concrete", ACI 207.5R-80.

3. Brand, B., Nappe Pressures on Gravity Dam Spillways, Dam Engineering Vol X,
Issue 2 , August 1999

4. Bureau of Reclamation, "Cavitation in Chutes and Spillways" Engineering

Monograph No. 42, 1990

5. Bureau of Reclamation, "Design of Small Dams", 1974 edition.

6. Cannon, Robert W.,"Design Considerations for Roller Compacted Concrete and

Rollcrete in Dams", Concrete International Magazine, Dec. 1985.

7. Cedergren, Harry R., "Seepage, Drainage and Flow Nets", John Wiley & Sons,
2nd Ed.

8. Chopra, A. K., Fenves, G.,"Simplified Earthquake Analysis of Concrete Gravity

Dams," ASCE Structural Journal August 1987.

9. Clough, Ray W., "The Finite Element Method in Plane Stress Analysis", ASCE
Conference Papers, Sept. 1960.

10. "Handbook of Dam Engineering", A. R. Golze (editor), Van Nostrand Reinhold

Co., C 1977.

11. Hansen, K. D. & Reinhardt, W.G., "Roller Compacted Concrete Dams" 1991,
McGraw-Hill

12. Housner, G. W., "The Behavior of Inverted Pendulum Structures During

Earthquakes" Bulletin of the Seismological Society of America, Vol 53, February
1963.

13. Jansen, R. B., "Advanced Dam Engineering" 1988, Van Nostrand Reinhold

3-37
14. Lund, Boggs, & Daley, "Soda Dam: Influence of Reservoir Silt Deposits on Uplift
Load" Proceedings of the International Conference on Hydropower, 1993

15. Newmark. N. M . and Rosenblueth, E. Fundamentals of Earthquake Engineering,"

Prentice Hall, 1971.

16. Raphael, J. M., "Tensile Strength of Concrete", Journal of The American Concrete
Institute, Mar/Apr 1984, p158.

17. Terzaghi & Peck, "Soil Mechanics in Engineering Practice," John Wiley & Sons,
copyright 1948. 2nd Edition.

18. U. S. Army Corps of Engineers, "Earthquake Design and Analysis for Corps of
Engineers Dams", ER-1110-2-1806, April 30, 1977.

19. U. S. Army Corps of Engineers, "Gravity Dam Design". EM-1110-2-2200, June

1995.

20. U. S. Army Corps of Engineers, EM-1110-2-1603 "Hydraulic Design of

Spillways," 31 March 1990.

21. U. S. Army Corps of Engineers, EM-1110-2-1612 "Ice Engineering" 15 October

1982.

22. U. S. Army Corps of Engineers, EM-1110-2-2502, "Retaining Walls," 29 May

1989.

23. U. S. Army Corps of Engineers, "Rock Testing Handbook (Standard and

Recommended Methods)" August 1980.

24. US Army Corps of Engineers, EM 1110-2-2006, "Roller Compacted Concrete", 20

August 1985.

25. U. S. Army Corps of Engineers, EM 1110-2-1901 "Soil Mechanics Design," Ch. 1,

Seepage Control, February 1952.

26. USCOLD "Anthology of Dam Modification Case Histories", April 1996 - M orris
Sheppard Dam

27. Westerdaard, W. M. ,"Water Pressures on Dams During Earthquakes",

Transactions ASCE, Vol 98, 1933.

3-38
28. Wiegel, R. L., "Earthquake Engineering" 1970, Prentice-Hall

29. Zienkiewicz, O. C., "The Finite Element in Structural and Continuum Mechanics",
McGraw-Hill, London.

3-39
 APPENDIX 3A
NAPPE PRESSURES

This appendix presents a simplified method for the determination of nappe

pressures given the following assumptions:

1) Streamlines are concentric and parallel to the spillway surface.

2) The curvature of streamlines changes gradually with respect to distance

along the streamline.

3) Flow is irrotational.

4) Energy is not dissipated by friction or aeration.

Figure 1A shows that the velocity and pressure exerted by the water are a function
of position and spillway curvature. The generalized equation for unit discharge is shown
below 3/:

Where:

q= Unit discharge

E= Total Energy

Y= Elevation of the point on the

spillway under consideration

A= Depth of flow measured

perpendicular to the spillway
surface

N= The angle of the outward

directed normal to the
spillway with respect to
horizontal
Figure 1A
6= Curvature of spillway surface
at point under consideration, Positive for flip buckets, negative for crests
 For a given q, equation 1 can be solved for A. A numerical procedure is required.
With the flow depth A at a point determined, the velocity of flow at the spillway surface
can be found using equation 2:

The pressure head at the spillway surface is then:

Using equations 1, 2, and 3, the

pressure at any point on a spillway by
the overflowing nappe can be
determined. Figure 2A shows the
application of this procedure to a
typical overflow spillway section.
Note that at the design discharge, the
nappe exerts almost no pressure on the
downstream face of the dam until flow
direction is changed by the bucket.
Bucket pressures are large, and tend to
overturn the dam since they are
exerted downstream of the base
centroid. Crest pressures are typically
small and can be negative. When they
are negative, they also tend to overturn
the dam. Figure 2A

The net external hydraulic resultant forces are as follows:

F X = 97.5 kips (Downstream) @ Y= 21.6'

F Y = 28.4 kips (downward) @ X= 43.6'

Note that the net nappe force on the dam is totally independent of the tailwater
elevation. This is a consequence of the fact that flow downstream of the crest is super-
critical, and thus not subject to downstream control. This is true as long as the tailwater
elevation is less than the conjugate depth.
Further treatment of hydrodynamic effects is given in reference 3.
 APPENDIX 3B
ROCKING RESPONSE OF BLOCKS

This appendix describes a method for the investigation of the response of rigid
blocks in response to seismic excitation. Dynamic moment equilibrium of the block
shown in figure 1B requires the satisfaction of equation
1.

EQ 1

Figure 1B
Where:
AB Acceleration of the block base. This is the ground acceleration if the block
is sitting on the ground. It is the acceleration modified by structural
response if the block is sitting on top of a structure.

W Weight of the block.

M Mass of the block.

(w Dw Weight and mass density of water.

R, ", N and d are as shown in figure 1B

The first term of the equation represents the moment about the pivot point
produced by horizontal forces resulting from horizontal acceleration of the center of
gravity of the block. The second term represents the moment about the pivot point
produced by vertical accelerations of the center of gravity of the block. The third term
represents the static moment produced by the weight of the block. The fourth term
represents the static moment produced by the reservoir. The HydroDyne term represents
the moments produced by hydrodynamic reservoir pressure. There are 2 components of
the HydroDyne term, one due to the horizontal acceleration of the dam and one due to the
rotational acceleration of the block.
 EQ 2

Where:

Equation 1 and 2 can be combined as shown below:

EQ 3

Where IP is the polar moment of inertia about the

pivot point and Dc. is the mass density of concrete.

Equation 3 can be solved numerically. Each time

the block rocks from one pivot point to another, equation 3
must be modified accordingly. In addition, significant
energy loss occurs each time the block changes pivot points
12/. Figure 2B shows the application of this technique for
a top block rocking in response to a sinusoidal block base
acceleration of 3 Gs at various frequencies. In this case,
the block is not subject to reservoir forces.

Figure 2B
 As can be seen, excitation
frequency has a large effect on the
stability of a rocking block.

The result of these analyses

is often a finding that while
seismic forces may crack the
concrete, they can not topple the
free blocks that result.

Figure 3B
 APPENDIX C

EXAMPLE GRAVITY ANALYSIS

Relevant parameters are as

follows:

Unit Weight of Concrete (

.150 kcf

Friction Angle N
45 /

Drain Effectiveness E
50%

Slope of Dam Foundation "

-7.125/

Figure 1C

DETERMINATION OF CRACK LENGTH

Initially assume that the base of the dam is not cracked, and that the uplift is
distributed as shown in figure 5 of this chapter, with T=0.

The location of the dam with respect to vertical and horizontal datum is un-
important. To illustrate this point, the global coordinate system will not be placed at the
toe of the dam. The forces applied to the dam are as shown in figure 2-c.
 Figure 2C

ITERATION # 1, ASSUMED CRACK LENGTH= 0

FORCE DESCRIPTION F-> ARM F^ ARM M @ 0,0

)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
DAM DEAD LOAD--> -630.00 126.55 79725.00
RESERVOIR LOAD--> 312.00 133.33 41600.00
TAILWATER LOAD--> -3.12 93.33 -2.18 177.67 96.82
UPLIFT--> 22.23 96.43 177.84 128.60 -20726.06
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
TOTAL FORCE = 331.11 -454.34 100695.77

RESULTANT LINE Y AXIS INTERCEPT @ 304.1158, X AXIS INTERCEPT @ 221.6289

RESULTANT INTERSECTS BASE @ 153.64 , 93.30
 The horizontal distance from the toe of the dam to the resultant/base intersection
point is:
180 - 153.64 = 26.36'

Since the base pressure distribution is assumed to be triangular, the resultant acts
at the 1/3 point of the base pressure distribution. Thus the length of the base pressure
distribution is:

3*26.36 = 79.08'

Note that 79.08' is less

than the horizontal base length
(80'), and therefore a crack must
be assumed to initiate at the dam
heel. This crack will effect the
uplift distribution since it must be
assumed that full reservoir head
will occur along the crack length.

The new crack length can

be assumed to be the difference
between the full base length and
the length of the base pressure
distribution, and the proceedure
repeated. The process concludes
when the assumed crack length
no longer changes. Figure 3-C
shows the results of the final
iteration.

Figure 3C
ASSUMED CRACK LENGTH= 5.26

FORCE DESCRIPTION F-> ARM F^ ARM M @ 0,0

))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
DAM DEAD LOAD--> -630.00 126.55 79725.00
RESERVOIR LOAD--> 312.00 133.33 41600.00
TAILWATER LOAD--> -3.12 93.33 -2.18 177.67 96.82
UPLIFT--> 24.17 96.54 193.33 127.70 -22354.66
))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
TOTAL FORCE = 333.05 -438.85 99067.17

RESULTANT LINE Y AXIS INTERCEPT 297.46, X AXIS INTERCEPT 225.74

RESULTANT INTERSECTS BASE @ 155.075 , 93.12
 As can be seen, the change in uplift force resulting from a different crack
assumption caused the location of the resultant/base intersection to change. The
horizontal distance from the toe of the dam to the resultant/base intersection point is:

180 - 155.75 = 24.92'

The length of the base in compression is:

3*24.925 = 74.775'

The new horizontal crack length is:

80 - 74.775 = 5.225'

The new crack length as measured along the base is:

The difference between the assumed crack length of 5.26 and 5.27 is negligible,
indicating that the correct crack length has been determined within a to lerance of .01'.

DETERMINATION OF SLIDING STABILITY

With the crack length and uplift forces determined, the sliding stability calculation
proceeds as follows:

The angle of the resultant force with respect to the vertical is:

The failure cone of the dam/foundation interface is shown in figure 4-C. The factor
safety for sliding is defined as:
In this example, this results in:

Figure 4C
 APPENDIX 3D
Dynamic & Post Earthquake Analysis

The flow chart below depicts the seismic analysis process applicable to concrete
gravity dams.

EXAMPLE

1. Dynamic Stress Analysis

The dynamic analysis of the RCC dam depicted below indicates that seismic
stresses in the vertical direction (across RCC lift joints) is approximately 4 times the lift
joint tensile strength when subjected to a .6g base excitation. The large top block of the
non-overflow section is of special concern. In all probability, the lift joint at elevation
7175.5 will fail in tension and the block will begin to rock back and forth in response to
seismic ground motion.
2. Block Rocking Analysis

The dynamic stress analysis

indicates that there is a likelihood of
tensile lift joint failure. Rather than try
to quantify how much cracking will
occur, block rocking analysis assumes
that the base of the block is completely
broken and that the block is free to pivot
about either upstream or downstream
corner. The free block rocking analysis
is described in Appendix 3B. The .6g
base acceleration will be amplified by the
response of the structure. To account for
this, the free block will be subjected to a
sinusoidal base acceleration of twice the
.6g dam base acceleration.

The response of the block subjected to 1.2 g sinusoidal acceleration is depicted

below. Time history analysis of the dam can be used to provide a block base
accelerogram, however, a continuous sinusoidal excitation with an amplitude equal to the
structurally amplified peak ground acceleration is conservative. 2 hz was selected for a
sample excitation frequency for this example. Typically, several excitation frequencies
should be used to check the
sensitivity of the result to
frequency.

As can be seen, the block

subjected to excitation of this
type will experience significant
rocking. The adjacent figure
shows that the crest will tilt
downstream as far as .9' and
upstream as far as .5'. (Upstream
deflections are positive)
However, the block will not
topple .
3. Post Earthquake Stability Analysis

Since the block rocking analysis shows that significant rocking could take place,
the block must be analyzed to determine if it can still resist static loads in a damaged
condition. For the post earthquake analysis, a residual shear strength of 30 / with no
cohesion will be assumed. Stability analysis procedures are outlined in Appendix 3C.
FORCE DESCRIPTION F-> ARM F^ ARM M @ 0,0
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
DAM DEAD LOAD--> -124.20 12.00 1490.40
RESERVOIR LOAD--> 18.73 183.67 3439.67
UPLIFT--> 18.35 8.00 -146.78
)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))
TOTAL FORCE = 18.73 -105.85 4783.30

CRACK LENGTH= 0 100 % OF BASE IN COMPRESSION

HEEL OR CRACK TIP STRESS= -2.053036 TOE STRESS= -6.768102 (KSF)

SLIDING SAFETY FACTOR= 3.263296

The analyses above indicate that while significant seismic damage may occur, the
as-damaged top block will continue to maintain the reservoir.
 APPENDIX 3E
APPLICATION OF PSEUDO DYNAMIC PRINCIPLES FOR FINITE ELEMENT
ANALYSIS

Consider the hypothetical gravity dam shown in Fig 1E.

Assume that the PGA of the MCE is 0.25 g with a
frequency spectrum that peaks at 5 Hz.

Step 1 Determine natural frequency and mode shape of the

dam independent of reservoir interaction

Natural frequencies are available through almost all

standard finite element programs. The mode shape of the
first mode is as shown in Fig. 2E. For Young's modulus of
3,500,000 psi the natural frequency calculated by a standard
finite element modal analysis is 7.0 Hz.

Plotting the modal deflection of the upstream face as shown

the finite element normalized mode shape can be compared
to Chopra's 8/ generalized mode shape. From Fig. 3E, it is Fig. 1E
clear that Chopra's generalized mode shape is almost
identical to that calculated for this section by the finite element method . This will generally be
the case. It should be noted however, that while the Chopra mode shape matches the finite
element horizontal upstream face deflections very well, it does not include the vertical
components of the mode shape. These vertical components can be seen by noticing that the
horizontal lines of the model are no longer horizontal. ( Fig. 2E.) In this case, the vertical
component of dam response constitutes about 8% of the total generalized mass for this mode.

Fig. 2E Fig. 3E
Step 2 Determine the effect of the reservoir on the natrual frequency and the modal participation
factor

The generalized modal vibration equation is as follows:

2
M * (d Z ) + K * (Z ) = L ag Eq. 1E
dt 2
Where:
M*= m(uX2 + uY2), uX , uY being the modal deflection of a node in the X
(horizontal) and Y (vertical) directions, and m being the mass associated
with the node. The sum is over all nodes in the finite element model.

K*= Generalized stiffness. The change of total elastic strain energy with respect
to a differential variation in modal amplitude.

Z= Modal amplitude, function of time only. Total nodal motion is given by

Z•(uX), Z•(uY)

L= Effective driving force factor, m(uX + uY). If only horizontal ground

accelerations are considered then m(uX ).

ag= Ground acceleration. Function of time.

Using equation 1E, it can be seen that the angular frequency of the mode N given by:

ω N = K*
M * Eq. 2E
Any finite element program that is capable of modal analysis calculates N , M* , K*, L , but
some may not output M* and L . If they do not, M* and L can be calculated from the modal
displacement output as follows:
TABLE 1E - M*, L
NODE MODAL MODAL MASS MODAL EFFECTIVE
DISPL, X DISPL, Y on NODE MASS, M* DRIVING
DIRECTION DIRECTION FORCE, L
1 uX uY m m(uX2 + uY2) m(uX )
2
N

m(uX2 + uY2) m(uX )

Results for this example > 1.00 3.64
The reservoir effects can now be added in. The effect of the reservoir is to increase the amount
of total mass in the system and to provide additional driving force to the system, thus the M*
term and L term must be modified.

The dynamic reservoir pressure has 2 separate components;

1) A rigid body component that completely independent of the dam's dynamic response.
This pressure distribution was originally derived by Westergaard 27/

2) A mode shape conformal component that is a result of the dam's dynamic response.

The relative magnitudes of these

two pressure distributions are a
function of the dams structural
response. If the structural response
amplifies the ground acceleration by
a factor of 2 or 3, which is often the
case, the mode shape conformal
pressure becomes significant.

Both of these distributions are

determined by evaluating infinite
sine series. (See section 3-4.4.6)
Because Chopra's generalized mode
shapes fits most dams very well, a
generalized mode shape conformal
pressure distribution can be derived
from it. (See Fig. 5E) Both of these
plots are normalized assuming a 1' Fig. 4E
high dam with an acceleration of 1g.
In the mode shape conformal case, the acceleration varies from 1g at the water surface, to 0 at the
base. To get pressures in ksf, the plot values must be multiplied by the reservoir depth (H), the
actual acceleration at the water surface elevation (a ) in g's , and the weight density of water
(0.0624 kcf).
Fig. 5E
Because the mode shape conformal pressure distribution represents additional mass that
participates in the modal vibration of the dam, M* must be increased by :

∫P
0
msc u X dy Eq. 4E

Where Pmsc is the mode shape conformal pressure per unit acceleration. Also, effective driving
force factor, L must be increased by:

∫P
0
rigid u X dy Eq. 5E

Using the finite element method requires these integrals become sums of nodal forces times
nodal displacements. Tables 2E and 3E demonstrate this process.

TABLE 2E- HYDRODYNAMIC COMPONENT OF M*

A B C D E F
2
Height Finite elem. Value from Pressure/unit accel. (1'/sec ) Nodal force
above modal displ.. conformal (C)(uxI152)(152)(.0624/32.2) (D)(trib. area) (E)(B)
base (ux) plot (Fig.5E)
0 0.000 0.085 0.0130 0.052 0.0000
8 0.002 0.086 0.0132 0.105 0.0002
16 0.007 0.088 0.0135 0.108 0.0007
24 0.013 0.091 0.0139 0.111 0.0014
32 0.022 0.094 0.0144 0.115 0.0025
40 0.032 0.098 0.0149 0.119 0.0038
48 0.045 0.102 0.0155 0.124 0.0056
56 0.060 0.106 0.0162 0.130 0.0078
64 0.078 0.111 0.0170 0.136 0.0106
72 0.098 0.116 0.0177 0.142 0.0139
80 0.122 0.121 0.0185 0.148 0.0180
88 0.148 0.126 0.0192 0.153 0.0228
96 0.179 0.130 0.0198 0.158 0.0283
104 0.213 0.132 0.0203 0.162 0.0344
112 0.251 0.134 0.0204 0.163 0.0409
120 0.294 0.131 0.0201 0.159 0.0469
128 0.343 0.123 0.0189 0.149 0.0512
136 0.398 0.108 0.0165 0.128 0.0511
144 0.457 0.076 0.0117 0.084 0.0385
152 0.520 0.000 0.0000 0.016 0.0081
160 0.582 0.000 0.0000 0.000 0.0000
Total additional modal mass —> 0.3867
 TABLE 3E - HYDRODYNAMIC COMPONENT OF L
A B C D E F
2
Height Finite elem. Value from Pressure/unit accel.(1'/sec ) Nodal force
above modal displ.. Westrgrd (C)(152)(.0624/32.2) (D)(trib. area) (E)(B)
base (ux) plot (Fig.5E)
0 0.000 0.74 0.2187 0.874 0.0000
8 0.002 0.74 0.2183 1.745 0.0036
16 0.007 0.74 0.2171 1.735 0.0113
24 0.013 0.73 0.2150 1.719 0.0224
32 0.022 0.72 0.2121 1.696 0.0366
40 0.032 0.71 0.2084 1.666 0.0536
48 0.045 0.69 0.2037 1.628 0.0732
56 0.060 0.67 0.1981 1.584 0.0952
64 0.078 0.65 0.1916 1.531 0.1191
72 0.098 0.62 0.1840 1.470 0.1444
80 0.122 0.59 0.1753 1.400 0.1704
88 0.148 0.56 0.1654 1.321 0.1960
96 0.179 0.52 0.1541 1.231 0.2198
104 0.213 0.48 0.1414 1.129 0.2401
112 0.251 0.43 0.1269 1.013 0.2544
120 0.294 0.37 0.1104 0.880 0.2591
128 0.343 0.31 0.0913 0.726 0.2492
136 0.398 0.23 0.0690 0.545 0.2167
144 0.457 0.14 0.0413 0.312 0.1429
152 0.520 0.00 0.0000 0.055 0.0286
160 0.582 0.00 0.0000 0.000 0.0000
Total additional driving force factor --> 2.5365

The M* from the computation depicted in Table 1E is 1.000. Because M* is a function of

modal amplitude, the finite element code used in this example sets modal amplitude at the value
that causes M* to be 1.00., but this may not always bethe case. Adding the result from Table
2E:
M* = M*structure + M*water. = 1.00 +.3867 = 1.3867

Since the modal mass has increased, the natural frequency must decrease as can be seen from
Equation 3E. The new natural frequency is:

*
M structure 1
fN = fN = 7 = 5.94Hz
*
M structure *
+ M water 1.3867

Also, the L term from the computation depicted in Table 1E is 3.64. Adding the result from
Table 3E:

L = L structure + L water = 3.64 + 2.5365 = 6.1765

The modal participation factor L / M* is then 6.1765/1.3867 = 4.45. Note that the reservoir has
caused the natural frequency to drop by 15% and the modal participation factor to increase by
22%.

Step 3 Application of pseudo dynamic loads to finite element model

The magnitude of structural

response to a given earthquake is a
function of where the natural
frequency of the structure lies on the
spectral acceleration plot, and the
modal participation factor
( L / M*). For this example, the
spectral acceleration plot shown in
Figure 6E will be used. This plot is
not representative of any particular
earthquake, rather it is the spectral
acceleration resulting from the
ground motion .0082sin[(2p)(5)(t)].
This ground motion produces a
0.25g peak ground acceleration at 5
Hz. Fig. 6E

The example structure had a natural frequency of 5.94 Hz, or a period of 0.168 seconds. From
Figure 6E, the spectral acceleration is 0.85 gs, (27.37 ft/sec2) which is 3.4 times the peak ground
acceleration of .25 gs.( 8.05 ft/sec2) .

The Y axis intercept of the spectral acceleration curve in Figure 6E is the peak ground
acceleration. This is the acceleration applied to a perfectly rigid body. The acceleration over and
above the peak ground acceleration is that which is contributed by the structure's dynamic
response. The dynamic response portion of the acceleration in this therefore:

0.85gs - 0.25gs = 0.6 gs (19.32 ft/sec2)

This partition of the spectral response into rigid body and modal response components is
necessary not only because of the partition of the hydrodynamic pressures, but to account for the
structures own response. The pseudo dynamic loads derived from this process are applied to a
static finite element model. The partition shows that if the structure is completely rigid, and
there is no modal response, the applied forces reduce to those of the old pseudo static method;
the peak ground acceleration times the structures mass plus the Westergaard pressure, which is
exactly what one would expect.
The pseudo dynamic loads can now be applied to the static finite element model to determine
earthquake induced stresses. Pseudo dynamic nodal loads Fx, Fy, are as follows:

Structure –> Fxs = [(uX )(19.32 ft/sec2)( L / M*) + ( 8.05 ft/sec2)](m)

dynamic response rigid body

Fys = [(uY )(19.32 ft/sec2)( L / M*)](m)

Water
Rigid –> Fxr = (E3 )( 8.05 ft/sec2)

Mode shape conformal > Fxmsc = (E2 )( 19.32 ft/sec2)( L / M*)

Where E2 E3 are the values from column E from Tables 2E and 3E respectively.

Figure 7E shows the base stress distribution resulting from the application of the pseudo dynamic
loads. Note that the pseudo dynamic method predicts a base pressure distribution that is
everywhere within 7% of the exact solution obtained from a fully coupled reservoir model which
includes all vibrational modes.

Fig. 7E

Step 4 Evaluation of results

The pseudo dynamic analysis indicates that tensile stresses of up to 80 ksf, (550 psi) will be
produced by the seismic loading at the dam/foundation interface. Significant cracking is likely to
occur even under the most optimistic assumptions regarding material strengths. In addition, it is
likely that cracking will eventually progress over the entire base as cyclic acceleration continues
due the to amplification of stresses at the crack tip. The question of dam stability in not resolved
however, by considering stresses alone. It remains to be seen if the dam can retain the reservoir
in a static post earthquake condition given the cracking that this analysis has indicated. A post
earthquake static stability analysis should be performed assuming a cracked base.
 CHAPTER IV

EMBANKMENT DAMS

APRIL 1991
 Chapter IV

Embankment Dams

4-0 Contents

Title Page

4-1 Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

4-1.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

4-1.2 Depth of Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
4-1.3 References 4-3

4-2 Sources of Data and Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3

4-3 Review of Existing Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-4

4-4 Need for Supplemental Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-5

4-5 Evaluation of Embankment Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6

4-5.1 Embankment Zoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-8

4-5.2 Seepage Control Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-9
4-5.3 Deformation, Predicted or Recorded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
4-5.4 Erosion Control Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
4-5.5 Structural Stability Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-12
4-5.6 Potential for Liquefaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-13
4-5.7 Soil Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-14
4-5.8 Embankment Overtopping Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-14

4-6 Static Stability Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15

4-6.1 General .................................................... 4-15

4-6.2 Review Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-16
4-6.3 Conditions to be Investigated . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-17
4-6.4 Shear Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-18
4-6.4.1 Laboratory Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-19
4-6.4.2 Q-Unconsolidated Undrained Shear Strength . . . . . . . . . . . . . . . . . . . . . . . 4-20
4-6.4.3 R and R -Consolidated Undrained Shear Strength . . . . . . . . . . . . . . . . . . . 4-20

4-i
 4-0 Contents

Title Page

4-6.4.4 S-Consolidated Drained Shear Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20

4-6.5 Types of Stress Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20
4-6.6 Loading Conditions for Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20
4-6.6.1 End of Construction Loading Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-21
4-6.6.2 Sudden Drawdown Loading Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-21
4-6.6.3 Steady Seepage Loading Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-23
4-6.6.4 Partial Pool Loading Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-25
4-6.6.5 Earthquake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-25
4-6.7 Factors of Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-26
4-6.8 Static Stability Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27
4-6.8.1 Limit Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27

4-7 Seismic Stability Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-29

4-7.1 General Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-31

4-7.2 Modes of Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-31
4-7.3 Methods of Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-32

4-8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-37

4-9 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-41

Appendix 4-A Engineering Data

4-ii
 Chapter IV

Embankment Dams

4-1 Purpose and Scope

4-1.1 General

The guidelines presented in this chapter provide staff engineers with recommended procedures and
criteria to be used in reviewing and evaluating the safety of existing and proposed earth and rockfill
(embankment) dams. The review performed by staff engineers will be conducted to ensure that all
decisions, methods, and procedures performed by licensees/exemptees, or their consultants, are sound
regarding dam safety, and to ensure that the Commission's Dam Safety Program objectives as stated in
Part 12 of the Commission's Regulations are consistent with accepted. up-to-date state-of the-art
procedures (the term licensees also refers to applicants for license where appropriate).

The evaluation of safety of both new and existing embankment dams presents special and unique
problems. Existing dams may prove difficult to analyze especially in those instances where the dam was
designed before the development of modern design and construction technology or where adequate
records are not available. Even for a relatively new dam, where records are extensive. Evaluation can
be cumbersome for the following reasons: (a) various levels of completeness of records, (b) different
site conditions, (c) varying degrees of quality in design and construction, and (d) differing depth of
evaluation required for each dam. The objective set forth in this chapter is to provide systematic
procedures for performing staff evaluations.

4-1.2 Depth of Review

The review of existing dams will generally not be as detailed as the procedures involved in the design of
new dams. Some critical areas may require detailed review. Primarily, the review is intended to
evaluate procedures and methodology of design and analysis to ensure that safe and adequate
embankment dams were constructed. The licensee's/exemptee's or its consultant's investigations and
evaluations should be examined to determine if all areas of importance were considered and
appropriate design criteria have been used.

For proposed dams, the licensee will be required to submit a design report in accordance with the
Commission's Regulations. This report will be thoroughly examined to determine if all appropriate
design criteria have been met.

During the investigation and evaluation for both proposed and existing dams, important areas to
consider are as follows:
• The embankment must be safe against excessive overtopping by wave action especially during
pre-inflow design flood conditions.

• The slopes must be stable during all conditions of reservoir operations, including rapid
drawdown, if applicable.

• Seepage flow through the embankment, foundation, and abutments must be controlled so that
no internal erosion (piping) takes place and there is no sloughing in areas where seepage
emerges.

• The embankment must not overstress the foundation.

• Embankment slopes must be acceptably protected against erosion by wave action and from
gullying and scour against surface runoff.

• The embankment, foundation, abutments and reservoir rim must be stable and must not
develop unacceptable deformations under earth quake conditions. 1/ See Section 4-7 and
Reference 36 for seismic design.

Existing dams should be viewed in light of knowledge of studies and reports on similar dams of the
same vintage to gain an understanding of probable design and construction methods. For existing
dams, an independent analysis of the embankment stability or adequacy need not necessarily be
performed by staff. The data presented by the licensee should be reviewed to determine if they appear
reasonable and if the latest information has been considered. The criteria used by the licensee or its
consultant should be consistent with any changed conditions discovered during onsite examinations such
as loadings, seepage, increased pore pressures in the dam or the foundation, erosion, etc.

For proposed dams, an analysis of the stability and adequacy is required unless specifically exempted
by the Commission. The methods and procedures used in the evaluation of any embankment should be
consistent with the latest, accepted state-of-the-art methods and criteria, and with guidance contained
in this chapter of the Guidelines.

4-1.3 References

Criteria and methods of evaluation and analysis used in reviewing licensee's reports should be based on
criteria and procedures established in literature published by such agencies as the Corps of Engineers,
U. S. Bureau of Reclamation, or other recognized engineering references. Selected references are
listed in Section 4-8.

1/ Reference 1, pg. 192

4-2
4-2 Sources of Data and Information

To properly evaluate all information and data presented in the licensee's design report, various available
FERC reports should also be reviewed. Available reports include:

• Prelicense Inspection Reports of existing dams and/or Site Inspection Reports of proposed
damsites

• Operation Reports

• Construction Reports

• Independent Consultant's Safety Inspection Reports

One or more of the above listed reports should be available for licensed projects. If a license has not
previously been issued, the staff engineer performing the review should refer to the Prelicense
Inspection Report prepared by the staff engineer responsible for the project in the Regional Office.

For existing dams, additional data may be available from the facility owner, previous owners, state or
local agency if the facility is a publicly owned project, and from the state agency responsible for dam
safety, such as Department of Water Resources, Department of Environmental Resources, Division of
Dam Safety or Department of Natural Resources. Also, technical information may be available from
Corps of Engineers Phase I Inspection Reports of public or private entities having impounding
structures upstream or downstream of the facility.

For proposed dams, the source of information will generally be the licensee and/or its consultants and
engineers. For all proposed dams, the licensee will be required to provide staff with those data
necessary to evaluate whether the design of the structure is safe and adequate.

Data that may be available from the sources referenced should include:

• Logs of drill holes, test pits, and exploratory trenches

• Site geologic reports

• Site seismicity reports

• Materials exploration and testing reports

• Reservoir area-capacity curves, rim conditions, and drainage basin information

4-3
• Dambreak analyses and reports

• Construction reports

• Correspondence that may highlight design changes or problems

• Design drawings and specifications

• Design reports including assumptions used and the reasons therefore

• Inspection records

• Maintenance records

• Aerial photography

• Licensee's reports

• Construction photographs

• Concrete materials and mix design

• As built drawings

4-3 Review of Existing Data

Appendix 4-A is a listing of various engineering data related to the design, construction, and operation
of an embankment dam. Prior to review and analysis of existing data, this appendix may be useful in
organizing the data as discussed in the U.S. Bureau of Reclamation's "Safety Evaluation of Existing
Dams" (SEED) Manual." 2/

The engineer performing the review should examine all data to determine if problem areas have been
recognized and, if appropriate methods are proposed for correction. Additionally, the data should be
examined to determine if the source of any current conditions or problems, such as seepage, settlement,
cracking, etc., are evident from existing data. The methodologies and criteria used in the design should
be examined and compared to accepted state-of-the-art procedures and criteria.

2/ Reference 2

4-4
Advances in accepted state-of-the-art methodologies may require a reevaluation of the original design
or of these guidelines. The SEED Manual discusses in greater detail specific information to look for in
the reports and data that may be available.

4-4 Need for Supplemental Information

The objective of reviewing existing data is to be in a position to use as much information as is available
to evaluate the structural adequacy of existing or proposed embankment dams. Data and analyses
should be the prevalent basis for judgments on dam safety. If potentially hazardous conditions are
believed or determined to exist and the existing data are insufficient to resolve the problem, it may be
necessary to request supplemental investigations, analyses, or information to complete the evaluation.
The information could involve additional visual inspections, measurements, foundation exploration and
testing, materials testing, seismic information, hydrologic and hydraulic data. Conditions that may
require supplemental information are as follows:

• Significant cracking, settlement or sloughing of an existing embankment and the potential for
such in any proposed structure

• Uncontrolled seepage conditions through the embankment, the abutments, or at the toe area,
and the potential for such in any proposed structure

• Available data is not adequate to perform accepted state-of-the-art analytical methods that are
necessary

• Increase in settlement rate

• Increase in measured seepage

• Rise in internal seepage pressures

4-5 Evaluation of Embankment Dams

The two principal types of embankment dams are earth dams and rock-fill dams, depending on the
predominant fill material used.

a. Earth Dams - An earth dam is composed of suitable soils obtained from borrow areas or
required excavation which are then spread and compacted in layers by mechanical means.
Earth dams may be constructed as homogeneous or zoned dams. Zoned dams are generally
preferred since zoning permits the use of several different material types in the embankment that
may be available from borrow areas or required excavations. Homogeneous embankments are
usually not considered except when free-draining materials are not readily available.

4-5
 Some older dams have been placed by hydraulic means. These hydraulic fill dams frequently
contain large masses of loose to very loose soils in them because of the dumping and sluicing of
the soils during construction. Adequate soil data (e.g. SPT blow counts, gradation analysis,
phreatic surface, etc.) must be available to evaluate the liquefaction potential and stability of
these dams.

b. Rock-fill Dams 3/ - A rock-fill dam is an embankment composed largely of fragmented

rock with an impervious earth core. The core is separated from the shells by a series of
transition zones built of properly graded materials. The impervious core may be central or
inclined. The core, transition zones, filters, etc. should be evaluated as discussed in Section
4-5.1. In some cases an impervious upstream membrane of concrete, asphalt, or steel plate
may exist or be used in lieu of an impervious core.

Rock-fill zones are generally compacted in layers, 12 to 24 inches thick by 10 to 15 ton

steel-wheel vibratory rollers. Layer thicknesses of 6 inches up to 36 inches have been also
used and may be appropriate. The largest particle diameter generally should not exceed .9 of
the compacted layer thickness. Dumping rock-fill and sluicing with water, or dumping in water
is generally not acceptable for embankment dam construction today. However, the application
of some water before compaction, on dirty rock-fill, and on the zone adjacent to an impervious
upstream membrane, to achieve better compaction, is acceptable.

The structural safety of an embankment dam is dependent primarily on the absence of

excessive deformations under all conditions of environment and operation, the ability to safely
pass flood flows, and the control of seepage to prevent migration of materials and thus preclude
adverse effects on stability.

To properly evaluate the stability of an embankment dam, the following areas should be reviewed.

• Embankment zoning and cross section

• Seepage control measures and records

• Deformation, predicted or recorded

• Erosion control measures

• Structural stability analyses

3/ Reference 3, Chapter 1

4-6
 • Liquefaction potential

• Overtopping potential and the ability to resist overtopping

• Foundation and embankment material properties and strengths

• Erodibility indices

• Adequacy of freeboard

For existing dams, the review should also include summarizing the past behavior of the dam, with
attention given to any problem areas noted.

4-5.1 Embankment Zoning

For zoned embankments, the zoning geometry and properties of the materials placed in the zones
should be reviewed to determine: (1) the structural design, and (2) the types of internal features such as
chimney drains, blanket drains, toe drains, etc., that are proposed or were used to provide for and
maintain embankment stability. One should keep in mind that embankment zoning is also established
for economic reasons according to the availability of materials. 4/ The embankment zoning should
provide an adequate impervious zone, transition zones between the core and the shells, and seepage
control zones. Desirable characteristics that these zones should have or provide are as follows:

• In general, the width of the core at the base of cutoff should be equal to, or greater than, 25 per
cent of the maximum difference between the maximum reservoir and minimum tailwater
elevations. The minimum top width of the core should not be less than 10 feet. 5/ The
coefficient of permeability of the core material should preferably be 10-4 cm/sec or less. More
permeable core material may be acceptable if seepage is still adequately controlled and
appropriate factors of safety are still met. 6/

• Transition zones must meet accepted filter criteria, e.g. see References 1, 4, & 5, to protect the
adjacent zones from piping. The transition zones should be sufficiently wide to ensure that they

4/ Reference 1, Chapter 6

5/ Reference 3, pg. 5-3

6/ Reference 1, Chapter 6

4-7
 are continuous and constructable with a minimum of contamination at the contact. 7/8/ The
range of gradation of the transition zones should be limited to avoid segregation of materials
during placement.

• Seepage control features within the embankment should be sized adequately to contain all
seepage flows. The features should also be sufficiently pervious to ensure that all seepage will
be intercepted and controlled without excessive pressure head losses. 9/10/

• Zoning of an embankment that places the more pervious material on each side of the core zone
is preferable. This placement improves the stability of the embankment during rapid drawdown
conditions and keeps the downstream slope drained for greater effective weight. 11/

Homogeneous dams should also have seepage control features such as chimney drains, blanket drains,
etc. including a transition zone between the main embankment material and the drain.. Even if a
homogeneous embankment has no specific seepage control features, these embankments must have
adequate internal drainage capability to ensure against seepage outbreak on the downstream slopes or
abutments. Desirable characteristics listed above also apply to the features of this type of structure.
The homogeneous structure is generally more massive and usually has flatter slopes than a zoned
embankment of the same height. These characteristics compensate for a tendency toward a higher
phreatic line in the homogeneous embankment. They also tend to provide better slope stability during
rapid drawdown. 12/

4-5.2 Seepage Control Measures

All embankment dams are subject to some seepage passing through, under, and around them. 13/ If
uncontrolled, seepage may be detrimental to the stability of the structure as a result of excessive internal

7/ Reference 4, pg. 57, 606

8/ Reference 1, Chapter 6

9/ Ibid.

10/ Reference 3, pg. 5-3

11/ Reference 5, pg.7

12/ Reference 1, Chap. 6

13/ Reference 5, pg.1

4-8
pore water pressures or by piping. 14/ For existing dams, records or evidence that seepage flows have
removed any significant degree of fine grained material must be evaluated. Any such record requires
further field investigation.

Seepage should be effectively controlled to preclude structural damage or interference with normal
operations.

In the evaluation of seepage reduction or seepage control measures as they pertain to dam safety, one
should review and evaluate the following:

• Protective control measures such as relief wells, weighted graded filters, horizontal drains, or
chimney drains which prevent seepage forces from endangering the stability of the downstream
slope. 15/

• Filters and transition zones designed to prevent movement of soil particles that could clog
drains or result in piping. 16/17/

• Drainage blankets, chimney drains, and toe drains designed to ensure that they control and
safely discharge seepage for all conditions. The design of these features must also provide
sufficient flow capacity to safely control seepage through potential cracks in the embankment
impervious zone. 18/

• Contacts of seepage control features with the foundation, abutments, embedded structures,
etc., designed to prevent the occurrence of piping and/or hydrofracturing of embankment
and/or foundation materials. 19/ If conduits or pipes exist through the embankment, they
should be inspected to ensure that they are functional or have been properly sealed.

• Grouting, cut-off trenches, and impervious blankets.

14/ Reference 3, pg. 1-6

15/ Ibid., pg. 1-6, b

16/ Reference 4. pg. 57

17/ Reference 1, pg. 218

18/ Reference 3, pg. 1-6

19/ Reference 1

4-9
• Construction records for foundation shaping, treatment and grouting at the contact between the
impervious core and foundation.

• Measures such as compaction requirements, seepage collars, placement of special materials, or

other similar features to prevent internal erosion from seepage at the interface with concrete
structures. 20/21/ If seepage collars are present, special attention should be given to
compaction requirements around them. The use of seepage collars is not recommended in new
construction.

• For existing embankments, all seepage records compiled during the existence of the structure
should be reviewed for significant trends or abnormal changes. The causes of any
abnormalities should be determined as accurately as possible.

4-5.3 Deformation, Predicted or Recorded

The type, amount, and rate of deformation of an embankment, either vertical or horizontal movement,
must be estimated during the design stage and should be recorded during the operation of the structure.
For proposed embankments, the structure should generally be cambered to allow for the estimated
settlement during the life of the structure. For existing embankments, any evidence or records of
unusual settlement, cracking, or movement should be reviewed to determine whether these conditions
are detrimental to the continued safe operation of the structure. Field investigations may be required to
determine the causes of these abnormalities. These investigations may involve such items as surveying
the structure, installing movement detecting instruments, or excavating test pits for examination. 22/ The
embankment history, height, foundation conditions, hazard, etc. are factors to be considered in
determining field investigation needs.

As a result of deformation, cracking can develop through the impervious core section below the line of
saturation which may result in piping. Adequately sized and graded filter zones located downstream
from the impervious core can prevent piping. 23/ Corrective measures or instrumentation may be
needed if adequate filter zones do not exist or are not correctly located.

4-5.4 Erosion Control Measures

20/ Ibid.

21/ Reference 3, Chapter 2

22/ Reference 4, Chapter 12

23/ Reference 4, Chapter II

4-10
Upstream and downstream slopes, the toe area, groin areas of the abutments, approach and discharge
channels, and areas adjacent to concrete structures should be protected against excessive erosion from
wave action, surface runoff, and impinging currents. Inadequate erosion protection can result in slope
instability. 24/ Some common types of protection used are riprap, gabions, paving (concrete or
asphalt), and appropriate vegetative cover.

The slope and toe protection of all embankment dams should be reviewed to determine if the dam is
adequately protected against erosive forces. If the slope protection is being continually displaced,
heavier protection is required. Additionally, if embankment materials, consisting of silty and sandy soils,
are being moved into the slope protection, measures must be taken to correct this condition before
erosion becomes detrimental to the embankment. If riprap is required, a bedding layer must be
designed according to established filter criteria and placed under the riprap protection. 25/

4-5.5 Structural Stability Analyses

The evaluation of the stability of embankment dams shall be based on the available design information
for proposed structures and on design and construction information and records of performance for
existing embankments. The Corps of Engineers Guidelines for Safety Inspection of Dams 26/ can be
used as a guide in performing the review.

Stability studies and analyses for proposed embankments will be conducted during design in
accordance with methods discussed in Section 4-6.8. Quality control testing during construction will be
used to confirm that the design values are being achieved. For existing embankments, the initial stability
studies and analyses will normally be acceptable if they were performed by approved methodologies.
Additional stability analyses should be performed if initial design analyses do not exist or are
incomplete, if existing conditions have deteriorated, if hazard potential of the project has increased, if
the embankment has been subjected to loading conditions more severe than designed for, if existing
analyses are not in agreement with current accepted state-of-the-art methodologies, or if assumed
design parameters cannot be satisfactorily justified. Satisfactory behavior of the embankment under
loading conditions not expected to be exceeded during the life of the structure should generally be
indicative of satisfactory stability, provided adverse changes in the physical condition of the
embankment have not occurred. 27/

24/ Reference 3, Chapter 5

25/ Reference 1, Chapter 6

26/ Reference 6

27/ Reference 6, pg. 10

4-11
Evidence of any adverse changes which could affect the stability of an embankment may be obtained
from visual inspection and observation of available instrumentation data covering such items as changes
in pore water pressures, displacements, changes in loading conditions, seepage, etc. Review of
maintenance records and related information may also provide a reference to structural behavior data
for a particular structure. Should a review of project records indicate possible deficiencies in the
stability of an embankment, additional information may be required regarding the foundation and
embankment materials. The Corps of Engineers Guidelines for Safety Inspection of Dams 28/ and
other available literature 29/30/31/32/ 33/34/35/ can be referred to in establishing the information
necessary to determine the condition and material properties of the foundation and embankment.

4-5.6 Potential for Liquefaction

The phenomenon of liquefaction of loose saturated sands, gravels, or silts having a contractive structure
may occur when such materials are subjected to shear deformation with high pore water pressures
developing, resulting in a loss of resistance to deformation.

The potential for liquefaction in an embankment or its foundation must be evaluated on the basis of
empirical knowledge and engineering judgment supplemented by special laboratory tests when
necessary. Simplified methods for evaluating soil liquefaction potential are used by Seed and Idriss 36/
and Castro 37/ to relate blow counts values from standard penetration tests to safe, unsafe, and
marginal conditions. These empirical charts relate to observations of manifestations of increase of pore
pressure under level ground, such as sand boils. The empirical charts should be considered only as a

28/ Ibid.

29/ Reference 3

30/ Reference 5

31/ Reference 7

32/ Reference 8

33/ Reference 9

34/ Reference 10

35/ Reference 15

36/ Reference 12

37/Reference 13

4-12
guide for identifying zones within the dam and its foundation that may require further study. Further
discussion of liquefaction is presented in Section 4-7.

4-5.7 Soil Properties

Soil properties including strength and seepage parameters to be used as input data for stability analyses
should be realistic and representative of the range and variation that exist in the foundation, abutment,
and embankment materials. 38/ For information concerning the characteristics and strengths of
foundation and embankment soils and rock, refer to the procedures established in the Corps of
Engineers and U.S. Bureau of Reclamation Guidelines, 39/40/41/42/ and other literature. 43/44/45/46/
The selection of the proper input parameters and their correct use in a stability analysis are generally of
greater importance than the method of stability analysis used.

4-5.8 Embankment Overtopping Potential

All embankment dams, either proposed or existing, should be evaluated for overtopping potential under
the most extreme conditions expected for which the dam is determined to be a hazard to life or
property. Chapter 2 of these Guidelines discusses the Spillway Design Flood and provides freeboard
criteria. The maximum reservoir elevation determined for the design flood and expected wave runup
are conditions that should be considered. However, a less severe storm with lower reservoir elevation
but greater wave propagation may result in conditions that are more critical than those produced by the
design flood. In general, overtopping of an embankment is not acceptable. However, for existing dams
should minor or intermittent overtopping be determined to be a possibility, an evaluation of the
acceptability of this condition must be made based on such information as the characteristics of the

38/ Reference 14

39/ Reference 8

40/ Reference 9

41/ Reference 2

42/ Reference 15

43/Reference 16

44/Reference 4

45/Reference 17

46/Reference 34

4-13
flood hydrograph, embankment materials, prevailing wind direction, fetch, slope and crest protection,
hazard potential at that time, etc.

4-6 Static Stability Evaluation

4-6.1 General

As discussed in Section 4-1.2, a new, independent stability analysis by staff is not necessarily required
for a proposed or existing embankment. Spot checks of analyses may be required to verify that
application of the specific analytical approach is correct. The analysis and evaluation of the structural
adequacy of an embankment dam by the licensee and/or its consultant should be reviewed based on
information formulated by the licensee and information developed by the Regional Office staff from
various project inspections and data requests resulting from the licensing or inspection program. For
embankment dams, stability analyses should be examined to determine if the criteria used and loading
conditions analyzed are appropriate. This review should be based on the above information to
determine if the methods of analyses used are based on accepted state-of-the-art and that proper types
of failure surfaces have been analyzed (e.g., wedge, circular, or noncircular).

An independent stability analysis should be performed by staff if actual conditions differ from those
assumed in the licensee's analysis, if soil parameters are inconsistent with material types, if soil strength
parameters or pore water pressures are inconsistent with the method being used, or if the critical failure
surfaces do not appear to have been determined.

Staff has several stability programs for computers available. 47/ These programs may be used by staff
in reviewing the results of the licensee's analyses. It should, however, be understood that the results
obtained by these methods of analyses may not necessarily agree exactly with the licensee's results
based on another method; however, it will provide an indication as to the adequacy of the analysis
being reviewed. Staff is not limited to the use of these computer programs. Other accepted programs
may also be used. The staff should verify that the licensee has checked the analysis by hand
calculations for potential critical cases that have marginal factors of safety.

A brief discussion is included in this section of the Engineering Guidelines concerning some methods of
stability analysis and why the results obtained from each method may vary to some degree.
Additionally, references are listed in Section 4-8 that analyze the various methods of stability analyses in
detail. An historical development of methods of stability analyses is presented in Reference 16. 48/

47/ Reference 37 and 38

48/ Reference 16, pp. 323-326

4-14
4-6.2 Review Approach

Stability analyses should be reviewed to determine if input data appear appropriate based on a
knowledge of the embankment and foundation materials, on pore pressures in the embankment and its
foundation, or if the method of analysis chosen by the licensee is being used correctly. The literature
provides several publications, textbooks, and other sources of information that discuss in detail the
various methods of analyses available. Refer to Section 4-8 for references that can be used in obtaining
information for use in reviewing a particular method of stability analysis. 49/50/

A review of the stability analysis presented by the licensee shall include an evaluation and summary of
the data used in the analysis and an evaluation to determine if the critical conditions have been
investigated. The items to be evaluated include:

• Densities of soils

• Shear strength parameters

• Pore water pressures, estimated or existing

• Loading conditions

• Trial failure surfaces

• Method of analysis

The soil densities and shear strengths to be used for the various loading conditions investigated can be
evaluated by studying available laboratory test data and/or comparing data presented to that known for
similar materials based on past experiences and on data available from other dams consisting of similar
materials and construction methods.

Pore water pressures used in the analysis of the various loading conditions investigated should be
reviewed to determine if they are realistic based on available instrumentation data or estimates based on
such methods as those proposed by Casagrande 51/ and Carstens and May. 52/

49/ Reference 20

50/ Reference 26

51/ Reference 18

52/ Reference 19

4-15
When field explorations and laboratory testing are required to provide additional information concerning
the strength characteristics of the embankment materials, the sampling and laboratory testing
procedures should be reviewed to determine if they were adequately accomplished and are
representative of the conditions analyzed. Corps of Engineers and U.S. Bureau of Reclamation
technical guidelines concerning sampling and laboratory testing procedures can be used to complete this
review. 53/54/55/

4-6.3 Conditions to be Investigated

An embankment and its foundation are subject to shear stresses imposed by the weight of the
embankment and by pool fluctuations, seepage, or earthquake forces. Loading conditions vary from
the commencement of construction of the embankment until the time when the embankment has been
completed and has a full reservoir pool behind it. The range of loading conditions encompasses the
following conditions at various stages from construction through the operational stage of the completed
embankment:

• End of Construction

• Sudden drawdown

• Partial pool with steady seepage

• Steady seepage, normal pool

• Earthquake

• Appropriate flood surcharge pool

In all loading cases, the shear strength along any potential failure surface must be defined. The shear
strength available to resist failure along any particular failure surface depends on the loading conditions
applied.

4-6.4 Shear Strength

53/ Reference 9

54/ Reference 10

55/ Reference 15

4-16
Generally, the shear strengths of materials used in stability analyses are determined from laboratory
testing procedures which attempt to duplicate the various loading conditions to which the embankment
is expected to be subjected. 56/57/58/

From the time construction begins until the reservoir has been filled and a state of steady seepage has
been established, three different conditions of drainage will have occurred. Shear strength values used
in stability analyses for each condition of drainage are determined from laboratory tests on specimens of
the material which are compacted to the density and water content that simulates the conditions
anticipated in the dam. 59/ Tests corresponding to the three conditions of drainage are: 60/61/62/63/

• Q or unconsolidated-undrained (UU) test in which no initial consolidation is allowed under the

confining pressure and the water content is kept constant during shear.

• R or R consolidated-undrained (CU) tests in which consolidation is allowed under initial stress

conditions but in which the water content is kept constant during application of shearing
stresses. The R test is identical to the R test except that pore water pressure measurements
are made during the R test.

• S or consolidated-drained (CD) test in which consolidation is permitted under the initial stress
conditions and also for each increment of loading during shear.

4-6.4.1 Laboratory Testing

Testing procedures for determining the shear strengths of soils to be used in stability analyses, as well as
determining other engineering properties of soils, such as density, moisture content, consolidation,

56/ Reference 11

57/ Reference 16

58/ Reference 20

59/ Reference 16

60/ Reference 11

61/ Reference 10, pp. 328-338

62/ Reference 16

63/ Reference 20

4-17
permeability, gradation, etc., can be found in Corps of Engineers and U.S. Bureau of Reclamation
manuals. 64/65/ When reviewing-analyses of existing embankments the R, R , and S shear strength
parameters may be considered. In situations where unconsolidated soils may still exist for years after
construction a strength envelope between the Q and R may be appropriate in evaluating the stability of
the embankment dam. For proposed dams, shear strength parameters obtained from the Q test will
also be used.

4-6.4.2 Q - Unconsolidated-Undrained Shear Strength

The Q test is performed on specimens of impervious materials under simulated loading conditions
expected to occur during construction of embankments and results in an approximation of the
end-of-construction shear strength of the material.

4-6.4.3 R and R - Consolidated-Undrained Shear Strength

The R and R tests apply to conditions in which impervious or semipervious soils that have been fully
consolidated under one set of stresses are subjected to a stress change during the test without time for
consolidation to take place (soil is sheared without allowing dissipation of pore pressures).

4-6.4.4 S - Consolidated-Drained Shear Strength

The shear strength resulting from an S test is obtained by fully consolidating the soil specimen under the
applied confining stress and, when drainage is complete, applying shear stresses slowly enough to allow
full drainage to occur during the shearing process under each loading increment.

4-6.5 Types of Stress Analyses

In general there are two types of stress analyses that are used in the evaluation of existing and proposed
embankments. These are the total stress analysis and the effective stress analysis. The total stress
analysis is used in the design of embankments for loading conditions during construction, rapid
drawdown, and earthquake. The effective stress analysis should be used only in cases where the soils
behave drained and piezometer data are available. The cases that can be analyzed by the effective
stress method are partial pool and steady seepage.

4-6.6 Loading Conditions for Analysis

64/ Reference 10

65/ Reference 15

4-18
As outlined in Section 4-6.3, an embankment may be subjected to several loading conditions during its
life, ranging from construction to full pool operation. The loading conditions for which an embankment
must be analyzed are presented in detail in the following paragraphs.

4-19
4-6.6.1 End of Construction Loading Condition

At the end of construction, an embankment dam is still undergoing internal consolidation under its own
weight. For homogeneous dams or for zones in dams constructed from impervious materials, pore
water pressures will be built up during construction due to the inability of the impervious soil mass to
drain rapidly during consolidation.

The shear strength applicable to the impervious dam or zones within the dam during the construction
loading condition, is determined by the Q test conducted at field moisture contents and at field confining
stresses. The type of stress analysis that applies to this loading condition is the total stress analysis.
Because of the difficulty in estimating pore water pressures within the embankment during this stage of
loading, an effective stress analysis is not generally used. The analysis may, however, be conducted
using pore pressure responses in previously constructed dams that used materials, construction
methods, and construction schedules similar to those for the proposed dam. For pervious zones in the
embankment where drainage can occur rapidly, S strengths should be used in the analysis.

The end of construction analysis using shear strengths obtained from the Q test as representative of the
strength available in the impervious zones of an embankment, represents a lower limit of stability since
consolidation is progressing during the course of construction. If there are any serious questions about
stability during construction, the only positive method to determine the stability is to install piezometers
and evaluate the stability during construction.

4-6.6.2 Sudden Drawdown Loading Condition

In the sudden drawdown loading condition the structure has been subjected to a prolonged high pool
during which time a steady seepage condition has been established through the embankment. The soil
in the embankment below the phreatic surface is in a completely saturated state and is fully consolidated
under the weight of the overlying material. If subsequently the reservoir pool is drawn down faster than
pore water can escape, excess pore water pressures develop. Consequently, the reduced factor of
safety following a reservoir drawdown is due primarily to the existence of high residual pore water
pressures (drawdown pore water pressures) acting inside the upstream slope. 66/ The shear strength is
governed by the state of stress developed during consolidation under buoyant weight before
drawdown. 67/

The shear strength parameters required for an analysis under this loading condition are obtained from
the R test. An expression is then determined for relating consolidation pressure to the undrained shear

66/ Reference 16, pg. 370

67/ Reference 20, pg. 26

4-20
strength. Laboratory tests are performed under consolidated -undrained conditions, in which the
samples are consolidated under stresses corresponding to the conditions immediately preceding the
drawdown. 68/69/ If the material being investigated can drain so rapidly as to dissipate practically all
the excess pore water pressure as the drawdown progresses, 70/ the drained or S strength is the
strength used in the analysis. This type of analysis is referred to as a total stress analysis.

If an effective stress analysis is conducted, one method of measuring the effective stress parameters is
to perform consolidated-undrained triaxial tests on the soil with the measurement of pore pressure.
This type of test is referred to as an R test. The accuracy of this type of analysis rests in how well the
pore pressures can be estimated. If R tests are run on undisturbed samples retrieved from an existing
embankment, results of pore-pressure observations in the field can be used in determining pore
pressure coefficients to be used in the R testing procedure.

For further discussion on differences between total and effective stress analyses refer to References 4
and 16. Laboratory procedures for the R and R tests are discussed in Reference 10.

When conducting a sudden drawdown analysis the Corps of Engineers uses shear strength based on
the minimum of the combined R and S envelopes (figure 1). 71/ Shear strengths of free-draining
materials where dissipation of pore water pressure can proceed as the reservoir pool is drawn down
will be based on the S shear strength envelope of the material.

68/ Reference 4, pg. 258

69/ Reference 20, pp. 23-27

70/ Reference 4, pg. 258

71/ Reference 11

4-21
The unit weights of the soils to be used in analyzing the "before drawdown" condition will be the moist
weights above the line of saturation and submerged weights below. In analyzing the "after drawdown"
condition, moist unit weights will be used for the zone above the original phreatic surface, saturated unit
weights will be used within the drawdown zone, and submerged weights will be used below the level of
drawdown.

4-6.6.3 Steady Seepage Loading Condition

Steady seepage develops after a reservoir pool has been maintained at a particular elevation (e.g.,
maximum storage pool) for a sufficient length of time to establish a steady line of saturation through the
embankment. The seepage forces which develop in the steady state condition act in a downstream
direction. The condition of steady seepage throughout an embankment may be critical for downstream
slope stability. 72/ The seepage forces can be conservatively estimated by assuming a horizontal
phreatic line through the impervious zone at the elevation of the storage pool intersected by zones of
free-draining material. However, high abutment groundwater tables may cause the phreatic surface to
be higher in the vicinity of the abutments. In homogeneous impervious embankments, the line of
seepage can be estimated by various methods. 73/74/ Examples of estimating the line of seepage
through an embankment are given in Reference 5. If sufficient instrumentation is available, piezometer
levels in both the embankment and foundation can be reviewed and phreatic surfaces can be developed
accordingly.

72/ Reference 11, pg. 19

73/ Reference 18

74/ Reference 19

4-22
The pore water pressures which exist within an embankment at any given time are generated as the
result of two actions which can be considered independent for practical purposes: (1) gravity seepage
flow, and (2) changes in pore volume due to changes in the total stresses. 75/ The full reservoir stability
condition is nearly always analyzed using the effective stress method of analysis and the pore water
pressures acting are assumed to be those governed by gravity flow through the embankment. 76/

For design purposes, the Corps of Engineers generally uses the shear strength of impervious soils
corresponding to a strength envelope midway between the R and S test envelopes when the S strength
is greater than the R strength. The S envelope is used when the S strength is less than the R strength
(figure 2). The shear strength of freely draining cohesionless soils should be represented by the S test
envelopes. 77/

Th
e
unit weights to be used in the analysis will be the moist unit weight above the line of saturation and
submerged weights below this line.

In the case where a steady seepage condition exists in an embankment, an additional horizontal thrust
may be imposed by a surcharge pool up to the probable maximum pool elevation, generally not for a
prolonged period of time. Thus the impervious zone would not become saturated above the steady

75/ Reference 16

76/ Ibid.

77/ Reference 11, pg. 18

4-23
state condition established under normal reservoir conditions. The shear strengths to be used in the
stability analyses should be the same as those used in the steady seepage case with maximum storage
pool.

4-6.6.4 Partial Pool Loading Condition

The same information applies to the partial pool loading condition as to the steady seepage loading
condition except that the upstream slope is also analyzed. The upstream slope should be analyzed for
various pool elevations to determine which pool elevation creates the lowest factor of safety.

4-6.6.5 Earthquake

Evaluations of seismic effects for embankments located in areas of low or negligible seismicity (0.05g or
less) may be accomplished using the seismic coefficient in the pseudostatic method of analysis. Seismic
coefficients at least as large as shown in figures 6, 6a, 6b, and 6c of Reference 11 shall be employed as
applicable. 78/ The pseudostatic method assumes that the earthquake causes additional horizontal
forces in the direction of potential failure. This investigation need only be applied to those critical failure
surfaces found in analyzing loading conditions without earthquake loading. An analysis of earthquake
loading is seldom necessary in conjunction with sudden drawdown stability analysis. However, if
earthquake loading is possible during reservoir drawdown associated with a pumped storage project
where frequency of drawdown occurs on a daily cycle, earthquake effects during sudden drawdown
should be investigated. The selection of shear strengths to be used in the analysis are discussed in
Section 4-7.

For embankments located in areas of strong seismicity, a dynamic analysis of embankment stability
should be performed based on present state-of-the-art procedures. Refer to Corps of Engineers ER
1110-2-1806, "Earthquake Design and Analysis for Corps of Engineers Dams," for the earthquake
loading to be used in dynamic analyses and for guidance in performing seismic evaluations.

In general, an embankment dam should be capable of retaining the reservoir under conditions induced
by the maximum credible earthquake where failure would cause loss of life. The following investigations
should be accomplished for all proposed and existing embankments, with the exception that existing
confirmed "low" hazard potential dams may be exempted from these investigations.

• A seismic stability investigation using a dynamic analysis for proposed and existing dams
located in Seismic Zones 3 and 4 of Reference 33.

78/ Reference 11, change 1, dated 17 February 1982

4-24
• An evaluation of the liquefaction potential for all dams that have or will have liquefiable
materials either in the embankment or foundation.

• A geological and seismological review of existing dams in Seismic Zones 2, 3, and 4 of

Reference 33, to locate faults and ascertain the seismic history the of region around the dam
and reservoir.

• A seismic stability investigation of existing dams by dynamic analyses, regardless of the seismic
zone in which the dam is located where capable faults or recent earthquake epicenters are
discovered within a distance where an earthquake could cause significant structural damage.

4-6.7 Factors of Safety

The factor of safety includes a margin of safety to guard against ultimate failure, to avoid unacceptable
deformations, and to cover uncertainties associated with the measurement of soil properties or the
analysis used. 79/ In selecting a minimum acceptable factor of safety an evaluation should be made on
both the degree of conservatism with which assumptions were made in choosing soil strength
parameters and pore water pressures, and the influence of the method of analysis which is used.80/
The latter concerns the method of calculation in which side earth forces are considered and how
assumptions of directions of side earth forces affect stability analysis results.

A qualitative estimate of the factor of safety can be obtained by examining conditions of equilibrium
when incipient failure is postulated, and comparing the strength necessary to maintain limiting equilibrium
with the available strength of the soil. 81/

Therefore, the slope stability analysis of soils requires measurements of the shear strength and
computation of the shear stress. Appropriate minimum values of factors of safety to be used in the
stability analysis of a slope depend primarily on the measurement of strength. Factors influencing the
selection of minimum factors of safety include:

• Reliability of laboratory shear strength testing results

• Embankment height

• Storage capacity

79/ Reference 22, pg. 48

80/ Reference 16, pp. 368-371

81/ Reference 23

4-25
• Thoroughness of investigations

• Construction quality, construction control of embankment fills

• Judgment based on past experience

• Design conditions being analyzed

• Predictions of pore water pressures used in effective stress analyses

FERC minimum factors of safety are listed in Table 1. Final accepted factors of safety may depend
upon the degree of confidence in the engineering data available. In the final analysis, the consequences
of a failure with respect to human life. property damage, and impairment of project functions are
important considerations in establishing factors of safety for specific investigations.

4-6.8 Static Stability Analysis

Various analytical methods for evaluating the static stability of an embankment dam exist. The method
utilized in the licensee's analysis should be consistent with the anticipated mode of failure, dam cross
section, and soil test data.

4-6.8.1 Limit Equilibrium

Many methods of stability analyses exist that use the same general approach of employing the "limit
equilibrium method" of slope stability analysis. In this type of approach a qualitative estimate of factor
of safety can be obtained by examining the conditions of equilibrium when incipient failure is postulated,
and comparing the strength necessary to maintain limiting equilibrium with the available strength of the
soil. The factor of safety (F.S.) is thus defined as the ratio of the total shear strength available (s) on the
failure surface assumed to the total shear stress mobilized T along the failure surface to in order maintain
equilibrium. 82/

F.S. = s
J (1)

A state of limiting equilibrium exists when the shear strength mobilized is expressed as:

82/ Reference 24

4-26
 J = 1 (s) (2)
F.S.

F.S. is a factor of safety with respect to shear strength and 1/F.S. is the degree of mobilization of the
shear strength. It may be shown that the definition of F.S. given by equation (1) is equivalent to the one
used in the Ordinary Method of Slices, where the factor of safety is defined as the ratio of the resisting
moment to the over turning moment. 83/

The shear strength of a soil is expressed by the following expression:

s = c + F tan N

in which c and N represent the intercept and slope of the Mohr-Coulomb shear diagram and F
represents the normal stress on the shear surface. Thus, to determine the shear strength along a
potential failure surface the normal stress on the shear surface must be known. In analyzing both force
and moment conditions of equilibrium it becomes apparent that the problem of determining the
distribution of the normal stress on the shear surface is statically indeterminate, that is, there are more
unknowns than there are equations of equilibrium. 84/ An approach to this situation is to make
assumptions to reduce the number of unknowns in order that the problem is statically determinate, such
as is done in the "limit equilibrium" analysis procedure. Different procedures use different assumptions.
Some methods do not satisfy all conditions of equilibrium, such as moment equilibrium or vertical and
horizontal force equilibrium. Table 2 shows equilibrium conditions satisfied by various methods of
analysis.

Table 2

Equilibrium Conditions Satisfied

Individual
Overall Slice Vertical
Horizontal
Procedure Moment Moment Force Force

Ordinary Method of Slices Yes No No No

83/ Reference 25, pg. 784

84/ Reference 25

4-27
Bishop's Modified Method Yes No Yes No

Janbu's Generalized Yes Yes Yes Yes

Procedure of Slices

Spencer's Procedure Yes Yes Yes Yes

Morgenstern and Price Yes Yes Yes Yes

Studies have been performed to examine the accuracy of the equilibrium methods of slope stability
analysis. 85/86/

4-7 Seismic Stability Evaluationy

Various methods of analyses are available for evaluating the seismic stability of an earth dam. These
may be classified as:

• Pseudostatic methods

• Simplified procedures

• Dynamic analyses of embankment stability and deformability

Regardless of the method of analysis, the final evaluation of the seismic safety of the embankment
should be based on all pertinent factors involved in the investigation and not solely on the numerical
results of the analysis. 87/ References presented in the Corps of Engineers ER 1110-2-1806 can be
used in determining the scope of analysis required for properly assessing the seismic stability of an
embankment dam.

Table 1 (1)

Minimum
Factor of Slope to be
Loading Condition Safety Analyzed

85/ Reference 25, pp. 783-791

86/ Reference 26, pp. 475-498

87/ Reference 21

4-28
End of construction 1.3 upstream and
condition downstream

Sudden drawdown from >1.1* upstream

maximum pool

Sudden drawdown from 1.2* upstream

spillway crest or top of gates

Steady seepage with 1.5 upstream and

maximum storage pool downstream

Steady seepage with 1.4 downstream

surcharge pool

Earthquake (for steady >1.0 upstream and

seepage conditions with downstream
seismic loading using
the seismic coefficient
method)

Earthquake (for all <2 feet of Newmark-type

dynamic analyses using deformation along the
a deformation method) potential failure
plane (3)

The degree of safety against ultimate failure may be defined as:

Factor of Safety = strength

stress
Jf
or F.S.'
J

where F.S. = factor of safety

Jf = shear strength along the trial shear surface

J = equilibrium shear stress along the same trial shear surface

4-7.1 General Approach

4-29
• Analyses for earthquake loading should begin with simplified procedures and proceed to more
rigorous methods of analyses as a particular situation may warrant. Projects with well
compacted embankments and dense foundation soils located in Seismic Zones 1 or 2, 88/ and
all confirmed low hazard potential projects, may be evaluated by the pseudostatic method using
the seismic coefficient assigned to the seismic zone the project is in.

• In areas of severe and/or frequent seismic loading such as in seismic Zones 3 and 4 or where
foundation liquefaction potential exists, more rigorous dynamic methods of analyses will be
necessary. 89/ Site specific seismic evaluations will be performed for all projects not covered in
the paragraph above. These studies will identify earthquake source areas, the maximum
credible earthquake, and the distance from the site of each relevant source area. Potential for
fault rupture in the dam foundation and in the reservoir will be assessed. The modes of failure
that need to be investigated and the appropriate methodology are described in the following
subsections.

4-7.2 Modes of Failure:

a. Loss of Stability

The dam becomes unstable as a result of loss in strength in the dam or foundation - Liquefaction Slide -
typical examples: Lower San Fernando Dam and Ft. Peck Dam.

b. Excessive Deformations

The dam remains stable during and after the earthquake; however, deformations can accumulate. The
accumulated deformation needs to be estimated and evaluated with respect to its effects on the
likelihood of an uncontrolled release of water from the reservoir.

c. Other Mechanisms:

• Overtopping due to seiches

• Movements along a fault passing under the dam

• Landslides in abutments causing direct damage to the dam or due to wave in reservoir (Vaiont
dam)

88/ Reference 33

89/ Reference Ibid.

4-30
4-7.3 Methods of Analyses

a. Pseudostatic Analysis Procedures

For many years the standard method of evaluating the safety of embankment dams against sliding during
earthquakes has been the pseudostatic method of analysis. In using this approach no special
consideration has been given to the nature of the slope-forming or foundation materials and if the
computed factor of safety was larger than unity, it has generally been concluded that the seismic stability
question has been satisfactorily resolved. 90/ In Terzaghi's opinion, depending on the nature of the
slope-forming materials, a slope may remain stable if the factor of safety is less than unity or may fail if
the factor of safety has been found to be greater than unity based on the pseudostatic approach. 91/
This has been confirmed by embankment performances in recent earthquakes.

In general, therefore, earthquake analyses using the seismic coefficient method may be performed only
for structures proposed or existing in Seismic Zones 1 and 2. Seismic coefficients at least as large as
shown in the Corps of Engineer ER 1110-2-1806, should be employed in the analysis. 92/ In Zones 3
and 4 and in other zones where the pseudostatic method of analysis does not necessarily evaluate
appropriately the safety of an embankment, more sophisticated analyses should be performed.

b. Simplified Analysis

Following a detailed study of embankment dam performance during earthquakes, 93/ Seed observed
that the seismic resistance of dams constructed of clayey soils is much higher than that of embankments
constructed of saturated sands or other cohesionless soils. Thus for embankments which do not involve
saturated cohesionless soils, the pseudostatic method of analysis may still be used; alternatively,
methods for evaluating deformations in such dams have been developed. The computed displacements
can be compared to allowable displacements to determine the adequacy of the embankment (See 4-
7.3.d). Methods for evaluating deformations have been developed by Seed and Newmark. 94/95/

90/ Reference 27, pg. 220

91/ Ibid.

92/ Reference 33

93/ Reference 27, pg. 227

94/ Reference 28

95/ Reference 29

4-31
When embankments and/or their foundations are composed of loose sands, silts, or gravels, the
pseudostatic method may not be applicable. Therefore, analyses must be performed to determine (a) if
liquefaction potential exists and (b) whether such a liquefied condition can lead to failure or excessive
deformations of an embankment. There are various simplified methods available for evaluating soil
liquefaction potential 96/97/98/ based on empirical correlations between in situ behavior of sands and
standard penetration resistance. In addition, methods exist to assess the liquefaction potential of a soil
by determining whether the soil is contractive or dilative. 99/100/ Under cyclic loading of sufficient
magnitude and duration, a loose saturated sand, silt, or gravel having a contractive structure will
develop high pore water pressures, lose a large portion of its resistance to deformation, and flow.

c. Loss of Stability

The potential for loss of stability can be analyzed using a conventional stability analysis (Section 4-6)
incorporating minimum strength values corresponding to the degree to which pore water pressures are
generated in the soils by the earthquake shaking. Where the pore pressure ratio in the soil builds up to
a value close to 100%, the soil is considered to have developed a condition of liquefaction.

The determination of those zones where liquefaction or pore pressure build-up will occur must be made
using a dynamic analysis to determine the stresses and strains induced in the embankment by the
maximum anticipated earthquake motions and a knowledge of the pore pressure generation
characteristics of the soils comprising the embankment and its foundation. 101/ In general clayey soils
do not appear to develop increases in pore pressure due to earthquake shaking. However cohesion-
less soils are highly vulnerable to pore pressure development depending on their relative density and
other characteristics which should be considered in the seismic evaluation.

Once the degree of pore pressure build up has been evaluated, and zones of potential liquefaction
identified, soil may be assigned strength values for use in a stability analysis as follows:

96/ Reference 29

97/ Reference 30

98/ Reference 13

99/ Reference 31

100/ Reference 32

101/ Reference 27

4-32
 Soil Type Saturated Unsaturated

Impervious (clayey) Sup Sup

Pervious (sands) lower of Sus Sd-u

with ru = 100% or Sr and Sds

Pervious (sands) Sd-u Sd-u

with ru <100%

where: Sup = undrained peak strength

Sus = undrained steady-state strength

Sds = drained steady state strength

Sr = residual strength of liquefied soil

Sd-u = shear strength determined by effective stresses corresponding to induced pore

pressure.

For soils which develop a condition of ru = 100% the value of Sus or Sr is likely to control the stability
of the slope and appropriate values may be determined as follows:

• Based on empirical information from liquefaction slides in similar soils. There is a general
correlation between values of Sr and values of (Nl)60, the normalized standard penetration
resistance of sands and silty sands, presented in Reference 39. However, it is important to be
guided in the choice of values of Sr by empirical information from previous failures involving
soils similar to the ones under study.

• Based on laboratory tests using the procedures described in Reference 21. In interpreting the
test data it should be noted that values of Sus are very sensitive to void ratio changes and thus it
is necessary to apply corrections to laboratory measured strengths to obtain in-situ values and
for possible void ratio redistribution during the period of earthquake shaking, and to interpret
the results conservatively.

If the stability analysis indicates no potential for a liquefaction (flow) failure, then a deformation analysis
should be performed.

d. Deformations

4-33
Deformation computations are applicable only to dams not subject to a liquefaction (stability) failure.

Deformations can be assumed not to be a problem if the dam is well-built (densely compacted) and
peak accelerations are 0.2g or less. 102/ If this condition is not satisfied, a deformation analysis should
be made. This analysis can he made using the Newmark approach or a simplified Newmark
procedure. 103/ The deformation calculated along the failure plane by these methods should not
generally exceed 2 feet. Larger deformations may be acceptable depending on available freeboard,
ability of the embankment to heal cracks and other considerations.

The basic steps involved in conducting a deformation analysis are as follows:

• Determine the magnitude and source of the earthquake or earthquakes that should be
considered

• Determine the time-history or time histories of the ground motion associated with the
earthquake or earthquakes

• Determine the yield strength of the embankment and foundation materials

• Determine the dynamic response of embankment and foundation materials

• Predict the extent of structural deformations resulting from earthquake shaking

• If predicted deformations are not tolerable, explore design alternatives that would provide a
tolerable response

e. Other Methods of Analysis

Other failure mechanisms identified in Section 4-7.2 require special methods of analysis which would
need to be adapted or developed for the special circumstances of the project. Generally dams located
over faults that could potentially move during an earthquake should not be permitted unless filter
transition zones are provided which are at least twice the maximum potential fault movement both
horizontally and vertically.

4-8 References

102/ Reference 8

103/ Reference 4

4-34
1. U.S. Bureau of Reclamation, "Design of Small Dams," 1987.

2. U.S. Bureau of Reclamation, "Safety Evaluation of Existing Dams (SEED) Manual," 1983.

3. U.S. Army Corps of Engineers, "Earth and Rock-Fill Dams General Design and Construction
Considerations,"
EM 1110-2-2300, 10 May 1982.

4. Terzaghi, K. and Peck, R.B., "Soil Mechanics in Engineering Practice," 1967.

5. U.S. Army Corps of Engineers, "Seepage Control," EM 1110-2-1901, Feb. 1952.

6. U.S. Army Corps of Engineers, "National Program of Inspection of Dams," Vol. 1, App. D.,
Recommended Guidelines for Safety Inspections of Dams.

7. U.S. Army Corps of Engineers, "Settlement Analysis,"

EM 1110-2-1904, Jan. 1953.

8. U.S. Army Corps of Engineers, "Subsurface Investigations, Soils," EM 1110-2-1803, Mar.

1954.

9. U.S. Army Corps of Engineers, "Soil Sampling,"

EM 1110-2-1907, Mar. 31, 1972.

10. U.S. Army Corps of Engineers, "Laboratory Soils Testing,"

EM 1110-2-1906, Nov. 30, 1970.

11. U.S. Army Corps of Engineers, "Engineering and Design Stability of Earth and Rock-Fill
Dams," EM 1110-2-1902,
Apr. 1, 1970.

12. Seed, H.B. and Idriss, I.M., "Simplified Procedure for Evaluating Soil Liquefaction Potential,"
Journal of the Soil Mechanics and Foundations Division, ASCE, Vol. 97, No. SM9, Proc.
Paper 8371, Sept. 1971, pp. 1249-1273.

13. Castro G., "Liquefaction and Cyclic Mobility of Saturated Sands, Journal of the Geotechnical
Engineering Division, ASCE, Vol. 101, No. GT6, Proc. Paper 11388, June 1975, pp.
551-569.

14. Federal Coordinating Council for Science, Engineering and Technology, "Improving Federal
Dam Safety," July 1, 1978.

4-35
15. Bureau of Reclamation, "Earth Manual," 1974.

16. Sherard, J.J., Woodward, R.J., Gizienski, S.F., and Clevenger, W.A., "Earth and Earth-Rock
Dams," New York, John Wiley and Sons (1963).

17. Sowers, G.B. and Sowers, G.F., "Introductory Soil Mechanics and Foundations," New York,
The Macmillan Company (1970), 3rd Edition.

18. Casagrande, A., "Seepage Through Dams," Contributing to Soil Mechanics 1925-1940,
Boston Society of Civil Engineers, Boston, 1940.

19. Carstens, M.R. and May, G.D., "Graphs for Locating the Line of Seepage in an Earth Dam,"
Civil Engineering, ASCE, Aug. 1967.

20. Lowe, J. III, "Stability Analysis of Embankments," Journal of the Soil Mechanics and
Foundations Division, ASCE,
Vol. 93, No. SM 4, Proc. Paper 5305, July 1967, pp. 1-33.

21. Poulos, S. J.; et al, "Liquefaction Evaluation Procedure," Journal of Geotechnical Engineering,
ASCE, Vol. III, No. 6, June 1985.

22. Janbu, N., "Slope Stability Computations," Embankment Dam Engineering, Casagrande
Volume, 1973.

23. Bishop, A.W., "The Use of the Slip Circle in the Stability Analysis cf Slopes," Proceedings:
European Conference on Stability of Earth Slopes (Stockholm) and Geotechnique,
Vol. 5, No. 1, 1955, pp. 7-17.

24. Spencer, E., "A Method of Analysis of the Stability of Embankments Assuming Parallel
Inter-Slice Forces," Geotechnique, Vol. 17, No. 1, 1967,.pp. 11-26.

25. Wright, S.G., Kulhawy, F.H. and Duncan, J.M., "Accuracy of Equilibrium Slope Stability
Analysis," Journal of the Soil Mechanics and Foundations Division, ASCE, Vol. 99, SM 10,
Oct. 1973, pp. 783-791.

26. Whitman, R.V. and Bailey, W.A., "Use of Computers for Slope Stability Analysis," Journal of
the Soil Mechanics and Foundations Division, ASCE, Vol. 93, No. SM4, July 1974, pp.
475-498.

27. Seed, H.B., "Considerations in Earthquake-Resistant Design of Earth and Rockfill Dams,"
Geotechnique 29, No. 3, Proc. Paper 5327, 1979, pp. 215-263.

4-36
28. Makdisi, F.I. and Seed, H.B., "A Simplified Procedure for Estimating Earthquake-Induced
Deformations in Dams and Embankments," Report EERC-77/19, Earthquake Engineering
Research Center, University of California, Berkeley,
Aug. 1977.

29. Newmark, N.M., "Effects of Earthquakes on Dams and Embankments," Geotechnique 15, No.
2., pp. 139-160.

30. Seed, H.B., "Soil Liquefaction and Cyclic Mobility Evaluation for Level Ground During
Earthquakes," Journal of the Geotechnical Engineering Division, ASCE, Vol. 105,
No. GT2, Feb. 1979.

31. Castro, G., "Liquefaction of Sands," Harvard Soil Mechanics Series, No. 81, Jan. 1969.

32. Wisa, Anwar; Martin, E.Z.; Torrance, R.; and Garlanger, John E., "The Piezometer Probe,"
Proc. ASCE Conference on In Situ Measurement of Soil Properties, Vol. 1, June 1-4, 1975
(Source A).

33. U.S. Army Corps of Engineers, "Earthquake Design and Analysis For Corps of Engineers
Project," ER 1110-2-1806, dated 16 May 1983.

34. Leps, Thomas M., "Review of Shearing Strength of Rockfill," Journal of the Soil Mechanics and
Foundations Division, ASCE, vol. 96, SM4, July 1970, pp 1159-1170.

35. Poulos, S. J.; Robinsky, E. I.; and Keller, T. 0., "Liquefaction Resistance of Thickened
Tailings," Journal of Geotechnical Engineering, ASCE, Vol. III, No.12, 1985.

36. U.S. Bureau of Reclamation, Design Standard No. 13, Chapter 13, April 1987.

37. TSLOPE/TSTAB Computer Programs for Limit Equilibrium Slope Stability Analyses, TAGA
Engineering Software Services, 1984.

38. Wright, S. G., " U TEXAS 2 (University of Texas Analysis of Slopes -- version 2): A
Computer Program for Slope Stability Calculations". Geotechnical Engineering Software
GS86-1, Geotechnical Engineering Center, Civil Engineering Department, The University of
Texas at Austin, Feb 1986.

39. Seed, H. Bolton, "Design Problems In Soil Liquefaction", Journal of Geotechnical Engineering
Division, ASCE, Vol 113, No. 8, Aug 1987.

4-37
4-9 APPENDICES

4-38
 APPENDIX 4-A

ENGINEERING DATA
 APPENDIX 4-A

ENGINEERING DATA

This appendix lists engineering data which should be collected relating to the design, construction, and
operation of an embankment dam to be used in establishing the adequacy of embankment structures.

1. General Project Data

a. Construction dates.

b. Design of structures.

c. As-built drawings indicating plans, elevations, and sections of embankment and appurtenant
structures.

d. Information on any modifications made, if applicable, such as dam raising.

2. Geotechnical Data

a. Regional and site seismicity.

b. Foundation data and geological features including logs or borings, geological profiles and cross
sections, and reports of foundation treatment.

c. Engineering properties assigned to construction materials and the foundation for design
purposes including results of laboratory tests, field permeability tests, construction control tests,
and assumed design properties for materials.

3. Construction History

a. Construction procedures and methods used.

b. Properties and characteristics of construction materials.

c. How was quality control measured and maintained?

d. Final foundation and embankment reports.

4-A-1
4. Operation and Maintenance Records

a. Performance record to date based on instrumentation observations and surveillance reports.

b. Comparison of conditions to which embankment has been subjected, to those assumed in the
original design.

c. Remedial measures undertaken during life of project.

d. Known deficiencies and any work underway to correct deficiencies.

5. Inspection History

a. Operation inspections reports.

b. Safety inspections reports.

4-A-2
 CHAPTER V

GEOTECHNICAL INVESTIGATIONS AND STUDIES

APRIL 1991
 Chapter V

GEOTECHNICAL INVESTIGATIONS AND STUDIES

(Dams, Dam Sites or Appurtenant Structures)

5-0 Contents

Title Page

5-1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

5-2 Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

5-3 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2

5-4 Sources of Pre-existing Data

and Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2

5-5 Need for Supplemental Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4

5-6 Geological Investigation and Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5

5-7 Intensity of Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-5

5-7.1 Preliminary Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6

5-7.2 Initial Design Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6
5-7.3 Final Design Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-7

5-8 Methods of Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-7

5-8.1 Constructed Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-7

5-8.2 Unconstructed Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
5-8.2.1 Types of Exploration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
5-8.2.1.1 Geologic Reconnaissance and Mapping . . . . . . . . . . . . . . . . . . . . . . . 5-8
5-8.2.1.2 Borings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
5-8.2.1.3 Special Excavations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
5-8.2.1.4 Geophysical . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
5-8.2.2 Location of Explorations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9
5-8.2.3 Laboratory Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9
5-8.2.4 Field Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9

5-i
 5-0 Contents
(Cont.)

Title Page

5-9 Instrumentation and Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-10

5-9.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-10

5-9.2 Geotechnical Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-12

5-10 Special Geologic Hazard Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-13

5-10.1 Volcanic Hazards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-13

5-10.2 Seismic Hazards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-14

5-11 Submittal of Geotechnical Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-15

5-12 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-17

5-13 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-17

5-14 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-19

Appendix V-A Types of Exploration

Appendix V-B Field Permeability Tests
Appendix V-C FERC Parameters for Developing Stability Analyses

List of Tables

5-1 Causes of Deficient Behavior, Means of Detection . . . . . . . . . . . . . . . . . . . . . . . 5-11

5-2 Inventory of Geotechnical Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-13

5-ii
 Chapter V

GEOTECHNICAL INVESTIGATIONS AND STUDIES

(Dams, Dam Sites, or Appurtenant Structures)

5-1 GENERAL

An adequate assessment of site geologic and geotechnical conditions is one of the most important
aspects of a dam safety evaluation. Evaluation of the safety of either a new or an existing dam requires,
among other things, that its foundation has been adequately examined, explored, and investigated so
that it is as fully understood as possible. Foundation explorations should be directed towards obtaining
only such information as may be important to an evaluation of the dam. The exploration program
should identify the factors that critically affect the safe performance of the dam, and not develop
extraneous information. The following sections of this chapter briefly identify the principal methods
commonly employed in foundation investigations, and are intended to suggest approaches and scopes
of investigations which, when properly implemented, should comply with FERC requirements and
expectations. It is emphasized that, because of the almost infinite variety of geologic conditions from
site to site, it would be unreasonable and impractical to attempt to set forth in these Guidelines specific
investigation programs.

While this chapter is principally directed toward dams and dam sites, the types of investigations and
studies discussed are also applicable to other water retention structures and appurtenant structures of
hydropower projects.

5-2 PURPOSE AND SCOPE

The purpose of this chapter is to present guidelines for use by FERC staff for determining the
appropriateness and level of geotechnical investigations and studies for dams, water retention structures
and other appurtenances. Reports on such investigations and studies must be included in the technical
exhibits supporting; (a) applications for license, (b) final design reports, and (c) Part 12 reports (FERC
Regulations, 18 C.F.R. Part 12) by independent consultants, when such reports have not been
previously submitted or referenced.

The scope of this chapter is intended to outline the desirable quantity and quality of those investigations
required to support design or evaluation conclusions. It is fully recognized that the specific investigation
programs and studies for individual projects cannot realistically be standardized, and will vary widely
according to site conditions, type of dam, hazard classification, and design phase.

5-3 REFERENCES
Selection of field investigation procedures, and use of data evaluation procedures supporting geologic
or geotechnical reports, are acknowledged to be best guided by criteria and procedures available in
qualified literature, as well as by proven local practice. Reliable sources include publications and
manuals authored by the U.S. Corps of Engineers, U.S. Bureau of Reclamation, American Society of
Civil Engineers, U.S. Committee on Large Dams, and other widely recognized engineering
organizations. In each particular case, variations based on good judgment and experience are
encouraged and differing approaches of scope and detail between government and private practice are
realistically inevitable. Selected references are listed in Section 5-13

5-4 SOURCES OF PRE-EXISTING DATA AND INFORMATION

The geotechnical information and data presented in a licensee's or applicant's design report and in
Independent Consultant's (Part 12) report, should incorporate references, where applicable to available
FERC reports. The following reports may be useful:

• Operation Inspection Reports

• Construction Inspection Reports

• Independent Consultants' Safety Inspection Reports

• Other Inspection and/or Special Reports on existing dams or sites that are available

One or more of the above listed reports can be expected to be available for licensed projects. If a
license has not previously been issued, the FERC staff engineer performing the review may have
available the Prelicense Inspection Report prepared by the responsible FERC regional office.

For existing dams, geologic and geotechnical data may be available from the facility owner, previous
owners, state or local agencies if the facility is a publicly owned project, and from the state agency
responsible for dam safety. Also, geological information may be available from Corps of Engineers
Phase I Inspection Reports of public or private entities having impounding structures upstream or
downstream of the facility.

For proposed dams, the source of geotechnical information will generally be the licensee and/or the
applicant's consultants and engineers.

For all proposed dams, the applicant will be required to provide those data necessary to evaluate
whether the foundation of the proposed structure is adequate to safely construct the structure proposed
for that site.

Data to be made available should include, where applicable:

5-2
• Logs of borings, test pits, and exploratory trenches

• Site geologic maps and reports

• Site seismicity reports

• Site geophysical reports

• Materials exploration and testing reports

• Reservoir rim conditions

• Reports and papers published by Geological Societies and Departments in their Bulletins

• Correspondence that may highlight geological changes or problems in the foundation

• Design drawings and specifications for foundation excavation and support

• Local landslide history

• Inspection records

• Maintenance records

• Aerial photography

• Seepage history

• Licensee's reports

• Construction photographs

• Concrete materials and mix design

• As built drawings

• Instrumentation and monitoring data

5-5 NEED FOR SUPPLEMENTAL INFORMATION

5-3
The preferred approach for assessing the adequacy of dam foundations should be to minimize the use of
general assumptions as to foundation conditions and strength parameters. The objective of reviewing
existing data is self-evident. Site-specific information and data-based analyses should be the prevalent
basis for judgments on dam safety. If potentially hazardous foundation conditions are believed or
determined to exist, and the existing data are insufficient to resolve the problem, it will be necessary to
conduct supplemental investigations and analyses, or develop additional information to complete the
evaluation. Appendix V-C provides step-by-step procedures for developing the parameters needed in
a stability analysis. The supplemental information will usually involve additional explorations and testing,
materials testing and seismic information.

Typical conditions that would require additional foundation explorations for existing dams are suggested
below:

• Significant cracking, settlement or sloughing of dams or related nearby structures.

• Increase in settlement rates or indications of downstream movement.

• Uncontrolled seepage conditions under, through, and around foundations or abutments, and at
the toe area of any water retention structures.

• Sudden or steady increase in observed seepage.

• Credible foundation data is insufficient to support stability analysis.

• Unexplained high or rapid rise in piezometric pressures either in the foundation or abutment
material or within an embankment structure.

• Highly fractured, jointed rock.

• Rock formations that are known to be susceptible to seepage problems, solution activity or
erodible material.

• Rock formations that are conducive to weak seams or planes with low strength characteristics
and adverse orientation (i.e. downstream dip).

• Use of assumed high shear-strength parameters in a stability analysis that are not justified or
supported.

• Request by an owner to allow use of a reduced fact

or of safety from the criteria normally required in Table 2 of Chapter 3.

5-4
5-6 GEOLOGICAL INVESTIGATION AND REVIEW

Geological Investigations should be conducted for new projects and reviewed for existing structures to
determine the following:

• The general geologic setting of the area at and near the project.

• The geologic conditions related to selection of the site.

• The characteristics of the foundation soils and rocks.

• Any other geologic conditions that may influence design, construction, and long term operation.

• Seismicity of the area.

• The sources of construction material.

The extent of the investigations will depend on whether the project is proposed or existing and/or the
design and the complexity of the local geology. The methods used for the investigations are dependent
on the data that needs to be obtained to fully understand the foundation for both constructed and
proposed projects. These investigative methods also depend on the types and size of the structures
involved, and on the extent and quality of the information needed.

5-7 INTENSITY OF INVESTIGATIONS

The extent of required investigations should be dictated by hazard classification, nature of structures,
and quantity of data already available. Existing dams without adequate data should be evaluated as
carefully as proposed structures; not to do so is to be dangerously presumptive.

Geotechnical investigations for proposed sites should be generally divided into three separate phases to
minimize costs and for developing the necessary data at each stage of the approval, design, and
construction of a project:

• Preliminary Investigations (Adequate information to justify site selection and preliminary cost
estimates).

• Initial Design Investigations (Information necessary to obtain regulatory approvals, refine cost
estimates, and develop engineering and environmental data).

• Final Design Investigations (Information necessary for developing plans and specifications,
obtaining bids, and constructing the project).

5-5
For existing dams, the extent of data needed may be relatively limited, depending upon the adequacy of
existing data and construction documentation. Evaluation of an existing structure generally requires
detailed foundation data that may only be obtained by drilling, sampling, and testing that is concentrated
on specific site areas or problems. Such investigations, when needed, should be planned to provide the
engineer with information and data to answer questions on specific dam safety problems and to perform
dam safety analyses.

5-7.1 Preliminary Investigations (Adequate information to justify site selection and preliminary
cost estimates).

This investigation should provide a first general impression of the engineering and geological aspects of
the proposed site, and should determine if further study of the site is warranted. The field work
generally would include preliminary field geologic mapping, some preliminary hand auger holes for soil
and overburden sampling, a limited number of core holes into rock and possibly some preliminary
seismic refraction lines. This information would be used to answer questions raised by an office study.
The data would also be used to plan the type, location, and amount of explorations and laboratory
testing required for future, more detailed investigations.

5-7.2 Initial Design Investigations (Information necessary to obtain regulatory approvals,

refine cost estimates, and develop engineering and environmental data).

These investigations would be undertaken to provide more detailed information on foundation

characteristics on a particular site or several sites, and to provide data for preliminary considerations of
the design requirements and construction methods. This type of information is usually developed for
inclusion in the license application or in reports providing conceptual analyses of existing project
structures. This phase of field investigation should include surface and subsurface exploration and
sampling through borings, test pits, test trenches, material testing, geologic mapping, and additional
geophysical surveys to supplement drilling. Data developed from these activities should be used to
compare alternative sites, to analyze different types of structures that might serve the same purpose, and
to develop economic evaluations of the sites. An end product of this investigation usually is an
application for license, which includes a specifically identified site and appurtenant structures.

5-7.3 Final Design Investigations (Information necessary for developing plans and
specifications, obtaining bids and constructing projects).

These investigations would be primarily composed of detailed drilling, sampling, and testing
concentrated on specific features at the selected project site; and should be specifically planned to
provide the engineer with information that is necessary to design structures, estimate quantities,
determine rates of construction progress, develop cost estimates, prepare plans and specifications, and
obtain bids.

5-6
5-8 METHODS OF INVESTIGATIONS

The adequacy of the analysis of an engineered structure will normally be primarily dependent on the
extent of the information known about foundation conditions of the site and the physical properties of
the foundation materials. To evaluate these properties, the type and application of sampling methods is
important. There is no single sampling method or sampling device that will guarantee the recovery of
satisfactory samples in all materials, but the less disturbance to the sample, the more accurate the results
will be from testing that sample. Different devices and techniques have been developed for drilling and
sampling a wide variety of material types. Proper sampling is a combination of science and art.
Although many procedures have been standardized, the alteration and adaptation of techniques are
often dictated by specific investigation requirements.

5-8.1 Constructed Projects

For constructed projects, including evaluation of structures under Part 12 of the Commission's
Regulations, the methods of investigation generally consist of researching available information as
described in paragraph 5-4. However, if questions of project safety cannot be properly addressed by
the use of existing data, then additional site specific field investigations should be required. This work
will generally include explorations of the foundation, abutments, and structures themselves, using
equipment and methods discussed in paragraph 5-8.2 below.

5-8.2 Unconstructed Projects

5-8.2.1 Types of Exploration

The general types of explorations used to investigate potential project sites fall into four categories (1)
geologic reconnaissance and mapping (2) borings, (3) special excavations, and (4) geophysical
measurements. These types of exploration methods are discussed in U.S. Corps of Engineers EM
1110-1-1804, and EM 1110-2-1907 (See References, Paragraph 5-13), and briefly presented in
Appendix V-A.

5-8.2.1.1 Geologic Reconnaissance and Mapping

Geologic reconnaissance and mapping is crucial for understanding critical items influencing siting,
design, and construction. It gives the overall picture. The geologic map is frequently the only "as built"
drawing of the foundation conditions and is very useful in evaluating any stability, settlement or seepage
problems that may occur during the operation of the project.

5-8.2.1.2 Borings

5-7
Of the different types of explorations, borings are the most practical and accurate method of obtaining
sub-surface information. The most important aspect of the drilling procedures is the recovery of the
material penetrated. A boring with low recovery is of limited value, and will generally raise more
questions than it answers. (See Appendix V-A).

5-8.2.1.3 Special Excavations

Special excavations are defined as those openings made with machinery other than drill rigs for the
purpose of obtaining soil or rock samples, or conducting insitu testing. They consist of test pits, test
trenches, large diameter borings, tunnels, shafts, drifts or adits. (See Appendix V-A).

5-8.2.1.4 Geophysical

Geophysical explorations are an indirect method of obtaining generalized sub-surface geologic

information by using special instruments to make certain physical measurements. Geophysical
observations in themselves are not geologic facts, but are statistical and orderly measurements.
Geophysical explorations complement core drilling, test pits, or other direct methods of sub-surface
exploration and can provide a rapid evaluation of certain geologic conditions. However, their reliability
is only as good as their confirmation by conventional means of exploration. (See Appendix V-A).

5-8.2.2 Location of Explorations

Adequate information about foundation properties and characteristics is critical to a full understanding
of the adequacy of any design or in the evaluation of an existing structure. Therefore explorations
should be adequately distributed over the dam site, including abutments and dam foundation, and in
special cases at appurtenant structures, including penstocks, tunnels, spillways, intakes and outlets, at
the powerhouse site, (whether surface or subsurface) along the reservoir rim, and at the material
borrow sites.

5-8.2.3 Laboratory Tests

Laboratory testing of foundation material may include the performance of such routine tests as direct
shear, unconfined and triaxial compression, sliding friction, modulus of elasticity, tensile strength, natural
and dry density, moisture content, consolidation, Atterberg limits, grain-size analysis, and permeability.
Less frequently, and where unusual geological conditions exist, tests for foundation rebound, slaking,
collapsibility, dispersive characteristics, permeability, compaction, and determination of the mineral and
chemical composition of the rock and ground water may be required. In addition, where liquefaction
potential may need to be evaluated, dynamic laboratory tests such as cyclic direct shear and/or cyclic
triaxial compression tests may be appropriate. (See References, Paragraph 5-13).

5-8
Laboratory tests may be needed to provide information regarding the behavior of foundation rock
under the various construction conditions to which it will be subjected, such as, rebound due to removal
of load, application of load, scour, exposure to weather, wet-dry, and freeze-thaw cycles. Laboratory
tests are also necessary to establish the quality of construction materials such as concrete aggregate,
impervious material, rockfill, and riprap. The interpretation, evaluation and application of the test results
to the design of the structures is a highly important phase and depends to a great degree upon
experience and judgment in correlating and weighing the data accumulated in the test program.

5-8.2.4 Field Tests

Two of the most important field tests performed are permeability and grouting. Permeability tests can
be done either by pumping out or hydraulic pressure. These tests are discussed in the U.S. Army
Corps of Engineers EM 1110-2-3506. (See References, Paragraph 5-13), and briefly presented in
Appendix V-B.

The pumping-out test consists of bailing or pumping water from wells or boreholes and observing the
effect of this operation on the water level in these and nearby holes. The test usually is performed in
one or more of the exploratory borings.

The hydraulic pressure test consists of pumping water under pressure into an isolated zone in the rock
or overburden through a borehole and noting the quantity of water pumped at any given pressure.
Descriptions of pressure testing equipment and procedures are contained in EM 1110-2-3506.

Test grouting may be useful. It consists of performing experimental grouting operations on exploratory
boreholes to determine, during the design stage, the extent to which subsurface materials are groutable.

While the above field tests may be used to provide information on the foundation, additional field (in-
situ) tests for evaluating the physical characteristics of the rock mass as a whole may be justified as
follows: Test blasting, rock bolt pull-out tests (RTH 323-80, Reference 16), flat jacking tests (RTH
365-80 & References 11,15 & 16), Goodman jacking tests (Reference 18), chamber tests, and direct
shear strength (RTH 321-80, Reference 16).

5-9 INSTRUMENTATION AND MONITORING

5-9.1 General

Instrumentation is used to document preconstruction site conditions, to monitor the performance of a

structure both during and after construction, and to provide evidence that design criteria are satisfied.
In addition, analysis of seepage, leakage, pressure and movement data furnishes valuable background
information for use in future design work. The extent of instrumentation for monitoring potential
deficiencies at existing dams should take into account the hazard potential of a dam, or threat to life and

5-9
property should there be a failure or sudden release of water from the project. Instrumentation should
always be installed because it is a necessary supplement for monitoring the performance and long-term
structural integrity of all significant and high hazard potential projects. If visual distress in the structure
cannot be adequately monitored by existing instrumentation, additional supplemental instrumentation
properly located and installed, should be provided to ensure proper monitoring. Some level of
instrumentation may be useful on a case-by-case basis for low hazard potential projects. The extent of
internal distress in a dam cannot always be directly measured. However, diagnostic procedures are
available to assist in identifying problems. Table 5-1 can be used as an aid to assessing the causes of
deficient behavior, and the means of detecting such problems. A key factor relevant to these
procedures is effective instrumentation. A prudently located array of instrumentation specifically
tailored to the given dam might include some of the following: Piezometers, observation wells, strain
gages, extensometers, accelerographs, seismographs, inclinometers, tiltmeters, and alignment
monuments (survey networks). The basic layout should be installed early so that monitoring can start
during construction and continue after the project becomes operational. A comparison of operating
conditions with design assumptions will help to determine whether the structure is performing
satisfactorily. In selecting equipment, specific requirements are ruggedness, reliability over a long
period, and simplicity of construction, installation, observation, and evaluation.

5-10
5-11
Since there are many types of dams and many different site conditions, each dam will have its own
instrumentation needs that are specific to the type of structure. Some types of instruments are
recommended for most dams, while others are recommended only in special cases. The number of
instruments and frequency of observations should be carefully optimized. There should be reasonable
balance between the level of surveillance sought and economy. Otherwise, the project could be
burdened by unnecessary accumulations of data that may interfere with sensible problem analysis. As a
minimum, alignment monuments should be installed at all existing high hazard potential dams.

5-9.2 Geotechnical Instrumentation

Table 5-2 below provides an inventory of frequently used geotechnical instruments and the usual factors
measured and monitored. The following list of reference material covers details for installation of
instrumentation basic to monitoring dams:

• U. S. Army Corps of Engineers Manual EM 1100-2-1908, Part 1 of 2, 31 Aug 71:

Instrumentation of Earth and Rockfill Dams (Groundwater and Pore Pressure Observations);
Chapter 5, Installation, Maintenance of Piezometers and Observations.

• U. S. Army Corps of Engineers Manual, EM 1110-2-1908, Part 2 of 2, 19 November 76:

Instrumentation of Earth and Rockfill Dams (Earth Movement and Pressure Measuring
Devices); Chapter 2, Movement Devices for Embankments and Foundations.

• U. S. Army Corps of Engineers Manual, EM 1110-2-4300, 15 September 80:

Instrumentation of Concrete Structures, Chapter 3, Uplift and Leakage, and Chapter 4,
Plumbing Instruments and Tilt Measuring Devices.

• The National Research Council Book, 1983: Safety of Existing Dams, Evaluation and
Improvement, Chapter 10 Instrumentation.

• United States Bureau of Reclamation Manual, Embankment Instrumentation.

Adding instrumentation to existing dams often requires specialized equipment and drilling techniques for
boreholes. During the drilling of boreholes, samples of materials and logs should be obtained of the
borings. These data may be of significant value for subsequent data evaluation and predictions con-
cerning the ongoing safety of the structure under operating conditions. The drilling, sampling and
installation methods, and procedures must be specified and carefully monitored to prevent damage from
these activities to the structure and the foundation.

5-12
 Table 5-2 Inventory of Geotechnical Instruments

Instrument Phenomena Measured

Piezometer, closed & open Pore water and ground
systems, observation wells water measurements

Weirs, flow meters and flumes Seepage measurements

Temperature sensors Temperature measurements

of groundwater, seepage
and indirectly rock

Extensometer, inclinometers Internal deformation

measurement

Tiltmeters Rotational and tilting

measurement of embankment
and rock concrete dams
and their foundations

Survey equipment Deformation measurements

horizontal and vertical

Crack Monitor Crack movement measurements

5-10 SPECIAL GEOLOGIC HAZARD STUDIES

Special geological hazard studies may be required for a project that is to be located in an area likely to
be subject to active natural geologic forces, such as volcanic and seismic activity. Evaluation of
geologic hazards is extremely important in determining if a site is safe for the construction of a project.
Therefore, it is important that the owner of a project demonstrate that the site is geologically suitable
and that natural geologic conditions have been considered and evaluated in the project design.

5-10.1 Volcanic Hazards

Potential hazards from volcanoes are significant in the western states and Alaska. The U.S. Geological
Survey has published a map showing volcanic hazards in the United States (Mullineaux 1975); this map
shows no hazard east of New Mexico.

Volcanic hazards can be separated into two categories on the basis of distance from the volcano. For
sites close to volcanoes, there is a high potential hazard for hydro projects to be affected by debris
flows or mudflows, ash falls and lava flows. In addition, volcanic activity may also result in earthquakes
which may cause severe ground shaking. For projects distant from volcanoes, the most significant

5-13
hazard may be ash falls resulting from major explosions. Secondary effects may include volcanically
induced seiches and floods caused either by the melting of ice on the volcanic cone (Mt. St. Helens) or
failure of dams in valleys draining the slopes of the volcano. Debris or mud flows into nearby full
reservoirs could cause extensive flooding and endanger the structure.

5-10.2 Seismic Hazards

Many regions of the world are subject to potentially destructive earthquakes. The west coast of the
United States is best known for seismic activity. However there are other significant seismic areas in
the Central, Midwest, Northeast and Southeast sectors. In highly seismic regions, where earthquakes
occur frequently and are actively studied, the seismic hazard is better understood. In regions of low
seismic frequency and/or intensity, where destructive quakes occur infrequently, the actual danger to
structures may be much greater because the seismic hazard is often not well understood or is not given
the attention it deserves. Therefore, it is imperative that more than just a cursory evaluation be given to
data to be used in performing stability analysis for a project. The occurrence of an earthquake in the
vicinity of a dam can cause damage or even failure if earthquake loadings have not been given adequate
consideration during the design phase.

The study to define the seismic hazard at a proposed project site should include the following:

• Seismological investigations. Studies are made of the past occurrence of earthquakes

in the general region of the site, and on that basis estimates are made of the probability of future
earthquakes. In order for this approach to be valid, a sufficiently long seismic history must be
available.

• Geological investigations. In this investigation an evaluation is made of the tectonic

processes in the general site region. Faults in the general region are identified and the degree of
activity of the faults is estimated.

• Site soils and geology investigations. Investigations are made of geological

formations, soil deposits and rock at the site area to assess their possible behavior during
earthquake shaking, and how they might affect the ability of a structure to resist earthquakes.

• Liquefaction Investigations. Where liquefaction potential may need to be evaluated,

field
and laboratory tests may be appropriate. These tests can be an aid to determining the cyclic
stress levels which may cause liquefaction of a soil.

Judgment based on the information provided by the above investigations must then be used to establish
appropriate earthquake design criteria for the project. This study should be detailed in the project
geotechnical report.

5-14
5-11 SUBMITTAL OF GEOTECHNICAL REPORT

The geotechnical investigative data, including results of laboratory and field tests, should be submitted in
the Geotechnical Report which is part of the supporting design report for new projects. The
Geotechnical Report is required because the design of all structures depends on the strengths and
weaknesses of the material they are founded on or in. In the independent consultants' reports for
existing dams, and particularly for those that have a long, satisfactory service record, such data as may
be available should be presented in the summary of geological conditions pursuant to Part 12 of the
Commission's Regulations. All geotechnical reports should consist of a succinct presentation of those
geological conditions that contribute to characterization of the project site and determination of the
design of the various structures.

The Geotechnical Report required for new projects should present a comprehensive assessment and
description of the geology of the project. It should be limited, however, to an effective combination of
brief discussions, tabulated data, and geological illustrations to depict the conditions that are of
engineering significance. The information in the reports should focus on the following topics:

1. Significant and controlling topographic conditions.

2. Description of all aspects of bedrock and recent geology, including discussions of: (a)
composition and structure of the rock, (b) engineering description of soils and of their
relationship to the bedrock, (c) principal engineering properties of the rocks and soils as
determined by field and laboratory investigations, (d) geologic conditions that present special
engineering problems, (e) remedies proposed or used for the special problems, and (f) sources
and characteristics of construction materials.

The surface and subsurface investigations, laboratory tests, and geological illustrations in geotechnical
reports should be sufficiently comprehensive to supply reliable information on all geological conditions
that can influence the design, construction and cost of the project. Unless a separate seismological
report is required, the geotechnical report should review the earthquake history of the region.

Following is a list of illustrative material that should be included in the geotechnical report, and, for the
most part, included or referenced in the geological summary in an independent consultant's report.

• Project Location Map

• Reservoir-Geology map

• Plan of Explorations

• Logs of Exploratory Borings

5-15
• Laboratory Test Plots and Tabulations

• Site Geology Map

• Photographs

• Top of Rock Contour Map

• Geologic Structure Map

• Geologic Sections and Profiles

Geologic sections and profiles should show correlation of soil and rock units together with such
significant features as water levels, water losses, faults, shear zones, foliations, jointing, and solution
zones. The sections should also emphasize geologic structure and show depths of primary and
secondary weathering. All sections and profiles should be superimposed with outlines of the principal
structures and the depth of foundation excavation for existing or proposed structures.

Further, all geologic investigations and tests required for developing information on any of the following
construction items should be completed and included in the report:

• Excavation slopes

• Special rock excavation methods for structural excavations.

• Rock bolting for slope stabilization or tunnel rock support.

• Foundation treatment by grouting or dental concrete filling.

• Protection of weather-sensitive foundations, such as shale, pending their burial.

• Special design and construction problems related to elastic rebound in foundation materials.

5-12 SUMMARY

In summary, it must be remembered that no matter how well a project's structures have been
engineered, if the foundation conditions are not understood and taken into account, dam safety
problems could occur. This chapter is not intended to be a detailed text on engineering geology, drilling
techniques and program planning, sampling and laboratory testing procedures, or monitoring. It is
intended as a guide for the reviewing engineer to determine if the quantity and quality of the

5-16
investigations and studies performed support the design and/or conclusions presented. Section 5-13
provides a sampling of reference material for the above activities.

5-13 REFERENCES

1. U. S. Army Corps of Engineers, "Geotechnical Investigations," EM 1110-1-1804, Feb. 1984

2. U.S. Army Corps of Engineers, "Geophysical Exploration", EM 1110-1-1802, May 1979

3. U.S. Army Corps of Engineers, " Laboratory Soils Testing," EM 1110-2-1906, November
1970

4. U. S. Army Corps of Engineers, "Soil Sampling," EM 1110-2-1907, Mar 1972

5. U. S. Army Corps of Engineers, "Tunnels and Shifts in Rock," EM 1110-2-2901, Sept 1978

6. U. S. Army Corps of Engineers, "Grouting Technology," EM 1110-2-3506, Jan 1984

7. U. S. Army Corps of Engineers, "Instrumentation of Earth and Rockfill Dams," EM 1110-2-

1908, Part 1 of 2, Aug 1976

8. U. S. Army Corps of Engineers, "Instrumentation of Earth And Rockfill Dams," EM 1110-2-

1908, Part 2 of 2, Nov 1976

9. U. S. Army Corps of Engineers, "Instrumentation of Concrete Structures," EM 1110-2-4300,

Sept 1980

10. U. S. Bureau of Reclamation, Dept. of the Interior, "Dams and Public Safety," a Water
Resources Technical Publication, 1983

11. Wahlstron, Ernest E. "Tunneling in Rock," Elsevier Scientific Publishing Co. 1973

12. Alfred R. Golze, "Handbook of Dam Engineering," Van Norstrand Reinhold Company, 1977

13. National Research Council (U.S.) Committee on the Safety of Existing Dams," Safety of
Existing Dams - Evaluation and Improvements," National Academy Press, 1983

14. Peck, Ralph; Hanson, Walter E.; Thornburn, Thomas H.; "Foundation Engineering," 2nd
Edition, John Wiley & Sons, 1973

5-17
15. Goodman, Richard E., "Introduction to Rock Mechanics," John Wiley & Sons, 1980 (second
edition)

16. U.S. Army Engineering Waterway Experiment Station, "Rock Testing Handbook," 1980 with
updates

17. Electric Power Research Institute. "Guidelines for Drilling and Testing Core Samples at
Concrete Gravity Dams," EPRI GS-6365, May 1989

18. ICOLD Bulletin 21 and 22: USCOLD "General Considerations Applicable to Performance
Monitoring of Dams," December 1986

19. Sherard et Al "Earth and Earth Rock Dams: Engineering Problems of Design and
Construction," John Wiley and Sons, 1963

5-18
5-14 APPENDICES

5-19
 APPENDIX V-A

Types of Exploration
APPENDIX V-A

Types of Exploration

Exploration is needed to refine the overall picture obtained from geologic mapping, for collecting
samples to obtain specific design parameters, and to define specific features identified in geologic
mapping that require specific attention as they relate to design and construction. The general types of
field explorations used to investigate sites are:

Borings

Of the different types of explorations, borings are the most practical and accurate method of obtaining
sub-surface information. Borings are divided into soil and rock types or a combination of both. The
most important aspect of the drilling procedures is the recovery of the material penetrated. A boring
with low recovery is of little value, and will generally raise more questions than it answers.

In soil borings the two prevalent methods for obtaining samples are the earth auger and the 2-inch split
spoon drive sampler. For investigative sampling it is better to use a bucket auger as the continuous
flight auger can provide misleading results in many situations by mixing the material as it moves up the
flights. The 2-inch split spoon sampler is also used in performing the Standard Penetration Test (SPT).
While the samples obtained by these two methods are considered disturbed, they provide the basis for
determining if, where, and what types of "undisturbed" samples are needed. The primary method for
obtaining undisturbed soil samples is by the Shelby tube sampler. See Reference No. 4, EM 1110-2-
1907 (Soil Sampling) for detail on soil sampling.

For rock sampling, core borings are the most common method. Size of the rock core borings can
range in diameter from 1 1/2 inches (EX) to 7 3/4 inches (6 x 7 3/4 inches), but the most common size
used for exploratory work is 3 inches (NW). The NW size drilling usually produces good core
recovery. Large diameter holes and special drilling equipment and methods may be justified in some
types of rocks if better recovery and/or identification, sampling and testing of material are required, and
in extracting the concrete/rock interface intact. The equipment and procedures for drilling and sampling
are given in detail in reference No. 1. (Geotechnical Investigations), EM 1110-1-1804.

Equally important as obtaining as complete a sample as possible, with little disturbance, is the need to
maintain the natural moisture content of the rock or soil sample. Material that may possibly be used for
laboratory testing should be wrapped and/or waxed immediately upon removal from the sampler to
preserve the natural moisture content.

5-A-1
 Standard Sizes for Core Drills
Diamond Core Drill Manufacturers Assoication (DCDMA)

Designation Core Size (in) Hole Size (in)

EX/EW 0.845 1.485

AX/AW 1.185 1.890
BX/BW 1.655 2.360
NX/NW 2.155 2.980
AQ* 1.062 1.890
BQ* 1.432 2.360
NQ* 1.875 2.980
HQ* 2.500 3.782
PQ* 2.344 4.828
2-3/8 X 3-7/8** 2.375 3.875
4 X 5-1/2** 4.000 5.500
6 X 7-3/4** 6.000 7.750

* Wire line series

** Large diameter series

Special Excavations

Special excavations are defined as those openings made with machinery other than drill rigs for the
purpose of obtaining soil or rock samples, or conducting in-situ testing. They consist of test pits, test
trenches, large diameter borings, tunnels, shafts, drifts or adits. These excavations expose large areas
of subsurface material thus permitting examination of in-situ subsurface conditions, recovery of large
undisturbed samples, in-situ testing, installation of instrumentation, and evaluation of abnormalities.
Some of these excavations require sheeting and shoring to protect the investigators as required by
OSHA.

Test pits are openings excavated vertically from the ground surface to expose the sub-surface material
for in-situ examination. Excavation is generally performed by backhoe, clam bucket or by hand. They
are also used as a means of obtaining undisturbed samples of soil materials. Test pits are most often
used in connection with soils exploration and testing. Test pits may also be used to study the character
of the overburden-bedrock contact, and the position, characteristic, and condition of the bedrock
surface and strata.

5-A-2
Test trenches are similar functionally to test pits, except that they are usually limited to relatively shallow
depths and extend over a greater length. Excavation of trenches is usually done by bulldozer. Trenches
are particularly useful for continuous exploration, examination and sampling of soils foundations; and for
examining and correlating bedrock surface conditions that cannot be defined as accurately by
conventional drilling and sampling methods. Combined with test pits, trenches are a more reliable
method of determining the occurrence, composition, distribution, structure, and stability of
unsatisfactory materials in deep alluvial and residual soil foundations for high dams.

Tunnels and drifts are nearly horizontal underground openings and passages which are excavated by
standard mining methods. They vary in size and shape depending on the purpose and type of material
being tunneled. The principal function of tunnels as an exploratory device is to permit detailed
examination of the composition and geometry of rock structures such as joints, fractures, faults, shear
zones, and solution channels where these conditions affect foundation stability, excavation, and
treatment. Although a slower investigatory approach, excavation of exploration tunnels should be used
when other methods do not supply adequate information. Geologic logging of the exploration tunnels
should be done concurrently with the excavation when possible. Sampling and insitu rock testing may
also be done during the excavation.

Geophysical

Geophysical explorations are an indirect method of obtaining generalized sub-surface geologic

information by using special instruments to make certain physical measurements. Geophysical
observations in themselves are not geologic facts, but are statistical and orderly measurements. The
geologic information is obtained indirectly through analysis or interpretation of these measurements.
The technology for such investigations has improved in recent years; therefore, geophysical testing has
gained acceptance among design engineers and geologists as accuracy has improved. However, since
these results are not subject to direct visual verification, geophysical exploration requires boreholes or
other direct geological exploration for references and control of measurements. Geophysical
explorations complement core drilling, test pits, or other direct methods of sub-surface exploration and
can provide a rapid evaluation of certain geologic conditions. The cost of geophysical explorations is
generally low compared with the cost of core borings or test pits, and considerable savings may often
be affected by judicious use of this exploration method in conjunction with other methods.

5-A-3
 APPENDIX V-B

Field Permeability and

Groutability Tests
APPENDIX V-B

Field Permeability and Groutability

Field tests to measure the permeability/groutability of the foundation material and the approximate yield
of water are:

1. Pumping-Out Test

The pumping-out test consists of bailing or pumping water from wells or boreholes and observing the
effect of this operation on the water level in these and nearby holes. The test usually is performed in
one or more of the exploratory borings. Records are kept of the water levels before pumping, the time
at which pumping is started, rate of pumping, amount of drawdown in the borehole, the time of each
drawdown measurement, and the rate of water level rise after pumping has stopped.

2. Pumping-In Test (Hydraulic Pressure)

The hydraulic pressure test consists of pumping water under pressure into an isolated zone in rock or
alluvium in a borehole and noting the quantity of water pumped at any given pressure. In general, water
pressure should be limited to 1 pound per square inch per foot of depth in rock and to 0.5 pound per
square inch per foot of depth of overburden to avoid damaging the foundation. The information
obtained is used in appraising the leakage potential of the foundation, and in estimating grouting
requirements for reducing seepage and controlling uplift pressures.

3. Field Grouting Test

Test grouting consists of performing experimental grouting operations on exploratory boreholes to

determine, during the design stage, the extent to which subsurface materials are groutable. Grout is
pumped into boreholes to full-depth or in sections isolated by packers. Detailed records are kept of
the grouting operations. The results are evaluated on the basis of the consistencies and quantities of
grout injected in relation to the pressures used, rate of injection, and time. Test grouting affords a
method of estimating grout requirements, and in many instances gives more reliable data than pressure
testing. Direct correlation between permeability and groutability has not been reliably established.

5-B-1
5-B-2
 APPENDIX V-C

FERC Parameters for Developing

Stability Analyses
APPENDIX V-C

FERC Parameters for Developing Stability Analysis

The necessary steps required to adequately address the parameters required in developing stability
analyses are presented for constructed and unconstructed dams. Determining potential modes of failure
is common to both.

Constructed Projects Shear Strength Parameters

Determine the need for additional field studies:

• Determine if condition of dam, as specified by Section 5-5, warrants supplemental foundation

information.

• Perform preliminary stability analyses with the friction angle (N) parameter selected from the
available records or from the literature and find the cohesion (c) values necessary to meet
FERC safety factors for all loading conditions. Use effective uplift/pore pressure information if
available, otherwise use full uplift pressure or total pore pressure, as mentioned in the
Guidelines.

• Perform a sensitivity analysis for N using the most conservative value for c that resulted from the
previous analysis. If the condition of the dam is acceptable and the most conservative c and N
values are within a defensible range, no field investigations would be necessary.

Plan for investigations:

• Identify the potentially weak shear planes by using the existing information. If no information is
available, use engineering judgement to define reasonable criteria for selecting such planes.

• Identify the types of tests and their scale (lab or field) as specified by section 5-8.2.1.

• Select the location of explorations as indicated by section 5-8.2.2.

• Prepare the specifications for the investigations.

Exploration and Testing

Dams on Bedrock Foundations

5-C-1
• Recover intact cores of concrete, bedrock and at the dam/foundation interface for classification
and testing. It is recommended that three borings be taken as a minimum at the center or deep
section of the dam and one at each abutment. For long structures, more borings may be
advisable at 100 to 200 feet intervals. More borings should be taken as necessary to document
any anticipated foundation problems (i.e. presence of clay seams).

• Drilling must be core type and the use of double or triple split barrels is highly recommended.
The diameter of the samples should be no less than that obtained from NW size equipment.
Preferably, for concrete the diameter should be approximately six inches depending on
maximum size of aggregate.

• Core samples should be prepared for performing direct and triaxial shear tests on intact rock
cores and cores containing potential weaker failure planes. These tests will provide
approximate values of cohesion and angle of internal friction to be used in sliding stability
analyses of the dam. Test results of intact samples will give upper bound shear strength values
while tests on smooth surfaces give lower bound results.

• Unconfined compressive tests should be performed on intact rock samples to determine

unconfined compressive strength of the bedrock. This data will be used to index rock
characteristics and determine if the foundation has adequate bearing capacity to resist the loads
imposed by the dam under all credible loading conditions.

Dams on Soil Foundations

• Recover a series of "undisturbed" soil samples from all zones of the embankment and the
foundation, for index, classification and engineering property tests. It is recommended that test
borings be generally located as above with the exception that all zones of material must be
sampled.

• The types of index and classification tests for the soil will be selected based on the nature of the
material. Cohesive soils should be tested for liquid and plastic limits. Clean sands and gravels
are tested for gradation limits. Based on these tests, the need for further investigations for
liquefaction potential should be assessed.

• Undisturbed soil samples should be tested for shear strength by either triaxial compression or
direct shear apparatus. The tests should be based on the proposed loading conditions and
drainage condition under which the dam will be subjected.

Constructed Projects Uplift Pressure Parameters

Determine the need for additional information

5-C-2
• If the preliminary analysis using full uplift and/or pore pressure indicated in section 5-13.1.1 will
result in defensible shear strength parameters, no additional information will be necessary;
otherwise:

• Perform a sensitivity analysis to find the range of a reduction in the uplift pressure that would
warrant defensible shear strength parameters. If the uplift reduction would have results within
an acceptable range, engineering judgement should be used to determine if the collection of
additional information would be beneficial.

Plan for collecting additional information

• Locate the measurement sections as mentioned in section 5-8.2.2.

• Review the instrumentation discussion in section 5-9 and if so determined, install piezometers in
the embankment at the interface and within the foundation bedrock or soil foundation to monitor
pore pressures and define the phreatic surface in the embankments and to measure uplift
pressures at the base of or below the dam at a possible foundation sliding plane. Readings
should be taken over a period of several months, at periods of high flows if possible, to assess
the functioning of internal drainage systems.

Unconstructed Projects Shear Strength Parameters

Field Studies

• Perform geological investigation as mentioned in section 5-6 to determine design criteria for the
required investigations.

• Plan preliminary investigations (section 5-7.1), initial design investigations (section 5-7.2) and
final design investigations (section 5-7.3) using the appropriate types of exploration (section 5-
8.2.1), locations of exploration (section 5-8.2.2) and types of laboratory tests (section 5-
8.2.3).

Proposed Dams on Bedrock Foundations

• Inspect and map the bedrock foundation noting joints or any other structural discontinuities that
may require remedial measures.

• Recover a series of intact rock samples of the foundation to determine the orientation of
bedding planes and for index and classification testing.

5-C-3
• Perform direct or triaxial shear tests on the intact core samples to determine appropriate design
values for cohesion and angles of internal friction. This data will be used to assess the sliding
stability of the dam along potential failure planes within the foundation rock below the dam.

• A series of unconfined compressive tests should also be made to determine the bearing capacity
of the bedrock below the dam, occasionally the bearing capacity of the foundation bedrock will
control the design of the proposed dam.

Proposed Dams on Soil Foundations

• If the situation requires more detailed analysis, a field reconnaissance to photograph and map all
major foundation features should be done under the direction of an experienced geologist
and/or geotechnical engineer.

• A series of borings should be taken to define the geologic stratigraphy and to obtain
"undisturbed" samples for index and in-situ tests.

• If feasible, test pits or trenches should be excavated to determine in situ subsurface conditions
and recover larger soil samples for testing.

• In-situ testing is performed to determine appropriate shear strength parameters. Available test
procedures include the Standard Penetration Test, direct shear, field vane shear, dutch cone.

• In-situ testing to determine stress conditions is important to assess the effect on the foundation
from construction of the proposed dam. Available test procedures include overcoring, flatjack
and hydrofracture.

• An evaluation of deformation characteristics of the foundation and consolidation due to imposed

loadings of the dam under all credible loading conditions should be made to determine the need
for any special remedial measures.

Unconstructed Projects Uplift Pressure

Instrumentation and Monitoring

• Determine the type and location of the instrumentation (section 5-9.2).

• Design the monitoring program.

Reporting and Controls

5-C-4
• Perform Quality Assurance/Quality Control of the specified investigations and specified
instrumentation.

• Prepare the Geotechnical Report (section 5-11), including the distribution of the foundation
shear parameters.

• Summarize the monitoring program in the final Geotechnical Report.

5-C-5
 CHAPTER VI

EMERGENCY ACTION PLANS

NOVEMBER 1998
 CHAPTER VI

EMERGENCY ACTION PLANS

6-0 Contents

Title Page

6-1 PURPOSE AND SCOPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

6-1.1 Purpose ..................................................... 6-1

6-1.2 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1
6-1.3 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6-1.4 Changes to the Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3

6-2 ANALYTICAL REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4

6-2.1 General ..................................................... 6-4

6-2.2 Dambreak Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
6-2.3 Inundation Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6
6-2.4 Exemption Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-9

6-3 EAP PREPARATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10

6-3.1 General .................................................... 6-10

6-3.2 Basic Considerations for Preparing Emergency Action Plans . . . . . . . . . . . . . . . . 6-10
6-3.2.1 Purpose. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10
6-3.2.2 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-11
6-3.2.3 The Six Basic Elements of an EAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
6-3.2.4 Coordination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13
6-3.2.5 Evacuation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-13
6-3.2.6 Emergency Duration, Security, Termination, and Follow-up . . . . . . . . . . . . 6-13
6-3.2.7 Maintaining an Emergency Action Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-14
6-3.3 EAP Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-15
6-3.3.1 The Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-15
6-3.3.2 Format Items Defined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-18

6-i NOVEMBER 1998

 6-0 Contents

Title Page

6-4 EAP EXERCISES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-42

6-4.1 General .................................................... 6-42

6-4.2 Annual Drill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-42
6-4.3 Comprehensive Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-43
6-4.3.1 Tabletop Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-46
6-4.3.2 Functional Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-50
6-4.3.3 Full Scale Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-52
6-4.4 Role of Licensee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-54
6-4.5 FERC Goals and Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-55
6-4.6 Results from an Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-57
6-4.7 Availability of Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-58
6-4.8 Licensee Initiative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-59

6-5 RADIOLOGICAL EMERGENCY RESPONSE PLAN . . . . . . . . . . . . . . . . . . . 6-59

6-6 EAP AT A GOVERNMENT DAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-61

6-7 APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-62

Appendix VI-A Dambreak Breach Parameters

Appendix VI-B FEMA "Exercise Design Course" Manuals
Appendix VI-C Sample Title Page, Approval Page and Table of Contents
Appendix VI-D Format For Critique of EAP Exercise

6-ii NOVEMBER 1998

 Chapter VI

EMERGENCY ACTION PLANS

6-1 PURPOSE AND SCOPE

6-1.1 Purpose

As required by Section 12.20 (a) of the Commission's Regulations, every applicant for license or
licensee/exemptee must develop and file an EAP with the Regional Director unless granted a written
exemption in accordance with Section 12.21 (a) of the Regulations.

The purpose of this chapter of the Federal Energy Regulatory Commission's Engineering Guidelines is
three-fold. First, recommended procedures and criteria are provided for performing or reviewing the
analytical studies required for an EAP. Second, a format is provided for preparing an EAP document
in a manner consistent with Federal guidance. Third, information is provided on the basic requirements
for conducting comprehensive exercises for testing an Emergency Action Plan (EAP). These
Guidelines provide a basis to develop and maintain an effective EAP. Since every EAP is unique,
individual project features and downstream effects will govern the content of the plan. The goal is to
develop the best EAP possible.

6-1.2 Background

The "Guidelines for Preparation of Emergency Action Plans" were established in November 1979. The
Guidelines were subsequently included as the Appendix to Order No. 122 of the Commission's
Regulations, issued January 21, 1981. Then, in accordance with the provisions of Section 12.22 (a) (1)
of the Commission's regulations, which states that "an emergency action plan must conform with the
Guidelines established, and from time to time revised, ...", the guidelines were revised on April 5,1985,
to provide more specific comprehensive guidance in the development of an EAP. Although the revised
Guidelines established a specific format to assist in preparing an effective, workable EAP, it was not
mandatory at that time that EAPs on file prior to April 5, 1985, comply with this format.

The EAP Guidelines were further revised on February 22, 1988, to provide a more workable EAP that
included a notification flowchart located at the front of the EAP and more clear, concise, easy-to-read
inundation maps depicting the dam break scenario. In addition, a need existed for a periodic reprinting
and redistribution of the EAP to improve this aspect of its dam safety program.

Since that time, an initiative was developed to provide national (Federal, State, local) consistency in the
content of Emergency Action Plans at dams throughout the country. As a result, the ad hoc

NOVEMBER 1998
Interagency Committee on Dam Safety (ICODS) prepared and approved federal guidelines for
emergency action planning at dams which was published by FEMA in October 1998. As a result of
the federal initiative, the FERC EAP Guidelines are further revised.

6-1.3 Scope

The FERC EAP Guidelines (Guidelines) establish a specific format to assist in preparing an effective,
workable EAP. The format was developed to include all the pertinent information required for the EAP
and accompanying appendices. In providing the appropriate information, the EAP should be site
specific, reflecting mode of operation, internal and external means of communication, and interaction
with appropriate agencies and owners of other sites. The format should be used in conjunction with the
instructions contained in Part 12, Subpart C of the Commission's Regulations.

All EAPs, except for those at government dams, shall conform with the format and criteria established
in these Guidelines. In order to ensure every EAP currently on file with the Regional Director complies
with the established format, every EAP filed prior to the date of issuance of these Guidelines must be
revised, as necessary, to conform with the format and then be completely redistributed to all
participants, with three copies resubmitted to the Regional Director. Subsequently, a completely
reprinted copy of the most up-to-date EAP must be redistributed to all participants, including three
copies to the Regional Director, on a five year cycle (as a maximum). The licensee/exemptee/applicant
for license has the option to place Appendix A of the Guidelines (Investigation and Analyses of
Dambreak Floods) in a separate volume which only has to be provided to the Commission. This
volume would need to be reprinted only when analyses are updated. All other sections of the EAP
must be reprinted at least every five years. During the intervening years, the licensee must maintain a
line of communication with all parties involved in their EAP. Regular exchanges of information will
assure that the EAP remains current and workable during an emergency. Information concerning
changes in organizations, personnel, phone numbers, emergency response responsibilities, or other site
specific information should be exchanged on a regular basis. Once notified of a change that would
affect the EAP, the licensee is required, within 30 days of the notification, make the necessary changes
to the EAP and issue revised pages, sections, maps, as appropriate, to all parties identified in the EAP.
If no interim changes are necessary, annual updates (which are to be submitted by December 31st of
each year) may be made by issuing to all plan holders only those pages that contain updated
information. Nevertheless, total reprintings of the EAP on more frequent basis are acceptable and
commendable.

When the applicant for license is not the owner of the dam nor is otherwise responsible for the
maintenance, operation and monitoring of the dam, the applicant for license should coordinate with the
owner of the dam to develop an EAP. In the event that an owner refuses to cooperate, the applicant
for license should prepare the EAP to the best of its ability with the information available to it and
provide it to the owner. If the owner indicates that it will not implement the EAP in the event of an
emergency, the applicant for license should provide a copy of the EAP to the State agency responsible

6-2 NOVEMBER 1998

for dam safety and explain the situation to the agency. The applicant for license should also advise the
Regional Director of the owner's lack of cooperation.

In the event of competing applications, if one of the applicants for license is the owner of the dam, it is
that applicant's responsibility to develop an EAP. If none of the competing applicants is the owner of
the dam, then it is the responsibility of the applicant first having its application on file to prepare the
EAP.

Under the provisions of Section 12.22 (c) of the Commission's Regulations, each hydroelectric project
under the jurisdiction of the Commission and located within a 10-mile radius of a nuclear power plant
reactor must have a radiological emergency response plan to be implemented in the event of a severe
accident or incident resulting in the release of radioactive materials from a nuclear plant. The guidelines
for preparation of a radiological emergency response plan (Section 6-5, page 6-60) should be used in
conjunction with the instructions contained in Section 12.22 (c) of the Commission's Regulations.

When a project is located at a Federal dam, a procedure for notifying the appropriate representatives
of the Federal agency of an emergency condition must be available (Section 6-6, page 6-61). The
EAP at a government dam must also include the requirement that the Commission's Regional Director
be notified immediately of the occurrence of an emergency situation.

6-1.4 Changes to the Guidelines

The Guidelines have been modified to be consistent with the "Federal Guidelines for Dam Safety:
Emergency Action Planning for Dam Owners", Mitigation Directorate FEMA 64, October 1998 (6-
3.3, page 6-12).

6-3 NOVEMBER 1998

6-2 ANALYTICAL REQUIREMENTS

6-2.1 General

The analytical requirements for an EAP include the information necessary to conduct dambreak
analyses and to prepare inundation maps. Since most of the engineering studies and requirements
related to flood studies pertinent to an EAP are discussed in Chapter II of the Engineering Guidelines
they will not be repeated in this Chapter. Therefore, out of necessity there will be frequent references
to Chapter II of the Engineering Guidelines. Thus, the reader should also have a copy of Chapter II for
reference.

The process of developing a workable EAP must necessarily begin with the knowledge of what areas
will be flooded as a result of a dam failure, or unusually large spillway release, so that the jurisdictions
and other agencies and individuals involved in the implementation of the EAP can be identified. The
tools for identifying the areas flooded and developing the notification procedures are primarily
dambreak analyses and inundation mapping. For EAP implementation, evacuation agencies will most
likely refer to the inundation maps to identify evacuation zones in the event of an emergency.

6-2.2 Dambreak Analyses

Information that should be provided regarding the dambreak analyses that are necessary to determine
the extent of inundation due to a dam failure should be included in Appendix A of the EAP document.
The dambreak analyses should be reviewed by the Regional Office staff for acceptability. Appendix II-
A of Chapter II of the Engineering Guidelines provides references and criteria which may prove useful
as indicators of reasonableness for the breach parameter, peak discharge, depth of flow, and travel
time determined by the licensee.

When large discrepancies in compared values exist or questions arise about assumptions, an
independent dambreak sensitivity analysis should be performed by Regional Office staff. For these
analyses, it is recommended that Regional Office staff use the dam break software package that utilizes
the National Weather Service's (NWS) DAMBRK Program (or most recent program). If necessary,
supplemental information should be requested from the licensee. Although the NWS Dambreak
program is preferred, it is acceptable to use other widely accepted models.

Several different inflow conditions may need to be investigated to determine the appropriate condition
prevailing at the time of a dam failure in order to ensure that the EAP includes all communities that need
to be notified. A "fair weather" dam failure (reservoir at normal full pool elevation, normal stream flow
prevailing) is generally considered to have the most potential for loss of human life, primarily due to the
element of surprise. However, a failure of a dam during flood flow conditions will result in flooding
downstream areas to higher elevations than during a "fair weather" failure. The result could be
additional loss of human life that otherwise would not occur during a "fair weather" failure. In addition,

6-4 NOVEMBER 1998

as discussed in Section 6-2.3 of this Chapter, experience has shown that the emergency preparedness
agencies will use the inundation maps to develop their evacuation procedures. To assist the agencies,
both the "fair weather" breach and a failure during a flood level approaching the inflow design flood
(IDF) are typically analyzed to bracket the full extent of the area potentially impacted by a failure. If
inundated areas for the "fair weather" breach and the IDF breach are essentially the
same or are too close to be shown separately on the inundation maps, then a single
inundation area for the two breach conditions may be shown. Otherwise, both the
"fair weather" breach and the flood flow breach should be clearly shown on the
inundation map because response agencies depend on the maps to implement an
evacuation and, therefore, need both dam failure boundaries shown on the maps.

These two failure conditions will greatly assist emergency preparedness agencies in their evacuation
responsibilities. Experience has shown that a failure during flood conditions can create special flooding
problems requiring changes and/or additions to the notification procedures that are developed for a "fair
weather" breach. If there is good reason not to include both boundaries, such as, the inundation
boundaries are essentially the same or are too close to be shown separately, or the agencies do not
want them on the maps, then only one boundary is necessary. (See discussion under inundation maps,
Section 6-2.3.)

For the flood condition failure, as discussed in Chapter II of the Engineering Guidelines, the dam should
be assumed to remain in place until the peak reservoir elevation for the assumed flood inflow condition
is attained before the postulated dam breach occurs. It may also be necessary to conduct a sensitivity
analysis of the breach parameters (i.e. varying the breach width and time to failure) for the various flood
inflow conditions in order to investigate the impacts for a range of possible failure scenarios. The
Regional Director may require, on a case-by-case basis, an investigation of other flood flow scenarios
in addition to those selected by the licensee to ensure that all communities requiring evacuation by local
officials have been identified. Pursuant to the Commission's Regulation (Part 12, Subpart D), most
dams that require an inspection and evaluation by an independent consultant have already been
investigated for the "fair weather" breach, as well as failure at the IDF. Therefore, most licensees have
available the necessary information to show the inundation boundaries for the "fair weather" and IDF
failure conditions.

Regardless of the flood condition analyzed, it should be remembered that the assumed flood condition is
not the only factor affecting the results of a dambreak study. Computer models, breach assumptions,
dam size and location, downstream terrain, map scales, and plotting of inundation boundaries inherently
place limitations on the accuracy of an inundation study. Therefore, the licensee should provide a
summary in Appendix A to the EAP of the possible limitations on the accuracy of the study (e.g.
computer generated elevations are expected to be within a certain accuracy).

The need to consider the domino effect of multiple dams should be made on a case-by-case basis. If
the assumed failure of a dam would cause the failure of any downstream dams, the licensee or dam

6-5 NOVEMBER 1998

owner has the responsibility to consider the domino effect in its routing of the floodwave downstream.
The flood wave should be routed to the point where it no longer presents a hazard to downstream life
or property, which includes downstream dams. Therefore, the licensee, after assuming a hypothetical
failure of its dam, should make an engineering judgment regarding the potential for failure of the
downstream dams from the flow condition under consideration or as a result of the failure of the dam
being investigated to determine whether it would be prudent to consider failure of any downstream
dams during the routing of the dambreak flood wave. The licensee may coordinate with the FERC
Regional Office staff to decide whether downstream dams should be considered to fail from the domino
effect. Coordination of such studies with downstream dam owners should be undertaken by the
licensees, where feasible. Of course, if the downstream dams are owned by the same licensee, this
should not be a problem.

Appendix VI-A contains guidance for the selection of acceptable breach parameters.

6-2.3 Inundation Maps

The elevations and travel times resulting from the dambreak analyses that clearly indicate the potential
hazard to downstream life and property should appear on an inundation map. Therefore, the inundation
map must be reviewed in conjunction with the dambreak study when determining acceptability. The
information on the inundation map must be up-to-date and adequate for the
development of a workable EAP. Therefore, it is recommended that the EAP text
describe the areas affected by a dam failure. The licensee should annually verify the accuracy
of the information provided in the text describing the areas affected by inundation. The text could then
be updated annually to reflect changes in the level of downstream development. Inundation maps
should be up-dated to reflect changes in downstream areas in accordance with the five year cycle
required in the EAP Guidelines for the complete reprinting and distribution of EAPs. Of primary
importance, however, is up-to-date inundation maps, whereas, the text is supplemental interim
information.

Inundation maps must conform to the requirements for mapping established in these Guidelines. The
inundation map should clearly indicate the areas subject to flooding. The maximum elevation,
increase in water surface elevation (rise), peak discharge, and arrival time of the
leading edge and peak of the flood wave at critical locations should be indicated on
the inundation map and/or in a table. Based on experience, it is extremely important
to show the arrival time of the leading edge of the flood wave on the inundation map.
If the map shows only the time of the peak elevation of the flood wave, the emergency preparedness
agencies could be led into a false sense of security believing they have more time available to evacuate
areas than actually exists. It is also important that the map be developed at a scale sufficient to be used
for identifying downstream inhabited areas (including habitable structures, recreational areas, etc.)
within the area subject to possible danger and that the inundated areas be clearly identified.

6-6 NOVEMBER 1998

It is best to terminate flood routing after non-damaging levels are obtained. However, to avoid
unnecessary study and costs, the routing may cease at a point where real-time flood warning
information can be provided on a preplanned basis. For example, if it is known that the time of failure
was 12:05 p.m. on Day 1 and the floodwave can be monitored, it may be possible to determine that the
floodwave will reach Town X at approximately 4:20 p.m. on Day 3; hence, real-time flood warning
information could be used. This may require coordination with the National Weather Service. If the
licensee terminates the flood routing before non-damaging levels are obtained, then it should be clearly
indicated that real-time flood warning information can be issued. Since real time data are
difficult to determine, this is not a recommended approach. It will most likely require
more than a day or two to obtain the information to model the actual failure, and known depths of
flooding and travel times, to be able to reproduce and predict real-time situations. Thus, real time flood
warning must be used selectively because of inaccuracies in predicting flood wave travel times. Again,
for these and other reasons, the use of real-time flood warning is not encouraged.

In order for an EAP to be a workable and usable document for the jurisdictions
affected, there are certain problems that need to be resolved by the jurisdictions.
For instance, road names used by local officials may be different from those used on
USGS maps or state route maps. The local agencies should be requested by the
licensee to specifically check road names so that the EAP includes the names
familiar to local residents. The agencies should be requested to furnish, by letter or
other documentation, the road names used locally so that EAP maps can be
appropriately modified. In addition, all bridges in the inundation area should be
highlighted. It is recommended that a note be included on the map advising the users of the map
that, because of the method and procedures used to develop the flooded areas, the limits of flooding
shown are solely to be used as a guideline for the establishment of evacuation zones.

Since local officials will likely use the inundation map for evacuation purposes, the accuracy, limitations,
and conditions for which the map was developed should be clearly understood. For example, when an
inundation map is developed for a "fair weather" breach and it is only supplemented by a note indicating
that failure under flood flow conditions would require evacuation to higher elevations, the evacuation
agency personnel are placed in a difficult position to accurately determine by extrapolation the areas
that may be impacted by a dambreak for some flood inflow condition greater than the one upon which
the inundation maps are based. For this reason, numerous emergency preparedness agencies are
requesting that the maximum flood level of potential adverse impact also be shown on the maps.

For consistency, the maximum flood level should be based on a hypothetical dam failure during the
inflow design flood (IDF). The IDF is the flood inflow condition above which the failure of the dam has
an insignificant effect on downstream flooding - see Chapter II. Therefore, unless there is good
reason to select other flood conditions under which the dam is assumed to fail,
subject to the discussion in Section 6-2.2 (page 6-3), licensees should show both the
"fair weather" and IDF dambreak flood levels as a means of developing a more

6-7 NOVEMBER 1998

effective EAP. The IDF level without failure should not be shown. The result will be inundation
maps that show the extremes in area that could be inundated. This will be of substantial assistance to
evacuation agencies that desire this information on the maps in order to better establish the evacuation
zones. The preferred method of representing the two dambreak scenarios on the same maps is to
identify the normal streamflow condition with one color, the area flooded by a "fair weather" breach
with another color, and the differential increase in flooding between the "fair weather" breach and a
failure during the IDF with yet another distinctive color. However, other methods that clearly identify
differences between the two failure conditions are also acceptable.

FERC staff has been advised by some agencies that the two dambreak conditions are helpful to the
evacuating agencies because they show the extremes in potential inundation, the difference in travel
times between the two conditions, and how far downstream evacuation is required for each condition.
If, through coordination with the evacuation agencies, it is their conclusion that
they need only one dambreak condition represented on the maps, or dambreak
conditions under flood conditions different from normal flow and/or the IDF, then
written documentation should be included in the EAP.

6-2.4 Exemption Requirements

In order to receive an exemption from filing an EAP, a licensee must demonstrate that no reasonably
foreseeable project emergency (i.e. failure of a dam or water retaining structure) would endanger life,
health or property. To satisfactorily demonstrate the consequences of a failure, the licensee will have to
submit a report that documents all reconnaissance and other studies performed to determine that failure
of the dam will not present a hazard to human life or cause significant property damage under all flood
flow conditions up to the Inflow Design Flood. Regional Office staff are to periodically review the
circumstances pertaining to those projects that have already been exempted from EAP requirements to
determine if additional documentation is necessary to verify the validity and continuation of previously
granted EAP exemptions.

If the results of a field reconnaissance study of the areas downstream of the dam are inconclusive in
determining the hazard potential of the dam, a dambreak analysis should be performed and results of
the analysis furnished in the report. The dambreak analysis should consider failure under normal
operating conditions and flood flows up to the point where no significant increase in hazard to
downstream life and property occurs as a result of failure, i.e. the inflow design flood. For each flood
event analyzed, it should be assumed that the failure is initiated when the peak flow or reservoir
elevation is reached. As discussed in detail in Chapter II of the Engineering Guidelines, dam failure
should be assumed to occur at the peak and not on the rising limb of the inflow flood hydrograph. A
sensitivity analysis should also be performed to establish the effect of breach width and time to failure
on downstream flood levels at various flood flow conditions.

6-8 NOVEMBER 1998

Chapter II, Appendix II-C, of the Engineering Guidelines discusses in detail the procedure for
performing a hazard evaluation. An inundation map and, if necessary, water surface profiles, should be
developed and furnished for the flow condition which results in the greatest potential for loss of life and
significant property damage. The method and assumptions utilized in the dambreak analysis should be
fully documented. The inundation map and water surface profiles should delineate the affected areas
and water surface elevations prior to failure with the dam in place and after the assumed failure. The
map and river profiles should also show the travel time for the arrival of the initial or leading edge of the
flood wave and the peak elevation of the flood wave at critical locations downstream of the dam. It is
important that the inundation map be developed at a scale sufficient to be used for identifying the
location of downstream inhabitants within the area subject to possible danger. The licensee should
annually verify the accuracy of the information on downstream development that appears on the
inundation maps. The licensee should perform a field reconnaissance to verify that the information on
the inundation maps is as accurate as possible and document this in writing to the Regional Director. If
there are any changes in downstream development, it will be necessary to evaluate whether the
exemption remains valid.

6-3 EAP PREPARATION

6-3.1 General

All EAPs, except for those at government dams, are to conform to the format and criteria established in
these Guidelines. This format should facilitate the preparation, updating, and annual review of an EAP.
It should be used in conjunction with the instructions contained in Part 12, Subpart C of the
Commission's Regulations. The format was developed to include all of the pertinent information to be
included in the EAP as required by the Commission's Regulations. It is important that the inundation
maps conform to the criteria discussed in these Guidelines so that they will be of sufficient scale to
clearly identify all impacted areas. The initial EAP shall be submitted in a loose-leaf binder, whereby
outdated pages or the entire EAP (every five years or less) can be easily removed and replaced by
updated information to ensure a complete, current, and workable plan.

6-3.2 Basic Considerations for Preparing Emergency Action Plans

-3.2.1 Purpose

There are many types of emergency events that could affect dams. Whenever people live in areas that
could be flooded as a result of failure of or operation at a dam, there is a potential for loss of life and
damage to property. The general purpose of these Guidelines is to encourage thorough and consistent
emergency action planning to help save lives and reduce property damage in areas that would be
affected by dam failure or operation.

6-9 NOVEMBER 1998

An Emergency Action Plan, or EAP, is a formal document that identifies potential emergency
conditions at a dam and specifies preplanned actions to be followed to minimize property damage and
loss of life. The EAP specifies actions the licensee should take to minimize or alleviate the problems at
the dam. It contains procedures and information to assist the licensee in issuing early warning and
notification messages to responsible downstream emergency management authorities of the emergency
situation. It also contains inundation maps to show the emergency management authorities the critical
areas requiring action in case of an emergency.

An emergency in terms of dam operation is defined as an impending or actual sudden release of

water caused by an accident to, or failure of, a dam or other water retaining structure, or the result of
an impending flood condition when the dam is not in danger of failure. The release of water may
endanger human life or downstream property.

The effectiveness of an Emergency Action Plan program is enhanced by promoting a uniform format for
EAPs which ensures that all aspects of emergency planning are covered in each plan. Uniform
Emergency Action Plans and advance coordination with local and state emergency management
officials and organizations should facilitate a timely response to a developing or actual emergency
situation.

6-3.2.2 Scope

These Guidelines are used for preparing or revising Emergency Action Plans and apply to all dams
unless exempted under Part 12, Subpart C, 12.21 (Section 6-2.4). Ownership and development of the
floodplain downstream from dams varies; therefore, the potential for loss of life as a result of failure or
operation of a dam will also vary. Every EAP must be tailored to site-specific conditions.

Emergency Action Plans generally contain six basic elements:

• Notification Flowchart
• Emergency Detection, Evaluation, and Classification
• Responsibilities
• Preparedness
• Inundation Maps
• Appendices

All of the elements should be included in a complete EAP. The licensee is responsible for the
development of the EAP. However, the development or revision of an EAP must be done in
coordination with those having emergency management responsibilities at the state and local levels.
Emergency management agencies will use the information in a licensee's EAP to facilitate the
implementation of their responsibilities. State and local emergency management authorities will

6-10 NOVEMBER 1998

generally have some type of plan in place, either a Local Emergency Operations Plan or a Warning and
Evacuation Plan.

6-11 NOVEMBER 1998

6-3.2.3 The Six Basic Elements of an EAP

This section lists and briefly examines why there is a need for the six basic elements of an EAP. The
requirements of these elements are discussed in detail in 6-3.3 of this chapter, which presents a format
for uniformity among EAPs.

A. Notification Flowchart. A notification flowchart shows who is to be notified, by whom, and in what
priority. The information on the notification flowchart is necessary for the timely notification of persons
responsible for taking emergency actions.

B. Emergency Detection, Evaluation, and Classification. Early detection and evaluation of the
situation(s) or triggering event(s) that initiate or require an emergency action is crucial. The
establishment of procedures for reliable and timely classification of an emergency situation is
imperative to ensure the appropriate course of action is taken based on the urgency of the situation. It
is better to activate the EAP while confirming the extent of the emergency, than waiting for the
emergency to fully develop.

C. Responsibilities. A determination of responsibility for EAP-related tasks must be made during the
development of the plan. Licensees are responsible for developing, maintaining, and implementing the
EAP. State and local emergency management officials having statutory obligation are responsible for
warning and evacuation within affected areas. The EAP must clearly specify the licensees
responsibilities to ensure effective, timely action is taken should an emergency occur at the dam. The
EAP must be site-specific, since conditions at and downstream of all dams are different.

D. Preparedness. Preparedness actions are taken to moderate or alleviate the effects of a dam failure
or operational spillway release and to facilitate response to emergencies. This section identifies actions
to be taken prior to any emergency.

E. Inundation Maps. An inundation map should delineate the areas that would be flooded as a result of
a dam failure. Inundation maps are used both by the licensee and emergency management officials to
facilitate timely notification and evacuation of areas affected by a dam failure or flood condition. These
maps greatly facilitate notification by graphically displaying flooded areas and showing travel times for
wave front and flood peaks at critical locations.

F. Appendices. The appendices contain information that supports and supplements the material used
in the development and maintenance of the EAP.

6-12 NOVEMBER 1998

6-3.2.4 Coordination

It is vital that development of the EAP be coordinated with all entities, jurisdictions, and agencies that
would be affected by a dam failure and/or flooding as a result of large operational releases, or that have
statutory responsibilities for warning, evacuation, and post-flood actions. The finished product should
be user friendly as it realistically takes into account each organization's capabilities and responsibilities.

Coordination with state and local emergency management officials at appropriate levels of management
responsible for warning and evacuation of the public is essential to ensure that there is agreement on
their individual and group responsibilities. Participation in the preparation of the EAP will enhance their
confidence in the EAP and in the accuracy of its components. Coordination will provide opportunities
for discussion and determination of the order in which public officials would be notified, backup
personnel, alternate means of communication, and special procedures for nighttime, holidays,
weekends, etc.

The tasks and responsibilities of the licensee and the emergency management officials that would be
implemented during a dam emergency incident need to be as compatible as possible.

To facilitate compatibility, the licensee should coordinate emergency response actions with the local
emergency management officials who have the responsibility to provide a timely warning and evacuation
notice to populations at risk. This should help prevent over, or under, reaction to the incident by
various organizations.

6-3.2.5 Evacuation

State and local officials who are charged with the safety of the public who live in areas that would be
inundated by failure of a dam or flood releases are responsible for evacuation planning and
implementation during a dam emergency. The licensee should not usurp the responsibility of the local
authorities responsible for evacuation. However, there may be situations where recreational facilities,
campgrounds, or residences may be located below a dam where local authorities would not be able to
issue a timely warning. In such cases, the licensee should coordinate with local emergency management
officials to determine who will warn these people and in what priority. Evidence of coordination
between the licensee and the alerting agencies should be provided in the EAP.

6-3.2.6 Emergency Duration, Security, Termination, and Follow-up

An Emergency Action Plan needs to address who in the licensee's organization issues status reports
during the emergency, when and how a declared emergency will be terminated, what security
provisions shall be maintained at the dam, and plans for a follow-up evaluation and report.
A. Emergency Duration. Emergency situations that occur at a dam will require that status reports and
situation assessments be provided by the licensee to appropriate organizations throughout the duration
of the incident.

6-13 NOVEMBER 1998

B. Security Provisions. An Emergency Action Plan should consider security provisions at, and
surrounding, the dam during emergency conditions in order to protect the public and permit effective
performance of emergency response actions.

C. Emergency Termination. There are two conditions requiring a termination of the emergency. One
has to do with emergency conditions at the dam and the other is related to the evacuation and disaster
response. The licensee is usually responsible for making the decision that an emergency condition no
longer exists at the dam. The EAP should clearly designate the responsible party. The applicable state
or local emergency management officials are responsible for termination of the evacuation or disaster
response activities.

The licensee and state and local officials should agree on when it is appropriate to terminate an
emergency. The licensee should cooperate with state and local officials to determine if a news release
is appropriate which can be used by the media for broadcast to the general public notifying them of
termination of the emergency condition. Such news releases are expected to be a supplement to other
methods of notifying the public that the emergency has been terminated.

D. Follow-up Evaluation. Following an emergency, an evaluation and review should be conducted by

the licensee that includes input from all participants. The following should be discussed and evaluated in
the after-action review:

• Events prior to, during, and following the emergency;

• Significant actions taken by each participant, and what improvements would be
practicable for future emergencies; and
• All strengths and deficiencies found in procedures, materials, equipment, staffing levels,
and leadership.

The results of the after-action review should be documented in an evaluation report chaired by the
licensee and used as a basis for revising the Emergency Action Plan.

6-3.2.7 Maintaining an Emergency Action Plan

After the Emergency Action Plan has been developed, approved, and distributed, the job is not done.
Without periodic maintenance, the EAP will become out-dated, lose its effectiveness, and no longer be
workable. If the plan is not exercised (verified), those involved in its implementation may become
unfamiliar with their roles and responsibilities, particularly if emergency response personnel change. If
the plan is not updated, the information contained in it may become outdated and useless. Maintaining
an Emergency Action Plan is addressed in greater detail in 6-4 of this chapter.

6-3.3 EAP Format

6-14 NOVEMBER 1998

6-3.3.1 The Format

A format is provided in these Guidelines to ensure all six basic elements are included in an EAP, to
provide uniformity, and to encourage thorough and consistent emergency action planning for levels of
preparedness that may save lives and reduce property damage in areas affected by dam operation or
failure. It is important that licensee and regulatory requirements be satisfied when selecting a format for
an EAP.

Although it is not necessary to exactly follow the format outlined below, it is necessary that all EAPs
within a given jurisdiction be similar and consistent to eliminate confusion when activating any EAP. To
the extent possible, an EAP should be organized in the format that is most useful for those involved in
the plan. The EAP must be user friendly so that it will actually be used during EAP exercises and actual
emergency events. Regardless of the format used, development of an EAP should consider the
elements described on the following pages to ensure all aspects of emergency action planning are
covered.

It is helpful to place the EAP in a loose-leaf binder, so that outdated pages (or the entire EAP) can be
easily removed and replaced with updated information, to ensure a complete, current, and workable
plan. It is also beneficial to place the date of the EAP or current revisions on each page.

6-15 NOVEMBER 1998

The format for an EAP appears below:

EAP FORMAT

Title Page/Cover Sheet

Table of Contents

I. Notification Flowchart
II. Statement of Purpose
III. Project Description
IV. Emergency Detection, Evaluation, and Classification
V. General Responsibilities Under the EAP

A. Licensee Responsibilities
B. Responsibility for Notification
C. Responsibility for Evacuation
D. Responsibility for Termination and Follow-Up
E. EAP Coordinator Responsibility

VI. Preparedness

A. Surveillance
B. Response During Periods of Darkness
C. Access to Site
D. Response During Weekends and Holidays
E. Response During Periods of Adverse Weather
F. Alternative Systems of Communication
G. Emergency Supplies and Information

VII. Inundation Maps

VIII. Appendices

A. Investigation and Analyses of Dambreak Floods

B. Plans for Training, Exercising, Updating, and Posting the EAP
C. Site Specific Concerns
D. Documentation
E. Approval of the EAP

The format separates an EAP into two distinct sections: the basic EAP and the Appendices. Together,
these sections constitute a complete EAP.

6-16 NOVEMBER 1998

A. The Basic EAP. Sections I through VII of the format constitute the basic Emergency Action Plan.
That is, they contain information that should be used by all parties (both the licensee and emergency
management officials) during an actual emergency. For example, the licensee will use the notification
flowchart to issue its emergency warning to the appropriate officials in a prioritized order. Similarly, the
emergency management officials should use the flowchart to contact other officials or the licensee, as
needed, throughout the emergency. As a second example, both the licensee and the emergency
management officials will use the inundation maps extensively in fulfilling their responsibilities.

It must be remembered that the responsibilities of the state and local emergency management authorities
and other organizations in the jurisdictions affected by a dam failure or flooding as a result of operation
of a dam are not included in an EAP. Information unique to state and local emergency management
authorities, and any other organizations who would have responsibilities for the warning and evacuation
of populations at risk, would be included in the portion(s) of the appropriate jurisdiction's Emergency
Operations Plan dedicated specifically to warning and evacuation of populations placed at risk as a
result of dam failure or flooding due to large operational releases. However, the information in the EAP
must be coordinated with the appropriate authorities since they will depend on and use the information
in the licensee's EAP to help them carry out their responsibilities.

B. The Appendices. The Appendices are also an important element which completes the EAP.
However, the information contained in the Appendices is not necessarily needed by all parties during an
actual emergency. They typically contain support materials used in the development of the basic EAP.
More specifically, the Appendices focus on such important issues as those that specifically address
maintenance requirements for the EAP and dambreak investigations and analyses, among others. This
information may be directly applicable to the actions of the licensee and possibly some of the
emergency management parties, but may not be critical to the actions and activities of other parties
during an actual emergency.

All emergency management officials should be offered the complete EAP. However, it may be left to
their discretion to decide whether they want to receive a copy of the complete EAP (basic EAP +
Appendices) or just the basic EAP. Those who opt to receive just the basic EAP should
understand that if it does not provide sufficient information for them to perform
their functions, then they should obtain the complete EAP.

NOTE: Every EAP must be tailored to site-specific conditions and to the requirements
of the organization that owns, operates, or regulates the use of the dam. This can be
accomplished under the format. Uniformity of EAPs is important because any one state
or local emergency management agency may be affected by a river system that has a
series of dams, the independent failure or operation of which may impact the
jurisdiction. Uniformity provides for clarity and better understanding of the
information in the EAP for each individual dam.

6-17 NOVEMBER 1998

6-3.3.2 Format Items Defined

This section follows the heading and numbering of the format and describes in detail each element of an
EAP.

Title Page/Cover Sheet

An EAP document's cover identifies it as an Emergency Action Plan and specifies the dam for which it
was developed. Since each dam must have its own EAP with its own specific procedures to be
followed, title pages or cover sheets are essential, so personnel can be sure that they are using the right
EAP for the circumstances. To assist state and federal dam safety personnel, include the National
Inventory of Dams number unique to each dam on the title page. A sample title page is included in
Appendix VI-C.

Table of Contents

List all major items in the Table of Contents, including flow charts, figures, tables, etc. A sample table
of contents is included in Appendix VI-C.

I. Notification Flowchart

The EAP should begin with one Notification Flowchart that clearly summarizes the following
information and is applicable to each of the emergency classification levels considered (See discussion
under item IV):

• Who is responsible for notifying each licensee representative(s) and/or emergency

management official(s).
• What is the prioritized order in which individuals are to be notified.
• Who is to be notified.

The notification flowchart should include individual names and position titles, office and home telephone
numbers, alternative contacts and means of communication (e.g., radio call numbers). The number of
persons to be notified by each responsible individual on the notification flow chart should be governed
by what other responsibilities the person has been assigned. It is usually recommended that individuals
not be responsible for contacting any more than three or four other parties.

The notification list should consider the following:

• Licensee.
• Local emergency management officials and other organizations.
• Appropriate federal and state emergency management agencies.
• Residents and property owners that are located immediately downstream of the dam
within the boundary of potential inundation where available warning time is very limited.
• Operators of other dams or water-retention facilities.
• Managers and operators of recreation facilities.

6-18 NOVEMBER 1998

 • National Weather Service
• News media. 1
• Others, as appropriate.

Although the list is not all inclusive, nor a prioritization of those entities listed, both the licensee and the
local, state, and federal emergency management authorities are typically given top priority in the
notification flowchart.

The Notification Flowchart should be easy to follow for each emergency classification level (see
Section IV). A single flowchart that represents all levels is preferred, for the sake of
effectiveness and simplicity. However, under certain conditions for clarity it might be necessary to
develop a flowchart for each classification level. Color coding (i.e. using different colored lines to trace
the proper sequence of notification under various emergency classification levels) may prove helpful. If
necessary, narrative information supplementing the flowchart may be provided on the page following the
flowchart.

NOTE: Information is exchanged both up and down the notification flowchart.

Copies of the flowchart should be readily available to each individual having responsibilities under the
plan, and should be kept up-to-date through exercises and revisions.

A sample Notification Flowchart is shown on the next page.

NOTE: This is only a sample flowchart. A flowchart must be tailored to the specific
needs and notification priorities of the dam to which it applies.

1
The news media, including radio, television, and newspapers, should be utilized to the extent available and
appropriate. Use of new media should be preplanned to the extent possible by the licensee or emergency
management officials. Notification to the news media may be by the licensee or emergency management
officials depending on the type of emergency. Notification plans should define emergency situations for
which each medium will be utilized and should include an example of a news release that would be the most
effective for each possible emergency. Information for media ordinarily should not be relied upon as the
primary means of warning.

6-19 NOVEMBER 1998

6-20 NOVEMBER 1998
 II. Statement of Purpose

Following the Notification Flowchart, briefly state the purpose and scope of the EAP. Two examples
of a statement of purpose are shown below.

Example 1: "This plan defines responsibilities and provides procedures designed to identify
unusual and unlikely conditions which may endanger Any Dam in time to take mitigative action
and to notify the appropriate emergency management officials of possible, impending, or actual
failure of the dam. The plan may also be used to provide notification when flood releases will
create major flooding."

Example 2: "The purpose of this Emergency Action Plan (EAP) is to safeguard the lives and
reduce damage to the property of the citizens of Alpha County living along Beta Creek, in the
event of failure of the Beta Creek Dam or flooding caused by large runoff."

III. Project Description

Provide a description of the project and its location. Include a project vicinity map and a simple
drawing showing project features. List any significant upstream or downstream dams. List
downstream communities potentially affected by a dam failure or by flooding as a result of large
operational releases. List and highlight critical site specific concerns (i.e., critical operating procedures
and material stockpiles) and refer the reader to more specific information contained in Appendix C of
the EAP document.

IV. Emergency Detection, Evaluation, and Classification

The EAP document should include a discussion of procedures for timely and reliable detection,
evaluation, and classification of an existing or potential emergency condition.

The conditions, events, or measures for detection of an existing or potential emergency should be
listed. Data and information collection systems (early warning system hardware, rule curves or other
information related to abnormal reservoir levels, inspection/monitoring plan, inspection procedures,
instrumentation plan, etc.) should be discussed. The process that will be used to analyze incoming data
should also be described.

Procedures, aids, instruction, and provisions for evaluation of information and data to assess the
severity and magnitude of any existing or potential emergency should be discussed.

Emergencies are classified according to their severity and urgency. An emergency classification
system is one means to classify emergency events according to the different times at which they occur
and to their varying levels of severity. The classification system indicates the urgency of the emergency

6-21 NOVEMBER 1998

condition. Emergency classifications should use terms agreed to by the licensee and
emergency management officials during the planning process, in order for the
system to work and to ensure organizations understand terminology and respond
appropriately to the event.

Titles for emergency classifications should be chosen carefully by the organizations who will use them so
that everyone will understand what each classification level means when notifications are issued and
received.

Declaration of an emergency can be a very controversial decision. The issue should

not be debated too long. An early decision and declaration is critical to maximize
available response time.

Some locations may require only two emergency classifications, while others may require more. For
the purpose of these EAP Guidelines, two dam failure emergency classifications and one
non-failure emergency classification are provided:

• Failure is imminent or has occurred (Condition A).

• Potential failure situation is developing (Condition B).

• Non-failure emergency condition.

The definition of these conditions follows:

• Failure is imminent or has occurred (Condition A)

Generally, this situation should convey the impression that "time has run out" with respect to the
failure of the dam. This is a situation where a failure either has occurred, is occurring, or
obviously is just about to occur. The question is often asked, "how much time is available when
failure is considered to be imminent?" It is impossible to determine how long it will take for a
failure to occur or for a complete breach to occur once failure begins. Therefore, once a
licensee determines that there is no longer any time available to attempt
corrective measures to prevent failure, the "failure is imminent or has
occurred" warning should be issued. Emergency management agencies, for evacuation
purposes, should conservatively interpret the phrase "failure is imminent" to mean that the dam
is failing. It should not be assumed that there is any time lag between "failure is imminent" and a
"failure has occurred." Therefore, "failure is imminent" and "failure has occurred" should
conservatively be interpreted as essentially the same condition for evacuation purposes.

6-22 NOVEMBER 1998

• Potential failure situation is developing (Condition B)

Generally, this situation should convey the impression that "some amount of time" is still
available for further analyses/decisions to be made before dam failure is considered to be a
foregone conclusion. This is a situation where a failure may eventually occur, but pre-planned
actions taken during certain events (such as major floods, earthquakes, evidence of piping, etc.)
may moderate or alleviate failure. Even if failure is inevitable, more time is generally available
than in a failure has occurred situation to issue warnings and/or take preparedness actions.

Is the time frame for this situation in hours, days, or weeks? When a dam safety situation is
observed that may lead to a failure if left unattended, but there is no immediate danger, the
licensee should issue a warning that a "potential failure situation is developing". The licensee
should assess the situation and determine the urgency of the emergency situation. Based on the
licensee's assessment (and as a result of prior coordination with the appropriate authorities), the
authorities should be placed on alert and it is up to the authorities to determine the appropriate
course of action.

If it appears that a situation may take days or weeks before it could develop into a failure
situation, the local authorities may decide on one course of action. Periodic status report
updates from the licensee are important because when it appears that the situation is continuing
to worsen at the dam, in spite of the actions being taken to moderate or alleviate failure, the
local authorities may decide to change their course of action. Depending on the location of
downstream residents with respect to the dam and the estimated warning time available, the
evacuating agencies should consider the prudence of early evacuation, or heightened
awareness, of certain downstream areas until the emergency has passed.

NOTE: It should be remembered that it may be appropriate to immediately declare a

Condition A. However, there should be smooth transition from Condition B to
Condition A when using Condition B initially.

To assist the evacuating agencies in selecting their appropriate course of action and to provide a proper
transition from Condition B to Condition A, the licensee should clearly communicate their assessment of
the situation to the agencies. The licensee should consider placing the agencies on an initial alert and
provide periodic updates on the situation as it develops so that the agencies can assess when they
should implement their evacuation procedures. For example, a licensee could issue an initial warning
and periodic updates on the reservoir level as it rises during flood conditions and eventually overtops an
embankment dam. As the reservoir rises, "a potential failure situation is developing" warning should be
implemented with periodic updates on how much time is available before the embankment overtops.
Immediately before the embankment overtops, a "failure is imminent or has occurred" warning should
be issued.

6-23 NOVEMBER 1998

• Non-failure Emergency Condition

Generally, this situation should be used when there is no danger of dam failure, but flow
conditions are such that flooding is expected to occur downstream of the dam. Non-failure
emergency conditions are more common than the failure emergency conditions. Use of the
EAP can provide an early warning to downstream areas during flood conditions or large
spillway releases. Based upon the severity of the flooding, local site conditions, consultations
with local emergency response agencies, and standard operating procedures used at the dam,
the EAP may not need to be activated during a non-failure emergency condition. However, it
may become necessary to fully activate the EAP if conditions escalate to levels agreed to
beforehand by all involved participants. Therefore, an important application of the EAP is
when there is a flood occurring on the river system, but there may be no apparent threat to the
integrity of the dam. In this situation, natural flooding is expected or is in progress upstream
from the dam site and an impending or actual release of water to downstream areas will result
from unusually large spillway releases or passage of unusually large flows at dams having
uncontrolled spillways. The licensee provides an important public service by notifying the
appropriate agencies of the expected release or passage of flood waters below the dam. While
the amount of flooding may be beyond the control the licensee, information on the amount of
releases from the dam is very helpful to the authorities in reaching any decisions on the need for
evacuation. Site specific concerns will dictate the level of notification necessary during a non-
failure emergency condition.

V. General Responsibilities Under the Plan

The plan should specify the person(s) or organization responsible for the maintenance and operation of
the dam and the persons or groups responsible for implementing various phases of the EAP. Some
specific responsibilities to be considered are discussed below.

A. Licensee's Responsibilities. The duties of the licensee or owner's designated representatives under
the EAP should be clearly described. Suggested information to include in this section include, but are
not limited to the following:

The operators should be advised of the importance of the Emergency Action Plan and why the EAP is
necessary. The operators' duties under the EAP should be described. Include pointers on how to
communicate the emergency situation to those who need to be contacted along with samples of typical
communications.

Specific actions operators are to take after implementing the EAP notification procedures should be
described. For example, opening spillway gates, especially if a certain sequence is desired, and
opening/closing water intakes, as appropriate. Instructions for the operation of the project during the
anticipated emergency should be provided.

6-24 NOVEMBER 1998

The chain of command should be described. Officials and alternates of the licensee that must be
notified should be designated and a priority of notification determined. Notification of supervisory
personnel on the licensee's staff is desirable, if time permits. Advice may be needed concerning
predetermined remedial action to delay, moderate, or alleviate the severity of the emergency condition.
The responsibilities required by the EAP should be coordinated with appropriate levels of management
to ensure full awareness of organizational capabilities and responsibilities. EAPs must always be
developed as a result of coordination and consultation with other entities and agencies that will be
affected by a failure of a dam, or large operational releases, or have statutory responsibilities in warning
and evacuation.

B. Responsibility for Notification. The person(s) authorized to notify state and local officials should be
determined and clearly identified in the EAP. If time allows in an emergency situation, onsite
personnel should seek advice and assistance. However, under certain circumstances, such as when
failure is imminent or has occurred, the responsibility and authority for notification may have to be
delegated to the dam operator or a local official. Such situations should be specified in the EAP.

The accurate and timely dissemination of emergency public information is very important to the overall
success of an EAP. The person who is responsible for disseminating information to the media and the
public on a periodic basis throughout the emergency should be designated. If resources are available,
an exclusive public information officer should be identified to disseminate all media briefs. Also, means
for keeping local authorities advised of continuing conditions at the dam should be described.

Licensees should develop procedures for dissemination of dam specific information to the media in
anticipation of questions the media may have about the incident as it applies to the dam. A procedure
like this should, in effect, help minimize the potential for dissemination of misinformation and spreading
of false rumors.

Throughout the United States, the National Weather Service (NWS) and/or other agencies have the
general responsibility for issuing flood warnings. Include the appropriate agency having this
responsibility on the notification chart so that its facilities can enhance warnings being issued.

Local agencies will usually establish an Emergency Operations Center (EOC), or Incident Command
System (ICS), to serve as the main distribution center for warning and evacuation activities. The
availability of specific local resources should be determined through discussion and orientation seminars
with local agency personnel. Proper coordination and communication among onsite technical personnel
at the dam, public information officer(s), and emergency personnel at the EOC is critical to a successful
EAP. Thorough verification during comprehensive EAP exercises will greatly assist in providing this
smooth interface.

C. Responsibility for Evacuation. Warning and evacuation planning are the responsibilities of local
authorities who have the statutory obligation. Under the EAP the licensee is responsible for notifying

6-25 NOVEMBER 1998

the appropriate emergency management officials when flooding is anticipated, a dam failure is imminent
or has occurred, or a potentially hazardous situation is developing.

Licensees should not assume, nor usurp, the responsibility of government entities for evacuation of
people. However, there may be situations in which routine notification and evacuation will not suffice,
as in the case of a resident located just downstream of the dam. In this case, the licensee should
arrange to notify that person directly. This procedure should be coordinated with the
appropriate public officials prior to an emergency situation developing.

D. Responsibility for Duration, Security, Termination, and Follow-Up. A person should be

designated for on-site monitoring of the situation at the dam and keeping local authorities informed of
developing conditions at the dam from the time that an emergency starts until the emergency has been
terminated.

Provisions for security measures at the dam during the emergency should be specified.

A person should also be responsible for declaring that the emergency at the dam is terminated. The
applicable state or local emergency management officials are responsible for termination of the disaster
response activities.

A follow-up evaluation after an emergency by all participants should be specified. The results of the
evaluation should be documented in a written report.

E. EAP Coordinator Responsibility. The licensee should specify in the EAP the designated EAP
coordinator who will be responsible for EAP-related activities, including (but not limited to) preparing
revisions to the EAP, establishing training seminars, coordinating EAP exercises, etc. This person
should be the EAP contact if any involved parties have questions about the plan.

VI. Preparedness

Preparedness actions are taken to prevent a dam failure incident, or to help reduce the effects of a dam
failure or operational spillway release and facilitate response to emergencies. A few of the
preparedness actions that a licensee may take include providing emergency flood operating instructions,
and arranging for equipment, labor, and materials for use in emergency situations.

The EAP should describe preparedness actions taken both prior to and following the development of
emergency conditions. Preparedness actions involve the installation of equipment or the establishment
of procedures for one or more of the following purposes:

• Preventing emergency conditions from developing, if possible, or warning of the development of

emergency situations.

6-26 NOVEMBER 1998

• Facilitating the operation of the dam to limit impacts in an emergency situation.

• Minimizing the extent of damage resulting from any emergency situations that do develop.

The need for timely action in an emergency situation cannot be overemphasized. The EAP should
contain a discussion of provisions for surveillance, and evaluation of an emergency situation and should
clearly indicate that emergency response procedures can be implemented in a timely manner. An
important factor in the effectiveness of the EAP is the prompt detection and evaluation of information
obtained from instrumentation and/or physical inspection procedures.

In the EAP, discuss the time factor from the actual occurrence of an emergency to awareness of the
emergency, and its effect on the workability of the EAP. Timely implementation of the EAP
and coordination and communication with downstream local authorities are crucial
elements in the effectiveness of emergency response to the incident.

There are several types of preparedness actions that should be considered when developing an EAP.
These actions include:

• Surveillance
• Response during periods of darkness
• Access to the site
• Response during weekends and holidays
• Response during periods of adverse weather
• Alternative systems of communication
• Emergency supplies and information

The following sections discuss each of these actions:

A. Surveillance. The EAP should contain a discussion of provisions for surveillance, detection and
evaluation of an emergency situation and should clearly indicate that the EAP can be implemented in a
timely manner.

When a dam is not continuously attended and dam failure or operational releases would endanger
human life or cause significant property damage, it is imperative that procedures be developed to
identify conditions requiring emergency actions, and to promptly alert emergency management officials
responsible for warning and evacuation of residents who would be affected in the event of an
emergency at the dam. In order to be able to promptly notify responsible officials of emergency
conditions, a licensee should be able to detect and evaluate developing emergency conditions. The
information system must be able to deliver clear, concise, and reliable data so that the responsible
official(s) may react with confidence and implement the EAP. While the EAP is being activated,
personnel should visit the site to verify conditions.

6-27 NOVEMBER 1998

For an unattended dam, a remote surveillance system that includes instrumentation and telemetering
facilities at the dam site should be considered to provide a continuous reading of headwater and
tailwater levels. If the licensee has an operations control center that is attended 24 hours a day, the
system should include a computer at the operations center to monitor the data, and to activate an
audible alarm whenever the rate of change of the headwater or tailwater over a given period of time
exceeds prescribed limits. The alarm also should be activated if the headwater or tailwater elevations
exceed prescribed maximum or minimum levels. The design must be site-specific. The limits
programmed in a system must account for changes in headwater and tailwater levels that would occur
during normal dam operation, floods, maintenance, etc.

Monitoring of the tailwater generally is more sensitive to changes resulting from a breach of the
structures than monitoring the headwater. Changes in tailwater will alert operators more quickly to site
conditions and help determine whether emergency management officials should be notified. If
continuous readings of both the headwater and tailwater are available, the operator can obtain a current
reading at any time and check conditions at the site after an alarm is sounded.

Provisions should be made for the alarm to sound when there is an interruption of power to, and loss of
communication with, the monitoring instrumentation. (When a dam tender lives close to the project, an
alarm should be installed in the dam tender's house.) When power to or communication with the site is
interrupted, the dam should be staffed until conditions are returned to normal. Operation of the alarms
should be checked and tested periodically. For instance, annual testing of the EAP might be initiated by
artificially tripping one of the alarms.

Reaction time must be minimized when inhabited structures are located immediately downstream of the
dam. When these conditions exist, special procedures may need to be included in the EAP to notify the
specific occupants involved. Local officials should be fully involved in the development of these special
procedures.

The EAP should describe any instrumentation for monitoring the behavior of unattended dams, and
explain how warning systems would be activated. Instrumentation responses should be instantaneous to
facilitate immediate action by operators.

Procedures should be described for providing continuous surveillance for periods of actual or
forecasted high flows. It may be necessary to send an observer to the dam during these periods, and
not rely on the instrumentation alone. It is very important that an observer, with a means of portable
communications, be at the dam when flood conditions or signs of serious structural distress have been
identified.

If a discussion of remote surveillance at the dam is not applicable, that fact should be stated in the EAP.

6-28 NOVEMBER 1998

B. Response During Periods of Darkness. Discussion in the EAP of the response to potential or actual
emergency conditions during periods of darkness should be addressed.

Actions to be taken to illuminate the spillway, operating deck, or observation of distressed areas of the
dam, and other actions that will facilitate the operation of gates or other emergency equipment should
be described.

Any special procedures for contacting or notifying the proper personnel, local officials, or others during
a power failure should be described.

The expected response time for verifying an emergency and implementing the EAP should be discussed
in detail.

Any other special instructions for the dam operators or local officials should be included.

C. Access to the Site. The description of access should focus on primary and secondary routes and
means for reaching the site under various conditions (e.g., foot, boat, helicopter, snowmobile, etc.).
Also discuss in detail the expected response (travel) time. Special attention should be given to access if
the main access road crosses the downstream channel and could be closed by flood waters.

D. Response During Weekends and Holidays. Discussion of emergency response during weekends
and holidays should be included in this section.

The actions to be taken should be described in detail. Actions should be based on the dam tender
schedule for attendance during this period.

Any special procedures for contacting or notifying personnel should be described.

E. Response During Periods of Adverse Weather. Discussion of emergency response under adverse
weather conditions should be included.

The actions to be taken should be described in detail. Action should be based on whether the dam is
attended or unattended.

Methods of access to the site (e.g., foot, boat, snowmobile) should be described.

The expected response time should be discussed in detail.

Any other special instructions for the dam operators or local officials should be listed.

6-29 NOVEMBER 1998

F. Alternative Systems of Communication. The description of the availability and use of alternative
communications systems at the site should be included.

Alternative channels of communication to be used in case of failure of the primary system or failure of
other systems immediately available should be listed.

Proper procedures for activating the alternative channels of communication should be described.

Any other special instructions should be included.

G. Emergency Supplies and Information. There are certain planning and organizational measures that
can help the licensee and local officials manage emergency situations more safely and effectively. These
measures include stockpiling materials and equipment for emergency use and coordinating information.
Also, alternative sources of power for spillway gate operation and other emergency uses should be
provided. The EAP should list the location of each power source, its mode of operation, and if
portable, the means of transportation and routes to be followed. The EAP should include the name and
day/night telephone numbers of each operator or other responsible person.

If any of these measures apply, they should be discussed in the EAP. Specific types of information to
include when describing these emergency supplies and information follow.

1. Stockpiling Materials and Equipment. Where applicable, document:

Materials needed for emergency repair, and their location, source, and intended use. Materials
should be as close as possible to the dam site.

Equipment to be used, its location, and who will operate it.

How the operator or contractor is to be contacted.

Any other people who may be needed (e.g., laborers, engineers), and how they are to be
contacted.

Also include any other special instructions. If stockpiling of materials and equipment is not
applicable to your dam, that fact should be stated in your EAP.

NOTE: For each applicable item, include specific contacts and their business and non-
business means of communication.

2. Coordination of Information. Where applicable, describe:

6-30 NOVEMBER 1998

 The need for coordination of information on flows based on weather and runoff forecasts,
failure, and other emergency conditions. Describe how the coordination is achieved and the
chain of communications, including names and day/night telephone numbers of responsible
people. Coordination with the National Weather Service (NWS) or other appropriate agency
is recommended to monitor storms, river stages, and flood waves resulting from a dam break.
The NWS or other appropriate agency may also be able to supplement the warnings being
issued by using its own communication system.

Additional actions contemplated to respond to an emergency situation or failure at an

unattended dam. Include periods of darkness, inclement weather, and non-business hours.

Actions to be taken to lower the reservoir water surface elevation, if applicable. Describe
when and how this action should be taken. If not applicable, that fact should be stated in the
EAP.

Actions to be taken to reduce inflow to the reservoir from upstream dams or control structures.
The EAP should provide instructions for operators or other persons responsible for contact
with other owners on when and how these actions should be taken. If such actions do not
apply, that fact should be stated in the EAP.

Actions to be taken to reduce downstream flows, such as increasing or decreasing outflows

from downstream dams or control structures on the waterway on which the dam is located or
its tributaries. The EAP should provide instructions for operators or other responsible persons
on when and how these actions should be taken. If such actions do not apply, that fact should
be stated in the EAP.

Also describe any other appropriate actions to be taken. If coordination of information on

flows is not applicable, that fact should be stated in the EAP.

3. Other Site Specific Actions.

Describe any other site-specific actions devised to moderate or alleviate the extent of possible
emergencies.

VII. Inundation Maps

Inundation maps are necessary and should be developed by the licensee in coordination with the
appropriate state and local emergency management agencies. Since those agencies will rely heavily on
the maps during an emergency it is important that the maps contain information required by those
agencies. The inundation map must show the following:

6-31 NOVEMBER 1998

The antecedent flow conditions on which the maps are based should be identified. Many local
emergency management and response organizations request maps showing both a "sunny day" failure
condition and a flood failure condition to show the expected extremes in peak water surface elevations,
travel times and distances downstream between the two scenarios. (For a further discussion see
Section VIII.C. - Investigation and Analyses of Dam Break Floods).

Describe how the inundation boundaries were plotted. As a minimum, show on the map and/or in a
table the peak discharge, maximum inundation elevation and the travel time (in hours and minutes) of the
leading edge and peak of the dam break flood wave to critical locations.

The map should be developed at a scale sufficient to be used for identifying downstream inhabited
areas within the area subject to possible danger. Inundated areas should be clearly identified. It may
be appropriate to supplement the inundation maps with water surface profiles showing the elevation
prior to failure, the peak water surface elevation after failure, and the location of structures at critical
locations.

A narrative description of the areas affected by the dambreak can be included to clarify unusual
conditions. It should describe the specific area threatened and include information on the size and
depth of expected flooding relative to known landmarks and historical flood heights. Whenever
possible, major streets, railroads, and other well known features should be used, using local names or
terms.

The best available topographic map should be used. The expected inundation following the assumed
failure should be delineated on the map. The lines delineating the inundated area should be drawn in
such thickness or form (solid line, dashed line, dotted line) as to readily identify the inundation limits as
the main features of the map but not bold enough to obliterate houses or other features which are to be
shown as being inundated by the flood waters. Clarity is important. When plotting inundation limits
between cross sections used for analysis, the lines should reasonably reflect the change in water levels
with consideration given to topographic patterns and both natural and manmade features. When
inundation lines enter the area of an existing lake or reservoir, they should be so drawn as to represent
an increase in the water level of such lake or reservoir. Should this increased water level overtop the
dam, the appropriate inundation lines should be drawn downstream of such dam to represent expected
inundation in the downstream channel up to a point where an increase in water level will no longer
represent danger to life or property. The area between the inundation lines representing the water level
may be shaded and or colored to distinguish the area of inundation. Care should be taken to select a
shading or colors which will not obliterate the background information shown on the map.

The accuracy and limitation of the information supplied on the inundation maps and how best to use the
maps should be described. Since local officials are likely to use the maps for evacuation purposes, a
note should be included on the map to advise that, because of the method, procedures, and
assumptions used to develop the flooded areas, the limits of flooding shown and flood wave travel times

6-32 NOVEMBER 1998

are approximate and should be used only as a guideline for establishing evacuation zones. Actual areas
inundated will depend on actual failure or flooding conditions and may differ from areas shown on the
maps. The licensee should review the inundation maps with the local jurisdictions to explain the
contents of the inundation maps. The licensee and local jurisdictions should discuss the mapping and
any suggested modifications should be made so that both parties fully understand the information
contained on the maps.

If inundation maps are to be shown on several pages, a map index should be included to orient the
individual pages.

Inundation maps should be updated periodically to reflect changes in downstream areas.

Include any other pertinent information as a result of coordination with the appropriate emergency
management authorities. Emergency management agencies may request the inundation maps highlight
evacuation routes and emergency shelters.

VIII. Appendices

Following the main body of the EAP (the basic EAP) an appendix section should be included that
contains information that supports and supplements the basic EAP.

Listed below are some of the specific topics that should be covered in the appendices accompanying
the EAP:

• Investigation and Analyses of Dambreak Floods

• Plans for Training, Exercising, Updating, and Posting the EAP

• Site Specific Concerns

• Approval of the EAP

Each of the these topics are described in detail below:

A. Investigation and Analysis of Dam Break Floods. The EAP appendices should identify and briefly
describe the method and assumptions selected to identify the potentially inundated areas.

Several factors usually have to be evaluated whenever dam failures are postulated. The type of dam
and the mechanism which could cause failure require careful consideration if a realistic breach is to be
assumed. Size and shape of the breach, time of breach formation, hydraulic head, and storage in the
reservoir contribute to the dam failure hydrograph. Most of the methods for estimating dam break

6-33 NOVEMBER 1998

hydrographs require the selection of these parameters. There are also several available procedures for
routing dam failure hydrographs to determine information on areas inundated by the flood as it travels
downstream.

Several different assumptions on inflow conditions should be made regarding the appropriate conditions
prevailing at the time of a dam failure in order to ensure that the EAP includes all communities that need
to be notified. A "fair weather" (often referred to as “sunny day”) dam failure (reservoir at normal full
pool elevation, normal stream flow prevailing) is generally considered to have the most potential for loss
of human life, primarily due to the element of surprise. A failure at the inflow design flood is considered
to show the upper limit of inundation.

Since emergency management agencies may use the inundation maps to develop their evacuation
procedures, both the "fair weather" breach and a failure during the flood level approaching the inflow
design flood (IDF) should be analyzed and shown on the inundation map. If inundated areas for the
"fair weather" breach and the IDF breach are essentially the same or are too close to be shown
separately on the inundation maps, then a single inundation area for the two breach conditions may be
shown.

Many methods for developing the dam failure hydrograph and routing dambreak flows downstream are
available. Many Federal agencies have developed dambreak computer programs that are available
upon request. They may be obtained from the National Weather Service, Bureau of Reclamation, Soil
Conservation Service, Corps of Engineers, Tennessee Valley Authority, Geological Survey, and
Federal Emergency Management Agency. The dambreak model developed by the National Weather
Service (NWS) is the most widely used and preferred.

Sensitivity analyses are recommended in order to fully investigate the effect of a failure on downstream
areas. Usually, an assumed failure during "sunny-day" conditions results in the worst-case condition for
EAP planning purposes since a failure during flooding conditions, when people are "on-alert", will
usually require no changes to the notification flowchart. When it is not obvious whether the same
notification list would be appropriate for a failure during major flood conditions, the sensitivity analysis
should be performed. The sensitivity analysis should vary key assumptions to identify their effect on
various failure scenarios in order to select the most appropriate failure mode for developing the EAP.
The sensitivity analysis is included for two primary reasons:

1. A sensitivity analysis should be performed when it is not obvious that failure during a "sunny-
day" condition would constitute the worst-case condition. For example, situations occur where
failure during a "sunny-day" condition will not result in a hazard to downstream life and
property. In this situation, a failure during flood flow conditions should be investigated to
determine if notification procedures are necessary in the event of an emergency. In addition, if
a failure during a flood condition will result in a different notification list or priority of notification
from that considered appropriate for a "sunny-day" failure, the EAP should be modified

6-34 NOVEMBER 1998

 accordingly. This condition often occurs in sparsely populated areas. A sensitivity analysis is
necessary in this case to ensure that all structures that could realistically be impacted are
included on the inundation map and all necessary local officials are included in the notification
procedures. However, as indicated above, in many cases only one failure scenario, whether it
be a "sunny-day" failure or a failure during a flood condition, requires analysis since the
notification list and the priority for notification usually remains the same regardless of the
antecedent condition investigated. In all cases, practical considerations should govern in
conducting dambreak analyses since the ultimate goal is to develop the best workable EAP.

2. A sensitivity analysis is also necessary when a licensee/exemptee/applicant for license desires to

demonstrate that a failure under any foreseeable failure scenario would not constitute a hazard
to life and/or property, and an exemption from EAP requirements may be justified. In
requesting such an exemption, a supporting sensitivity analysis is required.

The need to consider the domino effect should be made on a case-by-case basis. If the assumed failure
of a dam would cause the failure of any downstream dams, the licensee has the responsibility to
consider the domino effect in its routing of the floodwave downstream. The flood wave should be
routed to the point where it no longer presents a hazard to downstream life or property, which includes
downstream dams. Therefore, the owner, after assuming a hypothetical failure of its dam, should make
an engineering judgement regarding the potential for failure of the downstream dams from the flow
condition under consideration or as a result of the failure of the dam being investigated to determine
whether it would be prudent to consider failure of any downstream dams during the routing of the
dambreak flood wave.

B. Plans for Training, Exercising, Updating, and Posting the EAP. Plans should be developed for the
annual training of project operators and other responsible personnel, for conducting periodic EAP
exercises, for ensuring timely updating of the EAP, and for posting the Notification Flowchart. New
personnel should be trained immediately when they become responsible for EAP activities.

1. Training

Training of people involved in implementation of the EAP should be conducted to ensure that
they are thoroughly familiar with all elements of the plan, the availability of equipment, and their
responsibilities and duties under the plan.

Technically qualified personnel should be trained in problem detection and evaluation and
appropriate remedial (emergency and non-emergency) measures. This training is essential for
proper evaluation of developing situations at all levels of responsibility which, initially, is usually
based on onsite observations. A sufficient number of people should be trained to ensure
adequate coverage at all times.

6-35 NOVEMBER 1998

 A training plan could be included in the appendices to the EAP. Exercises simulating
emergency conditions are excellent mechanisms for ensuring readiness. Cross-training in more
than one responsible position for each individual is advisable in order to provide alternates. A
careful record by roster should be kept of training completed and refresher training conducted.

2. Exercising.

A proposed exercise schedule and the plans for the EAP exercise program should be included
in this portion of the appendices. It should also discuss plans for conducting a critique of the
exercise (both annual drills and periodic comprehensive exercises) and plans for updating the
EAP based on the comments from the critique.

This section should also include a form that can be used to document actions taken during any
actual emergencies.

The state of training and readiness of key personnel responsible for actions during an
emergency should be a part of any exercise to make sure that they know and understand the
procedures to be followed and actions required.

Any special procedures required for night time, weekends, or holidays should be included. The
exercises should involve an annual drill, as well as periodic tabletop and functional exercises.
Testing of remote sensing equipment at unattended dams should be included.

Coordination and consultation with state and local emergency management officials and other
organizations when developing a comprehensive EAP exercise program is important in order to
enhance the realism of the exercises. Their involvement will greatly improve the close
coordination necessary for a successful execution of emergency procedures during an actual
emergency. The exercises should include participation by both the licensee and the affected
state and local emergency management officials. The exercises should be evaluated both orally
and in writing and the EAP should be revised to incorporate the suggested improvements.

3. Updating.

All aspects of the EAP are subject to periodic review and updating in accordance with the
Guidelines and the specific and detailed instructions contained in Section 12.24 (a), (b), and (c)
of the Commission's Regulations. Although a licensee must conduct a review of the adequacy
of the EAP at least once a year, additional exchanges of information between the licensee and
all parties involved with the EAP is necessary. Information exchanges, such as an informal
phone call, will assure that personnel changes, phone number changes, or changes in emergency
response duties are promptly detected with interim changes made to the EAP. All parties
should be notified of any changes to the EAP. During the yearly review, the licensee should

6-36 NOVEMBER 1998

assure that project personnel are familiar with site-specific concerns identified in the EAP and
determine if any new developments or other changes suggest that revisions should be made to
the current EAP (including inundation maps). It is imperative that the licensee furnish the
Regional Director and all other holders of the EAP updates to the EAP immediately upon
becoming aware of necessary changes to keep the EAP workable. This includes revisions
when phone numbers and/or names change for Regional Office contacts.

The licensee must also annually furnish the Regional Director with a statement that the EAP has
been thoroughly reviewed and the date it was last tested, with inclusion of any needed revisions
and updates or a statement that no revisions and updates are needed. Provide all plan holders
copies of all revisions. Mark pages "Revised MO/DA/YEAR" and highlight revised material.

6-37 NOVEMBER 1998

4. Posting of the Notification Flowchart.

An up-to-date copy of the Notification Flowchart should be posted in prominent locations at

the dam site and local emergency operations center (essential for unattended dams), as
appropriate.

The flowchart should be posted at appropriate phones and radio transmitters at the dam,
powerhouse (if applicable), and any other desirable locations. The locations of the posted
flowcharts should be indicated in the EAP. Posting requirements for cellular phones present a
unique problem. If cellular phones are to be used in an emergency, all users should be familiar
with the locations of the flowchart or "pocket" sized, or other convenient form, flowcharts
should be carried by all cellular phone users.

A copy of the complete, up-to-date EAP should also be available to personnel at the dam and
to local officials. The location of each copy should be stated in this section of the EAP.
Consideration should also be given to having a copy of the EAP at the residences of key
personnel.

C. Site Specific Concerns. Each dam and downstream area is unique. As a result, each EAP is
unique. This section of the appendices should provide a discussion of any site specific concerns that
provide valuable information affecting the EAP. The EAP should emphasize where appropriate
structural drawings, flood data, etc. are maintained on-site. Quick access to this information is crucial
during emergency events.

D. Documentation.

1. Provide the most recent documentation of consultations with Federal, State and local agencies,
including public safety and law enforcement bodies. Only the most recent documentation
should be maintained in the EAP. Copies of the actual documentation sheets should be
submitted to the Commission. All other copies of the EAP need only contain general
statements pertaining to the documentation (e.g. a list of agencies involved, a statement that up-
to-date documentation is on file, a statement that necessary coordination meetings have been
held, etc.). Provide letters of acknowledgment from the contacted agencies.

C Letters should indicate that each agency involved understands its responsibility for
alerting and/or evacuating the public in those areas within its jurisdiction.

C Documentation should be updated on an annual basis to ensure that all participants have
received the updates to the EAP and have the most up-to-date EAP on file.

2. Provide letters or memoranda of contact.

6-38 NOVEMBER 1998

 C Coordination is essential to ensure that local officials responsible for warning and
evacuation of the public comprehend and accept their individual and group
responsibilities. Participation in the preparation of the plan will enhance their
confidence in the plan and in the accuracy of its components. Coordination will provide
opportunities for discussion and determination of the order in which public officials
should be notified, identification of backup personnel, alternate means of
communication, and special procedures for periods of darkness, inclement weather,
non-business hours, etc. Differences in procedures for notification for different
emergency situations should be coordinated prior to finalizing the notification plan(s).

C Advance preparations should include arrangements for such meeting(s) as are

necessary with local and county governments, law enforcement officials, and other
public officials who will be responsible for the warning and the evacuation of the
occupants of the affected areas. The licensee should discuss the accuracy of the
inundation maps or other means used to delineate the affected areas. Times available
for response should also be discussed. Public officials to be notified and their priority
of notification should be established. Special procedures should be developed for
periods of darkness, inclement weather, and non-business hours.

C All positions critical to the execution of the emergency action plan should be covered
24-hours a day, 7-days a week. Alternative or backup personnel should be identified
for all public officials to be notified. Alternative means of communication should be
identified.

C Describe the coordination efforts. Include all letters directed by you to agencies or
others and memoranda of meetings or conferences held.

E. Approval of the EAP. The EAP should include a section that is signed by all parties involved in the
plan, where they indicate their approval of the plan and agree to their responsibilities for its execution.
Including approval signatures helps to assure that all parties are aware of and understand the EAP and
agree to their assigned roles, should an emergency occur.

6-4 EAP EXERCISES

6-4.1 General

An annual drill (exercise) is required to be conducted to test the state of training and readiness of key
licensee personnel responsible for actions during an emergency. Therefore, more comprehensive, in-
depth exercises of EAP procedures will periodically be required in order to include active participation
by State and local emergency preparedness agencies and licensee's personnel. The purpose of an in-
depth exercise designed to test an EAP is to improve operational readiness and develop the

6-39 NOVEMBER 1998

cooperative spirit and coordination required between the licensee and the emergency preparedness
agencies. The exercise will not only reveal the strengths and weaknesses in the EAP itself, but may
also, among other things, reveal gaps in available resources, improve coordination requirements, clarify
roles and responsibilities, improve individual performance, and achieve public recognition of the EAP.
The licensee should also request that the local emergency response agencies notify them immediately of
any changes of key personnel.

6-4.2 Annual Drill

Each licensee is required to conduct an annual exercise known as the "in-house drill" to test the state of
training and readiness of key licensee personnel responsible for actions during an emergency to ensure
that they are fully cognizant of the procedures and actions required during an emergency. The
licensee must conduct an annual drill for each of its EAPs. It is acceptable for an
annual drill to concurrently test the EAP for several dams when an overlap in
notification is involved. Regional personnel should ensure that these drills are
completed. If a licensee does not complete its annual drill, the Regional Director
should request that a firm date be submitted for the next and subsequent drills at
least 60 days prior to that drill.

The drill should simulate an emergency condition. The licensee staff member responsible for conducting
the test should first develop a realistic scenario under which the EAP would be implemented.
Preferably, the scenario should be varied from year-to-year. Any special procedures required for
nighttime, weekends, and holidays should also be considered when developing the scenario. Testing of
remote sensing equipment at unattended dams should be included. In addition, different levels of
notification of internal hierarchy should be tested each time an emergency drill is conducted.

Coordination and consultation with local government, law enforcement officials, and other organizations
involved is desirable. This will enhance the realism of the drill and will ensure that telephone numbers
on the notification list are accurate. Licensees should be encouraged (not required) to consider the
merits of a surprise in-house drill versus a planned one. Of course, the licensee at the time it
implements a "surprise" drill should advise its employees that the drill is a test and not an actual
emergency. While a planned drill will allow participants to rehearse their roles in the EAP, a surprise
drill can be more educational, because of the probability of exposing more basic flaws in the EAP.

Immediately following the drill, the licensee should assess (evaluate) the results with all involved parties.
The responses to the emergency scenario at all levels should be reviewed. The purpose of the critique
is to identify deficiencies in the EAP, including notification, priorities, responsibilities assigned, etc.

After the critique has been completed, the EAP should be revised and the revisions disseminated to all
involved parties. The licensee must furnish the Regional Director within thirty days of
the date of the drill, a critique of the drill and any revision or updates to the EAP as

6-40 NOVEMBER 1998

a result of the drill. The critique must also include a list of the lessons learned. A
report should be filed with the Regional Director even if the evaluation reveals no
deficiencies.

6-4.3 Comprehensive Exercises

The involvement of State and local emergency preparedness officials in an exercise to test an EAP is
necessary to perfect the close coordination and cooperation that is necessary for a successful execution
of an EAP in an actual emergency.

Briefly stated, the licensee has the responsibility to provide warning and notification of a dam failure to
the agencies, and the emergency preparedness agencies have the responsibility to provide for
evacuation of the affected areas. The licensee must assume that if an emergency preparedness agency
is notified of an emergency, the agency will respond appropriately. However, to ensure that everyone
carries out their responsibilities, coordination, training and practice are necessary ingredients of a
workable EAP.

The comprehensive exercise is intended to bridge the gap between the warning and notification issued
by the licensee and the evacuation response of the emergency preparedness agencies. It is important
that both the licensees and emergency preparedness agencies understand the total picture so that both
parties can make the necessary changes to their plans. This will result in a smooth, confident response
should an emergency situation occur.

An exercise, as defined by FEMA in its Guide to Emergency Management Exercises, SM

170.2, January 1989, is "an activity designed to promote emergency preparedness; test or evaluate
emergency operation, policies, plans, procedures or facilities; train personnel in emergency management
duties; and demonstrate operational capability." The comprehensive, in-depth exercise that the FERC
requires is an exercise that tests, among other things, the licensee's warning and notification procedures,
the State and local agencies response to the notification, their knowledge of the EAP inundation maps,
and the cooperative spirit of the licensees and emergency preparedness agencies in a stress-induced
environment.

FEMA has identified five elements, or types of exercises, that constitute an exercise program, with each
one building on the concepts of the previous exercise. It is not a requirement that every exercise
program include all five exercises. However, it is advisable to build an exercise program upon
competencies developed from simpler exercises to achieve greater success with the more complex
exercises. This means that emergency exercises should be developed and conducted in an ascending
order of complexity. Also, sufficient time should be provided between each exercise to
learn and improve from the experiences of the previous exercise prior to conducting
a more complex exercise. The five exercise types, listed from simplest to most complex are:

6-41 NOVEMBER 1998

A. Orientation Seminar

This exercise is a seminar that involves bringing together those with a role or interest in an EAP (i.e.
licensee and state and local emergency management agencies) to discuss the EAP and initial plans for
an annual drill or more in-depth comprehensive exercise. The seminar does not involve an actual
exercise of the EAP. Instead, it is a meeting that enables each participant to become familiar with the
EAP and the roles, responsibilities, and procedures of those involved. An orientation seminar can also
be used to discuss and describe technical matters with involved, non-technical personnel.

B. Drill

A drill is the lowest level exercise and tests, develops, or maintains skills in a single emergency response
procedure. The in-house drill performed by licensees to test the validity of telephone numbers and
operator's responses is a drill. A drill is considered a necessary part of ongoing training.

C. Tabletop Exercise

The tabletop exercise is a higher level exercise than the drill. The tabletop exercise involves a meeting
of the licensee and the state and local emergency management officials in a conference room
environment. The format is usually informal with minimum stress involved. The exercise begins with the
description of a simulated event and proceeds with discussions by the participants to evaluate the EAP
and response procedures and to resolve concerns regarding coordination and responsibilities.

D. Functional Exercise

The functional exercise is the highest level exercise that does not involve the full activation of the
licensee and state and local emergency management agency field personnel and facilities or test
evacuation of residents downstream of the dam. It involves the various levels of the licensee and state
and local emergency management personnel that would be involved in an actual emergency. The
functional exercise takes place in a stress-induced environment with time constraints and involves the
simulation of a dam failure and other specified events. The participants "act-out" their actual roles. The
exercise is designed to evaluate both the internal capabilities and responses of the licensee and the
workability of the information in the EAP used by the emergency management officials to carry out their
responsibilities. The functional exercise also is designed to evaluate the coordination activities between
the licensee and emergency management personnel. Section 6-4.5 of this chapter lists several functions
that should be included in a functional EAP exercise.

The FERC will periodically be requiring, as a minimum, that a functional exercise of an EAP be
conducted by a selected number of licensees. A particular licensee should plan on being requested to
conduct a functional exercise about once every five years. It should be noted that the functional
exercises will be conducted on a licensee basis. Therefore, a licensee should not be expected to

6-42 NOVEMBER 1998

conduct a test more than once every five years for one of its projects. Unless there is good reason,
licensees will be requested to choose a different project for EAP testing during each five year cycle. Of
course, licensees with one or two projects will have to test the same EAPs.

E. Full Scale Exercise

The full scale exercise is the most complex level of exercise. It evaluates the operational capability of all
facets of the emergency management system (both licensee and state and local emergency management
agencies) interactively in a stressful environment with the actual mobilization of personnel and resources.
It includes field movement and deployment to demonstrate coordination and response capability. The
participants actively "play-out" their roles in a dynamic environment that provides the highest degree of
realism possible for the simulated event. Actual evacuation of critical residents may be exercised if
previously announced to the public.

Both the functional exercise and the full scale exercise are considered to be "comprehensive exercises."

For a detailed explanation of how to develop, conduct, evaluate, and follow-up an effective tabletop,
functional, and full-scale exercise, the licensee should attend the FEMA EAP Exercise Design Course
or the FERC EAP Exercise Design Course, held annually throughout the country. Summary
information on the three (3) higher level exercises has been extracted from the FEMA manuals and is
included below for convenience, with minor changes to address FERC's specific needs:

6-4.3.1 Tabletop Exercise

A tabletop exercise includes low stress, little attention to real-time, lower level of preparatory effort,
and only rough attempts to simulate reality. The focus is on training and familiarization with roles,
procedures, responsibilities, and personalities of the licensee and the emergency management agencies.

The methodology of tabletop exercises is by an open-ended discussion in a meeting format through a

facilitator. The discussion is allowed to be interrupted by questions and participant comments. The
effectiveness is determined by feedback from participants and the impact this feedback has on
evaluating and revising policies, plans and procedures.

There is no utilization of equipment or deployment of resources. Therefore, all activities are simulated
and participants interact through discussion. A "narrative" (or scenario) sets the scene for the simulated
event. It briefly describes what has happened and what is known up to the time of the exercise. A
sophisticated form of the tabletop exercise provides the participants with "messages" as a stimulus for
responses. The "messages" communicate detailed events to the participants as the exercise progresses.
The purpose of the "messages" is to provide sufficient information to the participants so that they will
respond with an action or a decision. New (or updated) messages are interjected throughout the

6-43 NOVEMBER 1998

exercise to evoke responses. The flow of the exercise depends on the quality of the messages and
upon their precise timing in the exercise.

The advantages of a tabletop exercise are that there is modest commitment in terms of time, cost and
resources. It provides an effective method of reviewing plans, implementing procedures and policies,
and it serves as an educational device to acquaint the licensee and key agency personnel on emergency
responsibilities and procedures. It also acquaints licensee and emergency response personnel with each
other on a personal basis.

The disadvantages of a tabletop exercise are that it lacks realism, and does not provide a true test of
participants' capabilities. It provides only a limited exercise of plans, procedures, and participants' staff
capabilities.

A facilitator (or controller) will monitor the pace and flow of a tabletop exercise by introducing the
scenario narrative and messages into play. Facilitators may also include individuals from the licensee's
organization and the emergency preparedness agencies. The facilitator must be able to stimulate
discussion, making sure that no one participant dominates the exercise. The facilitator leads the
conduct of a tabletop exercise and makes sure every participant responds to at least one message
during the exercise.

The process of developing a tabletop, functional, or full scale exercise involves similar steps which
largely differ in the level of complexity and realism desired. This involves assessing the needs for
an exercise, defining the scope of an exercise, writing a statement of purpose, writing
objectives, writing a narrative, and writing problem statements (tabletop) or messages
(functional and full scale). Greater realism and attention to detail are necessary for the development of
the functional exercise than for the tabletop, with an even greater amount of detail needed for the full
scale exercise. These steps are briefly discussed below, with allowances needing to be made as to the
level of detail desired in relation to the type of exercise being designed.

The first step in the process of developing an exercise is to assess the needs of the exercise by
identifying those areas most in need of an exercise.

In defining the scope of an exercise, six components need to be addressed in the developmental
stage: (1) the types of licensee and emergency management agency activities or procedures you want
to exercise; (2) the parties to be involved; (3) the kinds of personnel involved; (4) the degree of realism
desired; (5) the hazard or the selection of a high priority problem; and (6) the geographical area where
the problem could occur.

The statement of purpose is then developed. It should clearly and concisely explain why the
exercise is being conducted. At this point, the exercise should be announced, the necessary coordina-
tion should be accomplished, and the date and location should be established.

6-44 NOVEMBER 1998

The next step in developing an exercise is writing objectives; that is, defining what should be
accomplished by conducting the exercise. The needs assessment, scope, and purpose statement should
be examined very closely during objective writing to address expected benefits of the exercise and what
emergency actions are to be exercised. Exercise objectives should be specific and realistic, yet
challenging, results-oriented, and measurable.

The next step is to prepare a narrative. A narrative (or scenario) is a short written story that sets the
scene for the exercise. The job of the narrative is to get the exercise participants into the exercise as if
they were confronting a real situation. Simply put, the narrative is an account composed of a few
paragraphs that provides background information to the exercise participants. The narrative should be
written so that it helps participants to understand the exercise and reflect a sense of concern, urgency,
and excitement.

While setting the scene for the simulated emergency and providing some specific information, the
narrative should NOT provide participants with ALL the information necessary to respond to a
situation. Participants will gather additional details during the exercise as the events unfold. Obviously,
the narrative should NOT suggest possible responses to the simulated emergency.

The last step of exercise development is the preparation of a major sequence of events list, a detailed
sequence of events list, and problem statements (for a tabletop) or messages (for a functional or
full scale exercise). The exercise narrative provides the participants with only a certain amount of
information. The major sequence of events list itemizes the events from the beginning of the
exercise to the conclusion that will require a response by the licensee or the emergency preparedness
agencies. The major events can be developed from the statement of purpose. The detailed sequence
of events lists the details for each major event. The messages or problem statements are
developed from the major and detailed sequence of events list. Details of the exercise are
transmitted to the participants through the messages or problem statements. These provide sufficient
information to the participants so that they will be able to respond with an action or decision.

Once these steps are completed, the exercise can proceed. The flow of the exercise depends on the
quality of the messages and upon their precise timing in the exercise. A facilitator (for a tabletop) or a
controller (for a functional or full scale) monitors the flow of the exercise and supervises the input of
messages. A simulator creates the simulated emergency by sending pre-scripted messages and/or
spontaneous messages to players.

Once the exercise is completed it is necessary to evaluate the results of the exercise. A realistic
exercise provides the best opportunity to evaluate the emergency action plan and overall preparedness
to operate under emergency conditions. The extent and depth of the evaluation to be undertaken is
determined by the participants. Controllers' and participants' evaluations and observations are required
along with additional analysis by FERC observers.

6-45 NOVEMBER 1998

The purposes of evaluating an exercise are to identify needed improvements in the
EAP, identify needed improvements in the emergency management system and the
licensee's organization, identify needed training/personnel deficiencies, observe
whether the exercise has achieved its objectives, and identify areas requiring
additional coordination.

The outcome of an exercise consists of individual improvement through training and EAP improvement
through follow-up. Without evaluation, needed improvements will not be identified, improvements will
not be made, and the exercise will not be as worthwhile.

Evaluation begins in the early stages of exercise development when the objectives are planned. If the
objectives involve resource allocation by licensees or agencies, then it will be necessary to evaluate
communication between the licensees and agencies and the final resources allocated.

An immediate post-exercise critique should be held, followed by a more detailed

evaluation report. A critique is a debriefing of the participants. It is a time to gather and share
information about what happened during the exercise, to describe what went right, and identify what
went wrong. The critique should be both oral and written.

The oral critique is a group discussion led by one or more controllers. To minimize the defensiveness of
the participants and maximize the sharing of information, the oral critique should be structured to give
each participant an opportunity to share their observations and to encourage the participants to report
on both what went well and what went poorly. Each participant should typically be allotted a
reasonable length of time to present their observations in order to prevent one participant from
monopolizing the discussion. Having both oral and written critiques strengthen the evaluation report
data.

The formal evaluation of exercise performance consists of a brief written report that is based on
observations and recommendations that come out of the critique, as well as the reports of the
designated evaluators. Data needed for an evaluation report include one's own observations, the
participants' debriefing comments, the participants' written critique, comments from controllers and/or
simulators, any subsequent clarification or discussion with participants, and exercise plans, objectives,
expected actions, and procedures. Licensees should prepare the written report.

Follow-up is the final and critical stage of the exercise process and follow-up recommendations are the
purpose of the evaluation report. Follow-up is the process of implementing the recommendations.

6-46 NOVEMBER 1998

6-4.3.2 Functional Exercise

The functional exercise is intended to test or evaluate specific capabilities of the participants. As
discussed in Section 6-4.3.1, the Tabletop Exercise focuses on training and familiarization with roles,
procedures, responsibilities, and personalities. Therefore, the tabletop exercise provides an opportunity
to ask questions about the narrative, messages and the appropriateness of participants' responses. It is
an opportunity to discuss and establish appropriate responses to the situation. The functional exercise
tests and evaluates the reactions and responses of the participants in a stress induced environment with
time constraints. Whereas a tabletop exercise provides opportunities throughout the exercise to stop
and discuss what actions and responses would be appropriate, the functional exercise is a time
constrained test with limited opportunity for discussion. The functional exercise simulates actual
emergency situations and participants' responses without field deployment. Therefore, the exercise
should be conducted with the participants in one location or with the participants located at their own
facilities, with communications through expected emergency communication links. Those responses are
not evaluated until the conclusion of the exercise.

The functional exercise is based on a simulation of an emergency that includes a description of the
situation (narrative), a master sequence of events list (MSEL), a timed sequence of messages, and
communication between participants and simulators. If possible, the licensee should encourage the
activation of the emergency operations center (EOC) at the State or local level, as appropriate, so that
the EOC members can practice a coordinated, effective response in a time-pressured, realistic
emergency situation. Licensee, individual, agency, and system performances are evaluated during the
exercise. A functional exercise can involve policy, coordination, and operational response personnel of
the licensee and the effected community.

A functional exercise is designed to evaluate the response, the organizational, skills, and individual
efforts of both the licensee and local emergency management personnel. Functional areas that can be
tested include policy making, planning, decision-making, communication, coordination of resources,
management of personnel, and implementation of procedures. That is, it can include any function
needed for an efficient response or recovery from an emergency. See Section 6-4.5 for a list of
the five standard functions that should be included as a minimum in the exercise.

Conducted in a real-time environment (although after the initial hour of the exercise, compressed-time
may be necessary), a functional exercise is based on a scenario which comprises a predetermined
narrative, events list, and messages developed by an exercise team of one or more individuals. After
the initial stages, momentum of the exercise is determined largely by spontaneous interaction among
participants and simulators. Scenario-related events and messages of increasing complexity, threat, and
pressure are interspersed in an emergency situation designed to test the participants' skills, knowledge,
awareness, and ability to respond under simulated conditions. The functional exercise is followed by a
critique session that allows participants to evaluate their performance and lessons learned throughout
the training exercise.

6-47 NOVEMBER 1998

The purpose of the functional exercise is to exercise the coordination of the licensees and the
emergency management agencies under simulated conditions that provide realism and stress. The func-
tional exercise brings together the policy, coordination and operational officials of the licensee and
involved emergency preparedness agencies into one area, either a simulated or real EOC. The
functional exercise gives the participants a fully simulated experience of being in a major disaster. The
functional exercise allows one to assess the direction and control of the disaster management; the
decision-making process, communication and information among participants, allocation of resources
and staff; overall adequacy of resources to meet the disaster situation; and adequacy of current policies,
plans, and procedures. The functional exercise also encourages a spirit of cooperation and
coordination between the licensee, the emergency preparedness agencies, and the FERC.

Conducting a functional exercise should be the major goal of every exercise

program. It provides the greatest opportunity to observe, assess, and improve the coordination in
response to an emergency. The reason that a functional exercise is a general goal of an emergency
exercise program is that it offers the opportunity to test a participant's response in a full simulation under
"real-life" conditions. Since the functional exercise is a high-level exercise, it is strongly suggested
that orientation seminars, drills and tabletop exercises be conducted prior to the
functional exercise.

The individuals involved in the functional exercise should be those people who are responsible for the
coordination and implementation of the EAP. They should be those individuals from the licensee and
agencies that would be most active during a disaster.

It is sometimes difficult (because of busy schedules or other commitments) to get policy-level personnel
involved in a functional exercise, but their presence is beneficial. The licensee should attempt to involve
key personnel so that the appropriate level of importance is understood by management and other
personnel. Building the exercise program pays off. If reactions to earlier exercises are good, the
policy-level personnel will be more likely to participate in a functional exercise.

The preparation tasks for a functional exercise include developing specific objectives, developing the
narrative (or scenario), assuring adequate physical facilities, organizing displays and materials, recruiting
and training exercise participants, and planning for the exercise critique and evaluation. The level of
complexity needed for the functional exercise should be commensurate with the anticipated site
conditions and complexity of the notification procedures.

Because these tasks are so varied and dependent upon each other's completion, it is important to plan
this preparation time carefully. Milestones should be established along with responsibilities for each of
the major activities of preparation.

6-48 NOVEMBER 1998

Apart from the actual participants in the functional exercise, there are three roles that representatives of
the licensee and/or emergency preparedness agencies must fill. These are the exercise controller,
exercise simulators, and exercise evaluators.

The controller's responsibilities include monitoring the sequence of events as they unfold, the flow of
messages, the overall conduct of the exercise, controlling the spontaneous inputs by simulators,
coordinating information among simulators, and responding to unplanned situations.

The simulators' responsibilities include sending pre-scripted messages at the scheduled time, responding
to unanticipated actions by participants with spontaneous messages, and maintaining contact with the
controller about the progress of the exercise.

The evaluators have the task of observing the actions and decisions of the participants during the
exercise and contributing, along with the comments of exercise participants, to the formation of an
evaluation report. In particular, evaluators will be looking to see how participants react to the scenario
events and messages. Ideally, there should be an evaluation team with representatives from the
licensee, agencies, and FERC.

As with the critique for the tabletop exercise (Section 6-4.3.1), the licensee's evaluation report and
follow-up to the recommendations in the report are important aspects of the exercise.

6-4.3.3 Full-Scale Exercise

A full-scale exercise adds a field component that interacts with a functional exercise through simulated
messages.

A full-scale exercise is intended to evaluate the operational capability of licensee and agency
participants in an interactive manner over a substantial period of time. It involves the testing of a major
portion of the basic elements existing within emergency action plans and the participants in a stressful
environment. Full-scale exercises test the mobilization of personnel and resources and the actual
movement of emergency workers, equipment, and resources required to demonstrate coordination and
response capabilities.

A full-scale exercise should test a large portion of the expected actions needed by the licensee and
agencies to implement the EAP.

Full-scale exercises add an integration and coordination component to the functional exercise. They do
not substitute for simulation; instead, they complement it. Events and messages may be complex and
detailed. Many of the messages will be pre-scripted and scheduled, while others may be dynamically
input by controllers in response to the flow of the exercise.

6-49 NOVEMBER 1998

The major components of a full-scale exercise include adding the field component to the exercise
process; testing the deployment of seldom-used resources; involving policy, coordination, operational,
and field response personnel and resources; and testing a major portion of emergency action plans,
resources, and capabilities.

Full-scale exercises greatly expand the scope and visibility of the exercise program. As a result, a well-
done field exercise can result in substantial improvement in public attention and credibility. At the same
time, a poorly conducted exercise can create credibility problems for the entire program of emergency
action planning. Full-scale exercises should be the culmination of an exercise development program
that has grown with the capacity of the participants to conduct exercises. This should also include an
ongoing cycle of progressively more in-depth evaluations.

Full-scale exercises draw media and community attention to emergency preparedness; teach by doing;
test total coordination, not only among policy and coordination officials, but also field forces; test many
licensee and agency emergency management functions at one time; evaluate cooperation; and point out
physical resource capabilities. They can be a true test of the total emergency management system and
the effectiveness of a specific EAP.

For agencies or local communities, full-scale exercises require considerable preparation and can often
be aimed at practical tests of "first-in" responders, including police, fire, and medical personnel. They
can be used to test triage (dealing with casualties) procedures, on-scene management of resources, and
coordination through field command posts. Careful consideration should be given to selecting the day,
date, and time for any exercise. The inclusion of these types of considerations should be left to the
agencies since they can best assess the benefits and constraints of doing so.

Because a full-scale exercise requires the mobilization of personnel and resources, careful consideration
must also be given to the selection of an exercise site. The primary factor here is one of adequate
space, financial capability, and support.

In any exercise, a real emergency might occur, especially during a lengthy full-scale exercise. In
planning the exercise, both the licensee and emergency preparedness agencies should ensure there are
enough personnel and equipment not involved in the exercise to respond to a real emergency. In some
instances, it may be necessary to stop the exercise.

As with the functional exercise, the controller is responsible for assuring that the exercise starts on
schedule. Simulators and evaluators should keep a log of all significant events. Also, each participant
should log its actions as much as possible. Videotaping the exercise and critique can be beneficial.

A well-designed, full-scale exercise can be used to obtain a great deal of favorable media attention. In
fact, a full-scale exercise of any magnitude will draw media attention whether it is sought or not.

6-50 NOVEMBER 1998

Therefore, it is wise to include the media in any exercise plans. They can be extremely helpful in a
number of ways, and it will increase realism if they are present.

At the conclusion of the exercise, the critique and evaluation report required by the licensee are
important so that necessary follow-up action can be taken.

6-4.4 Role of Licensee

The design of an effective exercise depends on the coordination and cooperation of the licensee, the
FERC, and the emergency preparedness agencies. Ideally, the licensee should chair the exercise. It
may also be appropriate for an emergency preparedness agency representative, agreed to by the
participants, to co-chair the exercise. In other words, the licensee should take the responsibility for
coordinating the design of the exercise and holding the exercise. However, the licensee does not
necessarily have to serve as the Controller of the exercise, with the responsibility to monitor the flow of
the exercise and supervise the input of messages.

As chair, the licensee should oversee the development of the exercise. It has the responsibility to
coordinate the schedule for the actual exercise, including the orientation seminars, table top exercises,
etc. The licensee should advise the FERC Regional Director of the plan and schedule
for the exercise and date of each aspect of the exercise.

The functional or full scale exercise should test both the internal and external actions in response to
implementation of the EAP. The primary function of the exercise is to test the response to
a dam failure. The licensee, as chair, should ensure that this remains the primary
focus of the exercise.

Therefore, prior to contacting the State and local agencies to coordinate an exercise,
the licensee should establish its goals. It should clearly set forth for the agencies the
aspects of the EAP that it wants to examine and the level of involvement of the State
and local agencies. However, the local agencies may introduce other emergencies that could occur
at the time of the dam failure to test their capabilities to respond to several incidents at one time.

The FERC will provide assistance, as necessary. The FERC will participate in the exercise as an
observer and will participate in the follow-up critique of the exercise.

6-4.5 FERC Goals and Objectives

A full scale exercise of a simulated emergency is the ideal approach to evaluate every participant's
knowledge, understanding, and reaction to a dam failure event. There are practical
considerations that will indicate that full scale exercises may not be appropriate in
all cases. Due to the complexity and expense in terms of personnel and equipment

6-51 NOVEMBER 1998

committal, the full scale exercise will normally be executed at the option of the
licensee unless peculiar circumstances of a particular project or lack of confidence
in previously performed lower level exercises warrant a request by FERC staff for a
full scale exercise. Therefore, the FERC's goal is to have licensees periodically
conduct a functional exercise of an EAP.

The FERC will focus primarily on high hazard dams in identifying those projects that warrant a
functional exercise. Since a drill is one step in developing the higher level functional exercise, the annual
drill should be incorporated into the development of a functional exercise. No separate drill would be
required in any year when a functional exercise takes place.

Before a functional exercise can be conducted, it is necessary to lay the groundwork for that exercise.
This may require one or more orientation seminars, a drill to verify telephone numbers, and a tabletop
exercise, all of which should be scheduled before the planned date for the functional exercise so that
there is adequate opportunity to evaluate and improve the EAP. It is suggested that each of these
elements be held before the functional exercise is conducted.

Each EAP is unique. Thus, each exercise must be tailored to the EAP being tested. For example,
several unique applications to a dam failure event include the verification of failure, the moving or
expanding nature of the area in danger, the impacts on timing, the disruption of transportation, areas that
will become isolated due to flooding, alarms and sensors to detect a dam failure emergency, and the
concern for transients and recreationists. Other complications could include the extent of flooding
depending on the conditions at the time of failure, power and communication outages, and failure during
times of darkness and on weekends or holidays. In addition, there are site specific concerns and
complications that should be considered. The FERC's objective is to ensure that EAPs are periodically
reviewed and that each EAP is workable in an actual emergency.

The five standard functions or capabilities of the emergency preparedness agencies should be included
in an exercise. Therefore, when approaching agencies with the concept of a functional exercise to test
an EAP, the licensee should advise the State and local agencies that it would like the exercise to focus
on at least the following areas:

A. Alert, Notification, and Warning

This tests the communication system, the primary and/or alternate back-up systems, the messages to
determine if they are appropriate and clearly understood. It verifies the names and phone numbers on
the notification list and their order of priority. Remote sensing equipment should be tested at
unattended dams at the start of a functional exercise.

B. Direction and Control Function

6-52 NOVEMBER 1998

This tests and evaluates the emergency operations capability and timely response in a stressful
environment. It includes the response to health problems, fire, downed power lines and loss of life,
including drownings.

C. Evacuation

This is a key issue in the exercise as it tests the participants' understanding of the inundation maps.
Experience indicates the inundation boundaries and the road names thereon may not always be clear
and fully understood. Maps are often revised as a result of the exercise.

6-53 NOVEMBER 1998

D. Shelters

This reveals those shelters that should not be used because they are in the flood plain or access to the
shelters is affected by transportation through the inundation area.

E. Public Information

This tests the capability to issue accurate information for a dam failure event.

The licensee, in discussing these five areas with the State and local emergency preparedness agencies,
should provide the agencies with opportunities to identify other areas they believe should be exercised
to evaluate their effectiveness to respond to situations unique to a dam failure situation.

6-4.6 Results from an Exercise

The FERC has identified five major results that should be achieved through an exercise:

A. Develop a Spirit of Cooperation

This is to include the licensee, the State and local emergency preparedness officials, and the FERC.
Without a cooperative spirit, the EAP program will not be as successful.

B. Exchange of Knowledge

The licensee, the FERC, and the State and local officials will help each party to understand their
individual responsibilities and capabilities. It will also provide the opportunity to ensure that all parties
clearly understand the EAP, particularly critical matters such as the data presented on the inundation
maps. The exercise process should also reveal deficiencies in resources and information available to
the licensee and the state and local agencies.

C. Revision to EAPs

An exercise will most likely reveal areas of the EAP that require modification. This should reveal the
strengths and weaknesses of the EAP, including specified internal actions, external notification
procedures, and adequacy of other information, such as inundation maps.

6-54 NOVEMBER 1998

D. Expenditures

The cost to develop and conduct an exercise, as well as any follow-up action, should be kept to a
minimum. Innovative ideas and cost-effective implementation of an exercise, rather than major
expenditures, are the intent. Generally, the participants will be expected to pay for their costs in
participating in the exercise. If problems arise regarding funding, the Regional Director should be
contacted for guidance.

E. Written Critique

As one of the follow-up requirements to the exercise, a written critique, in the form
of a formal report, should be prepared by the licensee and submitted to the FERC
within sixty days of completing the exercise. See Appendix VI-D for the format that
should be followed. Comments from the licensee and the State and local participating agencies
regarding their respective participation in the exercise should be included in the critique. The critique
should document and evaluate the various aspects of the exercise, including the timeliness of responses
and areas of concern. It should include observations and recommendations that result from the
exercise, the debriefing comments, the participants' written critiques, any subsequent clarification or
discussions, and planned follow-up action with a plan and schedule. The report does not need to be
elaborate; it should be clear and concise in the presentation of the information required. The
licensee's report should also include a page summarizing the critique comments and
the lessons learned by both the licensee and the participating State and local
agencies and a plan and schedule to make the necessary changes.

It should be remembered that the purpose of the exercise is to identify areas for improvement of the
EAP. The licensee will not be held accountable for shortcomings identified exclusively in the state or
local agencies' domain.

6-4.7 Availability of Training

The Division of Dam Safety and Inspections (D2SI) of the FERC offers the "Emergency Action Plan
Exercise Design Course" at various locations throughout the United States several times a year. This
course is tailored for dam owners, and FERC licensees specifically. The course includes an invited
speaker from a FERC-licensed project to provide the "licensee perspective" related to the design of an
EAP exercise. The FERC endeavors to also invite other appropriate agencies, such as the National
Weather Service, state dam safety officials, and local response agency personnel to contribute to the
course instruction. The FERC Regional Offices should be contacted for availability of this course.

The "EAP Exercise Design Course for Dam Owners" developed by FEMA is sufficiently generic in
nature so that the knowledge learned about tabletop, functional, and full-scale exercises can be useful
for developing an EAP exercise for a simulated dam failure. The course is given nationwide and is

6-55 NOVEMBER 1998

conducted by FEMA in partnership with State emergency management offices. Licensees, as well as
Regional Office personnel, are encouraged to attend the course. Licensees should be encouraged to
suggest to local officials that they also participate in the course.

The State Training Office in each state periodically conducts an Exercise Design Course. It is
suggested that licensees avail themselves of this training. If it is not available in its home State, the
licensee should explore the availability of the course in other states. The State Office of Emergency
Management should be contacted for dates and other information on the "Exercise Design Course".
The FEMA training is currently offered free of charge.

Another source of "hands-on" training is to attend one, or more, tabletop, functional, or full-scale
exercises. As licensees develop and conduct their exercises, they are encouraged to invite other
licensees as observers. As a licensee observes an actual exercise, it may identify deficiencies in its own
plans and will be able to make improvements before it holds its own exercise.

6-4.8 Licensee Initiative

The licensee will have to rely on the State and local emergency preparedness agencies to respond to
notification that an emergency has occurred. Therefore, licensees should take the initiative
to hold periodic functional or full scale exercises with the appropriate agencies
rather than waiting for a FERC letter requiring them to do so. After a functional
exercise is undertaken and the licensees and agencies make changes to the EAP, although it is not a
FERC requirement, it may be advisable to conduct a follow-up functional exercise the
next year or as soon as practicable to verify that the changes to the EAP were
adequate.

6-5 Radiological Emergency Response Plan

Each owner of a hydroelectric project under jurisdiction of the Federal Energy Regulatory Commission
located within a 10-mile radius of a nuclear plant licensed to operate shall prepare a radiological
emergency response plan to be implemented in the event of a severe accident or incident resulting in the
release of radioactive materials. A plan is required if the 10-mile radius includes any project structures
such as the dam or powerhouse that are used in changing water flows, or project facilities that would be
affected by radioactive materials in such a manner that would interfere with project operations. The
plan will be a supplement to the Emergency Action Plan and made a part thereof. It should contain, but
not necessarily be limited to:

A. Detailed procedures for: The evacuation of power plant personnel when advised or directed to
do so by the appropriate State or local government official; setting of gate openings;
continuation, curtailment or cessation of generation; coordination with, and notification of,
customers, power pools, and other interconnected power suppliers; advance coordination with

6-56 NOVEMBER 1998

 operators of upstream and downstream reservoirs; and/or other actions as considered
appropriate.

B. A list of State and/or local government officials who are responsible for notification of
hydroelectric project personnel that nuclear accident or incident is developing (or has
occurred). This part of the plan should specifically identify the State or local government
officials responsible for notifying individual(s) in the hydroelectric power plant owner's
organization. It should also include provisions for keeping the owner's key personnel currently
informed on the developing situation to allow timely action or response at the affected
hydroelectric project. This portion of the plan should identify, if other than the officials noted
above, the State or local government agency representatives authorized to direct or advise
implementation of action, such as evacuation of the area, or other appropriate action.

C. Notification plans should be developed for alerting the following concerned individuals of
proposed plan implementation. Reference can be made to the notification procedures
contained in the main body of the emergency action plan if appropriate.

1. Local, State, and Federal government officials, including the FERC Regional Director or
alternate.

2. Operators of water-related facilities.

3. Residents and owners of properties that could be endangered by the change in project
operation.

4. Supervisors and other company officials.

The Radiological emergency response supplement to the emergency action plan shall be posted with the
main body of the emergency action plan in a prominent location accessible to operating and supervisory
personnel. Such personnel shall be familiar with their responsibilities under the plan. Training of these
personnel shall be conducted to assure adequate and timely performance of their duties in the event of
an emergency.

As with the other parts of the emergency action plan, all aspects of the plan are subject to continuous
review and updating. At least once a year, a comprehensive review shall be made of the plan. Any
revisions shall be made after consultation with Federal, State, and local agencies, and electric power
producers and users, as appropriate. The need for an update shall be reported to the Regional
Director no later than December 31, of each year.

The affected owner will be requested to file a plan no later than 3 months after the date of issuance of a
license to operate a nuclear plant.

6-57 NOVEMBER 1998

If the Regional Director determines that an emergency action plan is not required for the hydroelectric
project, the radiological supplement shall, nevertheless, be filed. Evidence of coordination with the
State or local director of civil defense, or the appropriate official responsible for emergency
preparedness, should be obtained and forwarded with the plan. Three copies should be submitted to
the Regional Office.

6-6 EAP at a Government Dam

When a project is located at a Federal dam, the licensee is to cooperate with the appropriate Federal
agency in any emergency action planning which would provide procedures to be followed the case of
an accident to or failure of water retaining structures or other structures under Commission jurisdiction
that may affect the integrity and/or operation of the Federal project. Therefore, a documented
procedure must be prepared for notifying the appropriate representatives of the Federal agency of an
emergency and should ensure that the operating personnel are familiar with these procedures. The EAP
is to include the requirement that the Commission's Regional Director is notified of the occurrence of an
emergency situation. Also, the procedure should discuss the licensee/exemptee's responsibilities and
plans to act under any EAP formulated by the Federal agency for that government facility. Three
copies of the procedure for notifying the Federal agency as well as a written statement, verified in
accordance with Section 12.13 of the Commission's regulation, indicating that the licensee/exemptee
will cooperate in the implementation of that Federal agency's EAP and that it has instructed its operating
personnel on how to respond to an emergency under the Federal agency's plan. The notification
procedure is subject to the requirements for training, exercising, updating and posting described on
pages 6-38 through 6-40; Section VII (Appendix B) of the Guidelines.

6-58 NOVEMBER 1998

6-7 APPENDICES

6-59 NOVEMBER 1998

 APPENDIX VI-A

DAMBREAK BREACH PARAMETERS

NOVEMBER 1998
6-A-1 NOVEMBER 1998
Comments:

1. BR is the average breach width, which is not necessarily the bottom width. BR is the bottom
width for a rectangle, but BR is not the bottom width for a trapezoid.

2. Whether the shape is rectangular, trapezoidal, or triangular is not generally critical if the average
breach width for each shape is the same. What is critical is the assumed average width of the
breach.

3. Time to failure is a function of height of dam and location of breach. Therefore, the longer the
time to failure, the wider the breach should be. Also, the greater the height of the dam and the
storage volume, the greater the time to failure and average breach will probably be.

4. The bottom of the breach should be at the foundation elevation.

5. Breach width assumptions should be based on the height of the dam, the volume of the
reservoir, and the type of failure, (e.g. piping, sustained overtopping, etc.).

6. For a worst-case scenario, the average breach width should be in the upper portion of the
recommended range, the time to failure should be in the lower portion of recommended range,
and the Manning's "n" value should be in the upper portion of the recommended range. If a
worst-case scenario is not used, a sensitivity analysis should be performed to fully investigate
the impacts of a failure on downstream areas since the actual breach parameters will not be
known. The sensitivity analysis will provide an estimate of the confidence limits and relative
differences resulting from varying failure assumptions.

a. To compare relative differences in peak elevation based on variations in breach widths, the
sensitivity analysis should be based on the following assumptions:

1. Assume a probable (reasonable) maximum breach width, a probable minimum time to

failure, and a probable maximum Manning's "n" value. Manning's "n" values in the vicinity of
the dam (up to several thousand feet or more downstream) should be assumed to be larger
than the maximum value suggested by field investigations in order to account for
uncertainties of high energy losses, velocities, turbulence, etc., resulting from the initial
failure.

2. Assume a probable minimum breach width, a probable maximum time to failure, and a
probable minimum Manning's "n" value.

6-A-2 NOVEMBER 1998

b. To compare differences in travel time of the flood wave, the sensitivity analysis should be based
on the following assumptions:

1. Use Criteria in a. 1.

2. Assume a probable maximum breach width, a probable minimum time to failure, and a
probable minimum Manning's "n" value.

Plot the results of both runs on the same graph showing the changes in travel time with
respect to distance downstream from the dam.

c. To compare differences in elevation between natural flood conditions and natural flood
conditions plus dambreak, the sensitivity analysis should be based on the following
assumptions:

1. Route the natural flood without dambreak assuming a maximum probable Manning's "n"
value.

2. Use criteria in a. 1.

Plot the results of both runs on the same graph showing changes in elevation with respect to
distance downstream from the dam.

d. Investigations under both normal and flood flow conditions should be considered, as
appropriate.

7. When dams are assumed to fail from overtopping, wider breach widths than those suggested on
Table 1 should be considered if overtopping is sustained for a long period of time.

6-A-3 NOVEMBER 1998

 TABLE 1
SUGGESTED BREACH PARAMETERS
(Definition Sketch Shown in Figure 1)

Parameter Value Type of Dam

Average width of Breach (BR) BR = Crest Length Arch

(See Comment No. 1)
BR = Width of 1 or Masonry, Gravity
more Monoliths,
usually BR < 0.5 W

HD < BR < 5HD Earthen, Rockfill

(usually between Timber Crib
2HD & 4HD

BR > 0.8 x Crest Slag, Refuse

Length

Horizontal Component of 0 < z < slope of Arch

Side Slope of Brach (z) valley walls
(See Comment No. 2)
z=0 Masonry, Gravity,
Timber Crib

¼<z<1 Earthen (Engineered,

Compacted)

1<z<2 Slag, Refuse

(Non-Engineered)

Time to Failure (TFH) TFH < 0.1 Arch

(in hours)
(See Comment No. 3) 0.1 < TFH < 0.3 Masonry, Gravity

0.1 < TFH < 1.0 Earthen (Engineered,

Compacted) Timber Crib

0.1 < TFH < 0.5 Earthen (Non Engineered,

Poor Construction)

0.1 < TFH < 0.3 Slag, Refuse

Definition: HD - Height of Dam

z - Horizonal Component of Side Slope of Breach
BR - Average Width of Breach
TFH - Time to Fully Form the Breach
W - Crest Length

Note: See page 6-A-1 for definition sketch

6-A-4 NOVEMBER 1998

Comments: See Page 6-A-2 - 6-A-3

6-A-5 NOVEMBER 1998

 APPENDIX VI-B

FEMA "EXERCISE

DESIGN COURSE" MANUALS

NOVEMBER 1998
FEMA Publications - Three publications have been prepared by FEMA that provide specific
information on how to design and conduct an effective exercise. Those publications, which are
excellent source materials and are used in the FEMA sponsored "Exercise Design Course", are titled:

1. Exercise Design Course, "Guide to Emergency Management Exercises", SM

170.2, January 1989

2. Exercise Design Course, "Student Workbook", SM 170.1, January 1989

3. Exercise Design Course, "Exercise Scenarios", SM 170.3, January 1989

6-B-1 NOVEMBER 1998

 APPENDIX VI-C

SAMPLE TITLE PAGE, APPROVAL PAGE

AND

TABLE OF CONTENTS

NOVEMBER 1998
 [Title Page]

EMERGENCY ACTION PLAN

[Name] of Development

Project No. [FERC No.]

National Inventory of Dams No.

Name of the licensee/exemptee/applicant for license:

Address:

Submitted [date]

6-C-1 NOVEMBER 1998

Verification:1

State of [ ],

County of [ ], ss:

The undersigned, being first duly sworn, states that [he, she] has read the following document and

knows the contents of it, and that all of the statements contained in that document are true and correct,

to the best of [his, her] knowledge and belief.

(Name of Person Signing)

(Title)

Sworn to an subscribed before me this [day] of [month], [year].

(Signature of Notary Public or other state

or local official authorized by law to
notarize documents).

1
The verification form is to be completed only by the licensee, exemptee, or applicant for
license that prepared the plan, not by agencies that received copies of the plan.

6-C-2 NOVEMBER 1998

SEAL

6-C-3 NOVEMBER 1998

 Contents of the Plan

Page No.

I. Notification Flowchart

II. Statement of Purpose

III. Project Description

IV. Emergency Detection, Evaluation, and Classification

V. General Responsibilities Under the EAP

A. Licensee Responsibilities

B. Responsibility for Notification

C. Responsibility for Evacuation

D. Responsibility for Termination and Follow-Up

E. EAP Coordinator Responsibility

VI. Preparedness

A. Surveillance

B. Response During Periods of Darkness

C. Access to Site

D. Response During Weekends and Holidays

E. Response During Periods of Adverse Weather

F. Alternative Systems of Communication

G. Emergency Supplies and Information

VII. Inundation Maps

6-C-4 NOVEMBER 1998

VIII. Appendices

A. Investigation and Analyses of Dambreak Floods

B. Plans for Training, Exercising, Updating, and Posting the EAP

C. Site-Specific Concerns

D. Documentation

E. Approval of the EAP

6-C-5 NOVEMBER 1998

 APPENDIX VI-D

FORMAT FOR CRITIQUE

OF EAP EXERCISE

NOVEMBER 1998
 Report on Functional EAP Exercise
Name of Project
Project Owner
Ferc Project Number
National Inventory of Dams Number

I. Purpose of Exercise

II. Date and Location

III. Design of the Exercise

A. Brief description of the physical set-up of the exercise and level of "play"
B. Selection process of participants
C. Expectations

IV. Exercise Critique

A. Summary of oral critique comments (debriefing)
B. Written comments (this should reference copies of participants' comments - refer to
Appendix C)
C. Discuss timeliness of responses during exercise Assessment of Licensee's capability to
notify agencies as necessary and of agencies capabilities to execute timely evacuation
(this is to include a review of necessary coordination for information, etc. between the
licensee and the agencies).

V. Results of Exercise
A. Lessons learned
B. Recommendations
1. Improvements to EAP
2. Ways to improve future exercises

VI. Follow Up Actions

A. Actions to be taken
B. Plan and Schedule

VII. Summary Page of Key Critique Comments, Lessons Learned, and

Recommendations (since this page summarizes the key points, it should
indicate the frequency of each critique comment - this summary should be on
a separate page of the report)

Appendices
A. List of Participants
B. Copy of Narrative and Messages

6-D-1 NOVEMBER 1998

C. Copies of Written Critiques
D. Copies of Other Pertinent Handouts

6-D-2 NOVEMBER 1998

 CHAPTER VII

CONSTRUCTION QUALITY CONTROL INSPECTION

PROGRAM 1

1
This Chapter of the FERC Engineering Guidelines has been prepared under
contract with R & H Thomas, Inc.

JANUARY 1993
 Chapter VII
Construction Quality Control Inspection Program

7-0 Contents

Title Page

7-1 Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1

7-1.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1

7-1.2 Review of QCIP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1

7-2 Quality Control Inspection Program Content . . . . . . . . . . . . . . . . . . . . . . . 7-3

7-2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3

7-2.2 Organization Chart For QCIP Staff . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
7-2.3 Number and Specialties of QCIP Staff . . . . . . . . . . . . . . . . . . . . . . . . . 7-4
7-2.4 Duties, Responsibilities, Qualifications and Authority . . . . . . . . . . . . 7-4
7-2.5 Field Tests and Frequency of Testing . . . . . . . . . . . . . . . . . . . . . . . . . 7-5
7-2.6 Field Laboratory or Commercial Testing Facilities . . . . . . . . . . . . . . . 7-5
7-2.7 Inspection Plan Including Documentation and Reporting . . . . . . . . . . 7-5
7-2.8 Planned Use of Consultants During Construction . . . . . . . . . . . . . . . . 7-6
7-2.9 Schedule of All Major Features of Construction . . . . . . . . . . . . . . . . . 7-7
7-2.10 Erosion Control and Environmental Compliance . . . . . . . . . . . . . . . . 7-7
7-2.11 Construction Inspection Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-7

7-3 Types of Quality Control Inspection Programs (QCIP) . . . . . . . . . . . . . . . 7-7

7-3.1 Category 1A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9

7-3.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
7-3.1.2 Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
7-3.1.3 Organization and Staffing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-9
Other Recommended Practices . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-10
7-3.1.4 Inspection Plan or Field Inspection Guidelines . . . . . . . . . . . . . . 7-10
7-3.1.4.1 Inspection Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-11
7-3.1.4.2 Contractor Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-11
7-3.1.4.3 QCIP Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-11
7-3.1.4.4 Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-12
7-3.1.4.4.1 Daily Inspection Report . . . . . . . . . . . . . . . . . . . . . . . 7-12
7-3.1.4.4.2 Nonconformance Report . . . . . . . . . . . . . . . . . . . . . . 7-12

7-i
 7-3.1.4.5 Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-13

7-0 Contents
(Continued)

Title Page

7-3.1.5 Field Testing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-13

7-3.1.6 Environmental Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-15
7-3.1.7 Construction Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-16
7-3.1.8 Planned Use of Consultants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-16
7-3.2 Category 1B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-16
7-3.2.1 Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-17
7-3.2.2 Organization and Staffing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-17
7-3.3 Category 1C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-17
7-3.3.1 Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18
7-3.3.2 Organization and Staffing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18
7-3.3.3 Field Testing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-18
7-3.4 Categories 2A, 2B and 2C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-19
7-3.4.1 Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-19
7-3.4.2 Organization and Staffing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-20
7-3.4.3 Inspection Plan or Field Inspection Guidelines . . . . . . . . . . . . . . 7-20
7-3.4.4 Field Testing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-20
7-3.4.5 Planned Use of Consultants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-21
7-3.5 Categories 3A, 3B and 3C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-21
7-3.5.1 Organization and Staffing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-21
7-3.5.2 Inspection Plan or Field Inspection Guidelines . . . . . . . . . . . . . . 7-22
7-3.5.3 Field Testing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-22
7-3.5.4 Planned Use of Consultants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-22
7-3.6 Small Construction Not Requiring a QCIP . . . . . . . . . . . . . . . . . . . . 7-22

7-4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-23

7-5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-25

7-ii
 7-0 Contents
(Continued)

Title Page

7-6 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-26

Appendix VII-A Construction Quality Control Inspection

Program Content and Proposed Outline for QCIP

Appendix VII-B Sample Organization Charts and Descriptions

of Duties and Responsibilities of Some Key Personnel

Appendix VII-C Sample Report Forms

Nonconformance Report
Environmental Deficiency Report

Appendix VII-D Sample Materials Testing Schedule and

Referenced Documents

Appendix VII-E Sample Civil Inspection Checklists

7-iii
 Chapter VII

Construction Quality Control Inspection Program

7-1 Purpose and Scope

7-1.1 General

The guidelines presented in this chapter provide staff engineers and geologists
with recommended procedures and criteria to be used in reviewing and evaluating
construction quality control inspection programs for FERC jurisdictional projects. These
guidelines are based on the FERC Regulations, FERC Operating Manual for Inspection
of Projects and Supervision of Licenses for Water Power Projects, and selected Quality
Control Inspection Programs (QCIPs) from each FERC Regional Office. The term
"licensee" in the remainder of this chapter refers to licensee, exemptee or applicant as
appropriate. The review performed by staff will be conducted to ensure that submitted
QCIPs comply with the Commission's Regulations and provide accepted construction
quality control inspection and documentation practices common to the hydroelectric
industry. Quality control has been defined as measuring conformance with the
requirements. 2 In addition it is deciding what will be measured and who will do the
measuring and documentation.

As stated in Section 12.40 of the Commission's Regulations and Article 4 of the

Standard L Forms for licensed projects, during any construction, repair or modification
of project works the licensee must maintain any quality control program that may be
required by the appropriate Regional Director, commensurate with the scope of work and
meeting any requirements or standards set by the Regional Director. Construction may
not begin until the QCIP has been approved by the Regional Director.3 The Regional
Director may decide not to require such a program for relatively minor renovation work;
however other conditions may be required upon review of the design and construction
plans and specifications.

The intent of Section 12.40 is to ensure quality construction. The regulations

require that quality control inspections must be conducted by the licensee, the design
engineer or an independent firm accountable to the licensee, and must not be performed
by a construction contractor or firm accountable to the construction contractor. The

2
Reference 1
3
Reference: 18 CFR, Part 12, Subpart E, Section 12.40
regulations recognize that construction contractors have interests in quality construction
work that lead them to establish their own quality control programs. The construction
contractor is not precluded from performing his own quality control inspections for his
own purposes. As stated in the preamble to Order 122, experience shows that
construction contractors may also have conflicting interests that may lead to neglect of
the quality of work. Because of the potential for conflict of interest, it is important to
provide for independent quality control inspections. 4 The desire for independent quality
control inspections is also evident in the requirement that, if the licensee's personnel are
performing the construction work, the licensee must provide for separation of authority
between construction personnel and quality control personnel.

Because of the varying types and complexities of projects under construction,

considerable engineering judgement must be used in evaluating QCIPs. Projects under
construction vary from new dams with surface or underground powerhouses, extensive
tunnels and spillways to projects with minor modifications such as tailrace scour repairs
and training wall modifications. The various types of construction have been grouped
into three categories and each category has been divided into three types of QCIPs. In
the following sections the categories are defined and examples of acceptable QCIPs for
each category are discussed.

7-1.2 Review of QCIP

The review of a QCIP by staff is to evaluate the adequacy of the QCIP relative to
the complexity of construction. Appendix VII-A contains required contents and a
proposed outline for a QCIP. The contents and outline should be considered when
reviewing a QCIP submitted for approval, keeping in mind the complexity of
construction. The contents of Appendix VII-A are discussed in this and subsequent
sections. Regardless of the complexity of construction, the QCIP should be clear on the
qualification, independence, responsibility, authority, number and specialty of personnel
responsible for quality control inspection. All QCIP reviews must include an evaluation
of the adequacy of erosion control and other measures to protect the environmental
quality of streams and other areas affected by construction. All QCIPs should have a
Materials Testing Schedule that specifies the types and number of tests for adequate
coverage for all materials included in the construction. On the larger and more complex
construction projects that extend over a period of years, the training and periodic
evaluation of quality control inspection staff should be reviewed.

4
Reference: FERC Order 122, Final Rule, Regulations Governing Safety of Water
Power Projects and Project Works, Issued January 21, 1981

7-2
 The necessary qualifications for quality control inspection personnel should be
established in the QCIP and resumes for personnel assigned to a specific function
included in the initial QCIP. At a minimum, resumes should be included for QCIP
personnel who have authority to stop work and personnel who have authority to
recommend stop work to the contractor and their supervisor. Emphasis should be placed
on previous experience, including type of construction and levels of responsibility. This
information should be supplemented on a continuing basis by submittal of qualifications
of personnel actually employed. The qualification standard for each position can be
established from existing standards, such as the ASCE Professional Grade Descriptions, 5
and the ACI Concrete Certification, 6. The qualification standard of education and
relative experience can also be specified in the QCIP.

It is important that personnel responsible for quality control inspection be

independent from personnel responsible for construction, and the responsibility and
authority of the quality control inspection personnel must be clear and specific. This
independence must be maintained for all types of construction, including turnkey design-
build construction and construction where the licensee is not only the designer and
constructor, but also is responsible for the quality control inspection. There must be a
separation of authority between construction personnel and quality control inspection
personnel. The responsibility and authority of the quality control inspection personnel,
e.g. the authority to require changes in construction or stop work, should be specifically
stated in the text and illustrated in the organization chart of the QCIP.

The QCIP should describe the number and specialities of proposed quality control
inspectors. Also, the number of inspectors proposed for each feature of construction,
including coverage of shift work, should be specified. Where full time inspection is not
proposed, the schedule for part time inspection should be described.

It is important that the contractor have an adequate erosion and sediment control
program to prevent environmental degradation of streams during construction. The
program should provide for the necessary inspection and monitoring to ensure that
required protective measures are implemented. If during construction, it is determined
that additional protective measures must be taken, the quality control inspection must
have adequate procedures for instituting the measures. The erosion control plan is
required by a license article and must be included in the QCIP, along with its inspection
requirements, to ensure that adequate reporting is in place.

5
Reference 2
6
Reference 3

7-3
 The QCIP should contain a Materials Testing Schedule that specifies the test method,
standard and frequency of tests for all materials. The Materials Testing Schedule will be
based on the testing requirements, standards and codes that are specified in the contract
plans and specifications.

A training and periodic evaluation program should be established as appropriate for

quality control inspectors in the QCIP. This program is especially important for the
larger and more complex construction projects that extend over a period of years and
where there is a turnover and reassignment of inspection personnel. For less complex
and short duration construction projects, an established training program may not be
required; however, the qualifications of any new or reassigned personnel should be
reviewed.

7-2 Quality Control Inspection Program Content

The QCIP should provide for an adequate and qualified construction inspection
force and should contain detailed information including, but not limited to, the
information contained in Appendix VII-A. In addition to the contents of Appendix VII-
A, consideration should be given to including a construction inspection checklist which
covers specific aspects of construction.

All of the items of the QCIP content in Appendix VII-A are discussed under the
various categories of construction in Section 7-3. The items are discussed in detail in the
suggested outline, with comments on items to be covered and pertinent issues to be
considered, for each category of construction and the QCIP. The following paragraphs
contain a brief discussion on each item listed in Appendix VII-A that pertains to all
QCIPs, regardless of the construction category.

7-2.1 Introduction

The introduction should describe the project and the purpose of the proposed
construction. Background information on the various construction contracts should be
discussed. The general goals of the QCIP should be discussed. The names of the
licensee, designer, construction contractor and organization(s) responsible for QCIP
should be stated.

7-2.2 Organization Chart For QCIP Staff

An organization chart should be prepared for each QCIP. The organization chart
should show the details of the relationships of the licensee, designer, QCIP personnel,

7-4
construction management personnel (if applicable), and the construction contractor. The
organization chart should contain the titles and names of all key personnel known at the
time of submittal. Personnel who have the authority to stop work due to adverse quality
conditions should be identified. Also personnel who have the authority to recommend
stop work to the contractor and to their supervisor should be identified. Appendix VII-B
contains sample organization charts that are discussed in Section 7-3.

7-2.3 Number and Specialties of QCIP Staff

The number and specialties of inspectors proposed for each feature of

construction should be included. The number of QCIP staff and the number of various
specialties should be determined by the type of construction and the construction
schedule. There should be adequate inspection whenever there is construction activity.
Where full time inspection is not proposed for certain personnel, the schedule and
approximate percentage of part time inspection should be described. If a qualified
inspector is proposed to cover more than one area of expertise, it should be demonstrated
that there will not be a conflict in scheduling the construction inspections.

7-2.4 Duties, Responsibilities, Qualifications and Authority

The necessary qualifications for QCIP staff should be established in the QCIP
and resumes for QCIP personnel assigned to specific functions included in the initial
submittal of the QCIP. As stated in Section 7-1.2, at a minimum, resumes should be
included for QCIP personnel who have authority to stop work or authority to recommend
stop work due to adverse quality conditions. The qualification standard can be based on
existing standards established by professional organizations such as ASCE and ACI or
the qualification standard can be specified in the QCIP. The information should be
supplemented on a continuing basis by submittal of qualifications of personnel actually
employed.

The responsibilities of the various organizations involved with the construction,

such as the licensee, designer, construction management organization, QCIP staff, testing
laboratory and construction contractor, should be addressed to provide an understanding
of the necessary coordination and relationship for construction of the project. The
services to be provided by each organization should be clearly delineated and discussed.

The responsibilities, duties and authority of key QCIP staff should be clear and
definitive, and should correlate with the organization chart. The position description
should contain such items as the job title; complete description of all duties and
responsibilities; authority, such as, authority to initiate a nonconformance report,

7-5
authority to stop work and authority to recommend stop work; responsibility relative to
contractor negotiations and scheduling of construction; and reporting authority. During
construction, there should be someone in the field at all times who has the authority to
stop work. There should be a separation of authority between the QCIP staff and
construction personnel. The principal QCIP supervisor in the field should have limited
or no involvement with contractor negotiations, scheduling of construction and cost
justification, except as described for Category 3 construction.

7-2.5 Field Tests and Frequency of Testing

The contract plans and specifications delineate testing requirements for the project
and the standards and codes to which the work will conform. The tests should be
conducted at a frequency which will ensure that elements of the work are in compliance
with the specified standards. In addition to the specifications, the type of testing required
should be addressed in the QCIP. One such example is hydrostatic testing of penstocks.
The FERC requires that hydrostatic or non-destructive testing be conducted on all
pipelines whose failure would result in a hazard to life, property or the environment. The
amount of testing, both hydrostatic and weld testing, should relate to the head and
physical size of the project facilities.

A Materials Testing Schedule and Referenced Documents (relative to the testing)

should be presented in a format similar to Appendix VII-D.

7-2.6 Field Laboratory or Commercial Testing Facilities

The supervision, equipment and location of the materials testing laboratory should
be described. For large construction projects it is common for a fully-equipped materials
testing laboratory to be maintained at the project site. The type of testing to be
performed at the laboratory should be described. If tests requiring special equipment are
to be performed by outside laboratories, the tests and laboratories should be described.

For small construction projects, there may not be a field laboratory at the project
site. Therefore, an independent commercial laboratory may be used for material testing
provided the licensee retains this service under a separate contract with the laboratory. If
an off-site laboratory is used, adequate on-site storage should be provided on an as
needed basis for such items as concrete cylinder molds and curing boxes, and other
required equipment. The names and qualifications of all off-site laboratories should be
provided in the QCIP.

7-6
7-2.7 Inspection Plan Including Documentation and Reporting

The Inspection Plan should be specific in providing guidance to QCIP staff and in
establishing inspection, reporting and documentation procedures. The essential elements
of an inspection plan are inspection criteria, contractor operations, QCIP operations and
documentation. A training program for field engineers and inspectors may also be
included in the inspection plan.

The criteria for inspection of contract work is in the executed contract between the
contractor and the licensee.

Normally, the contractor operates independently from the licensee and is

responsible for providing quality and schedule controls over materials, workmanship and
methods to assure meeting contract requirements.

QCIP staff are responsible for verifying that all contract work is performed in
conformance to contract documents and project procedures.

The purpose of reporting is to document the observation, investigation and

analysis of inspection work. There are numerous types of reports and each project
should use the inspection reporting that is appropriate for the type of construction and the
construction contract. The daily inspection, nonconformance, and environmental
deficiency reports are required for all QCIPs. The daily inspection report provides a
means of recording contractor daily operations. The nonconformance and deficiency
reports are used to identify, report and document all observed nonconformances and their
disposition. Appendix VII-C contains samples of nonconformance and environmental
deficiency reports.

For large and complex construction projects, there should be an established

training and periodic evaluation program for QCIP field engineers and inspectors. For
less complex and short duration construction projects, an established training program
may not be required; however, the qualifications of any new or reassigned personnel
should be reviewed to assure that the individual is fully qualified to oversee this assigned
area of responsibility.

7-2.8 Planned Use of Consultants During Construction

Depending on the size and complexity of construction and the downstream hazard
potential the licensee may be required to retain a Board of Consultants to review the
design, plans and specifications and construction of the project for safety and adequacy.

7-7
The Board should also review the initial QCIP and comment on any changes that are
considered necessary.

Also, certain projects may require special consultants such as grouting,

instrumentation and blasting experts. The qualifications and scope of work of the special
consultants should be included in the QCIP.

7-2.9 Schedule of All Major Features of Construction

The preliminary construction schedule should be included in the QCIP submitted

for approval. The schedule should contain milestone dates established for the
construction contractor. Modifications to the milestone dates should be included in the
licensee's monthly construction progress report, which is a separate item from the QCIP.

7-2.10 Erosion Control and Environmental Compliance

An Environmental Compliance Plan should be developed for all projects under

construction. The plan should include an approved erosion and sediment control plan to
prevent environmental degradation of streams during construction. The plan should also
include a listing of all permit and license requirements, and plans and programs that
require oversight by the licensee to ensure adherence to the documents.

7-2.11 Construction Inspection Checklist

Although not required, consideration should be given to providing the QCIP

inspectors with checklists to aid them in reviewing and inspecting the construction work.
The checklists could be generic or prepared for the specific construction project. The
checklists will help the inspectors plan their inspections and serve as a reminder in
review of work plans and inspection of installed work. Appendix VII-E contains three
sample civil inspection checklists for excavation, earthwork and concrete placement.
The checklists are prepared to indicate types of items to be covered and format and are
not intended to be complete for the categories discussed. Depending on the type of
construction, checklists for other categories, such as, mechanical, electrical and welding,
should be included.

7-3 Types of Quality Control Inspection Programs

The type of QCIP adopted will depend on the complexity of construction,

ownership of the project and contractual arrangements. Each program must be evaluated
on its ability to meet the FERC Regulations and its ability to provide for an adequate

7-8
inspection force. The primary goal is safety and not cost minimization. Cost is always
important; however, quality cannot be sacrificed for cost. The QCIP should assure that
the specified work is constructed in accordance with approved plans and specifications.
Included in this section is a discussion of the various types of QCIPs encountered under
the FERC's jurisdiction and what is considered to be an acceptable standard for each
type.

The various types of construction have been divided into three categories and each
category has been divided into three types of QCIPs. Each category and its attendant QC
arrangement are defined as follows:

• Category 1 - Construction of a major new hydroelectric project that includes a

new dam, powerhouse, spillway, tunnels and appurtenant facilities.

1A - QCIP by the licensee, design engineer or independent firm other

than the construction contractor.

1B - QCIP by the licensee who is also the designer and construction

contractor. This could also be a labor-broker construction
contract.

1C - Turnkey design-build construction. The same firm designs and

constructs the project with some quality control inspection
included in the contract. QCIP by licensee or independent firm
other than the design-build firm.

• Category 2 - Construction not as large and complex as Category 1. A typical

example would be an addition to an existing structure such as
construction of a powerhouse at an existing dam. QCIPs A, B and
C are as described under Category 1. It is recognized that there
have been and will be very large and complex projects that only
involve the construction of a powerhouse at an existing dam.

• Category 3 - Construction not as large and complex as Category 2. A typical

example would be the modification of an existing structure, such
as the installation of post-tensioned rock anchors in a concrete
gravity dam or major maintenance such as replacing gates or
resurfacing a spillway section. QCIPs A, B and C are as described
under Category 1; however, it should be noted that QCIPs 3A and
3B are more common than 3C.

7-9
 Routine maintenance that does not affect project safety would not normally
require a QCIP. If a licensee is unsure whether a QCIP is required, the Regional Director
or Director, Division of Dam Safety and Inspections should be contacted for further
guidance.

7-10
7-3.1 Category 1A

The construction of a new major hydroelectric project requires the most

comprehensive QCIP for hydroelectric projects under FERC jurisdiction. The QCIP
must contain all of the items in Appendix VII-A and be discussed in sufficient detail and
clarity for the document to be self contained. Described below is a suggested outline
with comments on items to be covered and pertinent issues to be considered for a
Category 1 construction where the QCIP is performed by the licensee, design engineer or
independent firm other than the construction contractor.

7-3.1.1 Introduction

The introduction should describe the project and the proposed construction. The
organization responsible for QCIP should be stated as well as the licensee, designer and
construction contractor. The general goals of the QCIP should be discussed. Specialized
construction techniques and equipment should be described.

7-3.1.2 Responsibilities

The responsibilities of the various organizations involved with the construction,

such as the licensee, designer, quality control inspection organization, testing laboratory
and construction contractor should be discussed. The services to be provided by each
organization should be itemized and briefly discussed.

7-3.1.3 Organization and Staffing

This section pertains primarily to the organization and staffing of the quality
control inspection personnel. However, personnel involved in construction management
should also be included to provide a better understanding of the necessary coordination
and relationship between personnel.

The responsibilities and duties of key QCIP staff should be clear and definitive.
Resumes for personnel assigned to specific functions should be included in the initial
QCIP and should be concise and specific on education and experience. Emphasis should
be placed on previous experience and involvement in the type of construction and the
level of responsibility. This information should be supplemented on a continuing basis
by submittal of qualifications of personnel actually employed. As stated in Appendix
VII-A, the QCIP should contain an organization chart of the construction inspection
force. It is helpful to relate the key personnel responsibilities and duties to an
organization chart. Appendix VII-B contains descriptions of duties and responsibilities

7-11
of some key personnel and sample construction management organization charts with
emphasis placed on the QCIP. The charts are identified by fictitious FERC project
numbers. Sample organization charts for Category 1A QCIPs are represented by FERC
Project Numbers 24,995 and 24,996.

Other Recommended Practices

In general there are other recommended practices to those discussed in Appendix

VII-B, relative to the organization and staffing of a QCIP, that should be encouraged or
required in some situations. These practices are as follows:

• In an effort to achieve a separation of authority for the QCIP, it is preferable to

place the quality control inspection personnel under a separate and equivalent
level of supervision, such as the Resident Engineer. However, if this organization
is not feasible, the principal QCIP supervisor in the field should have limited or
no involvement with contractor negotiations, scheduling of construction and cost
justification.

• It should be stated in the QCIP that all QCIP positions shown are intended to be
full time except where part time is specified. For part time positions, the
estimated time on the job should be specified and related to the construction
activity.

• In addition to the descriptions of personnel duties and responsibilities in the QCIP

text, consideration should be given to making a Key Project Personnel Summary
Table. The table would have the name, title with name of company, primary work
location, percent of time spent on site and a brief statement of responsibilities for
key project personnel. The key personnel should be both on-site and off-site
personnel of the QCIP and pertinent personnel from the licensee such as the
Project Manager. The table would provide, at a glance, a summary of the key
personnel involved in the construction along with a brief statement of their
responsibilities.

• In situations where the Design Engineer is not stationed at the project construction
site, there should be a statement in the QCIP establishing the frequency of
required field inspections of the ongoing construction and his involvement in
reviewing QCIP reports and test results. The field inspections should also be
correlated with critical stages of construction. For fast moving construction
projects such as a RCC dam, the inspections should be scheduled early and made
frequently.

7-12
7-3.1.4 Inspection Plan or Field Inspection Guidelines

Some large and complex construction projects have an inspection plan and others
have field inspection guidelines for the QCIP personnel. The purpose of the plan or the
guidelines is to provide guidance for the QCIP personnel and establish inspection,
reporting and documentation procedures. The contents can be varied between a plan and
guidelines or varied within either, depending on the licensee, design engineer or
contractor. The essential elements of an inspection plan or field inspection guidelines are
inspection criteria, contractor operations, QCIP operations, and documentation. A
training program for field engineers and inspectors may also be included in an inspection
plan. A construction inspection checklist covering specific aspects of construction may
also be included with the plan or guidelines, which is discussed in more detail in Section
7-2. A discussion of the essential elements described above are as follows:

7-3.1.4.1 Inspection Criteria

The criteria for inspection of contract work is the executed contract between the
contractor and the licensee and any amendments or change orders executed during the
work. The contract incorporates drawings, specifications, codes, standards and laws,
which are the basis of contract enforcement and must be available to inspection
personnel. The contract documents should be reviewed periodically to ensure that
current documents are being used and that all requirements are being met.

7-3.1.4.2 Contractor Operations

Normally, the contractor operates independently from the licensee and the QCIP
personnel and retains responsibility for satisfactory performance and site safety. This
allows the contractor, within the limitations of the contract, to choose his own methods,
schedules, materials and equipment. It is the contractor's responsibility to provide quality
and schedule controls over materials, workmanship and methods to assure meeting
contract requirements. It is the responsibility of the licensee, construction management
firm and QCIP personnel to verify that the contractor meets all contract obligations and
QCIP personnel inspect and verify, rather than direct or control, the contractor's field
construction operations.

7-3.1.4.3 QCIP Operations

The QCIP personnel are responsible for verifying that all contract work conforms
to contract documents and project procedures. Contractors should be advised

7-13
immediately upon detection of nonconforming work so that the work can be corrected.
General inspection duties should be listed, such as, becoming familiar with the contract
documents, when to contact the supervisor and guidance on when to stop work.
Procedures should be established for communications between QCIP personnel, the
contractor and construction management personnel.

7-3.1.4.4 Documentation

The purpose of reporting is to document the observation, investigation and analysis of

inspection work. There are numerous types of reports such as the daily inspection report,
nonconformance report, environmental deficiency report, field directive and clarification
report, concrete placement and test cylinder report, cadweld inspection report, compacted
fill density test report, field weld inspection form, pipe and components field inspection
form and shotcrete report, to name a few. Each project should use the inspection
reporting method that is appropriate for the type of construction and the construction
contract. A discussion of the proposed inspection reporting should be presented in the
QCIP along with sample report forms. The QCIP should define the review levels for all
reports. The daily inspection report, nonconformance report and environmental
deficiency report are required for all QCIPs. Discussions of the daily inspection and
nonconformance reports are contained in the following paragraphs. The environmental
deficiency report is discussed in Section 7-3.1.6.

7-3.1.4.4.1 Daily Inspection Report

The daily inspection report is prepared by the QCIP Inspectors and provides a
means of recording the contractor's daily operations. Daily reports are written when
there is construction activity. If there is no construction, there should be no daily reports.
For documentation, a daily report will note when construction stopped and another daily
report will note when construction began. The report should cover all important factors
affecting job conditions and progress of the work and can be used later as a basic
reference to determine the exact history of work at any given time. The daily report
should include such items as weather conditions, description of activities performed,
types of equipment used, materials incorporated into the work, description of any
problems requiring correction and corrective action taken, detailed description of any
instructions given to the contractor and any other information necessary to document the
contractor's activity and progress during the shift. For QCIPs structured similar to
Project 24,995, the original report would be filed with the Office Engineer and copies
retained by the Senior Civil Engineer and the inspector.

7-3.1.4.4.2 Nonconformance Report

7-14
 The nonconformance report is used to identify, report and document all observed
nonconformances and their disposition. A nonconformance is any observed deviation
from the intent of the construction contract documents. The report identifies the
condition and required action, and leaves space for future entry of the time and manner
of correction. The report is initiated by the quality control inspector. The inspector's
supervisor is responsible for seeing that disposition of the nonconformance is defined,
that corrective action is taken and the correction is documented. There is a distinction
between nonconforming work that is addressed on the spot and nonconforming work
that requires review and study. Failure to meet compaction criteria that results in the
immediate reworking or removal and replacement is an example of noncomforming work
that is addressed on the spot. This could be handled by the inspector and his supervisor
and would not necessarily require input, other than review, from the Design Engineer or
the Resident Engineer. Low strength concrete and inadequate foundation preparation are
examples of nonconforming work that require review and study. Such situations would
require input from the inspector's supervisor, Design Engineer, engineering or geological
specialists and the Resident Engineer and may require studies and follow-up reports. A
sample nonconformance report form is contained in Appendix VII-C. The disposition
for QCIPs structured similar to Project 24,995 would have the Senior Civil Engineer or
the Materials Engineer signing as QC supervisor in the description/affected area box and
the Resident Engineer signing as QC supervisor in the disposition box on the sample
form. To track action on all work observed to be deficient by the QCIP staff, a quality
tracking and reporting system should be developed and maintained. The system should
contain such information as, report number for tracking, date of issue, originator,
description of deficient work, disposition, technical basis for disposition, date of closure,
party responsible for closure and pertinent references.

7-3.1.4.5 Training

A project-specific training program for all incoming field engineers and inspectors
should be established. The Resident Engineer should assure that each engineer and
inspector has training in the area of expertise and quality control procedures for the
inspections that they will perform. Formal classroom training should be conducted by a
designated representative for the applicable category or discipline. Personnel such as
batch plant inspectors and field technicians with little or no experience should be given
on-the-job training and be required to pass a series of written tests on quality control
procedures, laboratory procedures, and test methods for either soils or concrete. They
should be supervised by an experienced lead engineer. Experienced supervisors,
engineers and inspectors should be required to have a specified number of years of
construction experience or be required to pass a series of written examinations relating to
the area of inspection, the contract documents and quality control procedures.

7-15
7-3.1.5 Field Testing Requirements

The contract plans and specifications have been developed to establish testing
requirements for the project and the standards and codes to which the work will conform.
The QCIP staff will use the specified tests to verify that the work is performed in
accordance with the contract. The sampling will be done and the tests will be conducted
at a frequency that will ensure that elements of the work are in compliance with the
specified standards.

7-16
 Appendix VII-D is a Sample Materials Testing Schedule and Referenced
Documents. The number of tests are not intended to be representative of the quantity of
tests required but rather an example of a preferred format for presenting the materials
testing schedule and frequency. The Referenced Documents provide a full description of
the tests referred to in the Sample Materials Testing Schedule.

7-3.1.6 Environmental Compliance

A detailed Environmental Compliance Plan should be developed for all projects

under construction. The plan should include a listing of all permit and license
requirements, and plans and programs which require oversight by the licensee to ensure
adherence to these documents. The listing could be presented in tabular form as
construction related environmental requirements. It is important that the plan contain an
adequate erosion and sediment control plan to prevent environmental degradation of
lands and streams during construction.

The erosion and sediment control plan is required as a license article and the
approved plan should be included in the QCIP to ensure that adequate inspection and
reporting is in place. The plan should address the protection of existing vegetation,
grading of slopes, control of surface drainage, sediment containment measures,
temporary topsoil stockpiling, storage and disposal of excess excavation and debris,
construction and upgrading of access roads, and clearing and construction of the
transmission line rights-of-way. Approved disposal sites should be indicated. The plan
should also outline a schedule for implementation of any mitigation measures proposed
and the monitoring and maintenance of the measures. The environmental requirements
should be outlined in the Technical Specifications.

For QCIPs structured similar to Project 24,995 (Appendix VII-B), the Resident
Engineer has overall responsibility for environmental compliance and the duties are
carried out by the Environmental Coordinator. The Environmental Coordinator monitors
the project under construction on a daily basis to assure compliance. The Environmental
Coordinator briefs the Resident Engineer and other key staff on environmental concerns.
In the event that environmental concerns arise from the Environmental Coordinator or
other QCIP staff, the Resident Engineer will take the appropriate action to notify the
contractor of his responsibilities and to correct any noncompliance. A sample
Environmental Deficiency Report is contained in Appendix VII-C. The deficiency report
should reference the requirement, the nature of the deficiency, the type of correction
required, and the time frame to complete the correction. If the deficiency violates a
license requirement, that should also be noted. The Resident Engineer or Environmental
Coordinator should make the required notification to outside concerns, such as resource

7-17
agencies, to comply with special permit requirements.

7-3.1.7 Construction Schedule

The preliminary construction schedule, based on the understanding of the project

at that time and the FERC license conditions, should be included in the QCIP submitted
for approval. The schedule should contain milestone dates established for the
construction contractor(s). After award of the construction contract, the contractor will
submit detailed construction schedules, which in some cases, will be updated monthly.
Modifications to milestone dates should be included in the licensee's monthly
construction progress report, which is a separate item from the QCIP.

The construction schedule in the QCIP should be presented in either tabular or

graphic form.

7-3.1.8 Planned Use of Consultants

In the order issuing license for most large and complex construction projects, the
FERC requires the licensee to retain a Board of Consultants to review the design, plans
and specifications and construction of the project for safety and adequacy. The Board
should also review the initial QCIP and comment on any changes that are considered
necessary.

Major areas of concern for certain projects may require special consultants such as
experts for blasting and vibration control to advise on the monitoring of shock and
vibration from blasting and pre/post blasting surveys; instrumentation specialists to
advise on various types of instruments and to interpret results and lake tap experts to
provide advice and experience on this highly specialized work. The special consultant's
qualifications and scope of work should be included in the QCIP. The need for special
consultants may be evaluated based on the actual conditions encountered during
construction. Immediately after selection of the consultant, his qualifications and scope
of work should be added to the QCIP.

7-3.2 Category 1B

The Category 1B QCIP also involves the construction of a new major

hydroelectric project; however, in Category 1B the licensee is the designer, construction
contractor and also provides the quality control inspection. The suggested outline,
comments and pertinent issues (as discussed in Section 7-3.1 and Category 1A of
Appendix VII-B) will apply in this section except where noted otherwise.

7-18
7-19
7-3.2.1 Responsibilities

Since the design, construction and quality control inspection will be done by the
licensee, it is important that the responsibilities of the various departments within the
licensee's organization be described. These will primarily be departments involved in the
design, construction and operation of the project. Of particular concern is the separation
of authority and the level of reporting. It is important that there be a separation of
reporting authority between construction staff and QCIP staff at the field level and that
the reporting comes together at as high a level as possible in corporate headquarters.

7-3.2.2 Organization and Staffing

This section is similar to the Organization and Staffing Section for Category 1A;
however, the primary difference is in the separation of authority between the QCIP staff
and the construction personnel. A sample organization chart and descriptions of duties
and responsibilities of some key personnel for Category 1B is presented as Project
24,997 in Appendix VII-B. The organization chart delineates the independent
relationship between the construction personnel and the quality control inspection
personnel. There are certainly other arrangements that could be considered; however, the
important item is the separation of authority.

7-3.3 Category 1C

As with Categories 1A and 1B, Category 1C also involves the construction of a

new major hydroelectric project. In Category 1C the project is constructed under a
turnkey contract by a design-build contractor. Frequently there is a built-in quality
control function by the design-build contractor that is required for proper production
controls. To comply with the FERC Regulations, there must be a separate QCIP by the
licensee or a separate engineering design firm under contract to the licensee. The
licensee's overall QCIP should integrate all available testing results developed during the
project by the QCIP staff, the design-build contractor's quality control activities and any
third party testing. An example of such a coordinated approach would be that the design-
build contractor must run tests to characterize potential concrete aggregate sources and to
identify the suitable portions of the given source. Based on the resulting data and
information, the licensee's QCIP staff would run selected tests to confirm that the
selected materials meet all requirements. In this manner, the contractor's production
control testing results would develop a base of quality control information that would
allow the licensee's QCIP testing program to be reduced in the number of tests and still
be specific and representative of the materials selected for construction.

7-20
7-21
 The suggested outline, comments and pertinent issues to be considered (as
discussed in Section 7-3.1 and Category 1A of Appendix VII-B) will apply in this
section except where otherwise noted.

7-3.3.1 Responsibilities

The responsibilities of the various organizations involved with the construction,

such as the licensee, design-build contractor and the quality control inspection
organization should be discussed. The services to be provided by each organization
should be itemized and discussed. An example would be that the coordination of the
testing laboratories, if more than one laboratory is used, should be discussed, such as the
sharing of test results and the QCIP inspector's authority to direct the type, location and
frequency of tests that the inspector deems necessary.

7-3.3.2 Organization and Staffing

This section pertains primarily to the organization and staffing of the quality
control inspection personnel. Since the licensee has a separate QCIP from the design-
build contractor, there should be communication with the design and construction
personnel of the design-build firm.

The statement on responsibilities, duties and resumes of key QCIP staff (as
presented in Section 7-3.1.3 and Category 1A of Appendix VII-B) apply to this category.

7-3.3.3 Field Testing Requirements

The Field Testing Requirements, as discussed in Section 7-3.1.5, are valid for the
Category 1C QCIP; however, the coordination of testing between the design-build
contractor and the QCIP testing (as discussed in Sections 7-3.3, 7-3.3.1 and in Category
1B of Appendix VII-B) should be taken into consideration for Category 1C construction.

Testing for foundations and materials during the advanced engineering and design
stage that would normally be done by the designer under Category 1A construction and
available to QCIP personnel prior to and during construction would normally be done by
the design-build contractor for Category 1C construction. Therefore, it is important that
this information be coordinated between the QCIP staff and the design-build contractor
and considered when designing the QCIP testing requirements.

7-22
7-3.4 Categories 2A, 2B and 2C

As stated in Section 7-3, Category 2 construction is not as large and complex as

Category 1. A typical example would be an addition to an existing structure such as
construction of a powerhouse at an existing dam. This could also include a private or
municipal powerhouse at an existing Federal dam.

The boundary line between one construction category and another is not always
distinct and it is not important that it should be. The purpose of establishing construction
categories is to provide a vehicle for discussing the relative magnitude of QCIPs
necessary for the project under construction. The QCIP policy statements, suggested
format and procedures discussed for Categories 1A, 1B and 1C also apply to Categories
2A, 2B and 2C.

The construction of a new powerhouse at an existing dam requires a

comprehensive QCIP. The QCIP must contain all of the items in Appendix VII-A and be
discussed in sufficient detail and clarity for the document to be self-contained.

The number of QCIP personnel and variation of QCIP disciplines in the sample
organization chart for Category 2C should be similar to that required for Categories 2A
and 2B. However, the number of QCIP personnel for any Category 2 construction will
probably be smaller than for Category 1 construction. Normally, a Category 2
construction such as construction of a new powerhouse at an existing dam is contained in
a smaller area than Category 1 construction, where a dam, powerhouse and spillway can
be spread out over a relatively large area, requiring more QCIP staff to inspect
concurrent construction activities. Also, fewer disciplines may be required for Category
2 construction. An example would be construction of a new powerhouse at an existing
dam, which may have little or no earthwork.

The suggested outline, comments and pertinent issues to be considered (as

discussed in Sections 7-3.1, 7-3.2 and 7-3.3 and in Categories 1A, 1B and 1C of
Appendix VII-B) will apply to Categories 2A, 2B and 2C except where otherwise noted.

The three Category 2 QCIPs will be discussed as one unit except for the
Organization and Staffing Section where a sample organization chart has been prepared
for a Category 2C QCIP.

7-3.4.1 Responsibilities

The responsibilities of the various organizations involved with construction (as

7-23
described in Sections 7-3.1.2, 7-3.2.1 and 7-3.3.1) are applicable to Category 2 and
should be used depending on the type of QCIP. If the project involves construction of a
new powerhouse at a Federal dam, the responsibilities of the Federal Agency that
operates the dam should be discussed. The FERC has Memoranda-of-Understanding
(MOU) with Federal Agencies, such as the Army Corps of Engineers and the Bureau of
Reclamation, relative to construction. The appropriate policy and procedures contained
in the MOUs should be considered in the discussion of responsibility.

7-3.4.2 Organization and Staffing

This section pertains primarily to the organization and staffing of the quality
control inspection personnel. The statement on responsibilities, duties and resumes of
key QCIP staff, as presented in Section 7-3.1.3, applies to this category.

Only one sample organization chart was prepared for Category 2. The
organization chart and the role of the principal QCIP supervisor in the field, for Project
24,999, are discussed in Appendix VII-B. Project 24,999 is considered to be a
representative QCIP for Category 2C.

7-3.4.3 Inspection Plan or Field Inspection Guidelines

The Inspection Plan or Field Inspection Guidelines discussed in Section 7-3.1.4

are also applicable to Category 2 construction.

For Category 2 projects, the training may need to be revised. Due to their
probable shorter construction time, formal classroom training may not be appropriate.

The Quality Control Engineer may rely on experience and on-the-job training to
assure that each engineer and inspector is trained in the area of expertise and quality
control procedures for the inspections that they will perform.

7-3.4.4 Field Testing Requirements

The Field Testing Requirements as discussed in Sections 7-3.1.5 are valid for
Category 2 construction. The coordination of testing between the design-build
contractor and the QCIP testing (as discussed in Sections 7-3.3, 7-3.3.1, 7-3.3.3 and in
Category 1C of Appendix VII-B) should be taken into consideration for the Category 2C
construction.

Depending on the size of construction, a field laboratory facility may not be

7-24
established at the project site and an independent commercial laboratory may be used for
material testing. Adequate on-site storage should be provided on an as needed basis for
such items as concrete cylinder molds, concrete cylinder curing box, and other required
equipment.

7-3.4.5 Planned Use of Consultants

Depending on the complexity of construction, uniqueness of design, downstream

hazard potential and other considerations, a Board of Consultants may not be required in
the license. There may be areas of concern for certain projects that will require special
consultants such as is discussed in Section 7-3.1.8.

7-3.5 Categories 3A, 3B and 3C

As stated in Section 7-3, construction for Category 3 is not as large and complex
as Category 2. Two examples of Category 3 construction would be the modification of
an existing structure such as the installation of post-tensioned rock anchors in a concrete
gravity dam or major maintenance of an existing hydroelectric project such as replacing
gates or resurfacing a spillway section. The post-tensioned rock anchor construction will
be used as an example in this discussion.

The installation of post-tensioned rock anchors in an existing concrete gravity

dam requires a QCIP that contains all of the items in Appendix VII-A. The QCIP should
be discussed in sufficient detail and clarity to be self-contained. The suggested outline,
comments and pertinent issues to be considered, (as discussed in Sections 7-3.1, 7-3.2, 7-
3.3, 7-3.4 and Categories 1A, 1B and 1C of Appendix VII-B) will apply to Categories
3A, 3B and 3C except where noted otherwise. Because of the single item, short duration,
and less complex type construction for Category 3 construction, the QCIP will be a
scaled down version of that required for Categories 1 and 2. There will be fewer QCIP
personnel and disciplines.

Category 3 will be discussed as one unit except for the Organization and Staffing
Section where a sample organization chart has been prepared for a Category 3A QCIP.
The number of QCIP personnel in the sample organization chart should be similar to that
required for Categories 3B and 3C.

7-3.5.1 Organization and Staffing

This section pertains primarily to the organization and staffing of the quality

7-25
control inspection personnel. The statement on responsibilities, duties and resumes of
key QCIP staff, as presented in Section 7-3.1.3 applies to all QCIP personnel in this
category.

Only one organization chart was prepared for Category 3. The sample
organization chart and the duties and responsibilities of the Project Manager and the
Resident Engineer, for Project 25,000, are discussed in Appendix VII-B. Project 25,000
is considered to be a representative QCIP for Category 3A.

7-3.5.2 Inspection Plan or Field Inspection Guidelines

The elements of the Inspection Plan or Field Inspection Guidelines as discussed in

Section 7-3.1.4 are applicable to Category 3 construction.

For Category 3 projects, the training section should be revised. Due to the short
construction time, the small number of QCIP personnel involved and the relatively fast
pace of construction, formal classroom training may not be appropriate. The practicality
of on-the-job training is also questionable.

It is important that qualified personnel with the appropriate experience be used to

staff the QCIP. If replacement personnel are required, they should also be qualified and
have appropriate experience.

7-3.5.3 Field Testing Requirements

The Field Testing Requirements as discussed in Section 7-3.1.5 are valid for
Category 3 construction. The coordination of testing between the design-build
contractor and the QCIP testing (as discussed in Sections 7-3.3, 7-3.3.1, 7-3.3.3, 7-3.4.4
and Categories 1A, 1B and 1C of Appendix VII-B) should be taken into consideration
for the Category 3C construction.

Normally, a field laboratory facility will not be established at the project site for a
Category 3 construction and an independent commercial laboratory will be used for
material testing.

7-3.5.4 Planned Use of Consultants

Normally, the FERC license will not require a Board of Consultants for a
Category 3 construction. There may be areas of concern for certain projects that will
require special consultants such as drilling and grouting consultants for rock tendons.

7-26
7-3.6 Small Construction Not Requiring a QCIP

As stated in Section 7-1.1, the Regional Director may decide not to require a
QCIP for relatively minor work. An example would be a low hazard project where the
construction consists of replacing an existing powerhouse, that is not a water impounding
structure, with a new powerhouse that would be constructed on the existing concrete
foundation. The design has been done by the owner and partially by the manufacturer of
the turbine and generator. Thus the owner is the partial designer, contractor and
operator. The owner and his associates would provide the quality control during
construction.

It is important that the design for such a project be reviewed in detail by a

professional engineer, in addition to the FERC staff. It should be stressed that the project
be constructed in accordance with approved plans and specifications. The FERC will
provide additional guidance when the Regional Office staff makes field inspections
during construction to verify that good construction practices are followed.

7-4 Summary

As stated in Section 12.40 of the Regulations and Article 4 of the Standard L

forms for licensed projects, during any construction, repair or modification of project
works, the licensee must maintain any quality control program that may be required by
the appropriate Regional Director, commensurate with the scope of work and meeting
any requirements or standards set by the Regional Director. The guidelines presented in
this chapter provide staff engineers and geologists with recommended procedures,
criteria and examples, to be used in reviewing and evaluating the licensee's QCIPs.

All QCIPs should provide for an adequate and qualified construction inspection
force and should contain detailed information including, but not limited to, the
information contained in Appendix VII-A and discussed in Sections 7-2 and 7-3. In
addition to the staffing responsibilities and authority described below, the QCIPs should
have an inspection plan, adequate documentation, training, materials testing schedule,
environmental compliance plan and construction schedule. Sample organization charts,
abbreviated descriptions of duties and responsibilities for some key QCIP personnel,
report forms, material testing schedules and referenced documents and inspection
checklists are found in the appendices.

QCIP staff must be independent from personnel responsible for construction. The
responsibility and authority of QCIP staff must be clear and specific. This independence
must be maintained for all types of construction. The principal QCIP supervisor in the

7-27
field should have limited involvement with contractor negotiations, scheduling of
construction and cost justification. Key QCIP staff should have the authority to stop
work due to adverse quality conditions. There should be someone in the field at all times
who has "stop-work" authority. QCIP personnel, such as the inspectors, should have
authority to recommend stop work to the contractor and to their supervisor.

There are numerous types of QCIPs depending on the complexity of construction,

ownership of the project and contractual arrangements. Each type of program must be
evaluated on its ability to meet the FERC Regulations and its ability to provide for an
adequate and qualified force for inspection of construction of the project works. In this
chapter, the various types of construction are grouped into three categories and each
category is divided into three types of QCIPs. The categories are defined and typical
QCIPs discussed with illustrations, such as the sample organization charts and
descriptions of duties and responsibilities of some key personnel in Appendix VII-B.

Where the licensee is the designer, construction contractor and also provides the
quality control inspection, it is important that there be a separation of reporting authority
between the construction staff and the QCIP staff at the field level. The reporting should
come together at as high a level as possible in the corporate headquarters and supervisory
structure.

For turnkey design-build construction, there is frequently a built-in quality control

function by the design-build contractor for production control. However, to comply with
the FERC Regulations, there must be a separate QCIP by the licensee or a separate
engineering design firm under contract to the licensee.

7-28
7-5 References

1. Fairweather, V., "The Pursuit of Quality: QA/QC", Civil Engineering,

February 1985, pp. 62-64.

2. American Society of Civil Engineers, "ASCE Professional Grade Descriptions",

ASCE Guide to Employment Conditions for Civil Engineers, ASCE - Manuals
and Reports on Engineering Practice, 2nd Edition, 1980, pp. 5-7.

3. American Concrete Institute, "Qualifications of Personnel for Inspection and

Testing Duties", Guide for Concrete Inspection, ACI 311.4R-88, Manual of
Concrete Practice, 1988, pp. C-12 and C-13.

7-29
7-6 APPENDICES

7-30
 APPENDIX VII-A

CONSTRUCTION QUALITY CONTROL INSPECTION

PROGRAM CONTENT

AND

PROPOSED OUTLINE FOR QCIP

 CONSTRUCTION QUALITY CONTROL INSPECTION PROGRAM CONTENT

The construction quality control inspection program should provide for an

adequate and qualified force for inspection of the construction of the project works. The
program description should contain detailed information including, but not limited to, the
following:

a. Introduction describing the project and proposed construction.

b. Organization chart of the construction inspection force.

c. Number and specialties of inspectors proposed. Information submitted with

this item, or item above, should include the number of inspectors proposed for
each feature of construction. Where full time inspection is not proposed, the
schedule for part time inspection should be described.

d. Description of duties, responsibilities, necessary qualifications, and scope of

authority of the QCIP staff. This information should be supplemented on a
continuing basis by submittal of qualifications of the personnel actually
employed.

e. Field tests to be performed and frequency of testing.

f. Field laboratory facilities or commercial testing services to be provided.

g. Description of Inspection Plan including documentation and reporting

procedures.

h. Planned use of consultants during construction.

i. Schedule of all major features of construction.

j. Description of erosion control and other environmental measures.

The QCIP should cover such items as: (1) water diversion during construction, (2)
underground and surface excavation, (3) production and placement of earth and concrete,
(4) powerhouse construction, (5) installation of penstocks and (6) installation of major
mechanical and electrical equipment. The information provided should be in sufficient
detail for the reviewer to determine that the proposed QCIP provides adequate
construction quality control.

7-A-1
 Some items, such as safety requirements and special construction techniques, may
be included in the contract specifications. If not, these items, especially those that may
involve the safety of personnel working in deep excavations, or in the vicinity of water
impounding structures (cofferdams), should be detailed in the QCIP or the temporary
Emergency Action Plan.

7-A-2
 PROPOSED OUTLINE FOR THE

QUALITY CONTROL INSPECTION PROGRAM1

A. Introduction
o Purpose
o Background
o Description of structures and types of construction
o Specialized construction techniques and equipment

B. Organization and Staffing Responsibilities

o Responsibilities of various organizations
o Number of staff and availability required
o Titles, duties and responsibilities of staff
o Specialty inspectors
o Lines of communication and authority
o Approval and rejection of work
o Authority to stop work
o Statement of independence
o Resumes

C. Inspection Plan and Field Practices

o Inspection criteria
o Knowledge of contract plans and specifications
o Inspection equipment and resources
o Contractor operations
o Coordination with contractor's schedule
o QCIP operations
o Frequency of inspections
o Documentation and follow-up action
o Training

D. Documentation
o Daily inspection reports
o Nonconformance reports

1
Bullet items are minimum suggested topics. Additional topics should be included to
fully describe the QCIP. A statement of "not applicable" is required for outline items not
needed in a particular QCIP.

7-A-3
 o Other periodic reports
o Maintenance of records
o Photographs

E. Training
o Study materials
o Classroom instruction
o On-the-job-training and supervision
o Proficiency testing and certification
o Resume update

F. Material Testing
o Testing schedule
o Testing standards
o Testing organization
o Adequacy of on-site laboratory
o Adequacy of off-site laboratory
o Evaluation of testing data and actions required
o Documentation

G. Erosion Control and Environmental Compliance

o Environmental compliance plan
o Erosion and sediment control plan
o License requirements
o Specialized plans, permits and approvals
o Frequency of inspections
o Documentation and corrective actions
o Environmental deficiency report

H. Schedule
o Start and finish dates
o Anticipated construction sequence
o Staged and phased construction

I. Planned Use of Consultants

o Areas of inspection and review
o Consultants names and resumes

J. Appendices
o Organizational chart

7-A-4
o Descriptions of duties and responsibilities of QCIP staff
o QCIP personnel resumes
o Project layout
o List of contract documents
o Materials testing schedule and referenced documents
o Example of reports to be used, e.g. nonconformance report
o Flow chart for tracking construction deficiency
o Contractor's schedule
o Record keeping procedures
 APPENDIX VII-B

SAMPLE ORGANIZATION CHARTS

AND

DESCRIPTIONS OF DUTIES AND RESPONSIBILITIES

OF SOME KEY PERSONNEL

7-B-1
7-B-2
Category 1A

Project 24,995

Project 24,995 has the construction management and QCIP being performed by an
engineering design firm under contract to the licensee and the construction contractor
under separate contract to the licensee.

As stated in Section 7-2.3, the number and specialities of inspectors proposed for
each feature of construction should be included and should be determined by the type of
construction and the construction schedule. This organization chart is for Category 1
construction, which requires the most comprehensive QCIP for projects under FERC
jurisdiction. Thus there are numerous QCIP engineers and inspectors shown on the
chart. When a qualified QCIP individual is proposed to cover more than one area of
expertise, it should be demonstrated that there will not be a conflict in supervision and
scheduling of construction inspections.

In this arrangement, the Senior Civil Engineer and the Materials Engineer (who is
in charge of the field and laboratory testing) are the principal QCIP supervisors in the
field and must coordinate activities with others such as the Environmental Coordinator
and the Quality Control Engineer, all of whom report to the Resident Engineer. Sample
abbreviated descriptions of duties and responsibilities for some key personnel are as
follows:

Resident Engineer

The Resident Engineer is responsible for the management and general direction of
the firm's site construction management organization. Will supervise the field inspection
and testing activities and engineering staff functions. Also will recommend progress
payment estimates, change order control, evaluation of claims and cost and scheduling of
construction activities. Has authority to stop work for nonconformance or potentially
unsafe work practices. Reports to the Project Manager, located off-site.

Senior Civil Engineer

The Senior Civil Engineer is responsible for the overall surveillance and
inspection of construction activities and any related testing required to confirm
compliance with the specifications. Through subordinates, prepares daily field
inspection reports and directives. First individual with line responsibility for requiring
the correction of any work by the contractor that does not conform to the specifications.

7-B-3
Has authority to issue nonconformance reports and notifies the Resident Engineer and
Quality Control Engineer for resolution. Reports to the Resident Engineer and
communicates to the Design Engineer any situation where the plans and specifications do
not appear to be appropriate for the conditions encountered. Has authority to stop work.
Is notified of any recommendation to stop work by QCIP personnel who have the
authority to recommend stop work to the contractor and their supervisor. Has limited or
no responsibility or involvement with contractor negotiations, scheduling of construction
and cost justification. Reports to the Resident Engineer.

Materials Engineer

The Materials Engineer is responsible for the operation of the materials laboratory
and for the inspection of production in the concrete batch plant. Supervises both
laboratory and field sampling and testing of concrete, embankment materials, penstocks,
conduits and other major facilities. Responsible for quality control inspection of the
batch plant and transportation to the point of placement. The first individual with line
responsibility for requiring the correction of any work, under his control, by the
contractor that does not conform to the specifications. Is notified of any recommendation
to stop work by QCIP personnel who have the authority to recommend stop work to the
contractor and their supervisor. Has authority to stop work and reports to the Resident
Engineer.

Quality Control Engineer

The Quality Control Engineer is responsible for assuring that all site activities for
field inspection and materials quality control testing conform to the QCIP requirements.
Reviews and audits the activities of the Materials Laboratory as well as the overall
activities of the field inspection staff. Has authority to stop work. Is notified of any
recommendation to stop work from QCIP personnel who have the authority to
recommend stop work to the contractor and their supervisor. Reports to the Resident
Engineer and communicates directly with the Quality Control Manager in Headquarters.

Environmental Coordinator

The Environmental Coordinator is responsible for reviewing and monitoring

project construction on a daily basis to assure environmental compliance. Reviews
contractor's environmental plans for compliance with approved license and permits.
Documents environmental compliance with erosion and sedimentation control
techniques, archeological monitoring, blasting and instream rock excavation techniques,
reservoir clearing activities and permit specifications for special concern areas. Provides

7-B-4
 environmental training to field inspectors and conducts environmental briefing
course for construction personnel. Coordinates with field inspectors on construction
activity relative to environmental compliance. Recommends modification of
environmental plans that are determined to be unsatisfactory during construction.
Collects quality control water samples on an as needed basis to verify the accuracy of the
contractor's water quality results. Provides input to the monthly report concerning
environmental and mitigation activities. Responsible for liaison with resource agencies
on environmental issues and takes measures to avoid permit violations. Has authority to
recommend stop work to the contractor and his supervisor and advises the Senior Civil
Engineer, Materials Engineer and the Quality Control Engineer of the action. Reports to
the Resident Engineer.

7-B-5
7-B-6
Project 24,996

Project 24,996 is similar to Project 24,995 in that the construction management

and QCIP is being performed by an engineering design firm under contract to the
licensee and the construction contractor is under separate contract to the licensee.
However, the QCIP differs from Project 24,995 in that the QCIP field staff for Project
24,996 are under the supervision of the Quality Control Supervisor since the Senior Civil
Engineer's primary function is as the Design Engineer. The Senior Civil Engineer and
his staff will coordinate the design with construction activities and provide the necessary
engineering support. Sample abbreviated descriptions of duties and responsibilities for
some key personnel are described in the following paragraphs.

Quality Control Supervisor

For Project 24,996, the Quality Control Supervisor is the principal QCIP
supervisor in the field and must coordinate activities with others such as the Senior Civil
Engineer, the Electrical and Mechanical Engineers and the Environmental Coordinator.
The Quality Control Supervisor is the first individual with line responsibility for
requiring the correction of any work by the contractor that does not conform to the
specifications. The testing laboratory and QCIP field inspectors are supervised by the
Quality Control Supervisor. Through subordinates, prepares daily field inspection
reports and directives. Reports to the Resident Engineer and communicates to the Senior
Civil Engineer any situation where the plans and specifications do not appear to be
appropriate for the conditions encountered. The Quality Control Supervisor, Senior Civil
Engineer and Resident Engineer have authority to stop work. Is notified of any
recommendations to stop work by QCIP personnel who have the authority to recommend
stop work to the contractor and their supervisor. The Quality Control Supervisor has
limited or no responsibility or involvement with contractor negotiations, scheduling of
construction and cost justification. Reports to the Resident Engineer and communicates
directly with the Quality Control Manager in Headquarters.

QCIP Inspectors

The QCIP Inspectors are responsible for performing the required verification of
the correctness and adequacy of the construction contractor's work in accordance with
applicable specification, drawing and procedural requirements. Inspectors will document
the results of each inspected function on the designated reporting form and inform
responsible personnel about unsatisfactory items, while ensuring that corrective actions
are taken to resolve the conditions. For defective work the inspectors will initiate a
nonconformance report and submit the report to the Quality Control Supervisor for

7-B-7
resolution and will verify that the defect has been corrected. The QCIP Inspectors have
authority to recommend stop work to the contractor and their supervisor. The QCIP
Inspectors report to the Quality Control Supervisor. The QCIP Inspectors are responsible
for observing and reporting on construction activities in their specific areas of
assignment. The principal areas of inspection are civil, mechanical, electrical and
welding. Brief descriptions of the responsibilities, in addition to those described in this
paragraph, and inspection functions for Civil and Mechanical Inspectors are contained in
the following paragraphs.

Civil Inspector

The Civil Inspector is responsible for such items as observing and recording the
results of all critical clearing operations; survey work; lines and grades; excavation;
blasting operations; instrumentation; foundation and concrete placement acceptance;
batching, mixing, testing, and transporting concrete; drilling and grouting operations;
earthworks such as placing, compacting and testing of embankments; rockfill; and
tunneling. The inspection functions are dependent on the type of construction.

The Civil Inspector should observe and assure the adequacy of the field and
laboratory tests. He should assure that the work is performed by qualified and, where
specified, certified personnel.

Inspection for conventional concrete batching and placement is relatively

straightforward and there is an abundance of information in the literature. Quality
control inspection for roller compacted concrete (RCC) construction is substantially
different than that of conventional concrete. RCC is discussed in Chapter 3 of the
Engineering Guidelines. RCC construction involves placing and spreading no-slump
concrete in horizontal layers and compacting with a smooth-drum, vibratory roller.
Foundation preparation and concrete mix designs are very important for a RCC dam.
Once construction starts, the rate is faster than conventional concrete or embankment
construction. Therefore, it is necessary to construct a test fill (prior to construction of the
RCC dam) to assess all of the required elements such as the mix design, speed of
placement, compaction effort, workability suitable for compaction, joint cleanup
requirement, segregation of coarse aggregates and contractor performance. As stated in
Chapter 3, the test fill should be constructed outside of the footprint of the proposed
structure. The test fill offers the QCIP staff an opportunity to gain useful experience in
the operation of monitoring equipment that will be used on the actual fill. The Civil
Inspector should be familiar with the results of the test fill and should use the design as a
basis for his inspections. The Civil Inspector should also be familiar with restrictions
during inclement weather.

7-B-8
Mechanical Inspector

The Mechanical Inspector is responsible for such items as observing and assuring
the correctness of the fit, acceptable tolerances, alignment, embedment and mating of all
critical parts of the field assembled turbine and generator. Assures that the contractor
performs sufficient inspections on all mechanical components and material. Makes
random and periodic inspections of the alignment, welding, flushing and hydrostatic
testing of powerhouse piping. Verifies by inspection that the correct installation,
alignment and final setting of mechanical components such as pumps, motors, pressure
vessels, valves and air compressors. Participates in completion inspections of installed
systems. Where appropriate, periodic site inspections should be made at fabrication
shops.

7-B-9
7-B-10
Category 1B

Project 24,997

In Project 24,997, the licensee is the designer, construction contractor and also
provides quality control inspection. In the field, the Resident Engineer supervises and is
responsible for construction and the Quality Control Supervisor supervises and is
responsible for the QCIP. It should be noted that there is a separation of authority in the
field between the QCIP staff and construction personnel and the reporting authority
comes together in the office of the Executive Vice President for Engineering,
Construction and Operations, a high level in the corporate structure. Sample abbreviated
descriptions of duties and responsibilities for some key personnel are described in the
following paragraphs.

Resident Engineer

The Resident Engineer is responsible for project management, production, costs

and overall quality of work. Responsible for the general direction of the discipline
activities, material and equipment coordination and contract coordination. Has authority
to stop work for nonconformance work or potentially unsafe work practices. Reports to
the off-site Project Manager.

Quality Control Supervisor

The Quality Control Supervisor is responsible for the overall surveillance and
inspection of construction activities and any related testing required to confirm
compliance with the specifications. Through subordinates, prepares daily field
inspection reports and directives. First individual with line responsibility for requiring
the correction of any work performed by the construction personnel that does not
conform to the specifications. Has authority to issue nonconformance reports and
notifies the Resident Engineer and Senior Civil Engineer for resolution. Communicates
to the Resident Engineer and Senior Civil Engineer any situation where the plans and
specifications do not appear to be appropriate for the conditions encountered. The
Quality Control Supervisor has authority to stop work. Is notified of any
recommendation to stop work from construction management personnel who have the
authority to recommend stop work. Has no responsibility or involvement with
scheduling of construction and cost justification. Reports to the Vice President for
Quality Control at the corporate level and communicates with the Resident Engineer and
his staff at the project site.

7-B-11
Testing Laboratory and Field Supervisor

The Testing Laboratory and Field Supervisor is responsible for the operation of
the materials laboratory and coordination with an outside laboratory, if appropriate, for
the inspection of production in the concrete batch plant. Supervises both laboratory and
field sampling and testing of concrete, embankment materials, penstocks, conduits and
other major facilities. Responsible for quality control inspection of the batch plant and
transportation to the point of placement. Has authority to recommend stop work to the
contractor and to his supervisor and reports to the Quality Control Supervisor at the
project site.

Civil Inspectors

The Civil Inspectors will verify that work in progress is being performed in
accordance with applicable specification, drawing and procedural requirements. They
will maintain an up-to-date status of construction progress and inform responsible
personnel about unsatisfactory items, while ensuring that corrective actions are taken to
resolve these conditions. For defective work, the Civil Inspectors will initiate a
nonconformance report and submit it to the Quality Control Supervisor for resolution and
will verify that the defect has been corrected. Have authority to recommend stop work to
the contractor and their supervisor. Reports to the Quality Control Supervisor at the
project site.

Senior Civil Engineer

The Senior Civil Engineer coordinates and approves project engineering design,
manages the overall project to meet specifications and supervises a specialty staff of civil
engineers and geologists. Approves engineering design and changes, resolves
engineering design conflicts and interface problems within the project and has authority
to stop construction if the work is deemed unsafe or in noncompliance with the
specifications. Reports to the Resident Engineer.

7-B-12
7-B-13
Category 1C

Project 24,998

The organization chart for Project 24,998 in Appendix VII-B is considered

representative of Category 1C.

Project 24,998 requires the construction of a new dam, powerhouse, tunnel,

spillway and appurtenant structures. The QCIP is performed by a combination of
licensee and separate engineering design firm personnel. The licensee has a contract
with the design-build firm to design and construct the project and a separate contract
with the engineering design firm for the QCIP. The QCIP could be staffed by all
licensee or all design firm personnel and still accomplish the same purpose.

In this arrangement, the Field Quality Control Supervisor and the Testing
Laboratory Supervisor are the principal QCIP supervisors in the field and report to the
Resident Engineer, who is responsible for the QCIP. Sample abbreviated descriptions of
duties and responsibilities for these personnel are contained in the following paragraphs.

Resident Engineer

The Resident Engineer is the licensee's senior project representative at the project
site. Responsible for all interface and coordination between the licensee and the design-
build contractor. Manages the activities of the QCIP staff and through the QCIP staff
monitors the quality of the design-build contractor's work. Monitors the design-build
contractor's engineering, construction reporting, quality control, progress and schedule.
Reviews and assures resolution of all nonconformances. Reviews design changes and
claims; however, normally, claims would be an internal matter with the design-build
contractor. Assures compliance with the FERC licensing requirements. Has authority to
stop work for nonconformance work or potentially unsafe work practices. Reports to the
licensee's off-site Project Manager.

Field Quality Control Supervisor

The Field Quality Control Supervisor is a licensee employee responsible for

assuring that the design-build contractor's site activities are carried out according to the
contract documents and the approved QCIP. He plans and directs the activities of a staff
of inspection and testing personnel. He assures that records of satisfactory completion of
site activities, equipment and material acceptability and qualifications of QCIP personnel
are maintained. Through subordinates, prepares daily field inspection reports and

7-B-14
directives. Reviews the design-build contractor's construction inspection and test
procedures and coordinates the results with the QCIP inspection and test results. Has the
authority to issue nonconformance reports and notifies the Resident Engineer, Project
Engineer, and Testing Laboratory Supervisor for resolution. Reports to Resident
Engineer and communicates to the turnkey designer any situation where the plans and
specifications do not appear to be appropriate for the conditions encountered. Has
authority to stop work due to adverse quality conditions. Is notified of any
recommendation to stop work from QCIP personnel who have the authority to
recommend stop work to the contractor and their supervisor. Has no responsibility or
involvement with contractor negotiations, scheduling of construction and cost
justification. Reports to the Resident Engineer.

Testing Laboratory and Field Supervisor

The Testing Laboratory Supervisor is an employee of the engineering design firm.

Responsible for the establishment of certified testing facilities, equipment and personnel.
Develops test procedures and instructions to personnel for testing performance.
Coordinates the QCIP testing with the design-construct contractor testing to obtain
maximum use of the information. Responsible for the performance testing of soils,
aggregate, concrete, rockbolts, soil and rock anchors, conduits, penstocks, field welds
and other required testing. Has authority to recommend stop work to the contractor and
to his supervisor and has no responsibility or involvement with contractor negotiations,
scheduling of construction and cost justification. Reports to the Field Quality Control
Supervisor.

7-B-15
7-B-16
Category 2C

Project 24,999

Project 24,999 requires the construction of a new powerhouse at an existing dam.

There are numerous combinations of licensee/owner relationships relative to owner of
the dam and owner of the powerhouse. For Project 24,999, the dam is operated by a
Federal Agency and the licensee/owner of the new powerhouse is a non-public
organization. The project is being constructed by a design-build contractor under
contract to the licensee and the QCIP is being done by an engineering design firm under
separate contract to the licensee.

Quality Control Engineer

The Quality Control Engineer is the principal QCIP supervisor in the field and is
responsible for the QCIP. As a representative of the licensee, he is responsible for all
interface and coordination between the licensee and the design-build contractor at the
project site. Is also responsible for all interface and coordination between the licensee
and the Federal Agency inspectors. Reviews any design changes or claims involved with
the design-build contractor and assures compliance with the FERC licensing
requirements. Plans and directs the activities of a staff of inspection and testing
personnel. Communicates to the turnkey designer any situation where the plans and
specifications do not appear to be appropriate for the conditions encountered. Through
subordinates, prepares daily field inspection reports and directives. Has authority to
issue nonconformance reports and to stop work due to adverse quality conditions or
potentially unsafe work practices. Has limited responsibility and involvement with
contractor negotiations, scheduling of construction and cost justification. Reports to the
off-site Quality Control Manager.

The Federal Agency inspector will provide inspection of those items that affect
the structural integrity or operation of the Federal project and will report to the Quality
Control Engineer for verification of any discrepancy or correction.

Normally, technical support for disciplines such as geotechnical, structural

engineering and blasting would be supplied from an off-site location on an as needed
basis for construction of this type. These personnel would report to the Quality Control
Engineer.

7-B-17
7-B-18
Category 3A

Project 25,000

The sample Construction Management Organization Chart for Project 25,000 is

for the installation of post-tensioned rock anchors in concrete gravity dam. The licensee
has separate contracts with a general contractor and an engineering design firm. The
design firm is also responsible for the QCIP.

Project Manager

The Project Manager, who is also the Design Engineer for the engineering design
firm, is located off-site. It should be stated in the QCIP that the Design Engineer will
inspect anchor tests to verify that load increments, for performance and proof tests,
conform to the design. For other types of Category 3 construction, it is important for the
Design Engineer to inspect the construction frequently and at critical stages. Normally,
for smaller jobs, the Design Engineer is located off-site.

Resident Engineer

The Resident Engineer, who is employed by the engineering design firm, is the
principal QCIP supervisor in the field and is responsible for the QCIP. All site project
activities are under the direction of the Resident Engineer. Has responsibility for the
complete project including the construction, installation, coordination of testing, cost and
schedule control, safety and material flow. Assures compliance with FERC
requirements. Plans and directs the activities of a small staff of inspectors. Coordinates
the necessary testing with a commercial laboratory, located off-site. In conjunction with
his staff prepares daily field inspection reports and directives. Reports to the Project
Manager any situation where the plans and specifications do not appear to be appropriate
for the conditions encountered. Has authority to issue nonconformance reports and to
stop work due to adverse quality conditions or potentially unsafe work practices.
Reviews any design changes or claims involved with the contractor. Involved in
contractor negotiations, scheduling of construction and cost justification. It may not be
practical, on a small job such as this, for the principal QCIP supervisor in the field to
have limited or no involvement with contractor negotiations, scheduling and cost
justification. The Resident Engineer reports to the Project Manager, who is located off-
site.

Due to the small QCIP staff the Resident Engineer and/or Inspectors will monitor
and inspect such procedures as drilling, watertightness testing, grouting and bolt tension

7-B-19
tests, check for proper alignment of drill holes, proper materials and placement
procedures for grout and compliance with anchor load test procedures.

APPENDIX VII-C

SAMPLE REPORT FORMS

NONCONFORMANCE REPORT

AND

ENVIRONMENTAL DEFICIENCY REPORT

7-C-1
 APPENDIX VII-D

SAMPLE

MATERIALS TESTING SCHEDULE

AND

REFERENCED DOCUMENTS

7-D-1
 SAMPLE
Noname Hydroelectric Project
Materials Testing Schedule
Field Testing
Material Test Test Method and/or Standard Test Frequency and/or
Certification

River banks and run off areas Compliance with erosion Control Visual/Daily turbidity Daily
plan and effectiveness of erosion
control measures

Rock anchors Fabrication and installation PTI Manual, ASTM A416 and A421 As directed by the Quality Control
Engineer

Fresh concrete Temperature ASTM C172, metal dial type First batch produced
thermometers each day and every 50 cubic yards

Air content ASTM C231 First batch produced each day and
every 50 cubic yards

Slump ASTM C143 First batch produced each day and

every 50 cubic yards

Hardened Compressive ASTM C31, C39 and C172 (7 and Six test cylinders from first 100
concrete strength 28 day) amd ACI 214 cubic yards, four cylinders from
each 150 cubic yards thereafter for
each class of concrete in any one
day

7-D-2
 SAMPLE
Noname Hydroelectric Project
Materials Testing Schedule
Field Testing
Material Test Test Method and/or Standard Test Frequency and/or
Certification

Concrete Aggregate Gradation, Fineness modulus, ASTM C33, C127, C131, C136 and Prior to delivery to batch plant. To
Abrasion, Specific gravity and C289 be done by testing laboratory
absorption

Grout holes Pressure test As directed by contract Prior to pressure grouting

specifications

Random fill Moisture content ASTM D1557 At time of placing

Density ASTM D1557 Before use

In-place density ASTM D1556, and D2216 A minimum of one test for every
three lifts or three tests for every
200 cubic yards

Structural steel Field-welded connections AWS D1.1 At completion and as directed by

the
Quality Control Engineer

Rock core Direct shear strength RTH 203-80 Minimum of three test specimens
for each rock type to determine
cohesion and the angle of internal
friction. As directed by the Quality
Control Engineer.

7-D-3
 SAMPLE
NONAME HYDROELECTRIC PROJECT
REFERENCED DOCUMENTS

American Concrete Institute (ACI)

ACI 214-77 1983 Recommended Practice for Evaluation of Strength Test Results
of Concrete

American Society for Testing and Materials

ASTM C 31 1985 Making and Curing Concrete Test Specimens in the Field

ASTM C 33 1986 Concrete Aggregates

ASTM C 39 1986 Compressive Strength of Cylindrical Concrete Specimens

ASTM C 127 1984 Test for Specific Gravity and Absorption of Concrete
Aggregate

ASTM C 136 1984 Method for Sieve Analysis of Fine and Coarse Aggregate

ASTM C 143 1978 Slump of Portland Cement Concrete

ASTM C 172 1982 Sampling Freshly Mixed Concrete

ASTM C 231 1982 Air Content of Freshly Mixed Concrete by the Pressure Method

ASTM C 289 Test Method for Potential Reactivity of Aggregate (Chemical

Method)

ASTM D 1556 1982 Density of Soil In Place by the Sand Cone Method

ASTM D 1557 1978 Moisture Density Relations of Soil and Soil-Aggregate

Mixtures Using 10-lb (4.54-kg) Rammer and 18-in. (457-mm)
Drop

ASTM D 2216 1978 Method for Laboratory Determination of Water (Moisture)

7-D-4
Content of Soil, Rock, and Soil-Aggregate Mixtures
 American Welding Society (AWS)

AWS D1.1 1988 Structural Welding Code

U.S. Army Corps of Engineers

RTH 203.80 1980 Direct Shear Strength of Rock Core Specimens, Rock Testing
Handbook (RTH), Geotechnical Laboratory, Waterways
Experiment Station

Value Your Business Engineering Co.

GC Spec. 1990 Specifications for General Construction Contract, Noname

Hydroelectric Project (Engineer's Specification No. 23456-
GC001)

7-E-1
 APPENDIX VII-E

SAMPLE

CIVIL INSPECTION CHECKLISTS

7-E-2
 SAMPLE
INSPECTION CHECKLIST
EXCAVATION

The following characteristics of excavation should be reviewed:

1. Perform a thorough review of all site exploratory reports made during

design and during construction.

2. Peview site plans to note any underground structures to be avoided, such as

pipes and utilities.

3. Prior to commencing excavation, clearing, grubbing and stripping

operations should be completed to the areas shown on the site excavation
drawings. Provisions should be provided for dust abatement.

4. For mass excavation remove soil, boulders, coal and any other unclassified
materials to the lines and grades shown on the drawings.

5. For structural excavation remove in-situ materials for structures,

underground utilities, pipes, culverts, drains or diversion channels to the
lines, grades, elevations and dimensions shown on the drawings.

6. Sumps and wellpoints should be constructed and adequate pumps should

be provided to prevent groundwater accumulation in the excavation.

7. Random sampling and testing should be performed on excavated material

to note changes in soil classification or physical properties.

8. Borrow areas should be properly maintained to provide effective drainage

and erosion control.

9. Excavation should be made to the lines, grades, elevations and dimensions

as shown on the drawings or as directed by the foundation engineer or
geologist to obtain a suitable foundation.

10. Completed excavation should be inspected, tested and accepted prior to

placement of mud mats, slabs, pipes or structural backfill.

7-E-3
 SAMPLE
INSPECTION CHECKLIST
EARTHWORK (BACKFILL)

The following characteristics of earthwork should be reviewed:

1. Determine the material requirements for the different types of earthwork on

the construction project.

2. Check that material sources have been approved.

3. If method specifications are used, determine the compaction requirements.

Determine what degree of compaction is needed to satisfy design criteria.
Determine the type and weight of compaction equipment required and the
number of equipment passes at a specified lift thickness necessary to meet
density requirements.

4. Check that provisions are adequate for dust abatement.

5. Foundation should be inspected, tested and accepted prior to start of fill

operations.

6. Provisions should be adequate for control and disposal of surface and

subsurface water.

7. Fill and borrow areas should be maintained to provide effective drainage

and are protected against erosion.

8. Field and laboratory tests should be conducted at the frequency specified to

verify physical requirements of the fill material.

9. Fill material should meet moisture, compaction and density requirements

and be placed in the specified lift thickness.

10. Moisture and density tests should be performed at random locations and at
specified frequency.

11. Fill material should be brought to final grade and inspected, tested and
accepted, if acceptable.

7-E-4
 SAMPLE
INSPECTION CHECKLIST
CONCRETE PLACEMENT

The following characteristics of concrete placement should be reviewed:

1. Placing equipment is clean and free of loose concrete, mud, and other
debris that could jeopardize the quality of the structure.

2. Reinforcing steel and embeds are clean and free of loose rust, grease or
other matter that may adversely affect concrete bond.

3. Embedded piping has been tested as specified.

4. Joints and surfaces to receive concrete are free of deleterious materials.

5. Forms are clean and free of foreign material.

6. Provisions for hot or cold weather concrete protection are provided.

7. Concrete is placed in a manner to prevent segregation.

8. Placement of concrete is made in lift thickness as specified and within time

restrictions between lifts for high lift placements.

9. Concrete is properly vibrated.

10. Placement is made to avoid excessive drying of fresh concrete before next
lift is placed.

11. Concrete is sampled and tested at specified frequency for strength, slump,
temperature and unit weight.

12. Concrete is brought to final grade and finished as specified.

7-E-5
 CHAPTER VIII

DETERMINATION OF THE
PROBABLE MAXIMUM FLOOD

September 2001
 Chapter VIII

Determination of the Probable Maximum Flood

Table of Contents

8-1 Background and Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

8-1.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
8-1.3 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
8-1.4 Limitations of PMF Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

8-2 Preliminary Review of Project and Hydrologic Data . . . . . . . . . . . . . . . . . . . . . 8

8-2.1 Identify and Obtain Preliminary Data . . . . . . . . . . . . . . . . . . . . . . . . . 8
8-2.2 Information About Upstream Dams . . . . . . . . . . . . . . . . . . . . . . . . . 11
8-2.3 Field Visit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
8-2.3.1 Dam, Spillway, Outlet Works, and Power Plant . . . . . . . . 11
8-2.3.2 Operating Personnel Interviews . . . . . . . . . . . . . . . . . . . . 12
8-2.3.3 Drainage Basin Assessment . . . . . . . . . . . . . . . . . . . . . . . 13

8-3 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

8-3.1 Information from Previous Studies . . . . . . . . . . . . . . . . . . . . . . . . . . 15
8-3.2 Streamflow Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
8-3.2.1 Continuous Streamflow Hydrograph . . . . . . . . . . . . . . . . 16
8-3.2.2 Peak Flow and Volume Data . . . . . . . . . . . . . . . . . . . . . . 17
8-3.3 Precipitation Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
8-3.4 Applicable Hydrometeorological Reports . . . . . . . . . . . . . . . . . . . . . 18
8-3.5 Physical Characteristics of the Drainage Basin . . . . . . . . . . . . . . . . . 18
8-3.6 Snowpack W ater Equivalent and Temperature Data . . . . . . . . . . . . . 18
8-3.7 Data on Existing Reservoirs, Spillways, Outlet Works,
and Operation Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

8-4 Review and Assessment of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

8-4.1 Unit Hydrographs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
8-4.2 Flood Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
8-4.3 Precipitation Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
8-4.4 Snowpack Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
8-4.4.1 Water-Equivalent Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

ii September 2001
 8-4.4.2 Temperature Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
8-4.5 Data on Reservoir Volume, Spillway and Outlet-Works
Capacity, and Operation History and Policy . . . . . . . . . . . . . . . . . 24
8-4.5.1 Reservoir Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
8-4.5.2 Spillway and Outlet-Works Capacity . . . . . . . . . . . . . . . . . 24
8-4.5.3 Operation History and Policy . . . . . . . . . . . . . . . . . . . . . . . 25

8-5 Approach to Tasks for Probable Maximum Flood Development . . . . . . . . . . . . 27

8-5.1 Subdivision of Drainage Basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
8-5.2 Gaged and Ungaged (Sub) Basin(s) . . . . . . . . . . . . . . . . . . . . . . . . . 29
8-5.3 Approach and Identification of Tasks . . . . . . . . . . . . . . . . . . . . . . . . 30

8-6 Unit Hydrograph for Gaged (Sub) Basins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

8-6.1 Historical Floods for Calibration and Verification . . . . . . . . . . . . . . 31
8-6.2 Determination of Basin Average Rainfall . . . . . . . . . . . . . . . . . . . . . 32
8-6.3 Cold Season Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
8-6.3.1 Snowmelt Considerations . . . . . . . . . . . . . . . . . . . . . . . . . 33
8-6.3.2 Infiltration Characteristics of Potentially Frozen Soils . . . 34
8-6.4 Base-Flow Separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
8-6.5 Time of Concentration and Clark's Storage Coefficient for
Each Subbasin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
8-6.6 Rainfall Sequence for Recorded Storms . . . . . . . . . . . . . . . . . . . . . . 36
8-6.7 Infiltration for Unit-Hydrograph Development . . . . . . . . . . . . . . . . . 36
8-6.8 Calibrate Unit Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
8-6.8.1 Cases Where a Single Basin Unit Hydrograph
is Sufficient (No Subdivision) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
8-6.8.2 Unit Hydrographs for Subbasins and Channel Routing . . . 38
8-6.9 Hydrograph Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

8-7 Unit Hydrographs for Ungaged (Sub)Basin(s) . . . . . . . . . . . . . . . . . . . . . . . . . 43

8-7.1 Applicable Unit Hydrograph Procedures for Each Basin
(Subbasin) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
8-7.2 Regional Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8-7.2.1 Data Required . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8-7.2.2 Rainfall Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
8-7.2.3 Development of Generalized Regional Relationships . . . . 48
8-7.3 Empirical Coefficients for Synthetic Unit-Hydrograph Procedures . . 49
8-7.3.1 Snyder Unit Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . 50
8-7.3.2 Clark Unit Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
8-7.3.3 SCS (NRCS) Dimensionless Unit Hydrograph . . . . . . . . . 52

iii September 2001

8-8 Loss Rates for Subbasins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
8-8.1 Basin-Averaged Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
8-8.2 Distributed Loss Rate Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
8-8.2.1 Application of Distributed Loss Rate Method . . . . . . . . . . 56
8-8.3 Verification and Model Adjustment . . . . . . . . . . . . . . . . . . . . . . . . . 57
8-8.4 Infiltration Characteristics of Soils Under Frozen Conditions . . . . . . 59

8-9 Probable Maximum Flood Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

8-9.1 Spatial Distribution and Disaggregation of the
Probable Maximum Precipitation . . . . . . . . . . . . . . . . . . . . . . . . . 61
8-9.1.1 Storm Duration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
8-9.1.2 Storm Spatial Distribution . . . . . . . . . . . . . . . . . . . . . . . . 61
8-9.1.3 Temporal Distribution of the Probable Maximum
Precipitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
8-9.2 Antecedent and Coincident Conditions . . . . . . . . . . . . . . . . . . . . . . . 63
8-9.2.1 Antecedent Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
8-9.2.2 Coincident Hydrometeorological Conditions . . . . . . . . . . 65
8-9.2.3 Snowmelt Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
8-9.3 Reservoir and Channel-Routing Approach . . . . . . . . . . . . . . . . . . . . 67
8-9.4 Base Flow Coincident with Probable Maximum Flood . . . . . . . . . . . 68
8-9.5 Inflow PMF Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
8-9.6 Review and Sensitivity Analysis of Representative
PMF Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

8-10 Reservoir Routing to Obtain the Outflow PMF Hydrograph . . . . . . . . . . . . . . . 69

8-10.1 Initial Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
8-10.2 Routing Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
8-10.3 Reporting Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

8-11 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

8-12 Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

8-13 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Appendix A Determining the PMF for Civil Works Flow Chart

Appendix B Probable Maximum Flood Study Report Outline
Appendix C HEC-1 Data-Analysis Techniques of Infiltration Rate
Estimate Methods
Appendix D Loss Rates for Subbasins – Distributed Loss Modeling

iv September 2001
Figures

Figure 8-1.1 Regions Covered by PMP Studies . . . . . . . . . . . . . . . . . . . . . . . . . -7-

Figure 8-6.1 Baseflow Simulation in HEC-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . -43-

Figure 8-6.2 Estimation of Clark Unit-Hydrograph Parameters . . . . . . . . . . . . . . -44-

Tables

Table 8-8.1 Minimum Infiltration Rates for Hydrologic Soil Groups . . . . . . . . . -56-

-v-
 Chapter VIII

Determination of the Probable Maximum Flood

8-1 Background and Purpose

This chapter of the Engineering Guidelines is primarily intended to provide procedures

for the development of the Probable Maximum Flood (PMF) for use in the evaluation of
proposed and existing dams and other impounding structures. The purpose of these
guidelines is to provide consistency in PMF determinations. The guidelines are not a
substitute for good engineering judgment and experience when available data clearly call
for a departure from recommended procedures. Therefore, the recommended procedures
should not be rigidly applied in place of other justifiable solutions.

For about the last 50 years, the PMF has received general acceptance as the design flood
for dams in the United States, whose failure would pose a threat to public safety [Myers
1967]. M ore recently, the PMF has received acceptance as the design flood for large
dams in many other countries as well [ICOLD 1991].

The definition of the PMF contained in these Engineering Guidelines is:

...the flood that may be expected from the most severe combination of
critical meteorological and hydrologic conditions that are reasonably
possible in the drainage basin under study.

A PMF is generated by the probable maximum precipitation (PMP), which is defined as:

...theoretically, the greatest depth of precipitation for a given duration that

is physically possible for a given size storm area at a particular geographic
location at a certain time of year.

Developing a PMF hydrograph for a dam safety evaluation generally involves two steps,
which are, respectively, hydrologic and hydraulic in nature:

C Modeling of runoff through the project drainage basin to produce an

inflow PMF for the project reservoir.

C Routing of the inflow PMF through the project reservoir and dam outlet
works to obtain the outflow PMF and the maximum reservoir elevation
at the dam.
These steps involve considering several coincident or sequential events, each of which
may have a strong effect on the resulting PMF. This chapter attempts to address the use
and estimation of those events to avoid compounding of conservatism and to provide a
reasonable PMF hydrograph given the limitations of basic hydrologic and meteorological
data.

Some important features of a specific project, such as the operation of the reservoir, the
outlet works, etc., which are relevant to routing the PMF through the reservoir and dam
outlet works, also need to be addressed. The safety of existing or designed dams is the
primary concern in adopting the PMF as the criterion for safeguarding the public. This
chapter provides guidance on the determination of the PMF. Additional guidance on
developing inflow design flood's (IDF's) is included in Chapter 2 of these Guidelines.

8-1.2 Objectives

There is little chance that hydrology will ever become the precise science that designers,
owners, and regulators would like to see. So many parameters define the basin
characteristics and hydraulics of runoff that the hydrologic engineer will always need to
rely on experience and good judgment. This chapter is intended to provide systematic
procedures that will consistently produce a reasonable PMF hydrograph and appropriate
reservoir flood levels for evaluation of project safety.

While keeping the inherent uncertainty of hydrologic calculations in mind, the objectives
of this chapter of the Guidelines are:

C To recommend a preferred method for developing PMF hydrographs.

C To present procedures which, if implemented by two or more qualified

and experienced hydrologic engineers, would result in reasonably close
or consistent estimates of the PMF.

C To make recommendations regarding the assumptions that must

normally be made in developing a PMF hydrograph for gaged and
ungaged sites.

C To produce an approach that will minimize the total effort and cost of
required studies, while ensuring that the developed hydrograph is
reasonable and pertinent for use in the design or safety analysis of civil
works.

C To provide guidelines for choosing appropriate hydrologic and hydraulic

parameters.

-2-
 C To provide greater consistency nationally for procedures used in PMF
development, while recognizing the wide variety of hydrologic
conditions present across the United States.

8-1.3 Overview

It is the responsibility of owners of dams not under FERC jurisdiction to ensure the safety
of their projects by using the best available technology. The procedures recommended in
this chapter for determining the PMF for FERC jurisdictional projects assume that all
dams in the basin upstream of the project will not fail during floods up to the PMF.
Therefore, the PMF at the project site will not be a combination of the naturally occurring
flood and the flood resulting from a failure of an upstream dam. The PMF at the site is
the result of routing the PMF through upstream dams assuming they remain in place.
However, this does not preclude owners of FERC jurisdictional dams from considering
the failure effect of upstream dams in PMF evaluations.

Previously accepted PMF studies are not required to be reevaluated in accordance with
the new Guidelines, unless it is determined that a re-analysis is warranted. Potential
reasons to re-analyze an existing PMF include, but are not limited to: significant errors
found in the original study or new data becomes available that may significantly alter
previous study results; significant changes in the conditions of the drainage basin such as
basin development or changes in upstream control structures; changes in the state-of-the-
art technology, etc. All new studies should comply with the requirements of these
Guidelines.

As PM F determinations are completed using these Guidelines for a project that could
affect nearby non-FERC-jurisdictional upstream dams, the FERC will advise the
appropriate State Dam Safety Office (State) of the PMF study. The State will be
informed that a new PMF study has been done for the FERC-jurisdictional dam,
assuming all upstream dams do not fail and that the PM F study is available to the State
for its review and information at the FERC Regional Office.

This chapter proposes the use of unit-hydrograph theory as the preferred runoff model for
developing an inflow PMF hydrograph. Unit-hydrograph theory was developed by
Sherman in 1932 based on five basic assumptions, which are: lumped-parameter model
instead of a distributed-parameter model, stationary basin characteristics, uniform rainfall
distribution in space and time, constant hydrographic base time based on lag time, and a
linear relation between rainfall excess and produced flood discharge. The development
of the unit hydrograph is of primary importance in the ultimate development of the PMF
hydrograph, because its use will determine both the temporal distribution and peak rate of
runoff. The use of the United States Army Corps of Engineers (COE) computer program
HEC-1 Flood Hydrograph Package [HEC 1990] is recommended because of the

-3-
widespread use and experience with that program. A new software package, entitled
Hydrologic Modeling System (HEC-HMS), has been developed by the Hydrologic
Engineering Center for rainfall-runoff simulation to supersede HEC-1. Since Windows-
based HEC-HM S retains most of the capabilities of HEC-1, all references to HEC-1 in
this chapter also apply to HEC-HMS. Many United States water-resource agencies have
developed models for their own regional use in developing hydrographs for gaged and
ungaged basins, including dimensionless unit hydrographs, expressions for lag times,
parameters for shaping unit hydrographs, runoff models, etc. This chapter recommends
that any special methods be evaluated for applicability before being used.

Development of unit hydrographs for both gaged and ungaged basins is discussed in this
chapter. The inherent uncertainties in developing PMF hydrographs are significant, even
for locations where quality data are available. Ungaged basins involve even more
uncertainties. Final review of a PMF hydrograph should include a sensitivity analysis for
parameters having a significant effect on the inflow hydrograph. Section 8-12 contains a
glossary that defines technical terms used herein, which are part of the professional
language of hydrologic engineering but may have slightly different meanings depending
on the user. Cautionary statements have been provided throughout the text where care
should be taken in the use of the recommended procedures, or where there are limitations
to their application. These statements appear throughout the text in italics. The
following appendices are included in this chapter:

C Appendix A includes a flowchart that summarizes the methodologies in

this chapter for determining the PMF for gaged and ungaged drainage
basins.

C Appendix B includes PMF Study Report Outlines for gaged and

ungaged basins.

C Appendix C includes a table of HEC-1 data-analysis techniques of the

methods for estimating infiltration rates.

C Appendix D presents a detailed explanation of the use of the distributed

loss rate method, with an example of application of STATSGO (State
Soil Geographic Database) data as discussed in Section 8-8.2 for use in
assigning loss rates.

8-1.4 Limitations of PM F Simulation

No single method of PMF analysis is without limitations. The information and/or

methodologies provided herein are recommended as guidelines that set forth the
engineering criteria and procedures rather than standards. The appropriateness of any

-4-
procedure depends on specific hydrologic and hydraulic characteristics of the watershed
and the availability of rainfall, stream flow, and snowmelt records needed to estimate the
parameters used in those procedures. Any type of analysis or procedure that is used must
be documented and verified. This chapter does not prohibit the use of any method so
long as its use is justified for the basin under study. The following are three general
limitations in the process of PMF simulation.

Drainage Basin Size. In general, this chapter covers drainage basins up to about 10,000
square miles, although the size of the drainage area is not necessarily a limiting factor.
Consideration has also been given to PMF’s produced by local storms that would cover
only part of a large basin, or all of a small one, and to general storms that could cover in
excess of 10,000-square-miles. The upper limit to the basin size is arbitrary but was
made to cover conditions applicable to many dams while still including basins requiring
subdivision for analysis. This chapter applies to most basins with multiple
FERC-licensed projects; however, procedures for very large drainage basins—such as the
lower Missouri or the Columbia Rivers—cannot easily be generalized, since even general
storms may not cover the entire basin.

PMP. It is assumed that complete details of the depth-area and duration of the PMP are
available and no attention has been given to development of the PM P. However,
references are made to developing the isohyetal pattern of the PMP and its use. Often
this information can be obtained from the National Weather Service (NW S)
Hydrometeorological Reports (HMRs). Because of the storm and flood data they
include, the HMR series are important references, but other site-specific PMP studies also
may be available. Figure 8-1.1 shows the geographic regions to which each HMR
applies. The most recent HMR should be used unless a site-specific PMP study has been
approved such as the 1993 EPRI PMP study for the Wisconsin/Michigan area.

Hydrograph Development. Ranges of recommended values of several parameters that

must be assumed for developing hydrographs are given throughout the text. The values
of these parameters were taken from available material developed by government
agencies and other organizations. However, the material cited or quoted does not
represent an exhaustive search of the literature, and each section suggests potential
sources of additional data. The methods recommended were chosen from those widely
recognized and accepted by the hydrologic engineering profession, and for which
considerable information is available.

Because the state-of-the-art in hydrology is constantly changing, the procedures

suggested herein may require future changes. Therefore, this is a “dynamic” document -
one that is subject to review and change as the state of hydrologic engineering is refined
or approved.

-5-
Where there are limitations to the recommended procedures, or where care should be
taken in their use, cautionary statements are provided throughout the text.

-6-
-7-
8-2 Preliminary Review of Project and Hydrologic Data

Prior to a site visit, the hydrologic engineer should become familiar with the project and
the pertinent hydrologic and topographic information, which will help identify special
features that should be observed and the types of data that should be pursued in the field.
This section is intended to be an aid in obtaining and reviewing preliminary data.

8-2.1 Identify and Obtain Preliminary Data

General information about the project should be acquired to identify items that should be
checked or obtained from the field reconnaissance. Generally, the greater the body of
available data, the more confidence in the reliability of the final PMF hydrograph. Each
project will dictate the level of required data acquisition. Information should include but
not be limited to:

C Topographic or site-specific maps. The maps should show the project

location, access roads, layout, and drainage area. Topographic
quadrangle maps can be obtained from the United States Geological
Survey (USGS) as well as from private vendors and some internet sites.
In some areas, topographic maps also are available in digitized form
from USGS Earth Science Information Centers (ESIC). Special
topographic maps, used during dam design and construction or for other
studies, often are available from the dam owner. Satellite imagery,
available through the National Aeronautics and Space Administration
(NASA), can be useful in addressing conditions within the drainage
basin.

C Aerial photographs of the drainage basin. These are sometimes

available from the dam owners, the district offices of the United States
Forest Service (USFS), and the Farm Service Administration (FSA)
(formerly the Agricultural Stabilization and Conservation Service
(ASCS)), local or state transportation agencies, or upstream dam
owners.

C Drainage basin soil types for estimating infiltration rates. If soil surveys
have been developed within the basin, a State Soil Geographic Database
(STATSGO), which provides soil association maps and related data,
will be available in digital form from state offices of the Natural
Resources Conservation Service (NRCS, formerly the Soil Conservation
Service (SCS)). These data also are available from the NRCS National

-8-
 Cartographic and GIS (Geographic Information System) Center in Fort
Worth, Texas.

Note: The NRCS established three soil geographic databases

representing kinds of soil maps. These include the Soil Survey
Geographic (SSURGO) on a county basis at a scale of 1:24,000, the
STATSGO mostly at the scale of 1:250,000, and the National Soil
Geographic (NATSGO) at a scale of 1:5,000,000.

C Stream gage locations and flow data. These data are available from the
USGS Water Supply Papers and Water Resources Data Reports for the
state in which the project is located, and from the USGS web site. Unit
values (short time interval) data for continuous flood hydrograph can be
obtained from the USGS website or district offices of the USGS. Data
for many stream gages usually are reported to the USGS even though
the gage may be operated and maintained by another federal, state, or
local agency; the dam owner; or another private party. If the historical
data for the gages are not obtainable from the USGS, the owner of the
gage should be contacted. Historical ratings for the gages will be
needed and can be obtained from a USGS district office, since they
usually are not contained in the annual Water Resources Data Reports.
Privately owned firms also may provide digital data containing
streamflow information from USGS records. In addition, the dam
owner may have streamflow data not available through other sources.

C Precipitation data from gages that are or were operated in or near the
basin. Rainfall data and snowpack or snow water equivalent data are
generally obtainable from the National Climate Data Center (NCDC)
and/or the National Water and Climate Center under NRCS (NWCC).
Rainfall data often are available from state water-resource agencies or
other federal, state, or local flood-control agencies. Much of the
NWS-stored climatological data are available on compact disks from
private vendors. In addition, the dam owner may have streamflow and
rainfall data that are not reported elsewhere. Upstream dam owners also
may have data. Climatological data as required, including temperature,
wind speed, and solar radiation also may be available from these
organizations.

C Remote sensing data. Remote sensing data can be used as an additional

source of information to topographical maps or aerial photographs to
provide a source of input data for hydrologic modeling. For instance,
these data can be used to determine land use, which affects infiltration

-9-
 rates (Maidment, 1993).

C Hydrologic data for historical storms and associated floods. A search

should be made for this information, which will include the rain gage
data (particularly from recording gages) within and near the drainage
basin and corresponding flood hydrographs. Offices that may have
performed special flood studies for severe floods and have such data on
file include the COE; USGS; FEMA; NW S district offices; NRCS state
offices; TVA; and state or local flood-control agencies. Local
newspapers and other media sources can sometimes provide useful
information, but any such data must be verified before being used.

C Engineering reports that provide information on dam height and type,

reservoir capacity-elevation, spillway type and rating, outlet type and
capacity, and power-intake capacity. The dam owner usually is the best
source for this information; much of it generally is contained in past
safety-analysis reports (in the case of existing projects), which may be
available from the dam owner or from state or federal dam-safety
agencies.

C Information on project operation during past extreme floods. This

information can be obtained from the project owner. Obtaining the
information may require reviewing project operation records and
interviewing project operators.

C Cross sections for the channels through which the PMF hydrograph may
need to be routed. These may be available from FEMA or a local
flood-control agency if flood studies have been made for the area. In
some cases, cross sections of sufficient accuracy can be taken from
7½-minute USGS maps, but field surveys of critical cross sections may
be needed to increase the accuracy of the hydraulic-routing
computations.

C Information on land use. Such information may be obtained from USGS

topographic maps and local land use maps. Aerial photos also are very
helpful for this purpose and are sometimes available from USFS district
offices, local and state transportation agencies, or the NRCS state
offices. The National Aerial Photography Program (NAPP) is available
from the USGS office in Reston, Virginia. Satellite image analysis
should be given consideration for cost-effective derivation of these data.
Field observations also are desirable. Information on future land use

-10-
 can also be important in rapidly developing urban areas - future runoff
conditions may need to be considered in a PMF study.

C Information on geologic conditions within the drainage basin . Geologic

maps frequently are available from the USGS district offices, the NRCS
state offices, and state departments of natural resources.

8-2.2 Information About Upstream Dams

All existing upstream dams must be identified and information obtained to determine
whether or not they create sufficient storage to affect the PMF timing and peak flow.
Up-to-date topographic maps of the drainage basin generally will show the locations of
all upstream dams with large enough reservoir storage to require consideration. The
National Dam Inventory (NATDAM)—available through the FEMA and the COE — lists
the height, length, dam type, reservoir volume, date of construction, and ownership for
dams in each state. Information desired for each dam includes:

C Type and height of dam, type of outlet works or spillways and rating
curves, and a cross section and crest profile of the dam. These data may
be necessary for routing of a PMF hydrograph.

C Tables or graphs of surface area and volume versus reservoir elevation.

C Reservoir operation rules that could possibly affect the timing and peak
flow of the PMF hydrograph. Operators should be interviewed to obtain
historical and proposed information on operation of the reservoir,
spillway, outlet works, and power plants during extreme floods to
determine how flood operations have been performed in the past. This
information also may be of interest in identifying historical floods for
development of unit hydrographs.

8-2.3 Field Visit

Once the preliminary information has been obtained and reviewed, an experienced
hydrologic engineer should visit the dam, spillway, outlet works, power plant, and the
drainage basin to check or confirm information developed in the preliminary review and
to obtain firsthand information about the dam, its facilities, and the drainage area.

-11-
8-2.3.1 Dam, Spillway, Outlet Works, and Power Plant

The dam, spillway, outlet works, and power plant should be visited to obtain information
not available in reports dealing with the site. Such information may include:

C Characteristics of spillways, outlet works, and power intakes.

C Discharge rating curves for each structure. Rating curves should be

checked in the field to ensure that they take into account limitations in
gate opening, such as orifice flow occurring because a radial gate cannot
be opened high enough to provide adequate clearance of the water
surface during passage of the PMF.

C Pertinent elevations on the rating curves of spillways and outlet works .

Elevations provided should be confirmed.

C Gate operation. It is particularly important to ascertain that the gates are

operable to the elevations indicated in the rating curves, and that the
gates have been recently operated under full head.

C Flashboards. If an uncontrolled spillway is equipped with flashboards,

information should be obtained on their height and the dates on which
the flashboards are placed on the spillway and removed. It also is
desirable to determine whether or not the flashboards will fail or can be
readily released at their design flood elevation.

C Available power and backup systems. Availability of power and the

existence of backup systems for operating spillway gates should be
ascertained.

C Remote or local operation. It will be necessary to determine if the dam

is operated remotely or by local operators, and to obtain details and
schedules for operation during extreme floods. Access to the spillway
and outlet facilities, and the reliability of remotely located equipment
and instrumentation under flooding conditions should be evaluated.

C Physical features of the dam and its appurtenances. Such information

will be necessary for routing the inflow PMF through the reservoir and
possibly for reverse-reservoir routing of releases to obtain inflow
hydrographs from historical floods.

-12-
8-2.3.2 Operating Personnel Interviews

Operating personnel should be interviewed. Items of particular interest in addition to

operation and maintenance records include:

C Procedures and operation rules for normal and emergency gate operation
during extreme floods. An assessment should be made of the reliability
inherent in the operation of spillway gates and flashboards, particularly
if the project is remotely operated.

C Rule curves for seasonal operation of the reservoir.

C Information on historical floods. Such information includes flood peaks

and hydrographs, reservoir levels, maximum rates of reservoir rise, and
rainfall depths and timing.

C High-water marks and eyewitness accounts of operations and events

occurring during past floods.

C Procedures and results of spillway and outlet-works gate testing.

8-2.3.3 Drainage Basin Assessment

The primary purpose of this assessment is to obtain quantitative information on the

drainage basin, with special emphasis on identifying all portions that contribute to runoff.
To the extent possible, the drainage basin should be observed by road. Photographs
should be taken to establish a record to aid in later recollection. Previously obtained
topographic, soil, and geologic maps; aerial photographs; and satellite imagery should be
taken to the field for reference. If there are no roads, or if the drainage basin is very
large, it may be desirable to fly over the area. Drainage-area observations should include
confirming or identifying the following:

C Location of rain gages and stream gages.

C Existing upstream dams.

C Special features within the drainage basin such as marshes, lakes, and
closed basins that may delay or reduce runoff.

C Constrictions such as bridge abutments or channel modifications that

may influence flood-routing characteristics.

-13-
 C Manning's "n" and general hydrologic and hydraulic characteristics of
stream channels.

C Areas where soil or geologic features or climatological conditions could

result in locally different rates of infiltration. These areas include large
exposures of rock; areas of high permeability such as karst formations,
deep sand, or fractured basalt; cultivated areas; areas of dense forest or
managed forest cover; high-altitude meadows; and areas where surface
ice conditions are developed by mid-winter thaw and refreeze.

C Large natural constrictions that could act as hydraulic control structures.

C Any significant changes in urbanization, hydrologic use, or land use and

cover that may have occurred since surveys for the available
topographic maps were conducted, or since the historical floods have
occurred.

The following may be necessary if peak flow data from historical floods are incomplete:

C High-water marks along the streams on bridge piers or abutments or

along banks. These may be useful for computing a flood peak flow.

C Eyewitness accounts of long-time residents. These will be helpful to

obtain information on historical flooding. Verify the accuracy of
accounts, if possible.

C Visits to local newspapers and television and radio stations. News

reports on historical flooding may be available.

8-3 Data Acquisition

Hydrologic and meteorologic data are necessary to develop unit hydrographs. The
primary objectives of data collection are as follows:

C To obtain basic precipitation and streamflow data for use in subsequent

analyses.

C To enable the engineer to understand the hydrologic response of the

basin for the season when the critical PMF would occur, to increase
confidence in simulating the runoff process, and to make appropriate
judgments.

-14-
In general, data recommended for use in developing a unit hydrograph for a given basin
are as follows:

C Streamflow records for major historical floods. It is desirable to have

records for at least three major floods and concurrent rainfall data to
provide confidence in the representative unit hydrograph.

C Precipitation records for the storms that produced the historical floods
and the location and history of all rain gages in the basin.

C Physical characteristics of the watershed including topography, soil

types, and land use.

C Snowpack and temperature records in the basin if snowmelt was a factor

in historical floods.

In addition, it is necessary to understand the project's physical features, as well as those

of upstream dams, to properly route flood hydrographs through the reservoir.

To meet the data acquisition objectives, this section describes the specific data needs and
information sources that may be available.

Caution: Delays may be experienced in data collection. These can take place due to
time needed to retrieve data from storage and for field data collection. Appropriate time
should be allotted (i.e., four to six months) for data collection.

8-3.1 Information from Previous Studies

Since unit hydrographs are commonly developed and used in flood-control studies, local,
state, or federal agencies with flood-control responsibilities may have already developed
one for the basin of interest. If available and applicable, the use of such unit hydrograph
may save considerable time and cost in developing the inflow PMF. If a unit hydrograph
is not based on current streamflow or rainfall information, it may be necessary to develop
a new unit hydrograph. Previous flood studies performed for nearby dams should also be
evaluated for relevant information. All information obtained must be reviewed for
quality and applicability; the required review, assessment, and justification procedures
are described in Section 8-4. Sources of information for regional flood studies include:

C Local flood control districts

C COE district and division offices
C USBR regional and area offices and the Technical Service Center
C TVA

-15-
 C NRCS state and district offices
C USGS district offices
C NW S River Forecast Centers
C State water resource agencies
C State dam safety agencies
C State departments of transportation
C Regional planning commissions or agencies
C Dam owners
C FEMA

8-3.2 Streamflow Data

8-3.2.1 Continuous Streamflow Hydrograph

The location of USGS stream gages, along with daily average flows for the water year,
are given in the annual Water Resource Data Report issued for each state by the USGS.
Historical daily average flows for all streamflow gages ever operated by the USGS also
are available on the world wide web at http://water.usgs.gov/. The USGS NAWDEX
system catalogs sources and types of streamflow data that may not be listed in the Water
Resource Data Reports but may be available on its website. A non-recording gage is
manually read on a daily basis. Depending upon stream size, a recording gage is read at 5
- 15 minute intervals and is recorded on either a continuous graph or more modern media.
The search for streamflow data varies depending on whether the basin is gaged or
ungaged. For gaged sites, collection is concentrated on the gages within the basin of
interest; for ungaged sites, the collection effort is extended to gaged basins in the region.
Daily flow records and maximum flows of record for gages in and near the basin can be
obtained from USGS annual Water Resource Data Reports and the previously mentioned
web site. Such data are needed to identify historical floods, and should be used when
developing a unit hydrograph.

To develop a unit hydrograph, streamflow hydrographs are needed for the identified
major historical floods. Continuous streamflow hydrographs can be obtained from USGS
district offices, where stage records for the historical floods and rating curves for the
pertinent stream gages are also available if questions about the accuracy of the historical
flood records arise during the data review. To determine the annual exceedance
probability of each flood event, a frequency curve of annual instantaneous peak flows is
required. Frequency curves are typically a part of Flood Insurance Studies prepared for
FEM A and flood studies prepared by other federal agencies.

Unit hydrographs should be developed from continuous flood inflow hydrographs. It is

preferable that stream gages be located on all tributaries entering the reservoir. However,

-16-
in the absence of flood inflow hydrographs, a continuous flood inflow hydrograph needed
for the unit-hydrograph determination can be developed by reverse-reservoir routing.
This requires knowledge of project outflow and headwater elevations during major
floods. Project outflow can be estimated from downstream stream gage records or project
discharges (gate operations and power releases). However, since reverse routing in HEC-
1 assumes the use of a level pool, this method may result in a less conservative inflow
hydrograph since the “wedge” storage in the reservoir is ignored. For large or long
riverine-type reservoirs, the inflow hydrograph may have to be determined using dynamic
routing methods. In addition, if the project outflow and headwater level data is not given
in hourly or smaller increments of time, then the accuracy of this method may be
questionable. However, if reverse-reservoir routing produces a relatively accurate inflow
hydrograph, it should also provide an acceptable unit hydrograph. This is because the
effect of the reservoir impoundment on flood flows is directly taken into account
[Maidment 1993, Newton 1983].

8-3.2.2 Peak Flow and Volume Data

As discussed in Section 8-3.1, the effort in collecting streamflow data will be greatly
reduced if a previously developed unit hydrograph is available for the project basin that
satisfies the guidelines in this chapter. In that case, the only streamflow data required
will be those necessary to identify the occurrence of antecedent floods and to verify the
assumed loss rate function. Development of antecedent floods could require data on both
annual-flood peak-flow rates and flood-hydrograph volumes. The necessary streamflow
data can be obtained from the USGS. In constructing or checking flood-peak frequency
curves, flood peaks should be segregated according to cause (e.g., thunderstorm,
hurricane, snowmelt, or rain-on-snow). It is particularly important to exclude floods
caused by ice jams or dam breaks.

Information about peak rates of flow and the time of peak of past large floods often is
helpful when evaluating the reliability of a unit hydrograph. Such information can be
obtained from staff gages or crest stage recorders, or from flood marks and other informal
flood records often available in special reports about major floods.

8-3.3 Precipitation Data

To develop the unit hydrograph, it is necessary to obtain precipitation data for the storms
that caused the identified historical floods. Precipitation data for rain gages within and
near the project basin can be obtained from the National Climatic Data Center or from
private vendors. Data from continuous recording gages (both within and near the basin)
are particularly important in assessing the temporal distribution of rainfall within the
basin when developing the unit hydrograph. The altitude and the period of record for all

-17-
rain gages should be noted. An isohyetal map of annual precipitation should be obtained,
if available. NEXRAD precipitation data can complement rain gage records.

Precipitation data for the periods preceding the historical floods will be required if a
special study is made to assess antecedent conditions. Special flood and PMP
studies—which may have been performed by the COE, USBR, NWS, or other federal or
state agencies— usually contain precipitation data that are more detailed and, in general,
more thoroughly reviewed and analyzed than that available from other sources (for which
supporting documentation are needed). Therefore, it may be beneficial to search for
information from such studies.

8-3.4 Applicable Hydrometeorological Reports

Knowledge of the hydrometeorology of the basin and its surrounding areas is necessary
to calculate the PMF. Applicable HMR’s (Hydrometeorological Reports) providing PMP
estimates for the region often include useful information on record storms and the
resulting floods. Sources of these data include:

C NWS
C FEMA
C State natural or water-resources related agencies
C Local flood-control districts
C Privately funded regional or site-specific studies that may have been
done for some nearby dams. The results of such studies might be
obtained from the dam owners.

8-3.5 Physical Characteristics of the Drainage Basin

Some of the parameters commonly used to define a watershed's runoff characteristics

include area, elevation, basin slope, land use, basin orientation, and slope and shape of
the major watercourse. Most of these parameters can be estimated using topographic
maps published by the USGS. Current and past aerial photographs can be very useful in
assessing land use or changes in land use. A site visit to the basin should be made to
support the parameters chosen for use in the PMF hydrograph development.

Information on soils classification within the basin is desirable for use in estimation of
applicable infiltration rates and can be determined from soil survey maps for the areas
published by the NRCS. A digital form of the STATSGO data for all 50 states is
available from, or is being prepared by, the NRCS National Cartographic and GIS Center
in Fort Worth, Texas. Land use data can be obtained from local government agencies, the
USFS (Forest Service), or the United States Bureau of Land Management if federal land

-18-
is involved. Future land use plans should be obtained and considered in the runoff
analysis if it is apparent that potential changes could have a significant effect on runoff
characteristics. Other resources are available such as Internet downloadable GIS
resources, e.g., Digital Elevation Models (DEMs), GIS software programs used to
develop basin characteristics from digital files, etc.

8-3.6 Snowpack Water Equivalent and Temperature Data

For sites where snowmelt contribution to extreme floods is possible, snowpack water
equivalent, wind speed, and temperature data must be obtained. Locations of snow
courses, snow pillows, and weather stations in and near the project basin need to be
identified and the altitude and period of record for these stations noted. If snowmelt
must be considered in the development of a unit hydrograph, both the snowpack water
equivalent and hourly and daily temperature data should be obtained for the periods
preceding and concurrent with the major historical floods identified in Sections 8-2.1 and
8-3.3. These data also may be necessary to develop snowpack and temperature sequences
for use in computing PMF runoff.

Aerial photographs showing the snow cover pattern throughout the winter and spring are
desirable for periods preceding the major historical floods identified in Sections 8-2.1 and
8-3.3, since it will be necessary to define the extent of snow cover for the runoff analysis.
The NW S has used aerial photographs to identify the extent of snow-covered areas in
some of the north-central states.

Snowpack water equivalent data, as well as NRCS SNOTEL data, may be obtained from
NRCS district or state offices or state water resource agencies. Daily temperature data
are available from the NWS NCDC. For modeling purposes, the maximum and minimum
daily temperatures are used to estimate an hourly temperature distribution.

8-3.7 Data on Existing Reservoirs, Spillways, Outlet Works, and Operation Policy

For an existing project, reservoir water levels, spillway gate operation, turbine releases,
and tailwater elevations recorded during passage of the identified historical floods should
be obtained—particularly if reverse-reservoir routing will be required to obtain an inflow
hydrograph. The operating policies for passage of extreme floods, which were in force
when the historic floods occurred, also should be obtained. To route the inflow PMF
through the reservoir, reservoir area-capacity data, rating curves for spillways and outlet
works, and flood-operation policy must be obtained from the dam's owner.

The rate of sediment deposition in the reservoir should be assessed to determine whether
the flood-storage capacity of the reservoir has been reduced. Historical information on

-19-
sediment deposition may be used to predict future loss of active storage if sediment
accumulation has been significant and is anticipated to continue.

Caution: It is important to note the date when this information was developed since
changes in active reservoir storage capacity or modifications to spillway and outlet
works may have since occurred.

8-4 Review and Assessment of Data

Before using the data obtained as described in Section 8-3 to develop the PMF for the
project basin, the data must be reviewed for accuracy and adequacy. This section
discusses the review process and acceptance criteria.

8-4.1 Unit Hydrographs

Any unit hydrograph available from a previous study for the project basin or from a
regional study must be reviewed and tested for its ability to accurately reproduce major
historical flood hydrographs. The best means of proving applicability of the unit
hydrograph is to use it to reconstitute the largest of the historical flood hydrographs
chosen for review. If the reconstituted flood hydrograph agrees well with the historical
flood hydrograph, the unit hydrograph normally can be accepted without adjustment.
Acceptance will depend on the historical flood magnitude, as discussed in Section 8-9. If
the available unit hydrograph does not reasonably reproduce major floods or is judged not
to do so due to changes in basin characteristics or error in the assumed time distribution
of rainfall excess, a new unit hydrograph will need to be developed. Unit-hydrograph
development is discussed in Sections 8-6 and 8-7.

Caution: It is important to determine the magnitude and importance of the flood

hydrographs that were used in producing the unit hydrograph. If the floods used were
not of major significance, the unit hydrograph may not accurately predict the peak and
timing of major floods. For this reason, small floods should not be used to develop the
unit hydrograph.

The predicted peak flow of the inflow PMF may be too low (or too high) as a result of
nonlinear effects in the runoff and the channel flow process that violate the unit-
hydrograph assumption of linearity between streamflow and excess rainfall. Studies
related to these nonlinear effects have been inconclusive [Pilgrim 1988]. If the historical
floods used in developing the representative unit hydrographs are large enough to be out-
of-bank, the nonlinear effects should not be significant.

-20-
8-4.2 Flood Data

The first task in the review of the flood data is to ensure that the historical floods used are
the largest for which records are available. They should be the largest floods of record
and should preferably have occurred during the season of the critical PMP. However,
floods caused by ice jams, debris blockage, or dam break should not be used in
unit-hydrograph analysis. It is important to note the cause of the floods
(e.g., thunderstorm, general storm, hurricane, snowmelt, or rain-on-snow).

The annual exceedance probability of each of these floods should be determined, as it is

preferred that the floods used for unit hydrograph development are significant flow
events. To determine the annual exceedance probability of each flood event, a frequency
curve of annual instantaneous peak flows is required. Frequency curves are typically a
part of Flood Insurance Studies prepared for FEMA and flood studies prepared by other
federal agencies. If a previously developed frequency curve cannot be found, one should
be developed. Flood flow frequency analyses should be made in accordance with the
latest methodology presented in Bulletin No. 17B. The Engineering Manual “Hydrologic
Frequency Analysis” EM 1110-2-1415 is a good reference to develop a frequency curve
in accordance with Bulletin 17B. The frequency curve(s) should be included in the
report.

Flood data must be reviewed for accuracy. The flood hydrograph should be plotted to
detect discontinuities and suspicious peaks or lows in the recorded flows. Historical gage
ratings, including methods used to extend the range for extreme floods, should be
reviewed to make certain that the conversion of recorded stage to discharge was correctly
done. Original stage records usually can be obtained from the local USGS district office
or the gage owner if questions arise regarding accuracy of recorded flood flows. The
following are two situations which need attention during review and assessment of flood
data:

C If a slope-area method was originally used to extend the rating curve, a

check should be made to ensure that the hydraulic control did not shift
to another location during the flood. This may require a computed
water-surface profile for the reach.

C If changes in watershed characteristics have occurred since the time of

the historical flood, adjustments may be necessary to adequately model
the new situation. For example, if the percentage of a watershed's
impervious area has changed, the input to the runoff model can be
adjusted to reflect the new percentage. Clear cutting of large areas of
forests may require changes in both initial abstractions and constant

-21-
 infiltration rates to reflect the changes. Such land use changes will
affect the unit hydrograph as well as losses.

Ideally, unit hydrographs should not be developed from storms that produced less than
1 inch of runoff or are not clearly overbank. Unit hydrographs for typical design storms
for conventionally engineered projects often are calibrated to flows mostly confined to
stream channels or stream channels with some overbank flows. However, PMF-type
floods often significantly exceed channel capacity and may become largely conveyed in
the overbank areas in which case Manning's roughness coefficients for submerged
overbank flows may affect flow travel times, flow depths, etc.

The available flood data should be presented in tabular form and include, but not be
limited to, the following: Date of Flood, Peak Flow (cfs), Rainfall associated with the
Flood, and Recurrence Interval of the flood event. If no floods have been recorded within
the basin of interest, flood records from other basins in the region will need to be
evaluated for applicability to unit-hydrograph development. This procedure is covered in
Section 8-7.

Caution: If questionable aspects of the flood data cannot be resolved, the data should
not be used in unit-hydrograph development.

8-4.3 Precipitation Data

Hyetographs for each storm at each recording rain gage should be plotted and examined
for consistency, continuity, accuracy, and completeness. Storm totals and the time
distributions for all rain-gage records should be compared to detect obvious
inconsistencies. Gaps in records can usually be filled by using regression and correlation
analysis with records from nearby gages. If a sufficient number of neighboring gages are
available, an average of several gages near the gage with the missing data, double-mass
analysis, or other methods may be used to fill the gaps in the record. An isohyetal map of
total rainfall for the storms of interest should be prepared using all acceptable rain-gage
records. The location of individual isohyets, for zones obviously influenced by
orographic effects, can be drawn parallel to elevation contours when the density of rain
gages is insufficient to clearly define the rainfall pattern throughout the area. The general
pattern should be compared to mean annual or 100-year isohyetal patterns, which can be
obtained from Technical Paper 40 or NOAA Atlas 2, published for individual states by
the NW S.

Comparisons of the hyetographs and the flood hydrograph should be made to identify
suspicious differences in timing between a storm's beginning and end and the rise,
recession, and peak of the flood hydrograph. The following are some situations which
should be resolved during review and assessment of precipitation data.

-22-
 C If a major timing difference is noted, additional study of the original
data records should be performed.

C The hyetographs from nearby rain gages should be checked to determine

if the timing difference is due to a clock problem with the rain gage or
the stage recorder.

C Rainfall records at the gage should be analyzed to detect any trends that
may coincide with changes in locations of gages or in conditions around
them.

C Double-mass analysis or regression methods may be used to adjust

rain-gage records to remove spurious trends and produce a
homogeneous rainfall record.

C All supporting data and information, including the hyetographs and

flood hydrographs, should be included in a graphical format to support
their use in the PMF determination.

Caution: Timing adjustments should not be made to the records unless the irregularity is
minor or the source of the error can be positively identified.

For this type of study, it is preferable to define the lag time as the elapsed time between
the centroid of the hyetograph and the peak of the flood hydrograph. Other definitions of
the lag time may be used with appropriate justification. The definition of lag time used in
a particular analysis should be consistent with the unit hydrograph method applied.

Because most rain gage records will be available only as daily totals, the records from the
most appropriate recording gage(s)— usually the nearest gage with a complete
record—should be used in disaggregating daily records to the required temporal
distribution. In assembling daily records, it is important to note the time at which each
daily gage was read, so that all daily totals can be adjusted to a common daily total.

8-4.4 Snowpack Data

Snowpack data will be required for those basins where snowmelt has been or may be a
contributing factor to major floods. The required snowpack-related data include the
portion of the basin covered by snow, water equivalent of the snow depth, and hourly or
daily minimum and maximum temperatures and wind speed.

-23-
8-4.4.1 Water-Equivalent Data

Snowpack water-equivalent data for snowcover that existed during historical storms
should be reviewed for completeness, consistency, and adequacy. Adequacy is
determined by plotting the recorded snowpack water-equivalent depths against elevation.
It is necessary to decide if data are sufficient to define an altitude-depth relationship for
the basin, including the lowest elevation of snowcover for mountainous regions. The
following are some situations for which additional data are required to estimate snow
water-equivalent data.

C If data are available from only one snow course in the basin, which
often is the case, data from other basins with a similar orientation and
exposure should be obtained.

C If applicable data from other snow courses are not available in sufficient
quantity at different altitudes, undefined portions of the
altitude-snowpack estimate can be proportioned in accordance with the
isohyetal maps for annual basin rainfall.

C It is possible to reconstitute snowpack data for historical floods through

the use of runoff models such as the Hydrological Simulation
Program-Fortran or the Sacramento Model [Burnash et al. 1973]. If no
snowpack data are available, but are required to study the historical
floods, such a procedure may be necessary.

8-4.4.2 Temperature Data

Temperature data can directly reflect the resulting snowmelt. Those data should be
reviewed for accuracy and for applicability in analyzing historical snowmelt events.

8-4.5 Data on Reservoir Volume, Spillway and Outlet-Works Capacity, and

Operation History and Policy

Data on the operating history and performance characteristics of the spillway and outlet
works, as well as on the reservoir storage volume, are required. Knowledge of operation
policies during extreme floods also will be required for routing the inflow PMF
hydrograph. The effect of floating debris on spillway gate operation with the potential of
a plugged gate must be considered for all dams that experience a significant amount of
debris under normal operating conditions.

8-4.5.1 Reservoir Volume

-24-
Data for reservoir area and volume should be reviewed for accuracy and possible changes
which may have occurred since the relationship was formulated. The following are
appropriate actions for various situations.

C Available data on sediment deposition in the active storage of the

reservoir should be reviewed to assess the need for adjustment of the
reservoir area and volume characteristics.

C If measured data are not available, visual observations of the reservoir's

upper reaches should be made.

8-4.5.2 Spillway and Outlet-Works Capacity

The discharge-capacity relationships of spillways and outlet works should be checked in

accordance with available discharge coefficients for tested hydraulic structures, such as
those given in the COE Hydraulic Design Criteria [COE 1989]. For unusual spillway
crest shapes, the USBR publication "Discharge Coefficients for Irregular Overfall
Spillways" [Bradley 1952] and the "Handbook of Hydraulics" [King and Brater 1954]
provide additional guidance. Because approach conditions and site-specific geometry can
affect the magnitude of the discharge coefficients, precise agreement should not be
expected but should be estimated within an acceptable allowance as described below.

C If differences of 10 percent or more are apparent, the source of the

original discharge-capacity estimates should be reviewed.

C If adequate physical model studies have been made for the structures,
experimentally determined discharge relationships can be accepted.

C If model studies have not been made, values from verified references for
discharge coefficients should be used for routing of the PM F inflow.
Determine if any structural modifications have been made that could
have produced a change.

C Ensure that a common datum has been used for elevations of reservoir
and the dam's appurtenances.

C The consideration of powerhouse discharges during a PMF may be

reasonable in some cases because of a wide variety of factors such as
general unit availability, headwater and tailwater levels, losing the load
through transmission outages, and sluicing through the units under no-
load conditions.

-25-
8-4.5.3 Operation History and Policy

Data on historical operation should be reviewed for correctness, especially if the data will
be used to determine inflow floods by reverse-reservoir routing. The location of the
reservoir stage recorder should be evaluated to ensure that measured stages are not
influenced by drawdown due to spillway or outlet works operation or wind-generated
waves. If stage records are available for any other location on the reservoir, the records
should be compared to detect any inconsistencies. This will aid in assessing the degree to
which the reservoir surface is sloped during passage of extreme floods.

It is necessary to review operation policy and procedures for the passage of extreme
floods to develop criteria for routing the inflow PMF hydrograph. The following
scenarios describe appropriate actions required for various situations.

C If it is possible for operators to be present at the project and to perform

the required operations during the PMF, and if redundant operation
systems exist, assume that gates and valves that have been tested under
head can be operated as proposed during flood passage.

C If gates and valves that would be operated during passage of an extreme

flood have not been tested under head to ensure their operation, it will
be necessary to make a detailed evaluation of their condition and
reliability. Assumptions on the operation of the gates during passage of
a PMF should be made based on the evaluation.

C If the gates are operated remotely, it is necessary to assess the reliability

of gate operation that can be expected during an extreme flood.
Operations during historical floods should be reviewed to determine
whether the operational policies have been consistently applied.

Spillways equipped with flashboards or stoplogs must be reviewed to determine the

operation policy relative to their installation and removal. In addition, if the flashboards
are designed to fail or collapse, it will be necessary to obtain detailed information on their
structural design. The head at which the flashboards will fail or collapse must be
checked.

C If the flashboards are designed to be tripped, the tripping operation

should be reviewed to ensure that it can be accomplished at the planned
time during passage of the design flood or larger floods.

-26-
 C If the spillway is sometimes blocked with stoplogs that must be removed
manually, it will be necessary to determine if there would be sufficient
warning time and available equipment to remove the stoplogs.

C It is important to consider the possibility that a spillway or outlet works

may be at least partially blocked by debris. The handling of debris
during past major floods should be assessed. If a debris-handling plan
that has worked successfully in the past is in place and there will not be
potential debris production in the areas previously untouched by flood
scouring, it is acceptable to assume that blockage will be insignificant
during passage of the PMF. If no debris plan exists, the potential loss of
spillway capacity must be evaluated with respect to the loss of spillway
capacity during the PMF event.

Caution: If deviations from the existing, approved reservoir operating plan are
proposed, the changes must be in accordance with the terms and conditions of the project
license. The appropriate Regional Office Engineer should be contacted to discuss the
proposed changes and obtain guidance concerning the potential for need to amend the
project license.

8-5 Approach to Tasks for Probable Maximum Flood Development

The approach and identification of tasks for PMF development depend on whether
available hydrologic and meteorologic data records for stations within the basin are
sufficient to provide for confidence in developing the PMF hydrograph. If not, the
available records must be supplemented with data or unit-hydrograph information from
other sources. The basin hydrometeorologic and runoff characteristics also have a role in
defining the types of analyses required for the PMF development. The choice of
procedures is governed by data availability and an understanding of the hydrologic
processes of the project basin developed through review and interpretation of the data
collected (Sections 8-3 & 8-4).

Unit-hydrograph theory is recommended for use in developing the PMF hydrograph. It

may be desirable to subdivide the basin to adequately treat hydrologic differences within
the basin. Some of the required subbasins may not have stream gage records at their
outlets which can be used to develop unit hydrographs for the subbasins. For cases where
the basin is subdivided, a runoff model must be developed that will incorporate the unit
hydrographs constructed for each subbasin, as well as the computations necessary to
route and combine flood flows from the subbasins to produce the required PMF
hydrograph.

8-5.1 Subdivision of Drainage Basin

-27-
Subdivision may be necessary for large basins that are not hydrologically homogeneous
or are drained by more than one major tributary. When records for the identified
historical floods of interest are available for more than one stream gage in the basin,
subdivision usually is advisable. If the reservoir area is relatively large compared to the
size of the basin, it should be considered a separate subbasin to allow consideration of
direct precipitation on the reservoir surface.

In the process of developing a unit hydrograph, and ultimately a PMF inflow hydrograph,
the calculations are made using average lumped conditions for the area. If parts of the
drainage basin have hydrologic conditions that differ significantly from the basin average,
subdivision should be considered. In such cases, separate analysis of the subbasins can
improve the confidence that an appropriate PMF inflow hydrograph has been developed.
Subdivision of large basins also is required to properly simulate the effects of spatial
distribution of precipitation. Subdivision also may be necessary if the methodology used
to simulate the flood event is limited to use on certain size watersheds. Subdivision also
should be considered if there are subbasins in the drainage basin that:

C Possess hydrologic characteristics obviously different from the average

characteristics of the total basin. Examples include shape; large urban
sections in an otherwise undeveloped drainage area; areas of unusually
high infiltration rates such as those of fractured basalt; closed subbasins;
and large areas of dense or managed forest in an otherwise clear
drainage area. Such hydrologic characteristics can be identified from
examination of soil maps, geological maps, topographic maps, aerial
photos, and land use maps, and from field visits.

C May contribute to delays in flood passage such as marshes, lakes, or

high-altitude meadows.

C Experience significantly greater or less rainfall than the basin average

due to orographic effects or spatial characteristics of local storms. Such
areas are best identified through study of isohyetal maps for individual
storms and average-annual rainfall.

C Are controlled by large natural constrictions that can act as hydraulic

control structures by restricting cross-sectional area and attenuating
water flow.

C Are upstream of dams with sufficient storage to affect the peak flow rate
and the timing of floods at the point of interest. Subdivision should

-28-
 definitely be considered if operational and streamflow records exist for
the upstream dam for the historical floods of interest.

C Have a total drainage area large enough that it may not be covered by a
single storm.

C Do not contribute to runoff from the basin.

C Have significantly steeper or flatter slopes than are typical for the basin.

C Have additional functional stream gages with good historical data.

C Have areas that are covered by snowpack when snowmelt is known to

be important for both historical floods and the PM F. The subbasin
covered by snow may have different infiltration rates than the rest of the
drainage basin.

8-5.2 Gaged and Ungaged (Sub) Basin(s)

It will be necessary to assess the available data and determine whether a basin can be
considered as "gaged" or "ungaged" to establish the recommended methodology to be
used in computing the inflow PMF hydrograph. For the purposes of this chapter, a gaged
basin (or subbasin) is defined as:

One for which available stream flow data (recorded at stations within the basin) and
precipitation data are sufficient in quantity and quality to provide for the development of
applicable unit hydrographs by enabling accurate calibration and verification with large
historical storms. A gaged basin should meet the following requirements:

C If the basin is not subdivided, at least one stream gage with available
flood records should be located within the basin, preferably at the inlet
to the reservoir. If the gage is located downstream of the dam, sufficient
historical operational data must be available to allow reverse-reservoir
routing to develop an inflow hydrograph for each recorded historical
flood.

C At least one rain gage—preferably a recording gage with complete,

correct, and consistent data, should be located within the project basin.
In the absence of rain gages within the basin, gages just outside the
basin may provide valuable information in regions not affected by

-29-
 orographic precipitation. If records for only one rain gage are available,
the catch of that gage should be representative of average basin rainfall.

C Concurrent records of runoff and basin rainfall for at least three severe
historical storms is preferred. The historical storms should occur in the
same season as the critical PMF, and should have the following
characteristics:

— All runoff-producing parts of the watershed should have contributed

runoff.

— The floods selected for analysis should not be snowmelt dominated,

unless it is apparent that the PMF also will be dominated by
snowmelt.

— The historical storms should have generated substantial runoff.

Ideally, the flood hydrograph should have at least one inch of runoff
from the contributing area and have generated significant overbank
flow along most reaches.

For the purposes of this chapter, a basin or subbasin should be treated as "ungaged" if it
does not meet this criteria. However, if available data include less than the desired
number of storms and corresponding flood hydrographs, all available data from within the
basin should still be used to the extent possible in the unit-hydrograph development.
Data from other drainage basins in the region that can be justified should also be used to
supplement the analysis. If no rain gages are located within the basin but flood data are
available, rainfall data from nearby stations can be used if a review indicates that the
data—and the results of their use in reconstituting historical flood hydrographs— are
acceptable. The general rule is that all site-specific data are potentially valuable and
should be evaluated for use.

If a basin is determined to be gaged , Section 8-6 should be used to develop the unit
hydrographs. Section 8-7 should be used to develop unit hydrographs for ungaged basins.
If a basin does not meet the criteria for gaged basins, it may have individual subbasins
that do meet this criteria. For these subbasins, the criteria of Section 8-6 should be
applied, and the criteria of Section 8-7 should be applied for the remaining subbasins.

8-5.3 Approach and Identification of Tasks

Once the basin is judged as "gaged" or "ungaged," the approach to developing the PMF
inflow hydrograph will be defined accordingly. However, there will be different degrees

-30-
to which available data within the basin can be used. The following briefly describes the
approach, depending on available data:

C Sufficient streamflow and rainfall data of satisfactory quality are

available for confidence in developing a unit hydrograph (gaged basin).
In this case, the approach will be to subdivide the basin as necessary and
to use available data to develop the necessary unit hydrograph and the
PMF inflow hydrograph. Details of the approach are given in Section
8-6.

C Stream gage records for major historical floods are available, but
available rainfall data are insufficient to develop a unit hydrograph. In
this case, rainfall data from gages adjoining the study basin, which
recorded the same storm, may be transposed. However, a test for
applicability of this transposed rainfall data will be whether or not it
allows satisfactory reconstitution of historical flood hydrographs.

C Available streamflow data are insufficient to provide confidence in

developing a unit hydrograph. In this case, it will be necessary to
follow the guidelines for "ungaged" sites as described in Section 8-7. If
any data for major historical floods are available in the basin
(e.g., gages, flood marks, informal flood records), they may be valuable
in verifying the unit hydrograph's applicability.

C In some cases, where the "ungaged" approach is indicated, it may be

possible to use a unit hydrograph developed in other studies or
generalized unit-hydrograph parameters developed in regional studies.
This possibility is discussed in Section 8-7.

8-6 Unit Hydrograph for Gaged (Sub) Basins

The methods described in the following paragraphs denote the preferred methodology for
developing unit hydrographs for gaged basins. Other methods may be applicable, but
they must be fully described, justified, and documented.

Section 8-5.2 discusses the criteria for a basin or subbasin to be considered gaged. The
COE-developed computer program HEC-1 Flood Hydrograph Package, (or the
subsequent COE-developed Hydrologic Modeling System HEC-HMS) is recommended
for use in developing unit hydrographs for gaged basins and PMF inflow hydrographs.

Programs with capabilities similar to HEC-1 have been developed by other agencies.
Some of these programs have unique capabilities, or incorporate data or relationships

-31-
applicable to specific regions of the United States. For example, the Tennessee Valley
Authority (TVA) has developed computer programs that are specific to the Tennessee
River Basin. Similarly, the Los Angeles District of the COE has developed a
preprocessor program for HEC-1 that incorporates unit hydrographs for the District's
entire region. Other programs may be used but must be fully documented and verified.

If regional studies that have produced accepted results are available, the methods
presented in those studies may be used, if justified. Use of the regional unit hydrographs
in developing the PMF inflow hydrograph is described in Section 8-7.2.

8-6.1 Historical Floods for Calibration and Verification

Data from severe historical storms and the resulting floods that are available from
systematic gaging stations should be considered for use in developing unit hydrographs.
Flood hydrographs resulting from single extreme rainfall events with uniform temporal
and spatial distributions are the most desirable for use in unit-hydrograph computation. A
unit hydrograph developed from a complex storm (i.e., multiple events occurring
back-to-back) can be in error and is very difficult to compute, primarily because of
problems associated with baseflow separation. However, HEC-1 and HEC-HMS do
provide the means to satisfactorily analyze flood hydrographs that are not single-peaked,
if required.

It will be difficult to develop a unit hydrograph that generally reproduces all portions of
all historical flood hydrographs. The adopted unit hydrograph should be the one judged
to best predict the magnitude, shape, and timing of the PMF. Normally, the adopted unit
hydrograph should be the one that most faithfully reproduces the largest floods of record
without under-prediction of the historical peak flows. If only historical flood peak
discharge and time-to-peak data are available, it may be advisable to attempt calibration
to that data, assuming a triangular-shaped hydrograph. This may be appropriate if
application of historical rainfall with synthetic unit-hydrograph parameters do not provide
a good match with the available data.

The greater the number of storms and floods that can be used, the greater the confidence
in the developed unit hydrograph. If data from at least three historical floods are
available, two should be used for calibration of the unit hydrograph and one for
verification. For calibration, unit-hydrograph parameters are computed by analyzing the
largest floods with the best (i.e. most reliable) data to develop a representative unit
hydrograph; the degree to which the representative unit hydrograph provides for
duplication of the verification flood(s) is then assessed. If not, then the unit-hydrograph
parameters must be reviewed and modified to improve the fit. When computing the
average depth of runoff for unit-hydrograph analysis, care must be taken to exclude those

-32-
areas that do not drain to the river system. The runoff-contributing areas for each flood
should be identified.

Ideally, the floods calibrated for unit-hydrograph development should have occurred
during the season when the critical PMP is likely to occur. In choosing the floods to be
used for calibration and verification, the distinction between rain-on-snow and
rainfall-generated floods should be noted. For the same basin, a rain-on-snow flood will
exhibit a longer lag time than an equivalent flood produced by rainfall alone. If the
critical PMP will occur during a month when a significant part of the basin will be
covered by snow, the calibration floods should include historical floods generated by
rain-on-snow. However, if the critical PMP will occur during summer months when
snow cover is unlikely, the calibration floods should be selected from rainfall-dominated
floods. In analyzing major floods that occurred during a cold season, it will be desirable
to judge whether or not the ground was frozen, since frozen ground may have reduced
infiltration rates.

8-6.2 Determination of Basin Average Rainfall

Basin average rainfall must be determined for each storm used in developing a unit
hydrograph. The method to be used in determining basin average rainfall depends on
whether orographic effects are present in the basin.

C If orographic effects are not important, either the Thiessen polygon or

the distance-averaging method can be used to calculate the basin
average precipitation using recorded rainfall at each gage.

C For basins where orographic effects are important, an isohyetal map

provides the best means to determine basin average rainfall. For
watersheds having drainage areas in high altitudes, it is important to
define the runoff contributing area on the basis of the rain/snow
interface line. The basin average rainfall is determined by integrating the
areas between isohyets in the subbasin. W hen orographic effects could
be significant, a meteorologist may need to be involved in the
development of the basin average rainfall depth.

HEC-1 will compute basin average precipitation from individual gage records, if a
weighting factor is entered for each rain gage. W hen multiplied by the recorded rainfall
depth at the gage, the weighting factors yield the portion of the basin average (or subbasin
average) rainfall contributed by the gage reading. The weighting factors must be
externally computed from the results of either the Thiessen polygon or isohyetal methods.
The height-balance polygons method may be needed for a mountainous drainage basin.

-33-
Caution: Separate weights may be required to (1) determine total storm volume and (2)
develop a temporal distribution of the rainfall depending on the averaging method used.

8-6.3 Cold Season Considerations

It should be determined if at least part of the basin had snowpack or ground subject to
frost during historical floods.

8-6.3.1 Snowmelt Considerations

If the basin is one for which at least part of the drainage area is subject to snowpack, and
if the historical rainfall-generated floods were influenced by snowmelt, snowmelt
calculations must be included in the rainfall-runoff simulation process. The area covered
by snow at the time of the flood-producing storm must be determined from the acquired
data.

To use the snowmelt function of HEC-1, the temperature at the base elevation of the
snowpack is required along with a temperature - lapse rate. For mountainous areas, the
elevation usually is taken in increments of 1,000 feet and the lapse rates are given in
increments of degree change per 1,000 feet.

C If sufficient temperature information is not available to construct a lapse

rate for each storm, a rate of 3°F per 1,000 feet may be used. A 3.5°F
lapse rate is used by the COE in the Northwest as a recommended value
for the SSARR program. A 3°F lapse rate is typically used in
California.

C The energy-budget method of snowmelt computation is recommended

for calibration of historical floods. Alternative methods exist and may
be used if properly documented and justified. Recommended values for
use in snowmelt calculations can be obtained from the U.S. Army Corps
of Engineers Snowmelt Manual EM 1110-2-1406 [COE 1960].

C Precipitation should be assumed to fall as snow above the elevation at

which it is 34°F. HEC-1 makes this assumption.

Snowmelt from large, relatively flat areas such as the northern Midwest are calculated by
HEC-1 in the same manner as for mountainous areas, but temperatures will be more
uniform across the area. Areas covered by forests, which will be covered by humus
beneath the snow cover, will tend to have higher retention and infiltration rates. HEC-1
provides the capability to consider snowmelt in up to 10 zones of equal increments of
elevation.

-34-
8-6.3.2 Infiltration Characteristics of Potentially Frozen Soils

In some basins, extreme historical floods result from rain on frozen soils. It may be
important to consider these events in unit-hydrograph analysis, especially if the PM F is
considered to have a high probability of occurring with frozen-soil conditions. Loss rates
for frozen soil conditions can vary considerably depending on the type of soil and the
presence of other factors such as forests, wetlands, and high groundwater tables. This is
discussed in detail in Section 8-8.4.

8-6.4 Base-Flow Separation

Separation of baseflow from direct runoff in unit-hydrograph analysis has been done in
several different ways, none of which are exact. For these guidelines, the procedures
specified in HEC-1 should be used. Three parameters must be determined from the
recorded flood data and used as input to separate direct runoff and baseflow:

C The flow rate at the beginning of runoff simulation, STRTQ.

C The value of flow at which direct runoff ceases, QRCSN.

C The recession characteristic, RTIOR.

As an aid in calculating these parameters, logarithms of recorded flows during the

hydrograph recession should be plotted against the time at one-hour intervals (semilog
plot). QRCSN is taken as the flow rate at which the plot of the recession deviates from a
straight line and RTIOR is taken as the slope of the straight line portion of the plot.

Caution: Choosing QRCSN can have an important effect on the ordinates of the unit
hydrograph and will involve judgment, since the plots are not always smooth and the
deviation often is gradual. Figure 8-6.1 shows the way in which baseflow and surface
runoff are separated in HEC-1.

8-6.5 Time of Concentration and Clark's Storage Coefficient for Each Subbasin

HEC-1 will calculate values of the time of concentration T C and storage coefficient R to
provide a unit hydrograph which yields, by transformation, an optimized fit to a recorded
flood hydrograph [HEC 1990]. R is a coefficient reflecting the effect of storage in the
basin and is described in Clark's original paper [Clark 1943]. The time of concentration,
T C , is defined as the time between the cessation of runoff-producing precipitation and the
time of the inflection point on the recession limb of the direct runoff hydrograph at which
the minimum value of R occurs. As shown in Figure 8-6.2, the value of R may be

-35-
estimated by dividing the discharge by the rate of change of the discharge at the inflection
point on the recession limb of the direct runoff hydrograph.

The ratio R/(T C + R) tends to be approximately constant for hydrologically similar

drainage basins in a region. Values for R and T C can be computed for input into HEC-1.
Using the optimization capability of HEC-1, rainfall and resulting flood flows can be
input to the program and values of R/(T C + R) and T C are automatically computed so that
the unit-hydrograph shape is optimized to produce a best fit between recorded and
simulated flood flows. HEC-1 also computes separate values for R and T C , which should
be checked against those estimated from drainage-basin characteristics. If the agreement
is good, the value of R/(T C + R) should be kept constant in the hydrograph analysis. To
check HEC-1-derived values for T C , the time between the end of rainfall excess and the
point of inflection (as plotted on the recession hydrograph) should be scaled for each
storm and related flood hydrograph. If the value of T C computed by HEC-1 differs
significantly from the scaled value, both should be reviewed and the calculations verified.
The scaled value should control, unless a clock-synchronization error is found in either
the rainfall or streamflow records.

In addition, a check can be made by calculating T C using hydraulic theory. This is done
by dividing the watercourse from the basin outlet to the top of the basin into segments of
approximately uniform slope; USGS quadrangle maps are adequate for this purpose. The
time of travel through the various portions of the flow path can be estimated using
methods developed by the NRCS [SCS 1986]. Average velocity of flow through each
channel reach can be estimated using the Manning equation. Appropriate flow depths can
be assumed and Manning's "n" values can be estimated using the USGS publications
“Roughness Characteristics of Natural Channels” Barnes [Barnes 1967] and Water
Supply Paper No. 2239 “Guide for Selecting Roughness Coefficients for Natural
Channels and Flood Plains” [USGS 1988], or the U.S. Department of Transportation’s
report entitled “Guide for Selecting Manning’s Roughness Coefficients for Natural
Channels and Flood Plains” [FHWA 1984]. Time of travel in each reach is calculated as
the length of the reach divided by the average velocity in the reach.

A value for R, the storage coefficient in Clark's unit hydrograph, can be calculated by
examination of the observed flood hydrograph as illustrated in Figure 8-6.2. This value
of R is not required for the unit-hydrograph determination but should be estimated for
comparison with the value calculated by HEC-1 after the unit hydrograph has been
optimized using the constant value of R/(T C + R).

8-6.6 Rainfall Sequence for Recorded Storms

The maximum time increment for the rainfall to be used in the unit-hydrograph analysis
is usually calculated as T L/5.5 rounded down to an even number, where T L is lag time.

-36-
This limitation will generally ensure numerical accuracy in the development of the unit
hydrograph and the flood hydrograph.

Caution: Sensitivity studies on the effect of the time increment on computational

accuracy should be performed if there is any indication that a shorter time increment
would result in a higher peak.

Temporal distribution of the basin-average rainfall must be developed for input to HEC-1.
This should be done by distributing the calculated basin average rainfall in accordance
with records from the nearest recording gage.

Caution: For basins where there may be more than one recording gage, it may be
appropriate to subdivide the basin and use the temporal distributions for each gage as
input to HEC-1 for the respective subbasin. Averaging recording gage readings usually
is not appropriate and must be justified.

8-6.7 Infiltration for Unit-Hydrograph Development

The initial-abstraction and uniform-loss-rate method of simulating infiltration is

recommended since it is easy to use, approximates an exponential loss function, and
provides sufficient precision. The value of uniform infiltration calculated by HEC-1 for
the historical floods should be checked against those expected for the soil types in the
basin. This check will provide an indication as to whether the values determined by
HEC-1, in the unit-hydrograph optimization process, are consistent with the basin
characteristics.

A detailed discussion of the selection of loss rates for the PMF runoff calculations may
be found in Section 8-8.

8-6.8 Calibrate Unit Hydrograph

Unit hydrographs must be generated for each historical flood chosen for calibration. The
way in which this is accomplished will depend on whether or not the basin is subdivided
and the number of stream gages present in the basin.

The flood peak estimate is very sensitive to T C . Also, the shape of a unit hydrograph can
change depending on the magnitude of a flood event. Therefore, for best results, unit
hydrographs should be calibrated to floods with overbank flow for most channels in the
basin. In general, 2-year floods or less are dominated by mostly channel hydraulics, 10 to
20-year floods will have some overbank hydraulics, and 50 to 100-year floods will have
substantial overbank and valley storage hydraulics. The hydraulics of overbank flow is

-37-
significantly different from channel flow due to increased surface roughness of the flow
boundary.

Caution: There are several sources of error that can affect the acceptability of a unit
hydrograph. A major potential source of error is the estimate of the temporal
distribution of rainfall excess. This estimate depends on the validity of the assumption of
basin average rainfall, the estimated temporal distribution of rainfall, and the selection
and variability of the infiltration rate. The adopted temporal and/or spatial distribution
of rainfall may be erroneous because of clock-synchronization errors, or because of an
insufficient number of rain gages to allow for accurate assessment. Given the temporal
distribution of rainfall, estimates of the precipitation rainfall excess for a given time
depend on the selection of the infiltration rate for that period. All of these assumptions
may make it difficult or impossible to develop a unit hydrograph that satisfactorily
reconstitutes a major historical flood hydrograph that then may be verified by
reproducing another historical flood. The hydrologic engineer needs to be alert to such
problems and use engineering judgment as appropriate.

8-6.8.1 Cases Where a Single Basin Unit Hydrograph is Sufficient (No Subdivision)

The rainfall input sequences, as calculated in Section 8-6.6, should be used with the
corresponding streamflow sequence and the hydrograph parameters computed in Sections
8-6.3, 8-6.4, and 8-6.5. The value of R/(T C + R) is calculated from the estimated values
of T C and R or adapted on the basis of available regional values. The parameter can be
fixed or allowed to vary when using HEC-1 to develop unit hydrographs. HEC-1 should
be programmed to optimize all parameters [HEC 1990a, Section 5] of the hydrographs.
For each calculated unit hydrograph, check the HEC-1-calculated values for T C , R, and
the uniform infiltration rate with the values estimated.

Caution: Since HEC-1 makes only a limited number of iterations in this optimization
process, more than one trial may be necessary to enable the program to reach an
optimum fit. The value of R/(T C + R) produced by HEC-1 should be input into
subsequent runs to ensure that a best fit, in terms of HEC-1 capabilities, has been
obtained.

Caution: If the estimated values of T C or R differ substantially from those calculated

(Section 8-6.5), review the calculation of those values. Calculated values for T C , because
of its physical relevance, should be a guide to the final value of R/(T C + R) chosen as
correct for the unit hydrograph.

If the reconstituted historical hydrographs compare well with the recorded hydrographs,
no further adjustment of the unit-hydrograph parameters will be necessary. However, if
the computed peak is too low, the hydrograph shape is poor, or the calculated values of R

-38-
or T C differ greatly from the original estimates, the input parameters should be revised
and HEC-1 should be rerun to compute a new hydrograph. This process should be
repeated until the fit between the reconstituted and recorded hydrographs can no longer
be improved.

A representative unit hydrograph should be prepared using the individual unit

hydrographs developed with HEC-1. In general, the representative unit hydrograph
should be based on the largest historical flood that occurred during the season of the
critical PMP. The representative unit hydrograph can be obtained by adopting
appropriate values (T C and R) from calibrations as opposed to manually adjusting the
unit-hydrograph ordinates.

Cautions: If adjustments to the representative unit-hydrograph peak and base are made,
the ordinates of the unit hydrograph will need to be adjusted to preserve a runoff volume
of 1 inch of rainfall excess.

8-6.8.2 Unit Hydrographs for Subbasins and Channel Routing

If the drainage area basin is to be subdivided, it will be necessary to compute runoff from
each subbasin and to route and combine runoff from the subbasins in the downstream
direction to develop the hydrograph at the basin outlet.

C If streamflow records are available for each subbasin, the entire process
of optimizing the unit hydrograph for each subbasin is the same as
described in Section 8-6.8.1.

C If streamflow records are not available at the outlet of some subbasins, it

will be necessary to estimate unit hydrographs for these ungaged
subbasins. For subbasins smaller than 20 mi2 , this can usually be done
with sufficient accuracy by using, for example, SCS dimensionless
synthetic unit hydrographs, which requires only an estimate of lag times
for the subbasins. For subbasins larger than 20 mi2 , a unit hydrograph
can be developed following the procedures described in Section 8-7.

C If a regional value for R/(T C + R) is available, it can be used to estimate

T C at the outlet of each subbasin.

The Muskingum-Cunge method of routing, as incorporated in HEC-1, is recommended

for channel routing of outflow from each subbasin. Channel cross sections required for
the routing can usually be obtained with sufficient accuracy by scaling measurements and
elevations from 7½ -minute USGS quadrangle maps.

-39-
Manning's roughness coefficients, required as input to the routing process, must be
estimated on the basis of field observations of the streams. Particular care should be
taken to select appropriate "n" values for overbank flow areas. The USGS publications
"Roughness Characteristics of Natural Channels" [Barnes 1967] and Water Supply Paper
No. 2339 "Guide For Selecting Roughness Coefficients for Natural Channels and Flood
Plains", or the U.S. Department of Transportation's report, FHWA-TS-84-204, entitled
“Guide for Selecting Manning's Roughness Coefficients for Natural Channels and Flood
Plains” can be used to aid in evaluating roughness coefficients. Also, Ven Te Chow's
"Open-Channel Hydraulics" (1959) provides guidance for choosing Manning's "n" values.
HEC-1 includes the capability of combining hydrographs in the downstream direction.
The combining and routing of the unit hydrographs forms a single-event runoff model for
the basin.

Caution: The Muskingum-Cunge routing method uses a single representative cross

section defined by eight coordinate points for each routing reach. The method cannot
account for backwater effects and should not be used when attenuation of the hydrograph
is expected. An example of where this technique might be used is to translate the
hydrograph from gages downstream. Where the intention is to properly model the
attenuation of the hydrograph, dynamic-wave routing is the preferred method (e.g., when
the river is expanding or contracting or where there is natural storage).

Calibration with the historical outflow hydrograph is accomplished differently when

routing is involved, because the runoff from each subbasin must be routed and/or
combined in the downstream direction to produce the total inflow hydrograph. The
agreement between the recorded and reconstituted hydrographs should be examined; if
differences are unacceptable, adjustments must be made to the routing parameters and/or
the unit hydrograph parameters for each subbasin. The unit hydrograph for each
subbasin, if the subbasin is gaged, is also calibrated by checking the accuracy with which
its use reproduces the recorded historical floods.

8-6.9 Hydrograph Verification

Once calibration of the unit hydrographs has been completed, the representative unit
hydrograph (or the runoff model consisting of the subbasin unit hydrographs and routing
calculations) is used with the corresponding basin average rainfall in an attempt to
reproduce the historical flood or floods chosen for verification. If the historical
hydrographs are duplicated well, the representative unit hydrograph can be accepted.
Checking between the historical hydrograph and the generated verification hydrograph
can be done automatically with HEC-1 in terms of statistical differences.

C A plot showing a comparison of the hydrographs should be included in

the study.

-40-
 C For the case where a single representative unit hydrograph is involved,
only adjustments to the unit-hydrograph parameters will be required.
Parameters should only be adjusted within appropriate ranges that can
be justified.

C For subdivided basins where the hydrograph generation involves a

runoff model, adjustments to both unit-hydrograph parameters and the
routing parameters may be needed to achieve better agreement with the
historical flood hydrograph.

It is important to be certain that any adjustments to the unit hydrographs or other

runoff-model parameters do not significantly decrease the degree of fit achieved in
Section 8-6.8 for the historical flood hydrograph. However, the verification process
should be continued until an acceptable fit is achieved.

-41-
Figure 8-6.1 Baseflow Simulation in HEC-1
[HEC 1990a]

-42-
Figure 8-6.2 Estimation of Clark Unit-Hydrograph Parameters
[HEC 1982]

-43-
8-7 Unit Hydrographs for Ungaged (Sub)Basin(s)

When a basin does not meet the criteria listed in Section 8-5.2 for gaged basins, it is
considered to be an ungaged basin, and a unit hydrograph must be developed
synthetically. One of the following approaches should be followed.

C A search should be conducted for regional studies that have developed

synthetic unit-hydrographs applicable to the basin.

C A regional study should be performed to develop synthetic

unit-hydrograph procedures. The study could develop either a new
approach or coefficients for an existing one.

C If no suitable data are available for a regional study, one of the existing
approaches should be used, such as those developed by Snyder, Clark,
the NRCS , or others. In this situation, the required coefficients must be
selected empirically based on coefficients developed for other regions.
The applicability of the adopted coefficients must be justified and
documented.

C For drainage areas smaller than 20 square miles, it is acceptable to use

the SCS (NRCS) dimensionless unit hydrograph; however, adjustments
may be necessary depending on basin characteristics (e.g., flat slopes).
For basins larger than 20 square miles, an aggregate method can be used
if justified and documented.

In a regional analysis, unit hydrographs are developed for gaged drainage basins in the
region. A representative unit-hydrograph model is adopted. Relations between the
parameters of the unit-hydrograph model and the physical characteristics of the basin are
developed. Synthetic unit hydrographs are developed for ungaged basins by means of
these established relationships between parameters of the unit-hydrograph model and the
physical characteristics of the basin.

Caution: The applicability of any method to an ungaged basin is always subject to

question because of the fundamental uncertainty in predicting basin response in terms of
defined physical characteristics. In general, any synthetic unit hydrograph should not be
used unless the parameters for the unit hydrograph are well defined and correlated with
quantifiable basin characteristics, and the unit hydrographs used in developing the
relationships have been verified by reproducing the largest historical floods in the
records.

-44-
Any historical rainfall or peak flow data from within the basin must be used to verify
regional synthetic hydrographs and determine their applicability to the basin. Thus, it is
always important to use all data available from stations within the basin when developing
a PMF hydrograph.

8-7.1 Applicable Unit Hydrograph Procedures for Each Basin (Subbasin)

Many general studies have been performed by local, state, and federal agencies to
develop synthetic unit-hydrograph procedures, or coefficients for existing synthetic unit-
hydrograph procedures, applicable to a particular region. The following are a few
examples of regional studies available from federal, state, and local agencies for
developing synthetic unit-hydrograph procedures for ungaged basins.

The COE has developed coefficients for use in computing Snyder and Clark unit
hydrographs for many areas in the United States. There is no single source for the
COE-developed information, but district offices of the COE can provide information on
the results of any studies conducted in their districts.

The USBR has developed a set of lag-time equations, dimensionless unit hydrographs,
and S-graphs for different parts of the western states [Cudworth 1989].

The USGS has performed a number of statewide regional studies for the development of
unit hydrographs in cooperation with state departments of transportation. These are
published as USGS water resources investigation reports. Some of these are referenced
in Section 8-11 [USGS 1982, 1986, 1988, 1990].

Caution: Any information obtained must be carefully reviewed to determine if it is

applicable to the project basin.

C A first check is to assess whether the basin of interest is hydrologically

similar to those used in the regional study. For instance, if the available
regional study was developed for basins in a rural setting, the study's
applicability to watersheds in an urban environment would be
questionable, or vice versa.

Caution: The reviewer must keep in mind that adjoining basins often
are not hydrologically similar even though they may adjoin. Any
differences in drainage area, cover, soil type, orientation, or geology
should be identified.

C Storm and flood data used in the regional study should meet the same
quality requirements as set forth in Section 8-6 for the development of

-45-
 unit hydrographs for "gaged" basins, including the consideration of
adjusting unit hydrographs for possible nonlinearity.

C The terminology used to define the various unit hydrographs and basin
parameters in the regional study should be clearly
understood—particularly the definitions of lag time and channel slope,
since a misunderstanding could lead to development of an invalid unit
hydrograph.

Caution: Lag time and channel or basin slope often are defined
differently in the various methodologies. The definition of the
parameters must be consistent with the methodology used.

C In the Snyder unit hydrograph (Equation 8-7.2), the lag time is defined
as the elapsed time from the centroid of the rainfall excess to the
unit-hydrograph peak, which is the same definition used by the NRCS.

C The USBR defines the lag time as the time from the center of the unit
rainfall excess to the time that 50 percent of the volume of the unit
runoff from the basin has passed the concentration point.

C The Los Angeles District of the COE defines the lag time as the time
from the beginning of the unit rainfall excess to the instant the resulting
hydrograph reaches 50% of the ultimate discharge.

Caution: The hydrologic engineer must have a clear understanding of the definitions of
all parameters involved, if using methodologies or studies developed by others. For
instance, since many unit hydrographs prepared in the past by federal agencies are
based on 6-hour durations, it will be necessary to change the unit duration for the
specific duration of the PMP under study.

The capability of a developed unit hydrograph to reconstitute major historical flood

hydrographs must be assessed. If reconstitutions were successfully performed in the
available study, the unit hydrograph may be acceptable for application to the basin of
interest. It also will be desirable to use the unit hydrograph to reconstitute a major
historical flood hydrograph on the basin of interest if data are available. If the results of
that reconstitution are satisfactory, the unit hydrograph may be acceptable.

Upon obtaining parameters from an acceptable regional study, unit hydrographs for each
subbasin should be developed in accordance with the application of the regional study or,
in the absence of specific directions, according to common unit-hydrograph theory.

-46-
-47-
8-7.2 Regional Analysis

If the search for applicable synthetic unit-hydrograph procedures for the basin of interest
proves to be fruitless, and the drainage basin is larger than about 100 square miles, a
regional analysis will be required.

A regional study could be either relatively easy or require a substantial effort, depending
on available data. For regions where systematic records of both rainfall and streamflow
have been carefully kept and are readily available, the effort may be as simple as plotting
graphs of peak-flow rate and lag time against drainage area; otherwise, the effort can
involve significant time and expenditure.

Regional unit-hydrograph studies generally are performed by developing unit

hydrographs for historical storms on "gaged" basins within the region. The process of
developing unit hydrographs for gaged basins is described in Section 8-6 for basins with
adequate data. In the final analysis, the parameters defining the developed unit
hydrographs are correlated with measurable basin characteristics to determine if an
analytical relationship can be formulated. If the hydrograph parameters correlate well
with basin characteristics, the results can then be used to generate unit hydrographs for
the ungaged basin of interest.

Caution: Similarity of the study basin to the "gaged" basin is required for a regional
analysis to produce reasonable results. The study basin must be similar to the "gaged"
basin in topography, slope, soil type, infiltration rates, elevation, and land cover. If
the "gaged" basin differs significantly from the study basin in physical properties, it is
not an appropriate "gaged" basin to be used in a regional analysis.

To conduct a regional study, "gaged" basins in the region need to be identified. The need
for and sources of data for development of unit hydrographs for such basins in the region
are the same as given in Section 8-6. Data review should follow the procedures given in
Section 8-4. Unit hydrographs used in regional studies should be developed only at
gaged sites, and not by some form of transfer or inference from a gaged site to an
upstream, downstream, or similar site.

8-7.2.1 Data Required

To evaluate the hydrograph parameters needed for input to HEC-1, an analysis of data for
"gaged" basins in the region is required. Rainfall and flood records for all basins in the
region should be obtained and examined. Since the objective is to develop a unit
hydrograph that can be used to determine the inflow PMF hydrograph, the data obtained
should include those indicated in Section 8-3 Data Acquisition. Also, the basins should
be visited to obtain information on land use, cover, and the physical characteristics of any

-48-
dams and reservoirs. If there are dams in any of the basins, information on reservoir area
and volume, spillway and outlet works capacity, and operation during historical floods
should be obtained.

The following parameters have been found to be useful for correlation of unit-hydrograph
parameters in regional analyses:

C Drainage area (A).

C Length of the longest watercourse in miles from the basin outlet to the
upper limit of the basin (L).

C Length of the main watercourse in miles from the basin outlet to the
point nearest the centroid of the basin area (L ca).

C Channel slope (S).

C Percent impervious area (A I).

C Percent of area covered by forest.

C Percent of area covered by lakes or marshes.

For each basin analyzed, the following parameters should be computed.

C An estimate of lag time TL and time of concentration TC for each basin

based on applicable equations obtained from the local flood-control
agencies, or calculated as described in Section 8-6.

C The maximum time increment of rainfall to be used in the

unit-hydrograph analysis is TL/5.5 rounded to the next lower even
number.

C Infiltration rates for each basin/subbasin using methods described in

Sections 8-6.3.2 and 8-6.7.

Caution: Because it does not increase the accuracy of the unit hydrograph for the basin,
subdivision to areas smaller than that represented by a recording stream gage should not
be done.

-49-
8-7.2.2 Rainfall Analysis

Basin average rainfall should be computed using the procedures described in

Section 8-6.2.
Temporal distribution of rainfall for each storm should be developed for each basin using
the procedures described in Section 8-6.6.

8-7.2.3 Development of Generalized Regional Relationships

HEC-1 and the Clark unit-hydrograph method should be used to develop representative
unit hydrographs for the selected basins with available data. The selection of the basins
should be justified. In general, it is desirable to have gage data for at least four basins in
the region. Parameters for use with the Clark unit hydrograph should be developed from
the basin data, including Clark's storage coefficient R, and the time of concentration TC .
In addition, it will be necessary to evaluate the HEC-1 baseflow separation parameters
STRTQ, RTIOR, and QRCSN. Procedures for determining these parameters are given in
Sections 8-6.4 and 8-6.5. Once all input information has been entered, HEC-1 should be
used to optimize a unit hydrograph for each selected basin. The HEC-1 runs for each
basin should be programmed to optimize the hydrograph parameters while allowing
R/(T C + R) to vary. A representative unit hydrograph must be developed for each basin
analyzed.

Once a representative unit hydrograph has been developed for each basin analyzed, the
values of R/(T C + R) for all of the basins should be used in a regression analysis against
basin parameters. A very simple regression analysis could be performed by plotting
values of peak flow and lag time against drainage area on semi-log or log-log paper. If a
well-defined relationship is found, the results can be used to develop a representative unit
hydrograph for the project basin.

If a well-defined relationship is not found in the simple regression analysis, it may be that
parameters other than drainage area have a strong influence in determining the peak flow
rate and lag time for basins in the region. In that case, it will be necessary to perform a
multiple linear regression of T C and R/(T C + R) against identifiable basin parameters, such
as S, L, Lca, and A, or combinations of these parameters. If a portion of the basin is
impervious, a measure of that parameter—such as the basin's percentage of impervious
drainage area— should be included in the regression analysis. If lakes or marshes exist in
the basins, it may also be necessary to include the percent of drainage area occupied and
controlled by lakes and marshes as an independent parameter.

A multiple linear regression program will yield values of the coefficient of determination.
The coefficient of determination provides a measure of the degree to which the
independent variables influence the value of the dependent variable. The regression

-50-
analysis should be started using all independent parameters and then eliminating those
with little influence on the value of the dependent parameter. For basins where
impervious areas are small enough to be considered insignificant, the resulting equation
for T C or (T C + R) may have the form

(8-7.1)

where C 1 and C 2 are constants determined in the regression. Ideally, the value of the
coefficient of determination will be equal to or greater than 0.9; a perfect correlation
would yield a value of 1.0.

Caution: In actuality, the value of the coefficient of determination will often range from
0.6 to 0.8. Different values of the regression constants will be determined for each set of
independent variables included in the regression. The hydrologic engineer should review
the derived relationships for consistency and use the equation that yields the smallest
value of standard error of estimate and the largest value of the coefficient of
determination.

Caution: Since R/(T C + R) tends to be constant for a region, it may not be statistically
significant in a regression analysis. In that case, an average value for the region should
be computed from the regional results and used for the analysis of the project basin. In
either event, the selected values should be justified.

Once the regression analysis has been completed, the values of T C , R, and R/(T C + R) can
be computed for the project basin in terms of the basin characteristics identified as
important in the regression analysis. All parameters then are available for use in the
Clark unit-hydrograph option in HEC-1 and can be used to develop the inflow PMF
hydrograph.

8-7.3 Empirical Coefficients for Synthetic Unit-Hydrograph Procedures

Failing to find applicable procedures or data to perform a regional analysis, consideration

should be given to using empirical coefficients for one of the existing procedures.
Empirical coefficients for computing a synthetic unit hydrograph often are presented in
technical literature as being applicable to basins described only in general terms, such as
rolling hills or coastal plains. These unit hydrographs often are used to design minor civil
works projects. However, these unit hydrographs and empirical equations for lag time
and time to peak are not acceptable for use in PMF-hydrograph computations, unless

-51-
there is documented evidence of their applicability, or proof that applicability can be
developed. Such justification may exist in the form of special regional studies.

C In this chapter the Clark, Snyder, and SCS unit hydrographs are the only
ones recommended, but only because the HEC-1 program includes these
methods.

C Other synthetic unit hydrographs may be available from other studies or

technical references and may be applicable to the project. If they are
used, full documentation must be provided and their use justified.

C Always check and explain regional results by comparison to the time of

concentration calculated with the TR 55 program [SCS 1986].

C Most synthetic unit hydrographs have been developed for a given storm
duration in keeping with unit-hydrograph theory. It will be necessary to
know the duration for any unit-hydrograph considered and to adjust that
unit hydrograph to fit the duration required for the basin being
considered (required duration must not be more than the lag time
divided by 5). Methods for making such adjustments, such as use of the
S-Curve, are covered in standard hydrology textbooks. The Snyder
parameters employed by HEC-1 are the "standard" lag, t p , and peaking
coefficient, C p. HEC-1 sets the unit duration of a developed unit
hydrograph equal to the computation interval ( )t) using equations based
on the Snyder "standard" parameters.

8-7.3.1 Snyder Unit Hydrograph

Many regional studies performed in the United States have concentrated on computing
coefficients for the Snyder unit hydrograph in terms of measurable basin parameters. The
equations used for the Snyder unit hydrograph are [HEC 1990a]:

(8-7.2)

(8-7.3)

where: tp = Time lag measured from the centroid of precipitation

excess to the time of peak flow (hours)

-52-
 L = Length of the main watercourse (miles)

Lca = Length along the main watercourse measured from the outlet
upstream to a point nearest the basin centroid (miles)

Qp = Peak flow rate of the unit hydrograph (cfs)

A = Drainage area (square miles)

The coefficients Ct and C p are strictly empirical values often recommended as applicable
to specific regions. C t accounts for storage and slope of the watershed, and C p is a
function of flood-wave velocity and storage.

Snyder unit-hydrograph parameters may be entered in the HEC-1 program if acceptable

generalized values are available for the region. The Snyder unit-hydrograph relationships
define only the unit-hydrograph peak discharge and the time lag, tp . Recommended
widths of the unit hydrograph at 50 percent and 75 percent of the peak flow can be
computed in terms of estimated values of C t and C p for the basin [COE 1946]. However,
when using HEC-1, this is not required since the program computes a Clark unit
hydrograph by estimating T C and R from the tp and C p values of the Snyder unit
hydrograph.

Caution:

(1) Unless a regional study has been performed for the selection of appropriate tp and C p
values as a function of definable basin characteristics, their selection would be entirely
judgmental based on the hydrologic engineer's personal impression of basin
conditions—a procedure which is not recommended. Selected values for tp and C p should
be documented and justified.

(2) Snyder's original development was performed for large basins in the Appalachian
region [Snyder 1938]. If information from detailed regional studies gives values of C t
and C p in terms of definable parameters for regional drainage basins, use of the Snyder
equations may provide satisfactory results. The acceptability of the Snyder method and
parameters, or any other method, must be documented and justified.

8-7.3.2 Clark Unit Hydrograph

The Clark unit hydrograph uses a time-area curve for the basin. Since the unit
hydrograph appear to be relatively insensitive to the shape of this time-area curve unless
the basin is one with little storage, the automatic generalized curve in HEC-1 can be used.
Values for T C and R should be estimated as described in Section 8-6.

-53-
Caution: The means of estimating TC and R are by no means infallible; it is extremely
important that the hydrologic engineer doing this estimation have substantial experience
and understand the hydrologic behavior of the basin. Although analytical techniques are
indispensable when working on ungaged basins, the judgment of the experienced
hydrologic engineer is important. The values selected for T C and R should be justified.

8-7.3.3 SCS (NRCS) Dimensionless Unit Hydrograph

If applicable data for regional studies are not available, the SCS (NRCS) unit-hydrograph
method for ungaged sites—which is described fully in the NRCS National Engineering
Handbook [SCS 1985]— may be used for basins with total areas not exceeding 100 square
miles. (This upper limit on total area only applies to ungaged sites.) However, subbasins
should not exceed 20 square miles if the SCS method is used. The only analytical
requirement for application of this method is estimation of the lag time for the basin. In
HEC-1, the SCS dimensionless unit hydrograph is fully defined by one parameter—the
SCS lag time—and is assumed equal to 0.6 T C .

Caution: Many empirical equations have been published for estimating T C , but all are
subject to large uncertainties; the hydraulic method of calculating T C , as recommended
in Section 8-6, should be used. The value, method, and equation selected for
computation of T C must be justified and consistent with the respective methodologies.

8-8 Loss Rates for Subbasins

This section pertains to assigning loss rates in the PMF hydrologic model. It will be
necessary to assume an infiltration rate representative of saturated conditions in
computing the PM F. The infiltration rate should be assumed in accordance with
recognizable characteristics of the drainage basin. The HEC-1 model offers five methods
for modeling losses, or abstractions. These are the SCS (NRCS) Runoff Curve Number
(CN), initial-and-uniform, Green-Ampt equation, Holtan equation, and exponential loss
function. Of these, the traditional approach for PMF computations is a basin averaging
method using initial and uniform losses, although the SCS method is often used. (Of the
remaining three methods, the Holtan method was designed primarily for croplands; the
Green-Ampt equation reduces to a uniform loss rate equal to the soil saturated hydraulic
conductivity when the soil is saturated; and the exponential equation is based entirely on
empirical calibrations.)

When using the initial and uniform loss method to compute the PMF, the peak flow will
almost always be insensitive to the initial loss, as will the flood volume in the vicinity of
the peak. Therefore, it may be appropriate to set the initial loss to zero, unless a specific
hydrologic condition, such as substantial depression storage, justify otherwise. If the SCS
loss function is used, Antecedent Moisture Condition (AMC) II is normally assumed

-54-
when establishing the runoff CN. The CN is determined based on the hydrologic soil-
cover characteristics including land use, treatment and hydrologic condition.

Any loss method can be applied in either a basin-averaged or distributed mode. The
HEC-1 model, in which the spatial unit of computation is the basin or subbasin, uses loss
functions in a basin-averaged mode. Typically, distributed loss calculations require more
effort than basin-averaged calculations. However, they should yield more representative
runoff estimates than basin-averaged parameters. A discussion of distributed loss
modeling using the STATSGO or SSURGO databases appears in Appendix D.

Infiltration losses can be quite variable depending upon the rainfall intensity and the
accuracy with which other inputs to HEC-1 (particularly rainfall distribution) are known.
In addition, antecedent conditions for the PMF will be different from the conditions
existing prior to historical storms. Regardless of the method used to compute losses, the
model must be verified with available historical storm data in accordance with Section 8-
8.3. Since some historical events may not be of sufficient size to prevent significant
nonlinear effects, the historical floods used for verification should be clearly out-of-bank
or it should be shown that saturated soil conditions existed in a significant part of the
basin prior to the storm.

For basins where verification is not possible, loss rates at the minimum values from Table
8-8.1 may be used in either the basin-averaged or the distributed loss rate methods.
When soil infiltration rates are selected or derived from databases that provide the
information as a range, such as the range for hydraulic conductivity presented for each
soil layer in the STATSGO or SSURGO databases , the infiltration rate of the least
permeable layer of the soil should be assumed to control the soils loss rate. A loss rate
that is justified based on site-specific information, such as a review of the geological
make-up of the soils, the review of soils information such as county or local soils maps,
or actual data obtained from any site investigations within the basin, should be chosen
from within this range. The justified infiltration rate may be used in the distributed loss
rate method (See Section 8-8.3 and Appendix D). The STATSGO or SSURGO data
should not be used to develop a basin-averaged loss rate since the high permeability
values for some sandy soil classes in the databases will raise the basin-averaged loss rate
to unrealistic values.

8-8.1 Basin-Averaged Methods

This method is recommended because it is relatively simple to use. If other methods are
used, they should be justified and, if possible, verified for several large historical floods.

For PMF runoff computations, the soil should be assumed to be saturated with infiltration
occurring at the minimum rate applicable to the area-weighted average soil type covering

-55-
each subbasin. Soil data for the drainage basin should be examined and the major soil
classifications delineated. An area-weighted average soil classification should be
established for each subbasin that can be identified with a NRCS Hydrologic Soil
Classification (A, B, C, or D). The minimum infiltration rate for the average hydrologic
soil classification should be selected from the information provided in the 1955 Yearbook
of Agriculture [USDA 1955] unless a larger infiltration rate can be justified based upon a
review of available soil and geologic maps of the watershed or other technical reports or
field investigations. The percentage of the area of each subbasin that is impervious
should include areas of open water, wetlands, frozen soils with high silt content, etc.
Table 8-8.1 provides the general soil characteristics and minimum infiltration rates taken
from the USDA reference.

Table 8-8.1 Minimum Infiltration Rates for Hydrologic Soil Groups [USDA 1955]

Minimum*
Hydrologic Infiltration
Group Rates Soil Description
(in./hr)
A 0.30 to 0.45 Deep sand, deep loess, aggregated
silts
B 0.15 to 0.30 Shallow loess, sandy loam
Clay loams, shallow loam, soils low in
C 0.05 to 0.15 organic content, soils usually high in
clay
Soils that swell significantly when
D 0 to 0.05 wet, heavy plastic clays, certain saline
soils
*
For each hydrologic group, use lowest value unless a higher
value can be justified.

8-8.2 Distributed Loss Rate Method

The use of basin-averaged loss parameters, especially for large watersheds, can give
inconsistent results for basins with spatially diverse characteristics. As the availability of
terrain and spatial data becomes common place, the use of geographic grids for
hydrologic analyses may become more practical. The use of a grid system will permit a
more detailed modeling of hydrologic processes than is possible with lumped parameter
methods. Currently, the HEC-GeoHM S computer program utilizes GIS terrain and
spatial information for input into hydrological models such as HEC-HMS. Future

-56-
advancements in hydrologic modeling software is expected and the use of new modeling
software is acceptable provided adequate information concerning the modeling process is
presented.

The distributed loss rate method described below gives the advantages of simulating
losses in a distributed fashion in a fairly simple, economical model structure such as the
HEC-1 model. Although it has been developed for use with the uniform loss rate method,
the principle of separating basins or subbasins into sections with homogeneous runoff
characteristics could be applied to any loss rate method.

Within a watershed, factors affecting the generation of runoff from rainfall include type
and depth of soils, land cover, and the presence of saturated soils. “Partial area” theory,
which is now generally accepted as a model for runoff generation on natural watersheds,
holds that only a portion of a watershed contributes direct runoff during a storm. The
size and location of the runoff-producing portion or “contributing area” can vary as a
result of the progressive saturation of the watershed soils.

The physical complexity of the watershed response leads to difficulties in using a single,
time-invariant, basin-averaged loss rate to represent conditions during all storms.
Suppose a watershed contains soils with a wide range of permeabilities , and a few areas
that are essentially impervious due to clays and high water tables. Any precipitation
event will generate some runoff, because of the impervious areas. If uniform basin loss
rates are calculated by subtracting the runoff rate from the rainfall rate during an observed
storm, the calculated loss rate during any event must always be less than the rainfall
intensity because some runoff will occur from the impervious areas. For example, a 1-
inch-per-hour storm must always yield a calculated loss rate less than one inch per hour -
even if 90 percent of the basin has soils with 10-inch-per-hour permeability. Another way
to view this is to consider that there is no way to measure a 10-inch-per-hour
permeability, unless the rainfall rate is at least 10 inches per hour. This means that for a
watershed with spatially diverse loss rates, the calibrated loss rates of the contributing
areas depend on the intensity of the storm.

As a simple example, consider a watershed with 70 percent of its area having a loss rate
of 4 inches per hour and 30 percent of the area having a loss rate of 0.5 inches per hour.
For a 1-hour, 1-inch rainfall event, the area-averaged rainfall excess will be 0.15 inches
((1.0 - 0.5)*0.30), and the effective basin-averaged loss rate will be 0.85 inch per hour
(1.0 - 0.15). For a 1-hour, 2-inch rainfall event, the area-averaged rainfall excess will be
0.45 inches and the effective basin-averaged loss rate will be 1.55 inches per hour. For a
1-hour, 4-inch rainfall, the area-averaged rainfall excess will be 1.05 inches and the
effective basin averaged loss rate will be 2.95 inches per hour, which is equal to the
actual area-averaged basin loss rate. For all three cases, 30 percent of the basin is
contributing all of the runoff and the remaining 70 percent is contributing no runoff. If

-57-
the storm intensity exceeds 4 inches per hour, then 100 percent of the basin will be
contributing runoff.

It's important to note that the effective basin-averaged loss rate will be equal to the actual
basin-averaged loss rate if the storm intensity is at least as large as the highest distributed
loss rate. However, for storm intensities less than the highest distributed loss rate, the
effective basin-averaged loss rate will always be greater than the actual basin-averaged
loss rate.

It follows that using a basin-averaged loss rate calibrated on one storm to calculate the
rainfall excess during a storm of a different intensity will not correctly predict the excess.
Neither will using a spatial average of the actual basin loss rates. In our example, if the
watershed model is given a 2.95-inch-per-hour basin averaged loss rate, any modeled
storm with an intensity less than 2.95 inches per hour will yield no runoff at all. The
effective basin average loss rate for a moderate-sized storm (for example, a 5-year or 10-
year storm) is unlikely to be the same as that for a 100-year storm or, especially, a
Probable Maximum Storm (PMS). It is important to recall that standard references on
inferring loss rates from soil characteristics (such as Table 8-8.1 of the Guidelines) were
developed and tested with more common design storms in mind - not the PMS.
Therefore, depending on the antecedent soil moisture conditions and the intensity of the
storm chosen for use in selecting the loss rates, the hydrologic model will likely yield
reasonable results for storms of similar size and antecedent conditions, but may not
correctly predict runoff from larger storms such as the PMS.

8-8.2.1 Application of Distributed Loss Rate Method

Using distributed loss rates avoids the particular problems associated with spatial
averaging. Certain assumptions are still necessary. The two main assumptions are that
(1) any unit of soil has a representative loss rate that does not vary over time, and (2)
appropriate loss rates for a soil/land cover combination can be inferred from maps or
other published data. One drawback of the method is that, if a distributed loss rate model
is verified with an observed event and found not to predict the rainfall excess well, there
is no single parameter that can be adjusted to provide a “fit.” Instead, it is necessary to
re-evaluate all of the assumptions and data sources that were used in developing the
distributed model.

A detailed, step-by-step description of the application of the distributed loss rate method
using STATSGO or SSURGO soils data is provided in Appendix D. The method
generally relies on digital soils and land cover databases that can be converted to a GIS
format and analyzed to identify areas of intersection between land cover and soil types.
From the spatial data and other information within the GIS (such as layer-by-layer soil
permeability contained in the NRCS’s STATSGO soils database), loss rates are assigned

-58-
to each combination of soil type and land cover occurring in the basin. A table or
database is then constructed listing the area percent of each subbasin having each loss
rate. Finally, for the storm being modeled, hourly rainfall rates are applied to each loss
rate category separately to compute rainfall excess; and the rainfall excess amounts are
weighted by area and summed over the basin or subbasin to give an hourly sequence of
rainfall excess generation. Note that in this procedure, the rainfall excess is area-
averaged, which is necessary to apply it in a conventional lumped-basin model. This
excess amount is storm-specific. This is a very different procedure, arithmetically, from
area-averaging the loss potential of the basin soils, which may or may not be fully
utilized during a given storm.

8-8.3 Verification and Model Adjustment

Regardless of the method used to compute losses, the model should be verified with
available historical storm data. Since some historical events may not be of sufficient size
to prevent nonlinear effects, the historical events used for adjusting loss rates should be
clearly out-of-bank floods or saturated soil conditions must have existed in a significant
part of the basin prior to the storm. The model should be run at a location with historical
time sequence data for several historical storms, and the results compared. A plot of
these results should be included with the study. If the modeled peak flow and runoff
volume underestimates the observed peak flow and volume of the historical storms, then
adjustments to the loss rates should be done. The emphasis should primarily be placed on
achieving a good fit of the runoff volume since it is determined primarily by the loss
rates, whereas the peak flow is effected by the loss rates and the unitgraph parameters.
Adjustments should consider the amount and reliability of the information available on
the physical characteristics of the soil in the study basin. Other sources of hydrologic
information should be explored, such as:

(a) depth-to-bedrock or depth-to-water-table maps: are subsurface

characteristics shown in the soils or land cover maps affecting
infiltration and resulting runoff?

(b) more detailed soil maps: for instance, do county soil maps support
the distribution of soil types indicated by the initial analysis or is actual
field data available?

(c) does rapid subsurface flow (interflow) to rivers and streams occur?

(d) other site- and storm-specific conditions: did land use, land cover,
groundwater, soil behavior, etc. affect the runoff?

-59-
 (e) lack of detail (either spatial or temporal) in the rainfall data. Hourly
rainfall is essential to accomplish the verification, and if the rain gage
locations do not accurately represent the full range of storm intensities
over the basin, it is very difficult to verify that the model is performing
correctly.

(f) are unit hydrographs obtained from other sources still valid?

If the minimum values from the SCS Minimum Infiltration Rates for Hydrologic Soil
Groups found in Table 8-8.1, or the minimums of the saturated hydraulic conductivity
range of the least permeable layer (as provided in the STATSGO or SSURGO databases)
are used to model the losses in the PMF model (with the appropriate adjustments for
impervious areas), and if there are no historical floods that can be used to verify the
model, then no further adjustments are necessary to verify the model. This would apply
to all ungaged basins and the portions of the gaged basins that were non-contributing
during the historical events.

If however, a loss rate other than the minimum is selected based on adequate
justification, a sensitivity analyses should be performed. Supporting data for justification
purposes may include a review of the geological make-up of the soils, the review of soils
information such as county or local soils maps, or actual data obtained from any site
investigations within the basin. Sound engineering judgment must be used to select and
justify the loss rates if they are significantly larger than the intensity of the historical rain
event used for verification. The sensitivity analysis should compare the runoff
hydrograph developed by using the justified loss rates versus a runoff hydrograph
developed by using the minimum value of the range of the least permeable layer of the
soil column. The results of this sensitivity analysis should be plotted on a single graph
for ease of review.

Loss rates may be adjusted as necessary if the model can be verified using historical
events. However, loss rates unaffected by the verification process should not be adjusted
from the minimum values unless additional physically based information is available to
support selecting a loss rate higher than the minimum values.

As additional information becomes available, such as flood events larger than historical
events, the model should be re-run to see if it adequately predicts the new flood event. If
the model does not adequately predict the new flood event, adjustments should be made
to the loss rates, unitgraph parameters, or other parameters. When drainage basins do not
contain adequate rainfall/runoff data, the installation of rain gages and flow gages should
be considered.

-60-
Caution: In the HEC-1 computer program, the Snowmelt Loss Rate (LM) overrides the
Uniform Loss Rate (LU) in a rain-on-snow analysis when the ground is covered with
snow. When the snowpack is depleted, the HEC-1 program then uses the LU variable for
the overall loss rate. Although it is possible for the snow-covered ground loss rate to be
less than the snow free loss rate, no literature on this subject quantifies this relationship.
Care must be used to ensure that the verification of the model adequately considers the
effect of changes in the snowpack during storm events. If possible, the LU and LM
variables should verified independently of each other.

8-8.4 Infiltration Characteristics of Soils Under Frozen Conditions

Many researchers have identified the effects of soil freezing on the infiltration capacity of
soils. Types of frost, soil structure, and antecedent soil moisture content have all been
noted as factors influencing frozen soil infiltration capacity. The structure type of soil
frost has a strong influence on the rate of infiltration of a soil [Trimble et al. 1987].
Because of different vegetation cover and surface soil characteristics, soils will respond
differently to freezing, producing different types of soil frost structures. These structures
are most commonly classified as either concrete or granular frost. Soils with concrete
frost, which allow very little infiltration, are identified by dense thin ice lenses and ice
crystals.

Granular frost, typically found in woodland soils, consists of small frost particles
intermingled with soil particles. Frost structures are related to the moisture content of the
frozen soil [Post and Dreibelbis 1942]. Soils frozen at low moisture content may become
granulated and provide little impediment to infiltration. Conversely, soils frozen at high
moisture contents often freeze into massive, dense, concrete-like structures that are nearly
impermeable to water [Zuzel and Pikul 1987]. Reduced levels of moisture content are
found in forested areas because of interception and evapotranspiration [Kane and
Stein 1983]. These low moisture contents result in granular frost structures in the winter.

Frozen sandy soils typically do not develop a concrete, or impermeable, frost. For
example, in Engelmark's [1987] set of laboratory experiments, infiltration rates were
measured in a fine sand. The grain-size curve of the fine sand indicated 84 percent
passing a #40 sieve and 5 percent passing a #200 sieve. Infiltration rates obtained for this
soil in the frozen state were between 1-2 mm/min. (2.4-4.7 in./hr). Another experiment
executed by Blackburn and Wood [1990] in a sandy soil provided a range of infiltration
rates of 0.42-1.08 mm/min (1-2.4 in./hr), depending on the type of frost.

When coarse soil types are combined with the vegetation, a low soil moisture content can
be predicted. Even heavy rainfall rates may not exceed the rate of infiltration in soils and
they will not become saturated. W ith these conditions, a granular soil frost will
predominate in the winter. Granular soil frost is far from impervious; it typically has

-61-
infiltration rates the same as, or even higher than, the soil in an unfrozen condition
[Blackburn and Wood 1990].

Based on these observations, the following guidelines should be followed in estimating

cool-season loss rates:

C Wetlands should be modeled as impervious elements. These soils, even

if they are sandy, may intersect the seasonal high water table and thus
have a higher potential to produce a concrete type of frost.

C Soils with high silt content associated with high groundwater tables
should be assumed to be impervious.

C Clays should also be assumed to be impervious.

C Forested soils or soils with a minimum 4-inch humus depth should have
unfrozen condition infiltration rates applied [Kane and Stein 1983].

C Nonforested soils, other than sands or sandy loams, should be

considered impervious when they occur within the historical maximum
frost depth.

C Infiltration rates for normal granular soils, such as sand and sandy loam,
should be assumed equal to the unfrozen condition. If, however, the
granular soil exhibits a high moisture content, this assumption may not
be appropriate.

Caution: Situations may exist where an ice layer can form on the soil surface as a result
of a mid winter thaw and refreeze. This condition may significantly reduce the
infiltration rate obtainable by granular soils.

8-9 Probable Maximum Flood Development

Sections 8-5, 8-6, and 8-7 described the process of developing the necessary runoff model
for use in computing the inflow PMF hydrograph. Section 8-8 provides guidance for
selecting loss rate parameters and verifying the model. For simple basins, this runoff
model will consist of a single representative unit hydrograph. For more complex basins,
the runoff model will consist of a combination of unit hydrographs for subbasins and a
streamflow-routing process. The runoff model is used to calculate the inflow PMF
hydrograph. This section provides guidelines for calculating the PMF including
parameters related to the PMP, antecedent hydrologic conditions, snowmelt, base flow,

-62-
and channel routing. In addition, guidelines for a sensitivity analysis of the calculated
inflow PMF are provided in Section 8-9.6.

8-9.1 Spatial Distribution and Disaggregation of the Probable Maximum

Precipitation

To compute the inflow PMF, it is necessary to determine both a temporal and spatial
distribution of the PMP on the project basin.

8-9.1.1 Storm Duration

A primary assumption on which this chapter is based is that complete depth-duration

information is available for the PMP for both general and local storms, so that the
necessary design storms can be constructed. A local storm is one with a relatively small
area of influence such as a thunderstorm. Local storms of short duration and high
intensity can produce a critical PM F for dams located on very small drainage basins (up
to about 1,000 mi2 ), or where the antecedent operating level of the reservoir can be higher
(such as due to flashboard installations or closure of spillway gates) during the late spring
and summer months. In addition, the local season PMP may govern for larger basins with
unusual shapes. However, for other small basins, the inflow PMF produced by a
long-duration general storm, when routed through the reservoir, will result in higher
reservoir levels and may produce the largest rate of outflow. Thus, it is usually necessary
to develop inflow hydrographs for both general and local seasonal PMPs to establish the
PMF.

8-9.1.2 Storm Spatial Distribution

Basin-average or subbasin-average rainfall must be developed for the PM F model. This

will require establishment of a spatial distribution for the PMP within the basin. Rainfall
data are seldom available from a large enough number of rain gages to allow construction
of an accurate isohyetal map for each historical storm. If a historical storm has been
studied by the COE, USBR, USGS, or NWS, isohyetal maps may have been developed
from rainfall depth information obtained during "bucket surveys." If isohyetal maps are
available for any of the historical extreme storms that have occurred in the area, or if they
can be constructed from data available, they could be used in defining the spatial
distribution of storm rainfall for the PMP.

Individual storm distribution may be biased because of a singular feature of the storm.
For this reason, this chapter recommends that the elliptical isohyetal map produced by the
NWS in Hydrometeorological Report No. 52 [NWS 1982] be used in the region east of

-63-
the 105th meridian. For other areas, refer to the appropriate HMR or site specific study
(Figure 8-1.1).

The storm pattern on the basin should be adjusted so that the maximum rainfall volume
falls on the drainage basin. In general, this will require that the area of greatest rainfall
depth be approximately centered on the basin, and that the storm pattern be rotated so that
the basin is covered to the greatest extent possible by the isohyets of greatest rainfall
depth. If the basin is subdivided, the peak runoff rate might result from a different
centering. A sensitivity analysis is required to determine the critical centering of the PMS
to optimize the storm's spatial distribution. Generally, a storm centered over the middle
of the basin will produce the PMF with the largest volume. A storm centered closer to
the dam may produce a higher peak discharge and may result in a higher maximum water
surface elevation. Other basin characteristics reflected in the model, such as loss rates,
unit hydrograph parameters, basin subdivision, and subbasin or basin shape, may have an
impact on the optimal storm centering and orientation.

The computer program HMR52, which was developed by the Hydrologic Engineering
Center of the COE, can be used to apply the procedures contained in HMR 52 [COE
1984]. In Wisconsin and Michigan, the computer program WM PMS (which is a
modified version of HMR52) is available through the Electric Power Research
Institute (EPRI). These programs automatically produce a 72-hour storm. However, the
storm totals are balanced so that lesser durations are also PMP values for the storm size.

For other locations such as the western states, the areal distribution of the storm cannot
be generalized as readily due to orographic influences or unique storm patterns.
Dependence must be placed on the patterns produced by the historical storm, mean
annual precipitation patterns, or 50-year or longer return period precipitation patterns
such as those found in NOAA Atlas 2 (Miller et. al., 1973). If insufficient data exist to
provide for development of an isohyetal pattern, a uniform distribution over the basin
may be assumed. The method used by the USBR, known as successive subtraction, can
be used to advantage [Cudworth 1989]. The successive subtraction technique allows for
centering a storm over a subbasin when an isohyetal pattern is not available. This
situation is common in the mountainous areas of the western U.S.

8-9.1.3 Temporal Distribution of the Probable Maximum Precipitation

The depth-duration relationship for the PMP should be taken from the envelope curve
included in the PMP data. In general, if the peak period of rainfall is placed at the
beginning of the storm, the peak rate of runoff will be minimized because the largest rates
of infiltration and initial abstraction will act to reduce the peak rate of runoff. For this
chapter, it is recommended that the peak 6–hour period of rainfall be placed between the
half and two-thirds point of the storm and that the remaining 6–hour increments be

-64-
arranged in alternating descending order on each side of the peak, beginning with the time
period that precedes the peak 6–hour period. Hourly increments of rainfall should be
taken from the PMP envelope curve and distributed so as to provide a smooth temporal
curve. Reference should be made to the appropriate HMR or site specific study.

8-9.2 Antecedent and Coincident Conditions

The inflow PMF hydrograph that produces the critical conditions within the reservoir and
at the dam may depend on either the peak inflow rate or the timing and volume of PMF
inflow, depending on the spillway capacity and reservoir storage available at the
beginning of the flood. Thus, the inflow PMF hydrograph could result from a
high-intensity local storm, or a general storm with a long duration.

Caution: Although it may be possible to assess in advance whether the peak outflow
and/or the maximum reservoir water-surface elevation will be produced by a local or a
general storm, an inflow hydrograph should be generated and routed through the
reservoir for each storm.

8-9.2.1 Antecedent Conditions

What reservoir level is reasonable as the starting elevation when routing the inflow PMF
through the reservoir, considering the possibility of antecedent storms? It is advisable to
determine if a water resources agency has conducted special regional studies related to
antecedent storms. If so, the results should be considered for application. In the absence
of antecedent storm information, the following four approaches are recommended as
acceptable alternatives:

(1) Consider that the reservoir surface is at a predefined annual maximum level at the
start of PMF inflow. It will be necessary to determine the annual maximum
reservoir level for each dam, depending on the characteristics of the dam, its
spillway and outlet works, and the historical and specified operation plans. For
most hydroelectric projects, the annual maximum reservoir level should be defined
as the annual maximum normal operating level. If flashboards are normally used
on the dam during the time of the PMF, they should be assumed to be in place for
the determination of the annual maximum reservoir level. Routing of the PMF
through the reservoir should assume that flashboards fail or collapse at their design
level.

(2) Use an operating rule curve, when available, to identify the reservoir surface
corresponding to the maximum storage level for the season of the controlling
PMP. A 100-year, 24-hour storm—using the percentages of the 24-hour
maximum temporal distribution developed for the PM P—should be assumed to

-65-
 end three days prior to the PMP. The runoff hydrograph from this 100-year storm
should be routed through the reservoir using established project operating rules,
with the beginning reservoir level at the normal maximum storage level for the
season. The reservoir level at the beginning of inflow from PMP runoff should be
taken as the level produced by the routed inflow from the 100-year storm, but it
need not be greater than the annual maximum reservoir level.

(3) Use or develop a wet-year rule curve to establish the reservoir level that would
exist at the start of the inflow PMF. To develop this rule curve, assume that the
reservoir level at the beginning of the inflow PMF is at the average of the five
consecutive, highest wet-year reservoir levels occurring during the season of the
critical PMP. The assumed starting level need not be higher than the annual
maximum reservoir level.

(4) Analyze historical extreme floods and antecedent storms for the region. A
possible procedure can be found in HMR 56 [NW S 1986]. If the analysis shows it
is probable that antecedent storms do occur in the region and could significantly
influence the maximum reservoir level and the magnitude of the routed PMF
outflow, develop a storm that could reasonably be expected to occur antecedent to
the PMP as follows:

(a) Prepare an arithmetic plot of the antecedent storm rainfall expressed as a

percentage of the principal storm versus the principal storm rainfall in
inches. Draw an envelope line of the maximum values and extrapolate
to the estimated PMP depth.

(b) Plot the number of dry or zero-precipitation days preceding the principal
storm rainfall versus the principal storm rainfall amount (in.). Draw a
line enveloping these numbers (of days), extending it to the range of the
estimated PM P.

(c) Read a total rainfall depth for the antecedent storm from the plot
obtained in step (a) by multiplying the value of the total PMP depth.

(d) Set the time between the antecedent storm rainfall and the PM P equal to
the extended value found in step (b).

(e) Use both the antecedent storm and the PMP to develop an inflow PMF
hydrograph.

Average monthly flow should be obtained for the months during the season when the
critical PMP would occur. Tabulated monthly average data are available in USGS water

-66-
data reports and its web site. The average monthly flow for the month of the critical PMP
should be added to the inflow PMF hydrograph before routing through the reservoir.
When using HEC-1 this initial flow is the parameter STRTQ. For the case when the
basin has been subdivided, the initial flow will already have been added as described in
Section 8-9.4. For "ungaged" basins, the average monthly flow per square mile of
drainage area, obtained from records for nearby "gaged" basins, should be used to
compute the initial flow.

In summary, the hydrologist should make sure that antecedent conditions represent
reasonable meteorologic conditions. The report should include meteorological
justification for assumed antecedent conditions. The antecedent assumptions should be
compatible with the initial reservoir elevation used for routing the PMF. For example,
selecting a starting reservoir level based on the annual maximum or a wet-year rule curve
assumes that a large storm or snowmelt has occurred prior to the PMF. This may
preclude using an initial abstraction to determine excess precipitation during the PMP
event since the antecedent condition would have left the soil saturated.

Caution: A reservoir cannot be assumed to be drawn down at the beginning of the PMF
unless a drawdown is documented as the normal operating procedure prior to an
impending storm for that season. In this case, a lower inflow PMF during a different
season may produce a higher reservoir elevation. This should be checked.

8-9.2.2 Coincident Hydrometeorological Conditions

Assume the pertinent physical conditions of soil-moisture content, frozen ground (see
Section 8-8), and snowpack water equivalent that could reasonably be expected to occur
antecedent to the PMP. If snowpack is apt to exist in at least part of the drainage basin in
the season when the critical PMP would occur, an antecedent 100-year snowpack
(covering the area that could be subject to snowpack) should be assumed to exist at the
time when the PM P occurs (see Section 8-9.2.3).

For basins and seasons where the PM F will have a snowmelt contribution, it is necessary
to adopt temperature and snowpack criteria for use in developing the PMF. The
following steps should be followed:

C Identify the area that may be covered by snowpack at the time the PMP
begins by considering the data on historical snowpack coverage
obtained in Section 8-3.

C Assume a 100–year snowpack water equivalent and snowpack areal

distribution.

-67-
 C Develop the coincident temperature sequence and temperature–elevation
distribution from data analyzed in Section 8-4. In California and the
Northwestern states, the temperature sequence coincident with the PMP
can be found in NWS HMR Nos. 58 and 57, respectively. For other
areas, the maximum temperature sequence observed in the area for the
season of the critical PMP is recommended.

C In areas east of the 103rd Meridian, seasonal PMP values can be

obtained from HMR Nos. 33 and 53 where an updated site-specific
study of seasonal PMP values is not available.

8-9.2.3 Snowmelt Estimates

Snowmelt during the PMF should be computed using the energy-budget method available
in the HEC-1 Flood Hydrograph Package. The energy-budget method is preferable to the
degree-day (temperature index) method because the degree-day method was developed
specifically for rain-free periods. The energy budget method, on the other hand, was
developed for either rain-on-snow or rain-free periods. In the case of a PMS, the heat
added to the snowpack by the rain is an important (and sometimes even dominant) melt
factor.

The HEC-1 model input calls for several variables, such as shortwave radiation and
dewpoint temperature, which may be difficult to estimate. However, the rain-on-snow
equation makes several simplifications, leaving only a few input variables that are
important to estimate. These are:

• snowmelt temperature

• temperature sequence

• wind speed

• snowpack water equivalent

• rainfall sequence

The snowmelt temperature may be taken to be 32° F. The temperature sequence is

selected from historical temperature sequence data, with the qualification that the
sequence was associated with the simultaneous occurrence of rainfall and snow on the
ground. The maximum historical daily temperature sequence meeting these requirements
is assumed to coincide optimally with the PMS. Depending on the depth of the
snowpack, the maximum historical temperature sequence data may need to begin as much

-68-
as 72 hours prior to the start of the PMS, or it may need to begin sometime during the
PMS, in order to determine the highest reservoir level. Sensitivity studies of the start of
the maximum historical temperature sequence data should normally be done.

In some cases snowpack records will not be available. W ater equivalence data are rarely
recorded. If total snowpack depth is available, assume a 100-year snowpack for the month
of the cool-season probable maximum storm and a starting water equivalence of 30
percent (Gray and Prowse, 1992). If no historical information on snowpack is available,
an unlimited water equivalent may be assumed.

Seasonal 100-year, 3-day flood discharges may be used in lieu of the snowmelt
component in non-mountainous regions, if temperature sequence data and snowpack
depths are absent. This flood is the annual maximum three day consecutive average
discharge exceeded with a probability of 0.01. This flow should be added to the normal
base flow covering the entire time base of the hydrograph and be combined with seasonal
rain on frost-conditioned soils.

Note: The evaluation of two PMF scenario are required in the area west of the
Continental Divide. This includes (a) PMP on 100-yr snowpack, and (b) 100-yr
precipitation on Probable Maximum Snowpack.

8-9.3 Reservoir and Channel-Routing Approach

Routing of the inflow flood hydrographs from subbasins to the dam site will generally be
through natural channels and upstream reservoirs. The following procedures should be
used:

C Since level pool routing is less data intensive and simpler to use to route
through an upstream reservoir, it is recommended for use in the HEC-1
program. Although dynamic routing is more precise, it is more data
intensive to use and must be done outside of HEC-1. Either routing
technique is considered appropriate.

C The Muskingum-Cunge method, as incorporated in HEC–1, should be

used to perform any channel routing from subbasins to the basin outlet.
Cross sections of the channels, along with Manning's roughness
coefficients, will be required to use the Muskingum-Cunge routing
method. For most cases, cross sections for routing the PMF can be
obtained from 7½-minute USGS quadrangle maps. HEC–1 has the
capability to compute and combine hydrographs from side areas with
the routed channel hydrograph.

-69-
 Caution: Muskingum-Cunge uses a single (representative) cross
section defined by eight coordinate points for each routing reach. The
method cannot account for backwater effects and should not be used
when attenuation of the hydrograph is expected. An example of where
this technique might be used is when translating a hydrograph from an
upstream location to a downstream point where off-channel storage is
insignificant. Where the intention is to properly model the attenuation
of the hydrograph, dynamic-wave routing is the preferred method
(e.g., when the river is expanding or contracting or where there is
natural storage).

C If evidence is available with regard to channel loss rates occurring

during passage of floods, those rates may be used in the routing process.
However, their effect is usually small compared to PMF flow and often
can be neglected.

C Large natural constrictions should be used as control points for channel

routing.

8-9.4 Base Flow Coincident with Probable Maximum Flood

The flow rate in the river for basins or subbasins at the time the PMP begins should be
consistent with the antecedent approach selected from Section 8-9.2.1. Average monthly
flow should be obtained for the months during the season when the critical PMP would
occur. Tabulated monthly average flow data are available in USGS water data reports or
from the USGS web site. The average monthly flow for the month of the critical PMP
should be used and added to the inflow PMF hydrograph before routing through the
reservoir, or combining or routing subbasin hydrographs. When using HEC-1, the initial
flow is the parameter STRTQ. For "ungaged" basins, the average monthly flow per
square mile of drainage area, obtained from records for nearby "gaged" basins, should be
used to compute the required initial base flow. If the 100-year, 3-day snowmelt option,
as delineated in Section 8-9.2.3, is used, there is no need for an additional base flow
component as that component is already included in the data record used for the statistical
analysis.

8-9.5 Inflow PMF Hydrograph

Use the input developed in Sections 8-8 and 8-9.1 through 8-9.4, and run HEC-1 to
compute the inflow PMF hydrograph. Whole model verification should be done using
historical data as discussed in Section 8-8.3. The procedures of the inflow PMF
hydrograph development outlined in this chapter rely on model calibration and
verification using historical data. The flood data used for calibration will usually have

-70-
return periods of less than 100 years. Lag times should be adjusted to account for PMF
conditions. A PMP event will produce rainfall intensities much greater than anything
previously experienced in the study area. This will shorten lag times so that appropriate
adjustments to them are needed for the severe conditions which could be expected in
generating the PMF.

8-9.6 Review and Sensitivity Analysis of Representative PMF Hydrograph

The first computed inflow PMF hydrograph should be considered as preliminary. A

review of the assumptions considered to have a significant effect on the PMF should be
made to assess the sensitivity of the individual parameters on the magnitude of the PMF.
The following steps should be performed for each study.

C A sensitivity analysis should be made to determine the degree the PMF

is effected by key parameters, such as the time of concentration, loss
rates, etc., even if conservative values for those parameters were
assumed.

C If the PMF is particularly sensitive to the magnitude of a parameter, the

source of the parameter determination should be reviewed to ensure that
the chosen value is reasonable.

C The results of the sensitivity analysis and the selection of the sensitive
parameters should be documented and justified.

8-10 Reservoir Routing to Obtain the Outflow PMF Hydrograph

The preceding sections led to the development of the inflow PMF hydrograph, which
must be routed through the reservoir to determine the maximum reservoir elevation and
peak discharge at the dam. Assumptions of reservoir starting elevation and initial outflow
must be made.

8-10.1 Initial Assumptions

The following assumptions should be made to route the inflow PMF through the
reservoir:

C Use the reservoir area-volume-elevation information as obtained and

reviewed in Sections 8-3 and 8-4, respectively.

C Use spillway and outlet-works capacities established in Section 8-4.5.2.

-71-
 C Use the existing gate operating policy as established in Section 8-4.5.3.
Any deviations from the accepted reservoir operations must be
reviewed to assure the deviations are within the terms and conditions of
the project license. Proposed changes to the operating plan should be
discussed with the appropriate Regional Office prior to incorporating
the changes into the routing of the PMF. Some of the available
computer models are not set up to deal with a gated control spillways.
Sensitivity analyses of the simulated gated spillway releases should be
developed to determine the outflow discharges due to gate operating
changes.

• Considerations regarding reservoir starting elevations were given in

Section 8-9.2.1 and should be considered simultaneously with the gate
and flashboard operations discussed in Section 8-4.5.3 to determine the
critical reservoir starting elevation. If the considerations regarding
operation of the gates or failure or removal of flashboards indicate a
higher reservoir starting elevation than would be given by the
considerations in Section 8-9.2.1, the higher elevation should be used.

8-10.2 Routing Procedures

Level-pool-routing procedures can generally be used. Whether or not level-pool-routing

procedures are satisfactory will depend on the unit hydrograph used to develop the inflow
PMF hydrograph and the dynamic effect of the reservoir on flood flows. The dynamic
effects of the reservoir could be pronounced for large or long and narrow reservoirs, and
is usually negligible for small reservoirs. Some adjustment for so-called "wedge storage"
during rapidly rising pool levels may be necessary.

Caution: If reverse-reservoir routing was used to develop the inflow hydrograph to the
reservoir during passage of the historical floods used in the unit-hydrograph analysis,
some of the dynamic effects will have already been implicitly included in the developed
inflow PMF hydrograph. Although dynamic effects during passage of a PMF may be
more dramatic than during the analyzed historical floods, they are satisfactorily
approximated in the reverse-reservoir routing process. Level-pool-routing procedures
can be used in these situations.

An alternative is to use a distributed inflow procedure where all inflows to the reservoir
rim are estimated. This requires developing inflow PMF hydrograph at all major
tributaries and the direct rainfall on the reservoir. The inflows are then routed through
the reservoir using dynamic routing procedures or simple translation with timing based on
wave celerity calculations. Dynamic reservoir flood routing procedures—although
mathematically complex and sometimes difficult because of numerical instability—can be

-72-
accomplished using the NWS unsteady routing program DAMBRK (Fread 1989) or
FLDW AV (Fread 1997).

The flood-passage operations should be reviewed after the initial routing of the inflow
PMF to assess the sensitivity of the resulting maximum outflow rate and reservoir
elevation to the assumed reservoir starting elevation.

8-10.3 Reporting Requirements

As a general rule for preparing reports, sufficient documentation should be provided to

allow the FERC staff to verify the reasoning and check the analyses of the PMF estimate.
Using the recommended report format (Appendix B) will help provide the necessary level
of information on each component of the study. Input and output files for computer
analysis should be provided as printout in submittals and also on 3.5-inch diskettes or
CD-ROM. If programs are used that are not readily available or not in common use, the
FERC staff might request code documentation, users manuals, and an executable version.

-73-
8-11 References

American M eteorological Society (1959), Glossary of Meteorology, Boston,

Massachusetts.

American Public Health Association, etc. (1981), Glossary Water and Wastewater
Control Engineering, Washington, D. C. 20037

American Society of Civil Engineers (ASCE) (1996), Hydrology Handbook, New York,
New York

Bailey, S.M. and G.R. Schneider (1939), "The Probable Maximum Flood and its Relation
to Spillway Capacity," Civil Engineering, Vol. 9, No. 1, American Society of Civil
Engineers, New York, New York.

Barnes, H.H. (1967), Roughness Characteristics of Natural Channels, Geological Survey

Water Supply Paper 1849, United States Department of Interior, Geological Survey,
Branch of Distribution, Arlington, Virginia.

Blackburn, W.H. and M.K. Wood (1990), "Influence of Soil Frost on Infiltration of Shrub
Coppice Dune and Dune Interspace Soils in Southeastern Nevada," Great Basin
Naturalist 50(1):41-46.

Bradley, J.N. (1952), "Discharge Coefficients for Irregular Overflow Spillways," United
States Bureau of Reclamation Engineering Monograph No. 9, Denver Federal Center,
Denver Colorado.

Burnash, J.C., R.L. Ferral, and R.A. McGuire (March 1973), "A Generalized Streamflow
Simulation System; Conceptual M odeling for Digital Computers," Joint Federal-State
River Forecaster Center, United States Department of Commerce, National Oceanic and
Atmospheric Administration, National Weather Service, and State of California,
Department of Water Resources.

Chow, V.T. (1959), "Open-Channel Hydraulics," McGraw Hill, Inc.

Clark, C.O. (1943), "Storage and the Unit Hydrograph," Transactions of American
Society of Civil Engineers, Vol. 108.

Crawford, N.H. and R.K. Linsley (July 1966), "Digital Simulation in Hydrology:
Stanford Watershed Model IV," Technical Report No. 39, Stanford University,
Department of Civil Engineering.

-74-
Cudworth, A.G. (1989), Flood Hydrology Manual, United States Department of the
Interior, Bureau of Reclamation, Denver Federal Center, Denver, Colorado.

Electric Power Research Institute (EPRI), (1993). Probable Maximum Precipitation

Study for Wisconsin and Michigan (Research Project 2917-29). Volume 1–Prepared by
North American Weather Consultants. Volume 2–Prepared by Mead & Hunt, Inc.

Ely, P.B. and J.C. Peters (1984), "Probable Maximum Flood Estimation - Eastern United
States," Paper No. 84017, Water Resources Bulletin, Vol. 20, No. 3.

Engelmark, H. (1987), "Rate of Infiltration into Frozen and Unfrozen Fine Sand,"
Canadian Journal of Earth Sciences, 25:343-347.

Federal Emergency M anagement Agency (1991), Probable Maximum Precipitation and

Probable Maximum Flood Workshop, Proceedings of the Workshop, May 1-4,
Washington, D.C.

Fetter, C. W . (1988), Applied Hydrogeology, Macmillan Publishing Company, New

York, New York

Fread, D. (1989), The NWS DAMBRK M ODEL: Theoretical Background/User

Documentation, Hydrologic Research Laboratory, Office of Hydrology, United States
National Weather Service, National Oceanic and Atmospheric Administration, Silver
Spring, Maryland.

Gray, Don M . and Terry D. Prowse (1992), "Snow and Floating Ice," Chapter 7 in
Handbook of
Hydrology, David Maidment, Ed. McGraw-Hill Publishers, 1982.

Hathaway, G. (1939), "The Importance of Meteorological Studies in the Design of Flood

Control Structures," Bulletin of the American M eteorological Society, Vol. 20, pp.
248-253.

Hathaway, G. (1944), "Discussion of the Primary Role of Meteorology in Flood Flow

Estimating," Transactions of American Society of Civil Engineers, Vol. 109.

Henderson, F. M. (1966), Open Channel Flow, University of Canterbury, Christchurch,

New Zealand

Hershfield, David M. (1961), Technical Paper No. 40 - Rainfall Frequency Atlas of the
United States, Engineering Division, Soil Conservation Service, United States
Department of Agriculture.

-75-
Hoggan, D.H. (1989), Computer Assisted Flood Plain Hydraulics, McGraw-Hill
Publishing Company, New York, New York.

Hydrologic Engineering Center (1990), HEC-1, Flood Hydrograph Package, User's

Manual, U.S. Army Corps of Engineers, Davis, California.

Hydrologic Engineering Center (1990), HEC-2, Water Surface Profiles, User's Manual,
U.S. Army Corps of Engineers, Davis, California.

Hydrologic Engineering Center (1984), HMR-52, Probable Maximum Storm (Eastern

United States), U.S. Army Corps of Engineers, Davis, California.

Hydrologic Engineering Center (1982), Training Document No. 15, Hydrologic Analysis
of Ungaged Watersheds Using HEC-1, U.S. Army Corps of Engineers, Davis, California.

International Commission on Large Dams (1991), "Selection of Design Flood," Bulletin

82, Paris, France.
Kane, D.L. and J. Stein (1983), "Water Movement into Seasonally Frozen Soils," Water
Resources Research, 19(6):1547-1557.

King, H.W. and E.F. Brater (1954), Handbook of Hydraulics, McGraw Hill, Inc., New
York, New York.

Linsley, Ray K., Jr. and Joseph B. Franzini (1979), Water Resources Engineering, 3rd
edition, McGraw-Hill Inc., New York, New York

Linsley, Ray K., Jr., Max A. Kohler and Joseph L. H. Paulhus (1975), Hydrology for
Engineers, 2nd edition, McGraw-Hill Inc., New York, New York

Maidment, David R., Editor in Chief (1993), Handbook of Hydrology, McGraw Hill Inc.,
ISBN 0-07-039732-5. See page 10.3, Routing Model Selection, by D.L. Fread.

McCuen, Richard H. (1998), Hydrologic Analysis and Design, Prentice-Hall, Inc., Upper
Saddle River, New Jersey

Mead, D.W. (1908), Water Power Engineering, McGraw Hill Book Company, New
York, New York.

Miller, J.F., Frederick, R.H. and Tracey, R.S. (1973), NOAA ATLAS 2,
Precipitation–Frequency Atlas of the Western United States, Volume IX–Washington,
United States Department of Commerce, National Oceanic and Atmospheric
Administration, National Weather Service, Washington, D.C.

-76-
Myers, V.A. (1967), "The Estimation of Extreme Precipitation as the Basis for Design
Floods, Resume of Practice in the United States," Proceedings of the Leningrad
Symposium, Floods and Their Computation, International Association of Scientific
Hydrology, August.

Morgan, A.E. (1914), "Discussion on Flood Flows," Transactions of American Society of

Civil Engineers, Vol. 77, American Society of Civil Engineers, New York, New York.

National Cartography and GIS Center (1994), State Soil Geographic (STATSGO) Data
Base - Data use information, United States Department of Agriculture, Natural Resources
Conservation Service, Fort Worth, Texas

National Weather Service (1956), Seasonal Variation of the Probable Maximum

Precipitation East of the 105 th Meridian for Areas from 10 to 1,000 Square Miles and
Durations of 6, 12, 24, and 48 Hours, Hydrometeorological Report No. 33, United States
Department of Commerce, National Oceanic and Atmospheric Administration, United
States Weather Bureau, Washington, D.C.

National Weather Service (1963), Hydrometeorological Report No. 39, Probable

Maximum Precipitation in the Hawaiian Islands, United States Department of Commerce,
National Oceanic and Atmospheric Administration, United States Weather Bureau,
Washington, D.C.

National Weather Service (1964), Hydrometeorological Report No. 40, Probable

Maximum Precipitation over the Susquehanna River Basin above Harrisburg,
Pennsylvania, United States Department of Commerce, National Oceanic and
Atmospheric Administration, United States Weather Bureau, Washington, D.C.

National Weather Service (1965), Hydrometeorological Report No. 41, Probable

Maximum Precipitation Over the Tennessee River Basin Above Chattanooga, United
States Department of Commerce, National Oceanic and Atmospheric Administration,
United States Weather Bureau, Washington, D.C.

National Weather Service (1973), Hydrometeorological Report No. 47, Meteorological

Criteria for Extreme floods for Four Basins in the Tennessee and Cumberland River
Watershed, United States Department of Commerce, National Oceanic and Atmospheric
Administration, United States Weather Bureau, Washington, D.C.

National Weather Service (1973), Hydrometeorological Report No. 48, Probable

Maximum Precipitation and Snowmelt Criteria for Red River of the North above
Pembina, and Souris River above Minot, North Dakota, United States Department of

-77-
Commerce, National Oceanic and Atmospheric Administration, United States Weather
Bureau, Washington, D.C.

National Weather Service (1977), Hydrometeorological Report No. 49, Probable

Maximum Precipitation Estimates, Colorado River and Great Basin Drainages, United
States Department of Commerce, National Oceanic and Atmospheric Administration,
United States Weather Bureau, Washington, D.C.

National Weather Service (1978), Hydrometeorological Report No. 51, Probable

Maximum Precipitation Estimates–United States East of the 105th Meridian, United
States Department of Commerce, National Oceanic and Atmospheric Administration,
United States Weather Bureau, Washington, D.C.

National Weather Service (1982), Hydrometeorological Report No. 52, Probable

Maximum Application of Precipitation Estimates–United States East of the 105th
Meridian, United States Department of Commerce, National Oceanic and Atmospheric
Administration, United States Weather Bureau, Washington, D.C.

National Weather Service (1980), Hydrometeorological Report No. 53, 1980 Seasonal
Variation of 10-Square-Mile Probable Maximum Precipitation Estimates, United States
East of the 105th Meridian, United States Department of Commerce, National Oceanic
and Atmospheric Administration, United States Weather Bureau, Washington, D.C.

National Weather Service (1983), Hydrometeorological Report No. 54, Probable

Maximum Precipitation and Snowmelt Criteria for Southeast Alaska, United
States Department of Commerce, National Oceanic and Atmospheric Administration,
United States Weather Bureau, Washington, D.C.

National Weather Service (1988), Hydrometeorological Report No. 55A, Probable

Maximum Precipitation Estimates–United States Between the Continental Divide and the
103 rd Meridian, United States Department of Commerce, National Oceanic and
Atmospheric Administration, United States Weather Bureau, Washington, D.C.

National Weather Service (1986), Hydrometeorological Report No. 56, Probable

Maximum Precipitation Estimates with Areal Distribution for Tennessee River Drainages
Less Than 3,000 Square Miles in Area, United States Department of Commerce, National
Oceanic and Atmospheric Administration, United States Weather Bureau,
Washington, D.C.

National Weather Service (1994), Hydrometeorological Report No. 57, Probable

Maximum Precipitation for the Pacific Northwest, United States Department of
Commerce, National Oceanic and Atmospheric Administration, United States Weather

-78-
Bureau, Washington, D.C., October 1994.

National Weather Service (1998), Hydrometeorological Report No. 58, Probable

Maximum Precipitation for California, United States Department of Commerce, National
Oceanic and Atmospheric Administration, United States Department of the Army, Corps
of Engineers, Silver Spring, MD, October 1998.

National Weather Service (1999), Hydrometeorological Report No. 59, Probable

Maximum Precipitation for California, United States Department of Commerce, National
Oceanic and Atmospheric Administration, U.S. Department of the Army, Corps of
Engineers, Silver Spring, MD., February 1999.

National Weather Service (1961), Technical Paper 42, Generalized Estimate of Probable
Maximum Precipitation and Rainfall Frequency Data for Puerto Rico and Virgin Islands,
United States Department of Commerce, National Oceanic and Atmospheric
Administration, United States Weather Bureau, Washington, D.C.

National Weather Service (1963), Technical Paper 47, Probable Maximum Precipitation
and Rainfall–Frequency Data for Alaska, United States Department of Commerce,
National Oceanic and Atmospheric Administration, United States Weather Bureau,
Washington, D.C.

Newton, Donald W. (1983), "Realistic Assessment of M aximum Flood Potentials,"

Journal of the Hydraulics Division, American Society of Civil Engineers, Volume 109,
No. 6, June 1983, figures 4 and 5, page 911.

Pilgrim, D.H., I.A. Rowbottom, and G.L. W right (1988), "Estimation of Spillway Design
Floods for Australian Dams," Transactions, 16th Congress of the International
Commission on Large Dams, San Francisco, California.

Post, F.A. and F.R. Dreibelbis (1942), "Some Influences of Frost Penetration and
Microclimate on the W ater Relationships of Woodland, Pasture and Cultivated Soils,"
Soil Science Proceedings, 7:95-104.

Snitter, N.J. (1979), Transactions of the 13th Congress, New Delhi, India, International
Commission on Large Dams, Paris, France.

Snyder, F.F. (May 1964), "Hydrology of Spillway Design: Large Structures - Adequate
Data," Proceedings of the American Society of Civil Engineers, Journal of the Hydraulics
Division, New York, N.Y.

-79-
Snyder, F.M. (1938), "Synthetic Unit Graphs," Transactions of the American
Geophysical Union, Vol. 19.

Soil Conservation Service (1985), National Engineering Handbook, "Section 4,

Hydrology," United States Department of Agriculture, Washington, D.C.

Soil Conservation Service (1986), "Urban Hydrology for Small Watersheds," Technical
Release No. 55, (TR-55), Washington, D.C.

Trimble, G.R., T.S. Sartz, and R.S. Pierce (1987), "How Type of Soil Frost Affects
Infiltration," Journal of Soil and Water Conservation, 13(1):81-82.

U.S. Army Corps of Engineers (1946), Engineering Manual for Civil Works, Part 2,
Chapter 5, Washington, D.C.

U.S. Army Corps of Engineers (1960), Runoff from Snowmelt, Superintendent of

Documents, Washington, D.C.

U.S. Army Corps of Engineers (1987), Hydraulic Design Criteria, Waterways Experiment
Station, Vicksburg, M ississippi.

United States Department of Agriculture (1955), Yearbook of Agriculture,

Washington, D.C.

United States Department of the Interior, Bureau of Reclamation (1987), Design of Small
Dams, Denver, Colorado

United States Geological Survey (1982), Water-Resource Investigations 82-13, "Time of

Concentration and Storage Coefficient Values for Illinois Streams," (J.B. Graf, G.
Garklavs, and K.A. Oberg), Water Resources Division, Urbana, Illinois.

United States Geological Survey (1988), Water Supply Paper 2339, "Guide for Selecting
Roughness Coefficients for Natural Channels and Flood Plains," Washington, D.C.

United States Geological Survey (1986), Water Resources Investigations Report 86-4192.
"Techniques for Simulating Flood Hydrograph and Estimating Flood Volumes for
Ungaged Basins in Central Tennessee," (C.H. Robbins), Water Resources Division,
Nashville, Tennessee.

United States Geological Survey (1986), Water Resources Investigations Report 86-4004.
"Simulation of Flood Hydrograph for Georgia Streams," (E.J. Inman), Water Resources
Division, Doraville, Georgia.

-80-
United States Geological Survey (1988), Water Resources Investigations Report 88-404.
"Estimating Flood Hydrograph and Volumes for Alabama Streams," (D.A. Olin and
J.B. Atkins), Water Resources Division, Tuscaloosa, Alabama.

United States Geological Survey (1990), Water Resources Investigations Report 89-4087.
"Determination of Flood Hydrograph for Streams in South Carolina," (L.R. Bohman),
Water Resources Division, Columbia, South Carolina.

World M eteorological Organization (1986), "Manual for Estimation of Probable

Maximum Precipitation," Operational Hydrology Report No. 1, Secretariat of the W orld
Meteorological Organization, Geneva, Switzerland.

Zuzel, J.F. and J.L. Pikul, Jr. (1987), "Infiltration into a Seasonally Frozen Agricultural
Soil," Journal of Soil and W ater Conservation, 42:447-450.

-81-
8-12 Glossary

Some hydrologic terms have slightly different definitions depending upon the agency
using them. These terms have been defined on the basis of the meaning used in these
Guidelines.

Accuracy - The state of being free from errors, i.e. the absolute nearness to the truth. In
physical measurements, it is the degree of agreement between the quantity measured and
the actual quantity. For example, clock records of rainfall and streamflow can be out of
synchronization, implying that measured time is not accurate. The prediction accuracy of
a rainfall-runoff model should not be confused with "precision," which denotes the
reproducibility of the measurement or computation and its refinement, e.g. 0.001, 0.01,
0.1, or 0.

Active Storage - That portion of reservoir storage which is filled and emptied from year
to year as the reservoir is operated normally.

Altitude-Depth Relationship - A relationship between snowpack water equivalent and

elevation for a given drainage basin.

Antecedent Moisture Condition (AMC) - The degree of wetness of a watershed at the

beginning of a storm. Three levels of AMC are designated as AMC-I, AMC-II, and
AMC-III. AMC-I is the lower limit of moisture; AMC-II is the average moisture; and
AMC-III is the upper limit of moisture with relatively small, medium, and large curve
numbers (CNs), respectively.

Antecedent Storm - A storm that occurred prior to the storm of interest.

Baseflow - The streamflow rate occurring during recession of a hydrograph. Baseflow is

separate from direct runoff.

Basin - The surface area within a given drainage system.

Basin-Averaged (Uniform Infiltration Loss Rate) Method (or Basin-Approximate

Method) -
The method is the most practical approach to estimate the area-weighted constant loss
rate of soils for a (sub)basin. Particularly for PMF runoff computations, the soil should
be assumed to be saturated with infiltration occurring at the minimum rate that has been
empirically determined in relation to the hydrologic soil group from USDA [1955]
literature. The spatially average soil classification should be established for the drainage
area that can be identified with a SCS Hydrologic Soil Classification (A, B, C, or D).

-82-
Basin Average Rainfall - The spatially averaged rainfall depth within a drainage basin for
a particular given total storm or time increment of that storm.

Basin (or Watershed) Characteristics - The physical characteristics of a drainage basin

that control its average hydrologic response in terms of runoff. These characteristics
include watershed relief (the most common parameters are channel slope, watershed
slope, and hypsometric curve; the greater the relief, the shorter the time of concentration,
Tc), watershed shape, direction, altitude, drainage pattern, land use and soil type, time of
flow parameters (commonly including Tc, time lag, reach travel times, etc.), storage and
vegetation within channels, etc.

Bucket Surveys - Supplemental "unofficial observations" (e.g., observations made by

individuals, radio and TV stations, and city and county public works departments)
surveys of precipitation data conducted immediately after the occurrence of severe storm
and flood events. The bucket surveys are beneficial as the network of precipitation
stations is still far from sufficient to provide the necessary temporal and spatial data for
detailed analyses of observed storm precipitation.

Calibration - A process of adjusting input parameters of a rainfall-runoff model within

physical limits using a 'trial-and-error" procedure to compare the output of the computed
hydrograph with the observed hydrograph or measured values until the model
satisfactorily or most closely simulates a hydrologic system it represents.

Channel Slope - The gradient measured by drop in elevation over channel distance, in
foot per foot. The application should be consistent with the methodology.

Channel Storage - The total volume of flowing water in the stream channel under
consideration during a period of time, or the in-channel stored water volume depending
on the stage of the water surface in the channel under consideration at any time.

Clark Unit Hydrograph - A synthetic unit-hydrograph developed by C.O. Clark for which
two parameters Tc and storage coefficient (R), and a time area curve are estimated to
account for storage in the basin as well as movement of runoff by translation of the flood
wave. This method uses the concept of the instantaneous unit hydrograph to define a
unique unit hydrograph for a gaged or ungaged basin.

Coefficient of Determination (r2 ) - A measure of the degree to which a regression line

explains the variance in the dependent variable.

Composite Unit Hydrograph - The unit hydrograph developed from the unit hydrograph
generated from historical storms and flood data. It is the unit hydrograph judged to be
representative of the hydrologic response of the drainage-basin system.

-83-
Consistency - The status of agreement or compatibility among hydrologic data if no
unusual changes or are present in the data.

Continuity - An uninterrupted succession of a record if the record contains no periods for

which data are missing. If data collection is based on a certain frequency, this process
may result in missed observations of significant events; such as daily flow measurements
at set times may miss peak values.

Continuous Streamflow Hydrograph - A hydrograph formed from continuous stage

recording at a streamgage.

Contributing Area - The total area from which surface water commonly is removed by
gravity to the outlet in a drainage basin.

Cross Section - A vertical section taken across a stream channel or a reservoir, usually
used to determine flow area and hydraulic radius for flow routing.

Daily Flow Records - A record of average daily flows at a streamgage.

Degree-Day Method - A method to calculate snowmelt in terms of a degree-day factor

[HEC 1990] determined from measured snowpack, runoff, and temperature for a
historical storm. Degree-day is defined as a day with an average temperature on degree
above 32 degrees F. The average is usually obtained by averaging the maximum and
minimum for the day.

Design Flood - The flood hydrograph for which a given project and its appurtenances are
designed.

Dimensionless Unit Hydrograph - A unit hydrograph whose vertical and horizontal

coordinates have been made dimensionless by dividing by the hydrograph peak flow and
the time to peak, respectively.

Disaggregation - The process of converting rainfall depths for one increment of time to
the incremental depths for smaller increments of time.

Distributed (Uniform Infiltration Loss Rate) Method (or Detailed Method) - The method
is an approach to estimate loss rates for the (sub)basin based on physical soil properties
using spatially detailed soils and land cover maps of the basin. Theoretically, any
measure of loss estimation may be applied in a distributed fashion. For instance, the
STATSGO's hydrologic soil groups or soil series or digital data of similar (or finer if
available) spatial resolution can be used to apply the distributed loss method. This

-84-
method allows the hydrologist to calibrate loss rates (if the (sub)basin gaged) on the basis
of permeability of each soil type, modified as necessary for other factors, such as
bedrock, groundwater conditions, AM C, land cover, land features, etc., that may affect
runoff.

Distributed-Parameter Models - Rainfall-runoff simulation models that account for spatial

variations in hydrologic parameters from point to point throughout a drainage basin.
These models can also minimize the effect of lumpimg watershed characteristics such as
soil types, soil profiles, impervious areas, and land uses into single parameters
representing the entire catchment.

Double-M ass Analysis - A plot of accumulated rainfall depth for one raingage against
average accumulated depth at another gage (or group of gages) in the same climatic area
used to detect trends or inconsistencies within the data.

Drainage Area - The area of a drainage basin.

Drainage Basin - The area contributing direct runoff to a stream. Specifically, the
delineated land bounded by a hydrologic surface drainage divide (or topographic divide,
i.e., the line that follows the ridges or summits forming the exterior boundary of a
drainage basin), from which surface runoff is drained to a point of interest (i.e., the
outlet) on a watercourse. Also called a drainage area or a catchment (i.e., the land
tributary to a stream) for some cases or, on a large scale, a watershed.

Drainage Pattern - An indicator of the drainage flow characteristics of storm runoff in a

basin, which can be represented by a number of parameters such as drainage density,
Horton's laws, etc.

Duration - The length of an actual or assumed period of time over which rainfall or
rainfall excess occurs.

Dynamic Effects - The effects on a channel or reservoir inflow hydrograph caused by

several factors which generally are considered in an unsteady flow's continuity and
momentum equations, including flood-wave wedge storage, water surface rate-of-rise,
lateral local inflow per unit distance, local acceleration, convective acceleration, etc.

Dynamic Wave - The wave resulting from a change in flow rate in an open channel with
the movement properties principally following the continuity and momentum (i.e., inertial
influences also considered) equations. It is a wave whose behavior is dependent not only
on depth, but on effects of local and convective acceleration.

Dynamic (i.e. unsteady) Reservoir Flood Routing - The flood routing procedures through

-85-
a reservoir using momentum and continuity equations by considering dynamic effects as a
large inflowing flood wave passes through the reservoir pool.

Emergency Gate Operation - The operation of gates on a controlled spillway when there
is danger of the dam being overtopped if the gates are not opened sufficiently or for an
emergency situation.

Energy-Budget Method - A method to calculate snowmelt due to heat transferred from

rain and the environment to the snowpack. Energy inputs to the snowmelt process are
longwave and shortwave (solar) radiation, convection, release of heat due to
condensation, ground heat, and heat introduced by rain (U.S. Army Corps of Engineers,
1956).

Envelope Curve - A smooth curve covering all peak values of rainfall plotted against
other factors, such as area or time.

Extreme (or Major) Flood - A flood whose peak flow is significantly larger than most
historical floods.

Flashboards - Structures which temporarily raise the crest of an overflow spillway.

Usually the flashboards are made from wooden planks supported by structural members.

Flood - A runoff event that causes a river or reservoir to rise above normal nondamaging
limits

Flood Hydrograph - A record of continuous streamflow versus time for a given flood at a
selected location on a stream.

Flood Peak - The highest flow discharge attained during the passage of a flood wave

Flood Routing - The process of progressively determining the timing and shape of a flood
wave at successive points along a river to estimate the outflow flood at a downstream
point from the inflow flood at an upstream point. Flood routing methods may be
classified as either hydrologic routing (i.e., lumped flow routing) or hydraulic routing
(i.e., distributed flow routing). Basically, in hydrologic routing, the flow is computed as
a function of time at one location along the watercourse. However, in hydraulic routing,
the flow is computed as a function of time simultaneously at several cross sections. Some
typical, well accepted hydrologic routing methods include Muskingum River and level-
pool reservoir routing methods and hydraulic routing methods include Muskingum-Cunge
and dynamic-wave routing methods.

Flood Storage - That portion of reservoir storage which is expressly reserved for storage

-86-
of flood water.

Flood W ave - A large moving swell of water on the surface of a water body.
Specifically, it is a distinct rise in streamflow to a crest in response to runoff generated by
precipitation or melting snow and its subsequent recession after the precipitation or
melting snow ends.

Front - The interface between two air masses of different density. When a front is moving
in the direction from cold air to warm, a cold front results, and vice versa.

Gaged Basin - A watershed where available hydrologic data, recorded at stations within
the basin, are sufficient in quantity and quality to provide confidence in development of a
hydrograph at the drainage-basin outlet.

General Storm - A storm caused by a frontal movement which generally covers a large
area (ranging from over 500 up to 60,000 mi2) and has a duration longer than 6 hours.

Geographic Information System (GIS) - An electronic system of maps (points, lines, and
polygons) connected to tables of data that describe the features on the maps with the
ability of managing (capture and storage), manipulating (retrieval and analysis), and
displaying spatial data. For instance, the integration of GIS with the curve number model
is an example of a GIS application to determine curve numbers for runoff analysis
through processing spatial data such as land use, land cover, hydrologic soil group, slope,
and other factors varying across a drainage basin by applying GIS overlay with the
NRCS' STATSGO.

High-Water Mark - A mark which identifies the maximum stage which occurred at a
particular location during an historical flood.

Homogeneous Data - Hydrologic data that all comes from the same phenomena during
the same time period.

Hydraulics - The physical science and technology of the static and dynamic behavior of
water or other fluids; about dealing with fluid properties, or the mechanism (i.e., a
system of governing physical laws) of fluid flows or forces and its applications in
engineering.

Hydrograph - Rate of flow in a stream plotted against time for a particular section.

Hydrology - The scientific study of water on and within the earth and related
applications; about dealing with water's occurrence, quantification, spatial and temporal
distributions, circulation, interactions with its environments such as ground surface,

-87-
underground media, atmosphere, etc.

Hydrologic Characteristics - The physical characteristics of precipitation (e.g. mean or

isohyetal, intensity, duration, frequency, etc.), evapotranspiration, and streamflow (e.g.
monthly and annual volumes, low-flow rates, floods, etc.) in a given drainage basin. The
primary controlling features of these characteristics are the basin's physiographic regions
and climatic patterns.

Hydrologic Condition - The feature (or factor) of land cover that can influence infiltration
based on the status of treatment or practice of the surface vegetation when the SCS CN
method is used to estimate runoff. The hydrologic condition usually is classified as good,
fair, or poor to reflect a relatively low, average or high CN or runoff, respectively.

Hydrologic Soil Groups - The classified four groups of soils (A, B, C, D) that are
designated by SCS to indicate the degree of runoff potential, very low, low, moderate,
and high as their infiltration rate are high, moderate, slow and very slow infiltration rates
under similar storm and cover conditions, respectively.

Hydrologic Units (HU) - Subbasins or subwatersheds. Each HU is the drainage area of a

minor tributary flowing into the main stream or a major tributary.

Hydrometeorology - The interdisciplinary science of meteorology and hydrology related

to the occurrence of extreme rainfall and extreme floods.

Hydrometeorological Report - Name given to a set of National Weather Service

publications. These publications contain generalized studies of extreme rainfall for a
given region. Such reports provide generalized information for estimating probable
maximum precipitation of a particular duration for given locations within the region.

Hyetograph - A graph of incremental rainfall depth, rainfall excess, or both versus time at
a sampling point or for a drainage basin.

Infiltration Capacity - The maximum rate at which the soil, when in a given condition,
can absorb falling rain or melting snow. The infiltration capacity depends on (a) basin
characteristics such as soil type, land use, and vegetation cover; (b) climate characteristics
such as rainfall intensity, temperature; and (c) underlying geological conditions.

Infiltration Rate - The rate at which rainfall is absorbed through the soil surface and into
the subsoil. It must equal the infiltration capacity or the rainfall rate, whichever is lesser.
It is expressed in depth of water per unit time (usually inches per hour). The infiltration
rate is determined by the smaller of either the entry (or penetration) rate at which water
enters the surface of the soil or the transmission rate at which the water percolates the soil

-88-
in either the vertical or horizontal direction in a given drainage area. The infiltration rate
is varied until it reaches the given soil's ultimate minimum infiltration capacity (an
approximate constant rate) under a fully saturated condition.

Inflow PMF Hydrograph - The hydrograph which represents PMF runoff entering a
reservoir.

Initial Abstraction (or Loss) - The portion of rainfall on a basin that is intercepted by
vegetation, held in derpressions, or evaporated.

Initial (or starting) Flow - The streamflow equal to the subsurface flow (i.e. without
direct runoff) at the time when a hydrograph simulation begins. In HEC-1, the initial
flow is the variate (i.e. random variable) designated as the STRTQ that does not
necessarily start from the streamflow without direct runoff.

Interflow - Water that infiltrates the soil and reappears as seepage or spring flow during
the period of runoff. Interflow is also called a rapid subsurface flow or subsurface runoff.

Isohyet - A line on a map, connecting points of equal rainfall amounts. Isohyets are used
to develop an isohyetal map of single storm or annual rainfall depths for given time
intervals.

Isohyetal Pattern - Spatial distribution of rainfall represented by lines of equal rainfall

depth (isohyets).

Kinematic Wave - The wave resulting from a change in flow rate in an open channel with
the movement property principally following from the equation of continuity. The
velocity of the wave is proportional to the change in depth.

Lag Time (T L) - The time which locates the runoff hydrograph relative to the occurrence
of a storm. Lag time generally is determined as the difference in time between the
centroid of rainfall excess and the peak of the runoff hydrograph, for instance, T L applied
in the Snyder or the NRCS dimensionless unit hydrograph methodologies. However,
definitions of T L differ depending on the methodology used. In the SCS dimensionless
unit hydrograph method for example, the average relation of lag time to time of
concentration is TL = 0.6 T c .

Land Cover - The extent and type of vegetation covering the drainage basin.

Lapse Rate - The rate at which air temperature decreases with increasing altitude within a
given drainage basin.

-89-
Level-Pool (Reservoir) Routing - The reservoir flood routing technique which assumes
the reservoir always has a horizontal water surface throughout its length. That is, the
reservoir is sufficiently large that the inflow has a negligible effect on the outflow.

Local Storms - Thunderstorms resulting from local convection which covers a limited
area, generally not more than 500 mi2 .

Loss Rate - The rates of infiltration for a given soil type.

Lumped-Parameter Models - Rainfall-runoff simulation models that ignore spatial

variations in hydrologic parameters throughout a drainage basin. A typical example of a
lumped parameter is the time of concentration which is held constant for all storms to
compute a unit hydrograph.

Manning Equation - The following equation for calculation of the average uniform
velocity in an open channel V = 1.486/n * R H b * S½ where V is the average velocity, R H
is the hydraulic radius for the section, S is the average slope of the channel, and "n" is a
coefficient reflecting the roughness of the channel. The equation should be applied to
segments of the channel that have constant slopes and gradually varying geometric
characteristics.

Manning's "n" - The coefficient used in the denominator of the Manning equation to
represent the effect of channel roughness.

Maximum Normal Operating Level - The maximum reservoir water-surface elevation

which a hydroelectric project is normally operated during the year.

Maximum Possible Flood - An earlier term used to describe the Probable Maximum
Flood.

Maximum Probable Precipitation - An earlier term used to describe the Probable

Maximum Precipitation.

Minimum Infiltration Rate - The minimum rate at which infiltration occurs after the soil
is saturated. This minimum rate is governed by the rate at which precipitation can enter
the soil surface and percolate to the subsurface.

Model - A physical process-simulation system that accounts for all of its known
properties by relating known inputs with outputs. In mathematical (or digital) models (or
formulations), the behavior of the system is represented with a set of equations that
numerically simulate the system behavior to predict hydrologic events resulting from
representative future hydrologic inputs. Alternative mathematical models of hydrologic

-90-
processes from various approaches have been produced to reproduce the historical record
in a statistical sense.

Nonlinear Effects - The tendency for a drainage basin to yield peak flows for greater
depths of storm runoff which are larger than a linear proportion would indicate.

NEXRAD (Next Generation Weather Radar) Precipitation Data - The precipitation

measurement obtained by using a radar system with hydrometeorological capabilities of
collecting information for gage-adjusted radar-rainfall estimate. The most important
advantage of using radar measurement of rainfall is the coverage that the radar system
provides of a large area for precipitation estimation with high spatial and temporal
resolution as small as 6 minute time interval and 4 km 2 area, respectively.

Operation Rules - The rules by which controlled spillways and outlet works are operated.

Optimization Process - The process of determining a set of parameters that best

replicates an observed runoff hydrograph for a drainage basin. The HEC-1 program has
an automatic process built into the program which performs such an optimization.

Orographic Effects - The effects of topographic variations on precipitation.

Overland Flow - Runoff flowing over the surface of a drainage basin prior to reaching a
channel.

Peak Flow - The maximum flow rate on a runoff hydrograph.

Permeability (or Permeability Coefficient) (k) - The capacity (or ability) of a geologic
material (e.g., soil) to transmit water through it or the quality of the soil that enables
water to move through it while overcoming surface tension and any other capillary
actions under given water and soil properties. It is expressed in water transmission
distance per unit time (usually inches per hour). Historically SCS' soil survey has used
"vertical" permeability as a term for saturated hydraulic conductivity that is an estimate of
vertical water movement in the given soil column under a saturated condition at a
temperature of 60 degrees F. The degree of permeability depends upon the size and
shape of the opening and the extent of the interconnections of permeable substance. For
example, a high permeability occurs for soils having a large porosity, such as sands and
gravels.

Preliminary Data - Physical and hydrologic data collected for a given project and its
drainage basin prior to making a visit to the site.

Probable Maximum Flood (PMF) - The flood that may be expected theoretically from the

-91-
most severe combination of critical meteorologic conditions that usually produce the
PMP and critical hydrologic conditions that are reasonably possible in the drainage basin
under study.

Probable Maximum Precipitation (PMP) - The greatest depth of precipitation

theoretically for a given duration that is physically possible over a given size storm area
at a particular geographical location during a certain time of year.

Probable M aximum Storm (PMS) - A total design storm that has been adjusted,
realistically patterned (spatially-temporally) on the basis of recorded storms, to a single
PMP area size and duration deemed critical.

Rainfall Excess - The part of the rainfall appearing as direct runoff which is not lost to
infiltration, depression storage, and interception. (In comparison, effective rainfall
includes rainfall excess and interflow)

Rainfall Sequence - The sequence of incremental rainfall depth used to develop a runoff
hydrograph resulting from the storm.

Rating Curve - A relationship between stage and flow rate developed for a particular
streamgage location.

Recession - The portion of the hydrograph showing a decline in the rate of runoff.

Reconstitution - The analytical process of using a developed unit hydrograph and

historical storm rainfall to reproduce a historical flood hydrograph.

Redundant Operating System - An additional system for operating spillway and outlet
works gates which is independent of all other systems.

Regional Studies (Analyses) - Studies of hydrologic data from drainage basins in a

hydrologically homogeneous or representative region to develop generalized information
for calculation of a unit hydrograph for an ungaged area.

Regression - The mathematical analysis performed to assess the statistical correlation and
linear relation between a hydrologic parameter and physical or other hydrologic
parameters for the drainage basin. Through logarithmic transformations some forms of
nonlinear relations can be evaluated.

Remote Sensing Data - The measurement of the electromagnetic spectrum to characterize

the landscape by using aerial photos, airborne sensors, or other like data collected by
satellite.

-92-
Representative Unit Hydrograph - The unit hydrograph which represents the hydrologic
response of the drainage basin. It is the same as the composite unit hydrograph.

Reservoir Starting Level - The reservoir water-surface elevation assumed to be present at

the beginning of the inflow PMF.

Reverse-Reservoir Routing - The reservoir routing technique that assumes the use of a
level pool to develop an inflow unit hydrograph based on the recorded reservoir outflow
and headwater elevations during one or more major floods. This assumption will be more
accurate only if the size of the reservoir is relatively large and/or wedge storage during
the passage of the flood wave is relatively small.

River Basin - The drainage basin for a river upstream from a selected point.

Routed Outflow - The downstream hydrograph that results from routing of a flood
hydrograph through a reservoir using the relevant capacities of the spillway and outlet
works.

Routing - The analytical process of computing the movement of a flood wave as it passes
through a reservoir or a channel.

Runoff Curve Number (CN) - The number (up to 100) assigned by SCS to hydrologic
soil-cover complexes (i.e., a combination of a hydrologic soil group and a land use and
treatment class) to indicate the runoff potential, i.e., the larger the CN, the greater the
runoff potential.

Runoff Modeling - The analytical process of computing the portion of rainfall from a
given storm, and/or snow melt that runs off the land into surface waters. In these
Guidelines, the unit hydrograph is used as an essential component of the runoff model.

Safety Evaluation (As applied to a dam.) - The process of determining the ability of dam
and its appurtenances to pass a given flood.

Sediment - The produced materials of wearing away of the land surface by erosive agents
such as water, wind, ice, and gravity in a process of natural geologic erosion or
accelerated erosion resulting from land-use alterations.

Sensitivity Analysis - The process to find the rate of change of one hydrologic factor with
respect to change in another factor to either measure the effect of one factor on another or
explore the importance or influence of one element with respect to some criterion.
Mathematically, sensitivity is simply defined as the derivative of model results with

-93-
respect to the model's input parameter of interest. Sensitivity of input parameters of
numerous variables in the categories of runoff and reservoir operations can be analyzed
with respect to the peak flow discharge.

Slope-Area Method - A process of determination, by indirect measurement, of the flood

peak discharge by field survey of a reach of channel and high-water marks, usually after a
flood has passed (the formulas referring to Chow, 1959, p147). Usually, discharge is
computed by the Manning formula with modified "n" values (relatively, small "n" values
for high discharge but large for low discharge) to account for nonuniform flow through
regular valley channels free from bends.

Snow Course - A defined line on a drainage basin, laid out and permanently marked,
along which depths of snowpack and water content are determined from the sampled
snow at definite distances or stations at appropriate times during a snow survey and
recorded on a regular basis.

Snow Cover -. The accumulated snow and ice on the surface of the ground in a drainage
basin at any time.

Snowfall - Precipitation assumed to fall as snow if the zone temperature is less than the
base temperature (i.e. freezing temperature varied from 320 F) plus two degrees.

Snowmelt - Melt of snow occurring when the temperature is equal to or greater than the
base temperature that varies with the zone atmospheric pressure.

Snowmelt Calculation - Estimation of the snowmelt occurring for a given snowpack and a
given set of meteorologic conditions.

Snowpack - The depth of existing snow in a drainage basin expressed in equivalent water
content.

Snow Pillow - A device for the measurement of snow pack water equivalent through a
process of weighing the overlying snow.

Snyder Unit Hydrograph - A synthetic unit hydrograph for an ungaged basin developed
by F.M. Snyder for which the peak flow and lag time are estimated in terms of regional
parameters (standard lag, tp and storage coefficient, C p).

Soil Map - A map identifying and showing the areal distribution of soil types.

Soil Moisture Content - The quantity of water present in the soil, usually expressed in
percentage of wet weight. The quantity at the beginning of a historical storm is of

-94-
primary interest.

Soil W ater Storage - The amount of water the soils of a watershed which is stored at a
given time. The amount for a given watershed is continually varying as rainfall or
evapotranspiration takes place.

Spatial Rainfall Distribution - The variation of rainfall with location in a drainage basin.

Spillway - The structure provided to pass flows which are not passed through the outlet
works or the power plant. The spillway may be an overflow type or an orifice type.

Spurious Trend - A trend in hydrologic data with time that appears in the data but is
actually is the result of data errors or other anomalies rather than a real climatic effect.

Standard Error of Estimate - The square root of the variance between values of a given
hydrologic data set and values estimated with a statistical, mathematical, or other model
of the process of interest.

STATSGO - The State Soil Geographic Database at a scale of 1:250,000 that provides
soil association maps and related data, including soil hydraulic conductivity, available in
digital format from the NRCS.

Storage Coefficient (R) - A coefficient used with the Clark unit hydrograph which is
related to storage effects of the basin. For estimation of this parameter see Figure 8-6.2.

Storage of Watershed - The total water volume stored in a watershed of interest; the
difference between inputs (e.g., rainfall, base flow) and outputs (e.g., runoff, infiltration
and other losses) of the hydrologic cycle during a period of time. Watershed storage
directly affects the shape and the time distribution of the runoff hydrograph.

Storm Transposition - The analytical process of moving historical storm data from the
location where it occurred to the location of interest.

Streamflow - The record of flow rate at a given point in a stream.

Streamgage - A gage which measures and records the water-surface elevation (stage) in a
stream. The recorded stage is converted to streamflow by use of a rating curve.

Subbasin - A subdivision of a drainage basin.

Subdivision - The process of dividing a drainage basin into subbasins.

-95-
Synthetic Unit Hydrograph - A unit hydrograph for an ungaged basin that has been
developed based on unit hydrograph developed at gage sites within a hydrologically
homogeneous or representative region. Synthetic unit hydrograph are estimated for
ungaged basins by means of relations between parameters of the unit-hydrograph model
and the physical characteristics of the basin.

Temporal rainfall distribution - The variation of rainfall depth with time in order for a
given storm.

Time of Concentration (T c) - There are two commonly accepted definitions depending

upon the method used: (1) the rational method defines Tc as the time required for runoff
or water to travel from the most hydraulically distant point in the watershed to the outlet
or point of interest, or (2) the Clark unit hydrograph method defines T c as the time
between the end of rainfall excess from a rainfall hyetograph and the inflection point on
the recession of the direct runoff hydrograph.

Thiessen Polygon Method - The method of dividing a drainage basin into individual
polygons each raingage represents within these subdivisions the basin-average rainfall for
a given storm is estimated by calculating a mean rainfall amount at each gage station
based on each subdivision area's weight in proportion to the basin area. The subdivision
is made by developing polygons whose boundaries are defined by lines bisecting the lines
connecting adjacent gage locations.

Uncontrolled Spillway - A spillway where overflow is not controlled.

Ungaged Basin - A basin for which available hydrologic data, recorded at stations within
the basin, are insufficient in quantity and quality to provide confidence in development of
an inflow hydrograph, or a basin for which input and output measurements necessary for
calibration are not available.

Uniform Loss Rate - The constant rate of infiltration, or called the minimum infiltration
rate, assumed to occur after initial losses and soil saturation have been satisfied. It is can
be estimated from average soil characteristics or calibrated from significant historical
flood events.

Unit Duration - The time increment of rainfall to be used in the unit-hydrograph analysis
which is usually calculated as TL /5.5, where TL is a lag time, rounded down to an even
number.

Unit Hydrograph - The direct runoff hydrograph from a given drainage basin representing
one unit (inch or mm) of rainfall excess for a specified duration and areal distribution.
Typically the rainfall excess should be spatially and temporally uniform.

-96-
Urban Area - An area which has been developed for urban use.

Validation - The process of demonstrating that estimated model parameters or basic

theory behind a model is correct or valid. This can only be done by carefully controlled
experiments with new detailed and accurate input and output data.

Verification Hydrograph - A computed hydrograph of a historical flood which is

generated using the corresponding rainfall data and the developed unit hydrograph as a
means of checking the suitability of the unit hydrograph and/or the runoff model by
comparing with the observed hydrograph or measured values.

Variability - The randomness with respect to the mean in the indicated physical process.

Watercourse - The path which runoff follows during passage from a drainage area.

Water Equivalent - The depth of water (in inches), that results from melting a given depth
of snow.

Watershed - Another term meaning drainage basin.

Water Table - The upper surface of ground water.

Wave Celerity - The velocity of the waveshape in relation to the body of fluid through
which the waves are propagated.

-97-
 Index
hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
National Weather Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
STATSGO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102, 121
accuracy . . . . . . . . . . . . . . 10, 14, 16, 19, 21, 22, 24, 36, 39, 40, 47, 53, 80
active storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19, 24, 80
annual maximum reservoir level . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63, 64
Antecedent Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . 17, 53, 56, 63, 65
Antecedent Moisture Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52, 80
antecedent storm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63, 64, 80
Average Rainfall . . . . . . . . . . . . . . . . . . . . . . . . . . 32, 33, 36, 37, 40, 48, 81
Baseflow . . . . . . . . . . . . . . . . . . . . . . . 31, 34, 35, 41, 48, 80, 114, 115, 117
Basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5, 27, 44, 54, 81
basin characteristics . . . . . . . . . . 2, 3, 18, 20, 35, 37, 43, 46, 49, 51, 62, 86
basin loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55, 56
basin-average rainfall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Basin-Averaged Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
bucket surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61, 81
Bureau of Reclamation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72, 73, 78
Calibration . . . . . . . . . . . . . . . . . . . . . 29, 31, 32, 34, 37, 38, 40, 52, 68, 81
channel flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20, 37, 73
Channel Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38, 39, 60, 67, 68, 114
channel slope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45, 47, 81
channel storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68, 81
Clark Unit Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . 44, 48, 51, 81, 93, 94
Coincident Hydrometeorological Conditions . . . . . . . . . . . . . . . . . . . . . . 65
Cold Season . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32, 33
component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Composite Unit Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81, 91
Continuous Streamflow Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . 16, 82
contributing area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32, 33, 55, 82
cross section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10, 11, 39, 67, 82
daily flow records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16, 82
data acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8, 14, 15, 46
degree-day method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
design flood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1, 12, 26, 82

-98-
Dimensionless Unit Hydrograph . . . . . . . . . . . . . 4, 43, 44, 52, 82, 87, 116
disaggregation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61, 82
Distributed Loss Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53-56, 102, 121
drainage area . . . . . . . . . . . . . . . . . . . . . 5, 28, 33, 38, 46-48, 65, 80, 83, 87
drainage basin . . . . . 1, 3-5, 8, 10, 11, 13, 18, 27, 35, 43, 46, 52, 54, 61, 83
drainage pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81, 83
dynamic effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70, 83
dynamic reservoir flood routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
emergency gate operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12, 84
energy-budget method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66, 84
envelope curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62, 63, 84
exponential loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36, 52
extreme flood . . . . . . . . . . . . . . . . . . . . 10-12, 18, 19, 21, 24-26, 64, 75, 86
extreme flood . . . . . . . . . . . . . . . . . . . . . . . . . . 12, 18, 19, 21, 24, 25
Farm Service Administration (FSA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
FEMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10, 11, 16, 20
flashboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12, 26, 61, 63, 84
flood data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5, 20, 21, 30, 34, 68
flood hydrograph . . . . . 3, 9, 10, 15, 20-23, 29-32, 35-37, 40, 45, 66, 67, 84
frozen soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34, 54
gaged basin . . . . . . . . . . . . . . . . . . . . . 16, 29-31, 46, 85, 114, 116, 121, 122
general storm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63, 85
GIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8, 18, 56, 85
HEC-1 . . . . . . . . . . . . . . . . . . . . 4, 16, 41, 52, 53, 55, 58, 66, 67, 123, 125
historical floods . . . 11-17, 19, 20, 24, 26, 27, 30-34, 36, 40, 43, 47, 53, 70
historical storms . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 23, 29, 31, 46, 53, 57
hydraulics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37, 85
hydrograph . . . . . . . . . . . . . 4, 5, 8, 10, 11, 13-16, 18, 19, 22, 27, 35, 38, 67
hydrologic characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28, 86
Hydrologic data . . . . . . . . . . . . . . . . . . . . 8, 9, 82, 85, 89, 90, 93, 114, 116
Hydrologic Modeling System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4, 31
Hydrologic Soil Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54, 80
hydrologic soil groups . . . . . . . . . . . . . . . . . . . . vi, 54, 58, 82, 86, 121, 122
hydrology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2, 5, 50, 72, 85
Hydrometeorological Report . . . . . . . . . . . . . . . . . . . . . . . . . . 5, 18, 61, 75
hyetograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22, 23, 86, 125

-99-
Infiltration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4, 8, 9, 13, 18, 114
infiltration characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34, 59
infiltration rates . . . . . . . . . . . . . . 4, 8, 9, 18, 21, 28, 32, 34, 46, 47, 59, 60
inflection point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35, 94
inflow design flood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Inflow PMF Hydrograph . . 3, 24, 25, 29, 46, 49, 60, 63, 64, 68-70, 87, 115
initial abstraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21, 62, 65, 87
initial flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65, 68, 87
interflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57, 87, 90
isohyetal map . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17, 22, 24, 28, 33, 61, 87
isohyetal pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5, 22, 62, 87
Kinematic Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
lag time . . . . . . . . . . . . . . . . . . 3, 4, 23, 32, 36, 39, 45, 47-50, 52, 68, 87, 92
land cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55-57, 82, 83, 85-87, 121-123
lapse rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33, 34, 87
layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53, 56, 60, 121, 122, 124, 126
level-pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84, 88
local storms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
local storms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5, 61, 88
loss class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
loss rates . v, 4, 34, 37, 52, 53, 55-57, 60, 62, 68, 69, 82, 115, 117, 121-124
lumped-parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3, 88
Manning equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36, 88
Map unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
maximum normal operating level . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63, 88
minimum infiltration rate . . . . . . . . . . . . . . . . . . . . . . vi, 54, 58, 88, 94, 122
model adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
model verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68, 123, 124
MUID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Muskingum-Cunge method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
National Aerial Photography Program . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
National Climatic Data Center . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
National Dam Inventory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
National Weather Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
NATSGO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 121
Natural Resources Conservation Service . . . . . . . . . . . . . . . . . . . . . . . 8, 75

-100-
NEXRAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17, 89
nonlinear effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20, 89
NWS . . . . . . . . . . . . . . . . . . . . . . . 5, 9, 10, 15, 17-19, 22, 61, 64, 66, 70, 73
operation rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 12, 89
optimization process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37, 38, 89
orographic effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22, 28, 32, 33, 89
outlet works . . . . . . . . . . . . . . . . . . . . 1, 2, 11, 19, 24-26, 47, 63, 89, 90, 93
overbank flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21, 29, 37, 39
peak flow . . . . . 11, 14, 16, 17, 20, 21, 28, 32, 43, 48, 50-52, 57, 82, 89, 92
permeability . . . . . . . . . . . . . . . . . . . . . . . . 13, 53, 55, 56, 83, 89, 121, 122
precipitation data . . . . . . . . . . . . . . . . . . . . . . . . . 9, 17, 22, 29, 81, 89, 114
probable maximum flood . . . . . . . . . . . . . . v, 27, 60, 68, 72, 102, 115, 117
probable maximum precipitation . . . 1, 61, 62, 73, 76, 86, 88, 90, 114, 116
probable maximum storm . . . . . . . . . . . . . . . . . . . . . . . . 56, 67, 74, 90, 116
rain-on-snow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32, 58, 66
rainfall analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
rainfall excess . . . . . . . . . . . 3, 20, 35, 37, 38, 45, 55-57, 83, 86, 87, 90, 124
rainfall sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36, 66, 90
rating curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 12, 16, 19, 21, 90
recession . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22, 34, 35, 80, 85, 90
reconstitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30, 45, 90
regional studies . . . . . . . . . . . . . . . . . . . . . . . . . 31, 43, 44, 46, 50-52, 63, 90
regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22, 48, 49, 81, 90
remote sensing data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 90
reservoir routing . . . . . . . . . . . . . . . . . . . . . . . 12, 16, 19, 25, 29, 70, 84, 91
reservoir starting elevation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69-71
reservoir storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 19, 24, 63, 80
reverse-reservoir routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12, 16
river basin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31, 91, 121
routing . . . . . 1-3, 10-13, 16, 24, 25, 38-40, 60, 63-65, 67, 70, 71, 114, 116
Runoff Curve Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52, 91
saturated hydraulic conductivity . . . . . . . . . . . . . . . . . . 52, 58, 89, 121, 122
sediment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19, 24, 91
sensitivity analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 4, 62, 69, 91, 115, 117
simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi, 4, 5, 24, 33, 34, 41, 124
snow course . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18, 23, 92

-101-
snow cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19, 32, 34, 92
snowmelt . . 5, 15, 17, 18, 20, 23, 24, 28, 29, 33, 34, 58, 65-68, 75, 82, 84,
92, 114, 115, 117, 125
snowpack . . . . . . 9, 15, 18, 19, 23, 28, 33, 59, 65-67, 80, 92, 114, 115, 117
Snyder Unit Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45, 50, 51, 92
Soil Conservation Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8, 73
soil map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 28, 57, 92, 121, 126
soil moisture content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59, 92
soil series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82, 121, 124, 126
soil structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
spatial distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . 27, 31, 37, 61, 62, 87
spillway . . . . . . . . . . . . . . . . . . 10, 12, 13, 19, 24-26, 47, 61, 63, 69, 70, 93
spurious trend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22, 93
SSURGO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 53, 56, 58, 102, 121
STATSGO . . . . . . . . . . . 4, 8, 9, 18, 53, 56, 75, 82, 93, 102, 121, 122, 126
STATSGO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Storage Coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . 35, 36, 48, 78, 92, 93
storm duration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50, 61
storm transposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
streamflow record . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14, 28, 35, 39
Subbasin . . 27, 29-31, 33, 35, 36, 38, 40, 44, 45, 47, 52, 53, 55, 61, 62, 67,
68, 86, 93, 114, 116, 121, 123, 124
subdivision . . . . . . . . . . . . . . . . . . . . . . . . . . . 5, 27, 28, 37, 62, 93, 94, 114
temporal distribution . . . . . . . . . . . . . . 3, 17, 23, 33, 36, 37, 48, 62, 63, 85
Tennessee Valley Authority . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Thiessen polygon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33, 94
time lag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50, 51, 81
time of concentration . . . . . . . . . . . . . . . . 35, 47, 48, 50, 69, 81, 87, 88, 94
travel time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21, 81
uncontrolled spillway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12, 94
ungaged basin . . . . . . . . . . . . . . . 4, 30, 43, 44, 46, 52, 58, 78, 94, 103, 116
uniform loss rate . . . . . . . . . . . . . . . . . . . . . . . . . . . 52, 55, 58, 94, 122, 123
unit duration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45, 50, 94
unit hydrograph . . 3, 4, 14, 15, 17, 19, 20, 23, 27, 29-32, 35-40, 43-46, 48,
50-52, 60, 70, 94, 114, 116
USFS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8, 10, 18

-102-
USGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-10, 15, 16, 18, 36, 67
verification . . . . . . . . . . . . . . . . . . . 29, 31, 32, 40, 53, 57, 58, 114, 122-124
water equivalent . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 18, 19, 23, 65-67, 95
water table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34, 55, 60, 95, 124
watercourse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18, 35, 47, 51, 83, 84, 95
watershed . . 4, 14, 18, 21, 29, 33, 44, 51, 54-56, 72, 81, 95, 114, 116, 121,
125
wave celerity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70, 95

-103-
8-13 Appendices

Appendix A - Determining the PMF for Civil Works Flow Chart

Appendix B - Probable Maximum Flood Study Report Outline

Appendix C -Hec-1 Data-Analysis Techniques of Infiltration Rate Estimate Methods

Appendix D - Distributed Loss Rate Methods Using STATSGO or SSURGO Databases

-104-
 Appendix A

Determining the PMF for Civil Works Flow Charts

As an aid for determining the PMF for gaged and ungaged basins, these flow charts show
the sequence of required decisions and analyses. Chapter and section references are
shown for each flow chart element.

-105-
-106-
-107-
-108-
-109-
-110-
-111-
-112-
-113-
-114-
 Appendix B

Probable Maximum Flood Study Report Outline

The following study report outline should assist the analyst in documenting PMF studies.
The outline parallels the reasoning in Chapter VII and the flow chart, except that some
subject areas are consolidated to avoid repeating information in the written report. When
subject headings are not applicable to the study, an explanation should be provided.

-115-
 PMF STUDY REPORT OUTLINE

GAGED BASINS

I. PROJECT DESCRIPTION

A. Project Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2.1 - 8-2.3, 8-4.5

B. Basin Hydrologic Data . . . . . . . . . . . . . . . . . . . . . . 8-2.1 - 8-2.3, 8-3.5
C. Upstream Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2.2, 8-3.7
D. Field Visit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2.3
E. Previous Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2.1 - 8-2.2, 8-3.1

II. WATERSHED MODEL AND SUBDIVISION

A. Watershed Model Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1.2

B. Subbasin Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5.1
C. Channel Routing Method . . . . . . . . . . . . . . . . . . . . . . . . . 8-6.8.2, 8-9.3

III. HISTORIC FLOOD RECORDS

A. Stream Gages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3.2

B. Historic Floods . . . . . . . . . . . . . . . . . . . . . . . 8-3.2, 8-4.1 - 8-4.2, 8-6.1
C. Precipitation Associated with Historic
Floods . . . . . . . . . . . . . . . . . . . . . . . . . 8-3.3, 8-3.4, 8-4.3, 8-6.2, 8-6.6
D. Snowpack and Snowmelt During Historic Floods . . . . . . . . 8-3.6, 8-6.3

IV. UNIT HYDROGRAPH DEVELOPMENT

A. Approach and Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5.3, 8-6.5

B. Baseflow Separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 8-6.4
C. Preliminary Estimates of Clark Parameters . . . . . . . . . . . . . ..... 8-6.5
D. Estimate of Infiltration During Historic Floods . . . . . . . . . . ..... 8-6.7
E. Subbasin Unit Hydrograph Parameters . . . . . . . . . . . . . . . . ..... 8-6.8

V. UNIT HYDROGRAPH VERIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6.9

VI. PROBABLE MAXIMUM PRECIPITATION

A. Probable Maximum Precipitation Data . . . . . . . . . . . . . . . . 8-3.4, 8-9.1

B. Candidate Storms for PMF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.1

-116-
VII. LOSS RATES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8.1 - 8-8.4

VIII. COINCIDENT HYDROM ETEOROLOGICAL AND HYDROLOGICAL

CONDITIONS FOR THE PROBABLE MAXIMUM FLOOD

A. Reservoir Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.2.1

B. Baseflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.4
C. Snowpack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.2
D. Snowmelt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8.4, 8-9.2.3

IX. PMF HYDROGRAPHS

A. Inflow PMF Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.5

B. Sensitivity Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.6
C. Reservoir Outflow PMF . . . . . . . . . . . . . . . . . . . . . . . . . 8-10.1 - 8-10.3

-117-
 PMF STUDY REPORT OUTLINE
UNGAGED BASINS

I. PROJECT DESCRIPTION

A. Project Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2.1 - 8-2.3, 8-4.5

B. Basin Hydrologic Data . . . . . . . . . . . . . . . . . . . . . . 8-2.1 - 8-2.3, 8-3.5
C. Upstream Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2.2, 8-3.7
D. Field Visit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2.3
E. Previous Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2.1 - 8-2.2, 8-3.1

II. WATERSHED MODEL AND SUBDIVISION

A Watershed Model Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1.2

B. Subbasin Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5.1
C. Channel Routing Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.3

III. HISTORIC FLOOD RECORDS

A. Stream Gages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3.2

B. Historic Floods . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3.2, 8-4.1 - 8-4.2

IV. UNIT HYDROGRAPH DEVELOPMENT

A. Approach and Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5.3, 8-7

B. Existing Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-7.1
C. Regional Analysis (include details as Appendix) . . . . . . . . . . . . . 8-7.2
(1) Gaged Basins Used in Analysis
(2) Cold-Season Considerations
(3) Regional Relationship for Unit Hydrograph Parameters
OR
C. Synthetic Unit Hydrographs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-7.3
OR
C. SCS Dimensionless Unit Hydrograph . . . . . . . . . . . . . . . . . . . . . 8-7.3.3

V. UNIT HYDROGRAPH VERIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . 8-6.9

VI. PROBABLE MAXIMUM STORM

A. Probable M aximum Precipitation Data . . . . . . . . . . . . . . . . 8-3.4, 8-9.1

B. Candidate Storms for PMF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.1

-118-
VII. LOSS RATES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8.1 - 8-8.4

VIII. COINCIDENT HYDROM ETEOROLOGICAL AND HYDROLOGICAL

CONDITIONS
FOR THE PROBABLE MAXIMUM FLOOD

A. Reservoir Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.2.1

B. Baseflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.4
C. Snowpack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.2
D. Snowmelt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8.4, 8-9.2.3

IX. PMF HYDROGRAPHS

A. Inflow PMF Hydrograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.5

B. Sensitivity Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9.6
C. Reservoir Outflow PMF . . . . . . . . . . . . . . . . . . . . . . . . . 8-10.1 - 8-10.3

-119-
 Appendix C

HEC-1 Data-Analysis Techniques of Infiltration Rate Estimate M ethods

Infiltration can vary temporally and spatially in a drainage basin as a very complex physical process.
Selection of data-analysis techniques or measurement techniques should consider these effects. Two
approaches are used to estimate loss rates of subbasins including basin-averaged methods and distributed
methods. Each approach basically should be based on physical soil properties and land covers. The most
practical (i.e. simplest) application is to use a constant loss rate rather than an unsteady rate. The constant
rate associated initial loss might be considered to represent the total loss due to surface factors and volume
infiltrated prior to attaining the soils relatively long-term infiltration rate.

Table VIII-C lists five models applied in the HEC-1 computer program. The use of relatively small,
homogeneous watersheds is recommended to minimize spatial variations of infiltration rates over larger
areas. In general, an upper limit of 25 mi2 for study-watershed drainage areas was suggested to minimize
the effects of lumping infiltration rates.

Parameters of simple empirical infiltration models or models with physically based or measurable
parameters need to be estimated. Parameter estimation techniques are categorized by application to gaged
or ungaged analysis below.

(A) Ungaged Parameter Estimation

Physical characteristics of the watershed may be the only information available for estimating parameters
on a theoretical basis.

(B) Gaged Parameter Estimation

Rainfall-runoff records are used to estimate infiltration model parameters. The basic element of a gaged
estimation is to utilize an optimization algorithm to choose model parameters so that some measure of the
difference between observed and predicted hydrographs is minimized. This approach to parameter
estimation is essentially a regression analysis based on recorded data of a number of events.

-120-
Table VIII.C Comparisons of Infiltration Models
INFILTRATION MODELS SOIL BASIS DOMINANT DISAD-VANTAGE APPLICA-
FACTORS BILITY

Rainfall Excess Models (const. -- -- --

or unif. loss rates)

A. Index Models (1) Four (1) hydrologic soil (1), (2) & (3) (1) ) Hydrologically
Data source for loss rates: hydrologic soil types; (2) & (3) Independent from similar watersheds
(1) Hydrologic soil groups groups soil series and its rainfall intensity and (2) & (3) mainly for
(USDA 1955) (2) & (3) permeability which volume, and cultivation of crops,
(2) NRCS' SSURGO data base Physical soil is affected by (1) Inaccurate basin- for engineering
(3) NRCS' STATSGO data base properties ground cover avaerage rate; (2) design applications
(4) Calibration/regional studies using spatially conditions; (4) Data from 1:250,000 if justified,
detailed soils rainfall excess maps and too crude (4) when design
and land cover soils associations; (3) storm is of similar
(4) None Incomplete database; magnitude to
(4) calibrated calibration storm
parameters are
related to the
intensity of the
calibration storm

B. NRCS Runoff Curve Number Four hydrologic soil- Minimal physical Small, agricultural
(RCN) Model: hydrologic soil cover com-plexes theory (conditions not frost-free
RCN decays as the total groups, land (cover includes considered such as watersheds
volume of accumulated cover & nd land use, near-surface bedrock)
infiltration increases and is treatment, treatment,
reflected as AMC I, II, or III) AMC hydrologic con-
dition), AMC

Empirical Models (unsteady loss -- -- --

rates)

A. Cumulative Loss-Dependent 2 parameters rainfall intensity, (a) At least three sets Engineering design
Loss Rate: related to accumulated losses of storm and flow applications
Exponential Loss Model basin (or soil moisture data required; (b) less
character., one storage), accuracy for
to antec. impervious area extrapolating rates
moisture
deficiency, one
to both

B. Time-Dependent Loss Rate:
Holtan Model- exponential
rate
Parameters for soil moisture Param.. fitting to Agricultural lands
cultivated soils storage, ground vegetative data rather
cover, surface pore than measurable soil
space and volume, characteristics
ultimate infiltration
capacity

Approximate Theory-Based -- -- --
Models (unsteady or constant
loss rates)

Green-Ampt Model (using Physical well-defined No adjustments for Small-scale crop or

Darcy's law) (The loss rate is properties of wetting front, watershed non- range lands
constant after the soil reaches the soil column constant volumetric homogeneity, surface
saturation - most of the Green- water contents, storage, or vegetation
Ampt equation deals with what constant soil-water effects
happens before that time.) suction

-121-
-122-
 Appendix D
Distributed Loss Rate Methods Using STATSGO or SSURGO Databases

Sections 8-8.2 and 8-8.3 discuss using the distributed loss rate method for developing
PMF studies. This appendix discusses in more detail the use of STATSGO or SSURGO
soil databases with the distributed loss rate method.

Digital soil maps are now available from the NRCS in the NATSGO, STATSGO, and
SSURGO series. The NATSGO series, which is designed to provide a regional overview
of soils on a national scale, is the least detailed of the three databases and does not
provide adequate detail for watershed studies. The STATSGO series was designed for
regional planning and river basin studies. STATSGO maps soils to the association level
and provides data on individual soil series within each association. SSURGO, which is
not complete as of 2000, is the most detailed of the three databases and will be a digital
form of county soil survey maps. In the distributed loss method, the STATSGO series or
digital data of similar (or finer) spatial resolution should be used.

Unlike the basin-averaged loss rate method - which uses spatially averaged minimum loss
rates that have been empirically determined in relation to the hydrologic soil groups - the
distributed loss rate method allows the hydrologist to assign loss rates on the basis of
physical properties of each soil unit, modified as necessary for other factors, such as
bedrock, groundwater levels, vegetation, etc., that may affect runoff.

Loss rates for a given soil unit can be derived from a soils database that contains
sufficient information to determine how much area within the subbasin each soil unit
occupies, and approximate areas of overlap between soil units and hydrologically
significant land cover types (such as wetlands and forests). For all but the smallest
basins, the application of the method will only be practical when the database is available
digitally and can be read into a GIS format that is easily superimposed on a digital
subbasin boundary map and land cover map. The discussion that follows will reference
the STATSGO database, but it is also applicable to basins where the SSURGO database
has been completed and digitized.

STATSGO classifies each layer of each soil unit within one of several standard
logarithmic ranges of permeability (saturated hydraulic conductivity). The least
permeable layer should be assumed to control losses for the area occupied by that soil
unit. For application purposes, it is necessary to represent the range of each layer as a
single value. The procedures for ungaged and gaged basins are as follows.

Ungaged (sub) basins. Since no historical data is available to verify the model, the

-123-
minimum value from either the Minimum Infiltration Rates for the Hydrologic Soil
Groups found in Table 8-8.1 or the minimum value of the given range of the permeability
(saturated hydraulic conductivity) of the least permeable layer, as provided in the
STATSGO database, should be used as the loss rate for each soil class. Deviations from
the minimum value are acceptable with adequate, physically based, justification.
Potential sources of information include a review of the geological make-up of the soils,
review of soils information such as county or local soils maps, or actual data obtained
from field investigations. However, the STATSGO or SSURGO data should not be used
to develop a basin-averaged loss rate since the high permeability values for some sandy
soil classes in the databases will raise the basin-averaged loss rate to unrealistic values.
Regional analyses are allowable with adequate support for the transfer of data from gaged
basins with similar hydrological properties. Factors to consider for transferring data
include the following:

C Basin size, slope, soil types (and distribution), soil column depth to
bedrock, land cover, etc., are similar to a gaged basin in the study area
(study basin);

C Adequate stream flow data are available from gages located downstream
of, but in close proximity to the study basin;

C Several large, single peaked storms are available that are centered in the
study basin close to the stream flow gages that provide for adequate
basin coverage. Since some historical events may not be of sufficient
size to prevent nonlinear effects, the historical events used for adjusting
loss rates should be clearly out-of-bank floods or saturated soil
conditions must have existed in a significant part of the basin prior to
the storm.

Gaged (sub) basins. A value within the range of the saturated hydraulic conductivity of
the least permeable layer can be selected as the preliminary loss rate if appropriately
justified. Physically based information, such as a review of the geological make-up of the
soils, review of soils information such as county or local soils maps, or actual data
obtained from field investigations, is necessary to support the selected rate. After
considering all hydrologic influences such as wetlands, open water, etc., the model and
selected loss rate should be verified with rainfall-runoff records for several large flood
events. Since some historical events may not be of sufficient size to prevent nonlinear
effects, the historical events used for adjusting loss rates should be clearly out-of-bank
floods or saturated soil conditions must have existed in a significant part of the basin
prior to the storm. If this verification does not support the selected loss rates, further

-124-
investigations and adjustments are required. The procedures for verifying the model and
adjusting the loss rates are discussed in Section 8-8.3.

The method presented here is designed for use with HEC-1 or a similar basin model.
Because assigned parameters other than USDA recommendations for basin-average
uniform loss rate (Section 8-8.1), are usually less conservative, it is important to verify
the results of a distributed model using large historical events. The assumptions for
assigning initial distributed loss parameters should be considered base line assumptions.
Deviation from the base line assumption is not justified when there is an inadequate rain
gage distribution within the study basin, or the proportion and/or location of the gaged
subarea relative to the entire basin size is small. The loss rate may require adjustment
based on model verification with historical storm hydrographs, geologic considerations,
groundwater elevation, land cover, or other parameters found to affect runoff. Testing of
the initial runoff model with historical events and careful evaluation of hydrologic factors
other than the baseline assumptions are necessary to ensure that the model adequately
represents runoff processes for extreme storms.

As additional information becomes available, such as flood events larger than historical
events, the model should be re-run to determine if it adequately predicts the new flood
event. If the model does not adequately predict the new flood event, adjustments should
be made to the loss rates and/or the unitgraph parameters. When drainage basins do not
contain adequate rainfall/runoff data, the installation of rain gages and flow gages should
be considered.

When using HEC-1 or a similar "lumped" basin model, the following steps should be
taken to develop the distributed runoff model:

A. Assigning Loss Rates

1. Digitally overlay the basin or subbasin boundary map, the soil unit/association
map (such as STATSGO), and the land cover map.

2. For each subbasin, determine the percentage of area covered by each soil unit.
STATSGO maps soils at the association level (not the unit level) and gives the
percentage of area represented by each soil unit. When using STATSGO, it is
necessary to assume that this typical distribution applies to the soil association as
it occurs in the subbasin.

3. For each subbasin, determine what percentage of each soil unit is occupied by
(a) open water and wetlands; (b) forests (if frozen soils are a consideration); and
(c) other land uses such as urban, agricultural, and rangeland/grassland areas.

-125-
 4. From the information gathered in steps 1-3, classify all of the basin or subbasin
area into loss rate groups, using the following assumptions:

a. As a baseline assumption, the infiltration capacity of the least

permeable layer in the soil profile should be assumed to control the
soil's loss rate. Each soil unit is assigned a uniform loss rate based upon
the review of the available soils data. Sources of data include soils and
geologic maps of the watershed, other technical reports, or field
investigations. These sources should be reviewed to support the
selection of the loss rate if it is larger than the minimum value of the
range.

b. For any area occupied by open water or wetlands, the uniform rate
should be changed to zero, regardless of underlying soil type. In basins
with high water tables or shallow impermeable layers, the loss rate may
also need to be set to zero - deviations from a zero loss rate must be
justified.

c. If a cool season analysis is performed, the loss rates should be further

adjusted in accordance with Section 8-8.3.

5. For each subbasin, tabulate the percent of area with each warm-season and
cool-season (if applicable) loss rate.

6. A STATSGO soil series map with the soil association designations should be
included in the study.

B. Determination of Rainfall Excess and Model Verification

1. Calculate precipitation to be modeled (historical rainfall for model verification,

or the PMP for PMF simulation) in hourly increments for each subbasin.

2. For each hour of the storm being modeled, calculate the difference between the
rainfall increment and the hourly loss rate for each of the loss classes defined in
(A)(4), above. If positive, the difference is the hourly rate of rainfall excess from
that loss class. If negative, there is no rainfall excess for that loss class and time
increment.

3. For each hour, multiply the rainfall excess from each loss class in each
subbasin by the area of the portion of the subbasin area occupied by that loss class.
Sum the rainfall excess over all loss classes in the subbasin for each subbasin.
This will produce the total rainfall excess for each subbasin. A spreadsheet

-126-
 program can be set up to perform this function, but care must be taken to ensure
the timing is correct.

4. Use the resulting subbasin total rainfall excess hyetographs as rainfall input in
the HEC-1 or other watershed model, setting losses in the watershed model equal
to zero. Run the model to produce the runoff hydrograph from each subbasin.

5. Verify the model and adjust loss rates as appropriate in accordance with Section
8-8.3.

6. Provide a table showing the precipitation, losses, excess, and snowmelt for
each time increment for each subbasin, similar to the HEC-1 program output.

7. For each loss rate category within each subbasin, provide a summary table
showing the percent area of the subbasin, the baseline loss rate values, and the loss
rate values after adjustments have been made.

8. A sensitivity analysis should be performed to compare the selected loss rates

and loss rates that are more conservative. At a minimum, an inflow hydrograph
developed by using justifiable loss rates and a hydrograph developed using a loss
rate equal to the minimum value of the least permeable layer of the STATSGO
database should be plotted. Intermediate values between the selected loss rate and
the minimum value may also be plotted for comparison purposes.

Note: Steps B2 through B4 can be directly modeled in HEC-1 or HEC-HM S by taking

advantage of the linear assumption of unit hydrograph theory. In the above steps, the
rainfall excess for each soil loss rate class are combined for each time step to form the
input rainfall excess hyetograph, which is then used in HEC-1 with the loss rate set to
zero to compute the runoff hydrograph. This hydrograph can also be developed by first
computing the rainfall excess hyetographs for each loss rate class, and then combining
them. All of this can be done using HEC-1 by subdividing a subbasin into pseudo-
subbasins corresponding to each loss rate class. For each pseudo-subbasin, the basin area
is set equal to the area of the portion of the subbasin area occupied by that loss rate class
similar to Step B3 above, and the uniform loss rate parameter (LU) is set equal to the loss
rate for that loss rate class. The subbasin areas for the pseudo-subbasins should add up to
the actual subbasin area in order to ensure that the proper volume of runoff is computed.
The unitgraph parameters and the precipitation values for the pseudo-subbasins are
equivalent to the values developed for that subbasin in order to preserve the proper
timing, shape, and volume of the subbasin hydrograph. The baseflow can be accounted
for by including it in one of the pseudo-subbasins. The outflow hydrographs from the
pseudo-subbasins can then be combined using HEC-1 to produce the outflow hydrograph
for that subbasin. The advantage of this method is that everything can be done within

-127-
HEC-1, and it is much easier to observe how each loss rate class contributes to the total
runoff for each subbasin. Consequently, sensitivity analyses and verification/model
adjustments required by Step B5 above are also easier to perform, and documentation of
the analysis required by Step B6 above can be done within HEC-1.

Major Terms of STATSGO Data Base

Several terms commonly used to describe soil types or components are defined as follows
and shown in Fig. 1:

1. Soil Profile

Soil Series - A specific soil type, for example: Massena, Sun, Mosherville.
Component - A specific soil series phase (i.e. soil properties).
Soil Association - A collection of soil series in a soil column. Each soil association in
STATSGO can contain up to 21 different soil series (i.e., components).
Soil Layer - A layer (essentially horizontal) of a soil series which defines the soil column.
Up to six different layers may be identified for a series. Each layer within a soil column
is identified with a layer number ("LAYERNUM"), starting from the top of the soil and
counting downward. Both the maximum and minimum k values for a soil layer are
provided.
Limiting Layer - The layer with the smallest minimum permeabilty of all layers of a soil
association.

2. Map Units

Soil Unit - The unit is usually represented by a single soil series. A soil map shows the
soil units over a drainage basin.
MUID - The mapping unit identifier which represents a particular soil association. The
soil associations are identified by "MUID" numbers, for examples, NY013, NY033, and
NYW (water body) in New York State.
MUID Sequence Number - The sequence number which is associated with a MUID for
different components and their percentages of the soil association.

-128-
-129-
 CHAPTER IX

INSTRUMENTATION AND MONITORING

 Chapter IX

Instrumentation and Monitoring

Contents
Title Page

9-1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1

9-1.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1

9-1.2 Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-1

9-2 Philosophy of Instrumentation and Monitoring . . . . . . . . . . . . . . . . . . . . . 9-2

9-2.1 Visual Observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-3

9-2.2 Purpose of Minimum Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . 9-4

9-3 Types of Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-4

9-3.1 Water Level and Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-4

9-3.1.1 Engineering Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-5
9-3.1.2 Water Level Gages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-8
9-3.1.3 Observation Wells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-9
9-3.1.4 Open Standpipe Piezometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-9
9-3.1.5 Closed Standpipe Piezometers . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-10
9-3.1.6 Twin-tube Hydraulic Piezometers . . . . . . . . . . . . . . . . . . . . . . . . 9-10
9-3.1.7 Pneumatic Piezometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-10
9-3.1.8 Vibrating Wire Piezometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-10
9-3.1.9 Bonded Resistance Strain Gage Piezometers . . . . . . . . . . . . . . . . 9-11
9-3.2 Seepage and Leakage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-11
9-3.2.1 Engineering Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-11
9-3.2.2 Weirs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-12
9-3.2.3 Parshall Flumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-13
9-3.2.4 Calibrated Containers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-13
9-3.3 Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-13
9-3.3.1 Engineering Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-14
9-3.3.2 Level Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-15
9-3.3.3 Alignment Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-15
9-3.3.4 Triangulation and Trilateration . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-16
9-3.3.5 Internal Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-17
9-3.3.6 Crack and Joint Measuring Devices . . . . . . . . . . . . . . . . . . . . . . . 9-18

9-i
 9-0 Contents
(Cont.)
Title Page

9-3.4 Stress and Strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-18

9-3.5 Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-19
9-3.6 Seismic Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-19
9-3.7 Loads in Post-Tensioned Anchors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-20

9-4 Minimum Instrumentation Recommendations . . . . . . . . . . . . . . . . . . . . . 9-20

9-4.1 Visual Observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-24

9-4.2 Existing Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-24
9-4.2.1 Low-hazard Potential Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-26
9-4.2.2 Significant and High-hazard Potential Dams . . . . . . . . . . . . . . . . 9-26
9-4.2.2.1 Water Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-26
9-4.2.2.2 Seepage and Leakage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-26
9-4.2.2.3 Pore/Uplift Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-26
9-4.2.2.4 Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-28
9-4.2.2.5 Seismic Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-30
9-4.2.2.6 Loads in Post-Tensioned Anchors . . . . . . . . . . . . . . . . . . . . 9-30
9-4.3 Proposed Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-30
9-4.3.1 Low-hazard Potential Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-31
9-4.3.2 Significant and High-hazard Potential Dams . . . . . . . . . . . . . . . . 9-31
9-4.3.2.1 Pore/Uplift Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-31
9-4.3.2.2 Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-33
9-4.3.2.3 Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-34
9-4.3.2.4 Seismic Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-34
9-4.3.2.5 Loads in Post-Tensioned Anchors . . . . . . . . . . . . . . . . . . . . 9-34
9-4.4 Additional Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-35

9-5 Instrumentation System Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-35

9-5.1 Project Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-35

9-5.2 Purpose of Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-36
9-5.3 Types of Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-36
9-5.4 Types of Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-36
9-5.5 Location and Number of Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . 9-37
9-5.6 Procurement and Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-38
9-5.7 Monitoring Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-39
9-5.8 Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-41

9-ii
 9-0 Contents
(Cont.)
Title Page

9-5.9 Maintenance and Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-41

9-6 Monitoring Schedules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-41

9-7 Data Processing and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-44

9-7.1 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-44

9-7.2 Data Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-45
9-7.3 Data Presentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-46
9-7.4 Data Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-46
9-7.5 Dam Performance Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-48
9-7.6 Adequacy of Instrumentation and Monitoring . . . . . . . . . . . . . . . . . . . 9-49

9-8 Automated Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-50

9-9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-51

9-10 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-56

Appendix 9-A Examples of Minimum Instrumentation

Appendix 9-B Sample Data Presentation

List of Tables

Table 9-3.1.1 Advantages and Limitations of Common Water Level and

Pressure Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-6
Table 9-4a Minimum Recommended Instrumentation for Existing
Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-22
Table 9-4b Minimum Recommended Instrumentation for Proposed
Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-23
Table 9-4c Typical Instrumentation and Monitoring Used in
Evaluating Causes of Common Problems/Concerns . . . . . . . . . . . 9-25
Table 9-6 Typical Monitoring Schedule for Significant and High-
hazard Potential Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-43

9-iii
List of Figures

Figure A-1 Example of Minimum Instrumentation for an Existing

Embankment Dam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-A-5
Figure A-2 Example of Minimum Instrumentation for a Proposed
Embankment Dam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-A-6
Figure A-3 Example of Minimum Instrumentation for an Existing
Concrete Gravity Dam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-A-7
Figure A-4 Example of Minimum Instrumentation for a Proposed
RCC Gravity Dam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-A-8

Figure B-1 Instrument Location Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-B-3

Figure B-2 Instrument Location Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-B-4
Figure B-3 Piezometer Data Collection Form . . . . . . . . . . . . . . . . . . . . . . . . 9-B-5
Figure B-4 Seepage Data Collection Form . . . . . . . . . . . . . . . . . . . . . . . . . . 9-B-6
Figure B-5 Data Reduction Spreadsheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-B-7
Figure B-6 Embankment Piezometer Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-B-8
Figure B-7 Foundation Piezometer Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-B-9
Figure B-8 Observation Well Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-B-10
Figure B-9 Seepage Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-B-11

9-iv
9-1 Introduction

9-1.1 General

The majority of FERC licensed dams have been evaluated and, where necessary, have
been or are being modified to meet current criteria. The next logical step is to concentrate
on dam safety monitoring programs, which consist of collecting data from visual
observations and instrumentation and evaluating the data with respect to dam
performance and safety. Visual observation consists of the thorough inspection of
conditions at the dam and appurtenant structures, noting any abnormal or unusual
conditions that could jeopardize the safety of the dam. Instrumentation consists of the
various electrical and mechanical instruments or systems used to measure pressure, flow,
movement, stress, strain, and temperature. Monitoring is the collection, reduction,
presentation, and evaluation of the instrumentation data. Instrumentation and monitoring
are tools that must be used with a vigilant inspection program to continually evaluate the
safety of dams.

9-1.2 Purpose and Scope

The purpose of this chapter is to provide staff engineers with recommended guidelines to
use in reviewing and evaluating the adequacy of instrumentation and monitoring
programs in license applications, supporting design reports, FERC Part-12 consultants
reports, or programs recommended by the regional office inspector following operation or
construction inspections. There are no simple rules for determining the appropriate level
of instrumentation and monitoring because it depends on the size and hazard potential
classification of the dam, the complexity of the dam and foundation, known problems and
concerns, and the degree of conservatism in the design criteria. Therefore, evaluation of
instrumentation and monitoring programs requires staff to apply engineering judgement
and common sense.

A major change in the extent of existing instrumentation programs at most projects is not
anticipated, though some dams will require additional instrumentation. Many monitoring
programs may need to be improved. Emphasis should be placed on timely collection and
evaluation of the instrumentation data. This chapter discusses the philosophy of
instrumentation and monitoring, commonly used instruments, minimum instrumentation
guidelines, instrumentation system design, monitoring schedule guidelines, data
processing and evaluation, automated data acquisition, and examples of acceptable
instrumentation and data presentation.

9-1
9-2 Philosophy of Instrumentation and Monitoring

The purpose of instrumentation and monitoring is to maintain and improve dam safety by
providing information to 1) evaluate whether a dam is performing as expected and 2)
warn of changes that could endanger the safety of a dam.

The causes of dam failures and incidents have been catalogued (ASCE 1975 and 1988,
Jansen 1980, National Research Council 1983, ICOLD 1992). The common causes of
concrete dam failures and incidents are:

• overtopping from inadequate spillway capacity or spillway blockage resulting in

erosion of the foundation at the toe of the dam or washout of an abutment or
adjacent embankment structure;

• foundation leakage and piping in pervious strata, soluble lenses, and rock
discontinuities; and

• sliding along weak discontinuities in foundations.

The principal causes of embankment dam failures and incidents are:

• overtopping from inadequate spillway capacity, spillway blockage, or excessive

settlement resulting in erosion of the embankment;

• erosion of embankments from failure of spillways, failure or deformation of outlet

conduits causing leakage and piping, and failure of riprap;

• embankment leakage and piping along outlet conduits, abutment interfaces,

contacts with concrete structures, or concentrated piping in the embankment itself;

• foundation leakage and piping in pervious strata, soluble lenses, and rock
discontinuities;

• sliding of embankment slopes due to overly steep slopes, seepage forces, rapid
drawdown, or rainfall;

• sliding along clay seams in foundations;

• cracking due to differential settlements; and

• liquefaction.

9-2
Instrumentation and monitoring, combined with vigilant visual observation, can provide
early warning of many conditions that could contribute to dam failures and incidents. For
example, settlement of an embankment crest may increase the likelihood of overtopping;
increased seepage or turbidity could indicate piping; settlement of an embankment crest
or bulging of embankment slopes could indicate sliding or deformation; inelastic
movement of concrete structures could indicate sliding or alkali-aggregate reaction.
Conversely, lack of normally expected natural phenomena may also indicate potential
problems. For example, lack of seepage in a drainage system could indicate that seepage
is occurring at a location where it was not expected or contemplated by the designer.

Instrumentation and monitoring must be carefully planned and executed to meet defined
objectives. Every instrument in a dam should have a specific purpose. If it does not have
a specific purpose, it should not be installed or it should be abandoned. Instrumentation
for long-term monitoring should be rugged and easy to maintain and should be capable of
being verified or calibrated. Instrumentation typically provides data to:

• characterize site conditions before construction;

• verify design and analysis assumptions;

• evaluate behavior during construction, first filling, and operation of the structure;

• evaluate performance of specific design features;

• observe performance of known geological and structural anomalies; and

• evaluate performance with respect to potential site-specific failure modes.

Installation of instruments or accumulation of instrument data by itself does not improve

dam safety or protect the public. Instruments must be carefully selected, located, and
installed. Data must be conscientiously collected, meticulously reduced, tabulated, and
plotted, and must be judiciously evaluated with respect to the safety of the dam in a
timely manner. A poorly planned program will produce unnecessary data that the dam
owner will waste time and money collecting and interpreting, often resulting in
disillusionment and abandonment of the program.

9-2.1 Visual Observation

Visual observation of all structures should be made in conjunction with instrumentation

monitoring to adequately assess the safety of a dam. Visual observation can readily

9-3
detect indications of poor performance such as offsets, misalignment, bulges, depressions,
seepage, leakage, and cracking. More importantly, visual observation can detect
variations or spatial patterns of these features.

Most visual observation provides qualitative rather than quantitative information, while
instruments provide detailed quantitative information. Visual observation and
instrumentation data are natural complements and when used together they provide the
primary means for engineers to evaluate the safety of a dam.

9-2.2 Purpose of Minimum Instrumentation

Though only a small percentage of dams develop problems, it is impossible to predict

those that will develop problems because of the highly indeterminate nature of the
structures and the infinite number of possible variations in conditions that could affect the
safety of a dam or appurtenant structures. Therefore, it is prudent that any dam that may
affect the public safety has basic instrumentation to monitor vital signs.

The minimum recommended instrumentation is limited to that which clearly provides

useful information for evaluating dam safety and is also readily installed and monitored.
In these guidelines, minimum instrumentation varies from visual observation of low-
hazard potential dams, to instruments for the measurement of pore pressures, uplift
pressures, surface movement, internal movement, and foundation deformation on
proposed, large, high-hazard potential structures. Minimum instrumentation should be
located where it will provide data that are representative of the entire structure.

9-3 Types of Instrumentation

Common types of instruments are summarized in this section. For each type of
measurement, basic engineering concepts and specific types of instruments are discussed.
Emphasis is placed on types of measurements recommended by these guidelines. For
other types of instrumentation, only a general discussion and a list of references is
provided. More detailed information for all types of instrumentation is available in the
literature (Dunnicliff 1981 and 1988, MESA 1973, Sherard 1981, USACE 1971, 1976,
and 1987c, and USBR 1976, 1977, 1987a, 1987b, and 1990).

9-3.1 Water Level and Pressure

Water level is commonly measured with staff gages, float-type water level gages, and
ultrasonic sensors. Water pressure is commonly measured with bubblers, observation
wells, and several types of piezometers, that are discussed below. Other types of water
level and water pressure measuring devices may be appropriate in special circumstances.

9-4
The USACE (1971 and 1987c) and the USBR (1987a, 1987b, and 1990) provide a more
detailed discussion of water level measuring devices. Advantages and limitations of
water level and water pressure instruments are listed in Table 9-3.1.1.

9-3.1.1 Engineering Concepts

Water pressure is a general term that includes pressure within a reservoir or other body of
water, pore pressure, and uplift pressure. Water pressure within soils and within concrete
is commonly referred to as pore pressure. Water pressure acting upward on the base of
concrete dams is commonly known as uplift pressure.

Water level and water pressure are directly related by the depth below the water surface
or phreatic surface. Thus, measurements or water pressure can be readily converted to
water level and vice-versa.

Water pressure usually varies from headwater level on the upstream side of a dam to
tailwater level, ground water level, or atmospheric pressure on the downstream side of a
dam. The headwater, tailwater, and varying pressure across the dam produce forces on a
dam that must be properly accounted for in stability analyses.

9-5
 TABLE 9-3.1.1
ADVANTAGES AND LIMITATIONS OF COMMON WATER LEVEL AND PRESSURE INSTRUMENTS

TYPE ADVANTAGES LIMITATIONS

Staff Gage Simple device, inexpensive, reliable. Cannot be automated.

Float-Type Simple device, inexpensive, reliable. Requires readout device. Sensor must be in water.
Water Level Easily automated. Must be protected from ice.
Gage
Ultrasonic Simple device, inexpensive, reliable. Requires readout device. Must be corrected for air
Water Level Sensor does not touch water. Easily temperature. Debris, foam, and ice can cause false
Sensor automated. readings.
Bubbler Simple device, inexpensive, reliable. Requires readout device. Sensor must be submerged in
Easily automated. water.
Observation Simple device, inexpensive. Easily Applicable only in uniform materials, not reliable for
Well automated. stratified materials. Long lag time in impervious soils.

Open Simple device, inexpensive, reliable.
Standpipe Simple to monitor and maintain. Standard
Piezometer against which all other piezometers are
measured. Can be subjected to rising or
falling head tests to confirm function.
Easily automated.
Long lag time in impervious soils. Potential freezing
problems if water near surface. Porous tips can clog
due to repeated inflow and outflow. Not appropriate for
artesian conditions where phreatic surface extends
significantly above top of pipe. Interferes with material
placement and compaction during construction. Can
be damaged by consolidation of soil around standpipe.

Closed Same as for open standpipe piezometers. Same as open standpipe piezometer but appropriate for
Standpipe artesian conditions.
Piezometer

Twin-tube Simple device, moderately expensive,
Hydraulic reliable, long experience record. Short lag
Piezometer time. Minimal interference with
construction operations.
Cannot be installed in a borehole, therefore, generally
not appropriate for retrofitting. Readout location must
be protected from freezing. Moderately complex
monitoring and maintenance. Periodic de-airing
required. Elevation of tubing and of readout must be
less than 10 to 15 feet above piezometric elevation.
Can be automated, but moderately complex.

Pneumatic Moderately simple transducer, moderately
Piezometer expensive, reliable, fairly long experience
record. Very short lag time. Elevation of
readout independent of elevation of tips
and piezometric levels. No freezing
problems.
Moderately complex monitoring and maintenance. Dry
air and readout device required. Can be automated, but
not over long distances. Sensitive to barometric
pressure. Automation is complex. Moderately
expensive readout.

Vibrating Moderately complex transducer. Simple to Lightning protection required. Expensive transducer
Wire monitor. Very short lag time. Elevation of and readout. Sensitive to temperature and barometric
Piezometer readout independent of elevation of tips pressure changes. Risk of zero drift, but some models
and piezometric levels. No freezing available with in-situ calibration check.
problems. Frequency output signal permits
transmission over long distances. Easily
automated.

Bonded Moderately complex device, expensive. Lightning protection required. Subject to zero-drift,
Resistance Simple to monitor. Very short lag time. therefore, not recommended for long-term monitoring.
Strain Gage Elevation of readout independent of Expensive transducer and readout. Voltage or current
(Electronic) elevation of tips and piezometric levels. output signal sensitive to cable length, splices,
Piezometer No freezing problems. Easily automated. moisture, etc.

9-6
The primary factors influencing the distribution of water pressures in soil are the
permeability of the soil, the ratio of horizontal to vertical permeability, and the variation
of permeability within different zones and strata. The primary factors influencing the
distribution of water pressures in rock are the joint permeability and the variation of the
permeability due to the variation of the orientation, spacing, persistence, interconnection,
and aperture of the joints. Where impervious strata exist in soil or rock, different
pressures may occur in adjacent strata. Water pressure distribution is also affected by
drains, abutment water tables, strata variations, and occasionally grout curtains. Rainfall
and regional water levels may change local water levels, which in turn may affect water
pressure distribution. All these aspects must be properly understood and accounted for
when selecting and locating piezometers.

Relatively high excess pore water pressures may develop in impervious zones and
compressible foundation strata during construction of embankment dams as the height of
the dam increases. The inability of the dam or foundation to maintain effective strength
during construction may lead to deformation or, in extreme cases, slope or bearing
capacity failures. Consolidation testing and analyses, and pore pressure measurements
during construction provide guidance for regulating the rate of fill placement and/or
moisture control in the fill during construction to prevent instability. These pressures
change to steady-state seepage pressures with time, depending on the permeability and
length of drainage paths of the system.

The location of the phreatic surface for steady state seepage conditions in embankment
dams is commonly established by theoretical analyses, and the variation of pressure
beneath the phreatic surface is estimated by flow nets or is assumed to vary
hydrostatically. Alternatively, pressures are estimated by finite element or finite
difference models. Steady state seepage conditions may take years to develop.

Uplift pressure beneath concrete structures is generally assumed to vary linearly from
headwater to tailwater or downstream ground surface. If foundation drains exist and are
adequately maintained, the uplift pressure is usually reduced at the line of drains in
accordance with the effectiveness of the drainage system. The linear pressure distribution
can be affected by the factors influencing the distribution of water pressures in soil and
rock that are discussed above. Common uplift pressure assumptions are illustrated in
Chapter III of these guidelines.

Seasonal water pressure variations can occur as a result of seasonal reservoir level and
temperature variations. Concrete dams and foundations deform slightly to adjust to these
changing loads. In some cases, the deformations are sufficient to alter the aperture and
permeability of foundation rock joints, which changes the pressure distribution. In a
closed, perfectly rigid hydraulic system, changes in water pressure are transmitted, nearly

9-7
instantaneously, by pressure waves. Piezometers are not perfectly rigid, or closed.
Therefore, some water must flow for a pressure change to be measured. The time
required for the flow to occur is known as lag time. Lag time is influenced by the degree
of saturation, the permeability of the materials surrounding the piezometer, the design of
the instrument, and the magnitude of change in pressure. Open standpipe piezometers
require a relatively large volume of water to fill the standpipe and, in low permeability
soils, lag time can range up to several months. Pneumatic and diaphragm type
piezometers installed in sealed and saturated zones require negligible flow, and lag time
for these types of piezometers is generally short. If the sensor is not sealed in a saturated
zone, the lag time is controlled by the filter pack or material surrounding the piezometer.
Lag time is usually only significant for piezometers installed in impervious materials.

Below the phreatic surface, soils are usually assumed to be saturated. Above the phreatic
surface, soils contain both gas and water within the pore spaces. In partially saturated
soils, piezometers measure pore air pressure rather than pore water pressure, unless high
air entry porous tips are used. In cohesionless materials, the difference between pore air
pressure and pore water pressure is minimal. In fine grained cohesive materials with high
capillary pressure, pore air is always greater than pore water pressure. In some instances
the difference can be significant with respect to evaluating the stability of a dam (Sherard
1981).

Piezometer tubing and cables should be installed to avoid development of seepage paths
along them, or through them as they deteriorate. Special attention must be paid to sealing
tubing and cables where they cross zones of an embankment dam. Adequate filters must
be used around tubing located outside of the core and where tubing exits from the dam to
prevent piping along the tubing or through damaged or deteriorated tubing.

9-3.1.2 Water Level Gages

Staff gages are the simplest method for measuring reservoir and tailwater levels. Staff
gages are reliable and durable. For automated monitoring, a float and recorder, ultrasonic
sensor, bubbler, or one of the other instruments discussed below is necessary.

Water level gages can be used to measure flow in rivers (e.g. minimum instream flow),
when the relationship between river flow and river stage is known. Stream bed erosion or
sedimentation can change the calibration and cause inaccurate measurements. Water
level gages used for flow measurements in channels with moveable beds should be
periodically re-calibrated.

9-8
9-3.1.3 Observation Wells

Observation wells are usually vertical pipes with a slotted section at the bottom or a tube
with a porous tip at the bottom. They are typically installed in boreholes with a seal at
the surface to prevent surface water from entering the borehole. The depth to the water
level is measured by lowering an electronic probe or weighted tape into the pipe.

Observation wells are appropriate only in a uniform, pervious material. In a stratified

material, observation wells create a hydraulic connection between strata. As a result, the
water level in the well is an ambiguous combination of the water pressure and
permeability in all strata intersected by the borehole. Observation well data may lead to
erroneous conclusions regarding actual water pressures within the dam and foundation.

9-3.1.4 Open Standpipe Piezometers

Open standpipe piezometers are observation wells with subsurface seals that isolate the
strata to be measured. Open standpipe piezometers are also known as Casagrande-type
piezometers and, in concrete dams, as pore pressure cells. The seals are usually made of
bentonite clay or cement grout and care must be taken during installation to develop a
good seal. Riser pipe joints should be watertight to prevent leakage into or out of the
pipe, which could change the water level in the pipe. The top of the standpipe should be
vented and the inside diameter should be greater than about 8 mm (0.3 inch) to be self de-
airing.

A common version of the open standpipe piezometer is a wellpoint, which is a

prefabricated screened section and riser pipe that is pushed into place. If the screened
section is not adequately sealed, it will act like an observation well rather than a
piezometer. Dunnicliff (1988) discusses methods of sealing well points.

The sensing zone (screened length or porous tip) of observation wells and open standpipe
piezometers is susceptible to clogging, which can increase lag time or result in failure of
the instrument. This susceptibility can be diminished by a properly designed filter pack
that meets filter criteria with the surrounding soil and properly sized perforations that are
compatible with the filter pack.

Open standpipe piezometers are the standard against which all other piezometers are
judged. They are simple, reliable, inexpensive, and easy to monitor.

9-9
9-3.1.5 Closed Standpipe Piezometers

Closed standpipe piezometers are identical to open standpipe piezometers, except that the
water level being measured is above the top of the standpipe (artesian condition) and the
pressure is measured with a pressure gage (or pneumatic, or vibrating wire piezometer)
fitted to the top of the pipe. In concrete dams they are also known as pore pressure cells.
Closed standpipe piezometers installed in concrete dams during construction usually have
riser pipes that are not vertical, but rather routed to a gallery for ease of monitoring.
Provisions for venting gas trapped inside of the riser pipe are often made, but are not
required on most common sizes of riser pipes.

9-3.1.6 Twin-tube Hydraulic Piezometers

Twin-tube hydraulic piezometers are similar in principal to closed standpipe piezometers.

They consist of a porous filter element connected to two flexible tubes. The tubes are
extended more or less horizontally in trenches through the fill or foundation to a readout
point. Two tubes are used to allow the system to be flushed to remove trapped air.
Water pressure is calculated using the average pressure head of the gages on each tube.

9-3.1.7 Pneumatic Piezometers

Pneumatic piezometers consist of a porous filter connected to two tubes which have a
flexible diaphragm between. The diaphragm is held closed by the external water
pressure. The end of one of the tubes is attached to a dry air supply and a pressure gage.
Air pressure is applied until it exceeds the external water pressure acting on the
diaphragm, which deflects the diaphragm and allows the air to vent through the other
tube. The air supply is shut off, and the external water pressure and internal pressure
equalize allowing the diaphragm to close. The residual internal air pressure is taken as
the external water pressure. Alternatively, the water pressure can be taken as the air
pressure required to maintain a constant flow through the tubes. Some constant flow
types use a third tube connected to a pressure gage to measure pressure at the diaphragm
rather than at the inlet to reduce potential errors and eliminate the need for individual
calibration curves.

9-3.1.8 Vibrating Wire Piezometers

Vibrating wire piezometers consist of a porous stone connected to a sealed metal chamber
with a diaphragm adjacent to the stone. Inside the chamber, a wire is stretched between
the diaphragm and a fixed point at the other end of the chamber. The chamber is
connected to an electronic readout device. Water pressure deflects the diaphragm, which

9-10
changes the tension and resonant frequency of the wire. Pressure is measured by
electronically vibrating the wire, measuring the frequency of vibration, and relating
frequency to water pressure using calibration data. Modern readouts perform the
calibration automatically.

9-3.1.9 Bonded Resistance Strain Gage Piezometers

Bonded resistance strain gage piezometers (a.k.a. electronic piezometers) consist of a

porous stone connected to a sealed metal chamber with a diaphragm adjacent to the stone,
similar to vibrating wire piezometers. Inside the chamber, a strain gage is bonded to the
diaphragm. Wires extend from the chamber to an electronic readout device. Water
pressure deflects the diaphragm and the magnitude of the deflection is measured by the
strain gage. Water pressure is determined by relating strain gage output to water pressure
using calibration curves. These piezometers are subject to zero drift, and therefore are
not appropriate for long-term monitoring.

9-3.2 Seepage and Leakage

Seepage is defined as interstitial movement of water through a dam, the foundation, or the
abutments. It is differentiated from leakage, which is flow of water through holes or
cracks. Seepage and leakage are commonly measured with weirs, Parshall flumes, and
calibrated containers. Other types of flow measuring devices such as flow meters may be
appropriate in special circumstances. Geophysical surveys can be used to determine flow
direction. USBR references (1987a, 1987b, and 1990) provide a more detailed discussion
of seepage and leakage measuring devices.

9-3.2.1 Engineering Concepts

The difference in water levels between the upstream and downstream sides of a dam
causes seepage and leakage. The primary factors influencing the amount of seepage and
leakage are the same as those influencing pressure distribution discussed in section 9-
3.2.1. The amount of seepage or leakage is directly proportional to permeability and
pressure. It is possible to have large flow with high pressure, large flow with low
pressure, low flow with high pressure, or low flow with low pressure.

Most of the factors that influence the amount of seepage or leakage do not change during
the life of a project. Usually the main variable is the reservoir level, and typically
seepage and leakage volume are directly related to the reservoir level. Any change in
seepage or leakage volume not related to reservoir level must be evaluated immediately.
Significant or rapid changes in seepage or leakage related to the reservoir level should
also be investigated. An increase in seepage or leakage may be an indication of piping.

9-11
A decrease in seepage or leakage may indicate clogged drains. A decrease in seepage
may also indicate that seepage is increasing at a location other than that being measured,
which could lead to piping. Cloudy or turbid seepage may indicate piping. New seeps or
leaks may also be indications of developing problems.

Another variable that affects the amount of seepage or leakage is the development of the
steady-state phreatic surface in a newly constructed project. The steady-state phreatic
surface can take years, during which, a gradual increase in seepage or leakage may occur.

For dams on soluble rock foundations (e.g. gypsum or halite), seepage may increase with
time due to dissolution of the rock. In these cases a slow steady increase in seepage may
indicate developing problems.

Water quality measurements can provide data to evaluate the dissolution of the
foundation rock, the source of seepage, or piping. Common water quality measurements
include field measurements of Ph, temperature, and conductivity, and laboratory
measurements of total dissolved solids, total suspended solids, and a variety of minerals
(e.g. sodium, potassium, carbonate, bicarbonate, sulfate, and chloride). Standard test
methods are given by the American Society for Testing and Materials (ASTM), the
American Water Works Association (AWWA), and the U.S. Environmental Protection
Agency (EPA). The USBR (1987) discusses application of the standard test methods to
evaluating seepage from dams.

9-3.2.2 Weirs

Weirs are usually metal or plastic plates with a notch in the top edge. They are installed
in a ditch, gutter, pipe, or in manholes in the relief well collection system. The quantity
of water flowing through the notch is calculated by measuring the depth of water from the
invert of the notch to the upstream water surface and using the measurement in the
appropriate hydraulic equation.

The notch can be triangular, rectangular, or trapezoidal. Triangular notches are

appropriate for low flows (less than about 0.05 m3/s [10 cfs]). Rectangular or trapezoidal
weirs are appropriate for larger flows. The crest of the weir should be thin enough that
the nappe springs clear. Standard weir dimensions and calibrations are readily available
(USBR 1984).

Weirs are simple, reliable, inexpensive, and require little maintenance. Limitations are
the severe restriction of the flow channel, relatively high head loss, and the need for
sufficient elevation change to prevent the tailwater from submerging the weir.

9-12
9-13
9-3.2.3 Parshall Flumes

Parshall flumes are specially shaped open channel sections. They consist of a converging
upstream section, a downward sloping throat, and an upward sloping and diverging
downstream section. They are usually permanent installations made of reinforced
concrete, metal, or prefabricated fiberglass and can be sized to measure a wide range of
flows. Throat widths from 25 mm (1 inch) to 10 m (33 feet) are common. Standard
flume dimensions are in USBR (1984). The quantity of water flowing through the throat
is calculated by measuring the depth of water upstream and using the measurement in the
appropriate hydraulic equation. Parshall flumes should be installed level and ideally at a
site free of downstream submergence.

Parshall flumes are simple, reliable, and require little maintenance. They cause minimal
restriction to the flow channel and low head loss. The primary limitation is the relatively
expensive installation.

9-3.2.4 Calibrated Containers

Containers of known volume can be used to measure low flows that are concentrated and
free-falling. The flow rate is computed as the volume of the container divided by the
time required to fill the container.

Extremely low flow rates can be measured accurately. The maximum flow rate is limited
by the size of the container that can be maneuvered quickly into and out of the flow or
into which flow can readily be diverted. Typically, calibrated containers are appropriate
for flows less than about 0.003 m3/S (50 gpm).
Calibrated containers are reliable for low flows and are inexpensive. They have limited
application because of the requirement for a free-falling flow, they are not accurate for
large flows, and are labor intensive.

9-3.3 Movement

Movement can be divided into three types: surface movement, internal movement, and
crack or joint movement. Since it can occur in any direction, measurements in three
mutually perpendicular directions are necessary to accurately determine vector
movement. Measurements are typically made in vertical, transverse horizontal, and
longitudinal horizontal directions. Movement in one or more of these directions is often
judged to be negligible and is not measured.

Surface movement is defined as horizontal or vertical movement of a point on the surface

of a structure relative to a fixed point off of the structure. It is usually determined by

9-14
some type of surveying. Modern surveying equipment has increased the number and type
of surveys that are available.

Internal movement is defined as horizontal or vertical movement within the structure. It

is usually determined relative to some point on the structure or in the foundation.

Joint or crack movement is defined as horizontal or vertical movement of one part of a

structure relative to another part of a structure. It is usually measured across block joints
or cracks in concrete structures or cracks in earth structures.

Tubing or cables for movement measuring devices should be installed to avoid

development of seepage paths along them, or through them, as they deteriorate. As for
piezometer tubing discussed on Section 9-3.1.1., special attention must be paid to sealing
tubing and cables where they cross zones of an embankment dam.

Commonly used techniques for measuring movement are summarized below. More
detailed information may be found in ICOLD (1993), ISRM (1981), USACE (1987a and
1987c) and USBR (1987a and 1987b).

9-3.3.1 Engineering Concepts

All structures move as the result of applied loads. Embankments settle and spread over
time as the result of consolidation and secondary settlement of the dam and foundation
from self weight. Embankments also deform due to external loads produced by reservoir
water, rapid drawdown, earthquakes, undermining, swelling clays, and piping. Concrete
structures deform due to internal loads such as pore pressure, cooling, and alkali
aggregate reaction of concrete; and external loads caused by air and reservoir
temperature, solar radiation, reservoir levels, uplift pressure, wind, earthquakes,
undermining, ice, overflowing water, swelling clay, and foundation settlement.

Movements in response to such loads are normal and acceptable, provided they are within
tolerable ranges and do not cause structural distress. Embankments are less brittle than
concrete structures and can undergo larger movements without distress. As a result,
measurements of surface movements of embankment dams are typically less precise than
those for concrete structures. Sudden or unexpected direction, magnitude, or trend of
surface movement could indicate developing problems. Internal movement measurements
of both concrete and embankment dams and their foundations should be detailed and
precise.

Measuring points for all movement surveys should be installed so that they are not
subject to movement from freeze-thaw action or traffic.

9-15
9-3.3.2 Level Surveys

Vertical surface movements are commonly measured by conventional differential leveling

surveys. Measuring points are established on the crest or slopes of the dam.
Embankment measuring points are usually steel bars embedded in concrete placed in the
fill. Concrete dam measuring points are usually bronze markers set in the concrete or
scratch marks. The change in elevation between the measuring points and survey control
monuments off of the dam are measured using levels and rods. Typically, survey
methods and equipment for measurements of embankments should be sufficiently
accurate to discern movement on the order of 30 mm (0.1 foot). A conventional level and
rod are usually adequate for embankment dams. Typically, survey methods and
equipment for measurements of concrete structures should be sufficiently accurate to
discern movement on the order of 3 mm (0.1 inch). Precision levels and rods equipped
with micrometer targets are usually used for concrete structures.

Level surveys are the simplest and most accurate method for determining vertical
movement of a dam. A limitation of level surveys is the labor cost, though modern
surveying equipment has reduced the time required to perform a survey and reduce the
data.

9-3.3.3 Alignment Surveys

Horizontal surface movements are commonly measured as offsets from a baseline. The
same measuring points used for the level surveys are normally used for alignment
surveys. The method and equipment used depends on the type of dam and the desired
accuracy.

For embankment dams, one or more lines of measuring points are established along the
crest and on the slopes parallel to the crest. Instrument and target monuments are
established at the ends of the lines on the abutments beyond the dam. To measure
movement, a theodolite is set up on the instrument monument on one abutment and
sighted to the target monument on the opposite abutment. Offsets from the line-of-sight
are then measured to each measuring point using a plumb bob and tape. Typically,
survey methods and equipment should be sufficiently accurate to discern movement on
the order of 30 mm (0.1 foot).

For concrete dams a similar procedure is employed, but with refinements to increase the
accuracy of the measurements. These surveys are also known as collimation surveys.
Measuring points are established along straight lines on the crest and, in some cases,
along the face of the dam. The measuring points are markers set in the dam concrete.
Instrument and target monuments are established outside the limits of the dam at the ends

9-16
of the lines of measurement points. The monuments are usually 200- to 250-mm- (8- to
10-inch-) diameter concrete-filled pipes buried at least 3 m (10 feet) into the ground. The
top of the instrument monument is fitted with a threaded plate to fit a theodolite. The
target monument is fitted with a threaded plate to fit a target. The line-of-sight is
established using a high precision theodolite set on the instrument monument and sighted
to the target on the target monument. Offsets from the base line are measured with a
micrometer attached to a moveable target leveled over each measuring point. Typically,
survey methods and equipment should be sufficiently accurate to discern movement on
the order of 3 mm (0.1 inch).

Alignment surveys are the simplest and most accurate method for determining horizontal
movement in straight dams. Their application is limited for curved dams, irregularly
shaped dams, or where the line-of-sight is limited, because the number of measurement
points along any one line is small. A limitation of alignment surveys is the labor cost,
although modern surveying equipment has reduced the time required to perform a survey
and reduce the data

9-3.3.4 Triangulation and Trilateration

Triangulation and trilateration use trigonometric principles of triangles to measure the

location of points on a dam. In triangulation surveys, angles to a measuring point on the
dam are determined from two locations on a base line. Using the known distance
between, and the elevation of base line monuments, the triangle between the three points
is solved trigonometrically to determine the location (horizontal and vertical) of the
measuring point. Angles are measured with precise theodolites.

In trilateration surveys, the distances between a measuring point on the dam and two
locations on a base line are determined. Using the known distance between, and the
elevation of monuments on the baseline, the triangle between the three points is solved
trigonometrically to determine the location (horizontal and vertical) of the measuring
point. Since distances can be measured more precisely than angles, trilateration surveys
are more precise than triangulation surveys.

Distances are measured with electronic distance measurement (EDM) equipment. EDMs
determine distance by measuring the time it takes for light to travel from the source to a
reflector and back and then multiplying by the speed of light. Extremely high accuracies
can be obtained with this equipment. Measurements must be corrected for barometric
pressure, temperature, and the curvature of the earth.

Baseline monuments are similar to instrument monuments used for alignment surveys of
concrete dams. Triangulation and trilateration are useful when measuring points do not

9-17
lie along a straight line or when lines of sight are obstructed. Vertical movements can be
measured with both surveys if the base line has a significant vertical component. The
surveys are highly accurate, but require an experienced crew. Disadvantages are the cost
of the survey crew labor, the cost of establishing the baseline, the need for specialized
equipment, and the relatively complex calculations.

9-3.3.5 Internal Movement

Internal settlement of an embankment or foundation can be measured with a variety of

instruments including settlement plates, cross-arm devices, magnetic- or inductance-type
probe extensometers, fluid leveling devices, pneumatic settlement sensors, vibrating-wire
settlement sensors, and various other mechanical and electrical sounding devices.
Internal horizontal and vertical movements are commonly measured with inclinometers
and extensometers. Internal movements of concrete structures are commonly measured
with plumblines, tiltmeters, inclinometers, and extensometers. The operation,
advantages, and limitations of these devices are well described in the references for
section 9-3.3. Some common types of internal movement instruments are described
below.

Plumblines consist of a plumb bob suspended from a wire in a vertical shaft in a dam.
Measurements of the location of the wire relative to the suspension point are taken at one
or more elevations along the shaft. They are simple, inexpensive, accurate, and reliable.

Tiltmeters consist of a base plate, sensor, and readout device. They are commonly
attached to a surface (internal or external) of a structure and measure vertical rotation of
the surface. They are portable, accurate, and precise.

Inclinometers consist of specially shaped casing, a probe and readout device. They are
commonly installed in vertical drill holes in dams, foundations and abutments, although
they may be installed in a dam during construction. Inclination of the casing is measured
at regular intervals and lateral movement with respect to the bottom of the casing is
calculated. They are reliable and accurate.

Extensometers consist of one or more rods anchored at different depths in a borehole and
a reference head at the surface. They are commonly installed vertically to measure
vertical movement of the reference head relative to the anchor zone(s), though they may
be installed in other orientations. They are accurate and can be used to measure small
movements.

9-18
9-3.3.6 Crack and Joint Measuring Devices

Movement of one side of a crack or joint in a concrete structure relative to the other side
of the joint or crack is commonly measured with reference points or crack meters. Grout
or plaster patches can be used to evaluate whether or not movement is occurring. Many
variations are used.

Reference points can be scratch marks in the concrete, metal pins, or metal plates on
opposite sides of a joint or crack. The distance between the scratch marks is measured
with a micrometer or dial gage to determine movement. Sometimes three points are used
in a triangle to measure both horizontal and vertical movement.

Crack meters are commercially available devices that allow movement in two directions
to be measured. A common device consists of two plastic plates. One plate is opaque
and contains a grid. The other plate is translucent and contains a set of cross hairs. One
plate is fixed on each side of the crack or joint with the cross hairs set over the center of
the grid. Movement is measured by noting the location of the cross hairs with respect to
the grid. A variety of other crack meters including Carlson and vibrating-wire sensors,
dial gages, and mechanics feeler gages may be used to measure movement of cracks.

All these devices are simple to install and monitor. The accuracy and reliability varies
depending on the details of the devices and measurements. Mineral deposits, iron
staining, or efflorescence obscuring the instruments are a common problem if seepage or
leakage flow is present.

9-3.4 Stress and Strain

Earth pressures within fill and against concrete structures are commonly measured with
earth pressure cells. These are also known as total pressure cells. They consist of two
flexible diaphragms sealed around the periphery, with a fluid in the annular space
between the diaphragms. Pressure is determined by measuring the increase in fluid
pressure behind the diaphragm with pneumatic or vibrating-wire sensors. Earth pressure
cells should have similar stiffness as the surrounding soil to avoid inaccurate
measurements of in-situ stress caused by arching. Soil pressures against structures can
also be measured with a Carlson-type cell. It consists of a chamber with a diaphragm on
the end. Deflection of the diaphragm is measured by a Carlson-type transducer and is
converted to stress. Stress in concrete structures can be measured with total pressure cells
or Carlson-type cells designed to have a stiffness similar to concrete. It can also be
measured by overcoring.

The modulus of elasticity, creep coefficient, and the Poisson's ratio for concrete can be

9-19
determined from the laboratory testing of concrete field cylinders. These values are
required for converting strain measurements to stress.

A variety of mechanical and electrical strain gages are used to measure strain in concrete
structures. Some of the instruments are designed to be embedded in the dam during
construction and others are surface mounted following construction. Strain gages are
often installed in groups so that the three-dimensional state of strain can be evaluated.

The operation and limitations of stress and strain instruments are discussed by Dunnicliff
(1988), USACE (1976, and 1987c), and USBR (1976, 1977, 1987a, and 1987b).

9-3.5 Temperature

Temperature measurements of a dam, foundation, or instruments, are often required to

reduce data from instruments, increase precision, or to interpret results. For example,
movements of concrete dams and changes in leakage at concrete dams are commonly
related to changes in temperature. Temperature is also commonly measured in concrete
dams under construction to evaluate mix design, placement rates, and block and lift sizes;
to time grouting of block joints; and to evaluate thermal loads.

Temperatures can be measured with resistance thermometers or thermocouples. The

operation and limitations of these devices are discussed by Dunnicliff (1988), USACE
(1987c), and USBR (1976, 1977 and 1987b).

Temperature measurements of seepage and leakage may indicate the source of seepage.

9-3.6 Seismic Loads

Seismic strong motion instrumentation records acceleration from earthquake shaking.

The data are used to evaluate the dynamic response of dams. Seismic acceleration and
velocity are usually recorded with strong-motion accelerographs. These devices typically
consist of three mutually-perpendicular accelerometers, a recording system, and
triggering mechanism. To prevent accumulation of unwanted data, the instruments are
usually set to be triggered at accelerations generated by nearby small earthquakes or more
distant, larger earthquakes. They are expensive, especially considering that multiple
instruments are necessary to record dynamic response at several locations on a structure,
a foundation, or abutments. The devices must be properly maintained, so that they
operate if an earthquake occurs. These devices are discussed by USACE (1987c) and
USBR (1987a and 1987b).

9-20
9-3.7 Loads in Post-Tensioned Anchors

Post-tensioned anchors consist of single or multiple wires, strands, or bars installed in

drilled holes. The bottom end is grouted in the dam or foundation and the top end is
fitted with a head that allows the anchor to be post-tensioned. The section between the
grouted end and the head is known as the stressing length. It may be free (ungrouted) or
grouted. Post-tensioned anchors are commonly used to improve the stability of concrete
dams. Installation, design, and testing of post-tensioned anchors in discussed by
Littlejohn and Bruce (1976), Hanna (1982), the FHA (1984), and the Post-Tensioning
Institute (1994).

There is no practical method of measuring loads in fully grouted anchors.

Loads in post-tensioned anchors that have a free length can be evaluated by lift-off tests,
load cells, extensometers, and fiber-optic cables.

In lift-off tests, a jack is attached to the head of an anchor and the pressure required to lift
the head is measured by a pressure cell. This type of test requires that the anchor head be
accessible and be capable of being connected to a jack.

Load cells can be located beneath the anchor head to measure the load in the anchor.
Hydraulic and vibrating-wire cells have been used successfully. Electrical resistance
strain gage load cells have not had good performance records. Load cells can be
permanently installed in new anchors and in some cases can be placed under the heads of
existing anchors.

Extensometers and fiber-optic cables can be installed integral with multiple strand or wire
anchors to measure change in length. The length can be converted to load with elastic
constants, assuming no slippage at the head.

Water pressure, seepage, leakage, movement, stress, and strain data taken before and after
installation of anchors may be useful in evaluating the response of the dam to anchor
loads.

9-4 Minimum Instrumentation Recommendations

Minimum instrumentation recommendations for all dams are established in this section.
The recommended minimums are considered to be generally applicable; however, since
each dam is unique, the recommendations should be applied using engineering judgement
and common sense.

9-21
Minimum recommended instrumentation is separated into categories of existing and
proposed dams and further subdivided depending on the hazard potential classification
and the type of structure. Tables 9-4a and 9-4b summarize the recommended minimum
instrumentation. For simplicity, types of measurements rather than specific instruments
are shown in the tables. Instruments for each type of measurement are discussed in
Section 9-3.

9-22
 TABLE 9-4a
MINIMUM RECOMMENDED INSTRUMENTATION FOR EXISTING DAMS 1

TYPE OF LOW- SIGNIFICANT AND HIGH-HAZARD POTENTIAL DAMS

MEASUREMENT HAZARD
POTENTIAL
DAMS — EMBANK- CONCRETE ARCH BUTTRESS SEPARATE INTEGRAL
MENT GRAVITY SPILLWAY POWER-
ALL TYPES
AND/OR HOUSE
OUTLET

VISUAL X X X X X X X
OBSERVATION 2

RESERVOIR X X X X X X
LEVEL

TAILWATER X X X X X X
LEVEL

DRAIN FLOW, X X X X X X
SEEPAGE, AND
LEAKAGE

PORE/UPLIFT X X X X
PRESSURE 3

SURFACE
SETTLEMENT

SURFACE X X X X X
ALIGNMENT

INTERNAL
MOVEMENT

JOINT/CRACK 4 X X X X X
DISPLACEMENT

FOUNDATION X X X X X X
MOVEMENT 5

SEISMIC X X X X X X
LOADS 6

LOADS IN POST- X X X X X
TENSIONED
ANCHORS 7

1
This table is provided to help explain the guidelines. Refer to the text of Section 9-4 for more detailed discussion of minimum
recommendations. Additional instrumentation should be used to address specific concerns.
2
Visual observation consists of walking tours of the crest, toes, abutments, etc.
3
Using existing piezometers, observation wells, or foundation drains; or using existing or new piezometers or observation wells from dams
with reduced uplift assumed in stability analysis or that do not meet criteria using conservative estimate of the phreatic surface.
4
Only on structurally significant joints or cracks that have visible displacement.
5
Should be considered for dams on compressible or weak foundations.
6
Should be considered on a case-by-case basis for dams in seismic zones 3 and 4.
7
For anchors that are required to meet stability criteria, loads should be measured wherever it is possible to measure anchor loads, or
anchors should be modified to measure loads.

9-23
 TABLE 9-4b
MINIMUM RECOMMENDED INSTRUMENTATION FOR PROPOSED DAMS 1

TYPE OF LOW- SIGNIFICANT AND HIGH-HAZARD POTENTIAL DAMS

MEASUREMENT HAZARD
POTENTIAL
DAMS - EMBANK- CONCRETE ARCH BUTTRESS SEPARATE INTEGRAL
MENT GRAVITY SPILLWAY POWER-
ALL TYPES
AND/OR HOUSE
OUTLET

VISUAL X X X X X X X
OBSERVATION 2

RESERVOIR X X X X X X
LEVEL

TAILWATER X X X X X X
LEVEL

DRAIN FLOW, X X X X X X
SEEPAGE, AND
LEAKAGE

PORE/UPLIFT X X X X
PRESSURE

SURFACE X
SETTLEMENT

SURFACE X X X X X X
ALIGNMENT

INTERNAL X X X
MOVEMENT 3,5

JOINT/CRACK 4 X X X X X
DISPLACEMENT

FOUNDATION X X X X X X
MOVEMENT 5

TEMPERATURE X X X

SEISMIC X X X X X X
LOADS 6

LOADS IN POST- X X X X X
TENSIONED
ANCHORS 7

1
This table is provided to help explain the guidelines. Refer to the text of Section 9-4 for more detailed discussion of minimum
recommendations. Additional instrumentation should be used to address specific concerns.
2
Visual observation consists of walking tours of the crest, toes, abutments, etc.
3
For concrete dams greater than about 100 feet high.
4
Only on structurally significant joints or cracks that have visible displacement.
5
Should be considered for dams on compressible or weak foundations.
6
Should be considered on a case-by-case basis for dams in seismic zones 3 and 4.
7
Loads should be measured in anchors that are required to meet stability criteria.

9-24
The minimum instrumentation applies to separate spillway and outlet works only if they
are substantial and independent water-retaining structures. Instrumentation for separate
spillway and outlet structures that retain only minimal water should be evaluated on a
case-by-case basis. In most cases, these types of structures will require little or no
instrumentation.

Instrumentation in addition to the minimum recommended should be used wherever it

will help to resolve a dam safety concern. The minimum instrumentation assumes a
geometrically simple dam on a sound foundation. Instrumentation in addition to the
minimum should be tailored to known or suspected site-specific conditions. For example,
a low dam on a pervious, weak, compressible, jointed, or similar problem foundation may
require more instrumentation than a higher dam on a sound foundation.

Possible causes of, and remedial measures for a wide variety of problems and concerns
are tabulated in reports by the EPRI (1986) and the National Research Council (1983).
Case histories of dam incidents and remedial measures are discussed in reports by the
ASCE (1975, 1988). Instrumentation is often installed to help evaluate causes of
problems and concerns. Table 9-4c summarizes typical instrumentation that can be used
to help evaluate common problems and concerns.

9-4.1 Visual Observation

Though not strictly instrumentation, visual observation is included as a type of

measurement in the tables to stress its importance. Visual observation of all structures
should be made in conjunction with instrument monitoring. It typically consists of
walking tours of the dam crest, toes, and abutments in order to identify any unusual or
abnormal conditions that could jeopardize the safety of the dam. Photographs or videos
are often useful to document existing conditions and to help evaluate whether or not there
has been any change from the previous conditions. Visual observation is discussed in
many publications including USBR (1983), USACE (1977), National Research Council
(1983), EPRI (1986), and ICOLD (1987).

9-4.2 Existing Dams

Minimum instrumentation recommendations for existing dams are less than for proposed
dams, because instrumentation to monitor construction and first filling is not appropriate,
retrofitting instrumentation can be expensive, and the performance of the dam is known.
Minimum recommended instrumentation for existing dams is listed in Table 9-4a.

Existing instruments should continue to be monitored if they still provide useful

information (even if they exceed minimum instrumentation recommendations). However,
if existing instrumentation no longer provides useful or meaningful information, it should
be abandoned as discussed in Section 9-7.4.

9-25
 TABLE 9-4c
TYPICAL INSTRUMENTATION AND MONITORING USED IN EVALUATING CAUSES OF COMMON
PROBLEMS/CONCERNS 1

PROBLEM/CONCERN TYPICAL INSTRUMENTATION

Seepage or leakage Visual observation, weirs, flowmeters, flumes, calibrated

Boils or piping
Uplift pressure, pore pressure, or phreatic surface
Drain function or adequacy
Erosion, scour, or sedimentation
Dissolution of foundation strata
containers, observation wells, piezometers
Visual observation, piezometers, weirs
Visual observation, observation wells, piezometers
Visual observation, pressure and flow measurements, piezometers
Visual observation, sounding, underwater inspection,
photogrametric survey
Water quality tests

Total or surface movement (translation, rotation) Visual observation, precise position and level surveys, plumb
measurements, tiltmeters

Internal movement or deformation in Settlement plates, cross-arm devices, fluid leveling devices,
embankments pneumatic settlement sensors, vibrating wire settlement sensor,
mechanical and electrical sounding devices, inclinometers,
extensometers, shear strips

Internal movement or deformation in concrete Plumblines, tiltmeters, inclinometers, extensometers, jointmeters,

structures calibrated tapes

Foundation or abutment movement Visual observation, precise surveys, inclinometers,

extensometers, piezometers

Poor quality rock foundation or abutment Visual observation, pressure and flow measurements,
piezometers, precise surveys, extensometers, inclinometers

Slope stability Visual observation, precise surveys, inclinometers,

extensometers, observation wells, piezometers, shear strips

Joint or crack movement Crack meters, reference points, plaster or grout patches

Stresses or strains Earth pressure cells, stress meters, strain meters, overcoring

Seismic loading Accelerographs

Relaxation of post-tension anchors Jacking tests, load cells, extensometers, fiber-optic cables
Concrete deterioration Visual observation, loss of section survey, laboratory and
petrographic analyses

Concrete growth Visual observation, precise position and level surveys, plumb
measurements, tiltmeters, plumblines, inclinometers,
extensometers, jointmeters, calibrated tapes, petrographic
analyses
Steel deterioration Visual observation, sonic thickness measurements, test coupons
1
Appropriate remedial measures should be taken for all problems and concerns. Possible remedial measures for a wide variety of
problems and concerns are discussed in EPRI (1986), National Research Council (1983), ASCE (1975 and 1988) and USACE (1986a).

9-26
9-4.2.1 Low-hazard Potential Dams

Extensive instrumentation on low-hazard potential dams is not required. Minimum

recommended instrumentation for low-hazard potential dams consists of visual
observation.

Some low-hazard potential dams are a critical source of municipal water, may cause
unacceptable environmental impacts if they fail (e.g. release of heavy metals in
sediments), or are important to public safety for other reasons. Instrumentation at these
dams should be reviewed on a case-by-case basis and increased as appropriate.

9-4.2.2 Significant and High-hazard Potential Dams

Minimum instrumentation for significant and high-hazard potential dams includes that
recommended for low-hazard potential dams plus additional instrumentation to monitor
headwater and tailwater levels, significant seepage and leakage, pore pressure or uplift
pressure, loads in post-tensioned anchors, and movement. Strong motion instrumentation
should be considered on a case-by-case basis for dams in seismic zones 3 and 4.

9-4.2.2.1 Water Level

One measurement site for headwater level and one for tailwater level is usually sufficient.
The instruments should be located where the levels are representative of project
conditions and where they are easily and safely accessed.

9-4.2.2.2 Seepage and Leakage

Seepage flow from embankment toe drains, concrete gravity or arch foundation drains,
and foundation relief wells are usually collected in a drainage system and routed to the
downstream channel. Flow measurements should be taken near the system outfall where
the combined flow can be measured. Additional measurements at intermediate locations
may be appropriate on a case-by-case basis. Measurements of seepage or leakage from
other sources should be made at locations where the flow is representative and readily
measured. If the foundation has soluble strata, water quality tests should be considered.

9-4.2.2.3 Pore/Uplift Pressure

Pore pressures and uplift pressures are typically the least certain loads on a dam.
Wherever reasonable, estimates of the pressures used in stability analyses should be
verified by measurements. Redundant piezometers should be considered as discussed in

9-27
Section 9-5.5.
Embankments. Pore pressures should be measured in all existing observation wells and
piezometers within an embankment and within the foundation. Installation of pore
pressure instrumentation at existing dams is not required unless they do not meet stability
criteria using a conservative estimate of the phreatic surface or anomalous conditions are
suspected.

Existing dams that do not meet stability criteria using a conservative estimate of the
phreatic surface should have instruments to locate the phreatic surface. If existing
instrumentation is not adequate to do so, additional instruments should be installed.
Instrumentation should be located based on the site-specific geotechnical characteristics
of the embankment, foundation, and abutments. Typically, a minimum of two or three
piezometers along a transverse line through the maximum section are sufficient.
Additional transverse lines may be appropriate for long dams or for complex foundations.

Concrete Gravity Dams. Uplift pressure should be measured in all existing instruments at
existing concrete gravity dams. If existing instruments no longer provide useful
information, they should be abandoned as discussed in Section 9-7.4.

Installation of uplift pressure instrumentation at existing dams is not required unless a

reduction in uplift is needed to meet stability criteria or anomalous conditions are
suspected.

Dams that require a reduction in uplift below a linear variation between headwater and
tailwater to meet stability criteria have instrumentation to verify the uplift reduction. If
existing instrumentation is not adequate to demonstrate the uplift reduction, additional
instruments should be installed. Instrumentation should be located based on the site-
specific geotechnical characteristics of the foundation and abutments. Typically, a
minimum of two or three piezometers along a transverse line through the maximum
section are sufficient. Additional transverse lines may be appropriate for long dams or
for complex foundations.

Uplift pressure measurements should be made in foundation drains of existing concrete

gravity dams if no other means to measure uplift exists and if additional instrumentation
is not required. Measurements should be made in selected drains that are judged to
provide representative uplift pressures beneath sections of the dam that are critical for
stability. Drains should be clear the full length of their depth. Pressure measurements
should be made in only one or two drains at one time because plugging a large number of
drains may significantly increase the uplift pressure and possibly jeopardize the stability
of the dam.

9-28
Uplift pressures measured in foundation drains may be greater or less than actual uplift
pressures on potential failure surfaces. In a uniform foundation they tend to be greater
than actual uplift pressures because the measuring device prevents the drainage and
pressure reduction normally provided by the drain. In stratified foundations pressures
measured in drains may be less than actual uplift pressures. This is because the pressure
in various strata intersected by the drain are combined in an ambiguous manner,
depending on the pressure and permeability of each strata. In this case, pressures should
be measured in isolated lengths of the drain to identify the location of high and low
pressure strata.

Arch and Buttress Dams, Spillways, Outlet Structures, and Powerhouses. Uplift pressure
should be measured in all existing instruments at existing arch dams, buttress dams,
separate spillways and/or outlet structures, and integral powerhouses. Installation of
uplift pressure instrumentation at existing dams is not required unless a reduction of uplift
is needed to meet stability criteria or anomalous conditions are suspected.

For thin arch dams and buttress dams that are not founded on slabs, uplift pressures
generally have a minimal effect on stability and need not be measured. All other dams
that require a reduction in uplift below a linear variation between headwater and tailwater
to meet stability criteria should have instrumentation to verify the uplift reduction. If
existing instrumentation is not adequate to demonstrate the uplift reduction, additional
instruments should be installed. Instrumentation should be located based on the site-
specific geotechnical characteristics of the foundation and abutments. Typically, a
minimum of two or three piezometers along a transverse line through the maximum
section are sufficient. Additional lines may be appropriate for long dams or for complex
foundations.

9-4.2.2.4 Movement

Movement is perhaps the most important indicator of structural distress. Though all dams
deform in response to applied loads, excessive movement may indicate developing
problems.

Embankments. For existing embankment dams, settlement of the crest or bulging of the
slopes might indicate developing problems. Visual observation by a trained inspector
should be sufficient to identify significant movements of embankments that have a
satisfactory performance record with respect to movement.

Concrete Gravity, Arch, and Buttress Dams. Concrete gravity, arch, and buttress dams
deform elastically in response to changing reservoir loads and seasonal temperature
changes. However, inelastic movements might indicate potential instability. Transverse

9-29
horizontal movement of the crest of the dam should be monitored. Longitudinal
horizontal and vertical movements are usually small and generally do not need to be
routinely monitored. The measurements should be sufficiently precise to clearly identify
cyclical seasonal movements from water level and temperature loadings and any inelastic
trends. One line of four or five measuring points along the crest is typically sufficient.
Measuring points should be spaced close enough to allow measurement of all significant
deformation. At least two survey monuments (survey control points) should be
permanently established off the dam structure.

Contraction joints between blocks of concrete gravity and buttress dams are a plane of
weakness and if not keyed or grouted, movement will first become visible at the joints.
Small elastic movements in response to seasonal temperature and reservoir levels changes
are normal and do not require instrumentation to monitor them. Wherever there is
indication of significant inelastic movement, instrumentation to measure relative
movement between blocks should be installed.

Almost all existing concrete dams have a variety of cracks. Most cracks are not
important with respect to the structural stability or integrity of the dam. Nevertheless,
large, recently formed, growing, or critically oriented cracks may be an indication of
structural distress. As with block joints, deformation will first become visible at cracks.
Instrumentation should be installed to measure relative movement across cracks that are
judged to be a potential indication of structural distress. The measurements should be
sufficiently precise to clearly identify seasonal trends and any inelastic movements.

Arch dams are designed to act monolithically and measurement of relative movements,
except at significant cracks, is usually not warranted. Arch dams can impose higher
stresses on foundations and abutments than other types of dams. Existing arch dams on
foundations or abutments that are expected to have significant deformation should have
instrumentation to measure the deformation of the foundation and abutments. The
appropriate number and location of instruments depends on the specific foundation
conditions.

Spillways, Outlet Structures, and Powerhouses. Recommended minimum movement

instrumentation for significant water-retaining spillway and outlet structures and integral
powerhouses consists of that appropriate to measure transverse horizontal surface
movements. Two measuring points are typically sufficient. At least two survey
monuments (survey control points) should be permanently established off the dam
structure.

9-30
9-4.2.2.5 Seismic Loads

Seismic strong motion instrumentation should be considered on a case-by-case basis for

dams in seismic zones 3 and 4 shown on the seismic zone maps in Chapters III and IV of
these guidelines. Dam design, foundation materials, and methods of construction, and
any site-specific seismotectonic studies should be weighed against any potential benefits
before seismic strong motion instrumentation is required.

9-4.2.2.6 Loads in Post-Tensioned Anchors

Loads should be monitored in post-tensioned anchors that are required to meet stability
criteria. The number of anchors to be monitored should be evaluated on a case-by-case
basis, but should typically be between 5 and 10 percent of the total number of anchors.

All existing post-tensioned anchor installations that do not have provisions to measure
loads in representative anchors should be modified, wherever possible, to have such
provisions. Existing post-tensioned anchors that cannot be modified to measure loads
should be evaluated on a case-by-case basis. The possibility of loss of load from
corrosion, creep of the grouted anchorage, and movements that may have locally yielded
the anchor should be evaluated.

All new post-tensioned anchor installations in existing dams should have provisions for
long-term measurement of loads in a representative number of anchors.

9-4.3 Proposed Dams

Minimum instrumentation for proposed dams includes that recommended for existing
dams plus additional instrumentation to monitor conditions during construction, first
filling, and the early life of the project. The minimum recommended instrumentation is
shown in Table 9-4b.

The recommended minimum instrumentation is limited to that which clearly provides

information useful in confirming key design assumptions and evaluating the stability of
the structures. Instrumentation is often included in proposed dams for reasons other than
dam safety such as to guide timing of construction operations or to provide information
for various studies. While this type of instrumentation is necessary and encouraged, it is
not included in the minimum.

9-31
9-4.3.1 Low-hazard Potential Dams

Minimum recommended instrumentation for proposed low-hazard potential dams is the

same as recommended for existing low-hazard potential dams — that is, visual
observation.

Instrumentation similar to that for proposed significant and high-hazard potential dams
should be considered for proposed low-hazard potential dams that are likely to become
high-hazard potential dams from downstream development.

9-4.3.2 Significant and High-hazard Potential Dams

Minimum instrumentation recommended for proposed significant and high-hazard

potential dams includes that recommended for existing significant and high-hazard
potential dams plus additional instrumentation to monitor uplift or pore pressures, internal
movement, foundation movement, and loads in post-tensioned anchors. Strong motion
instrumentation should be considered on a case-by-case basis for dams in seismic zones 3
and 4.

9-4.3.2.1 Pore/Uplift Pressure

All proposed significant and high-hazard potential dams should have instrumentation to
measure pore pressures or uplift pressures to confirm design assumptions, evaluate pore
pressures during construction, check seepage conditions, check performance of the
drainage system, and measure the uplift pressure distribution. The number and location
of the instruments depends on the type of dam and site-specific geotechnical
characteristics. Redundant piezometers should be considered as discussed in Section 9-
5.5.

Embankments. Embankments should have piezometers included in the dam and

foundation to confirm design assumptions during construction and first filling, and for
long-term monitoring during operation.

Embankments During Construction and First Filling. The need for piezometers in
impervious cores of embankment dams should be left to the discretion of the design
engineer. Problems have occurred in the past from seepage developing along the
piezometer cables and tubing. Piezometers are not essential in moderate to thin,
impervious cores of low- to moderate-height dams that are constructed with good
moisture control. The appropriateness of piezometers in thick core zones or unusually
high dams should be evaluated on a case-by-case basis. If used, special attention should
be given to sealing the piezometer cables and tubing to prevent development of seepage

9-32
paths and using filter material around tubing located outside of the core to protect against
piping.

Piezometers should be located downstream of the impervious core to monitor the

effectiveness of the drainage system during filling.

Where the foundation contains compressible, low permeability strata, piezometers should
extend into the foundation. Sufficient piezometers should be installed to measure
induced pore pressures in the foundation beneath the dam and also for an appropriate
distance upstream of and downstream from the dam.

Embankments During Operation. More piezometers are commonly installed to monitor

construction and first filling than are required for long-term operation. Instrumentation
should be planned such that some of the construction/first filling instruments will be used
for long- term monitoring. At least one line of pore pressure instruments located along a
transverse plane through the maximum section is recommended for long-term monitoring
of embankment dams. Additional transverse lines of instruments may be appropriate for
long dams or for complex foundations. Sufficient instruments should be installed along
each line to define the phreatic surface through the dam and in the foundation. Three or
four piezometers are usually sufficient. Specific locations should be based on the
embankment zoning, foundation, and abutments. If relief wells are used, the line of
piezometers should extend downstream from the line of relief wells and additional
piezometers should be located at midpoints between selected relief wells.

Concrete Gravity Dams. All proposed concrete gravity dams should have instruments
installed at the base to monitor uplift pressures. At least one line of piezometers located
along a transverse plane through the maximum section is recommended. Additional lines
may be appropriate for long dams or for complex foundations. Sufficient piezometers
should be located along the line to adequately measure the uplift pressure distribution.
Two or three piezometers are typically sufficient. If foundation drains are included in the
proposed dam, at least one piezometer should be located along the line of drains, midway
between two drains, to provide data to evaluate drain spacing.

Piezometers within the foundation are appropriate if the foundation is soil, soft rock, or if
it contains relatively impervious, adversely oriented strata that may act as an aquiclude or
be a potential plane of sliding. Sufficient piezometers should be installed to adequately
measure the uplift pressure distribution.

Arch and Buttress Dams, Spillways and Powerhouses. Uplift pressures beneath thin arch
dams and buttress dams not founded on slabs are less important than for concrete gravity
dams because of the smaller base area on which uplift can act. The appropriateness of

9-33
piezometers should be evaluated on a case-by-case basis for these structures. For thick
arch and buttress dams that are founded on slabs, piezometers should be installed in the
foundation and in the abutments to verify design assumptions. The number and location
of the piezometers depends on the specific conditions at the site. Piezometers within the
foundation are appropriate if the foundation is soil, soft rock, or if it contains relatively
impervious, adversely oriented strata that may act as an aquiclude or be a potential plane
of sliding. Enough piezometers should be installed to adequately measure the uplift
pressure distribution.

Separate spillways, outlets, and integral powerhouses should have one transverse line of
piezometers similar to that for concrete gravity dams.

9-4.3.2.2 Movement

Movement instrumentation recommended for proposed dams includes that recommended

for existing dams plus additional instrumentation to monitor surface movement of
embankment dams and internal movement of concrete dams.

Embankments. Horizontal (longitudinal and transverse) and vertical surface movement of

proposed embankment dams should be monitored. One line of measuring points along
the crest of an embankment dam is usually sufficient. For large embankments or those on
soft foundations, an additional longitudinal line(s) on the downstream slope is
appropriate. Measurement points should be spaced sufficiently close to allow
measurement of all significant deformation. Five to 10 measuring points are typically
sufficient. At least two survey monuments (survey control points) should be permanently
established off the dam structure.

The need for instruments to monitor internal movement should be evaluated on a case-by-
case basis. For example, there may be little value in monitoring internal movement of
small to moderate height embankment dams on good foundations or of rockfill dams on
good foundations. Internal movement instrumentation would be appropriate for high
dams, those with wide impervious zones, or those on compressible or weak foundations.

Proposed embankments on foundations that contain compressible strata should have

additional instrumentation installed to monitor foundation deformation. The appropriate
number and location of instruments depends on the specific foundation conditions.

All Other Structures. Concrete gravity, arch, and buttress dams should have transverse
horizontal surface movement instrumentation as recommended for existing concrete
gravity dams. In addition, proposed dams that are greater than about 100 feet high should
have instrumentation to monitor transverse horizontal internal movement. Internal

9-34
movement should be monitored along a transverse line through the maximum section.
For long dams, additional transverse lines may be appropriate. The need for internal
movement instrumentation on lower dams should be evaluated on a case-by-case basis.

All proposed concrete structures on soft or deformable foundations should have

instrumentation to measure horizontal and vertical movements of the dam with respect to
the foundation and deformation of the foundation. The appropriate number and location
of instruments depends on the specific foundation conditions.

Arch dams may impose higher stresses on foundations and abutments than other types of
dams. Proposed arch dams on foundations or abutments that are expected to have
significant deformation should have instrumentation to measure the deformation of those
foundation and abutments. The appropriate number and location of instruments depends
on the specific foundation conditions.

9-4.3.2.3 Temperature

No temperature measurements are recommended for embankment dams. Proposed

concrete gravity and arch dams should have an array of instruments to measure internal
and surface temperatures along a transverse plane through the maximum section. In
addition, concrete arch dams should have a string of instruments to measure reservoir
temperature along the height of the maximum section. The data should be collected until
the dam has been in satisfactory service for several years and the temperatures stabilize
and fluctuate between predictable values.

9-4.3.2.4 Seismic Loads

Seismic strong motion instrumentation should be considered on a case-by-case basis for

dams in seismic zones 3 and 4 shown on the seismic zone maps in Chapters III and IV of
these guidelines. Dam design, foundation materials, and methods of construction, and
any site-specific seismotectonic studies should be reviewed and any potential benefits
should be weighed against the life-cycle costs before seismic strong motion
instrumentation is required.

9-4.3.2.5 Loads in Post-Tensioned Anchors

Loads should be monitored in post-tensioned anchors that are required to meet stability
criteria. The number of anchors to be monitored should be evaluated on a case-by-case
basis, but should typically be between 5 and 10 percent of the total number of anchors.

9-35
9-36
9-4.4 Additional Instrumentation

Instrumentation, in addition to the minimum recommended, should be required wherever

there is a concern regarding a condition that may affect dam safety or other critical water
retaining structures. Typical reasons to require additional instrumentation are to check
design assumptions; to provide data to evaluate specific problems such as continuing
movement, excessive cracking, or increased seepage; to provide data to support design of
remedial modifications; and to provide data to evaluate effectiveness of remedial work.

Table 9-4c lists typical types of instrumentation that should be considered to evaluate
common problems and concerns. The table is not exhaustive, but it provides examples of
instrumentation and monitoring that can be used for general types of problems and
concerns. The instrumentation listed may not be appropriate for all cases.
Instrumentation and frequency of monitoring for specific problems and concerns should
be selected based on the specific circumstances. Once the problem or concern has been
resolved, the usefulness of the instrumentation should be re-evaluated. If it no longer
provides useful information, it should be abandoned as discussed in Section 9-7.4.

Appropriate remedial measures should be taken for all problems and concerns. Possible
remedial measures for a wide variety of problems and concerns are discussed in EPRI
(1985), National Research Council (1983), ASCE (1975 and 1988), USACE (1986a), and
Chapter V of these guidelines.

9-5 Instrumentation System Design

Instrumentation system design should receive the same level of effort as other features of
a dam. It should follow a logical step-by-step process beginning with establishing the
objectives and ending with pre-determined action based on the data obtained. General
considerations for design of instrumentation systems are discussed by ICOLD (1969),
USACE (1989a), and USCOLD (1986). Basic steps involved in the process are described
below. The process is discussed in more detail by Dunnicliff (1982, 1988, 1990) and
USCOLD (1986). Dunnicliff (1988, Appendix A) gives a checklist of steps for planning
an instrumentation system.

9-5.1 Project Conditions

The initial step in any instrument system design is to establish the site conditions. At
existing dams, all information available about the design, construction, and performance
of the dam should be reviewed. At proposed dams, groundwater levels, foundation
stratigraphy, dam design, and construction methods should be reviewed. Areas of

9-37
potential weakness and their potential effects on the stability of the dam should be
identified. Often some type of instrumentation, such as piezometers, will be used to help
identify site conditions.

9-5.2 Purpose of Instrumentation

The next step is to define the purpose of all existing and proposed instrumentation. All
instrumentation should have a specific purpose. The designer should understand what
data will be generated by the instrumentation and how the data will be used. If there is
not a valid purpose for the instrumentation, it should be abandoned or should not be
installed.

There are a variety of valid purposes for using instruments to monitor dams. Minimum
instrumentation provides the basic engineering measurements required to adequately
assess dam stability and to monitor indications of developing problems. Additional
instruments might be required to confirm design assumptions and to evaluate performance
by comparing measured to predicted behavior during construction, first filling, rapid
drawdown, and long-term operation. Instruments can be installed to support operations,
to evaluate specific conditions at a site, or to obtain data for design or evaluation of
remedial repairs.

Dams with complex foundations, known geologic anomalies, unique designs, marginal
design criteria, or unconservative assumptions usually require more instrumentation to
demonstrate satisfactory performance than dams without those features.

9-5.3 Types of Measurements

The types of measurements that are commonly monitored at a dam site are discussed in
Section 9-3. They include headwater level, tailwater level, pore pressure, uplift pressure,
seepage, leakage, surface movement, internal movement, crack or joint movement, stress,
strain, temperature, and seismic loads. The magnitude and expected ranges of the
parameters should be estimated to allow selection of the proper instruments. Levels that
indicate the development of potentially hazardous conditions should be established during
instrumentation system design.

9-5.4 Types of Instruments

Usually there are several commercially available instruments for each type of
measurement required. When selecting instrumentation, the major consideration should
be reliability and not cost. Reliability encompasses a variety of factors including

9-38
simplicity, durability, longevity, precision, accuracy, and a length of satisfactory
performance history. The relative importance of each of the factors depends on the
purpose of the instrument. Instruments appropriate for use during construction may be
different than those for long-term operation. For example, piezometers that have very
short time-lag, but limited life, may be appropriate for control of construction operations,
whereas longevity may be more important for long-term monitoring.

The type of data acquisition — manual or automated — should be considered when

choosing instruments. However, automated data acquisition systems should not be used
to justify the use of inaccessible electrical transducers. Transducers that are accessible
for calibration or replacement should be used wherever possible. For example, vibrating
wire piezometers should not be used in preference to open standpipe piezometers just
because the data will be acquired automatically. Open standpipe piezometers can be
automated with accessible electrical transducers. For conditions for which access could
be difficult (e.g. uplift pressures during floods), consideration should be given to using
remotely read instruments.

The total cost should be considered when comparing alternative instruments or instrument
systems. The total cost includes the instruments, installation, maintenance, longevity,
monitoring, and data processing. The least expensive instrument does not necessarily
provide the lowest life-cycle cost, especially if replacement instruments will have to be
installed. Allocation of sufficient funds to cover the total cost of instrumentation during
design phase can help avoid inadequate collection and evaluation of data due to lack of
funds.

9-5.5 Location and Number of Instruments

Minimum instrumentation should be installed where behavior is expected to be

representative of the dam as a whole. The number of instruments should be sufficient to
provide a complete picture of the parameter being measured. Usually, minimum
instrumentation should be installed along longitudinal or transverse sections of the dam.

Often, depending on access and equipment costs, it may be more cost-effective to install
redundant instruments to account for the possibility of malfunction, than to replace
inoperable instruments at a later date. For example, vibrating-wire piezometer
transducers are relatively inexpensive and are often installed in pairs to provide continuity
of data if one of the transducers should fail. If a sensor will be inaccessible for
calibration or replacement, multiple sensors should be considered to provide redundancy.

Redundant measurements are also useful for verifying and evaluating unusual readings.

9-39
Redundancy can be provided by using additional lines of instruments, more closely
spaced instruments, or different instruments to measure the same feature.

Instrumentation to monitor a particular area of concern should be placed along cross

sections where the suspected behavior will most likely manifest itself. The results of
structural analyses may indicate appropriate locations and numbers of instruments. Often
additional cross-sections should be monitored adjacent to areas of concern to provide data
for comparison and to aid in the evaluation of the extent and magnitude of the concern.

For new dams, the need to install instrumentation should be established early in the
design phase and before the preparation of final construction drawings to avoid
interference. For existing structures, as-built drawings and the location of equipment
should be reviewed. Potential interferences with rebars, pipes, and gates should be
identified prior to finalizing instrument drawings.

9-5.6 Procurement and Installation

Typically, once the type and quantity of instruments have been selected, a specification is
prepared for their procurement and installation. Because many organizations are required
by policy to select the lowest bid, it is important that the evaluation criteria are included
in the technical specifications so that the best equipment can be obtained at the lowest
cost. Consideration should be given to instruments that have been proven in the field in
similar applications and have a record of reliability (as defined in Section 9-5.4), ease of
testing and maintenance, suitable response time, and ease of installation in the specific
locations chosen. In some cases, delivery times, repair policies, and the number of
vendors providing the type of instrument, can be important. "Or equal" provisions in the
specifications should be avoided since there is often a considerable difference between
similar instruments made by different manufacturers. Proposed substitutions for specified
instruments should be carefully evaluated and compared to the selection criteria.

Installation specifications should include detailed step-by-step procedures for installing

and testing instruments. The installation should include an installation log to document
"as-built" conditions. Where appropriate, calibration measurement, performance testing,
and initial readings should be obtained during installation. Consideration should be given
to the level of the contractor's experience with installing similar instruments in similar
conditions that the contractor should have. A quality assurance program should be
included in the specification.

Drilling to install instruments can potentially damage existing structures.

Hydrofracturing of embankments can be caused by drilling and can lead to piping and

9-40
loss of the reservoir. Drilling can intercept and destroy filter or transition zones in
embankments, or damage drains, underground utilities, or other structures. Installation
specifications should be carefully written and enforced to prevent damage to existing
dams and appurtenances.

Instrumentation is often damaged by maintenance equipment or vandals. All

instrumentation should be enclosed in lockable covers or should be otherwise protected.
For example, the tops of embankment piezometers should be installed inside metal guard
pipes with locking covers. Embankment survey measuring points should be sturdy
enough so that they will not be damaged by mowing equipment or should be protected
with guard poles.

9-5.7 Monitoring Program

A monitoring program that includes responsibility assignments and procedures for data
collection, reduction, processing, and presentation should be developed and documented.
Responsibility for collecting, reducing, and evaluating the instrumentation data should be
assigned to specific groups or individuals. Specific step-by step procedures for setting up
equipment, taking measurements, recording data, and field screening data should be
included. This information may be provided by the supplier. The monitoring program
should include steps for reporting the monitoring results through responsible management
personnel and a system to ensure timely response to problems disclosed in the
surveillance and evaluation of data.

Personnel who perform visual observations, and collect, reduce, and evaluate data, should
be given basic dam safety training. The training should include as a minimum, common
causes of dam failures and incidents, identification of signs of potential distress by visual
observation, and actions to be taken when unusual conditions, signs of potential distress,
or emergency conditions occur. A series of videotapes and workbooks known as Training
Aids for Dam Safety (TADS) are available from the USBR.

For manual data acquisition, data sheets should be developed to use for recording
instrument data. All data sheets should show the project name, instrument type, and
instrument location. They should have places to record the date, time, operator, data, and
comments. The sheets should also have places to record complementary data such as
headwater and tailwater levels, weather, rainfall, snowfall, temperature, and any unusual
conditions. Examples of data sheets are included in USBR (1987a and 1987b) and
USACE (1987b).

Threshold limits and the criteria used to develop them should be documented in the

9-41
monitoring program. Threshold limits should be established based on the specific
circumstances. In some cases, they can be based on theoretical or analytical studies (e.g.
uplift pressure readings above which stability guidelines are no longer met). In other
cases, they may need to be developed based on measured behavior (e.g. seepage from an
embankment dam). Sometimes they may be used to identify unusual readings, readings
outside the limits of the instruments, or readings which, in the judgement of the
responsible engineer, demand evaluation. Both magnitude and rate of change limits
should be established.

Threshold limits are intended to provide checks and balances in a monitoring program.
Readings that exceed threshold limits do not necessarily mean that drastic action must be
taken, only that some action must be taken. The monitoring program should include
action to be taken if an instrument reaches its threshold limit. However, predetermined
actions are no substitute for situation specific responses and should only be used as a
guideline. Actions to be taken upon exceeding a threshold limit depend on the particular
circumstances, but might include a combination of the following.

• Notify the responsible engineer.

• Confirm the reading by retaking it, and where possible, confirm the instrument
calibration by the use of redundant readings.

• Inspect the dam.

• Evaluate the situation and revise the threshold limit.

• Increase the frequency of readings to monitor and provide data to further evaluate
the situation.

• Implement investigative measures such as the installation of additional

instruments.

• Implement remedial measures such as cleaning foundation drains, repairing

damage, or modifying the dam.

• Implement emergency measures such as drawing down the reservoir.

The monitoring program should include requirements for establishing initial or baseline
measurements. Since most data are compared against these measurements, it is important
that they are correct. A minimum of three and preferably more measurements should be

9-42
taken. The readings should meet expected values and accuracy. If they do not, the
equipment should be checked and additional readings should be taken until readings meet
expected values and accuracy or the measured values can be justified.

Some instruments such as piezometers, and some types of strain gages, take a significant
amount of time to stabilize after installation due to drilling effects, lag time, or
temperature. For these instruments, daily or more frequent readings should be taken until
the readings have stabilized.

9-5.8 Documentation

An instrumentation document should be developed that includes a discussion of the

purpose of each instrument, expected ranges of data, threshold limits, manufacturers'
literature, procurement and installation specifications, installation logs, calibration data
and initial readings. Plan and section drawings showing the number, location and details
of each instrument should be included in the document. Appropriate subsurface
stratigraphy should be shown on the drawings.

Details of subsurface conditions and construction should be documented for all proposed
dams and remedial work at existing dams.

9-5.9 Maintenance and Calibration

A routine of regular maintenance of instruments, readout devices, and field terminals

should be established. For many instruments, manufacturers will suggest maintenance
procedures and schedules that should be followed unless there is adequate justification to
alter them.

Periodic calibration of all instruments is necessary to provide accurate data. Detailed

measurements and careful evaluation of data has little value, and may be misleading, if
the data are inaccurate. The nature and frequency of calibration depends on the specifics
of the instrumentation and should be developed on a case-by-case basis.

Instruments that are suspected to be malfunctioning should be tested to evaluate whether

or not they are functioning properly, as discussed in section 9-7.4.

If the responsible engineer determines that an instrument no longer provides useful or

meaningful information, it should be abandoned as discussed in 9-7.4.

9-6 Monitoring Schedules

9-43
Typical monitoring schedules are discussed in this section. The schedules are considered
to be generally applicable for all significant and high-hazard dams; however, since each
dam is unique, the schedules should be applied using engineering judgement and common
sense. For example, more frequent readings of reservoir level should be taken at pumped
storage projects. Table 9-6 summarizes typical frequencies of measurements for various
stages in the life of a dam. Specific monitoring schedules should be developed on a case-
by-case basis.

The types of measurements listed in Table 9-6 are the same as used for minimum
instrumentation. Visual observation is listed as a type of measurement to emphasize its
importance. Frequent visual observation by trained personnel will often detect unusual
conditions, such as increased seepage, or cloudy seepage, or large movements, before
instrumentation readings taken at widely spaced points and at long intervals would reveal
such conditions or where no instrumentation exists.

During construction, the dam and foundation are adjusting to such factors as self-weight,
thermal loads, seepage, and any unusual conditions. Measurements should be taken
frequently to allow construction operations to be adjusted to changing conditions. Less
frequent measurements may be appropriate during construction shutdowns.

During first filling, the dam and foundation are adjusting to the reservoir load.
Monitoring frequencies should depend, in part, on the rate of filling. In the first few
years of operation following first filling most dams have not reached equilibrium with
respect to self-weight, concrete thermal load, reservoir load, seepage forces, and pore
pressure/uplift. Measurements should be taken frequently because most dam failures and
incidents occur during these periods.

Even though existing dams have generally reached equilibrium with imposed loads,
baseline data must be obtained to compare with subsequent measurements. Therefore,
the frequency of measurements shown for first, second, and third years apply to new
instrumentation installed at existing dams.

After a dam has substantially adjusted to imposed loads, the frequency of readings can be
reduced to that shown in Table 9-6. Further reduction of frequency may be justified in
some cases. For example, after surface settlement of an embankment dam has ceased or
become very small, surveys every 2 to 5 years may be adequate. The frequency of
measurements shown in the table for long-term operation assumes that the performance of
the project is satisfactory.

More frequent measurements than shown in the table should be made whenever an

9-44
 unusual situation develops or whenever they will help to resolve a dam safety concern.
For example, when the reservoir is abnormally raised or lowered (whether for a specific
reason or because of flood surcharge), frequent readings during the raising or lowering
should be made, plotted, and compared to expected behavior in order to identify any
potentially unusual behavior. Examples of other situations requiring more frequent
measurements include sustained high reservoir levels, earthquakes, unusual movements,
abnormal measurements, threshold measurements exceeded, new cracks, new seeps, and
new leaks. Following resolution of the problem or concern, measurements should return
to the normal schedule.
TABLE 9-6
TYPICAL MONITORING SCHEDULE FOR SIGNIFICANT AND HIGH-HAZARD POTENTIAL DAMS 1

TYPE OF FREQUENCY OF MEASUREMENTS

MEASUREMENT2
CONSTRUCTION FIRST FIRST YEAR SECOND AND LONG-TERM
FILLING AFTER THIRD YEARS OPERATION
FILLING

VISUAL Daily Daily Weekly Monthly Monthly

OBSERVATION
RESERVOIR - Daily to Semi-monthly Monthly and at Monthly to
LEVEL Weekly and at same same time as quarterly and
time as any any other at same time as
other measurements any other
measurements measurements

TAILWATER - Weekly Semi-monthly Monthly and at Monthly to

LEVEL and at same same time as quarterly and
time as any any other at same time as
other measurements any other
measurements measurements

DRAIN FLOW - Daily to Weekly to Monthly Monthly to

Weekly monthly quarterly
SEEPAGE/ Monthly Daily to Weekly to Monthly Monthly to
LEAKAGE FLOW Weekly monthly quarterly
PORE PRESSURE/ Daily to Weekly Daily to Monthly Monthly Monthly to
UPLIFT weekly quarterly
SURFACE - Monthly Quarterly Semi-annually Semi-annually
SETTLEMENT to annually to annually

SURFACE - Daily to Quarterly Semi-annually Semi-annually

ALIGNMENT monthly to annually to annually
INTERNAL - Weekly to Monthly to Monthly to Monthly to
MOVEMENT Monthly quarterly semi-annually annually
JOINT/CRACK - Weekly to Monthly to Monthly to Monthly to
DISPLACEMENT Monthly quarterly semi-annually annually

9-45
 TYPE OF FREQUENCY OF MEASUREMENTS
MEASUREMENT2
CONSTRUCTION FIRST FIRST YEAR SECOND AND LONG-TERM
FILLING AFTER THIRD YEARS OPERATION
FILLING

FOUNDATION Weekly Weekly to Quarterly Semi-annually Semi-annually

MOVEMENT Monthly to annually
TEMPERATURE Hourly to weekly Weekly Semi-monthly Monthly Typically not
required
LOADS IN POST- Typically not Typically not Annually Typically not Quinquenniall
TENSIONED required required required y
ANCHORS

1
Refer to Section 9-6 for a discussion of monitoring schedules. Readings should be taken following major events
such as earthquakes and floods. More frequent readings should be taken as needed to address specific concerns.
2
Refer to Sections 9-3 and 9-4 for discussion of types of measurements, and minimum instrumentation.
Schedules should be arranged so that all instrumentation data are collected at the same
time to facilitate correlation between the measurements. For example, measurements that
are taken quarterly should be taken on the same day as measurements that are taken
monthly.

Infrequent measurements should be scheduled to coincide with maximum and minimum

reservoir, thermal, or other significant loads on the dam. For example, semi-annual
measurements might be taken to coincide with low and high reservoir levels or minimum
and maximum temperatures, and annual measurements might be taken to coincide with
high reservoir levels.

Data should be collected from instrumentation immediately after installation. This is

especially true of movement measurement devices, because all subsequent measurements
will be subtracted from the initial reading to calculate movement.

For concrete dams where thermal loads can be significant, internal movement and
crack/joint movement measuring devices should be read early in the morning to reduce
the influence of daily temperature changes and solar radiation. Where it can be done
safely, precise surveys for surface movements should be done in the early morning, just
before sunrise, to avoid visual distortion due to heat waves and wind currents.

9-7 Data Processing and Evaluation

This section is written assuming data are collected manually. Automatic data acquisition
is becoming more common and is discussed in Section 9-8. The steps required to process
and evaluate data, whether collected manually or automatically, are the same.

9-46
Instrument data should be processed and evaluated according to the procedures
established by the monitoring program. Accumulation of instrument data by itself does
not improve dam safety or protect the public. Data must be conscientiously collected,
meticulously reduced, graphically summarized, and interpreted in a timely manner. Data
must be evaluated with respect to the safety of the dam. A poorly planned program will
produce unnecessary data that the dam owner will waste time and money collecting and
interpreting, often resulting in disillusionment and abandonment of the program. A
typical monitoring program would include the steps discussed below. Example data
collection forms, data tables, and data plots, are provided in Appendix B.

9-7.1 Data Collection

Data collected manually should be recorded on the data sheets prepared as part of the
monitoring program. Complementary data, such as air temperature, reservoir level,
reservoir temperature, recent precipitation, and other information or observations that
may be important in evaluating the instrumentation data should be noted on the data
sheets. Data should be compared against previous measurements and threshold limits in
the field to identify erroneous measurements. Measurements that are outside of normal
scatter or threshold limits should be immediately retaken.

Personnel collecting data should be trained in the operation of the instruments, the
importance of the data and the need for proper documentation. They should be trained to
identify improperly functioning instruments based on measured data or visual
observations. They should be aware of the procedures to follow, should unusual or
threshold measurements occur.

Personnel collecting data should visually observe the dam for indications of poor
performance such as offsets, misalignment, bulges, depressions, seepage, leakage, change
in color of seepage or leakage, and cracking.

All monuments and measuring points should be inspected during data collection for
evidence of damage or movement from external sources such as frost heave, impact from
maintenance equipment, or vandalism. The assumption is made during data reduction
and interpretation that the survey control monuments have not moved and that any
movements of the measuring points represent movement of the structure.

9-7.2 Data Reduction

9-47
Most instruments require raw data to be converted into useable engineering units. The
arithmetic calculations required to convert data are known as data reduction. The data
reduction may be done in the field or office. It should be done under the supervision of
the responsible engineer and it should be checked by someone other than the preparer to
reduce errors. Reduced data should be summarized in tabular form showing the date,
time, measurements, and comments. Spreadsheet type software is readily available and
can facilitate this step.

Reduced data should be reviewed for measurements that are significantly different from
previous measurements and for data exceeding threshold limits. Usually this step should
be taken within a few days of collecting the data. Any questionable measurements should
be retaken. If any of the data reach threshold limits, pre-planned actions established in
the written monitoring program should be taken.

9-48
9-7.3 Data Presentation

Plots facilitate screening of data and comparison with expected data. Plots are also useful
to summarize data. All reduced data should be summarized in graphical form. All plots
should include sufficient previous data to identify any long-term trends. Furthermore, the
plots should be self-explanatory. They should show the project name, type of instrument,
and what is being measured. If more than one set of data are on one plot, different
symbols or line types should be used to distinguish the data and a legend should be
provided. Scales should be consistent to allow comparison of data between plots and
they should be labeled. Threshold limits, scatter, and magnitude of significant changes
should be considered when selecting scales. Plan and section drawings showing the
number, location, and details of each instrument should be included with the plots.
Where practical, threshold limits should be shown on the plots. Plotting software is
available to facilitate this step.

The best type of plot depends on the purpose of the instrument(s) and should be selected
on a case-by-case basis. Generally data versus time plots are good for displaying
piezometric, seepage, and most movement data. Location versus movement graphs are
preferable for surface movement data and some internal movement data. Often more than
one type of plot is useful for evaluating data.

Factors that have significant influence on instrument data should be plotted or noted on
the data plots. For example, reservoir and tailwater levels should be included on all post-
construction piezometer plots, or they should be included as separate plots to the same
scale as the piezometer plots. Other factors that might be included on the plots are the
height of the dam during construction, daily temperature, rainfall, and seismic events.

9-7.4 Data Interpretation

Data should be reviewed for reasonableness, evidence of incorrectly functioning

instruments, and transposed data. Several checks for reasonableness can be made on all
data. The magnitude of data should be near the range of previous data. Data that are
significantly different may be incorrect. For example, water levels in piezometers should
not be above the reservoir level, except possibly during rapid drawdown or construction.
Data should be within the limits of the instrument. For example, data from open
standpipe piezometers must be below the top and above the bottom of the pipe. Open
standpipe readings at the top of the pipe are ambiguous because the phreatic surface
could be exactly at the top of the pipe, or it could be well above the pipe. The standpipe
must be raised or have a pressure gage added to it to clarify the reading. Whenever the

9-49
phreatic surface is above the top of a standpipe it may indicate a developing problem and
should be investigated.

It is important to distinguish between accuracy and precision when dealing with

measurements. Accuracy is the nearness to the true value. Precision is the degree of
refinement of the measurement. A measurement may be precise without being accurate
and vice-versa. For example, a pressure transducer may be capable of measuring water
depth to 1 millimeter, but the location of the transducer may be known to only several
tenths of a meter. Since attaining precision complicates data evaluation, the need for
precision and the level of precision should be carefully evaluated so unnecessary data are
not collected and/or costs are not increased.

All data will have scatter from instrument error, human error, and from changes in natural
phenomena such as temperature, wind, and humidity. The true accuracy of data will not
be apparent until a significant number of readings have been taken under a variety of
conditions.

All data will follow trends, such as decreasing with time or depth, increasing with time or
depth, seasonal fluctuation, direct variation with reservoir or tailwater level, direct
variation with temperature, or a combination of such trends. The trends are usually
evident in the plotted data. Statistical analysis of data may be useful in evaluating trends
that are obscured by scatter. However, such analyses are no substitute for judgment
based on experience and common sense. Data inconsistent with established trends should
be investigated. Readings deviating from established trends should be verified by more
frequent readings. Erroneous readings should be so noted on the original data sheets and
should be removed from summary tables and plots.

Constant measurements or widely varying measurements may indicate improperly

performing equipment. For example, a piezometer that reads a constant value and does
not change with headwater level, tailwater level, or season may not be functioning
properly.

Instruments that do not appear to be functioning properly should be further investigated.

For example, data should be checked against redundant data to determine whether or not
trends and magnitudes are the same. Accessible sensors or gages should be replaced to
see if the error remains. Calibration of the instruments should be checked. Often, tests
can be devised to evaluate proper functioning. For example, piezometers and observation
wells could be filled with water (or bailed out) and the rate at which the water returns to
its original level measured and compared to the results of similar tests done at the time of
installation, or expected behavior.

9-50
Improperly functioning instruments should be abandoned or replaced. Instruments that
are vital to the safety evaluation of a dam should be replaced. Instruments that provide
no meaningful information or that provide unnecessary redundancy should be abandoned.

Abandonment procedures should be evaluated on a case-by-case basis. They may consist

of a variety of actions such as:

• ceasing readings and maintenance;

• ceasing readings, but continuing minimal maintenance to keep the

instrumentation in a safe condition;

• plugging and sealing the instrument; and

• removing the instrument and repairing the hole.

If the abandoned instrument remains in place, it should be clearly marked as such to

avoid continued collection of data.

9-7.5 Dam Performance Evaluation

The purpose of instrumentation and monitoring is to maintain and improve dam safety.
The data should be used to evaluate whether the dam is performing as expected and
whether it provides a warning of developing conditions that could endanger the safety of
the dam.

The licensee's responsible engineer should evaluate dam performance for each set of data.
Additionally, during Part 12 Safety Inspections, the Independent Consultant will evaluate
the dam performance using the data.

All data should be compared with threshold levels established in the monitoring program
(Section 9-5.7). Trends of measurements toward threshold levels should be identified and
evaluated. If threshold levels will be reached within a short time, investigations and
remedial action should be implemented.

All data should be compared with expected behavior based on the basic engineering
concepts that were discussed in section 9-3. Variations from expected behavior may
suggest development of conditions that should be evaluated. For example, at a concrete
gravity dam, increasing uplift pressure, or decreasing drain flow may indicate that the
foundation drains may need to be cleaned.

9-51
All data should be compared with design assumptions. For example, measured pore
pressures and uplift pressures should be compared against those used in stability analyses.
If data are available for unusual load cases, such as rapid drawdown and floods, it should
be compared with assumed pressures.

More than one phreatic surface may exist where there are impervious strata in the
foundation. Piezometric data should be evaluated with geologic data to identify multiple
phreatic surfaces. If the phreatic surface for any strata is above the ground surface, the
stability of the dam should be evaluated using the elevated phreatic surface.

If no unusual behavior or evidence of problems is detected, the data should be filed for
future reference. If data deviates from expected behavior or design assumptions, action
should be taken. The action to be taken depends on the nature of the problem, and should
be determined on a case-by-case basis. Possible actions include:

• performing detailed visual inspection;

• repeating measurements to confirm behavior;

• reevaluating stability using new data;

• increasing frequency of measurements;

• installing additional instrumentation;

• designing and constructing remedial measures;

• operating the reservoir at a lower level; and

• emergency lowering of the reservoir.

9-7.6 Adequacy of Instrumentation and Monitoring

The last step should be to assess whether the instrumentation and monitoring program is
sufficient to evaluate if a dam is performing as expected and warn of developments that
could endanger the safety of the dam. The evaluation should include answers to the
following three questions.

1) Are the type, number, and location of instruments proper for the behavior being
monitored?

9-52
2) Is the frequency of readings appropriate?

3) Are the data being collected, processed, and evaluated in a timely and correct
manner?

The licensee's responsible engineer should evaluate the adequacy of instrumentation and
monitoring for each set of data. Staff will evaluate the adequacy of instrumentation
during annual Operation Inspections. Additionally, during Part 12 Safety Inspections, the
Independent Consultant will evaluate the adequacy of instrumentation and monitoring.

If there is a discrepancy between the measured and expected behavior of the dam, it may
indicate that data do not adequately represent the behavior of the dam, or that conditions
exist that were not accounted for in the expected behavior. In either case it is often useful
to perform field investigations and install additional instrumentation to evaluate the
behavior.

If trends or inter-relationships between data are not clear, it may be appropriate to take
more frequent measurements or collect additional complementary data.

9-53
If data are not being processed and evaluated in a timely and correct manner, personnel
involved in the instrumentation and monitoring program should be reminded, and further
trained if necessary, in the importance of each phase of the program and the potential
impacts with respect to dam safety. A dam safety program is inadequate if the
performance of a dam is not understood. Instrumentation provides the means for that
understanding.

9-8 Automated Data Acquisition

Automated data acquisition systems (ADAS) have evolved significantly over the last 10
years and are currently installed on over 40 dams in the United States. Most of the dams
are owned by federal agencies (USACE, USBR, and TVA). Design of successful ADAS
requires considerable effort. ADAS can range from the simple use of a datalogger to
collect data from a few instruments to computer based systems that collect, reduce,
present, and interpret data from a network of hundreds of different instruments. Most
types of water level, water pressure, seepage, leakage, stress, strain, and temperature
instrumentation can be readily monitored. Some types of instrumentation such as
movement surveys and probe inclinometers cannot be automated.

Advantages of automated data acquisition include: reduced manpower costs for collecting
data, remote collection of data, and data collection in electronic format suitable for
computer reduction, analysis, and printout. Rapid notification of potentially hazardous
performance and increased frequency of measurements can be taken on demand.

Disadvantages include high initial costs and complex equipment. By far the greatest
disadvantage is that visual observations normally made during routine manual data
collection will not be made.

Basic requirements of successful ADAS, according to USCOLD 1993, are listed below.

• Each instrument or sensor should maintain the ability to be read manually and
electrically prior to entering the automated network.

• There should be a redundant system for critical and very important instruments.

• The automated system should have a central network monitor station at the
project office to manage the field system and provide an external
communications interface. This station must also be able to collect, process,
validate, display, and produce hard copy of all data at the project.

9-54
 • A primary and backup external communications link for data transmission
should be provided.

• The system should have multiple, redundant systems of lightning protection

that isolate various components of the systems from the inevitable lightning
strike-induced power surge. Grounding and shielding of all cables is
mandatory.

• A primary and backup power source should be provided for uninterrupted data
collection (this is usually in the form of a rechargeable battery, together with
some form of reliable charging method such as solar panels).

• The systems should be protected from vandalism by protection inside a

structure or by burial in specifically designed units.

USCOLD (1993) discusses whether ADAS is appropriate; the design, installation, and
operation of ADAS; case histories; and provides a list of ADAS vendors. Other
references include ICOLD (1982) and USACE (1985, 1986b, 1986c, 1986d, 1987b, 1988,
1989b, and 1990).

9-9 References

ASCE (American Society of Civil Engineers), "Lessons from Dam Incidents, USA,"
1975.

ASCE, "Lessons from Dam Incidents, USA-II," 1988.

Dunnicliff, J., "Long-term Performance of Embankment Dam Instrumentation," Recent

Developments in Geotechnical Engineering for Hydro Projects, ASCE, 1981.

Dunnicliff, John, "Geotechnical Instrumentation for Monitoring Field Performance,"

National Cooperative Highway Research Program, Synthesis of Highway Practice 89,
Transportation Research Board, National Research Council, Washington D.C.,
April 1982.

Dunnicliff, John, "Geotechnical Instrumentation for Monitoring Field Performance," John

Wiley & Sons, 1988.

Dunnicliff, John, "Twenty-Five Steps to Successful Performance Monitoring of Dams,"

Hydro-Review, August, 1990.

9-55
EPRI (Electric Power Research Institute), "Inspection and Performance Evaluation of
Dams," EPRI AP-4714, September, 1986.

FHA (Federal Highway Administration), "Permanent Ground Anchors," FHWA-DP-68-1,

March, 1984

Hanna, T.H., "Foundations in Tension, Ground Anchors," Trans-Tech Publications and

McGraw-Hill Book Company, First Edition, 1982.

ICOLD (International Commission of Large Dams), BULLETIN 21: "General

Considerations Applicable to Instrumentation of Earth and Rockfill Dams,"
November 1969.

ICOLD BULLETIN 41: "Automated Observation for the Safety Control of Dams," 1982.

ICOLD BULLETIN 59: "Dam Safety Guidelines," 1987.

ICOLD BULLETIN 68: "Monitoring of Dams and Their Foundations, State of the Art,"
1989.

ICOLD BULLETIN 87: "Improvement of Existing Dam Monitoring, Recommendations

and Case Histories," 1992.

ICOLD BULLETIN 88: "Rock Foundations for Dams," 1993.

ISRM (International Society for Rock Mechanics), "Part 3 Monitoring," Rock

Characterization Testing and Methods, Pergamon Press, 1981.

Jansen, R. B., "Dams and Public Safety," Water Power and Resources Service (now U.S.
Bureau of Reclamation), 1980 (Revised Reprint, 1983).

Littlejohn, G.S. and Bruce, D.A., "Rock Anchors — State of the Art," Foundation
Publications, Ltd., England, 1976.

MESA (Mining Enforcement and Safety Administration), "Engineering and Design

Manual, Coal Refuse Disposal Facilities; Chapter IX, Monitoring, Instrumentation and
Maintenance," 1973.

National Research Council, "Safety of Existing Dams, Evaluation and Improvement,"

National Academy Press, Washington D.C., 1983.

9-56
Post-Tensioning Institute, "Recommendations for Prestressed Rock and Soil Anchors,"
Phoenix, Arizona, Second Edition, Fourth Printing, 1994.

Sherard, J. L., "Piezometers in Earth Dam Impervious Sections," Recent Developments in

Geotechnical Engineering for Hydro Projects, ASCE, 1981.

USACE (U.S. Army Corps of Engineers), "Instrumentation of Earth and Rock-fill Dams,
Part 1, Groundwater and Pore Pressure Observations," EM 1110-2-1908, August 31,
1971.

USACE "Instrumentation of Earth and Rock-fill Dams, Part 2, Earth Movement and
Pressure Measuring Devices," EM 1110-2-1908, November 19, 1976.

USACE, "Recommended Guidelines for Safety Inspection of Dams", National Program

of Inspection of Dams, Appendix D, 1977.

USACE, "Automated System for Monitoring Plumblines in Dams," REMR Technical

Note CS-ES-2.1, September 1985.

USACE, "Seepage Analysis and Control for Dams," EM 1110-2-1901, September 30,
1986a.

USACE, Waterways Experiment Station, "Instrumentation Automation for Concrete

Structures; Instrumentation Automation Techniques," Technical Report REMR-CS-5,
Report 1, 1986b.

USACE, Waterways Experiment Station, "Instrumentation Automation for Concrete

Structures; Automation Hardware and Retrofitting Techniques," Technical Report
REMR-CS-5, Report 2, 1986c.

USACE, Waterways Experiment Station, "Instrumentation Automation for Concrete

Structures; Available Data Collection and Reduction Software," Technical Report
REMR-CS-5, Report 3, 1986d.

USACE, "Method of Measuring the Tilt of Large Structures," REMR Technical Note CS-
ES-2.2, August 1987a.

USACE, "Methods of Automating the Collection of Instrumentation Data," REMR

Technical Note CS-ES-2.3, August 1987b.

9-57
USACE, "Instrumentation for Concrete Structures," EM 1110-2-4300, Change 1,
November 30, 1987c.

USACE, "Data Base for Automated Geotechnical Instrumentation," ETL 1110-2-316,

November 15, 1988.

USACE, "Earth and Rock-fill Dams-General Design and Construction Considerations,"

EM 1110-2-2300, Change 2, January 15, 1989a.

USACE, Waterways Experiment Station, "Instrumentation Automation for Concrete

Structures; Demonstration of Instrumentation Automation Techniques at Beaver Dam,
Eureka Springs, Arkansas," Technical Report REMR-CS-5, Report 4, April 1989b.

USACE, "Procedures for Automated Control and Monitoring Systems," ETL 1110-3-418,
January 30, 1990.

USBR (U.S. Bureau of Reclamation), "Design of Gravity Dams," 1976.

USBR, "Design of Arch Dams," 1977.

USBR, "Embankment Dam Instrumentation Manual," January 1987a.

USBR, "Concrete Dam Instrumentation Manual," October 1987b.

USBR, "Water Measurement Manual," Second Edition, Revised Reprint, 1984.

USBR, Design Standard No. 13, "Embankment Dams," Chapter 11, "Instrumentation,"
July 1, 1990.

USCOLD BULLETIN: "General Considerations Applicable to Performance Monitoring

of Dams," December 1986.

USCOLD BULLETIN: "General Guidelines and Current U.S. Practice in Automated

Performance Monitoring of Dams," May 1993.

Wilson, S.D., and Marsal, R.J., "Current Trends in Design and Construction of
Embankment Dams," ASCE, 1979.

9-58
9-10 APPENDICES

9-59
 APPENDIX IX-A

Examples of Minimum Instrumentation

 Appendix A
Examples of Minimum Instrumentation

Case histories of instrumentation are provided in Dunnicliff (1981), ICOLD (1989 and
1992), USCOLD (1986) and Wilson (1979). Illustrative examples are provided in
Dunnicliff (1988) and USBR (1990). Examples illustrating these instrumentation and
monitoring guidelines are given in the following appendices for an existing and proposed
embankment dam and an existing and proposed concrete dam.

Example 1. Minimum Instrumentation for an Existing Embankment Dam

Figure A-1 shows a plan and section of an existing 46-foot-high homogeneous

embankment dam. A small toe drain had been constructed previously in order to lower
the phreatic surface. The foundation is semi-pervious alluvium. There is no grout curtain
and no instrumentation was installed in the dam when it was constructed.

Minimum recommended instrumentation is shown on Figure A-1. The headwater and

tailwater levels are measured with a staff gage located on the separate spillway on the left
abutment (not shown in figure). Seepage from the toe drain is collected by a pipe and is
measured with a calibrated container.

Several observation wells had been installed in the embankment during the toe drain
construction. Since then, the phreatic surface has remained below the embankment
surface. Though installation of the observation wells would not be required by the
guidelines, monitoring of them provides useful information. Therefore, the existing
observation wells are used to monitor the phreatic surface and the performance of the toe
drain.

Well points were added at the toe of the embankment dam to evaluate seepage conditions.
Underseepage existed as a result of a more pervious layer in the foundation. Several
years after the well points were added, seepage was observed along the toe of the dam. A
weighted reverse filter was placed over the seepage area while plans for a new toe drain
were being developed.

Survey monuments and measuring points were established for a horizontal and vertical
movement survey of the crest of the dam. The survey monuments were established on
each abutment. The measurement points were located near the downstream edge of the
crest at approximately 250-foot intervals.

No foundation movement instrumentation was installed because the alluvium was judged

9-A-1
to be stronger than the embankment material, the factors of safety against sliding were
adequate, and there was no evidence of structural distress.

Example 2. Minimum Instrumentation for a Proposed Embankment Dam

Figure A-2 shows a plan and section of a proposed 180-foot-high embankment dam. The
dam is a zoned type with a central core, chimney and blanket drains, random fill shells,
and a semi-pervious upstream fill. The foundation is a horizontally interbedded
sandstone and siltstone with occasional clay seams overlain by alluvium. A single line
grout curtain is located beneath the core trench.

Minimum recommended instrumentation is shown on Figure A-2. The headwater and

tailwater levels will be measured with staff gages located on the separate spillway on the
left abutment (not shown in figure). Seepage from the chimney and blanket drain will be
collected by a pipe and measured with a portable flume.

Two transverse lines of instrumentation will be installed because of the length of the dam.
The designer anticipates that the core material will consolidate rapidly. An array of
pneumatic piezometers will be located in the core to verify this design assumption. These
piezometers will be used for construction control and will be abandoned after
consolidation of the core is complete.

Pneumatic piezometers will be located in the alluvium beneath the blanket drain to
monitor its operation. Several additional pneumatic piezometers will be located beneath
clay seams in the sandstone to monitor the pore pressures in the foundation. Two sensors
will be located in each piezometer for redundancy.

Survey monuments and measuring points will be established for a horizontal and vertical
movement survey of the crest and downstream slope of the dam. The survey monuments
will be established on each abutment. The measurement points will be located near the
downstream edge of the crest at approximately 250-foot intervals through the major
portion of the embankment and 500-foot intervals elsewhere.

The designer is concerned about potential movement of the dam along the clay seams in
the foundation, therefore, several inclinometers will be installed to monitor for this type
of movement. To avoid drilling through the blanket drain, in-place inclinometers will be
used. Shear strips will be installed adjacent to each in-place inclinometer for redundancy.

Example 3. Minimum Instrumentation for an Existing Concrete Gravity Dam

9-A-2
Figure A-3 shows a plan and section of an existing 88-foot-high concrete gravity dam.
The dam has a center overflow spillway and a gallery that extends through the tallest
sections. The foundation is horizontally interbedded sandstone and shale. A single-line
grout curtain and row of foundation drains extend into the foundation from the gallery.
No instrumentation was installed in the dam when it was constructed.

Minimum recommended instrumentation is shown on Figure A-3. The headwater and

tailwater levels are measured with staff gages located on the left side of the spillway.

Seepage from the internal formed drains and the foundation drains is collected by the
gutter along the gallery. Weirs were installed in the gutter adjacent to the gutter drain at
the center of the spillway.

Since a reduction in uplift pressure at the line of drains is needed to meet stability criteria,
standpipe piezometers were installed to verify the reduction. One piezometer was
installed midway between two drains in the maximum section near the center of the
spillway. At least one additional piezometer downstream of the line of drains would be
desirable. Installation of a downstream piezometer from the gallery would be difficult
because of the narrow confines of the gallery. Installation of a piezometer from the
surface of the spillway would leave it exposed to possible damage from spillway flows.
Therefore, a piezometer was installed from the downstream face in the adjacent non-
overflow section. Uplift pressures at the non-overflow section were judged to be
representative of the maximum section because the foundation geology of both sections is
similar. A second piezometer was installed midway between two drains in the non-
overflow section as a check on the assumption of similar uplift pressure between the two
sections and to provide some redundancy.

The borings for the piezometers were extended into the foundation to document
foundation conditions and obtain samples for laboratory testing. A horizontal shale seam,
found about 20 feet below the base of the dam, was judged to be a potential sliding plane.
Pressure measurements in the borings showed increased uplift pressures immediately
below the seam. Therefore, piezometers were installed immediately below the seam.

Survey monuments and measuring points were established for a horizontal alignment
survey of the crest of the dam. The survey monuments were established on each
abutment. The measurement points were located on either side of the spillway and at
approximately 100-foot intervals.

Crack meters were installed on the ceiling of the gallery to monitor potential movement in
the transverse direction. There were no cracks that were judged to be evidence of

9-A-3
structural distress so no additional crack meters were installed.

9-A-4
No foundation movement instrumentation was installed because even though the
foundation contains a shale seam, the factors of safety against sliding were adequate and
there was no evidence of structural distress.

Example 4. Minimum Instrumentation for a Proposed RCC Gravity Dam

Figure A-4 shows a plan and section of a proposed 120-foot-high RCC gravity dam. The
dam has a center overflow spillway and a gallery that extends through the tallest sections.
The foundation is massive granite. A single-line grout curtain and row of foundation
drains extend into the foundation from the gallery.

Minimum recommended instrumentation is shown on Figure A-4. The headwater and

tailwater levels will be measured with staff gages located on the left side of the spillway.

Seepage from the internal formed drains and the foundation drains will be collected by
the gutter along the gallery. Weirs will be installed in the gutter adjacent to the gutter
drain at the center of the spillway.

Since a reduction in uplift pressure at the line of drains is needed to meet stability criteria,
standpipe piezometers will be installed to verify the reduction. Two transverse lines of
two piezometers each will be located in the spillway. The granite foundation contains
several sub-horizontal exfoliation joints that could potentially be a path of seepage.
Piezometers will be installed in the foundation at the location of the exfoliation joints to
monitor pressure in the joints.

Survey monuments and measuring points will be installed for a horizontal alignment
survey and settlement survey of the crest of the dam. Survey monuments will be
established on each abutment and measurement points will be located on either side of
the spillway and at approximately 150-foot intervals.

Crack meters will be installed on any significant cracks to monitor potential vertical,
longitudinal, and transverse movement. No internal or foundation movement
instrumentation will be installed because the foundation is strong and competent. An
array of thermocouples will be installed to monitor the heat of hydration gradient during
construction.

9-A-5
9-A-6
9-A-7
9-A-8
9-A-9
 APPENDIX IX-B

Sample Data Presentation

 Appendix B
Sample Data Presentation

Sample data collection forms, tables, and plots are illustrated in this appendix to show
proper data presentation.

Instrument Plan

Figure B-1 is a plan of the dam showing instrument locations. There is a transverse line
of piezometers through the maximum section and several observation wells near the
abutments. Seepage is measured by a weir located approximately at the downstream toe
of the maximum section.

Figure B-2 shows a section through the embankment along the line of piezometers.
Pertinent embankment and foundation details are shown on the section.

Sample Data Collection Forms

Figure B-3 is a sample data collection form for the observation wells and piezometers.
The upper portion of the form is for pertinent and complementary data. The lower
portion is for the actual readings. Figure B-4 is a sample data collection form for the
seepage measurements.

Sample Data Reduction Spreadsheet and Summary Tables

Figure B-5 shows a spreadsheet that reduces the observation well and piezometer data,
checks the measurements against threshold levels and depth to the top and bottom of the
standpipes, and summarizes the data in tabular form. Threshold limits are based on
previous minimum and maximum readings.

The seepage weir data are reduced manually. Threshold limits for seepage data are based
on the range of the instrument.

Sample Data Plots

Figures B-6 through B-8 are plots of the observation well and piezometer data. The
instruments are grouped according to type and location. Embankment and alluvium
piezometer data are plotted on Figure B-6. Headwater level is plotted along with the
piezometer data. The piezometers respond to changes in headwater and piezometric
levels decrease with distance downstream. It was assumed there was no tailwater, so it

9-B-1
was not plotted. If there is tailwater, the licensee should plot it.

Embankment observation well data are plotted in Figure B-7. The data has slightly more
scatter that the piezometer data. OW-2 responds markedly to the heavy rainfall that
occurred in early May. Otherwise, the data are consistent with the piezometer data.

Foundation piezometer data are plotted in Figure B-8. The data respond to changes in
reservoir level and piezometric level in the sandstone decreases with distance
downstream.

Phreatic surface data for the highest reservoir level are plotted on Figure B-2. The
phreatic surface in the embankment and alluvium are essentially the same and are shown
by the long-dashed line. There is a distinct drop in the phreatic surface across the
concrete core wall. The sandstone is isolated from the embankment by the shale layer,
which acts as an aquiclude. The phreatic surface for the sandstone layer is shown by the
short-dashed line. It is not affected by the presence of the core wall.

The seepage data are plotted in Figure B-9. Seepage approximately follows headwater,
but is influenced by surface runoff from rainfall. Although rainfall was not plotted, it is
sometimes helpful to plot that data in order to understand changes in drainflows.

9-B-2
9-B-3
9-B-4
 SAM ADAMS DAM
OBSERVATION WELL AND PIEZOMETER DATA COLLECTION FORM
Date: Time: Personnel:
Weather:
Recent rainfall:
Headwater Elevation: feet Tailwater Elevation: feet
Visual Observations (unusual or abnormal conditions):

READINGS
INSTRUCTIONS: Measure depth to water from top of standpipe with water level meter. Record
depth to nearest 0.1 foot. Note under "COMMENTS" any difficulties in taking the measurements, need
for repair, maintenance, etc. Compare readings with threshold limits and previous readings. If readings
exceed threshold limits or are more than 3 feet different than previous readings, retake the readings to
confirm them.
PIEZOMETER/ THRESHOLD DEPTH TO COMMENTS
OBSERVATION READINGS, WATER,
WELL FEET FEET
(MIN/MAX)
OW-1 41.9/54.8
OW-2 38.9/53.6
P-1 34.4/43.4
P-2 44.1/54.3
P-3 43.7/55.3
P-4 20.8/28.0
P-5 30.0/35.9
P-6 30.6/36.5
P-7 0.3/7.4
P-8 9.3/12.4
FIGURE B-3. PIEZOMETER DATA COLLECTION FORM

9-B-5
 SAM ADAMS DAM
SEEPAGE WEIR DATA COLLECTION FORM
Date: Time: Personnel:
Weather:
Recent rainfall:
Headwater Elevation: feet Tailwater Elevation: feet
Visual Observations (unusual or abnormal conditions):

READINGS
INSTRUCTIONS: Measure depth of water upstream of weir from invert of weir notch. Record depth
to nearest 1/8 inch. Note under "COMMENTS" and difficulties in taking the measurements, need for
repair, maintenance, etc. Compare readings with threshold limits and previous readings. If readings
exceed threshold limits or are more than 2 inches different than previous readings, retake the readings to
confirm them.
WEIR THRESHOLD DEPTH OF COMMENTS
READINGS, WATER,
INCHES INCHES
(MIN/MAX)
W-1 0/6.0

FIGURE B-4. SEEPAGE DATA COLLECTION FORM

9-B-6
9-B-7
9-B-8
9-B-9
9-B-10
9-B-11
CHAPTER X

OTHER DAMS

October 1997
 Chapter X
Other Dams

Contents
Title Page

10-1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1

10-1.1 Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-1

10-1.2 Review Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-2
10-1.3 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-2
10-1.4 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-3

10-2 Buttress Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-4

10-2.1 Types of Buttress Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-4

10-2.2 Typical Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-6
10-2.3 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-7
10-2.3.1 Dead Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-7
10-2.3.2 External Hydrostatic Pressures . . . . . . . . . . . . . . . . . 10-7
10-2.3.3 Internal Hydrostatic Loads . . . . . . . . . . . . . . . . . . . . . 10-9
10-2.3.4 Earthquake Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-9
10-2.3.5 Ice Pressures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-11
10-2.3.6 Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-13
10-2.4 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-14
10-2.5 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-14
10-2.5.1 Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-15
10-2.5.2 Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-16
10-2.5.3 Analysis of Reinforced Concrete . . . . . . . . . . . . . . 10-18
10-2.5.4 Dynamic Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . 10-19
10-2.5.5 Post Earthquake Stability Analyses . . . . . . . . . . . . . 10-20
10-2.6 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-20
10-2.6.1 Static . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-20
10-2.6.2 Dynamic and Post Earthquake . . . . . . . . . . . . . . . . . 10-20
10-2.6.3 Foundation Stability . . . . . . . . . . . . . . . . . . . . . . . . 10-21
10-2.7 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-21
10-2.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-21

10-3 Concrete Dams on Pile Foundations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24

10-3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24

10-i
 10-0 Contents
(Cont.)
Title Page

10-3.2 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-24

10-3.3 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-25
10-3.4 Structure and Subsurface Conditions . . . . . . . . . . . . . . . . . . . . 10-25
10-3.5 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-26
10-3.5.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-26
10-3.5.2 Vertical Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-26
10-3.5.3 Lateral Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-26
10-3.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-33
10-3.7 Pile Analysis Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-35

10-4 Concrete Dams on Soil Foundations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-47

10-4.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-47

10-4.2 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-47
10-4.3 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-47
10-4.3.1 Bearing Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-48
10-4.3.2 Sliding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-48
10-4.3.3 Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-48
10-4.4 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-49
10-4.5 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-49
10-4.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-49

10-5 Timber Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-50

10-5.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-53

10-5.2 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-53
10-5.3 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-53
10-5.4 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-54
10-5.5 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-54
10-5.5.1 Timber Deterioration . . . . . . . . . . . . . . . . . . . . . . . . 10-54
10-5.5.2 Timber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-55
10-5.5.3 Rockfill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-56
10-5.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-56

10-ii
 10-0 Contents
(Cont.)
Title Page

10-6 Inflatable Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-57

10-6.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-59

10-6.1.1 Dead Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-59
10-6.1.2 External Hydrostatic Pressure . . . . . . . . . . . . . . . . . 10-59
10-6.1.3 Ice Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-60
10-6.1.4 Temperature Effects . . . . . . . . . . . . . . . . . . . . . . . . 10-60
10-6.2 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-60
10-6.3 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-60
10-6.3.1 Superstructure Analysis . . . . . . . . . . . . . . . . . . . . . . 10-60
10-6.3.2 Substructure Analysis . . . . . . . . . . . . . . . . . . . . . . . 10-61
10-6.4 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-61
10-6.4.1 Pool Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-61
10-6.4.2 Inflating Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-61
10-6.4.3 Superstructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-62
10-6.4.4 Substructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-63
10-6.5 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-64
10-6.5.1 Site Investigation . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-64
10-6.5.2 Tube . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-64
10-6.5.3 Inflation/Deflation System . . . . . . . . . . . . . . . . . . . . 10-65
10-6.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-67

10-7 Stone Masonry Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-68

10-7.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-68

10-7.2 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-69
10-7.3 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-69
10-7.4 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-69
10-7.5 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-69
10-7.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-70

10-8 Water Retaining Power Plant Structures . . . . . . . . . . . . . . . . . . . . . . . . 10-70

10-8.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-71

10-8.1.1 Dead Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-71
10-8.1.2 External Hydrostatic Loads . . . . . . . . . . . . . . . . . . . 10-71

10-iii
 10-0 Contents
(Cont.)
Title Page

10-8.1.3 Internal Hydrostatic Loads (Uplift) . . . . . . . . . . . . . 10-71

10-8.1.4 Compaction Residual Stresses . . . . . . . . . . . . . . . . . 10-71
10-8.2 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-72
10-8.3 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-72
10-8.3.1 Floatation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-72
10-8.4 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-73
10-8.4.1 Flotation Stability . . . . . . . . . . . . . . . . . . . . . . . . . . 10-73
10-8.5 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-74
10-8.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-74

10-9 Cellular Sheet Pile Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-75

10-9.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-76

10-9.1.1 Dead Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-76
10-9.1.2 Internal Hydrostatic Loads . . . . . . . . . . . . . . . . . . . . 10-76
10-9.1.3 External Earth Pressures . . . . . . . . . . . . . . . . . . . . . 10-79
10-9.1.4 Berm Pressures . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-79
10-9.1.5 Earthquake Forces . . . . . . . . . . . . . . . . . . . . . . . . . . 10-79
10-9.1.6 Surcharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-79
10-9.2 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-79
10-9.2.1 Case I Usual Loading Combination . . . . . . . . . . . . . 10-80
10-9.2.2 Case II Unusual Loading Combination . . . . . . . . . . 10-80
10-9.2.3 Case IV End of Construction Combination . . . . . . . 10-80
10-9.3 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-81
10-9.3.1 External Cell Stability . . . . . . . . . . . . . . . . . . . . . . . 10-81
10-9.3.1.1 Sliding . . . . . . . . . . . . . . . . . . . . . . . . . 10-81
10-9.3.1.2 Overturning . . . . . . . . . . . . . . . . . . . . . 10-82
10-9.3.1.3 Rotation . . . . . . . . . . . . . . . . . . . . . . . . 10-82
10-9.3.1.4 Bearing Capacity . . . . . . . . . . . . . . . . . 10-82
10-9.3.1.5 Settlement . . . . . . . . . . . . . . . . . . . . . . 10-83
10-9.3.1.6 Seepage . . . . . . . . . . . . . . . . . . . . . . . . 10-83
10-9.3.1.7 Scour . . . . . . . . . . . . . . . . . . . . . . . . . . 10-84

10-9.3.2 Internal Cell Stability . . . . . . . . . . . . . . . . . . . . . . . 10-85

10-9.3.2.1 Interlock Tension . . . . . . . . . . . . . . . . . 10-85
10-9.3.2.2 Tilting . . . . . . . . . . . . . . . . . . . . . . . . . 10-86

10-iv
 10-0 Contents
(Cont.)
Title Page

10-9.3.2.3 Pullout of Outboard Sheeting . . . . . . . . 10-85

10-9.3.2.4 Penetration of Inboard Sheets . . . . . . . . 10-86
10-9.3.3 Dynamic Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 10-86
10-9.3.4 Finite Element Analysis . . . . . . . . . . . . . . . . . . . . . . 10-86
10-9.4 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-87
10-9.4.1 Factors of Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-87
10-9.4.2 Cell Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-88
10-9.5 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-88
10-9.5.1 Foundation Properties . . . . . . . . . . . . . . . . . . . . . . . 10-89
10-9.5.2 Cell Fill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-89
10-9.5.3 Steel Sheet Piling . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-89
10-9.6 Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-90
10-9.7 Construction Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-91
10-9.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-92

10-10 Mechanically Stabilized Earth Dams . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-93

10-10.1 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-99

10-10.1.1 Dead Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-99
10-10.1.2 External Hydrostatic Loads . . . . . . . . . . . . . . . . . . . 10-99
10-10.1.3 Earth Pressures . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-99
10-10.1.4 Ice Pressures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-99
10-10.2 Loading Combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-100
10-10.3 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-100
10-10.3.1 External Stability . . . . . . . . . . . . . . . . . . . . . . . . . . 10-101
10-10.3.1.1 Sliding Resistance . . . . . . . . . . . . . . . 10-101
10-10.3.1.2 Overturning . . . . . . . . . . . . . . . . . . . . 10-102
10-10.3.1.3 Bearing Capacity . . . . . . . . . . . . . . . . 10-102
10-10.3.1.4 Seismic Stability . . . . . . . . . . . . . . . . . 10-102
10-10.3.2 Internal Stability . . . . . . . . . . . . . . . . . . . . . . . . . . 10-102
10-10.3.2.1 Tension Failure . . . . . . . . . . . . . . . . . 10-103
10-10.3.2.2 Pullout . . . . . . . . . . . . . . . . . . . . . . . . 10-104
10-10.3.2.3 Seismic Loading . . . . . . . . . . . . . . . . . 10-105
10-10.3.2.4 Facing Strength . . . . . . . . . . . . . . . . . 10-105
10-10.3.3 Back-to-Back Wall Design . . . . . . . . . . . . . . . . . . 10-105
10-10.3.4 Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-106

10-v
 10-0 Contents
(Cont.)
Title Page

10-10.3.5 Seepage . . . . . . . . . . . . . . . . . . . . . . . . . . 10-107

10-10.3.6 Computer Programs . . . . . . . . . . . . . . . . 10-107
10-10.4 Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-107
10-10.4.1 External/Internal Stability . . . . . . . . . . . . 10-107
10-10.4.2 Steel Reinforcement . . . . . . . . . . . . . . . . 10-107
10-10.4.3 Geosynthetic Reinforcement . . . . . . . . . . 10-108
10-10.4.4 Backfill . . . . . . . . . . . . . . . . . . . . . . . . . . 10-110
10-10.4.5 Seepage Control . . . . . . . . . . . . . . . . . . . 10-110
10-10.4.6 Corrosion/Deterioration . . . . . . . . . . . . . 10-111
10-10.5 Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-112
10-10.5.1 Site Investigation . . . . . . . . . . . . . . . . . . . 10-112
10-10.5.2 Reinforcement . . . . . . . . . . . . . . . . . . . . . 10-112
10-10.5.3 Corrosion Protection . . . . . . . . . . . . . . . . 10-113
10-10.5.4 Facing . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-114
10-10.6 Construction Considerations . . . . . . . . . . . . . . . . . . . . 10-114
10-10.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-114

List of Tables

Table 10-9.1 Recommended Minimum Factors of Safety . . . . . . . . . . 10-87

Table 10-10.1 Summary of Reinforcement and Face Panel
Details for Various MSE Systems . . . . . . . . . . . . . . . . . 10-96
Table 10-10.2 Comparison of MSE Systems . . . . . . . . . . . . . . . . . . . . 10-97
Table 10-10.3 Recommended Minimum Factors of Safety . . . . . . . . . 10-109

List of Figures

Figure 10-2.1 Angle of Contact Force Between Slab and Buttress . . . . . 10-7
Figure 10-2.2 Hydrodynamic Pressures on Buttress Dams . . . . . . . . . . . 10-8
Figure 10-2.3 Entrapped Air In Buttress Dams . . . . . . . . . . . . . . . . . . . 10-8
Figure 10-2.4 Increased Pressure Coefficients For Constant
Sloping Faces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-10
Figure 10-2.5 Initial Interaction Between Ice and Sloping Structure . . . 10-12
Figure 10-2.6 General Interaction Between Ice and Sloping Structure . 10-12
Figure 10-2.7 C1 Vs. Slope Angle and Friction . . . . . . . . . . . . . . . . . . 10-13
Figure 10-2.8 C2 Vs. Slope Angle and Friction . . . . . . . . . . . . . . . . . . 10-13

10-vi
Figure 10-2.9 Uplift Pressure Diagram for Buttress Dams . . . . . . . . . . 10-16
Figure 10-2.10 Unit Arch Strip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-17
Figure 10-3.1 Effects of Pile Top Fixity . . . . . . . . . . . . . . . . . . . . . . . 10-27
Figure 10-3.2 Typical Pile Nomenclature . . . . . . . . . . . . . . . . . . . . . . 10-28
Figure 10-3.3 Typical Soil Load-Deflection (p-y) Curves . . . . . . . . . . 10-30
Figure 10-3.4 Limit States for Ultimate Analyses . . . . . . . . . . . . . . . . 10-32
Figure 10-3.5 Ambursen Dam Pile Foundation . . . . . . . . . . . . . . . . . . 10-35
Figure 10-3.6 Force Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-36
Figure 10-3.7 Pile Top Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-40
Figure 10-3.8 Force Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-41
Figure 10-5.1 Typical Timber Crib Dam . . . . . . . . . . . . . . . . . . . . . . . 10-51
Figure 10-5.2 Typical Timber Buttress Dam . . . . . . . . . . . . . . . . . . . . 10-51
Figure 10-6.1 Inflatable Dam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-58
Figure 10-9.1 Typical arrangement of circular, diaphragm, and
cloverleaf cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-78

10-vii
10-1 Introduction

10-1.1 Purpose and Scope

The guidelines presented in this chapter provide staff engineers with recommended
procedures and criteria to be used in reviewing and evaluating the safety of the following
types of dams:

• buttress dams;

• concrete dams on pile foundations;

• concrete dams on soil foundations;

• timber dams;

• inflatable dams;

• stone masonry dams;

• water retaining power plant structures;

• cellular sheet pile structures; and

• mechanically stabilized earth dams;

The staff engineers' review will be conducted to ensure that all decisions, methods, and
procedures performed by licensees/exemptees, or their consultants, are sound regarding
dam safety, and to ensure that the Commission's Dam Safety Program objectives as stated
in Part 12 of the Commission's Regulations are consistent with accepted up-to-date
procedures (the term licensees also refers to exemptees or applicants for license or
exemptees from license where appropriate).

Safety evaluation of both new and existing dams presents special and unique problems.
Existing dams may prove difficult to analyze especially in those instances where the dam
was designed before the development of modern design and construction technology or
where adequate records of design, construction, and material properties are not available,
such as for timber dams, masonry dams, and piles. Existing dams should be viewed in
light of knowledge of studies and reports on similar dams of the same vintage to gain an
understanding of probable design and construction methods.

10-1
The objective set forth in this chapter is to provide systematic procedures for performing
staff evaluations. It is not intended to generate any new philosophy or theories on the
methods used to analyze these structures. Existing Bureau of Reclamation and Corps of
Engineers literature as well as other readily available references have been used.

10-1.2 Review Procedures

Staff review of analyses performed by licensees or their consultants should concentrate

on the justification of the assumptions used in the analysis. The basis for critical
assumptions such as allowable stresses, shear strengths, seepage control, and loading
conditions should be carefully examined. The consultant's reports, exhibits, and
supplemental information must provide justification for these assumptions by way of
foundation exploration and testing, material testing, instrumentation data, and records
maintained during the actual construction of the project. Methods of analysis should
conform to conventional procedures and standards used in the engineering profession.

As mentioned in the preface to the Engineering Guidelines, considerable engineering

judgement must be exercised by staff when evaluating procedures or situations not
specifically covered herein. Unique problems or unusual solutions may require
deviations from the criteria and/or procedures outlined in this chapter. In these cases,
such deviations must be evaluated on an individual basis in accordance with Chapter I,
Section 1-4.

10-1.3 Forces

Forces to be considered in the design of the various dam structures included in this
Chapter are discussed in detail in Chapter III - Gravity Dams of these Guidelines.

Forces generally consist of:

• dead loads (including post-tensioning);

• external hydrostatic loads;

• internal hydrostatic loads;

• earth and silt loads;

• earthquake loads;

• ice loads;

10-2
 • wind loads; and

• temperature loads.

Many of the forces that must be considered in the design of the dam structure are of such
a nature that an exact determination cannot be made. The intensity, direction, and
location of these forces must be estimated by the designer after consideration of all
available facts and, to a certain extent, must be based on judgement and experience.

Only exceptions, clarifications, and additions are included in the following sections.

10-1.4 Loading Combinations

Loading combinations and requirements suitable in general for gravity type dams are
discussed in detail in Chapter III of these Guidelines. These include:

• Case I Usual Loading Combination - Normal Operating Condition.

• Case II Unusual Loading Combination - Flood Discharge.

• Case IIA Unusual Loading Combination - Ice.

• Case III Extreme Loading Combination - Normal Operating with MCE.

• Case IIIA Extreme Loading Combination (Unconstructed Dams Only) -

Construction Condition with Earthquake.

Where a new dam is founded on compressible clay, a Case IV should be added.

• Case IV End of Construction Combination (Unconstructed Dams Only) -

Dams founded on soil, or rock with clay layers.

For embankment type dams, loading combinations are discussed in detail in Chapter IV -
Embankment Dams of these Guidelines. These include:

• end of construction;

• sudden drawdown;

• partial pool with steady seepage;

10-3
 • steady seepage;

• earthquake; and

• appropriate flood surcharge pool.

Only exceptions, clarifications, and additions are included in the following sections.

10-2 Buttress Dams

Concrete buttress dams consist of a sloping upstream face supported by a series of

triangularly shaped buttresses. Buttress dams evolved from concrete gravity dams. Since
concrete gravity dams require only about 25 to 35 percent of the concrete to safely carry
stresses to the foundation, buttress dams eliminate the extra concrete and at the same time
eliminate most of the uplift pressures. If there is a base slab that spans between the
buttresses, the uplift is not eliminated. Stability is attained by sloping the upstream face
so that the weight of water acts as a stabilizing load. The design of buttress dams is
discussed in several references (Boggs, Jansen, and Tarbox 1988; Burroughs 1969;
Copen, Lindholm, and Tarbox 1977; Corns, Tarbox, and Schrader 1988; Creager 1917;
Creager and Justin 1927; Creager, Justin, and Hinds 1945; Davis 1969a, b, c; Houk and
Wengler 1969; Legas 1988; Marcello 1969; Thomas 1976; Wegeman 1927).

Buttress dams are economical when 1) labor and forming are inexpensive, 2) concrete is
expensive, 3) materials need to be hauled to a remote site, or 4) construction water is
scarce. Since construction of new buttress dams is generally not economical in modern
times, this section focuses on evaluation of existing buttress dams.

10-2.1 Types of Buttress Dams

The most common types of buttress dams are discussed below. Many different names
have been used for various types of buttress dams and some of them are listed in
parentheses to help reduce confusion when consulting the literature.

Flat Slab Dam (a.k.a. Ambursen Dam, Deck Dam). A buttress dam in which the
upstream face is a relatively thin flat slab usually made of reinforced concrete. Typically,
the slabs are simply supported and are not integral with the buttresses making the
structure relatively flexible. As a result, ordinary foundation movements have little effect
on stress distributions. A compressible material and waterstop are normally provided

10-4
between the slab and the corbel of the buttresses to allow movement yet remain
watertight. The slab is continuous in some designs, making the structure more rigid.

Multiple Arch Dam. A buttress dam in which the upstream face is a series of arches
spanning between buttresses. The arches are generally semi-circular with central angles
between 100 to 180 degrees, though non-circular arches have been used. The arches may
be unreinforced or reinforced. The arches may be integral with the buttresses, making the
structure rigid and susceptible to damage from even small foundation movements.
Alternatively, the arches may be structurally independent of the buttresses, making the
structure somewhat flexible. One variation that has been used is a multiple dome buttress
dam, in which the arch is curved both in plan and section.

Hollow Gravity Dam. In modern usage, a dam that has the external appearance of a
gravity dam, but which has large open areas inside. These are usually buttress dams in
which the downstream portion is covered with a reinforced concrete slab. They are often
used for overflow spillway sections of buttress dams or as housing for a powerhouse.

Massive Head Buttress Dam (a.k.a. Round Head, Diamond Head, Massive
Buttress, Cored Gravity, Hollow Gravity [archaic definition], Tee Head, Hammer
Head). A buttress dam in which the upstream end of each buttress (head) is enlarged to
span the gap between buttresses. Although common in Europe, there are very few
massive head buttress dams in the United States and none under FERC jurisdiction.
Therefore, the guidance in this section applies specifically to slab and buttress or multiple
arch buttress dams and not necessarily massive head buttress dams.

There are many variations within the basic types of buttress dams that can have
significant influences on the structural behavior of the dams. Therefore, the structural
details of each dam must be established before the safety of the dam can be properly
evaluated. Typical buttress dam variations include:

• thin upstream faces with thin, closely-spaced buttresses to massive

upstream faces, with massive, widely-spaced buttresses;

• completely unreinforced upstream faces and buttresses to heavily

reinforced upstream faces and buttresses;

• constant thickness upstream faces and buttresses to upstream faces and

buttresses that increase in thickness with depth;

10-5
 • upstream faces structurally independent of the buttresses to upstream
faces integral with the buttresses;

• straight upstream faces to upstream faces that have a steeper slope near
the top;

• single wall buttresses to double wall buttresses with internal bracing;

• laterally unbraced buttresses to buttresses braced with pilasters, flanges,

struts, and shear walls;

• monolithic buttresses to buttresses with inclined or longitudinal joints

that separate the buttress into a series of inclined columns; and

• buttresses founded directly on rock to buttresses founded on spread

footings or continuous slabs.

10-2.2 Typical Problems

Many buttress dams were designed and built over 50 years ago using outdated methods,
loads, and acceptance criteria. As a result, a number of these structures have been, or
will need to be, modified to overcome inadequate stability or excessive stresses. The
most critical areas tend to be concrete deterioration from freeze-thaw damage, poor
quality concrete, inadequate design, and inadequate seismic stability. Case histories of
problems and resolutions to the problems are common in the literature (Garland, Waters,
Focht, and Rutledge 1995; Rohde and Zuccolotto 1995; Reynolds, Joyet, and Curtis
1993; Niziol and Paolini 1993; Lamar, Ivarson, and Tenke-White 1991; Wheelock and
Wilkins 1991; and Bengtson 1989).

Freeze-thaw damage is common in cold climates because the upstream face is generally
saturated and the downstream side of the upstream face is exposed to seasonal
temperatures. Non air-entrained concrete was commonly used in older buttress dams and
contributes to freeze-thaw damage. Severe freeze-thaw damage can jeopardize the safety
of the dam because of the relative thinness of the upstream faces.

The strength of the concrete has been found to be quite low (less than 2,000 psi) and
variable in some buttress dams. These conditions have been attributed to insufficient
cement, excessive water, and poor quality control during construction. Such low and
variable strengths significantly impact buttress dam safety because the concrete is
relatively highly stressed.

10-6
Thin, unreinforced, upstream faces and buttresses cannot tolerate significant deflection or
deterioration and are, therefore, highly susceptible to damage from a variety of common
conditions.

Multiple arch buttress dams, in which central angles of individual arches are less than 180
degrees, impose significant lateral load on buttresses. This load is resisted in part by
adjoining arches. As a result, failure of one arch could cause catastrophic failure of the
entire dam.

Buttresses, especially unreinforced or unbraced buttresses, are susceptible to damage

from lateral earthquake loading.

10-2.3 Forces

Forces for the analysis of buttress dams are generally the same as for concrete gravity
dams, which are discussed in Chapter III, Section 3-2 of these Guidelines. Only
exceptions, clarifications, and additions are discussed below.

10-2.3.1 Dead Loads

Dead loads include the weight of the upstream and downstream face and the weight of the
buttresses. Dead load is carried by the buttresses regardless of the connection detail,
because the normal pressure on the sloping contact times any reasonable friction angle is
sufficient to prohibit sliding of the slab
with respect to the buttress. Consider
the example shown in Figure 10-2.1.
Note that the angle of the contact force
between the slab and the buttress is
less than 14o. This means the contact
force alone will keep the slab pinned
against the buttress.

10-7
10-2.3.2 External Hydrostatic Pressures

The effects of hydrodynamic pressures on a slab and buttress dam must be considered
with more care than one would typically apply when considering a gravity dam. As water
flows over the surface of an ogee crest, hydrodynamic forces are produced which can be
much greater than the depth of flow, or which can exert a negative pressure which tends
to lift the slab off of the buttresses (Figure 10-2.2). Bucket pressures can approach the
value of the jet stagnation pressure, and negative pressures near the crest of the dam can
be sub-atmospheric. While these forces pose no problem to the mass concrete of a
gravity dam,
the thin slabs of
a slab and
buttress dam
may not have
been sufficie
ntly reinfor
ced to resist
the large hydrod
ynamic pressur
es associa
ted with flood
loading. In
addition, since
these dams
are hollow,
proper air venting of the interior of the dam must be assured. During a flood event with a
quickly rising tailwater, an air pocket can be formed within the dam that will exert
upward pressure on the dam (Figure 10-2.3). This also can cause the slabs to be pulled
off the buttresses, and this upward load has a detrimental effect on overall stability.

10-8
10-2.3.3 Internal Hydrostatic Loads

If a buttress dam is founded on a continuous slab, uplift pressures at the concrete-to-rock

interface should be treated as discussed for concrete gravity dams in Chapter III, Section
3-2.4 of these Guidelines.

If the buttress
es are founded
directly on rock or
on spread footings
, the uplift at the
concrete- to-rock
interface will be
substantiall y
reduced because
the open areas
between the
buttresses provide
drainage. For
these cases,

10-9
uplift may be assumed to vary from headwater pressure at the upstream face to tailwater
pressure at the downstream edge of the upstream face slab or arch. Uplift pressure
beneath the remaining portion of the buttress or buttress footing may be assumed to be
tailwater pressure.

When examining failure planes within the foundation, the uplift pressures should be
treated similar to uplift pressures for concrete gravity dams. If shallow, subhorizontal
discontinuities exist, uplift pressures should be calculated using cracked base type
analyses.

10-2.3.4 Earthquake Forces

Earthquake forces are the same as for concrete gravity dams. Separate analyses should be
run with the horizontal accelerations applied in the transverse (upstream-downstream)
and longitudinal (cross-valley) directions. Vertical acceleration should also be
considered. For application of the seismic coefficient method to buttress dams, the
upstream slope of the dam effects the hydrodynamic forces resulting from an earthquake.
Zangar (1952) provides curves for determination of increased pressure for various
upstream slopes.

Figure 10-2.4 is reproduced from Engineering Monograph No. 11, US Bureau of

Reclamation, Hydrodynamic Pressures on Dams Due to Earthquake Effects, by Zangar,
May 1952. This figure provides the maximum value of coefficient C and the values of
the coefficient at the base of dams for various inclinations of the upstream face from the
vertical, as defined on Figure 10-2-4.

10-10
 Fo
r
da
ms
wit
h
co
nst
ant upstream slopes, the increase in pressure due to a horizontal earthquake may be
estimated, according to Zangar (1952), from:

Pe ' ½%wh Cm ” y/h(2&y/h) % y/h (2&y/h) ]

Where:

Pe is the increase in pressure in pound per square foot,

% is the horizontal earthquake intensity,

w is unit weight of water, pounds per cubic feet,

h is the height of dam above the base in feet,

y is the depth at which the pressure increase is being determined, and

Cm is the maximum value of C obtained from Figure 10-2.4.

10-11
At the base of dam, y = h, and the expression for Pb (Pe at the base) becomes:

Pb = % w h Cm

And the total horizontal force, in pounds per foot, above the base is:

Vb = 0.726 Pb h

And the total over-turning moment, Mb, in foot-pound per foot, is:

Mb = 0.299 Pb h2.

10-2.3.5 Ice Pressures

The sloping upstream faces of the buttress dams tend to cause ice to fail by bending
rather than crushing and as a result the magnitude of the ice forces are significantly
reduced. A study of ice forces on sloping structures shows that ice forces depend on the
ice strength, ice thickness, coefficient of friction between ice and concrete, and
inclination of the slope as illustrated by Figures 10-2.5 and 10-2.6. Reduced ice forces
can be computed according to the simple two-dimensional theory presented by Croasdale
(1980):
H D g t 5 1/4
'Ff( w ) C1 % Z t Di g C2
b E

where:

H = horizontal force
b = width of structure
Ff = flexural strength of ice
Dw = density of water
g = force of gravity
t = ice thickness
E = elastic modulus of ice
Z = height of ice on slope
Di = density of ice

where, C1 and C2 are functions only of the slope angle ("), and the coefficient of friction
between ice and concrete (µ). Values for the coefficients C 1 and C2 are plotted in Figures
10-2.7 and 10-2.8 and for typical values of " and µ.

10-12
Typical properties of fresh water ice are listed below (Rice 1975):

Modulus of elasticity, E:
Strain rate (s-1) E, psi
1 x 10-8-6 400,000
1 x 10-4 700,000
1 x 10 1,000,000
Poissons ratio: 0.33
Tensile strength: 30 to 150 psi
Compressive strength:
Strain rate (s-1) Max. Stress (psi)
1 x 10-4-7 1,400
3 x 10 50
Shear strength: 50 to 150 psi
In the simple two-dimensional theory given above, the first term can be considered to be
the force necessary to break the ice, and the second term can be considered to be the
force necessary to push the ice pieces up the sloping structure. As a 2-D theory it might
be considered accurate for a very wide structure, but it is probably inaccurate for narrow
structures.

10-13
For example, let:

Ff = 700 kPa
Dwg = 9.8 kN/m3
t = 0.6 m
E = 7 x 106 kPa
Z = 1.5 m
Di g = 9.0 kN/m3
% = 45 degrees
F = 0.3

then, C1 = 1.25, C2 = 2.6

H 9.8(0.6)5 1/4 kN
' 700( ) 1.25 % 1.5(0.6)(9.0)(2.6) ' 16 % 21 ' 37
b 6 m
7x10

10-2.3.6 Temperature

Temperature effects are not significant for many buttress dams, because the relatively
flexible nature of buttress dams allows thermal stresses to be relieved through slight
deformations. Temperature effects need be considered only if the dam is not reinforced,
or there is physical evidence of overstressing from temperature loads.

10-14
Excessive thermal stresses will usually be apparent from cracking at the connection
between the buttress and the upstream face slab or arch. Thermal expansion of a buttress
tends to cause horizontal cracks in the upstream face slab or arch. Thermal expansion of
an upstream face slab or arch tends to cause cracks in the buttress perpendicular to the
upstream face.

10-2.4 Loading Combinations

Loading combinations for the analysis of buttress dams are the same as for concrete
gravity dams discussed in Chapter III, Section 3-3 of these Guidelines, except that the
need to include temperature effects should be evaluated. Where temperature effects may
be significant, temperature changes should be included in the load cases in the same
manner as is done for arch dams (USBR 1977a and b), except summer ambient on
downstream face can be reduced because of shading between buttresses.

10-2.5 Analyses

Analysis methods commonly used in evaluating buttress dams are discussed in this
section. Simplified methods such as the gravity method and unit width method or the
two-dimensional finite element method are generally sufficient for most structures. More
sophisticated methods such as three-dimensional finite element analyses may be justified
when simpler analyses indicate unsatisfactory behavior.

Depending on the details of design, a buttress arch or slab and buttress dam may more
closely resemble a building. Method of frame analysis used for buildings may be
appropriate. The strength and stability on individual components may be more critical
than the strength and stability of the dam as a whole.

Typical steps for the evaluation of buttress dams are:

1) Review available project drawings, reports, and data.

2) Inspect the dam for evidence of deterioration, cracking, and leakage.

3) Inspect the dam for offsets between members that could indicate movement.

4) Establish the geometry, loads, and make conservative assumptions regarding the
material properties.

5) Perform stability analysis using the gravity method.

6) Perform stress analyses using the gravity and unit width methods or finite element
method.

10-15
7) If appropriate, perform a dynamic analysis.

8) If the analysis indicates that the structure does not meet safety criteria, or if the
analytical results do not appear to be consistent with the historical behavior of the
dam, then develop more elaborate field and laboratory investigations to refine
assumptions.

10-2.5.1 Stability

Stability analyses of buttress dams are the same as for concrete gravity dams discussed in
Chapter III, Section 3-4 of these Guidelines, with only minor differences.

The stability of representative buttresses should be evaluated with loads from contributing
portions of the upstream face, rather than evaluating a unit thickness as for gravity dams.

Buttresses with longitudinal joints (or cracks) may not act monolithically and stresses for
each column formed by the joints (or cracks) should be analyzed separately.

Stability should be evaluated for potential failure planes within the structure (commonly
along lift lines), at the base of the structure, and within the foundation. Foundation
failure planes are often more critical for buttress dams than for gravity dams, because
buttress dams are lighter than gravity dams and thus mobilize less frictional resistance to
sliding for the same height of dam. Care should be taken to identify all potential failure
planes in the foundation.

The cracked base type of analysis is not applicable for buttress dams, unless they are
founded on continuous slabs or have shallow, subhorizontal discontinuities in the
foundation (Figure 10-2.9). This is because uplift is at tailwater level all around the
buttresses if the buttresses are founded directly on the foundation. If there is a base slab
or a horizontal joint in the foundation rock directly below the dam, the uplift distribution
then resembles that of a standard gravity dam. (Refer to Chapter III.)

10-16
10-2.5.2 Stress

Stress analyses of buttress dams should be more detailed than for concrete gravity dams,
because stresses are higher and often control the design. Principal stresses and maximum
shear stresses must be evaluated in all portions of the structure including the upstream
face, buttresses, corbels, upstream face footings, and lateral supports.

In many buttress dams, tensile stresses occur in the buttresses, near the upstream face,
due to the geometry of the buttresses and the applied loads.

Damage to the connection between the upstream face and the buttresses from
overstressing or inadequate reinforcing is common. Shear and tensile stresses at the
connections must be evaluated.

If buttresses have longitudinal joints (or cracks), the buttress may not act monolithically
and each column separated by the joints (or cracks) should be evaluated independently.

Gravity Method. The gravity method is generally sufficient for calculation of stresses in
most buttresses. The basis of the gravity method is that vertical stresses on horizontal
planes have a linear variation giving a trapezoidal distribution. This assumption, often
referred to as the "trapezoidal law", greatly simplifies the analysis of principal and shear
stresses and is applicable for buttresses less than about 200 to 250 feet high (Marcello
1969; Davis 1969c).

10-17
Once vertical and shear stresses on horizontal planes are found by the trapezoidal law,
principal and maximum shear stresses are calculated by balancing forces on elementary
sections and computing the stresses on the incremental areas of the sections. The method
is described in various references (Burroughs 1969; Copen, Lindholm, and Tarbox 1977;
Creager, Justin, and Hinds 1945; Corn, Tarbox, and Schrader 1988; Davis 1969b;
Marcello 1969). Because the method involves small differences in large numbers, the
thickness of the horizontal slices must necessarily be small; typically, between 1 and 10
feet have been used. The smaller spacing provides more accurate results and should be
used where stresses are critical, as is often the case just under the upstream haunch.

Unit Width Method. In most cases, the stresses in the upstream face slab or arch can
be adequately evaluated by analyzing an independent unit strip of the slab or arch
spanning between the buttresses as shown in Figure 10-2.10. The pressure applied to a
unit strip by the reservoir is not uniform because the upstream face is inclined. Unit
strips should be evaluated at the maximum depth of a given thickness or reinforcement
pattern. Flat slabs should be evaluated as either simply supported or continuous beams or
slabs. If this type of analysis shows that an arch or slab strip of unit width can resist all
applied loads, then no more rigorous analysis needs to be done.

Finite Element Method. The finite element method is discussed in Chapter III, Section
3-4 of these Guidelines. Two-dimensional finite element models can often be used to
evaluate the upstream face and buttresses separately. Three-dimensional finite element
models may be justified where simpler methods do not resolve all concerns or where
seismic loading is large.

10-18
For evaluation of reinforced concrete structures, the moments, shears, and axial forces of
structural members must be calculated by integrating the stresses computed by the finite
element model. The resultant loads in the member must be compared to the capacity of
the member to evaluate its adequacy.

For most static three-dimensional analyses, there is seldom a need to analyze more than
one bay of buttress dams, because cross valley distribution of load typically does not
occur and it should not be relied upon if it does occur.

The finite element method is the only practical method available for evaluating the
dynamic response of buttress dams. Although two-dimensional models have been used to
calculate the dynamic response of individual buttresses (Zienkiewicz and Anderson 1967
and Zienkiewicz 1969), these analyses are typically done with three-dimensional models.
Nuss (1995) compares the results of a three-dimensional model of an entire buttress dam
with simplified three-dimensional models of slices taken from the complete model.
Several boundary conditions should be used to bound a reasonable range of solutions
when slice models are used.

10-2.5.3 Analysis of Reinforced Concrete

In dams constructed before the 1940s, square, un-deformed reinforcing bars were
sometimes used. Because of the lack of deformation ridges, these bars have longer
development lengths and inferior post yield performance. For this reason, the ultimate
strength method of concrete analysis currently used by ACI is not appropriate. These
structures must be analyzed using the working stress method in ACI 318-95 Appendix A,
"Alternate Design Method." The following criteria shall apply for slabs of slab and
buttress dams with un-deformed bars:

Allowable Stresses for Concrete Slabs with Plain Bars*

Tensile Stress in Reinforcing 20 KSI (140 mPa)
Steel
Bond Stress Between Concrete 0.04 Fêc (160 psi maximum) (ACI 1936 and ACI
and Reinforcing 1963)

*
These values may be increased by 1.33 for unusual and extreme loading conditions.

The upstream faces of flat slab buttress dams and reinforced buttresses, with deformed
reinforcing bars, should be analyzed using the Corps of Engineers "Strength Design for
Reinforced Concrete Hydraulic Structures" USACE (1992).

For these analyses, the size, spacing, and strength of reinforcing steel in existing dams
should be established. As a first step, conservative assumptions may be used. Steel used

10-19
in many existing buttress dams may have a lower yield strength than modern steel. If the
results of the initial analyses based on conservative assumptions indicate unsatisfactory
behavior, or are ambiguous, field investigations may be needed to refine the assumptions.
The type, spacing, and condition of existing reinforcement can be established with field
investigations and its yield strength can be evaluated by testing.

Large deflections can crack the slabs and expose the reinforcing steel to water. Where
large deflections are observed, corrosion of the reinforcing steel should be evaluated.
Non-destructive testing can be used to locate reinforcing steel, but coring or excavating is
usually required to evaluate the condition of the reinforcing steel.

Slabs on the downstream face are evaluated in the same manner as those on the upstream
face.

Corbels or haunches used to support the upstream face on buttresses should have
sufficient reinforcement for water loads, upstream face loads, and frictional forces that
tend to prevent expansion or contraction of the upstream face from temperatures changes.

Where buckling is suspected of being a problem, buttresses should be evaluated as load

bearing plates (or columns if there are longitudinal joints or cracks).

A number of computer software programs are available for use in analyses. Refer to
Chapter III, Section 3-4.7 for further discussion.

10-2.5.4 Dynamic Analyses

The seismic coefficient method, discussed in Chapter III, Section 3-2.6.2 of these
guidelines is adequate for buttress dams located in seismic zones 0 and 1. Vertical
acceleration should also be considered.

Dynamic analyses are required for all buttress dams in zone 2 as well as for those in
zones 3, and 4, because of the susceptibility of buttresses to overstressing from lateral
loading.

The pseudo-dynamic method discussed in Chapter III, Section 3-2.6.3 of these Guidelines
was developed based on typical dynamic characteristics of concrete gravity dams and is
therefore not appropriate for buttress dams.

The finite element method is the only practical method available for performing dynamic
analyses of buttress dams. This type of analysis is discussed in Chapter III, Section 3-4.4
of these guidelines. Either time history or response spectrum analyses are acceptable.
The response of the dam should be evaluated for earthquake loads applied in both the
transverse (upstream-downstream) and longitudinal (cross-valley) directions. Dynamic,
three-dimensional finite element analyses of multiple arch buttress dams are quite
complex and require considerable experience and judgement to perform and evaluate.

10-20
Nuss, Payne, and Sozen (1994), Nuss (1995) and Boggs, Jansen, and Tarbox (1988)
describe several such analyses and some of the complexities involved.

10-2.5.5 Post Earthquake Stability Analyses

The effects of cross-valley loading and structural response must be considered in the post
earthquake stability analysis. Significant rocking of the buttresses with respect to the
foundation, may destroy cohesive bonds.

10-2.6 Acceptance Criteria

Criteria for buttress dams are generally the same as for concrete gravity dams discussed
in Chapter III of these Guidelines. Only exceptions, clarifications, and additions are
discussed below.

10-2.6.1 Static

Sliding stability criteria are the same as for concrete gravity dams discussed in Chapter III
of these Guidelines.

Stress criteria for unreinforced concrete are the same as for concrete gravity dams
discussed in Chapter III of these Guidelines, except that the major principal, minor
principal, and maximum shear stresses should meet the recommended factors of safety in
Table 2 "Recommended Factors of Safety" of Chapter III.

For buttresses, the maximum allowable compressive stress may be governed by the
slenderness ratios. If bracing, struts, stiffeners, or counterforts exist, they will increase
resistance to buckling and should be considered in any analysis.

Stress criteria for reinforced concrete should follow the Corps of Engineers "Strength
Design for Reinforced Concrete Hydraulic Structures" (USACE 1992).

10-2.6.2 Dynamic and Post Earthquake

Dynamic stability criteria for buttress dams are the same as for concrete gravity dams
discussed in Chapter III, Section 3-5.3 of these Guidelines.

The USBR has applied deflection criteria to evaluate the stability of multiple arches when
stress levels were near the allowable dynamic strength limits of the reinforced concrete
and substantial cracking was anticipated (Nuss, Payne, and Sozen 1994). The criteria is
applicable to reinforced concrete elements whose behavior is dominated by flexural and
axial stresses. The calculated deformation of the arches, in a cracked condition, was
found to be of the same magnitude as the estimated yield deformation. Since
experimental tests have shown that reinforced concrete beams can tolerate displacement
on the order of twice the yield displacement, the arches were judged to be adequate.

10-21
10-2.6.3 Foundation Stability

Foundation stability criteria for buttress dams are the same as for concrete gravity dams
discussed in Chapter III, Section 3-5.6 of these Guidelines. For Buttress dams, the
gravity and hydrostatic loads are transferred to the foundation through the buttresses.
While gravity dams can bridge over small areas of bad foundation material, each and
every buttress must be founded on a competent foundation.

10-2.7 Material Properties

In general, material properties for buttress dams are the same as for concrete gravity dams
discussed in Chapter III, Section 3-6 of these Guidelines. Specific concerns for buttress
dams are poor quality concrete, deterioration of the concrete, and the condition, size,
spacing and strength of any reinforcing steel.

Most buttress dams were constructed before construction quality control programs and
concrete air-entrainment were common. As a result, the quality of concrete is quite
variable, and often subject to deterioration, especially from freeze-thaw action. All
concrete should be thoroughly examined for evidence of deterioration. Where there is a
concern about the quality of the concrete, cores should be taken to evaluate the nature and
extent of deterioration and representative samples should be tested to evaluate strengths.

Deterioration and cracking of the concrete can expose the reinforcing steel to air and
water, resulting in corrosion. Where existing reinforcing steel is structurally necessary,
the condition, type, and extent of reinforcing steel should be verified with field
investigations. Such field investigations may include coring, geophysical surveys, and
excavation and exposure of reinforcing in areas of concern. Where appropriate, the
strength of the reinforcing should be evaluated by testing. Concrete faces coated with
gunite can be deceiving. Poor quality concrete and corroded reinforcement may be below
surface.

10-2.8 References

American Concrete Institute (ACI), "Building Regulations for Reinforced Concrete,"

1936.

ACI, "Building Code Requirements for Reinforced Concrete," 1963.

Bengtson, G. S., "Analysis and Rehabilitation of Ambursen Dams," Proceedings of the

International Conference on Hydropower, held August 23-25, 1989 in Niagara Falls, New
York, ASCE, New York, New York. 1989.

Boggs, J., Jansen, R., and Tarbox, G., "Arch Dam Design and Analysis," Chapter 17 of
Advanced Dam Engineering, edited by R. Jansen, Van Nostrand Reinhold, New York,
1988.

10-22
Burroughs, E., "Ambursen Dams," Chapter 13 of Handbook of Applied Hydraulics,
edited by C. Davis and K. Sorensen, 3rd edition, McGraw-Hill Book Company, 1969.

Copen, M., Lindholm, E., and Tarbox, G., "Design of Concrete Dams," Handbook of
Dam Engineering, edited by A.R. Golze, Van Nostrand Reinhold Company, 1977.

Corns, C., Tarbox, G., and Schrader, E., "Gravity Dam Design and Analysis," Chapter 16
of Advanced Dam Engineering, edited by R. Jansen, Van Nostrand Reinhold, New York,
1988.

Creager, W., "Engineering for Masonry Dams," John Wiley & Sons, London, 1917.

Creager, W. and Justin, J., "Hydro-Electric Handbook," John Wiley & Sons, London,
1927.

Creager, P., Justin, J., and Hinds, J., "Concrete Dams," Vol. 2 of Engineering for Dams,
John Wiley & Sons, New York, 1945.

Croasdale, K., "Ice Forces on Fixed, Rigid Structures," Working Group on Ice Forces on
Structures, edited by T. Carstens, U.S. Army, Corps of Engineers, Cold Regions Research
and Engineering Laboratory (CRREL), Hanover, New Hampshire, 1980.

Davis, C., "Concrete Dams, Basic Principles of Design," Chapter 9 of Handbook of

Applied Hydraulics, edited by C. Davis and K. Sorensen, 3rd edition, McGraw-Hill Book
Company, 1969a.

Davis, C., "Gravity Dams," Chapter 11 of Handbook of Applied Hydraulics, edited by C.

Davis and K. Sorensen, 3rd edition, McGraw-Hill Book Company, 1969b.

Davis, C., "Multiple Arch Dams," Chapter 15 of Handbook of Applied Hydraulics, edited
by C. Davis and K. Sorensen, 3rd edition, McGraw-Hill Book Company, 1969c.

Garland, J.D., Waters, R.H., Focht, J. A. Jr., and Rutledge, J. L., "Dam Safety, Morris
Sheppard Dam Rehabilitation," Proceedings of the International Conference on
Hydropower, held July 25-28, 1995 in San Francisco, California, ASCE, New York, New
York. 1995.

Houk, I. and Wengler, R., "Arch Dams," Chapter 14 of Handbook of Applied Hydraulics,
edited by C. Davis and K. Sorensen, 3rd edition, McGraw-Hill Book Company, 1969.

Lamar, B. H., Ivarson, W., and Tenke-White, K., "Rehabilitation and Replacement of
Slab and Buttress Dams at the Wissota Hydroelectric Project," Proceedings of the
International Conference on Hydropower, held July 24-26, 1991 in Denver, Colorado,
ASCE, New York, New York. 1991.

10-23
Legas, J., "Concrete Buttress Dams," Section 4 of Development of Dam Engineering in
the United States, edited by E. Kollgaard and W. Chadwick, Pergamon Press, Elmsford,
New York, 1988.

Marcello, C. "Hollow Gravity Dams," Chapter 12 of Handbook of Applied Hydraulics,

edited by C. Davis and K. Sorensen, 3rd edition, McGraw-Hill Book Company, 1969.

Niziol, J. S. and Paolini, E. M., "Case History of Sherman Island Hydro Buttress Dam,"
Proceedings of the International Conference on Hydropower, held August 10-13, 1993 in
Nashville, Tennessee, ASCE, New York, New York. 1993.

Nuss, L., Payne, L. and Sozen, M., "Case Study of Dynamic Analyses of an Existing
Multiple Arch Dam: Bartlett Dam, Phoenix, USA," Proceedings of the International
Workshop on Dam Fracture and Damage, held March 16-18, 1994 in Chambrey, France,
A. A. Balkema, Rotterdam and Brookfield, Vermont, 1994.

Nuss, L. K., "Comparisons Between Complete and Slice Finite Element Models of a
Multiple Arch and Buttress Dam," Proceedings of the International Conference on
Hydropower, held July 25-28, 1995 in San Francisco, California, ASCE, New York, New
York. 1995.

Reynolds, R. D., Joyet, R. A., and Curtis, M. O., "Victoria Dam Rehabilitation,"
Proceedings of the International Conference on Hydropower, held August 10-13, 1993 in
Nashville, Tennessee, ASCE, New York, New York. 1993.

Rice, E., “Building in the North,” Geophysical Institute of the University of Alaska,
Fairbanks, Alaska, 1975.

Rohde, M. W. and Zuccolotto, J. M., "Monitoring and Analysis of Dam Deformations,

Sherman Island Hydroelectric Development," Proceedings of the International
Conference on Hydropower, held July 25-28, 1995 in San Francisco, California, ASCE,
New York, New York. 1995.

Thomas, H., "The Engineering of Large Dams," John Wiley & Sons, London, 1976.

USACE (U.S. Army Corps of Engineers), "Strength Design of Reinforced Concrete

Hydraulic Structures," EM 1110-2-2104, June 30, 1992.

USBR (U.S. Bureau of Reclamation), "Design Criteria for Concrete Arch and Gravity
Dams," Engineering Monograph No. 19, 1977a.

USBR "Design of Arch Dams," 1977b.

10-24
Wheelock, D. C. and Wilkins, N. A., "Analysis and Stabilization of Ambursen Dams -
Two Case Studies," Proceedings of the International Conference on Hydropower, held
July 24-26, 1991 in Denver, Colorado, ASCE, New York, New York. 1991.

Wegeman, E., "Design and Construction of Dams," 8th edition, John Wiley & Sons, New
York, 1927.

Zangar, C.N., “Hydrodynamic Pressures on Dams Due To Earthquake Effects,”

Engineering Monograph No. 11, U.S. Bureau of Reclamation, Denver, Colorado, May
1952.

Zienkiewicz, O.C., Anderson, G., and Irons, B., "Buttress Dam Analysis for Earthquake
Loads," Water Power and Dam Construction, September, Surrey, United Kingdom,
September, 1967.

Zienkiewicz, O.C., "The Finite Element Method," Chapter 10 of Handbook of Applied

Hydraulics, edited by C. Davis and K. Sorensen, 3rd edition, McGraw-Hill Book
Company, 1969.

10-3 Concrete Dams on Pile Foundations

10-3.1 Introduction

Piles used to found concrete dams and other hydraulic structures are designed to
distribute foundation loads, both axial and lateral, into or through loose, soft or
compressible soils beneath the structures. This section provides an overview of the
evaluations of existing hydraulic structures founded with piles, primarily with respect to
lateral forces transmitted from the dams to the foundations. The response of pile
foundations to vertical loads is important and should be considered in safety evaluations
of hydraulic structures. However, detailed discussions concerning the evaluation of pile
resistance to vertical loads are not provided as vertical loading is usually not the critical
case in such evaluations. The design and construction of new pile foundations are also
not specifically addressed.

Piles are defined as driven, not drilled, members. Piles can include steel H piles, steel
pipe piles, precast concrete, cast-in-place and mandrel driven piles and timber piles. Pile
details, capacities and allowable stresses are provided in USACE 1991 and many other
publications.

10-3.2 Forces

Forces for the analysis of concrete dams on pile foundations are generally the same as for
gravity concrete dams discussed in Chapter III, Section 3-2 of these Guidelines.

10-25
10-3.3 Loading Combinations

Loading combinations for the analysis of concrete dams on pile foundations are generally
the same as for gravity dams discussed in Chapter III, Section 3-3 of these Guidelines.
Loading combinations should be established to produce critical combinations. For each
loading combination, the effect each load will have on pile forces and on internal forces
in the pile cap should be considered. Some loadings may control the internal design of
the pile cap even though they may not produce the critical pile forces. Generally, it is
important to analyze the load cases with the largest lateral loads in each direction and the
cases with the maximum and minimum vertical loads.

10-3.4 Structure and Subsurface Conditions

The evaluation of hydraulic structures on pile foundations involves the solution of

complex soil-pile-structure interaction problems. Details concerning the structure, piles
and subsurface conditions need to be established regardless of the method of analysis.
Simplifying assumptions concerning many of the parameters are required regardless of
the analytical technique used for the evaluations.

Information concerning the structure includes the following:

• Structure configuration and dimensions, including plans, sections,

expansion joints, etc.

• Pile foundation details, including pile types, sizes, lengths, locations,

spacing, depths, attachment conditions to the structure, etc.

• Structure and foundation performance, including horizontal and vertical

movements, seepage conditions, the condition of the structure and piles,
etc.

• Past, current and anticipated reservoir and tailwater levels and other
similar conditions that result in forces and loads on the structure.

Subsurface condition information includes:

• Types and depths of soils and bedrock beneath the structure

• Groundwater levels, uplift pressures and seepage conditions

• Soil and bedrock engineering properties, including unit weight,

plasticity, gradation, strength (angle of internal friction, cohesion),
consolidation, etc.

• Potential soil settlement, which can create voids under the structure.

10-26
Sources of such information include structure and geotechnical data and design reports,
construction plans and specifications, construction records, inspection reports, and
records documenting structure and reservoir monitoring. Such information may not be
readily available, particularly for older structures. Assumptions concerning structure, pile
and subsurface conditions must frequently be made for initial evaluations. If so,
parametric evaluations of the effect of differing conditions should be considered.

One concern related to hydraulic structures supported on piles is potential settlement of

soil under the structure. Existence of such condition, if suspected, should be investigated
by installation of piezometers and drilling of exploratory bore holes as necessary. If
existence of voids is confirmed, controlled grouting so as not to overstress the structure
and impact its design provisions would be required.

10-3.5 Analyses

10-3.5.1 General

Pile foundations are used to distribute foundation loads into or through soils beneath
structures to provide increased foundation resistance to anticipated loads and to reduce
lateral and/or vertical deformations. The load capacity of a pile is dependent on the
properties of the pile, the properties of the soil and/or bedrock around and beneath the
pile, and allowable pile movement. Attachment conditions of the pile to the pile cap also
affects the lateral load capacity of piles. Design (working) capacities are generally
governed by pile movement considerations and are less than ultimate (failure) capacities
of the piles.

10-3.5.2 Vertical Capacity

Piles beneath hydraulic structures should be evaluated assuming all vertical loads are
carried by the piles, and assuming no uplift (tension) loads in the piles. However, the
vertical capacity of piles is generally not the critical condition in safety evaluations as
actual pile foundation loads exceeding allowable pile vertical capacity will likely result in
increased pile settlement and the pile cap bearing on the underlying soils. While
settlement may cause problems, it does not necessarily constitute a safety problem.

Vertical capacity and settlement evaluations for pile foundations are discussed in ASCE
1984, CGS 1985, Davisson 1970, NAFAC 1982, Poulos and Davis 1980, USACE 1991;
and Vesic 1977, as well as many other publications.

10-3.5.3 Lateral Capacity

Resistance provided by piles to horizontal forces is the result of complex interaction of

the structure, pile and foundation soils. Structure forces are distributed through and by
the piles to the surrounding soils. The pile and the soils each provide resistance, the

10-27
amount of which is a function of the relative stiffness of the pile and the soils, and the
magnitude of the load applied to the pile. Soil resistance is a nonlinear function of soil
deflection and of the specific type and properties of the soil. Varying soil deflections and
resistances with depth cause pile bending which induces shear and bending stresses in the
pile. Pile top connections to the structure (pile caps) also influence the distribution of
stresses in the pile and soils. Rigid, fixed connections where pile top rotation is
prevented or limited cause more bending stresses at the pile top than do pinned
connections where the pile top is free to rotate (see Figure 10-3.1).

Several procedures have been used to model the response of vertical piles subject to
lateral loads. Linear or constant soil responses, rigid, non-bending piles, fixed points of
pile rotation and other similar simplifying assumptions have been included in most of the
models to facilitate their solution. A more rigorous model is based on the elastic bending
of the pile and a nonlinear soil response that is a function of soil properties and soil
movement.

Considerable judgement must be applied in selecting and applying analytical techniques

to the evaluation of pile foundations due to the complexity of the problems and the
simplifying assumptions involved. The techniques used should be selected on a case-by-
case basis. Parametric studies where input assumptions and values are systematically
varied to test the effect of the various assumptions on the evaluation results should be
considered.

The elastic pile method and an ultimate analysis method are discussed below.

10-28
Elastic Pile Method

Based on the assumption that the pile is a linearly elastic beam and that the soil reaction
can be represented as a line load (Reese 1977), the equation for pile-soil interaction is:

d 4y d 2y
EI %P x
&P'0
dx 4 dx 2

where, as indicated on Figure 10-3.2:

E = modulus of elasticity of the pile

I = moment of inertia of the pile
y = lateral deflection of the pile at point x along the pile
length
x = distance from top of pile
Px = axial load on the pile
P = soil reaction per unit length

For most conditions, horizontal forces induced into the pile by the axial load (Px) are
small due to relatively small lateral deflections (y) and can be neglected in the analysis.
The equation becomes:

10-29
 d 4y
EI ' P
dx 4

The following can be obtained by integrating the equation:

d 3y
EI 'V
dx 3

d 2y
EI 'M
dx 2

dy
EI 'S
dx

where: V = shear force in the pile at point x

M = bending moment in the pile at point x
S = slope of the deflected shape of the pile at point x

The soil reaction (P) in the equations is a nonlinear function of deflection (y) and varies
depending on soil type and properties (see Figure 10-3.3). Analytical procedures have
been developed (Reese 1977, Geosoft 1987, and others) to predict the soil reaction (P-y
curves) based on typical soil properties, including soil type, cohesion (c), friction angle
(ø), and soil unit weight ((). Computer programs (Reese 1977, Geosoft 1987, and others)
are available that develop P-y curves and use iterative techniques to solve the equations
for elastic bending of the pile. The fixity or amount or rotation of the pile top can be
specified, and layered soil profiles can be used. Output from the programs typically
includes moment (M), shear (V), lateral deflection (y) and soil reaction (P) relative to
distance below the pile top (x).

10-30
The above methods can be used to evaluate resistance of pile foundations to lateral
forces. Forces acting on piles can be developed based on the lateral forces acting on the
dam, and pile deflections, moments and shear forces can be estimated using the computer
programs. Theoretical factors of safety against pile failure or against pile stresses
exceeding allowable values can be obtained by dividing ultimate or allowable moments
and shear forces obtained from elastic pile properties by the maximum moments and
shear forces obtained from the lateral load analysis. Forces from varying loading
conditions (usual, unusual or extreme) can be evaluated, and potential total movements of
the dam under those forces can be obtained for further consideration in the safety
evaluations. If data is available documenting actual dam movements at known loading
conditions, the computer model can be calibrated using that information.

An example of the elastic pile method of analyses is summarized in the attached Pile
Analysis Example.

Ultimate Pile Strength Method

Another method to evaluate the ultimate capacity of piles and pile foundations is to
assume the foundations deflect sufficiently over their entire length to develop the ultimate
resistance of the soil surrounding the piles (Poulos and Davis 1980). The ultimate
resistance of the soil is (Broms 1964a and 1964b):

10-31
 For cohesive soils:

From the bottom of the pile cap to 1.5 pile diameters

C = 0

From 1.5 pile diameters to the bottom of the pile cap

C = 9cD

For cohesionless soils:

Kx = 3Kp(D

where:

C = ultimate resistance of cohesive soils, lb/ft

c = soil cohesion, lb/ft2
Kx = ultimate resistance of cohesionless soils at point x
along pile length
kp = coefficient of passive lateral earth pressure
= (1 + sin ø)/(1 - sin ø)
ø = angle of internal friction (degrees)
( = soil unit weight, lb/ft3
x = vertical distance from ground surface, ft
D = pile diameter, ft

From the limit states and pressures shown on Figure 10-3.4, the following ultimate loads
for laterally loaded piles can be derived:

For cohesive soils:

Short Pile:
V u'C(L&1.5D)

Medium Pile:
Vu' 4CMt%2(CL)2%4.5(CD)2&C(1.5D%L)

Long Pile:
Vu' 2C(Mt%M b)%2.25(CD)2&1.5D

For cohesionless soils:

10-32
 Short Pile:
KL 2
V u'
2

Medium Pile:
2
3M tL3 3 L2
Vu'K[( % ) & ]
2K 2 2

Long Pile:
1
Vu'[1.125K(Mt%M b) ] 2 3

where:

Vu = ultimate lateral resistance

M = ultimate moment capacity of the pile at the pile top (Mt) and pile
tip (Mb)

L = pile length

For a given soil type and fixity condition, the equation that yields the minimum load
governs. In cases where the pile is not sufficiently embedded in the cap to be considered
fixed, Mt should be assumed to be zero or an evaluation of the percentage of the fixed
head moment should be developed for the particuar embedment case.

A procedure for deriving theoretical factors of safety using the ultimate pile strength
method is to calculate ultimate pile moment capacities (MT, MB) and ultimate pile shear

10-33
capacity using ultimate strength values for the pile type and size being analyzed. Those
ultimate moment capacity values, along with the ultimate soil strength values (C,K) can
be substituted into the appropriate equations to calculate the minimum ultimate lateral
resistance (Vu) of the piles, as previously described. The factor of safety is the minimum
ultimate lateral resistance divided by the sum of the lateral forces acting on the dam, or
the maximum ultimate shear capacity of the piles, whichever produces the lower value.

10-3.6 References

ACI (American Concrete Institute), "Building Code Requirements for Reinforced

Concrete," ACI 318, Detroit, Michigan, 1995.

ACI, "Recommendations for Design, Manufacture and Installation of Concrete Piles,"

ACI 543, Detroit, Michigan, 1986.

AISC (American Institute of Steel Construction), Manual of Steel Construction, 9th ed.,
New York, 1989.

ASCE (American Society of Civil Engineers) Geotechnical Engineering Division,

Committee on Deep Foundations, "Practical Guidelines for the Selection, Design, and
Installation of Piles," American Society of Civil Engineers, New York, New York, 1984.

ASTM (American Society for Testing and Materials), "Method for Establishing Design
Stresses for Round Timber Piles," D2899, Philadelphia, Pennsylvania, 1974.

ASTM, "Standard Test Method for Individual Piles Under Static Axial Tensile Load,"
D3689, Philadelphia, Pennsylvania, 1990a.

ASTM, "Standard Test Method for Piles Under Lateral Loads," D3966, Philadelphia,
Pennsylvania, 1990b.

ASTM, "Standard Test Method for Piles Under Static Axial Compressive Load," D1143,
Philadelphia, Pennsylvania, 1987.

Baguelin, F., Jezequel, J.F., and Shields, D.H., "The Pressuremeter and Foundation
Engineering," Trans Tech Publications, Rockport, Massachusetts, 1978.

Bowles, J.E., "Pile Cap Analysis," Proceedings, Eighth Conference on Electronic

Computation, A.S.C.E., 1983, pp. 102-113.

Broms, B.B., "Lateral Resistance of Piles in Cohesive Soils," Journal of the Soil
Mechanics and Foundation Division, ASCE, Vol 90, SM2, 1964a.

Broms, B.B., "Lateral Resistance of Piles in Cohesionless Soils," Journal of the Soil
Mechanics and Foundation Division, ASCE, Vol 90, SM3, 1964b.

10-34
CGS (Canadian Geotechnical Society), "Foundation Engineering Manual," 2nd Edition,
1985.

Davisson, M.T., "Design Pile Capacity," Proceedings of the Conference on Design and
Installation of Pile Foundations and Cellular Structures, Lehigh Valley, Pennsylvania,
Envo Publishing Co., 1970a.

Davisson, M.T., "Lateral Load Capacity of Piles," Highway Research Record No. 333,
Highway Research Board, National Academy of Sciences - National Research Council,
Washington, D.C., 1970b.

Deep Foundations Institute, Inspection and Testing Committee, "Inspector's Manual for
Pile Foundations," Springfield, New Jersey, 1979.

Ensoft, Inc., "L Pile 1," 1989.

Ensoft, Inc., "Group Analysis of a Group of Piles Subjected to Axial and Lateral
Loading," 1990.

Geosoft, "PILED/G, Lateral Load Analysis of Drilled Piers and Piles with Internal
Generation of P-Y Curves," 1987.

Meyerhof, C.G., "Bearing Capacity and Settlement of Pile Foundations," Journal,

Geotechnical Engineering Division, American Society of Civil Engineers, New York,
New York, March, 1976.

NAVFAC (Naval Facilities Engineering Command), Foundations and Earth Structures,

Design Manual 7.2, May 1982.

Poulos, H.G. and Davis, E.H., Pile Foundation Analysis and Design, John Wiley and
Sons, 1980.

Reese, L.C., "Laterally Loaded Piles: Program Documentation," Journal, Geotechnical

Engineering Division, American Society of Civil Engineers, New York, New York, April,
1977.

USACE (U.S. Army Corps of Engineers), "Design of Pile Foundations," EM 1110-2-

2906, January 1991.

USACE, "Strength Design for Reinforced Concrete Hydraulic Structures," EM 1110-2-

2104, June 1992.

Vesic, Aleksandar S., "Load Transfer in Pile-Soil Systems," Proceedings of the

Conference on Design and Installation of Pile Foundations and Cellular Structures,
Lehigh Valley, Pennsylvania, Envo Publishing Co., 1970.

10-35
Vesic, Aleksandar S., "Design of Pile Foundations," National Cooperative Highway
Research Program, Synthesis of Highway Practice No. 42, Transportation Research
Board, Washington, DC, 1977.

10-3.7 Pile Analysis Example

Consider the pile supported Ambursen dam shown below. The dam is hollow with 2-foot
thick interior piers at 20-feet on center.

10-36
As with any stability analysis, the first step is to determine the loads acting on the
structure. In this case, since there is a joint between the dam and the spillway apron,
bending moments will not be transmitted between the two structures. For this reason, the
vertical loads will be independently tabulated for the dam and the spillway. The
horizontal loads, however, will be resisted by both structures together.

DETERMINATION OF LOADS:

UPLIFT

Uplift can be determined using a hand drawn flow net or a 2 dimensional finite element
solution. This example is not intended to demonstrate this procedure. The uplift head
values shown below will be assumed for this example.

Total Horizontal
Head Location Force At
(Ft) (Ft) (kips) (Ft)

POINT A 32.52 0 80.87 1.00

U.S. SHEETPILE 32.28 2

D.S. SHEETPILE 17.52 2

1109.20 31.69
POINT B 9.41 68

TOTAL-> 1190.07 29.60

POINT B 9.41 68
297.19 82.59
POINT C 5.47 100

10-37
 Now that the uplift has been determined, the other forces and moments can be summed to
determine the resultants acting on each pile group.

Note below that the horizontal reservoir force has been broken up into 2 portions; the 30
feet above the reservoir bottom and the 4 feet acting on the vertical face of the dam below
the reservoir bottom. This is necessary because the head loss due to vertical flow in the
soil causes the head at point A to be slightly less than the full 34 feet of reservoir head.
Note also that active lateral soil pressure is applied to the dam and passive to the
spillway. These are not the forces present at service load, but they will be the forces
present at failure; the failure condition is what we referenced to the safety factor.

Fx @ Fy @ M@A
(Kips) (ft) (Kips) (ft) (Kip-ft)
AMBURSEN DAM

Structure Wt. -1661.2 33.0 54819.6

Vert. reservoir -564.6 10.1 5702.5
Horiz. reservoir 561.6 14.0 7862.4
156.0 2.0 312.0
Active earth 3.6 1.33 4.8
Uplift 1190.1 29.6 -35227.0
___________________________________________________________

TOTALS-> 721.2 -1035.7 33474.3

VERT. RESULTANT @ R= 32.3

DOWNSTREAM APRON M@B

Structure Wt. -432.0 17.5 7560.0

Tailwater -13.7 1.33 -18.2
Passive earth -32.4 1.33 43.2
Uplift 297.2 14.6 -4339.1

TOTALS-> -46.1 -134.8 3159.5

VERT. RESULTANT @ R= 23.4

Total External Forces on

Dam and Apron 675.1 1170.5

10-38
INDIVIDUAL PILE AXIAL FORCES

To determine the axial forces on the piles, the pile group section properties must be computed.
The axial force on a pile is given by the following equation:

Fv F v(R&X)(X&X)
P x' %
N I

WHERE:

X' j ,I'j (X&X)2

X
N

PILE GROUP SECTION PROPERTIES

AMBURSEN DAM

# of piles Location
in row × (ft) 5*× 5*(×-0)2

5 5 25 4500
5 9 45 3380
5 13 65 2420
5 17 85 1620
5 21 105 980
5 25 125 500
5 29 145 180
5 33 165 20
5 37 185 20
5 41 205 180
5 45 225 500
5 49 245 980
5 53 265 1620
5 57 285 2420
5 61 305 3380
5 65 325 4500

N v 80 3 5* ×v 2800 27200 w I

GROUP CENTROID 0 = 35
Note that the sheetpile is ignored in the calculation of the pile group section properties. The
sheetpile was also ignored in the calculation of forces. This is conservative. If sheetpile walls

10-39
are attached the base slab sufficiently to develop the moment capacity of the pile, and if the
piles themselves are of substantial section, the lateral load resistance of sheetpile walls can be
significant.

Upstream row:
&1035.7 &1035.7(32.3&35)(5&35)
P x' % '&16 KIPS
80 27200

Downstream row:
&1035.7 &1035.7(32.3&35)(65&35)
P x' % '&9.9 KIPS
80 27200

Note that all forces are compressive. The magnitude of these pile loads is less important than
the sign. If the pile is overloaded in compression, the structure may settle, but it is not likely to
fail. If a pile is loaded in tension, the analysis should be re-done with the pile in tension
removed. Pile tension should not be relied upon for stability.

DETERMINATION OF PILE ULTIMATE STRENGTHS:

The ultimate moment capacity of a timber pile is a function of the extreme fiber bending
strength. For a 12-inch diameter pile with an extreme fiber bending strength (Fb) of 2500 psi,
the ultimate moment capacity is:

Mt'SFb

BD 4 D BD 3
S'I/C' / '
64 2 32

B(12)3 2500
Mt ' x ' 424.1 kip&inch ' 35.3 kip&ft
32 1000

Timber piles are tapered. Assuming the diameter of the pile tip is about 9.5 inches, the
maximum moment Mb at the pile tip is half of the ultimate moment at the pile top, or 17.6 ft-
kips:

B(9.5)3 2500
Mb ' x ' 210.4 kip&inch ' 17.6 kip&ft
32 1000

10-40
Piles in the dam are embedded 2 feet. Those in the apron are embedded 1 foot. It is necessary
to determine if the full moment capacity of the pile top can be developed based upon those
embedment lengths into the pile cap.

If the pile embedment is not sufficient to resist rotation (twisting) of pile in the pile cap socket,
then stress on the outside of the pile can be assumed to reach ultimate crushing stress (Fc) at the
pile centerline, and other stresses can be assumed as shown in Section A-A, Figure 10-3.7.

Fc = Ultimate crushing strength

FR = Radial crushing strength
Fx = Crushing strength normalized to x-direction

FR = Fc cos 1
Fx = FR cos 1 = Fc cos2 1

The total force per unit length (vertical) available in the direction of movement (x) should be:

m0
B/2
F ' 2 Fc cos21Rd1

B/2
1 sin2 1
F ' 2RFc %
2 4 0

BRFc DFcB
F ' 2RFc[B/4] ' '
2 4

10-41
If it is assumed that rotation/twisting of pile in pile cap is about the midpoint of the
embedment, then:

E
FR ' (F)( )
2

j Mo ' 0

FE 2
M ' (FR)(E/2) '
4

DFcB
Substituting: F '
4

(DFcB/4)E 2 DFcB E 2
Results in: M ' '
4 16

Mc M
Fb ' ' Where Fb is bending stress
I S

BD 3
Section modulus (S) for round pile '
32

FbBD 3
M ' Fb S '
32

10-42
If the bending strength is assumed to be three times the crushing strength:

Fb
Fc '
3

Equating the moments results in:

FbBD 3 FcDBE 2
'
32 16

Substituting for Fc results in:

FbB D 3 (Fb/3)DBE 2
'
32 16

BD 2 BE 2
'
32 48

1/2
48B 3
E ' D ' D
32B 2

E ' 1.22D

Therefore, if E/D is greater than 1.22, then the full moment capacity of the pile top can be
developed.

For a 12-inch diameter pile with an ultimate shear stress perpendicular to the pile axis of 500
psi, the ultimate shear strength is:

500 BD 2
V' ' 56.5 kips
1000 4

10-43
ELASTIC PILE ANALYSIS:

The lateral movement of the top of each pile should be the same as all piles are similarly
embedded into the concrete. Since all the piles and the soils surrounding the piles are the
same, the lateral resistance provided by each pile should also be the same provided they are all
similarly fixed into the dam concrete. However, the piles in the downstream apron are
embedded one foot into the concrete, so they may not be entirely "fixed." If so, each pile there
should contribute less lateral resistance than each pile beneath the dam as fixed head pile
provide more resistance than do pinned head or partially fixed head piles at the same amount
of pile head movement.

Assuming the piles beneath the dam and apron are all fixed head piles, the lateral load carried
by each pile should be:

Vo = (721.2 kips - 46.1 kips)/(80 piles + 40 piles)

Vo = 5.6 kips per pile

Assuming the piles beneath the apron are not fixed and do not provide significant lateral
resistance to horizontal dam forces, the lateral load carried by each pile beneath the dam would
be:

Vo = (721.2 kips - 46.1 kips)/(80 piles)

Vo = 8.4 kips per pile

Soils around the piles are assumed to be sands with the following properties:

(m = 130 pcf

(b = 67.6 pcf

ø = 30o

The properties of the timber piles used for the analysis are:

Length = 20 feet

Diameter at the top = 12 inches

Diameter at the tip = 9.5 inches

10-44
 Average diameter (Da) = 10.75 inches

Moment of Inertia (I) = (B Da4)/64 = 665.6 in4

Modulus of Elasticity (E) = 1.5(106) psi

EI = 0.998(109) pound - in2

Using the program PILED/G (Geosoft 1987) and the above properties, the following were
obtained:

Shear at Pile Top Moment at Pile Top Pile Top Movement

5.6 kips 14.5 kip-ft 0.2 inches

8.4 kips 24.7 kip-ft 0.4 inches

Theoretical factors of safety against failure in shear and bending are:

Factor of Safety Shear (FSs) = (56.5 kips)/(8.4 to 5.6 kips)

FSs = 6.7 to 10.1

Factor of Safety Bending (FSb) = (35.3 kip-ft)/(24.7 to 14.5 kip-ft)

FSb = 1.4 to 2.4

The lower theoretical factor of safety values so obtained assume only the piles beneath the dam
resist lateral loads. Based on the relatively small lateral movement indicated by the analysis,
regardless of the number of piles considered, it appears that the lateral resistance of each pile
beneath the dam and apron should be nearly the same and near 5.6 kips per pile.

ULTIMATE PILE STRENGTH METHOD:

The ultimate soil strength for the sandy soils surrounding the piles should be:

K = 3 Kp ( D

Kp = (1+sin ø)/(1-sin ø) = (1+sin 30o)/(1-sin 30o) = 3.0

K = 3(3)(0.0676 kips/ft3)(1 ft)

K = 0.608 kips/ft2

10-45
Substituting into the appropriate single pile equations, as follows:

Short Pile:
KL 2
Vu '
2

(0.608)(20)2
Vu ' ' 121.7 kip
2

Medium Pile:

2
3Mt L2 L2
Vu ' K % 3
&
2k 2 2

2/3
3 x 35.3 203 202
' 0.608 % &
2 x 0.608 2 2

' 32.3 kip

Long Pile:

Vu ' [1.125k(M t%Mb)2]1/3

' [1.125 x 0.608(35.3%17.6)2]1/3

' 12.4 kip

The piles in the downstream apron are embedded only 1 foot. It has previously been shown
that an embedment of 1.22 pile diameters into the pile cap is required to consider the top of a
round timber pile fixed. This means that the full moment capacity of the piles in the
downstream apron cannot be counted on.

10-46
Referring to a previously developed relationship for round timber piles:

3
E ' D
2

Which simplifies to:

2
2 E
' 1.0
3 D

Using this relationship, the following equation can be used to determine the ultimate moment
capacity at the top of a partially embedded pile. This equation is only applicable for round
timber piles and assumes that the bending strength is equal to 3 times the crushing strength.

2
2 E
M+T ' MT
3 D

Where:
MUT= Ultimate moment capacity at the top of a partially embedded pile.

E = Embedment distance into the pile cap

D = Diameter of pile

2
2 1
M+T ' 35.3 kip&ft ' 23.5 kip&ft
3 1

For these piles, the ultimate lateral capacity is 10.5 kip (based on the Long Pile formula and Mt
= 23.5 kip-ft).

The total lateral resistance capacity for the dam and apron based on individual pile strengths is
then:

80 piles x 12.4 kip/pile % 40 piles x 10.5 kips/pile ' 1412 kips

10-47
The sum total of the external lateral forces acting on the dam and apron is 675.1 kips, as
previously determined under DETERMINATION OF LOADS. The overall theoretical factor
of safety is:

j Vu 1412
j Vx
' ' 2.1
675.1

10-4 Concrete Dams on Soil Foundations

Concrete dams constructed on soil foundations are relatively small structures that exert low
bearing pressures on the foundation. Typically, such structures have been less than 50 feet in
height with less than 20 feet of headwater-tailwater differential. Large structures on soil
foundations are usually supported on piles as discussed in Chapter X, Section 10-3 of these
Guidelines.

10-4.1 Forces

Forces for the analysis of concrete dams on soil foundations are generally the same as for
gravity dams discussed in Chapter III, Section 3-2 of these Guidelines.

The resulting shear and moment from net pressure acting on the cutoff should be applied to the
structure. For flexible steel sheet piles, the unbalanced load transferred to the structure may be
negligible. For a continuous rigid cutoff, such as a concrete cutoff, the unbalanced load should
be accounted for. If a cutoff is assumed to transfer the load to the dam, it must have sufficient
strength.

10-4.2 Loading Combinations

Loading combinations for the analyses of concrete dams on soil foundations are generally the
same as for gravity dams discussed in Chapter III, Section 3-3. In addition, for any new dam
constructed on a soil foundation, the end of construction loading condition must be analyzed as
discussed in Chapter IV, Section 4-6 of these Guidelines.

10-4.3 Analyses

Analyses for existing concrete dams on soil foundations typically use the gravity, and dynamic
methods for concrete gravity dams discussed in Chapter III, Section 3-4 of these Guidelines.
The cracked base type of analysis discussed in Section 3-4.6 is not applicable. Generally, the
dam would be long in relation to height and transverse contraction joints neither keyed nor
grouted in order to allow independent monolith movement, therefore the trial load analysis
would not be appropriate. The foundation should be analyzed for static and seismic stability,
seepage and erosion as discussed in Chapter III, Section 3-5.6.2 and liquefaction potential as
discussed in Chapter IV, Section 4-5.6.

10-48
Analyses for new dams should consider the use of the finite element method in order to
adequately model the dam-foundation interaction to account for foundation settlement.

10-4.3.1 Bearing Capacity

Foundation bearing capacity must be checked since the concrete dam may impose fairly high
stresses on the foundation soil. The potential presence of weak soil beneath the dam should be
carefully investigated. It may be necessary to enlarge the base of the dam to reduce stresses
and improve the foundation bearing capacity.

10-4.3.2 Sliding

Passive resistance of the toe of the dam and foundation soil should be taken into account if the
toe area is amply protected from scour. For existing dams where the downstream toe is
inundated, adequate investigations should be made to evaluate existing scour. This is
extremely important because if the soil can be eroded, it cannot be relied upon for passive
resistance. If the scour protection is being continually displaced, heavier protection is
required.

Sliding analysis should use the limit equilibrium approach as described in USACE ETL 1110-
2-256 (1981). Special care should be taken to identify the presence of any thin weak lenses in
the soil foundation such as clay layers in alluvium.

10-4.3.3 Deformation

Foundation settlement of existing dams is usually complete within a few years after
construction. For existing dams, visual examination and surveys of the dam structure should
be conducted to identify continuing movement and excessive or unequal settlement of the dam
which would adversely impact the water tightness of the structure or the integrity of the dam
and foundation seepage control facilities.

For new dams, verifying the overall stability includes an estimation of the absolute and
differential settlements due to the soil foundation in order to make sure that deformation is
compatible with the proper behavior of structures and with the deformation capacities of the
rigid concrete dam. Adequate field studies are required to quantify the foundation
deformability characteristics. Consideration should be given to the proper placement of
keyed/unkeyed joints and waterstops in the rigid concrete dam to control differential settlement
and seepage. A 50-foot spacing of contraction joints is usually sufficient. Where foundation
conditions are such that undesirable differential settlement or displacement between adjacent
blocks can occur, shear keys should be formed in the contraction joints. These may be formed
vertically, horizontally, or in a combination of both, depending on the direction of the expected
displacement. Leakage through the contraction joints is controlled by imbedding waterstops
across the joints (USBR 1976).

10-49
10-4.4 Acceptance Criteria

Acceptance Criteria should be consistent with those given in Chapter III and Chapter IV of
these Guidelines. The minimum sliding factor of safety for the worst static case should be 1.5.

The design must include adequate scour control measures if water passes over the crest, in
order to prevent undermining the structure. Scour may be controlled by a concrete stilling
basin apron for a short distance below the dam, the end of the apron being protected by an
extension of rip-rap. The lower end of the apron is often further protected by a vertical
diaphragm of concrete or sheet-piling, which serves to retain the foundation under the apron if
the rip-rap is washed away. A vertical diaphragm is often also used at the toe of the dam to
provide for a possible failure of the apron.

The size and extent of the riprap required in the exit area depends upon the effectiveness of the
stilling basin, tailwater depth in the exit and configuration of the exit area. The maximum size
is dependant on velocity and turbulence of water exiting the spillway apron. General
guidelines for riprap design are given in USACE (1987). Model and computed results suggest
a D50 size of 1.5 to 2 feet for velocities of 15 ft/sec and 3 to 4 feet for velocities of 30 ft/sec.
Thickness for placement in the dry should be 1.5 D100 (max) or 2.0 D50 (max), whichever is
greater. Thickness for placement underwater should be increased 50 percent. A bedding layer
must be designed according to established filter criteria and placed under the riprap protection.

10-4.5 Material Properties

The construction materials for concrete dams on soil foundations are generally the same as for
concrete gravity dams discussed in Chapter III, Section 3-6, of these Guidelines.

The geotechnical and subsurface investigations at the site must be adequate to determine the
suitability of the foundation. FERC requirements are given in Chapter V of these Guidelines.
Emphasis should be to determine the in-situ shear strength of the foundation soils and the
presence of any weak layers.

Shear strength of the interface between concrete and soil can be assumed equal to the soil shear
strength if the concrete has been cast directly against soil. If the concrete has not been cast
directly against soil, such as in the case of the vertical surface of a retaining wall, the shear
strength of the interface may be assumed to be 2/3 of the soil shear strength.

10-4.6 References

USACE, (U.S. Army Corps of Engineers), "Engineering and Design, Gravity Dam Design,”
EM-1110-2-2200, September 1958.

USACE, "Hydraulic Design of Flood Control Channels," EM1110-2-1601, July 1991.

USACE, "Hydraulic Design of Navigation Dams," EM-1110-2-1605, May 1987.

10-50
USACE, "Hydraulic Design of Spillways," EM-1110-2-1603, March 1965.

USACE, "Retaining and Flood Walls," EM-1110-2-2502, September 1989.

USBR, "Design of Small Dams," 1987.

USBR, "Embankment Dams," Design Standard No. 13.

10-5 Timber Dams

Timber dams include all dams that rely on timber for structural support and include timber
buttress dams, timber crib dams, and embankment dams with timber cribbing used as
reinforcement to steepen the slopes. Timber faced rockfill dams that rely on the timber only to
limit seepage, or provide erosion protection, and embankments that have wooden trestles
buried in them do not rely on the structural integrity of the wood for stability and are not
considered to be timber dams.

Many of the early dams in the United States were built with timber. These were generally
small dams that were designed empirically. An early text by Wegeman (1905) discusses the
following six types of timber dams.

Brushwood Dams. Alternating courses of brushwood and gravel. Three- to five-foot thick
courses of saplings and trees (with branches upstream) are sunk by placing stone and gravel on
them. The dam was finished by covering the slopes with planking or rip rap.

Log Dams. Horizontal courses of logs with several inches of saplings, brush, stone, and earth
placed between them. The butt ends of the logs were place downstream so that the dam had a
triangular shape in cross section. The entire structure was covered with stone and earth.

Pile Dams. Consisted of one to three rows of timber piles driven vertically across the river
with logs and brushwood placed horizontally between or against the piles.

Plank Dams. Formed by placing 2-inch by 12-inch by 12-foot planks to form vertical arches,
convex upstream. Each course of planks was spiked or nailed to those beneath. On rock
foundations, a single arch was used with the lower planks anchored to the foundation by iron
bolts. On soil foundations, two arches were used with the space between filled with earth,
gravel, or stone.

Timber Crib Dams. Consisted of square timber cribs filled with stone or gravel. The cribs
were constructed by placing layers of timbers spaced 6 to 10 feet apart at right angles to the
previous layer. The structure was faced with wooden planks spiked into the timber cribs. If
placed on rock foundations, the bottom course of logs was fastened to the foundation by iron
bolts.

10-51
Timber Buttress Dams. Deck and frame timber dams. Timber buttresses supporting a
wooden plank face. These dams were typically spiked together. If founded on rock they were
bolted to the rock with iron bolts. If placed on soil foundations, wooden sheetpiling was
driven vertically from the heel (upstream toe) to form a seepage cutoff.

Few significant timber dams were built after the 1930s. Of the six designs discussed by
Wegeman, only the timber crib and the timber buttress dam designs were durable enough for
the structures to survive to the present. Many have been rehabilitated or modified to improve
their integrity. Typical problems include deterioration of timber, undermining, excessive
seepage, and wash out of rockfill. Modifications have included removal and replacement of
the timber, concrete capping, and buttressing. Remnants of the other types may still be found
in some dams.

Figures 10-5.1 and 10-5.2 show typical cross sections through timber crib and timber
buttress dams. Creager, Justin, and Hinds (1945) discuss the timber crib dam as follows:

10-52
 In this type of timber dam, cribs of round or squared timbers are drift-
bolted together, filled with rock fragments or boulders, and topped by
a plank deck. The timbers are usually spaced about 8 feet centers
both ways. The bottom timbers of the cribs are often pinned to the
rock foundation if the site is not submerged.

As much as 25 percent of a timber crib dam was wood. The logs were flattened at the
ends to keep the sides of the cribs vertical and were spiked together with 3/4 inch square
iron or steel bolts. Hardwood spikes were sometimes used in place of iron spikes.

On rock foundations, the upstream face was often sloped to improve stability by taking
advantage of the weight of the water on the sloping deck. The downstream face was
typically vertical and a timber apron extended downstream to prevent undermining.

On soil foundations, the section is usually reversed, with the upstream face vertical and
the downstream face sloping and frequently stepped to limit erosion and undermining
caused by overflowing water. The dams were typically wider than high to control
underseepage. Often, wooden sheetpiling was driven vertically from the heel (upstream
toe) to form a seepage cutoff.

Some timber crib dams were constructed in deep water by constructing the timber cribs
on land, floating them into position, and sinking by filling with rockfill.

Creager, Justin, and Hinds (1945) discuss the timber buttress dam as follows:

It is generally built of squared timbers and planks and is not rock-

filled. For its stability it depends on the weight of the water on its
deck and the anchorage of the sills to the foundation. .... The deck
makes an angle of 30 degrees or less with the horizontal.

The sill, a, (Figure 10-5.2) is first fastened to the ledge rock by wedge
bolts or anchor bolts, preferably grouted in. The struts, b, are then
framed to the sills and held in place by cross-bracing and batten
blocks. The wales, c, are then placed, the entire structure being
thoroughly drift-pinned together. These bents are placed from 6 to 12
feet apart, according to the height of the dam and the size of the
timbers used. Across the bents are placed the studs, d, to which the
lagging, e, is nailed. The lagging should be either tongued and
grooved or lapped and should not be less than 2-inch stuff.

Less than 120 timber crib or timber buttress dams are currently under the jurisdiction of

10-53
the FERC and nearly all of these are small, low hazard potential dams. Less than a dozen
are significant or high hazard potential dams.

Timber cribbing was commonly used for steepening slopes around the turn of the century.
Accurate figures on the number of significant and high hazard potential dams that used
this type of construction are not available. Though timber is still commercially available
and it may be appropriate in some situations such as for corrosive environments or
bulkhead type cofferdams, new timber dams are generally not considered to be
acceptable. The rest of this section is concerned only with existing dams.

10-5.1 Forces

Forces for the analysis of timber crib and buttress dams are generally the same as for
concrete gravity dams discussed in Chapters III of these Guidelines. Forces for the
analysis of timber crib supported slopes are generally the same as for retaining walls.

Uplift pressures for timber crib dams may be different from that commonly assumed for
concrete gravity dams, because of the relative permeability of the dam. The location of
the phreatic surface should be determined as would be done for an embankment dam.
Alternatively, buoyant weight of the wood and rockfill could be used.

Uplift pressures for timber buttress dams should be the same as for buttress dams
discussed in Section 10-2. Many timber crib structures have been grouted or capped with
concrete and may therefore act similarly to a mass concrete structure with respect to
uplift. In such cases, uplift must be considered.

10-5.2 Loading Combinations

Loading combinations for the analysis of timber crib and buttress dams are generally the
same as for concrete gravity dams discussed in Chapter III of these Guidelines. Loading
combinations for the analysis of timber supported slopes are generally the same as for
embankment dams discussed in Chapter IV of these Guidelines.

10-5.3 Analyses

Methods of analyses of timber crib and buttress dams are generally the same as for
concrete gravity dams discussed in Chapter III of these Guidelines. Methods of analyses
of timber supported slopes are generally the same as for embankment dams discussed in
Chapter IV Guidelines.

10-54
Low hazard potential timber dams and timber supported slopes whose failure would not
jeopardize the safety of the dam do not require extensive evaluation. A cursory
evaluation combined with vigilant surveillance for signs of distress should be sufficient
for these structures.

Significant and high hazard potential timber dams warrant detailed stability analyses.
The global stability of timber buttress and timber crib dams should be evaluated similar to
concrete gravity dams, except that cracked base analyses are not appropriate.

Stresses at connections and within timber members of timber crib and buttress dams and
timber crib supported slopes may be estimated with conventional structural analysis,
mechanics of materials, and appropriate simplifying assumptions.

10-5.4 Acceptance Criteria

Acceptance criteria for timber crib and buttress dams are generally the same as for
concrete gravity dams discussed in Chapter III of these Guidelines. Acceptance criteria
for timber crib supported slopes are generally the same as for embankment dams
discussed in Chapter IV of these Guidelines.

Connections between timbers should meet criteria in the National Design Specification
for Wood Construction (American Forest and Paper Association 1991).

10-5.5 Material Properties

The strength of timber is highly dependent on the amount of deterioration that has
occurred as well as the type of wood. For existing significant and high hazard potential
timber dams, field inspections and investigations are extremely important to evaluate the
extent of deterioration. Field investigations may also be appropriate to determine the
species of wood, dimensions, and spacing of timbers. Field investigations that are
appropriate may range from simple inspection by divers to exposing timbers in the fill.
Whenever possible, a complete internal inspection of timber buttress dams is essential.
Field investigations are also appropriate to determine the type, construction details, and
condition of the connections between timbers.

10-5.5.1 Timber Deterioration

Although timber may last indefinitely when permanently submerged in fresh water or
surrounded by saturated soil, untreated timber in air, unsaturated soil, partially saturated
soil, or brackish water is susceptible to damage from decay, insects, and wood borers.

10-55
Reservoir fluctuations can result in wetting and drying cycles that are extremely
detrimental to wood. Also, at times, deterioration of wood may not be visible.

Decay is caused by fungi or bacteria. Decay from fungi can completely destroy wood
(dry rot). Wood that is exposed to the atmosphere is usually too dry to support fungus
growth. Wood that is below water, below the phreatic surface, or embedded in clayey
material is usually too wet to support fungus growth. Decay typically occurs in wood
that is partially saturated. In some cases, wood submerged in water may be attacked by
bacteria resulting a slow decay process (American Institute of Timber Construction
1985).

Damage from termites is the most common type of damage from insects. They are found
throughout the United States, except for the north central states. They can completely
destroy timber above the water level. Other types of insects that can damage wood
include powder-post beetles and carpenter ants. Damage from these insects is usually
minor (American Institute of Timber Construction 1985).

Marine borers inhabit salt or brackish waters and can severely damage wood below water
level within a matter of months or years (American Institute of Timber Construction
1985). However, few if any hydroelectric projects under FERC jurisdiction are located in
salt or brackish water.

10-5.5.2 Timber

Wood exhibits a wide range of engineering properties between species and even within
species because of the differences in growth. Physical and mechanical properties of
many species of wood are summarized in the Establishing Clear Wood Strength Values
(ASTM 1981) and the Timber Construction Manual (American Institute of Timber
Construction 1985).

Allowable stresses in several types of timber are given in Design of Pile Foundations
(USACE 1991). Methods of establishing design stresses for timber piles and poles may
be adapted for timber members in dams (ASTM 1974 and 1979).

For initial evaluations, estimates of engineering properties of wood may be developed

knowing the species of wood, its condition, and tables of typical strengths. If
unsatisfactory results are obtained, material properties should be developed from
laboratory tests of a representative number of specimens (ASTM 1983a and 1984).

10-56
10-5.5.3 Rockfill

Testing of rockfill materials is often difficult and expensive because the large rockfill
particles require large scale equipment and tests. It is usually sufficient to assume
material properties based on tests on similar materials. Rockfill shear strength properties
are summarized by Leps (1970).

10-5.6 References

American Forest and Paper Association, "National Design Standard for Wood
Construction," Washington, DC, 1991.

American Institute of Timber Construction (Englewood, Colorado), Timber Construction

Manual, 3rd Edition, John Wiley & Sons, New York, 1985.

ASTM (American Society for Testing Materials), "Standard Specification and Methods
for Establishing Design Stresses for Round Timber Construction Poles," D-3200,
Philadelphia, Pennsylvania, 1974.

ASTM, "Method for Establishing Design Stresses for Round Timber Piles," D-2899,
Philadelphia, Pennsylvania, 1979.

ASTM, "Establishing Clear Wood Strength Values," D-2555, Philadelphia, Pennsylvania,

1981.

ASTM, "Small Clear Specimens of Timber, Methods of Testing," D-143, Philadelphia,

Pennsylvania, 1983a.

ASTM, "Specific Gravity of Wood and Wood-base Materials," D-2395, Philadelphia,

Pennsylvania, 1983b.

ASTM, "Static Tests of Timbers in Structural Sizes," D-198, Philadelphia, Pennsylvania,

1984.

Creager, P., Justin, J., and Hinds, J., Engineering for Dams, John Wiley & Sons, New
York, 1945.

Leps, T.M., "Review of Shearing Strength of Rockfill," Journal of the Soil Mechanics and
Foundations Division, American Society of Civil Engineers, July, 1970.

10-57
USACE (U.S. Army Corps of Engineers), "Design of Pile Foundations," EM 1110-2-
2906, January 15, 1991.

Wegeman, E., "Design and Construction of Dams," 4th Edition, John Wiley & Sons, New
York, 1905.

10-6 Inflatable Dams

An inflatable dam consists of a sealed, inflatable, rubber-coated fabric tube anchored to a

concrete foundation constructed across a watercourse as shown on Figure 10-6.1. It is
raised by inflating with air, water, or a combination of the two. When it is inflated to full
design height, it impounds water and acts like any other fixed dam in this respect.
However, it is capable of being completely deflated to allow maximum run-off during a
storm, thereby reducing upstream flooding, and to allow passage of sediment, debris, and
ice.

The first inflatable dam was installed in 1957 in the United States and in 1964 in Japan.
There are no known United States standards for inflatable dams. As a result, the Japanese
standards have been referred to for this section (LDTRC 1983).

Typical dimensions of an inflatable dam are generally less than 10 feet in height and 200
feet in length. However, heights up to 20 feet and unlimited lengths are possible. An
inflatable dam has a service life of approximately 30 years and is adaptable for various
installation. Typical uses are for diversion structures, check structures for flood control,
overflow weirs, flashboard and gate replacement, sluice gates, control gates, and barriers
for erosion.

10-58
The more important issues to be considered in their use are:

• Maximum overflow depth - if water overflow depth exceeds a certain

limit, overflow pulsation may cause detrimental vibration in the tube.

• V notch effect - when air is used for inflation, a V notch forms during
deflation of the tube resulting in flow concentration and difficulty in
controlling a desired water level and flow rate. The location of the V
notch should be controlled, otherwise it can happen anywhere and even
move around. Since all the flow is concentrated in the V, adequate
erosion protection must be planned for the area.

10-59
 • Water level change and tube deformation - change of upstream and
downstream water levels deform the tube and, consequently, change the
dam height.

• Wave effects - tension fluctuation is produced by waves and may result in

fatigue of the tube and fittings. Because of this, and the stress fluctuation
caused by inflating and deflating, fatigue resistance must be guaranteed by
the manufacturer.

• Flattening - under the conditions of a small upstream to downstream water

level difference and small flow velocity, the tube may not totally flatten,
thus increasing possible wear and/or damage to the tube.

• Prevention of damage to the tube - in a river with many rolling stones or

flowing debris, the tube and fittings are vulnerable to wear and damage.

• Effect of silt - if a significant amount of silt accumulates on the tube while

deflated, the tube may be difficult to inflate.

• Repair - when an inflatable dam is continuously submerged, repair in the

wet is difficult to impossible.

• Vandalism is often a problem at inflatable dams. People tend to shoot at

them. Compressors must have sufficient capacity to maintain the dam in an
inflated state even with some holes in it.

10-6.1 Forces

Forces on an inflatable dam are generally the same as for concrete gravity dams and
embankment dams, as discussed in Chapters III and IV of these Guidelines. Only
exceptions, clarifications, and additions are discussed below.

10-6.1.1 Dead Loads

The effect of sediment buildup against an inflatable dam located at riverbed level may be
significant and should be evaluated.

10-6.1.2 External Hydrostatic Pressure

The effect of dynamic forces caused by water flowing over the dam may require
consideration in special cases. During deflation it is possible for the water pressure to

10-60
pinch off sections of the dam trapping air in the middle making them impossible to fully
deflate.

10-6.1.3 Ice Pressure

Ice pressure is normally not of great importance in the analysis of inflatable dams.
Inflatable dams naturally pass ice (i.e., the ice pressure pushes the crest down slightly so
it can pass). The excellent performance of the inflatable dam under ice conditions is due
to its ability to deflect and absorb the thermal expansion of ice and the impact of ice
flows. The tube also presents a sloped front to the ice, thus allowing it to ride up and
over it, similar to the design of a sloped pier.

10-6.1.4 Temperature Effects

To consider a load caused by temperature change, a ±15EC temperature change and a

coefficient of thermal expansion of 1 x 10-5 per EC should be adopted (LDTC 1983).

10-6.2 Loading Combinations

Loading combinations are generally the same as for concrete gravity dams, as discussed
in Chapter III of these Guidelines.

10-6.3 Analyses

After the design parameters are established the tube suppliers generally perform the
design and analysis of the superstructure, which includes the tube and tube attachment
fittings. The substructure, which includes the piers, side walls, tube mounting base plate,
aprons, and revetment, is generally analyzed by the inflatable dam designer. The stability
of the substructure must be evaluated as with any other gravity structure. The
manufacturer may provide the force on the anchoring strip for the dam, but if not, all
vertical and horizontal forces must be determined as with any other structure.

10-6.3.1 Superstructure Analysis

Tube. The tube shape and tension are determined based on external pressure, internal
pressure, and weir height.

Tube Attachment Fittings. Design of the tube attachment fittings should account for
the tension in the tube's rubberized fabric and nut tightening forces. The fittings should
be constructed so that the force in the tube's rubberized fabric is evenly held.

10-61
10-6.3.2 Substructure Analysis

The substructure includes a concrete base slab, side walls, tube mounting bed plate, and
could include interior piers. The substructure stability should be analyzed by following
the applicable methods presented in Chapter III (Gravity Dams), Chapter X, Section 10-3
(Pile Foundations) and Section 10-4 (Soil Foundations) of these Guidelines. Metal parts
should be corrosion protected.

Tube Mounting Bed Plate. The width of the tube mounting bed plate required is
dependent upon the height of the dam and the inflation medium used. The bed plate
needs to be wide enough to allow sufficient room for the dam to lie flat when deflated.
The tube mounting bed plate is designed to support the upper load and ensure tube water
and air tightness. Stress of the tube mounting bed plate is calculated as a beam on an
elastic floor or a cantilever fixed on the pier or side wall, depending on how it is
constructed.

10-6.4 Acceptance Criteria

10-6.4.1 Pool Level

The tube does not deflate evenly, that is, the crest of the dam does not remain straight and
parallel in relation to the foundation as it lowers. Discharging water rushing over a
deflating dam increases the pressure acting on it, which tends to push it down in a
particular place, creating a depression shaped like a "V" notch as the dam deflates to
around 70-90% (depending on its height and length) of its normal operating height. The
V-notch has the effect of concentrating water discharge at the point where this depression
in the tube occurs, which makes water level adjustment difficult. To facilitate the start of
deflation, a "V" notch (1 vertical to 10 or 20 horizontal) is formed by the foundation pad.
Deflation then starts over this area.

10-6.4.2 Inflating Medium

Tube inflating medium should be selected from among water, air, and co-use of both
considering the following:

• Natural conditions of the site (ambient temperature, foundation quality, fill

water availability.

• Purpose and operation policy (required function, operation frequency).

10-62
 • Maintenance and control.

• Economics.

10-6.4.3 Superstructure

The tube should be designed to satisfy the following requirements:

• Necessary dam height for all combinations of design water levels and
design flow rates.

• Air and/or water tightness.

• Easily and adequately raised, and completely flattened in deflation.

• Sufficiently durable (service life of at least 30 years)

• No harmful vibrations.

• Convenient for maintenance.

Factors of Safety. The LDTRC (1983) recommends a factor of safety of 8 or more, for
usual loading, be used for the tension in the tube after rubberization and adhesion. The
attachment fittings should use a factor of safety of 3 or more. For earthquake design, the
usual factors of safety can be reduced by 2/3.

Vibration. Due to the flexible structure of the inflatable dam, vibration takes place in
the tube when the overflow and downstream water depth become greater. Generally, the
higher the internal pressure, the more difficult it is for the tube to vibrate. In the case of
the water-filled type, the tendency is that the larger the downstream water depth, the more
easily the tube vibrates. The maximum recommended overflow depths from the
viewpoint of vibration are (JIID 1989):

Air Inflation Type h = 0.2H

Water Inflation Type

downstream is exposed jet flow h = 0.5H

10-63
 downstream water level higher h = 0.4H, where (P/H 2.5 - 3.0)
h = 0.3H, where (P/H 2.0 - 2.5)
h = 0.2H, where (P/H 1.5 - 2.0)

where h = overflow water depth (ft)

H= tube height at time of overflow (ft)
P = Tube bottom internal pressure (water pressure in ft)

If the expected maximum overflow depth exceeds the above-mentioned maximum,

vibration is possible. A spoiler or deflector (fin) should be provided to separate the nappe
from the tube and reduce anticipated vibration.

External Damage. Adequate acceptance criteria should be established for external

damage. Inflatable dams are susceptible to damage by abrading, cutting, and puncturing.
Moving rocks, floating debris (such as trees), and vandalism are all sources for damage.
The inflatable dam lies flat on its foundation (producing no obstruction) when deflated,
which allows it to get out of the way of heavy flood-swept debris. When deflated, debris
may land on the tube causing damage. To prevent such damage, polyethylene-based
foam blocks may be placed inside the tube to act as cushioning against water borne
debris. The amount of cushioning varies depending on the site conditions. The minimum
and maximum amount of cushioning are 1.6 inch and 14.4 inches, respectively
(Sumitomo 1991/1992).

Inflatable dams are highly resistant to wear and puncture. However, with considerable
effort they can be cut and can suffer bullet damage. Knife cuts are more of a concern
than bullet holes because the damage is larger and more difficult to repair. If the tube is
punctured or cut it will usually not cause the dam to deflate since there is a sufficient air
or water supply to maintain inflation until the damage can be repaired.

To deal with the threat of vandalism in high-risk areas, one manufacturer has developed a
tube employing ceramic chips embedded in its outer lamination to provide protection
from knife cuts.

10-6.4.4 Substructure

The substructure of the inflatable dam should be designed to be safe for expected loads,
provide watertightness as required, and have adequate durability.

The acceptance criteria applicable to the substructure, such as sliding, settlement, bearing
capacity and seepage are discussed in Chapter III and Chapter X, Section 10-3 (pile
foundations), and Section 10-4 (soil foundations) of these Guidelines.

10-64
The foundation support should be selected considering the loading, effect of water flow,
ground conditions, construction work, environmental conditions, safety, and economy.
The foundation must not deform to a point where it impacts the inflation/deflation and
pipes embedded in the tube mounting bed plate.

10-6.5 Material Properties

Inflatable dams have relatively simple foundation and structural requirements. A bed
plate holding a rail and bolt assembly anchors the dam body with clamps and nuts.
Inflation and deflation involves piping that leads from the dam to an intake/exhaust valve
and pump system.

10-6.5.1 Site Investigation

The foundation investigations should follow the recommendations of Chapter V of these

Guidelines. Bends in the river channel should be avoided. The plan configuration of the
dam should be straight and normal to the flow.

10-6.5.2 Tube

The body of an inflatable dam must withstand tension, be air and watertight, be resistant
to ozone to prevent deterioration from sun exposure, be abrasion resistant to minimize
wear from rock transported in the stream, and be resistant to low and high temperatures.

Fabric. The fabric maintains, as a tension member, the tensile force caused by the
internal pressure of the tube and the external water pressure. The fabric is normally a
rubber and nylon laminated sheet. The stress bearing component is usually nylon.
Depending on the load and environmental requirements, there are various fabrics
available in terms of weaves, cord counts and weights.

A synthetic rubber is usually incorporated to provide air and watertightness and to protect
the fabric. The rubber used should be designed with weatherability, ozone resistance,
water tightness, and abrasion resistance taken into account. An Ethylene Propylene
Diene Monomer (EPDM) based rubber or Chloroprene Rubber (CR) base are commonly
used.

A number of fabric plies are laminated into a sheet to handle stress requirements.
Strength is achieved by the lamination of fabric plies, and wear and weather resistance by
the thickness. There are no known United States standards for inflatable dams.
Appendix Table 1 in JIID (1989) lists properties of the tube.

10-65
Most repair is not difficult, rarely necessary, and can, depending on the position of the
damage and river flow, normally be done without deflating the dam. In the case of minor
damage (less than about 0.4 inch), it can be repaired using the rubber plug repair method
that is used to repair the tubeless type tires. On the other hand, if the damage is relatively
large (about 0.4 inch or more), a patch type repair as used for repairing a conveyor belt,
etc. is carried out. In this instance, the repair has to be carried out with the bag deflated
and in the dry (LDTRC 1983).

Dimensions. There is no limit to the length of an inflatable dam span from the
standpoint of stress due to the fact that stress is evenly distributed along the entire length
of the anchor line rather than concentrated at selected points. The length of a inflatable
dam generally does not exceed 300 to 400 feet due to the economics of shipping and
handling during installation. For longer dams concrete piers are used between each dam
section.
The fabrics used for dam construction are capable of withstanding forces in dams up to
about 20 feet in height. Under special circumstances (considerable downstream head
against the dam), the total height of the dam may be safely increased.

Anchoring System. The clamping system acts as an anchor for the inflatable dam while
also creating an air and watertight seal. This type of system provides for simple, fast, and
assured installation at the dam site. The fixing method of the tube may include a single
clamping line system (upstream side) or a dual clamping line system (upstream and
downstream sides). In cases where there is a downstream water level or wave forces
from the downstream are significant, it may be necessary to provide a dual clamping line
system.

The fitting system consists of an embedded plate, clamping plate, and anchor bolts.
Galvanized steel is generally used. Stainless steel is used if salt water is expected.

10-6.5.3 Inflation/Deflation System

Inflation and/or deflation of the dam can be simple or complex, depending on the
requirements of the structure.

Safety apparatus that will prevent overinflation and ensure deflation should be provided.
Adequate provisions must be made to ensure that the inflation/deflation system is
operable at all times, particularly during adverse weather conditions. The system and
backup power must be operated at least once each year, either during regular project
operation or on a test basis.

10-66
The piping connecting the tube with the operation chamber includes the inflation medium
supply and exhaust piping, the internal pressure sensing piping, and the drain piping. The
piping connecting the operation chamber with the upstream and downstream sides of the
tube includes the upstream water level detection pipe and the water exhaust pipe (water
type only). The piping should be designed with sufficient capacity and durability based
on the application and place of installation.

The size of the operating room depends on the inflation/deflation system chosen, the size
of the dam, and the manufacturer of that system but is generally about 110 square feet.
The operating room should be designed for safe and easy access and dry conditions, be
equipped with lighting and ventilation, and display an instruction board of operating
procedures in a suitable place.

Inflation System. The inflation system supplies inflating media to the tube. An engine-
or motor-driven blower is used for air type tubes while an engine- or motor-driven pump
and ancillary devices such as valves are used for water type tubes. The type, number, and
capacity of units composing the power equipment should be selected taking dam size,
design inflation time, control reliance, operation frequency, and cost into consideration.
The types of valves and their operating forces should be selected based on the type of
power equipment and the size of the dam.

The time needed for inflation is determined by the capacity of the pump or blower,
diameter of feeding and discharging pipes, inner capacity, inner pressure, and loss rate of
the tube. The inflation time required is normally within 10 minutes to 1 hour.

Deflation System. The deflation system removes the inflation medium from the tube. In
addition to manual exhaust, an automatic deflation device that is linked with the upstream
water level is normally provided.

Automatic deflation systems include mechanical float and bucket types, and electrical
types. The float and bucket types introduce the river water into the operation chamber
through an upstream water level detector pipe, and mechanically opens the valves by way
of a buoyancy float or the weight of water. The electrical type detects the upstream water
level by way of a water level indicator and opens electric valves, which is often jointly
used with the float or bucket types, as a backup.

The deflation time differs depending on the size of dam, piping size and length, and
reservoir capacity. The time for deflating is determined by the upstream and downstream
water levels and internal pressure. Since these conditions are not constant, the deflating
speed is not constant. Deflation time should be in accordance with the design criteria,

10-67
generally from 10 minutes to about 1 hour in many cases. Normally the system is
designed so that the deflation time can be adjusted by changing the opening of the
discharge port.

10-68
10-6.6 References

Anwar, H.O., "Inflatable Dams," Journal of the Hydraulics Division, ASCE, Vol. 93, No.
HY3, Proceeding Paper S239, pp 99-119, May 1967.

Bridgestone Engineered Products, Co., Questions & Answers About the Bridgestone
Rubber Dam, July 1991.

Bridgestone Tire Co., Ltd., Bridgestone Inflatable Rubber Dam, Catalog No. R4720E-1.

Bridgestone Tire Co., Ltd., Bridgestone Rubber Dam: Cutting and Abrasion of Rubber
Dams by Rolling Stones.

Bridgestone Tire Co., Ltd., Construction of a Rubber Dam at River Indus, Sheung Shui,
N.T., Hong Kong.

Bridgestone Tire Co., Ltd., The Functions and Operating Installations of New Rubber
Dam, RX-001.

Bridgestone Tire Co., Ltd., New Bridgestone Rubber Dam: Method for Repairing, RT-
006.

Bridgestone Tire Co., Ltd., New Bridgestone Rubber Dam: Results of Hydraulic
Experiments, 1982.

Hunt, Rodney, Inflatable Rubber Dams, Sumigate, 1991.

Hunt, Rodney, Sumigate Inflatable Rubber Dam: 20 Questions, 1992.

JIID (Japanese Institute of Irrigation and Drainage), "Headworks - Volume 2", March
1989.

LDTRC (Land Development Technical Research Center), Committee for Preparation of

Technical Standard of Rubberized Fabric Inflatable Wiers (Second Draft), Technical
Standard of Rubberized Fabric Inflatable Weirs (Second Draft), August 1983.

Obermeyer Hydro Accessories, Inc. "Pneumatically Operated Crest Gates"

Sumitomo (Sumitomo Electric Europe SA), "The Inflatable Rubber Dam - 30 Years On",
The Commonwealth Minister Reference Book, Kensington Publications Ltd., London,
1991/1992 Edition.

10-69
Sumitomo Electric Industries, Ltd., "Inflated Ideas", New Civil Engineer, September
1990.

10-7 Stone Masonry Dams

Stone masonry dams are constructed of closely placed large stones with the spaces
between the stones filled with mortar. (Dry-laid-stone dams are not considered here.)
Stone masonry dams were common in the United States before about 1915.

Most stone masonry dams were made of unshaped quarried stone blocks of irregular sizes
known as "rubble masonry." The size of the stone blocks varied, depending on the nature
of the local rock, from several cubic feet to several cubic yards and weighed from several
hundred pounds to about 5 tons. Stone masonry dams were constructed without regular
lifts or vertical contraction joints. The stone blocks were fit together as tightly as
possible, but often had large gaps between. Small rock chips or "spalls" were often
placed with the mortar between the larger blocks. The mortar was composed of varying
proportions of sand and cement, and was either packed into place by hand or grouted in
place. Roughly 50 percent of a stone masonry dam was stone blocks, 25 percent spalls,
and 25 percent mortar.

Many stone masonry dams were faced with ashlar masonry - stone blocks hewn into
rectangular shapes of uniform sizes and fit tightly together. Typically every third or
fourth stone know as a "header" was longer than adjacent stones and extended
horizontally into the dam to interlock the facing with the rest of the dam.

Stone masonry dams are distinguished from cyclopean concrete dams in which large
stone "plums" are bedded in concrete with the spaces between the stones filled with
concrete. Typically about 25 percent of the dam is composed of large stones.

10-7.1 Forces

Forces for the analysis of stone masonry dams are generally the same as for concrete
gravity dams, which are discussed in Chapter III, Section 3-2 of these Guidelines or the
chapter on arch dams. Only exceptions, clarifications, and additions are discussed below.

For gravity stone masonry dams, uplift should be considered as it would be for concrete
gravity dams. The argument is sometimes made that the uplift pressures would be
relieved through unmortared or cracked mortared joints. Although there may be some
merit it this argument, reduced uplift should not be used unless it is verified by
piezometers.

10-70
10-7.2 Loading Combinations

Refer to Chapter III.

10-7.3 Analyses

Both gravity and arch dams were built of stone masonry. Stone masonry gravity dams
were designed to avoid tension, but uplift was not considered. They should be evaluated
similar to concrete gravity dams discussed in Chapter III of these Guidelines. Stone
masonry arch dams were designed before analytical tools to evaluate arch stresses were
developed. They should be evaluated the same as arch dams.

Stability should be evaluated for all kinematically possible failure surfaces through the
dam. Although the principal of avoiding potentially weak horizontal planes through stone
masonry dams by interlocking stones and avoiding horizontal courses was generally
understood (Wegeman 1905), the possibility that they may exist should be evaluated.
This is especially important for the analysis of flashboards on the upper courses of stone
masonry dams. For flashboard installations, it may be necessary to tie the upper course
of stones used for mounting flashboard pins to deeper courses within the structure.

By review of any construction photographs or drawings and data from similar dams, an
assessment should be made of the extent of the design provisions included in the dam to
prevent sliding, if any. Based on this assessment, conservative assumptions should be
developed for evaluating the sliding resistance of various sliding planes being
investigated.

10-7.4 Acceptance Criteria

Refer to Chapter III.

10-7.5 Material Properties

The unit weight of stone masonry dams can vary considerably and depends on the type
and percentage of rock. Creager (1917) gives typical weights of various types of stone
masonry dams. They vary from 130 pcf for sandstone rubble masonry to 165 pcf for
granite ashlar masonry.

Creager (1945) gives unit weight, compressive strength, modulus of elasticity, Poissons
ratio, shear strength, and coefficient of thermal expansion for a variety of building stones.

10-71
Generally the weakest material in a stone masonry dam is the mortar between the stone
blocks. In many stone masonry dams, the mortar between the stone blocks has
deteriorated resulting in leakage and occasionally deformation. Often the thin layer of
mortar is cracked due to the different properties of the stone and the mortar. Unless there
is a preponderance of evidence otherwise, the mortar should be assumed to be cracked
and have no cohesion.

Rehabilitation of stone masonry dams may require use of stone that is similar in
appearance to the original stone for aesthetic reasons. Care should be taken to select
stone which will limit water absorption so that freeze-thaw damage can be minimized.
Chinking repairs must be done with very hard stone.

10-7.6 References

Creager, W., "Engineering for Masonry Dams," John Wiley & Sons, London, 1917.

Creager, P., Justin, J., and Hinds, J., "Concrete Dams," Vol. 2 of Engineering for Dams,
John Wiley & Sons, New York, 1945.

Raphael, J., "Concrete Gravity Dams," Section 2 of Development of Dam Engineering in

the United States, edited by E. Kollgaard and W. Chadwick, Pergamon Press, Elmsford,
New York, 1988.

Taylor, F. Noel, "Masonry as Applied to Civil Engineering," D. Van Nostrand Co., New
York, 1915.

Veltrop J., "Concrete Arch Dams," Section 3 of Development of Dam Engineering in the
United States, edited by E. Kollgaard and W. Chadwick, Pergamon Press, Elmsford, New
York, 1988.

Wegeman, E., "Design and Construction of Dams," 4th edition, John Wiley & Sons, New
York, 1905.

10-8 Water Retaining Power Plant Structures

This section covers the evaluation of a hydroelectric powerplant structure when the
powerhouse and headworks form a part of the dam and retain water against them. The
design is covered in detail in USACE (1993).

10-72
10-8.1 Forces

Forces are generally the same as for concrete gravity dams as discussed in Chapter III of
these Guidelines. Only exceptions, clarifications, and additions are discussed below.

10-8.1.1 Dead Loads

Dead loads include the weight of the structure itself, including the walls, floors,
partitions, roofs, and all other permanent construction and fixed equipment. The
approximate unit weights of materials commonly used in construction can be found in
AISC (1989 and 1991) and ANSI (1989).

10-8.1.2 External Hydrostatic Loads

The pressure of water in the penstocks should be included as hydraulic thrust resulting
from wicket gate closure, depending upon the assumed conditions.

Since it is sometimes impracticable to protect the powerhouse against flooding at

maximum tailwater elevation, a level should be selected above which flooding and
equalization of interior and exterior water loads will occur.

10-8.1.3 Internal Hydrostatic Loads (Uplift)

Refer to Chapter III, Section 3-2.4, of these Guidelines for a detailed discussion on uplift
forces.

It is nearly always advisable to reduce the uplift pressures on the draft-tube floor by
means of a drainage system. When "floating" or relatively flexible floor slabs are used,
they are not considered in the stability analysis, either as contributing weight or resisting
uplift. When the floor slabs must take part of the foundation load, as is sometimes the
case when the foundation is soil or poor rock, uplift should be assumed and the slab made
an integral part of the draft-tube structure.

10-8.1.4 Compaction Residual Stresses

Compaction of backfill behind an unyielding wall tends to increase horizontal pressures

beyond at-rest values. Estimation of these compaction residual stresses can be done as
prescribed by EM-1110-2-2502, “Retaining and Flood Walls,” USACE 1989, Section 3-
17, “Earth Pressures Due to Compaction.”

10-73
10-74
10-8.2 Loading Combinations

Loading combinations are generally the same as for concrete gravity dams, as discussed
in Chapter III of these Guidelines. Case I and IA should be evaluated for normal loading
conditions.

Case I Usual - Head gates closed, normal power pool, minimum tailwater, draft tube
and spiral case open to tailwater.

Case IA Usual - Headgates closed, normal power pool, minimum tailwater, draft tube and
spiral case empty. For a multi-unit plant, not all units should be assumed
dewatered (50% would be acceptable).

Load cases IIA - Ice and III - Earthquake should be evaluated for both Cases I and IA.

Where only partial installation is to be made under the initial construction program,
consideration should be given to the temporary loading conditions as well as those
anticipated for the completed structures.

10-8.3 Analyses

Stability analyses are generally the same as for concrete gravity dams discussed in
Chapter III of these Guidelines. Selection of the type of analysis should be governed by
the design stage, the type and configuration of the structure considered, and the type of
foundation.

Refer to Chapter III, Section 3-5.6.2 and Chapter X, Section 10-5 of these Guidelines, for
analyses used for structures on soil foundations. If the foundation at the selected site is
entirely soil, or is a combination of soil and rock, special consideration should be given to
the possibility of unequal settlement.

Refer to Chapter X, Section 10-4 of these Guidelines for analyses used for structures on
pile foundations.

10-8.3.1 Flotation

The structure should be adequately stable with respect to buoyant forces. The flotation
safety factor, SFf, is defined as:

10-75
 Ws%Wc%S
SF f'
U&W g

where: Ws Weight of the structure, including weights of fixed equipment and soil
above the top surface of the structure.

Wc Weight of the water contained within the structure that is controlled by a

mechanical operator (i.e., a gate, valve, or pump).

S Any surcharge loads (such as take-off towers or other structures).

U Uplift forces acting on the base of the structure.

Wg Weight of surcharge water above top surface of the structure that is

totally controlled by gravity flow.

Vertical resistance mobilized by friction along the exterior faces of the structure should
be generally neglected (USACE 1987).

The weight of generating machinery should be included in Ws, unless there is reason to
believe that it will be removed and that it makes a significant contribution to the weight
of the structure. Estimates of the weight of the embedded and rotating part of the
generating machinery could be obtained from the equipment manufacturers for the unit
ratings and specific data.

10-8.4 Acceptance Criteria

Refer to Chapter III, Section 3-5, of these Guidelines for a discussion of acceptance
criteria. Only clarifications, exceptions, or additions are discussed below.

10-8.4.1 Flotation Stability

Concrete hydraulic structures should be designed to have the following minimum

flotation safety factors (USACE 1987):

Minimum
Loading Conditions Safety Factor

Case I - Normal Operation 1.5

10-76
 Case IA - Scheduled Maintenance 1.3
[structure dewatered with normal
tailwater(normal water pool)]
Extreme Maintenance 1.1
[structure dewatered with maximum
tailwater(max. water pool)]
Case II - Unusual Operation 1.3
Case IV - Construction 1.3

If the powerhouse has multiple units, the maintenance condition shall consist of 1 of the
units de-watered. It shall be assumed that the machinery remains in the powerhouse, and
therefore its weight shall be counted.

10-8.5 Material Properties

Refer to Chapter III, Section 3-6, of these Guidelines for a discussion of material
properties.

General guidance on foundation properties may be found in Chapter III, Section 3-6.4 of
these Guidelines.

10-8.6 References

ACI (American Concrete Institute), "Building Code Requirements for Reinforced

concrete," ACI - 318, Detroit, Michigan 1995.

AISC (American Institute of Steel Construction), "Manual of Steel Construction

Allowable Stress Design Ninth Edition," M016, 1989.

AISC, "Manual of Steel Construction Load and Resistance Factor Design First Edition,"
M015L, 1991.

ANSI (American National Standards Institute), "Building Code Requirements for

Minimum Design Loads in Building and Other Structures," A58.1, 1989.

AWS (American Welding Society Inc.), "Structural Welding Code Steel," Thirteenth
Edition, D1.1, 1992.

USACE (U.S. Army Corps of Engineers), "Sliding Stability for Concrete Structures,"
ETL 1110-2-256, June 1981.

10-77
USACE, "Floatation Stability Criteria for Concrete Hydraulic Structures," ETL 1110-2-
307, August 20, 1987.

USACE, "Strength Design of Reinforced Concrete Hydraulic Structures," EM 1110-2-

2104, June 1992.

USACE, "Planning and Design of Hydroelectric Power Plant Structure," EM 1110-2-

3001, April 1993.

USACE, “Retaining and Flood Walls,” EM-1110-2-2502, September 1989.

10-9 Cellular Sheet Pile Structures

Cellular sheet pile structures consist of gravity retaining structures constructed of

interlocking steel sheet piling forming adjacent cells that are filled with soil. They can be
built in the wet, thus eliminating the need for dewatering. Cellular sheet pile structures
are usually used as temporary cofferdams during construction. Occasionally they have
been used as permanent retaining walls, fixed crest dams and spillway weirs. The FERC
general policy is not to approve cellular sheet pile structures as permanent replacement
dams.

The design of cellular sheet pile structures is discussed in several references (Belz 1970;
Cummings 1957; LaCroix, Esrig, and Luscher 1970; Swatek 1966 and 1970; USACE
1989b; NAVFAC 1982; USS 1974). Planning, design, and construction of a temporary
structure must be accomplished by the same procedures and with the same high level of
engineering expertise as those required for permanent structures in order to protect
personnel, equipment, and completed work.

There are three general types of cellular sheet pile structures, each depending on the
weight and strength of the cell fill for its stability. The three common cell configurations
and arrangements are the circular cell, diaphragm cell, and cloverleaf cell as shown in
Figure 10-9.1.

Circular Cells. Consist of a series of individual large diameter circles connected by arcs
of smaller diameter. These arcs generally intercept the circles at a point making an angle
of 30 to 45 degrees with the longitudinal axis of the cofferdam. The primary advantage
of circular cells is that each cell is self-supporting and independent of the adjacent cells.
The circular cell can also be filled as soon as it is constructed, and it is easier to form by
means of templates.

10-78
Diaphragm Cells. Comprised of a series of circular arcs connected by 120 degree
crosswalls (diaphragms). The radius of the arc is often made equal to the cell width so
that there is equal tension in the arc and diaphragm.

The stress at the joint of a diaphragm cell is smaller than that at the joint of a circular cell
of an equal design. The diaphragm cell will distort excessively unless the various units
are filled essentially simultaneously with not over 5 feet of differential soil height in
adjacent cells. Unlike circular cells, diaphragm cells are not independently stable and
failure of one cell could lead to failure of the entire cofferdam.

Cloverleaf Cell. Consists of four arc walls, within each of the four quadrants, formed by
two straight diaphragm walls normal to each other, and intersecting at the center of the
cell. Adjacent cells are connected by short arc walls and are proportioned so that the
intersection of arcs and diaphragms form three angles of 120 degrees. The cloverleaf is
used when a large cell width is required for stability against a high head of water. This
type has the advantage of stability over the individual cells, but has the disadvantage of
being difficult to form by means of templates. An additional drawback is the requirement
that the separate compartments be filled so that differential soil height does not exceed 5
feet.

10-9.1 Forces

Forces for the analysis of cellular sheet pile structures are generally the same as for
concrete gravity dams discussed in Chapter III of these Guidelines. Only exceptions,
clarifications, and additions are discussed below.

10-9.1.1 Dead Loads

Dead loads should include the weight of the cell fill and the sheet pile shell.

10-9.1.2 Internal Hydrostatic Loads

The slope of the saturation line is dependent upon the type of fill, the presence of a berm,
and any positive measures taken to control the saturation surface in the cell or the berm
such as weep holes in the cell or drains and pumped wells in the berm. The saturation
level within the cell fill is perhaps the single most important consideration in the design
of the cells; therefore, its location must be estimated with extreme care.

10-79
The location of the saturation line in a cell is usually estimated using empirical
relationships based on the type of cell fill:

Slope Fill

1H:1V Free-draining coarse-grained

2H:1V Silty coarse-grained
3H:1V Fine-grained

These recommendations are conservative for most applications. Each design should be
evaluated for conditions that would tend to raise the saturation line. If both the quality of
the cell fill and the assurance of proper inspection cannot be guaranteed during
construction, full saturation of the cell should be considered for design purposes. Some
conditions that require evaluation are:

• possible leakage from pipelines crossing the cells;

• waves overtopping the outboard (upstream) sheet piles;

• excessive leakage through the outboard piles;

10-80
Figure 10-9.1 Typical Arrangement of Circular, Diaphragm, and Cloverleaf Cells

10-81
10-9.1.3 External Earth Pressures

For computing the external earth pressures, reference is made to EM-1110-2-2503,

USACE (1989b).

10-9.1.4 Berm Pressures

The passive force developed by a berm should be determined by a wedge analysis that
accounts for the intersection of the failure wedge with the back slope of the berm. The
Coulomb method of analysis or a Culmann graphical solution can be used when
appropriate. The resistance provided by the berm should be limited to a value consistent
with the berm reaction resulting from a sliding analysis.

10-9.1.5 Earthquake Forces

The seismic earth pressure and hydrodynamic pressures should be considered in the
design and analysis. In addition to these loads, a reduction in strength of the foundation,
cell fill, or berm can also simultaneously occur during an earthquake. Structures founded
on loose, saturated, cohesionless materials or cohesive soils that contain lenses of loose
saturated, cohesionless soil can lose much of their foundation support when subjected to
earthquake loading. Similarly, the cell fill or the berm can also liquefy, increasing the
lateral loading against the cell.

Further discussion of earthquake forces is given in Chapter III, Section 3-2.6 of these
Guidelines.

10-9.1.6 Surcharge

Generally, the effect of surcharge from equipment working on the top of the cell is not
significant because the horizontal earth pressure resulting from equipment is greatest at a
shallow depth below the top of the cell where the interlock tension is low.

10-9.2 Loading Combinations

Loading combinations for cellular sheet pile structures used as dams are generally the
same as for concrete gravity dams discussed in Chapter III of these Guidelines. Only
exceptions, clarifications, and additions are discussed below.

10-82
10-9.2.1 Case I Usual Loading Combination

A cofferdam structure should be evaluated for drawdown (dewatering). The pool level
inside the cofferdam is some specified distance below the pool level outside the
cofferdam; the cell fill saturation line varies uniformly between the outside pool level and
some specified distance above the pool level inside the cofferdam. This condition is
checked to determine the maximum rate of dewatering. This condition can be critical for
stability and interlock stress. Since the cell fill saturation level is critical, the actual
saturation level must be monitored in the field during dewatering to verify the assumed
conditions.

10-9.2.2 Case II Unusual Loading Combination

Cellular sheet pile structures are often used where they can be overtopped during
flooding. Loading should consider the outside pool at the top of the cell with the cell fill
saturation line assumed to slope from the top loaded face of the cell to the unloaded side
of the cell. Flood gates should be provided to allow the interior of cofferdams to be
flooded before the cells are overtopped.

10-9.2.3 Case IV End of Construction Combination

Forces acting upon a cofferdam can change significantly during construction and the
stability should be evaluated at various stages of construction. For example, overburden
may be present on the inboard side when it is initially dewatered; however, the
overburden may subsequently be excavated, thus perhaps adversely affecting the stability
of the cofferdam. For cellular fixed weir structures with flow over the weir, permanent
upstream and downstream rock berms extending the full height of the cells are usually
constructed for stability and scour prevention. Stability should be evaluated for the case
where the cell is filled before the berms are placed and after the berms are placed.
Maximum interlock stresses will probably occur in the construction condition when the
cells are filled and before the berms are built.

Where the structure serves as a construction cofferdam, evaluate balanced pools on both
the inside and outside of the cofferdam. For determination of maximum interlock stress,
cell fill is assumed to be completely saturated to the top of the cell unless positive
measures are taken to prevent fill saturation.

10-83
10-9.3 Analyses

The stability of a sheet pile cell results from the composite action of the soil fill and the
interlocking steel piling. Because of this composite action, cells cannot be classified as a
traditional concrete gravity monolith or a flexible earth embankment.

Analyses should evaluate the external stability (sliding, overturning, rotation, bearing
capacity, settlement, seepage, and scour) and internal stability (pile interlock tension,
tilting, pullout, penetration). A summary is provided below. A more detailed discussion
is given in USACE (1989b).

The equivalent width, B, of a sheet pile cellular structure is used for design purposes. It
is defined as the width of an equivalent rectangular section having a section modulus
equal to that of the actual structure and may be expressed as A/2L, where A is the area of
a main cell plus one connecting cell and 2L is the center to center distance between the
main cells.

10-9.3.1 External Cell Stability

10-9.3.1.1 Sliding

Sliding along the base or deep seated sliding below the base of most sheet pile cellular
structures can be adequately assessed using a limit equilibrium approach. A two-
dimensional analysis with an assumed plane failure surface is usually sufficient. A more
detailed analysis should be done if unique, three-dimensional geometric features and
loads critically affect the sliding stability of a specific structure.

Considerations regarding displacements are excluded from the limit equilibrium

approach. The relative rigidity of different foundation materials and the sheet pile
cellular structure may influence the results of the sliding stability analysis. Such complex
structure-foundation systems may require a more intensive sliding investigation than a
limit equilibrium approach. The effects of strain compatibility along the assumed failure
surface may be included by interpreting data from in situ tests, laboratory tests, and finite
element analyses.

The possibility of a deep-seated failure along any weak seams below a cellular structure
should be evaluated. Weak seams may exist as clay seams in alluvium or between
competent rock strata in sedimentary rock formations. Seams of pervious sand within a
clay deposit may permit the development of excess hydrostatic pressure, which reduces
the effective stress and subsequently reduces the shearing resistance. A weak seam may

10-84
also appear after excavation due to the removal of overburden pressure, which results in a
drop in the shear strength of clay shale to its residual strength.

If sufficient space is available, a soil berm may be constructed on the unloaded side to
increase the stability against sliding on the base. The berm will also serve to lengthen the
path of seepage and decrease the upward seepage gradients on the unloaded side.
However, the berm will require a larger cofferdam enclosure and an increase in the
overall length of the cofferdam, which will increase construction and maintenance costs.
Also, an inside berm inhibits inspection of the inside piling for driving damage and makes
cell drainage maintenance more difficult. In order for a berm to function as designed, the
berm must be constantly maintained and protected against erosion and the degree of
saturation must be consistent with design assumptions.

10-9.3.1.2 Overturning

A soil-filled cellular structure is not a rigid gravity structure that could fail by overturning
about the toe of the inboard side. Before overturning could occur, the structure must fail
from causes such as pullout of the sheet piles at the heel and subsequent loss of cell fill.
Nevertheless, a gravity-block analysis may serve as the starting point for determining the
required cell diameter. Considering that the cell fill cannot resist tension, the cell should
be proportioned so that the resultant of all forces falls within the middle one-third of the
equivalent rectangular base. This type of analysis will also serve to determine the
foundation pressures.

10-9.3.1.3 Rotation

Hansen's method considers cellular structures to act as rigid bodies (Hansen 1953;
Oveson 1962). Stability, as determined by Hansen's method, is directly related to the
engineering properties of the cell fill and the foundation, and properly considers the
saturation level within the cell as well as seepage forces beneath the cell. Design should
not be based on the Hansen method; rather, it should be emphasized as a sensitivity check
only.

10-9.3.1.4 Bearing Capacity

The bearing capacity of granular soils is generally good if the penetration of the sheet
piles into the overburden is adequate to control seepage of water underneath the cell base.
Clay should be stiff to hard for a good bearing capacity. However, even on relatively soft
soils, cellular structures have been successfully constructed using sand or rockfill berms
(Cummings 1957). The bearing capacity of both cohesive and granular soils supporting
cellular structures can be determined by Terzaghi's method of analysis. However, the

10-85
failure planes assumed for the development of the Terzaghi bearing capacity factors do
not appear to be as realistic as those developed specifically for cellular structures by
Hansen. Hence, for bearing capacity investigation, the Hansen method of analysis should
also be used (USACE 1989b).

The bearing capacity of rock is not readily determined by laboratory tests on specimens
and mathematical analysis, since it is greatly dependent on the influence of
nonhomogeneity and geologic defects on the behavior of rock under load. To allow for
the possibility of unsound rock, a higher factor of safety is generally adopted to determine
allowable bearing pressure.

10-9.3.1.5 Settlement

Two types of settlement can occur within a cellular sheet pile structure: settlement of the
cell fill and settlement of the sheet piles. In some cases, settlement can also be caused by
dewatering the cofferdam area. The impact of such settlement on the structure design and
performance should be evaluated.

The settlement of cell fill occurs under the self load of the fill placed within the cell. For
granular fill, generally a majority of the settlement will have been accomplished soon
after the fill placement. Hence, the postconstruction settlement of granular cell fill under
its own weight is, generally, insignificant. Any volume decrease of the cell fill due to
settlement can always be compensated by placing additional fill in the cell before any
other load is applied to the cell. The settlement of cell fill may also result from seismic
liquefaction of uncompacted sand.

A cellular structure underlain by compressible soils below its base will undergo
settlement due to the weights of the cell and berm fills. If the compressible soils below
the structure continue to consolidate after the reservoir loads have been applied, the
reservoir loads can create a moment on the structure producing an unequal distribution of
pressure at the base of the cell resulting in differential settlement. Dewatering of a
cofferdam may cause drawdown of water levels within soil layers located below nearby
existing structures or utility lines, potentially resulting in undesirable settlement.

10-9.3.1.6 Seepage

Foundation underseepage is generally not a problem for structures built on clay or good
quality rock foundations. Problems almost always are confined to coarse-grained soil
such as gravel and sand and sometimes silty materials. The most serious conditions occur
where undetected pervious seams exist in the foundation. Fill material should be of
sufficient coarseness to prevent material loss through interlocks.

10-86
Major problems associated with seepage below a sheet pile cellular structure are listed as
follows.

• Piping, boils, or heave of the soil mass in front of the toe may occur if the
exit gradient exceeds the critical hydraulic gradient. Boils and heave will
considerably lower the bearing capacity of the soil, potentially resulting in
toe failure of the cell. Piping causes loss of materials underneath the cell
foundation and may cause excessive settlement and eventual sinking of the
cell.

• Upward seepage forces at the toe may excessively reduce the passive
resistance of the soil. This loss of lateral resistance may cause sliding or
overturning failure of the cell.

• Seepage forces acting on the soils at the inboard face of the cell may
excessively increase the hoop stress in the sheet piles. This may increase
the possibility of interlock failure of the sheet piles and result in the loss of
cell fill.

Cofferdams on sand are often designed using a trial sheet pile penetration of two-thirds of
the upstream water head. A flow net analysis is most often used to estimate the seepage
forces. If the exit gradient at the toe of the structure is large, a loaded filter or a wide-
base berm should be considered. Seepage pressures should be estimated as discussed in
Chapter III, Section 3-2.4.4.

10-9.3.1.7 Scour

Scour has contributed to a number of cellular structure failures. By removing the lateral
earth support, the interlock stresses increase. Continued scour exposes sheet piles that
have not penetrated to rock or were driven out of interlock and result in loss of cell fill
and subsequent failure.

Damage by scour should be prevented either by protecting the outside of the cell with
riprap or by carrying the sheets to a greater depth. Deflectors designed to streamline flow
are effective in minimizing scour along the face of the structure.

10-87
10-9.3.2 Internal Cell Stability

10-9.3.2.1 Interlock Tension

A cell must be stable against bursting pressure; that is, the pressure exerted against the
sheets by the fill inside the cell must not exceed the allowable interlock tension. The
interlock tension developed in a cellular structure is a function of the internal cell
pressure and is proportional to the radius of the cell. It is common practice to calculate
the interlock tension in the main cell sheet piling based solely on the internal pressure.
The coefficient of internal pressure is dependent upon the type of cell fill material and the
method of placement. Table 4-2 in USACE (1989) recommends a coefficient in the range
of 1.2 Ka to 1.6 Ka where Ka is the active coefficient of horizontal earth pressure. The
location of the maximum horizontal pressure is dependent on cell restraint and is shown
in Figure 4-16 of USACE (1989). Further changes in the depth of overburden, removal
of berms, saturation level in the cell fill, and rate of dewatering must be anticipated when
determining the maximum interlock tension.

The interlock tension at the connections between the main cells and the connecting arcs is
increased due to the pull of the connecting arcs. For critical structures, special analyses
such as finite elements should be used to determine interlock tension at the connections.

Several methods to reduce interlock tension are given in USACE (1989).

10-9.3.2.2 Tilting

Tilting of cells is resisted by both the vertical and horizontal shear resistance of the soil in
the cell, to which the frictional resistance of the steel sheet piling is added. Vertical shear
resistance is determined by the theory developed by Terzaghi (1945). The horizontal
shear resistance is determined by the theory proposed by Cummings (1957). Both of
these methods of analyses should be used independently to determine the adequacy of the
cell to resist tilting. Additionally, tilting resistance of cells founded on overburden should
be investigated by the theory proposed by Schroeder and Maitland (1979). Refer to
USACE (1989) for further discussion.

10-9.3.2.3 Pullout of Outboard Sheeting

The penetration of sheetpiling is generally determined by the need to control seepage by

increasing the flow path. However, the penetration must also be adequate to insure
stability with respect to pullout of the outboard sheeting due to tilting. The calculated
overturning moments are applied to the sheet piles, which are assumed to act as a rigid

10-88
shell. Resistance to pullout is computed as the frictional or cohesive forces acting on the
embedded length of piling. Typical values of steel sheet piling against various soils are
given in Table 4-3 of USACE (1989).

10-9.3.2.4 Penetration of Inboard Sheets

The penetration of the sheet piles on the inboard side must be sufficient to prevent any
further penetration under loading. The factor of safety against sheet pile penetration is
defined as the ratio of the shear resistance on both sides of the embedded portion of the
piles on the unloaded side to the internal downward shear force on the unloaded side.

10-9.3.3 Dynamic Analysis

When seismic excitation is a design consideration, the structure should be sited, if

practicable, on competent rock. Such siting will automatically preclude the possibility of
strength reduction beneath the structure under earthquake conditions. For structures
situated on sand or clay, a refined analysis to evaluate the effect of cyclic loading on
strength is necessary.

The sliding stability of a sheet pile cellular structure for an earthquake-induced base
motion should be checked by assuming that the specified horizontal and vertical
earthquake acceleration act in the most unfavorable direction. The earthquake-induced
forces on the structure and foundation wedges can then be determined by a rigid body
analysis.

A general discussion of seismic design of cellular sheet pile structures is presented by

Chakrabarti et al. (1978).

10-9.3.4 Finite Element Analysis

The application of finite element model (FEM) analysis to date has been to develop its
state of the art to the point where it can be used to refine existing design techniques and
to analyze potential failure modes that cannot be checked by other methods. All studies
so far have been made by researchers or engineers who are extremely familiar with the
FEM techniques using specialized FEM programs for soil and structure modeling. The
FEM analysis does not yet lend itself to application to typical design engineers working
with currently available general-use programs. USACE (1989) provides further
discussion on the use of FEM.

10-89
10-9.4 Acceptance Criteria

10-9.4.1 Factors of Safety

Considering the temporary application of these structures, the recommended minimum

factors of safety for various potential failure modes are listed in Table 10-9.1.
TABLE 10-9.1
RECOMMENDED MINIMUM FACTORS OF SAFETY

Loading Combination
Failure Mode Usual Unusual Extreme
Sliding 1.5 1.5 1.3
Overturning (gravity block)1 Inside Inside Inside
Kern Kern Base
Rotation (Hansen)1 1.5 1.25 1.1
Bearing capacity
Sand 2.0 2.0 1.3
Clay 3.0 3.0 1.5
Interlock tension2 2.0 1.5 1.3
Vertical shear resistance 1.5 1.25 1.1
(Terzaghi)
Horizontal shear resistance 1.5 1.25 1.1
(Cummings)
Vertical shear resistance 1.5 1.25 1.1
(Shroeder-
Maitland)1
Pullout of outboard sheets1 1.5 1.25 1.1
Penetration of inboard sheets1 1.5 1.25 1.1
Piping 2.0 2.0 2.0

1
Design should not be based on these modes of failure, but rather these analyses should be employed as sensitivity
checks only.
2
The factor of safety against interlock tension failure should be applied to the interlock strength value guaranteed
by the manufacturer for the particular grade of steel. The guaranteed value for used piling should be reduced as
necessary depending upon the condition of the piling.

10-90
10-9.4.2 Cell Deformation

The cell is a yielding structure that can adapt to some deformation. When a cell is
initially filled, it bulges out under the imposed lateral stresses, which causes take-up of
slack in the interlocks and strain in the sheet piles. The maximum bulging usually occurs
at a depth approximately 2/3H below the top of the cell (where H = cell height above
dredge line) and measures on the order of 3 to 6 inches. Additional deformation is
created in circular cells by the pull of the adjacent cells when they are filled. Under the
application of the horizontal load, high cellular cofferdams may deflect at the top
anywhere from 3 to 18 inches. The above order of movements are not expected to lead to
rupturing of interlocks or excessive bending at connections (LaCroix 1970).

10-9.5 Material Properties

The material strength assumptions made during the design process must be verified by the
site investigations and attained in the field during the construction process. Specific
details concerning geotechnical investigations are contained in Chapter V of these
Guidelines. The following general guidance shall apply.

10-9.5.1 Foundation Properties

A detailed foundation investigation should be performed, because the foundation

conditions significantly impact the cost and degree of difficulty in construction and
eventual integrity of the cellular structure. Continuous soil sampling or rock coring
should extend to at least 15 to 20 feet below the design base level of the cell.

The program must be refined through analysis of the geologic details to provide specific
and reliable information on the character of the overburden:

• typical obstacles such as boulders in the overburden that may cause

difficulty in driving the cell sheets and lead to interpretations of a false top
of rock;

• the depth to and configuration of the top of rock;

• the depth and character of bedrock weathering and discontinuities;

• the physical properties of the foundation materials;

• the elevation and fluctuation limits of the ground water;

10-91
 • potential foundation problems and their treatment, such as leakage and
stability.

A reliable estimate of water inflow as well as an accurate determination of the elevation

and fluctuation of the groundwater table are primary concerns in the design and
construction. One method of obtaining information is the field pumping test, which may
be performed to determine the permeability of the foundation materials.

10-9.5.2 Cell Fill

The performance of the sheet pile structure is directly related to the drainage
characteristics of the cell fill. Clean, coarse-grained, free-draining granular (soils with
less than about 5 percent of the particles by weight passing the No. 200 sieve and 15
percent passing the No. 100 sieve) soils are preferred for cell fill because they lower the
seepage line. This improves the cell performance by reducing the sheet pile interlock
force, increasing the lateral sliding resistance, and increasing the internal shear resistance.

10-9.5.3 Steel Sheet Piling

New piling in good condition should be considered for major structures. Steel for sheet
piling should conform to the requirements of the following American Society for Testing
and Materials (ASTM) standards:

A328 Steel Sheet Piling

A572 High-Strength Low-Alloy Columbium Vanadium Steels of Structural
Quality
A690 High-Strength Low-Alloy Steel H-Piles and Sheet Piling for Use in
Marine Environments

A328 is the basic sheet piling and is satisfactory for most installations. A572 specifies
high-strength sheet piling and is applicable for use in large diameter (>70 feet) cells
where high interlock strength is required. A690 steel sheet piling provides greater
corrosion resistance than other steels and should be considered for use in permanent
structures in corrosive environments.

Cold-formed steel sheet piling is also available, although this piling has limited
applicability. Presently, there is no ASTM specification covering this piling.

An extruded wye, using A572, Grade 50 steel, is available on a limited basis. These
wyes have a small cross section and are extremely flexible, thus creating handling and

10-92
driving difficulties. As a result of this characteristic, together with their limited
availability, the use of these extruded wyes is not recommended.

Since tees and wyes are subjected to high local bending stresses at the connection, strong
ductile connections are essential. Welded connections do not always meet this
requirement because neither the steel nor the fabrication procedure is controlled for
weldability. Therefore all fabricated tees, wyes, and crosspieces shall utilize riveted
connections. In addition, the piling section from which such connections are fabricated
shall have a minimum web thickness of one-half inch. Only straight web pile sections
shall be used for cells as the hoop-tension forces would tend to straighten arch webs, thus
creating high bending stresses (USACE 1989b).

When cofferdams are used as permanent structures, especially in polluted brackish water,
or seawater, severe corrosion occurs from the top of the splash zone to a point just below
mean low water level. In these areas, protective coating, corrosion resistant steel, and/or
cathodic protection should be used.

10-9.6 Instrumentation

The kinds of instruments selected will depend on the purpose, project conditions, and the
variables that will be monitored. The following is a brief description of the more
common instruments used in a program to monitor steel sheet pile structures:

• Observation wells are mainly used to measure unconfined ground-water

levels and are monitored directly by a probe or tape. They may be installed
to monitor ground-water levels in the cell fill, backfill materials, and
stabilizing berms.

• Piezometers are used to monitor pore pressures in the cell fill and
foundation, in the stabilizing berms, and in the backfill material.

• Inclinometers can be used to monitor horizontal deformation within the cell

fill, along the length of a sheet pile section, in the cell foundation, and
within the stabilizing berm.

• Earth pressure measuring devices are designed to measure the total stress at
a point in an earth mass or to measure the total stress or contact stress
against the face of a structural element.

• Strain gages are used to observe the interlock tension within sheet pile
members.

10-93
 • Precise measurement systems are used to detect horizontal and vertical
surface displacement by making precise measurements of lengths, angles,
and alignments between reference monuments and selected points on the
structure.

A more detailed discussion of instrumentation is provided in Chapter IX of these

Guidelines.

10-9.7 Construction Considerations

The safety and performance of cellular sheet pile structures are very sensitive to site
conditions and construction practices. Great care must be taken to ensure that the effects
resulting from all potential construction and inservice site conditions, and construction
techniques are properly anticipated, considered, and accounted for in the design. In
addition, construction progress must be closely monitored by design personnel in order to
evaluate or verify design assumptions and to recognize any changed conditions that might
require a design modification.

• All handling holes in the sheet piling on the loaded side of the structure
should be plugged. This is necessary to prevent an objectionable amount of
water from entering the cell or loss of cell fill.
• Sheet piling should not be driven through overburden containing boulders.
Extremely dense overburden should be excavated to a depth such that it can
be penetrated without damaging the piling. Although dependent on the
nature of overburden, 30 feet is generally accepted as a maximum depth to
drive through overburden.

• Setting sheet piling on bare rock should be avoided wherever possible since
support from the overburden is beneficial in helping maintain the desired
cell configuration.

• Wherever cells and fill are placed against sloped or stepped faces of
existing concrete, care should be taken to seal the contact between the sheet
piles and concrete to prevent infiltration of water that could saturate the fill
or cause piping.

10-94
10-9.8 References

Belz, C.A., "Cellular Structure Design Methods," Design and Installation of Pile
Foundations and Cellular Structures, Proceeding on the Conference, Lehigh Valley, PA,
pp 319-338, 1970.

Chakrabarti, S., Husak, A.D., Christiano, P.P., and Troxell, D.E., "Seismic Design of
Retaining Walls and Cellular Cofferdams," Earthquake Engineering and Soil Dynamics,
Volume 1, Proceedings of the ASCE Geotechnical Engineering Division Specialty
Conference, Pasadena, CA, pp 325-341, June 1978.

Cummings, E.M., "Cellular Cofferdams and Docks," Journal of the Waterways and
Harbors Division, American Society of Civil Engineers, New York, NY, Vol. 83, No.
WW3, September 1957.

Grayman, Robert, "Cellular Structure Failures," Design and Installation of Pile

Foundations and Cellular Structures, Proceeding on the Conference, Lehigh Valley, PA,
pp 383-391, 1970.

Hansen, J.B., Earth Pressure Calculations, The Danish Technical Press, The Institution of
Danish Civil Engineers, Copenhagen, 1953.

Lacroix, Y. Esrig, M. and Luscher, U., "Design, Construction, and Performance of

Cellular Cofferdams," Lateral Stresses in the Ground and Design of Earth Retaining
Structures, 1970 Specialty Conference, Cornell University, Ithaca, NY, pp 271-328,
1970.

NAVFAC (Naval Facilities Engineering Command), "Foundations and Earth Structures

Design Manual 7.2," NAVFAC DM-7.2, pp 116-127, May 1982.

Ovesen, N.K., Cellular Cofferdams Calculation Method and Model Tests, Bulletin 14,
Danish Geotechnical Institute, Copenhagen, 1962.

Patterson, John H., "Installation Techniques for Cellular Structures," Design and
Installation of Pile Foundations and Cellular Structures, Proceeding on the Conference,
Lehigh Valley, PA, 1970.

Pile Buck, Inc., "Cellular Cofferdams," Jupiter, FL, 1990.

Pile Buck, Inc., "Steel Sheet Piling Design Manual," Jupiter, FL, 1987.

10-95
Pile Buck, Inc., "Steel Sheet Piling Specifications Chart," Jupiter, FL, 1989.

Schroeder, W.L. and Maitland, J.K., "Cellular Bulkheads and Cofferdams," Journal,
Geotechnical Engineering Division, American Society of Civil Engineers, New York,
Vol. 105, GT 7, Paper 14713, pp 823-837, 1979.

Swatek, E.P., Jr., "Cellular Cofferdam Design and Practice," Conference Preprint 393,
ASCE Transportation Engineering Conference, Philadelphia, PA, October 1966.

Swatek, E.P., Jr., "Summary: Cellular Structure Design and Installation," Design and
Installation of Pile Foundations and Cellular Structures, Proceeding on the Conference,
Lehigh Valley, PA, pp 413-423, 1970.

Swatek, E.P., Jr., “Steel Sheetpile Cellular Cofferdams,” in Chapter 7 of Advanced Dam
Engineering, edited by R. Jamsen, Van Nostrand Reinhold, New York, 1988.

Tennessee Valley Authority, Divisions of Engineering and Construction, "Steel Sheet

Piling Cellular Cofferdams on Rock," Technical Monograph No. 75, Knoxville, TN,
1957.

Terzaghi, K., "Bearing Capacity," Theoretical Soil Mechanics, John Wiley and Sons,
Inc., New York, NY, 1943.

Terzaghi, K., "Stability and Stiffness of Cellular Cofferdams," American Society of Civil
Engineers, Transactions, Vol. 110, Paper No. 2253, pp 1083-1202, 1945.

USACE (U.S. Army Corps of Engineers), "Retaining and Flood Walls," EM 1110-2-
2502, September 1989a.

USACE, "Design of Sheet Pile Cellular Structures," EM 1110-2-2503, September 1989b.

USACE, "Hydraulic Design of Navigation Dams," EM 1110-2-1605, May 1987.

USS (United States Steel), "Steel Sheet Piling Design Manual," February 1974.

10-10 Mechanically Stabilized Earth Dams

Mechanically Stabilized Earth (MSE) walls and slopes are cost-effective soil retaining
structures that can tolerate much larger settlements than reinforced concrete walls. By
placing tensile reinforcing elements in the soil, the strength of the soil can be improved
significantly such that a vertical face of the soil/reinforcement system is self supporting.

10-96
Use of a facing system to prevent soil raveling between the reinforcing elements allows
very steep slopes and vertical walls to be safely constructed.

MSE is defined as any wall or slope supporting system in which reinforcing elements are
placed in a soil mass to improve its mechanical properties. Reinforcing elements is a
generic term that encompasses all man-made elements incorporated in the soil to improve
its behavior. Examples are steel strips, geotextile sheets, and steel or polymeric grids.
The term reinforcement is used only for those elements where stress transfer occurs
continuously along the reinforcement element. Other inclusions may act simply as
tendons between the wall face and an anchorage element.

Facing is a component of the reinforced soil system used to prevent the soil from raveling
out between the rows of reinforcement. Common facings include precast concrete panels,
metal sheets and plates, gabions, welded wire mesh, shotcrete, wood lagging and panels,
and wrapped sheets of geosynthetics.

Geosynthetics is a generic term that encompasses flexible synthetic materials used in

geotechnical engineering such as geotextiles, geomembranes, geonets, and polymer grids
(also known as geogrids).

Suitability. MSE is an attractive design alternative when space is limited. MSE has
been used to raise dam heights, steepen slopes, and replace traditional sloped earth
embankments with vertical faces. In a few cases MSE walls have been used for spillways
however, this is discouraged.

MSE is a relatively new technology, having been used in dams in the United States since
about the 1980's. The use of MSE in design of new dams and the repair of existing dams
is growing rapidly, as is the development of new and different types and uses of
reinforcement. MSE has been used in few projects with installed lifetimes exceeding 30
years. Since the service life of dams should be considered indefinite, prudence demands
that the most durable materials be used. Experience with MSE is short and the long-term
performance of these products is still pretty much unknown.

The potential for catastrophic damage and loss of life in the event of a dam failure
suggests a cautious approach to use of MSE in dams. MSE should only be used:

• where it is not critical to the long-term performance of the dam, and

generally where it can be readily exposed, repaired, or replaced if
necessary;

10-97
 • in a configuration where it does not serve the sole defense against dam
failure. Particular attention must be given to the potential for a domino type
failure, such as one facing unit dislodging, resulting in failure of the
adjacent area.

Types. Table 10-10.1 provides a summary of many of the current systems by proprietary
name, reinforcement type, and facing system. MSE systems can be described by the
reinforcement geometry, the stress-transfer mechanism, the reinforcement material, and
the extensibility of the reinforcement material as shown in Table 10-10.2.

10-98
 TABLE 10-10.1
SUMMARY OF REINFORCEMENT AND FACE PANEL DETAILS FOR
VARIOUS MSE SYSTEMS

SYSTEM NAME REINFORCEMENT DETAIL TYPICAL FACE PANEL DETAIL1

Reinforced Earth: Galvanized Ribbed Steel Strips: 0.16 in Facing panels are cruciform shaped precast
The Reinforced Earth Company (4mm) thick, 2 in (50 mm) wide. Epoxy- concrete 4.9 ft x 4.9 ft x 5.5 in (1.5 m x 1.5 m x 14
2010 Corporate Ridge coated strips also available. cm). Half-size panels used at top and bottom.
McLean, VA 22102

VSL Retained Earth: Rectangular grid of W11 or W20 plain steel Precast concrete panel. Hexagon-shaped, (59-1/2
VSL Corporation bars, 24 in x 6 in (61 cm x 15 cm) grid. in high, 68-3/8 in wide between apex points, 6.5 in
101 Albright Way Each mesh may have 4, 5 or 6 longitudinal thick (1.5 m x 1.75 m x 16.5 cm).
Los Gatos, CA 95030 bars. Epoxy-coated meshes also available.

Mechanically Stabilized Embankment:
Dept. Of Transportation
Div. Of Engineering Services
5900 Folsom Blvd., P.O. Box 19128
Sacramento, CA 95819
Rectangular grid, nine 3/8 in (9.5 mm) Precast concrete; rectangular 12.5 ft (3.81 m) long,
diameter plain steel bars on 24 in x 6 in (61 2 ft (61 cm) high and 8 in (20 cm) thick.
cm x 15 cm) grid. Two bar mats per panel
(connected to the panel at four points).

Georgia Stabilized Embankment: Rectangular grid of five 3/8 in diameter (9.5 Precast concrete panel; rectangular 6 ft (1.83 m)
Dept. Of Transportation mm) plain steel bars on 24 in x 6 in (61 cm wide, 4 ft. (1.22 m) high with offsets for interlocking.
State of Georgia x 15 cm) grid 4 bar mats per panel.
No. 2 Capitol Square
Atlanta, GA 30334-1002

Hilfiker Retaining Wall: Welded wire mesh, 2 in x 6 in grid (5 cm x
Hilfiker Retaining Walls 15 cm) of W4.5 x W3.5 (.24 in x .21 in
PO Drawer L diameter), W7 x W3.5 (.3 in x .21 in), W9.5
Eureka, CA 95501 x W4 (.34 in x .23 in), and W12 x W5 (.39
in x .25 in) in 8-ft-wide mats.
Welded wire mesh, wraps around with additional
backing mat and 1.4 in (6.35 mm) wire screen at
the soil face (with geotextile or shotcrete, if desired).

Reinforced Soil Embankment: 6 in x 24 in (15 cm x 61 cm) welded wire Precast concrete unit 12 ft 6 in (3.8 m) long, 2 ft
The Hilfiker Company mesh: W9.5 to W20 - .34 in to .505 in (8.8 (61 cm) high. Cast-in-place concrete facing also
3900 Broadway mm to 12.8 mm) diameter. used.
Eureka, CA 95501

Websol: 5.3 in (135 mm) wide Paraweb: made from T-shaped precast concrete panel 34.4 sq ft
Soil Structures International, Ltd. high-tenacity polyester fibers by Imperial (3.2m2) area, 6.3 in (160 mm) thick.
58 Highgate High St. ChemicaL Industries.
London N65HX England

York Method: Galvanized mild steel or stainless steel or Hexagonal: glass fiber reinforced cement; 24 in (61
Transport and Road Research Laboratory glass fiber reinforced plastic or Paraweb or cm) across the flat; 9 in (23 cm) deep.
Crowthorne Terram.
Berkshire, England

Anda Augmented Soils: Fibretain straps (pultruded fiberglass Precast concrete crib units with 12-in (30 cm)-high
Anda Augmented Soils Ltd. reinforced plastic strip, developed by headers 4 ft (1.2 m) apart.
Oaklands House Pilkington Brothers, 1.6, 3.1 or 6.3 in wide,
Solarton Road, Farnborough .08, .1 or .16 in thick (40, 80 or 160 mm
Hants GU14 7QL England wide, 2, 2.5 or 4 mm thick).

Tensar Geogrid System: Non-metallic polymeric grid mat made from Non-metallic polymeric grid mat (wrap around of the
The Tensar Corporation high-density polyethylene or polypropylene. soil reinforcement grid with shotcrete finish, if
1210 Citizens Parkway desired), precast concrete units.
Morrow, GA 30260

Miragrid System: Non-metallic polymeric grid made of Precast concrete units or grid wrap around soil.
Mirafi, Inc. polyester multifilament yarns coated with
PO Box 240967 latex acrylic.
Charlotte, NC 28224

Maccaferri Terramesh System: Continuous sheets of galvanized double- Rock-filled gabion baskets laced to reinforcement.
Maccaferri Gabions, Inc. twisted woven wire mesh with PVC coating.
43A Governor Lane Blvd.
Williamsport, MD 21795

1
Many other facing types, as compared to those listed, are possible with any specific system.
Ref.: FHWA 1990

10-99
10-100
Three types of reinforcement geometry can be considered:

1) Linear unidirectional - strips; steel, plastic, and fabric; rods, and cables.

2) Composite unidirectional - grid strips or bar mats.

3) Planar bidirectional - continuous sheets of geosynthetics, welded wire mesh, and

woven wire mesh.

Stresses are transferred between soil and reinforcement by friction and/or passive
resistance, depending on the reinforcement geometry:

• Friction - Stresses are transferred from soil to reinforcement by shear along

the interface. This is the dominant mechanism with linear and planar
reinforcements (strips, rods, cables, fabrics, geotextile sheets).

• Passive resistance - Stresses are transferred from soil to reinforcement by

bearing between the transverse elements against the soil. This is the
dominant mechanism for reinforcement containing a large number of
transverse elements of composite reinforcements such as bar mats, grids,
and wire mesh.

The performance and durability considerations for the two classes of reinforcement vary
considerably. The distinction made between the characteristics of metallic and
nonmetallic reinforcement are:

• Metallic reinforcements - Consist of mild steel or aluminum.

• Nonmetallic reinforcements - Generally polymeric materials consisting of

polypropylene, polyethylene, or polyester polymers.

There are two classes of extensibility:

• Inextensible - The deformation of the reinforcement at failure is much less

than the deformability of the soil (metallic reinforcements).

• Extensible - The deformation of the reinforcement at failure is comparable

to or even greater than the deformability of the soil.

10-101
10-10.1 Forces

Forces for the analysis are generally the same as for concrete gravity dams and
embankment dams as discussed in Chapter III and Chapter IV of these Guidelines. Only
exceptions, clarifications and additions are discussed below.

MSE resists these forces under conditions somewhere between those of an embankment
dam and those of a concrete gravity dam (ICOLD 1993).

10-10.1.1 Dead Loads

The dead loads considered should include the effective weight of the fill, reinforcement,
and the facing.

10-10.1.2 External Hydrostatic Loads

The hydrodynamic forces on a MSE body subject to overtopping and on the lower portion
of MSE body subject to high tailwater levels can be severe. Every element of these parts
(backfill, facing units, reinforcements, attachments) has to be designed to withstand these
forces. The actions of high hydrodynamic pressures, their rapid changes, the shocks of
the floating bodies, the risk of fines being removed, and the alternate chemical action of
water and oxygen must be evaluated in a realistic manner and adequate measures must be
taken against such actions. Adequate secondary defensive design must be provided.

10-10.1.3 Earth Pressures

Earth pressures exerted on the MSE body by soil backfill should be calculated using the
appropriate coefficient of earth pressure based on the type of reinforcement, the character
of the backfill, along with the construction sequence. Refer to FHWA (1990) for further
discussion.

10-10.1.4 Ice Pressures

If the MSE body is to be used as a spillway and overflow is expected during the winter,
ice formation should be prevented, or the structure should be designed for external ice
load. The potential for ice pressures in the backfill behind facing panels should be
evaluated if the dam is exposed to severe frost.

10-102
10-10.2 Loading Combinations

Loading combinations for the analysis of MSE dams are the same as for embankment
dams discussed in Chapter IV, Section 4-6.6 of these Guidelines.

10-10.3 Analyses

A great deal of progress has been made in developing analytical techniques for MSE in
the past 10 years. However, the analytical techniques are relatively new, and are still
being refined. Details at interfaces, junctions, and boundaries are often the starting point
of failures due to factors such as strain incompatibility or edge effects and should be
carefully evaluated. All appurtenances behind, in front of, under, mounted upon, or
passing through the fill structure such as drainage structures, utilities, or other
appurtenances should be accounted for in the stability design of the MSE structure.

Many MSE systems are patented or proprietary. Some companies provide services
including design preparation of plans and specifications for the structure, supply of the
manufactured wall components, and construction supervision.

The design of MSE systems is discussed in Mitchell and Villet (1987), Koerner (1990)
and FHWA (1990). Design and Construction Guidelines — Reinforced Soil Structures
(FHWA 1990) was developed to assist designers in determining the feasibility of using
MSE systems for walls and embankment slopes, evaluating different alternative
reinforcement systems, and performing preliminary design of simple systems. It provides
a basis for evaluation and preliminary design of new MSE systems that may be proposed
in the future. The design methods provided in the manual are not meant to replace
private and proprietary system-specific design methods, but they should provide a basis
for evaluating such designs. The various systems have different performance histories,
and this sometimes creates difficulty in adequate technical evaluation. Methods for
handling the matter of specifications and for obtaining the most cost competitive and
technologically acceptable system are given in FHWA (1990).

The MSE structure should be designed to using three types of analysis: an analysis at
working stresses, a limit equilibrium analysis, and a deformation (or displacement)
analysis.

The analysis at working stresses is used to: 1) select reinforcement location and check
that stresses in the stabilized soil mass are compatible with the properties of the soil and
reinforcement, 2) evaluate local stability at the level of each reinforcement and predict
progressive failure, and 3) estimate vertical and lateral displacements.

10-103
The limit equilibrium analysis is used to check the overall stability of the structure.
Three types of stability must be considered, external, internal, and combined. The
external stability involves the overall stability of the stabilized soil mass considered as a
whole. The internal stability analysis consists of evaluating potential slip surfaces within
the reinforced soil mass. In some cases, the critical slip surface is partially outside and
partially inside the stabilized soil mass, and a combined external/internal stability
analysis may be required.

The deformation analysis is used to evaluate the performance of the structure with respect
to anticipated displacement. In addition, the influence of variations in the type and
density of reinforcement on the performance of the structure can be evaluated.
Deformation analyses are the most difficult and least certain of the three types of
analysis. In many cases, they are done only approximately or it is simply assumed that
adjusted factors of safety against external or internal stability failure will ensure that
deformations will be within tolerable limits.

10-10.3.1 External Stability

External stability of an MSE structure should be evaluated for sliding resistance,

overturning, bearing capacity, and seismic stability.

10-10.3.1.1 Sliding Resistance

The MSE body is subject to the lateral driving pressures on the upstream face that must
be resisted by the frictional and cohesive resistance along the base or a weak layer near
the base of the MSE body and the passive resisting pressures on the downstream face.
Passive resistance should be neglected due to the potential for erosion and deterioration
of the toe area. The shear strength of any facing system should also be neglected.

Three possible sliding planes must be evaluated: 1) foundation soil, 2) backfill soil, and
3) soil-reinforcing interface (sheet-type reinforcement).

Stability is determined using rotational or wedge analyses, as appropriate, which can be

performed using a slope stability analysis. Compound failures, passing through both the
unreinforced and reinforced zones, must be considered if complex conditions exist. In
the case of a modified structure such as a retaining structure on the crest of a dam or
steepened slopes, the modified loads on the embankment and any potentially higher
reservoir water surface should be used to check overall deep-seated slope stability.

10-104
10-10.3.1.2 Overturning

Due to the flexibility of MSE structures, it is unlikely that a block overturning failure
could occur. Nonetheless, an adequate factor of safety against this classical failure mode
will limit excessive outward tilting and distortion of a suitably designed structure.

10-10.3.1.3 Bearing Capacity

An MSE body, especially one with near-vertical slopes, and any surcharge loads
supported by it should be checked for bearing capacity failure of the foundation soil. The
maximum vertical earth pressure and the effective vertical stress exerted by the MSE
body on the foundation soil are calculated by traditional geotechnical methods. Due to
the flexibility of the reinforced fill mass, the safety factor with regard to bearing capacity
failure can be less than for more rigid structures.

10-10.3.1.4 Seismic Stability

Relatively large earthquake shaking could result in significant permanent lateral and
vertical deformations. Various methods of analyses are available for evaluating the
seismic stability of an MSE body — pseudostatic, simplified, and deformation. Refer to
Chapter IV, Section 4-7 of these Guidelines, FHWA (1990), and Richardson (1978) for
further discussion.

10-10.3.2 Internal Stability

Internal failure of an MSE structure can occur in two different ways:

• Tension - The tensile forces (and, in the case of rigid reinforcements, the shear
forces) in the reinforcements become so large that the reinforcements elongate
excessively or break, leading to large movements and possible collapse of the
structure.

• Pullout - The tensile forces in the reinforcements become larger than the pullout
resistance. This, in turn, increases the shear stresses in the surrounding soil,
leading to large movements and possible collapse of the structure.

The most critical slip surface in an MSE wall is assumed to coincide with the maximum
tensile forces line. The maximum tensile forces line has been assumed to be
approximately bilinear in the case of inextensible reinforcements, approximately linear in

10-105
the case of extensible reinforcements, and passes through the toe of the wall in both
cases.

The most critical slip surface may cross one or more layers of the reinforcements. A
slope stability analysis method that takes into account the effect of reinforcements
intercepted by the slip surface must be used in these cases.

In general, any shape of slip surface can be considered: plane, circle, multilinear. MSE
slope stability analysis methods have been published and computer programs are
available that allow for limit equilibrium analysis of reinforced soil masses.

Finite element methods (FEMs) have been used to study the performance of geotextile-
reinforced embankments in both analysis and design situations (Andrawes 1982, Rowe
1984). Although these sophisticated computer-based methods might not be routinely
used for noncritical situations, they do give great insight into the behavior of the system.
Finite element studies have been found to provide more realistic results than limit
equilibrium analysis, which have been found to overpredict stresses, loads and
movements (Jarrett 1988).

10-10.3.2.1 Tension Failure

The reinforcement is subjected to tensile forces and extension through the shear-transfer
that develops between it and the fill. The location of the maximum tensile forces line is
influenced by the extensibility of the reinforcement as well as the overall stiffness of the
facing. With inextensible reinforcements the maximum tensile forces line can be
modeled by a bilinear failure surface that is vertical in the upper part of the structure.
The state of stress is assumed to be at rest at the top and decreases to the active state in
the lower part of the structure. With extensible reinforcements the maximum tensile
forces line coincides with the Coulomb or Rankine active failure plane, and the stresses in
the fill correspond to the active earth pressure condition. The location of the maximum
tensile forces line may also vary due to external factors such as the shape of the structure
and surcharge conditions.

The maximum tensile forces line separates two areas in the MSE body: an active area in
which the shear stress exerted by the ground on the reinforcement is directed toward the
outside of the wall and a resistant area in which this stress is directed towards the inside.
For each reinforcement layer the "adherence length" is defined as the reinforcement
length located within the resistant area.

The lateral earth pressure to be resisted by the reinforcements should be calculated using
the appropriate coefficient of earth pressure based on the type of reinforcement used

10-106
times the vertical soil stress at each reinforcement layer. The vertical soil stress shall be
calculated using the Meyerhof method. The seepage force should also be taken into
account. The maximum tensile force must be checked to see if it is less than the
allowable reinforcement strength (not including the extra thickness provided for
corrosion).

At the connection of the reinforcements with the facing, check that tensile force is not
greater than the allowable tensile strength of the connection. The connection strength
will depend on the structural characteristics of the facing system used.

10-10.3.2.2 Pullout

The design of the soil reinforcement system requires an evaluation of the long-term
pullout performance with respect to three basic criteria:

• Pullout capacity, i.e., the pullout resistance of each reinforcement should be

adequate to resist the design working tensile force in the reinforcement with a
specified factor of safety.

• Allowable displacement, i.e., the relative soil to reinforcement displacement

required to mobilize the design tensile force should be smaller than the allowable
displacement.

• Long-term displacement, i.e., the pullout load should be smaller than the critical
creep load.

The pullout resistance of the reinforcement is mobilized through interface friction and/or
passive soil resistance. The load transfer mechanisms mobilized by a specific
reinforcement depends primarily upon its structural geometry. The soil-to-reinforcement
relative movement required to mobilize the design tensile force depends mainly upon the
load transfer mechanism, the extensibility of the reinforcement material and the soil type.
The long-term pullout performance is predominantly controlled by the creep
characteristics of the soil and the reinforcement material.

The pullout resistance of the reinforcement is defined by the ultimate tensile load
required to generate outward sliding of the reinforcement through the reinforced soil
mass. Several approaches and design equations have been developed and are currently
used to estimate the pullout resistance by considering frictional resistance, passive
resistance, or a combination of both. The design equations use different interaction
parameters, and it is, therefore, difficult to compare the pullout performance of different
reinforcements for a specific application. Refer to FHWA (1990) for further guidance.

10-107
10-108
10-10.3.2.3 Seismic Loading

The seismic event will lead to dynamic incremental loads in the strips. The distribution
of the corresponding incremental tensile forces is different from that occurring in the
static case. However, as long as the seismic intensity is not very high this will have very
little influence on the position of the resultant line of maximum tensile forces. Under
strong ground movements the line will tend to move away from the wall facing.

High MSE fills may not behave in a ductile manner near the base of the fill, because the
reinforcements in this location may fail in tension, rather than by sliding inside the
backfill.

The internal stability of the MSE body under earthquake loading conditions is analyzed
according to procedures presented by Richardson (1978). These procedures determine
the dynamic force magnitude in each reinforcing strip through computation of the
amplitude and of the distribution of dynamic earth pressures acting on the walls.
Sufficient reinforcement must be provided so that combined static and dynamic earth
pressures do not cause tensile or soil-strip frictional failure of the reinforcement.

Design strip forces are obtained by combining the dynamic and static forces. If changes
in reinforcing strip density or wall geometry are necessary to obtain a desired factor of
safety, the design procedure must be repeated since this will change the stiffness of the
wall. The magnitude and distribution of the dynamic forces and peak dynamic strain are
functions of the wall stiffness.

10-10.3.2.4 Facing Strength

A comparison must be made between the horizontal loads that the facing will have to bear
and the allowable loads for each type of facing (according to their nature, thickness, and
number of fixing points).

10-10.3.3 Back-to-Back Wall Design

The back-to-back design must be considered in the case of a double-faced wall used to
raise dam crests, which is actually two separate walls with parallel facings. This situation
can lead to a modified value of backfill thrust which influences the external stability
calculations.

For the first case, the overall base width is large enough so that each wall behaves and
can be designed independently. In particular, there is no overlapping of the
reinforcements. The active thrust may be mobilized without any inclination on the

10-109
horizontal or the active thrust may be reduced, depending on the distance between the
two walls.

For the second case, there is an overlapping of the reinforcements, so that the two walls
interact. Consequently, the two walls are designed independently with the same internal
local stability procedure, but assuming no active thrust from the backfill.

10-10.3.4 Deformation

Conventional settlement analysis for shallow foundations should be carried out to ensure
that immediate, consolidation, and secondary settlement of the structure are less than the
performance requirements of the project. An estimation of the total and differential
settlements due to the foundation ground, MSE, facing, and any other contiguous
structures must be made. This information is necessary in order to make sure that the
expected settlements are compatible with the proper behavior of the reinforcement
connections, deformation capacities of the facing, and the behavior of integral or adjacent
structures.

There is no standard method to evaluate the overall lateral displacement of MSE.

Loading of the MSE section and associated lateral deformation will primarily occur
during construction with the exception of post construction surcharge loads. Post-
construction movement could also occur due to settlement of the structure.

The major factors influencing lateral displacements during construction include

compaction intensity, reinforcement to soil stiffness ratio (i.e., the area of reinforcement
and deformability as compared to the modulus and area of the reinforced soil section),
reinforcement length, slack in reinforcement-to-facing connections, and deformability of
the facing system.

The total lateral displacement of simple structures on firm foundations that is anticipated
during construction can be estimated from Figure 32 in FHWA (1990), based on the
length of reinforcement (L) to height of the wall (H) ratio and the extensibility of the
reinforcement. This figure was empirically developed using data from actual structures
and computer simulation models. It provides a first order lateral deformation estimate
that could be used to establish appropriate face batter and to evaluate anticipated
horizontal alignments. It should be noted that as L/H decreases, the lateral deformation
increases. This is important when determining the suitability of the final reinforcement
length. For example, going from a length of 0.7H to 0.5H could essentially double the
lateral deformation anticipated during construction.

10-110
For critical structures requiring precise tolerances, the lateral displacement of the wall has
to be calculated more accurately. A finite element method of calculation is recommended
for this analysis.

10-10.3.5 Seepage

A seepage analysis should be performed to determine the amount of seepage and the
magnitude of the seepage forces. The simplified line-of-creep method, flownet analysis
and numerical methods, may be used to perform the seepage analysis. The potential for
the reinforcement creating a seepage path through the dam should be evaluated. Refer to
Chapter III, Section 3-2.4.4 and Chapter IV, Section 4-5 for further discussion.

10-10.3.6 Computer Programs

The ideal method for MSE design is to use a conventional slope stability computer
program that has been modified to account for the stabilizing effect of reinforcement.
Such programs should account for reinforcement strength and pullout capacity, compute
reinforced and unreinforced safety factors automatically, and have some searching routine
to help locate critical surfaces. The ideal method would also include the confinement
effects of the reinforcement on the strength of the soil in the vicinity of the reinforcement.
Very few of these programs are publicly available, and those are usually limited to
specific soil and reinforcement conditions.

Refer to Chapter III, Section 3-4.7 for further guidance on review of computer programs.

10-10.4 Acceptance Criteria

Adequate safety concerning overall stability, reinforcement tensile resistance, pullout,

facing stability, seepage, deformation of all parts of the structure, and the durability of all
materials must be demonstrated.

10-10.4.1 External/Internal Stability

Recommended minimum factors of safety are given in Table 10-10.3.

10-10.4.2 Steel Reinforcement

The allowable tensile stress in steel reinforcements and connections, at the end of service
life shall conform to the following:

10-111
 Ft = 0.55Fy at reduced gross section1
Ft = 0.50Fy at the net section at bolt hole
(applicable to bolted connections only)

Where Ft allowable tensile stress

Fy yield stress

An allowance must be made in the metal cross section area to account for the estimated
corrosion loss.

10-10.4.3 Geosynthetic Reinforcement

The tensile properties of geosynthetics are affected by creep, construction damage, aging,
temperature, and confining stress. Furthermore, characteristics of geosynthetic products
manufactured with the same base polymer vary widely.

Polymeric reinforcement, although not susceptible to corrosion, may degrade due to

physicochemical activity in the soil such as hydrolysis, oxidation, and environmental
stress cracking. In addition, it is susceptible to construction damage. Degradation most
commonly occurs from mechanical damage, loss of strength due to creep, and
deterioration from exposure to ultraviolet light.

10-112
 TABLE 10-10.3
RECOMMENDED MINIMUM FACTORS OF SAFETY

Loading Combination
Parameter Usual Seismic
External Stability Recommended minimum factors of safety should be in accordance
with Chapter III and Chapter IV of these Guidelines.

Deformation Maximum allowable total and differential, based on performance

requirements of the project.
Internal Stability

Pullout Resistance1

Granular soils 1.5 >1.1

Cohesive soils 2.0 >1.1
Breakage Strength Determine allowable tension in reinforcement.
(Do not forget the reinforcement/facing connection tensile strength.)
Durability Take into account the design life in the determination of the allowable
tension

1
minimum embedment length 3 ft.

Ideally the allowable tension should be determined by thorough consideration of

allowable elongation, creep potential and possible strength degradation using a complex
method that requires extensive long-term strength testing of the geosynthetic product. In
the absence of sufficient test data a simplified expression may be used. The expression
takes into account yield tensile strength, a creep reduction factor, a durability factor of
safety, a construction damage factor of safety, and an overall factor of safety. The
expression must be equal to or less than the long-term tension capacity of the
geosynthetic at a selected design strain (usually 5% or less) (FHWA 1990).

The yield tensile strength is obtained from wide strip tensile strength tests (ASTM
D4595). The creep reduction factor (CRF) is the creep limit strength obtained from the
creep test results divided by the yield tensile strength. If the CRF value of the specific
reinforcement is not available, the following recommendations are provided in FHWA
(1990).

10-113
 Polymer Type Creep Reduction Factors1
Polyester 0.4
Polypropylene 0.2
Polyamide 0.35
Polyethylene 0.2
1
Additional reduction should be made for applications in high-temperature environments (temperatures greater
than 90NF in the region of the reinforcement, e.g. at facing connection, in hot climates).

The durability factor of safety is dependent on the susceptibility of the geosynthetic to

attack by microorganisms and chemicals, thermal oxidation, and environmental stress
cracking and can range from 1.1 to 2.0. In the absence of product specific durability
information use 2.0.

The construction damage factor of safety can range from 1.1 to 3.0. In the absence of
product specific construction damage tests use 3.0.

The overall factor of safety accounts for uncertainties in the geometry of the structure, fill
properties, reinforcement properties, and externally applied loads. For permanent,
vertically faced structures use a minimum of 1.5.

10-10.4.4 Backfill

Backfill properties should be in accordance with the requirements of the MSE system
being used. The design and construction of the MSE system depends on the specific
properties of the soil used as backfill, and various types of soil (sand, silt, clay, gravel)
can be used. However, most MSE systems specify high-quality backfill in terms of
durability, drainage, and friction consisting of well graded, granular materials. Typical
ranges of soil types for MSE are summarized in Table 10-10.2. Many of the MSE
systems depend on friction between the reinforcing elements and the soil. In such cases,
generally a material with high friction characteristics is specified and required. Some
MSE systems rely on passive pressure on reinforcing elements and in those cases, the
quality of backfill is still critical. These requirements generally eliminate soils with high
clay contents. Corrosive conditions may be present due to acid or marine environments,
aggressive soils, presence of sulfates, etc. and require special measures.

10-10.4.5 Seepage Control

As a general rule, MSE is not to be used for dam imperviousness. If possible, the
waterproofing system should be located upstream of the MSE and drainage, including a
filter, should be provided between the impervious layer and the reinforced fill body.

10-114
When MSE materials are sufficiently permeable and self-filtering, they can serve as
drains. When the backfill is not highly permeable, a drainage layer should be included
behind the facing panels.

For particular applications such as a dam raising, the reinforced fill system might serve as
a watertight barrier for the design flood. Although water will be against the wall only
during remote events, the structure must be capable of providing the necessary protection
against seepage that could affect the integrity of the structure or embankment. A viable
means of providing a water barrier is to seal all joints on the upstream reinforced fill
facing. This could be accomplished by applying a waterproof membrane over the joints
on the inside face of the wall. Although the membrane should provide a positive water
barrier, deformations of the facing during normal operations or unnoticed damage to the
membrane during construction could disrupt its integrity. Assurance of a continuous
membrane-to-facing seal would be difficult to control during construction. Therefore, as
a defensive design measure, it is necessary to provide zoning of materials in the retaining
wall system to further control any seepage not stopped by the membrane. The zoning of
materials should provide:

• Sufficiently low permeability to restrict seepage flow to manageable

quantities should disruption of the membrane occur.

• Sufficiently high free-draining capability to prevent the retention of

moisture around reinforcing strips, and thereby, minimize the potential for
corrosion of the strips during normal operations.

• Filter protection between adjacent zones to prevent the loss of material

should seepage occur.

10-10.4.6 Corrosion/Deterioration

The service life and safety of MSE structures centers around embedding reinforcement in
fill materials and protecting that reinforcement from corrosion or loss of material that
could eventually compromise stability. The MSE should be designed with a service life
comparable with that of the rest of the dam.

Practically all the metallic reinforcements used in construction of embankments and

walls, whether they are strips, bar mats, or wire mesh, are made of galvanized mild steel.
Woven meshes with PVC coatings also provide corrosion protection, provided the coating
is not significantly damaged during construction. Epoxy coating can be used for
corrosion protection, but it is also susceptible to construction damage, which can
significantly reduce its effectiveness.

10-115
10-10.5 Material Properties

The material strength assumptions made during the design process must be verified by the
site investigations and attained in the field during the construction process. Specific
details concerning geologic investigations are contained in Chapter V of these Guidelines.
The following general guidance shall apply.

10-10.5.1 Site Investigation

If the MSE is to be used in conjunction with an existing structure, the stability and
condition of that structure must be investigated. If the MSE is to be a new structure, the
investigation should be sufficient to define the major geologic and hydrologic conditions
with emphasis on those that will affect design. In all cases a special emphasis must be
placed on the site factors that could aggravate the corrosion or the deterioration of any
type reinforcement. Important factors to be considered are the reservoir water quality,
underground water quality, and fill quality.

10-10.5.2 Reinforcement

The following information on the reinforcement materials is needed for the design:
geometric characteristics, strength and stiffness properties, durability, and soil
reinforcement interaction properties. The two most commonly used reinforcement
materials are steel and geosynthetics.

Two types of geometric characteristics can be considered:

• Strips, Bars, and Steel Grids: A layer of steel strips, bars or grids is
characterized by the cross-sectional area, the thickness and perimeter of the
reinforcement element and the center-to-center horizontal distance between
elements. A layer of geosynthetic strips is characterized by the width of the
strips and the center-to-center horizontal distance between them.

• Sheets and Geosynthetic Grids: A layer of sheet of geosynthetic grid is

characterized by the width of the sheet or grid component and the center-to-
center horizontal distance between the sheets or grid components. The
cross-sectional area is not needed, since the strength of a sheet and a grid
are expressed by a tensile force per unit width, rather than by a stress.

A variety of geosynthetics may be used as multilayer reinforcement to construct slopes

that are steep or even vertical. A high strength/high-modulus geosynthetic is usually
required (typically geogrids or multifilament woven geotextiles) for dam support. Three

10-116
basic properties that need to be established for the use of a geosynthetic are the tensile
resistance, elongation to rupture, and the tear resistance. The long-term durability of the
geosynthetic must also be established. The geosynthetics and other new reinforcement
materials have limited performance history compared to traditional dam construction
materials. While they offer the potential for economic design, the need for safety and
longevity in dam design is paramount.

10-10.5.3 Corrosion Protection

For metallic reinforcements the site factors that could aggravate their corrosion or
deterioration are mainly a marine environment, acid water, soil with high aggressive salts
contents, and pure water. When exposed to corrosive conditions galvanized steel
reinforcements are normally used. Means to control the effect of subsequent corrosion
include increasing the reinforcement thickness, finding another fill material, or shielding
the reinforced fill from the aggressive seepage with a watertight membrane. Shielding
has included:

• Providing cathodic protection so that the rate of metal loss for the
embedded metalwork could be effectively controlled and monitored.

• Insulating the connection between the metal embedded in soil and the metal
embedded in concrete to reduce the potential for corrosion.

• Using an epoxy coating on all reinforcing components embedded in soil or

concrete to reduce the potential for corrosion.

The corrosivity of the backfill material must be evaluated because of the corrosion of
buried metals. The corrosion of buried metals depends on the presence of dissolved salts
in the soil, pH, and degree of saturation. Highest corrosion rates are produced by a high
content of dissolved salts, a high chloride concentration, a high sulfate content, and acidic
or alkaline pH conditions in the soil.

The polymeric formulation and resin additive package of the geosynthetic must be
compatible with the chemistry of the backfill and the potential for the environment within
the backfill to change with time. The backfill should be checked for such items as high
and low pH, chlorides, organics, and oxidation agents such as soils that contain Fe2SO4,
calcareous soils, and acid sulfate soils that may result in deterioration of the geosynthetic
with time. Other possible detrimental environmental factors include chemical solvents,
diesel, and other fuels, active slag fills, and industrial wastes. FHWA (1990) provides
additional data on corrosion rates. Generally, site specific corrosion studies should be
performed to determine the appropriate metal loss rates.

10-117
10-10.5.4 Facing

The facing is not subjected to much stress and consists of relatively light units, except if
seepage forces may be present. The facing is connected to the reinforcement so that it
becomes part of the structure. It must be flexible enough to adapt to the settlements and
deformations of the structure and its foundation. The edges of the facing panels must be
shaped so that they are properly secured to one another.

Large deformations may lead to opening of the contacts between facing elements
resulting in a loss of backfill. When such deformations are expected, the design must
provide for them without loss of fill. Natural materials or filter cloth (geotextiles) can be
used to prevent the loss of material through the contacts between the facing elements.

No facing component is excluded from consideration but it is essential that the assembly
be rapid so that fill placement occurs smoothly. The major facing types are: segmental
precast concrete panels; case-in-place concrete, shotcrete or full height precast panels;
semi-cylindrical metallic facings; welded wire grids; gabion facings; fabric facing; plastic
grids; and postconstruction facings. The use of separate panels provide the flexibility to
absorb differential movements, both vertically and horizontally, without undesirable
cracking that could occur in a rigid structure. Metal facings have the disadvantage of
shorter life because of corrosion unless provision is made to compensate for it. Facings
using welded wire mesh or gabions have the disadvantages of an uneven surface, exposed
backfill materials, more tendency for erosion of the retained fill, possible shorter life from
corrosion of the wires, and more susceptibility to vandalism. The greatest advantages of
such facings are low cost, ease of installation, design flexibility, and good drainage
(depending on the type of backfill) that provides increased stability. The surface
deterioration of fabric facing and plastic grids must be evaluated.

10-10.6 Construction Considerations

Careful inspection during construction is important. Special attention must be paid to the
of joints, seams, anchorages, penetrations, and similar interface conditions as well as to
the reinforcement.

10-10.7 References

ASTM (American Society of Testing Materials), " Wide Strip Tensile Strength Test,"
D4595, Philadelphia, PA.

10-118
Andrawes, K.Z., McGown, A., Wilson-Fahmy, R.F., and Mashhour, M.M., "The Finite
Element Method of Analysis Applied to Soil-Geotextile Systems," Proc. 2nd Int. Conf.
Geotextiles, Las Vegas, NV, Vol. 2, pp. 690-700, 1982.

Cassard, A., Kern, F., and Mathieu, G., "Use of Reinforcement Techniques in Earth
Dams," International Conference on Soil Reinforcement, Vol 1, Paris, March 1979.

Christopher, B.R. and Holtz, R.D., Geotextile Engineering Manual, STS Consultants,
Ltd., Report No. FHWA-TS-86/203, pp. 1044, Washington, DC, 1984.

Darbin, M., Jailloux, Jean-Marc, and Montuelle, J., "Performance and Research on the
Durability of Reinforced Earth Reinforcing Strips," Proceedings of Symposium on Earth
Reinforcement, ASCE Annual Convention, Pittsburgh, Pennsylvania, April 27, 1978.

Duster, C.O., "Dam Crest Raising with Reinforced Earth Retaining Walls," Dam Safety
and Rehabilitation, 4th Annual USCOLD Lecture, Phoenix, Arizona, January 1984.

FHWA (Federal Highway Administration) U.S. Department of Transportation,

"Reinforced Soil Structures Volume I. Design and Construction Guidelines," FHWA-
RD-89-043, November 1990.

French Committee on Large Dams, "Use of Geotextiles in Earth Dams," October 1982.

Giroud, J.P., "Functions and Applications of Geosynthetics in Dams," Water Power &
Dam Construction, June 1990.

ICOLD (International Congress on Large Dams), "Reinforced Rockfill and Reinforced

Fill for Dams," Bulletin 89, 1993.

Jarrett, P.M. and McGown, A., eds., Proc. NATO Workshop on Application of Polymeric
Reinforcement in Soil Retaining Structures, June 1987, Royal Military College of
Canada, Kingston, Ontario, Canada, Kluwer Acad. Publ., 1988.

King, R.A., and H. Nabizadeh, "Corrosion in Reinforced Earth Structures," Proceedings

of Symposium on Earth Reinforcement, ASCE Annual Convention, Pittsburgh,
Pennsylvania, April 27, 1978.

Koerner, R.M., "Designing with Geosynthetics," Prentice-Hall, Englewood Cliffs, N.J.,

1990.

10-119
Mitchell, J.K. and Villet, W.C.B., "Reinforcement of Earth Slopes and Embankments,"
NGHRP Report No. 290, Transportation Research Board, Washington,D.C., 1987.

Richardson, Gregory N., "Earthquake Resistant Reinforced Earth Walls," Proceedings of

Symposium on Earth Reinforcement, ASCE Annual Convention, Pittsburgh,
Pennsylvania, April 27, 1978.

Rowe, R.K., and Soderman, K.L., "Comparison of Predicted and Observed Behavior of
Two Test Embankments," Int. J. Geotextiles Geomembranes, Vol. 1, pp. 143-160, 1984.

Schmertmann, G.R., Chouery-Curtis, V.E., Johnson, R.D., and Bonaparte, R., "Design
Charts for Geogrid - Reinforced Soil Slopes," Proc. Geosynthetics, Vol. 1, pp. 108-120,
New Orleans, 1987.

10-120
ENGINEERING GUIDELINES FOR THE EVALUATION OF HYDROPOWER
PROJECTS

CHAPTER 11 - ARCH DAMS

Federal Energy Regulatory Commission

Division of Dam Safety and Inspections
Washington, DC 20426

October, 1999
 TABLE OF CONTENTS

11-1 INTRODUCTION................................................................................................. 11-1

11-1.1 Purpose..................................................................................................... 11-1
11-1.2 Applicability............................................................................................. 11-1
11-1.3 Definition of Safety.................................................................................. 11-2
11-1.4 Evaluation Criteria ................................................................................... 11-3
11-1.4.1 Review of Existing Data and Site Inspection............................ 11-3
11-1.4.2 Method of Analysis................................................................... 11-4
11-1.4.3 Evaluation for Static Loading ................................................... 11-4
11-1.4.4 Evaluation for Seismic Loading................................................ 11-6
11-1.4.5 Sliding Stability ........................................................................ 11-7
11-2 FOUNDATION CONSIDERATIONS................................................................ 11-9
11-2.1 General ..................................................................................................... 11-9
11-2.2 Field Investigations .................................................................................. 11-9
11-2.2.1 Foundation Features that Create Stability Concerns and
Warning Signs ......................................................................................... 11-9
11-2.3 Material Parameter Selection ................................................................... 11-12
11-2.3.1 Shear Strength of Foundation Interface .................................... 11-12
11-2.3.2 Shear Strength of Potential Foundation Failure Planes &
Wedges .................................................................................................... 11-16
11-2.3.3 Foundation Modulus of Deformation ...................................... 11-16
11-2.4 Foundation Rock Erodibility.................................................................... 11-19
11-2.4.1 Field Investigations ................................................................... 11-19
11-2.4.2 Assessing the Erodibility of Rock ............................................ 11-20
11-2.4.3 Errosion Downstream from the Dam ....................................... 11-32
11-3 CONCRETE MATERIAL PARAMETERS...................................................... 11-39
11-3.1 Visual Inspection of the Concrete............................................................ 11-39
11-3.2 Ultrasonic Pulse Velocity Tests ............................................................... 11-39
11-3.3 Concrete Coring and Specimen Parameters............................................. 11-41
11-3.4 Petrographic Examination of Concrete .................................................... 11-41
11-3.5 Elastic Properties ..................................................................................... 11-42
11-3.6 Thermal Properties................................................................................... 11-42
11-3.7 Strengths of Concrete............................................................................... 11-43
11-3.7.1 Compressive Strength ............................................................... 11-43
11-3.7.2 Tensile Strength ....................................................................... 11-43
11-3.7.3 Shear Strength .......................................................................... 11-45
11-3.8 Dynamic Material Properties .................................................................. 11-46

i
11-4 LOADING............................................................................................................. 11-47
11-4.1 Dead Load ................................................................................................ 11-47
11-4.2 Hydraulic Loading ................................................................................... 11-47
11-4.2.1 Normal Water Loads................................................................. 11-47
11-4.2.2 Flood Loads............................................................................... 11-48
11-4.2.3 Uplift ......................................................................................... 11-48
11-4.2.4 Silt Load.................................................................................... 11-49
11-4.2.5 Ice Load..................................................................................... 11-50
11-4.2.6 Hydraulic Loading of Spillways ............................................... 11-51
11-4.3 Thermal Loading...................................................................................... 11-52
11-4.3.1 Temperature Distribution.......................................................... 11-52
11-4.3.2 Air Temperature........................................................................ 11-53
11-4.3.3 Reservoir Water Temperature................................................... 11-54
11-4.3.4 Solar Radiation.......................................................................... 11-55
11-4.3.5 Concrete Temperatures ............................................................. 11-55
11-4.4 Earthquake Loading ................................................................................. 11-56
11-4.4.1 Safety Evaluation Earthquakes and Associated Ground
Motions ................................................................................................... 11-56
11-4.4.2 Response Spectrum Earthquake Input ...................................... 11-56
11-4.4.3 Acceleration Time History Earthquake Input ........................... 11-56
11-4.4.4 Spatial Variation of Ground Motion ......................................... 11-58
11-4.5 Loading Combinations............................................................................. 11-60
11-4.5.1 Usual Loading Combinations.................................................... 11-60
11-4.5.2 Unusual Loading Combinations................................................ 11-61
11-4.5.3 Extreme Loading Combinations ............................................... 11-62
11-5 STATIC ANALYSIS ............................................................................................ 11-63
11-5.1 Overview.................................................................................................. 11-63
11-5.2 Finite Element Analysis ........................................................................... 11-63
11-5.2.1 Structural Modeling and Assumptions..................................... 11-63
11-5.2.2 Application of Loads................................................................ 11-67
11-5.2.3 Presentation and Interpretation of Results ............................... 11-70
11-5.2.4 Evaluation of Stress Results..................................................... 11-73
11-5.3 Alternative Continuum Models................................................................ 11-74
11-5.3.1 Trial Load Method ................................................................... 11-74
11-5.3.2 Other Methods ......................................................................... 11-76
11-5.4 Rock Wedge Stability .............................................................................. 11-76

ii
 11-5.4.1 Identification of Kinematically Capabale Potential Failure
Planes and Wedges................................................................... 11-77
11-5.4.2 Analysis by Stereographic Projection Procedures ................... 11-78
11-5.4.3 Vectorial Analysis ................................................................... 11-78
11-5.4.4 Loads to be Considered............................................................ 11-79
11-5.4.5 Appropriate Factors of Safety .................................................. 11-81
11-5.5 Parameter Sensitivity ............................................................................... 11-82
11-5.5.1 Effects of Foundation Modulus on dam Stresses.................... 11-82
11-5.6 Limit State Analysis................................................................................. 11-90
11-5.6.1 Sliding on the Abutment Contact.............................................. 11-90
11-5.6.2 Buckling Failure Modes............................................................ 11-92
11-6 DYNAMIC ANALYSIS........................................................................................ 11-94
11-6.1 Overview.................................................................................................. 11-94
11-6.2 Finite-Element Response Spectrum Analysis .......................................... 11-96
11-6.2.1 Structural Models...................................................................... 11-96
11-6.2.2 General Principles ..................................................................... 11-97
11-6.2.3 Presentation and Interpretation of Results ................................ 11-100
11-6.3 Finite-Element Time-History Analysis .................................................... 11-104
11-6.3.1 Structural Models...................................................................... 11-104
11-6.3.2 General Principles ..................................................................... 11-105
11-6.3.3 Presentation and Interpretation of results.................................. 11-107
11-6.3.4 Time-History Stability Analysis................................................ 11-118
11-6.4 Alternative Analysis Techniques ............................................................. 11-119
11-6.5 Reservoir and Foundation Effects............................................................ 11-120
11-6.5.1 Dam-Water Interaction ............................................................. 11-120
11-6.5.2 Dam-Foundation Interaction ..................................................... 11-126
11-6.5.3 Direction of Ground Motions.................................................... 11-129
11-6.6 Post-Earthquake Safety Evaluation.......................................................... 11-130
11-6.6.1 Evaluation for Static Loads....................................................... 11-131
11-6.6.2 Evaluation for Aftershock Events ............................................. 11-132
11-7 INSTRUMENTATION ........................................................................................ 11-134
11-7.1 Purpose and Need for Instrumentation..................................................... 11-134
11-7.2 Special Instrumentation Considerations for Arch Dams.......................... 11-134
11-7.3 Frequency of Measurements .................................................................... 11-135
11-7.4 Presentation of Data and Interpretation of Readings ............................... 11-136
11-7.5 Comparison of Predicted and Measured Deflections............................... 11-136

iii
 11-7.6 Long Term Instrumentation Performance ................................................ 11-137
11-7.7 Interpretation of Data ............................................................................... 11-137
11-8 HISTORIC FAILURES - PROBLEMS.................................................. 11-140
11-8.1 Overview.................................................................................................. 11-140
11-8.2 Landslide Case ......................................................................................... 11-143
11-8.2.1 Vajont Dam............................................................................... 11-143
11-8.3 Abutment Failure Cases ........................................................................... 11-146
11-8.3.1 Malpasset Dam.......................................................................... 11-146
11-8.3.2 Experimental Plum Dam........................................................... 11-149
11-8.4 High Discharge Induced Failures............................................................. 11-151
11-8.4.1 Failures of Arch Dams .............................................................. 11-151
11-8.4.2 Damage to Stilling Basins and Plunge Pools ............................ 11-152
11-8.5 Earthquake Induced Damage ................................................................... 11-153
11-8.5.1 Pacoima Dam ............................................................................ 11-153
11-8.5.2 Other Significant Cases............................................................. 11-158
11-8.6 Detrimental Chemical Reactions ............................................................. 11-159
11-8.6.1 Kouga Dam, South Africa......................................................... 11-162
11-8.6.2 Santa Luzia Dam, Portugal ....................................................... 11-162
11-8.6.3 Alto-Ceiro Dam, Portugal ......................................................... 11-162
11-8.6.4 Cahora-Basa Dam, Mozambique .............................................. 11-163
11-8.6.5 Gene Wash and Copper Basin Dmas, California...................... 11-163
11-8.6.6 Horse Mesa Dam, Arizona........................................................ 11-164
11-8.6.7 Owyhee Dam, Oregon............................................................... 11-164
11-8.6.8 N'Zilo Dam, Zaire ..................................................................... 11-165
REFRENCES .................................................................................................................. 11-166

iv
 ENGINEERING GUIDELINES FOR THE EVALUATION OF HYDROPOWER
PROJECTS

CHAPTER 11 - ARCH DAM

11-1 INTRODUCTION

11-1.1 Purpose
This chapter of the Guidelines provides guidance on the criteria and procedures used by
the FERC to evaluate the safety and structural integrity of existing arch dams under its
jurisdiction. The intent of this guidance is to outline criteria and evaluation procedures
including foundation considerations, material properties and testing, loading, methods of
analyses, and predicted and observed performance that provide the basis for review and
approval of the analysis and inspection studies submitted to the FERC.

The material presented in this chapter assumes that the reader has a general knowledge
and understanding of the basic principles of arch dams, i.e., how they are designed, con-
structed, operated, and maintained. For detailed discussions on design and a better under-
standing of the arch dam behavior, consult the US Bureau of Reclamation "Design of
Arch Dams," (USBR 1977) and the US Army Corps of Engineers EM 1110-2-2201 "Arch
Dam Design," (COE 1994) ," (USBR 1977). This chapter presents much information. The
intent of this chapter not to mandate new analyses and investigations regardless of
whether or not they are needed. Rather, the variety of issues addressed and computation
methods put forward are an attempt to anticipate the variety of problems that could be
encountered. This chapter should not be interpreted as requiring every test, analysis, and
investigation that it describes at every dam. It may well be that for a given dam, specific
failure mechanisms suggested in this chapter are not pertinent. Analysis of arch dam
safety should always start with simple analysis techniques and conservative assumptions.
If simple analyses indicate problems, more complicated and rigorous analyses may be in
order.

11-1.2 Applicability

This guidance is applicable to FERC engineers and licensees engaged in the safety evalu-
ation of existing arch dams. The design of new arch dams should follow the guidance and
criteria of the references (USBR, 1977 and COE, 1994) , but could also benefit from the
evaluationphilosophy presented in this chapter.

11-1
11-1.3 Definition of Safety

Safety is defined as their adequacy against an uncontrolled release of reservoir water.

The structural integrity is maintained and the dam is considered safe if overstressing,
sliding, and other possible modes of failure will not occur. A safety evaluation, therefore,
should identify all significant failure modes and conduct appropriate analyses to assure
that the structural stability of the dam is maintained.

Overstressing of concrete arch dams may exhibit a tendency toward developing a partial
failure, if large tensile stresses from the linear-elastic analysis indicate extensive joint
opening and cracking . Considering that the ultimate load-resisting capacity of an arch
dam is limited by the compressive strength of the concrete (unless foundation or other
mode of failures occur first), severe and widespread joint opening and cracking might
eventually exhaust the capacity of the concrete to carry compression due to subsequent
load redistributions, or might form surfaces along which partial sliding could occur.
Whether such partial failures could actually occur is unknown, because they have not
been observed previously, and also because of the inherent redundancy in arch dams and
the fact that arch action might restrain movements of the portions separated by joint
opening and cracking.

With respect to sliding failures, two types of potential foundation sliding instability cases
should be considered. The first type is potential sliding of rock wedges within the foun-
dation and in contact with the dam, and the second is potential sliding along the contact
between the dam and foundation rock. The sliding of rock wedges typically occurs along
one potential failure plane (plane sliding) or along the line of intersection of two of these
planes (wedge sliding). For a rock wedge to be kinematically capable of failure, the di-
rection of sliding must intersect or "daylight" a free surface downstream from the dam.
While an arch dam might be capable of bridging a small unstable foundation block at the
bottom, large, unstable wedges of rock in the abutments could endanger the safety of the
dam. In fact, the first failure of an arch dam at Malpasset Dam in 1959 resulted from dis-
placements of a large wedge of rock in the left abutment.

Sliding stability along the dam-foundation contact of a concrete arch dam is less likely
because of the wedging produced by arch action and embedment of the structure into the
rock. However, arch dams with relatively flat abutment slopes, or arch dams with abut-
ment thrust blocks supported by rock foundations with inadequate shear strength could be
susceptible to sliding along the foundation contact and should be considered.

11-2
Other cases requiring special considerations include structural deformations and deterio-
ration of concrete caused by alkali-aggregate reactions, and foundation or abutment ero-
sion due to overtopping, which if severe, could lead to instability.

11-1.4 Evaluation Criteria

Existing concrete arch dams should be evaluated by conducting a review and analysis of
all existing data, a field inspection, and any analyses necessary to determine the safety of
the dam for continued normal operation and resistance against the unusual and extreme
loading conditions.

11-1.4.1 Review of Existing Data and Site Inspection

A thorough knowledge must first be gained on a dam's original design and its perform-
ance history and records, to provide a basis for evaluation and any further studies that
might be required. The existing data for review can be obtained from the owner's files or
from the files maintained by the FERC. The review should reveal whether the original
design criteria and assumptions, materials investigations, geological and seismological
studies, and design analyses are satisfactory based on the current practice, and if not,
whether they are acceptable. The same should also be applied to any modifications or al-
terations in design and on any subsequent analyses, investigations, or reviews. Of par-
ticular value are data on instrument readings that provide information on actual perform-
ance of the dam.

It is sometimes difficult to locate historic data from older projects. In addition to the
owner’s and FERC’s files, other sources may include articles in periodicals and technical
journals, state dam safety engineer’s files, consultant’s records, construction contractor’s
records and personal files of individuals who worked on the project.

After completion of the review of existing data, a site inspection should be carried out to
observe the present condition of the dam, and to resolve any discrepancy that may exist
concerning available data such as drawings, instrument data, and other pertinent informa-
tion. The inspection should also provide the opportunity to identify any cracks, deterio-
rated joints, and other distress conditions that need to be considered in the evaluation of
dam safety.

In some cases, an adequate determination of dam safety might be possible from the re-
view and analysis of existing data and field observations. If the review and observations
indicate that additional analyses are required, then these analyses should be performed as
follows:

11-3
11-1.4.2 Method of Analysis

Three dimensional finite-element analysis is preferred for the static and dynamic analysis
of arch dams. Trial load method may be used for static stress analysis only, if the dam has
a simple geometry and uniform material parameters can be assumed for the concrete and
for the foundation rock. Other mathematical formulations and approaches can also be
employed, but the accuracy of such methods should be verified by comparison with the
finite-element analyses.

11-1.4.3 Evaluation for Static Loading

The performance of concrete arch dams under static loading conditions should be evalu-
ated using deflections and stresses. Concrete and foundation rock material parameters
used in the analyses should be determined on the basis of field and laboratory in-
vestigations (Sections 11-2.4 and 11-3). In situations where determination of certain ma-
terial parameters is neither cost effective nor conclusive, their effects on the dam response
should be evaluated by parameter sensitivity analyses (Section 11-5.5). All applicable
static loads should be considered and combined according to their probabilities of occur-
rence in three categories of Usual, Unusual, and Special loading combinations (Section
11-4).

The basic results of analyses should include both deflections and stresses developed in the
dam. Plots of computed deflections provide a visual means for checking the numerical
results, and whenever possible they should be correlated with the observed deflections
measured by instrumentation monitoring (Section 11-7), in order to verify and possibly
calibrate the mathematical model. The initial position and temperature of the dam is not
typically known. For this reason, when comparing computed deflections to observed de-
flections, it is the differential deflection rather than the absolute deflection that is mean-
ingful.

Stress results are used to evaluate the dam performance in the response to each loading
combination. The evaluation starts with comparison of the computed stresses with
strength of the concrete reduced by a factor of safety (Table 11-1.1), but will also involve
determination of location, magnitude, extent, and direction of high stresses should some
crack-inducing stresses be expected. If all factors of safety are met the dam is considered
to perform satisfactorily, even though some minor contraction joint opening may occur.
Otherwise joint opening and cracking could be significant and should be evaluated in ac-
cordance with procedures outlined in Sections 11-1.4.3.1, 11-1.4.3.2, and 11-1.4.4.

11-4
Table 11-1.1 Factors of Safety for Existing Arch Dams

Internal Shear
Compressive Tensile Sliding 1
Loading Combination Stresses
Stresses Stresses Stability
Usual (normal operating) 2.0 1.0 2.0 1.5
Unusual (flood condition) 1.5 1.0 1.5 1.5
Extreme (seismic) 1.1 1.0 1.1 1.1
1
Factors of safety valid for the assumption of no cohesion.

11-1.4.3.1 Performance for Usual and Unusual Loading Combinations

Arch dams resist applied static loads by developing primarily compressive stresses along
the arch sections. The adequacy of the dam under a given load combination should be
evaluated in accordance with the compressive stress and shear stress criteria listed in Ta-
ble 11-1.1.

The stress results produced by the linear-elastic finite-element analysis usually indicate
some areas of tensile stress in the dam. While tensile strength of the intact concrete can
reach several hundred psi, the fact that a typical arch dam is made of concrete blocks di-
vided by lift joints, vertical contraction joints, and pre-existing cracks should be consid-
ered in the evaluation of tensile stresses (Figure 11-1.1).

Fig. 11-1.1 Existing arch dam with vertical contraction joints, horizontal lift
joints, and pre-exiting cracks

The tensile strength in the direction normal to the contraction joints and cracks is very
small, and across the lift joints may be only a fraction of the tensile strength of the intact

11-5
concrete. In some cases, joints and cracks may have no tensile strength at all. For this
reason, it is not appropriate to evaluate the indicated tensile stresses of a finite element
model in terms of an allowable tensile stress for the intact concrete alone.

When a finite element model does indicate regions of tensile stress, the reviewer must
realize that these stresses are probably not an indication of the actual stress state of the
dam, but a consequence of the modeling assumptions of linear elasticity and structural
continuity. Thus, large areas of indicated tensile stress may reveal a problem with the
assumption of linear-elastic behavior and not necessarily a problem with the performance
of the dam.

The effect of tensile stress relief by joints and cracks is to increase compressive and pos-
sibly shear stress in other areas of the dam. If a finite-element model indicates large areas
of tensile stress, or tensile stresses that are high (see Section 11 -3.7.2), the finite element
model should be modified to account for the loss of tensile resistance due to joint opening
or cracking. The modified model should then be re-run and evaluated. This process will
require some judgement on the part of the reviewer and the analyst as to when indicated
tension is excessive thus requiring model modification. If the combination of linear finite
element analysis and engineering judgement is not sufficient to determine whether or not
a dam is safe, non-linear finite element analysis may be required.

11-1.4.4 Evaluation for Seismic Loading

The performance of concrete arch dams under earthquake loading should be evaluated by
conducting a three-dimensional linear-elastic dynamic analysis using the finite-element
method (Section 11-6). The FE model of the dam system should account for the dam-
water and the dam-foundation rock interaction effects. Material parameters for the con-
crete and foundation rock should be established by giving due consideration to the effects
of the rate of loading typical of earthquake response. The design earthquake for the safety
evaluation of arch dams is the maximum credible earthquake (MCE). The MCE is an
extremely rare event capable of producing the largest ground motion that could ever occur
at the dam site. An MCE should be considered to be an extreme loading condition, for
which significant damage would be acceptable, but the dam must not rupture and thus
threaten life and property downstream. Seismic input ground motions for the MCE
should be developed from a deterministic ground motion analysis, but may be supple-
mented by a probabilistic ground motion analysis should evaluation of the likelihood of
the MCE ground motions become desirable. The earthquake response of the dam may be
computed using the response-spectrum mode-superposition method, but if maximum

11-6
stresses exceed the allowable values a linear time-history analysis can be helful to as-
sessing the severity of joint opening and tensile cracking .

For evaluation of the earthquake performance of arch dams using response-spectrum

method, dynamic stresses are combined with static stresses due to the usual loading
combination and compared with the allowable values. The evaluation criterion for time-
history analysis, however, is more involved than simple stress checks. It considers not
only the maximum stress values, but also the sequencing, spatial extent, and number of
excursions beyond the allowable values. In addition, in cases where severe damage is
predicted, sensitivity analyses should be performed to account for uncertainties associated
with modeling assumptions, seismic input application, and material properties. Horizon-
tal lift joints and vertical contraction joints should be assumed to crack when subjected to
tensile stresses exceeding their tensile strengths. In situations where there is a net tensile
force across a vertical contraction joint, it should be assumed that the contraction joint
will open through the full thickness of the dam, possibly forming a free cantilever block.
(See 11-6.3.3.5) The evaluation of dam safety using linear elastic assumptions requires a
qualitative judgement of how stresses will be redistributed during joint opening and
cracking. This evaluation is done in lieu of more sophisticated nonlinear analysis. This
approach may not be sufficient for some situations and a more detailed analysis using
non-linear techniques may be required. The dam may be considered safe for the MCE if,
after the effects of crack and joint opening have been accounted for, it can be shown that
the concrete is not over-compressed and free cantilevers do not topple.

11-1.4.4.1 Post-earthquake Safety Evaluation

A post-earthquake safety evaluation is required to assure the safety of the dam if, a dam-
aging MCE should occur near the dam site, or the predicted performance of the dam due
to a postulated MCE should indicate substantial damage. This evaluation should consider
the effects of static loads as well as severe aftershock earthquakes that invariably occur
after any major quake. Factors of safety for the post-earthquake conditions are the same
as those given in Table 11-1.1 for the usual case.

11-1.4.5 Sliding Stability

To assure safety against sliding along identified kinematically admissible failure planes in
the dam , at the dam foundation/interface, or in the foundation, the shear friction factor of
safety assuming no cohesion shall be 1.5 for normal and unusual loading, and greater than
1 for extreme loading. These safety factors assume that stability has been evaluated with
respect to conservative shear strength parameters.

11-7
For major dam structures subjected to severe seismic loading, response history analyses
should be performed for abutment and foundation stability instead of the usual pseudo-
static analyses. In response history analyses, factor of safety varies with time and may be-
come less than 1.0 for one or more cycles provided that the resulting cumulative sliding
displacement is very small and can be tolerable.

11-8
11-2 FOUNDATION CONSIDERATIONS
11-2.1 General
The abutment foundations of an arch dam are particularly critical to the stability of the dam
because they are required to resist the majority of the reservoir forces that attempt to push
the structure in a downstream direction. The modulus of deformation of the abutments and
the foundation is also an important element in analyzing the performance of the dam since
the flexibility of the foundation directly affects the stresses in the dam. Foundation informa-
tion must provide sufficient geological detail to identify and locate any potential sliding
wedges of rock that could cause instability. If such features are found to exist, a stability
analysis must be performed to assure that there is an adequate factor of safety against abut-
ment sliding. For some existing dams, sufficient data and analyses are already available to
provide the necessary information. In other cases it will be necessary to perform field
investigations and conduct stability analyses. In either event, the staff must require that
sufficient abutment foundation information and analysis be provided to support a review that
verifies the findings with regard to the stability of the foundation and abutments. The same
requirement is true for the foundation modulus of deformation.

11-2.2 Field Investigations

Field investigations are well described in Chapter 5 of these Guidelines and in the U.S. Army
Corps of Engineers EM 1110-2-2201 (1994). Additional details are contained in EM 1110-
1-1804 (1984) and EM 1110-1-2908 (1994) (See References). The following narrative is
intended as a summary and for the purpose of additional elaboration where required to
specifically address the requirements of this Guideline.

It is imperative that the conditions of the foundation at the site be well defined. In particular,
geologic investigations as outlined in Chapter 5 should provide an interpretation of the rock
type and quality, identify the discontinuity (joint) pattern, locate any planes or wedges of rock
which could fail under the structural loading, provide samples and data for determining the
rock mass modulus of deformation, the bearing capacity and the shear strength available to
resist failure.

11-2.2.1 Foundation Features That Create Stability Concerns and Warning Signs

Jointing - A feature of primary concern is a large wedge of rock in an abutment foundation

created by a planar rock fracture or the intersection of two or more rock fractures whose
intersection trend daylights in a downstream direction. Refer to EM 1110-2-2201, pages
10-28 and 29, for examples. Because of the high intact strength of most rock formations,

11-9
failure is improbable unless it can occur along preexisting fractures. For a failure to occur,
movement of the rock wedge must be kinematically possible. In other words, the orienta-
tion of the trend of the intersection of the rock fractures or slide plane must normally daylight
in a direction which would allow movement to take place under the applied loads without
shearing a great deal of intact rock. A relatively small amount of intact rock may sometimes
be sheared when the trend comes close to daylighting without it actually occurring.

In addition to joint orientation, joint connectivity must be considered. Joint connectivity

determines whether kinematically possible wedges are small, and of little consequence, or
large and capable of compromising the stability of the dam.

Hydrostatic Pressure - The stability of an abutment rock wedge is affected by the hydrostatic
pressure in the joints that define the wedge as can be seen in the figure below. The drilling
of joint drainage holes to relieve hydrostatic pressure is often very effective in increasing
wedge stability.

Fig. 11-2.1

Faults - Zones of faulted or sheared rock within the foundation must be carefully considered.
A fault is a rock fracture distinguished from a joint by virtue of translational movement of
one wall relative to the other wall at some time during the geologic record. If a fault is found
to be present, the question as to whether it is active or inactive must be answered. If it is
determined to be an active fault, its effect upon the structure during movement must be very
carefully assessed and appropriately acted upon.

Next, the fault's effect upon the static stability of the foundation must be determined. Since

11-10
it is a presheared feature in the rock, it probably provides a plane of reduced strength to resist
movement. In many cases slickensides and clay gouge are formed which greatly reduce the
rock strength. Its orientation is significant in the effect it has on reducing stability against
sliding in the foundation.

Another concern is the permeability of a through-going fault. If the sheared rock was very
brittle and the shearing process formed a zone of primarily broken rock (breccia), it may
form a highly permeable path for water passage beneath the dam. If the shearing movement
forms a clay gouge within the breccia zone, the result may be a very impervious barrier in
the foundation. Such a barrier to seepage can in some configurations result in the develop-
ment of abnormally high uplift pressure in the foundation. These questions must be ad-
dressed, satisfactorily answered and incorporated into the stability analysis of the dam.
Treatment may consist of curtain and consolidation grouting of a highly pervious shear zone
or drains to relieve abnormal uplift created by a highly impervious shear zone. The possibil-
ity of erosion of gouge must also be considered.

The effect of a large faulted shear zone on the modulus of deformation of the foundation
must be taken into consideration. A large change in modulus over a very short distance may
result in the formation of concentrated stresses in the concrete of the dam if the shear zone
was not properly treated during construction. Treatment usually consists of excavation of the
sheared material to a depth from two to three times its width followed by backfill with dental
concrete.

Coal Seams - Coal seams or beds in the foundation of an arch dam are a feature of concern.
The clay layers associated with coal beds are an even greater concern. This combination in
the foundation of an arch dam can form a plane with significantly lower shearing resistance
than the surrounding rock. It should be evaluated both for planar failure and as a wedge in
combination with the fracture pattern existing in the rock mass.

Planar Features - Planar features such as bedding, fisility, shale or clay seams, schistocity,
foliation, cleavage, and stress relief features such as exfoliation and valley relief joints may
all form sides of a rock wedge and therefore are features of some concern to be included in
the abutment foundation stability analysis.

Sudden Changes in Stiffness - Adjacent rock beds with radically different moduli of defor-
mation are of some concern. This difference must be taken into account during the stress
analysis for the dam.

11-2.3 Material Parameter Selection

11-11
Experienced engineering and geologic judgement are very important in the selection of
foundation material parameters for use in the analytical procedures. Since it is often not
feasible to make prototype tests and measure directly the rock mass strength and deformation
properties, it is necessary to use laboratory tests of small samples as the basis for estimating
these properties. Laboratory testing may be required to identify rock types and to provide
shear strength and rock mass modulus of deformation data. The testing program must be
carefully developed by the Independent Consultant to provide sufficient tests to establish a
statistically sound basis for estimating shear strengths of all critical features as well as the
rock mass modulus of deformation. Refer to the Rock Testing Handbook (1990) and to
Chapter 10 of EM 1110-2-2201 (1994) for more detail on laboratory testing than is provided
in this Guideline.

Three parameters may be involved in providing shearing resistance to sliding at the interface,
in the foundation and in the abutments. These include angle of friction, cohesion, and the
angle of the rock asperities. Circumstances will dictate the applicability of using cohesion
versus asperity angle. With smaller structures where the normal force applied by the dam
and the weight of the overlying rock is insufficient to shear through the rock asperities during
sliding, no cohesion parameter should be counted in the analysis. Instead, the angle of the
asperities should be added to the angle of friction as a resisting parameter, since no move-
ment can occur without overriding the asperities. This results in dilation or lifting, thereby
requiring an increased driving force to slide the structure. Conversely, in large dams that
impose high normal forces, shearing through asperities may occur rather than riding over
them during sliding failure. In this circumstance it is proper to use cohesion as a resisting
force rather than the angle of the asperities. Further explanation is contained in the following
paragraphs.

11-2.3.1 Shear Strength of Foundation Interface

Factors to be considered in estimating the shear strength of the foundation interface include
strength of the bond of the concrete to the rock foundation, roughness or asperity angle "i"
of the interface, and embedment of the structure into the rock. Laboratory testing of intact
core samples of the interface can provide data on which to estimate the bond strength.
Sufficient core samples of the interface must be obtained to allow a statistically sound
appraisal of the percentage of the interface that can be reliably assumed to be bonded.

The roughness or asperity angle may be very difficult to estimate and because of this may
have to be ignored. In some cases it may be possible to estimate from photographs of the
foundation taken just before concrete placement. Another possibility is to estimate the

11-12
irregularity from closely spaced core borings. Refer to Figure 11-2.2b for a diagrammatic
representation of interface roughness.

Where information exists for determination of an asperity angle at the interface between the
structure and the foundation, this angle may be added to the friction angle as a resisting force
in the stability analysis if the least resistance to sliding includes overriding the irregularities.
It is not applicable where the least resistance is developed by shearing through the rock of
the irregularities. Refer to Figure 11-2.2b for an example to illustrate the effect that founda-
tion roughness may have in resisting sliding at the interface.

Embedment may possibly be determined from as-built drawings, construction photographs,

and borings. This factor can be very important for preventing sliding on the interface pro-
vided the concrete is placed against the embedded surface, which would mobilize the down-
stream rock strength before movement could occur.

The process of determining interface strength for arch dams is not as straight forward as is
the case of gravity dams. This is because failure mechanisms must be considered in 3
dimensions. As can be seen in Figure 11-2.2a, the direction of shear force at the
dam/foundation interface changes with respect to position along the valley. In one area, the
shear force may be parallel to the strike of the asperities, in another area it may be perpen-
dicular. Strength must be defined with respect to the direction of shearing force. As in the
case of indicated tensile stresses, local exceedance of the shear strength of the interface may
not be an indication of dam failure. Excessive shear stress may be able to be re-distributed.

11-13
 Fig. 11-2.2a Direction of shear stress at dam/foundation interface

Fig. 11-2.2b Concrete to rock foundation interface showing asperities interrupting poten-
tial failure plane

11-14
The strength of the concrete to rock interface where the concrete is bonded is represented by
the angle of friction plus cohesion. Data needed for determination of conservative values of
the angle of friction and cohesion are obtained by laboratory direct shear tests. Normal
stresses for these tests are determined by bracketing the stress level expected in the founda-
tion. The tests should include a suite of core samples where the concrete is bonded to the
foundation rock.

The strength of the concrete to rock interface where the concrete is not bonded to the rock
is represented by the angle of friction and either the apparent cohesion or the asperity angle.
To determine the angle of friction and the apparent cohesion, a suite of core samples should
be tested in which the interface is either not bonded naturally or the bond has been previously
broken. Judgement must be carefully exercised in determining whether apparent cohesion
may be allowed along unbonded portions of the potential failure plane. Test results will
frequently result in a cohesion intercept on the shear stress diagram even though the contact
is unbonded. This is probably explained by shearing of the concrete during testing due to the
roughness of the interface. If it is determined that apparent cohesion is to be used, it is
prudent to ignore the asperity angle since they both reflect shearing resistance due to inter-
face roughness. There may be some exceptions where it can be demonstrated that there are
two or more orders of asperity angles existing at the interface, i.e. fine versus coarse irregu-
larities. In this case, the tested samples would reflect shearing strength caused by the fine
asperities, while the structure itself would mobilize the resistance provided by both fine and
coarse asperities.

If it is decided that shear resistance is a combination of shear through asperities and asperity
ride up, strain compatibility may become an issue. It is likely that the shear through asperi-
ties will occur at lower strains than asperity ride up. Therefore, frictional resistance from
aspertity ride up may not be mobilized simultaneously, and can not be added to the ultimate
shear strength.

The strength parameters resisting failure along a plane at the interface should be weighted
according to the percentage of the plane where bonding of concrete to rock is expected, the
percentage where unbonded contact is expected, and the percentage where shear through rock
or dilation over the rock irregularities is expected. Where shear through rock is a factor, it
must be determined whether the failure would likely follow a natural rock fracture or would
be required to shear through intact rock. The latter determination can make a very large
difference in the estimation of resisting forces because of the large difference in the strength
of intact rock versus fractured rock.

Since there is likely to be a great deal of scatter in the results of the laboratory tests, it is

11-15
prudent to rationalize and select conservative strength parameters based on the test results.

11-2.3.2 Shear Strength of Potential Foundation Failure Planes and Wedges

The abutment foundations are particularly critical to the stability of an arch dam as previ-
ously discussed in Sections 11.2.1 and 11-2.2.1. Abutment instability can develop along
either a planar discontinuity or a combination of planar discontinuities which intersect to
form an unstable wedge. Procedures for determining the shear strength of potential foun-
dation failure planes and wedges are discussed in detail in Chapter 5 of these guidelines.

11-2.3.3 Foundation Modulus of Deformation

The deformability of the foundation of an arch dam can affect the behavior of the structure
because the dam and foundation function together as an integrated unit. The modulus of
deformation provides a measure of this property. It is a representation of the deformational
property of the rock mass as a whole, with all its discontinuities, as contrasted with the
modulus of elasticity of an intact specimen of the rock.

There are different approaches to developing an estimation of the modulus of deformation.

The most direct measurement can be made by performing static in situ jacking tests in
abutment adits. There are also procedures available for measuring the dynamic elastic
properties of a rock mass using seismic techniques. It is generally accepted that the lower
modulus values provided by the jacking tests are more appropriate for use in arch dam
foundation analysis because it appears that this technique better models the effect that dis-
continuities have on the foundation. However, the results obtained from a jacking test are
very local in nature, and may not be appropriate for the foundation in another area.

In the case of existing structures it is unlikely that an adit will be available in which to
perform either jacking or seismic tests. In some cases these tests may have been performed
for the original design and the results may still be available. In this case a review of the
existing data may be all that is required to develop an estimate of the foundation modulus of
deformation.

Where no data is available, it is possible to develop an estimate of the foundation modulus

by testing representative intact specimens of the rock obtained from core samples to deter-
mine modulus of elasticity of intact rock, then applying an appropriate reduction factor to

11-16
convert from the modulus of elasticity of the intact rock to the modulus of deformation of
the rock mass. Refer to Hendron (1968) for a study which demonstrated that the fracture
frequency in the rock mass is a primary factor in the reduction of the elastic modulus of a
rock mass from the modulus of an intact specimen. He provides examples of how rock
quality designation (RQD) and velocity ratio may be used to estimate the appropriate reduc-
tion factor.

TABLE 11-2.1 (From Bieniawski, 1990)

GEOMECHANICS CLASSIFICATION OF JOINTED ROCK MASSES

A. CLASSIFICATION PARAMETERS AND THEIR RATINGS

PARAMETER RANGE OF VALUES

1 Strength Point-load >10MPa 2 - 4 Mpa 2 - 4 MPa 1 - 2 Mpa For this low range, uniaxial com-
of strength index pressive test is preferred
intact
rack Uniaxial
compressive >250MPa 100-250MPa 50–100 MPa 25-50MPa 5-25 1 -5 <1 MPa
material MPa MPa
strength

Rating 15 12 7 4 2 1 0

2 Drill core quality RQD 90% - 100% 75% - 90% 50% - 75% 25%- 50% <25%

Rating 20 17 13 8 3

3 Spacing of discontinuities >2m 0.6 -2 m 200-600 mm 60 - 200 mm < 60 mm

Rating 20 15 10 8 5

Very rough Slightly rough Slightly Sickensided Soft gauge > 5 mm thick
surface Not surface Sepa- rough sur- surface OR OR
4 Condition of discontinui- continuous ration < 1 mm face Separa- Gauge < 5 Separation >5 mm
ties No separa- slightly tion < 1 mm mm thick Continuous
tion weathered Highly OR
Unweathered walls weathered Separation
wall rack walls 15 mm
continuous

Rating 30 25 20 10 0

inflow per None < 10 liters/min 10-25 li- 25-125 > 125 liters/min
10 m ters/min iters/min
Ground tunnel OR ____
5 Water length OR _ ___ OR _ OR ____ OR _
Ratio: 0.0 - 0.1
(joint water 0 0.1 - 0.2 0.2 - 0.5 > 0.5
pressure)/(major
principle stress)
OR ____
OR ____ OR __ OR __ OR _
Damp
General Dripping
Conditions Dry Wet Flowing

Rating 15 10 7 4 0

11-17
B. RATING ADJUSTMENT FOR JOINT ORIENTAITON
Strike and dip orientations for joints Very favorable Favorable Fair Unfavorable Very unfavorable
Tunnels 0 -2 -5 -10 -12
Ratings Foundations 0 -2 -7 -15 -25
Slopes 0 -5 -25 -50 -60

C. ROCK MASS CLASSES DETERMINED FROM TOTAL RATING

Rating 100←81 80←61 60←41 40←21 <20

Class No. I II III IV V

Description Very good rock Good rock Fair rock Poor rock Very poor rock

D. MEANING OF ROCK MASS CLASSES

Class No. I II III IV V

10 years for 6 months for 1 week for 10 hours for 2.5m 30 minute for
Average Stand-up time 15 m span 8 m span 5 m span span 1 m span
> 400 kPa 300-400 kPa 200-300 kPa 100-200 kPa <100 kPa
Cohesion of the rock mass

Friction angle of the rock mass >45 35 - 45 25 – 35 15 - 25 <15

Serafim and Pereira (1983) developed a relationship between Bieniawski's (1979) Rock
Mass Rating (RMR) system and the modulus of deformation of rock masses which has
been shown to be valid on other projects. This relationship is as follows:

(RMR-10)
40
For RMR’s < 58 E = 10

(11-2.1)
For RMR’s > 58 E = 2·(RMR) - 100

E = Modulus of Deformation measured in gigapascals (GPa)

1 GPa = 145,037.7 psi.

Factors included in Bieniawski's RMR are unconfined compressive strength or point load
strength index, RQD, spacing of discontinuities, condition of discontinuities, and ground
water. Refer to Table 11-2.1 for his Geomechanics Classification of Jointed Rock Masses.

The above described systems may be used to develop a rational basis for estimating the
modulus of deformation of the foundation of an existing dam. It must, however, be consid-
ered that different parts of a foundation may have very different moduli depending upon rock
type variations, abutment and valley location, geologic structure, etc. Refer to EM 1110-2-

11-18
2201 (1994) for additional guidance.

Borehole dilatometers and pressure meters are also available for performing in situ deforma-
tion tests in boreholes. Data from these tests may also be useful in estimating the rock mass
modulus of deformation

It must be realized that the modulus of deformation is often difficult to quantify. The
techniques discussed above can be useful, however it may be more prudent to run several
analyses with differing foundation modulii bracketing reasonable expected values rather than
to spend effort in laboratory testing and field investigations attempting to more precisely
quantify the modulus of deformation.

11-2.4 Foundation Rock Erodibility

The erosion of a plunge-pool downstream of a functioning arch dam spillway is a common

occurrence unless measures have been taken to prevent it. It is a natural way for the energy
of the falling water to be dissipated. Erosion of even very strong and massive rock can occur
at the location of impingement of the water falling from a high dam spillway. Two examples
of deep erosion of strong rock include the 79 ft. (24m) deep plunge-pool eroded in blocky
andesite at Alder Dam in the USA and the 65 ft. (20m) deep plunge-pool eroded in excellent
granite at Picote Dam in Portugal described by Mason (1984). Typically, the rate of plunge-
pool erosion decreases with depth until a stable configuration is reached. Plunge-pools are
sometimes planned for in the design of a spillway as a means of energy dissipation, but
plunge-pool formation can cause a stability problem if it continues to grow laterally, eroding
the dam foundation.

11-2.4.1 Field Investigations

Field investigations are conducted to determe the extent of previous erosion that may have
occurred and to provide data for determining the threat to the structure posed by possible
future erosion during flood events. The investigations should consider the erosion that has
occurred around the spillway(s) and at or near the toe of the dam with the intent of assessing
the long-term erodibility of the rock in these locations. Investigations may include hydro-
graphic surveys to establish the depth and extent of an existing plunge-pool, and by repeating
surveys after spillway flows it is possible to determine plunge-pool stabilization or continued
growth. Engineering geology investigations are conducted to provide data on the foundation
rock conditions including such things as rock type, fracture spacing and condition, bedding

11-19
frequency, zones of weaker rock such as softer beds and sheared rock zones, unconfined
compressive strength, orientation of beds and fractures, etc. Borings may be required to
provide data for the engineering geology investigations.

11-2.4.2 Assessing the Erodibility of Rock

The erodibility of rock has been the subject of numerous studies by both engineering geolo-
gists and hydraulic engineers. These studies have provided considerable insight into this
very complex problem. The complexity of the interaction of the water forces with the
endless variety of rock conditions encountered, however, makes each situation unique.

Rock erodibility is controlled by the following factors:

1. Intact rock strength (unconfined compressive strength).

2. Fracture frequency (size of individual rock blocks).

3. Orientation of fracture sets.

4. Shear strength and condition of fractures (continuity, roughness, aperture opening,

in-filling material and alteration or weathering condition of wall rock).

5. Weak planes in the intact rock (bedding, foliation, schistocity, fissility, etc.)

6. Faults and shear zones.

An assessment of rock erodibility must take these factors into account. The integration of
these factors is best accomplished by the use of one or more of the rock classification sys-
tems, such as the Geomechanics Classification developed by Bieniawski (1979 and 1990)
and Barton's Q System (1974 and 1988).

Plunge-pool growth is a function of hydrologic factors as much as erodibility of the bed

material. Mason (1984) presented an equation which related depth of scour to unit flow,
head drop, acceleration of gravity, tail-water depth and particle size of bed material. This
equation is lacking in that it looks at only one aspect of the bed's resistance to erosion, that
being particle size. He did, however, calibrate the equation using both model tests and actual
case histories and developed reasonable agreement with actual plunge-pools.

11-20
There have been studies made which related rock erodibility to an RMR System. Annandale
(1995) presents an erodibility index related to a rippability index developed by Kirsten (1982
and 1988) using Barton's (1974 and 1988) Q System as a basis. Another study was per-
formed by Cameron et al. (1986, 1988a, 1988b, and 1989) which relates rock erosion in
emergency chute spillways to Bieniawski's Geomechanics Classification System (1979 and
1990). These may be useful in assisting the independent consultant in his evaluation of the
likelihood of further development of a plunge-pool. This subject is still under investigation
by government agencies and other institutions. More reports further defining rock erodibility
may be expected in the literature in the future.

11-2.4.2.1 Quantitative Method of Assessing the Erodibility of Rock

There are two situations in which rock downstream from an arch dam can be subject to
erosion. Some arch dams are designed with an overflow spillway which is an integral part
of the dam. When the spillway operates, the rock in the area impacted by the jet can be
eroded. A plunge-pool is usually provided in the area where the jet falling from the spillway
impacts. The plunge-pool acts to dissipate the energy of the falling water and minimize
potential for erosion of the rock. If the rock in the impact area is hard and free of fractures,
an excavated but unlined plunge-pool may allow for satisfactory energy dissipation and
adequately limit erosion. However, if the rock in the impact area is fractured, a substantial
concrete lining of the plunge-pool may be required. For low arch dams with sound rock
under and downstream from the dam, the falling jet may not have the potential to erode the
rock.

An arch dam may be subject to overtopping during extreme floods if spillway capacity is not
sufficient. When overtopping occurs, water falling from the dam crest will impact rock all
along the downstream of the dam and the potential for erosion of the abutments becomes a
concern. Erosion of abutments is a particular concern for arch dams for which rock in the
abutments is not sound. Such overtopping can pluck out loose rock. Joints in the abutment
rock will make the abutments particularly susceptible to erosion. Thus, it is important to
consider the possible effects of erosion that would result from overtopping of the dam as that
resulting from operation of the spillway.

Two questions need to be answered relative to the erodibility of a given rock resulting from
overtopping of the dam or from spillway operation: 1.) Will the rock erode as a result of the
impact of the falling water?, and 2.) If the rock will erode, will the extent of erosion cause
a stability problem?. To answer the first question, information must be obtained on the

11-21
overall quality of the rock and the degree to which it is fractured; and a measure of the energy
with which the water impacts the rock must also be developed. At present, the second
question can only be answered in terms of past experience.

An article by Annandale (1995) provides a quantitative methodology by which an assessment

can be made of whether or not a given rock will erode under the action of falling water. The
method assumes that there is a relationship between the rate at which energy is dissipated in
the receiving pool of water (the stream power P) and the erodibility of the rock. The fol-
lowing Erodibility Index, developed by Kirsten (1982), was used as a quantitative estimate
of erodibility:

Kh = MsKbKdJs (11-2.2)

where Kh is the erodibility index, Ms is the mass strength number, Kb is the particle block
size number, Kd is the discontinuity or inter-particle bond shear strength-number and Js is the
relative ground-structure number. Higher values of the Erodibility Index indicate greater
resistance to erosion. In evaluating the Erodibility Index the following equations are re-
quired:

Kb = RQD/Jn (11-2.3)

where RQD is the Rock Quality Designation, a standard parameter in drill-core logging
(Deere 1988), and Jn is the joint-set number, which is a function of the number of joint sets
in a rock mass (Table 11-2.7). The particle block size number for cohesionless granular
materials can be determined directly with the following equation:

3
Kb = 1000 (D50) (11-2.4)

where, for a cohesionless granular material, D50 is the mean particle size. Kd is evaluated in
terms of the ratio Jr/Ja where Jr is the joint roughness number and Ja is joint alteration num-
ber. All of the independent parameters required to evaluate the Erodibility Index can be
assessed with standard geotechnical field techniques and laboratory tests. Tables 11-2.2
through 11-2.10, developed by Kirsten (1982), provide the means for quantitatively evaluat-
ing these parameters from the results of field observations and standard geotechnical tests.

11-22
A collection of data obtained for unlined spillways by the U.S. Soil Conservation Service and
the U.S. Bureau of Reclamation was used to develop a graphic relationship between the
erodibility index and rate of dissipation in the plunge-pool. The data included values for
materials ranging from 0.1 mm diameter through gravels, cohesive soils, vegetated soils,
weathered rock, and jointed and fractured rock. The resulting relationship between the
Erodibility Index and energy dissipation is plotted in Fig. 11-2.7, which indicates a division
between cases where erosion has and has not occurred.

Fig 11-3 Erodibility of Rock and Earth Materials by Hydraulic Jets (Annandale, 1995)

To use the method proposed by Annandale, the Erodibility Index must be evaluated for the
rock in which the plunge-pool will develop, which will require field observations and labo-
ratory tests. In general, the equation for Stream Power is:

P=QγE (11-2.5)

Where P is the stream power, Q is the flow rate, γ is the unit weight of water (62.4 Lb/ft3),
and E is the reduction in velocity head that occurs in the impact area (V2/2g). The rate at
which energy dissipation will occur in the impact area must then be evaluated. Techniques
and relationships for evaluation of the rate of energy dissipation can be found in The Hand-
book of Hydraulics by Davis and Sorensen (1969), Open Channel Flow by Henderson

11-23
(1966), Hydraulic Engineering by Roberson, Cassidy, and Chaudry (1988), and Open Chan-
nel Hydraulics by Chow (1959).

The method proposed by Annandale is at present (1997) the only available method by which
a quantitative estimate of erodibility can be made in terms of the hydraulic parameters of a
plunging spillway jet and characteristics of the rock in the impact area. It has had limited
application at the present time, and the results should be used with caution. Nevertheless,
Fig. 11-2.3 appears to indicate that the use of this method can give a reasonable conclusion
as to whether the rock downstream from a given dam should be expected to erode.

Table 11-2.2. Mass Strength Number for Granular Soils (M)

SPT Mass
Consistency Identification in Profile Blow Strength
Count Number
(M)

Crumbles very easily when scraped with geological

Very loose 0-4 0.02
pick

Small resistance to penetration by sharp end of geo-

Loose 4-10 0.04
logical pick

Considerable resistance to penetration by sharp end

Medium dense 10-30 0.09
of geological pick

Very high resistance to penetration of sharp end of

Dense 30-50 0.19
geological pick - requires many blows of pick for
excavation

High resistance to repeated blows of geological pick -

Very dense 50-80 0.41
requires power tools for excavation
Note: Granular materials in which the SPT blow count exceeds 80 to be taken as rock –
see Table 11-2.4.

11-24
 Table 11-2.3 Mass Strength Number for Cohesive Soils (M)
Vane Mass
Consistency Identification Shear Strength
Strength Number
(kPa) (M)

Pick head can easily be pushed in up to the shaft

Very soft 0 - 80 0.02
of handle. Easily moulded by fingers.
Easily penetrated by thumb; sharp end of pick can
Soft be pushed in 30 mm - 40 mm; moulded by fingers 80 - 140 0.04
with some pressure.
Indented by thumb with effort; sharp end of pick
can be pushed in up to 10 mm; very difficult to
Firm mould with fingers. Can just be penetrated with 140 - 210 0.09
an ordinary hand spade.

Penetrated by thumbnail; slight indentation pro-

duced by pushing pick point into soil; cannot be
Stiff moulded by fingers. Requires hand pick for exca- 210 - 350 0.19
vation.

Indented by thumbnail with difficulty; slight in-

Very stiff dentation produced by blow of pick point. Re- 350 - 750 0.41
quires power tools for excavation.

Note: Cohesive materials of which the vane shear strength exceeds 750 kPa to be
Taken as rock - see Table 11-2.4.

11-25
 Table 11-2.4 Mass Strength Number for Rock (M).

Unconfined Mass
Compressive Strength
Hardness Identification in Profile
Strength Number
(MPa) (M)

Very soft rock Material crumbles under firm (moderate) Less than 1.7 0.87
blows with sharp end of geological pick and
can be peeled off with a knife; is too hard to 1.7 - 3.3 1.86
cut triaxial sample by hand.

Soft rock Can just be scraped and peeled with a knife; 3.3 - 6.6 3.95
indentations 1 mm to 3 mm show in the
specimen with firm (moderate) blows of the 6.6 - 13.2 8.39
pick point.

Hard rock Cannot be scraped or peeled with a knife; 13.2 - 26.4 17.70
hand-held specimen can be broken with
hammer end of geological pick with a single
firm (moderate) blow.

Very hard rock Hand-held specimen breaks with hammer 26.4 - 53.0 35.0
end of pick under more than one blow.
53.0 - 106.0 70.0

Extremely hard Specimen requires many blows with geo- Larger than 280.0
rock logical pick to break through intact material.
212.0

11-26
 Table 11-2.5 Mass Strength Number for Detritus (M).

In Situ De- Mass

Consistency Identification in Profile formation Strength
Modulus Number
(MPa) (M)

Very loose Particles very loosely packed. High percent- 0-4 0.02
age voids and very easily dislodged by hand.
Matrix crumbles very easily when scraped
with geological pick. Raveling often occurs
in excavated faces.

Loose Particles loosely packed. Some resistance to 4 - 10 0.05

being dislodged by hand. Large number of
voids. Matrix shows small resistance to
penetration by sharp end of geological pick.

Medium dense Particles closely packed. Difficult to dislodge 10 - 30 0.10

individual particles by hand. Voids less ap-
parent. Matrix has considerable resistance to
penetration by sharp end of geological pick.

Dense Particles very closely packed and occasion- 30 - 80 0.21

ally very weakly cemented. Cannot dislodge
individual particles by hand. The mass has a
very high resistance to penetration by sharp
end of geological pick - requires many blows
to dislodge particles.

Very dense Particles very densely packed and usually 80 - 200 0.44
cemented together. The mass has a high re-
sistance to repeated blows of geological pick
- requires power tools for excavation.

Note: Determined by plate bearing test of diameter 760 mm.

11-27
 Table 11-2.6 Joint Count Number (Jc).

Number of Joints Per Ground Quality Des- Number of Joints Per Ground Quality Des-
Cubic Meter (Jc) ignation Cubic Meter (Jc) ignation
(RQD) (RQD)

33 5 18 55
32 10 17 60
30 15 15 65
29 20 14 70
27 25 12 75
26 30 11 80
24 35 9 85
23 40 8 90
21 45 6 95
20 50 5 100

Table 11-2.7 Joint Set Number (Jn).

Joint Set Number

Number of Joint Sets
(Jn)

Intact, no or few joints/fissures 1.00

One joint/fissure set 1.22

One joint/fissure set plus random 1.50

Two joint/fissure sets 1.83

Two joint/fissure sets plus random 2.24

Three joint/fissure sets 2.73

Three joint/fissure sets plus random 3.34

Four joint/fissure sets 4.09

Multiple joint/fissure sets 5.00

11-28
 Table 11-2.8 Relative Ground Structure Number (Js).

Dip Direction of Closer Dip Angle of Closer

Spaced Joint Set Spaced Joint Set
Ratio of Joint Spacing, r
(degrees) (degrees)
1:1 1:2 1:4 1:8

180/0 90 1.14 1.20 1.24 1.26

In direction of stream 89 0.78 0.71 0.65 0.61

flow 85 0.73 0.66 0.61 0.57
80 0.67 0.60 0.55 0.52
70 0.56 0.50 0.46 0.43
60 0.50 0.46 0.42 0.40
50 0.49 0.46 0.43 0.41
40 0.53 0.49 0.46 0.45
30 0.63 0.59 0.55 0.53
20 0.84 0.77 0.71 0.67
10 1.25 1.10 0.98 0.90
5 1.39 1.23 1.09 1.01
1 1.50 1.33 1.19 1.10

0/180 0 1.14 1.09 1.05 1.02

Against direction of -1 0.78 0.85 0.90 0.94

stream flow -5 0.73 0.79 0.84 0.88
-10 0.67 0.72 0.78 0.81
-20 0.56 0.62 0.66 0.69
-30 0.50 0.55 0.58 0.60
-40 0.49 0.52 0.55 0.57
-50 0.53 0.56 0.59 0.61
-60 0.63 0.68 0.71 0.73
-70 0.84 0.91 0.97 1.01
-80 1.26 1.41 1.53 1.61
-85 1.39 1.55 1.69 1.77
-89 1.50 1.68 1.82 1.91
180/0 -90 1.14 1.20 1.24 1.26

Notes: 1. For intact material take Ks = 1.0

2. For values of r greater than 8 take Ks as for r = 8

11-29
 Table 11-2.9 Joint Roughness Number (Jr).

Joint Separation Condition of Joint Joint Roughness Number

Joints/fissures tight or Discontinuous joints/fissures 4.0

closing during excava- Rough or irregular, undulating 3.0
tion Smooth undulating 2.0
Slickensided undulating 1.5
Rough or irregular, planar 1.5
Smooth Planar 1.0
Slickensided planar 0.5

Joints/fissures open and Joints/fissures either open or 1.0

remain open during ex- containing relatively soft
cavation gouge of sufficient thickness
to prevent joint/fissure wall
contact upon excavation.

Shattered or micro-shattered 1.0

clays

11-30
 Table 11-2.10 Joint Alteration Number (Ja).

Joint Alteration Number

(Ja) for Joint
Description of Gouge
Separation (mm)
1.01 1.0 - 5.02 5.03

Tightly healed, hard, non-softening im- 0.75

permeable filling

Unaltered joint walls, surface staining 1.0

only

Slightly altered, non-softening, non- 2.0 2.0 4.0

cohesive rock mineral or crushed rock
filling

Non-softening, slightly clayey non- 3.0 6.0* 10.0*

cohesive filling

Non-softening, strongly over-consolidated 3.0* 6.0** 10.0

clay mineral filling, with or without
crushed rock

Softening or low friction clay mineral 4.0 8.0* 13.0*

coatings and small quantities of swelling
clays

Softening moderately over-consolidated 4.0* 8.0* 13.0

clay mineral filling, with or without
crushed rock
Shattered or micro-shattered (swelling)
clay gouge, with or without crushed rock 5.0* 10.0* 18.0

Note: 1. Joint walls effectively in contact.

2. Joint walls come into contact after approximately 100 mm shear
3. Joint walls do not come into contact at all upon shear
4. * Values added to Barton et al's data
5. ** Also applies when crushed rock occurs in clay gouge without rock
wall contact

11-31
11-2.4.3 Erosion Downstream from the Dam

Jets that impact the downstream area with a large velocity (100 ft/sec or larger) will usually
produce erosion unless the rock is extremely hard and quite sound. Cases where erosion has
not occurred are rarely described in the technical literature since they do not create problems.
In actuality, erosion of rock has occurred in almost all cases where a high-velocity jet impacts
on rock, particularly if the rock in the impact area is of poor quality or is highly weathered
or fractured.

11-2.4.3.1 Historic Observations of the Depth and Extent of Erosion

Cases in which erosion downstream from spillways has occurred are abundant in the techni-
cal literature. However, information on erosion due to overtopping of a dam is scarce. Data
from the reported cases of erosion downstream from spillways has been used by several
engineers to provide an estimate of the depth of expected erosion due to impact of a falling
jet of water. The depth of erosion is a function of the velocity at which the jet enters the
plunge-pool, the angle at which the jet enters, the depth of water in the impact area, and the
physical character of the rock. Field and laboratory observations have shown that erosion
will continue with further operations, but at a decreasing rate. Bulletin Number 58 of the
International Commission on Large Dams (1985) states that significant plunge-pool erosion
has not been observed for many existing spillways. However, in many cases those spillways
have operated only for short durations and at small discharges since they were constructed.
Table 11-2.11 provides a summary of information on several cases where erosion down-
stream from a spillway has been observed and recorded.

Table 11-2.11 shows that in some cases deep erosion has been experienced even though the
rock in the impact area was apparently hard. In general the depth of erosion depends upon
the energy of the falling jet, the duration of the flow, and the character of the rock. The
extent of erosion depends mostly on the character of the rock. If the upper layers of rock are
more erodible than the deeper layers, the lateral extent of the erosion will spread to form a
wider hole since the harder lower layers will deflect the high velocities toward the more
erodible sides.

11-32
 Table 11-2.11 Historic Scour Depths of Plunge-pools
Material Country q1 Head3 Depth3

Alder Andesite USA 20,0002 300 79

Na- SS,MS4 USA 250
ciemento
Picote Granite Portugal 1250 213 118
Kariba Gneiss Zimbabwe 415 160
Tarbela Limestone Pakistan 570 320 121
854 323 160
Karakaya Turkey 1034 413 131
619 403 90
Keban Turkey 351 360 75
335 44
Killckaya Turkey 130 245 39
Elmali Granite 32 89 49
Kondo- Granite 149 39 21
poga
Cabora Mozam- 2957 300 223
Bassa bique 283 315 75
Ukai Basalt India 854 154 95
Guri Basalt Venezuela 52,8002 118
1
Discharge q is unit discharge in cfs/ft.
2
Only the total discharge for the spillway was available.
3
Head is given in feet.
4
Abbreviations are: SS-Sandstone, MS-Mudstone.

11-2.4.3.2 Analytical Methods for Calculating Erosion Depth Downstream from

Spillways

Maximum Depth of Scour. As pointed out in Section 11-2.5.2.1, erosion of concern can
occur downstream from an arch dam due to spillway operation if the spillway is located on
the dam.

Over the years, a large number of empirical equations have been developed for estimation
of the depth of erosion which should be expected in the area impacted by the jet produced
by operation of an overflow spillway. Mason (1985) has documented 25 different equations
developed for this purpose, dating from 1937. Figure 11-2.4 shows a typical cross section

11-33
of an arch dam with an overflow spillway and the parameters that have been found to be
important in the estimation of depth of erosion. Erosion of rock by an impacting jet occurs
when the rock is fractured. Turbulent eddies unsteady pressures produced by the impacting
jet attack, move, and eventually pluck out particles of rock. Tailwater depth has an important
influence on the maximum depth of erosion since deeper water in the impact area results in
greater energy dissipation within the pool. Deeper tailwater results in less erosion.

The parameters which have been found to be important in determining the depth of erosion
D are shown on Fig. 11-2.4. They include the head H (the vertical distance between the
reservoir surface and the tailwater surface), the tailwater depth h2, the angle at which the jet
enters the pool Θ, the discharge per unit width of the jet q, and the size of fractured rock
particles d.

Fig 11-2.4 Typical arch dam with overflow spillway and plunge-pool

The best known of these equations, and the simplest, is that by Veronese (1937):

0.225 0.54
D = 1.32 H q (11-2.6)

where D is the maximum scour depth (FT), H is the difference in elevation between the
reservoir and the tailwater surfaces (FT), and q is the unit discharge (CFS). The Veronese
Equation, as is true of all other forms of the equation, was derived by empirical analysis of
recorded data on erosion downstream of flip-bucket and overflow spillways. In the Veronese
Equation, the scour depth D is measured from the tailwater elevation. Thus, tailwater depth

11-34
is considered explicitly. This equation has been widely used by the design profession and
is given in the U.S. Bureau of Reclamation Design of Small Dams (1987).

Yildiz (1994) reasoned that the Veronese Equation had been developed for cases where the
spillway jet entered the tailwater at a nearly vertical angle as would occur for an overflow
spillway on a high arch dam as shown in Fig. 11-2.3. Yildiz suggested that the depth of scour
as given by Eq. 11-2.6 should be measured along a line tangent to the direction of the jet as
it enters the plunge-pool. Fig. 11-2.4 shows such a configuration. Yildiz then modified the
Veronese Equation as:

Sin Θ
0.225 0.54
D = 1.32 H q (11-2.7)

where Θ is the angle of inclination of the jet at the water surface in the plunge-pool. The
resulting scour-depth predictions using Eq. 11-2.7 give a better fit to the recorded erosion
depths downstream from spillways than is achieved using Eq. 11-2.8.

Although the size of fractured rock (the size of the fractured particles of rock that would be
plucked out by the action of the spillway jet) is considered to be important, neither Eq. 11-2.8
nor Eq. 11-2.9 considers rock size. Mason (1985), on the basis of data from model studies
as well as prototype experience, derived the following equation for depth of erosion:

q 0.60 H 0.05 h2.015

D = 3.72
g 0.30 d 0.10
(11-2.8)

In Eq. 11-2.8, g is the acceleration due to gravity, d is the mean rock size, and h2 is the
tailwater depth. Mason found that Eq. 11-2.8 gave scour depth predictions which agreed
well with prototype experience and also yielded maximum scour depths which agreed well
with the results obtained in model studies when H was considered to be the total energy in
the jet as it leaves the spillway. For prototypes, he found that the effect of accounting for
energy loss occurring on the spillway surface was negligible.

It appears that either the modified Veronese Equation 11-2.7 or Mason’s Equation 11-2.8 is
as good as is available for estimating depth of erosion downstream from a spillway. How-
ever, the user should always be aware of the fact that both equations are strictly empirical and
could either overestimate or underestimate the potential scour depth for a particular rock
condition or jet configuration.

11-35
Pattern of Erosion. The pattern of erosion downstream from a spillway is very much de-
pendent upon the characteristics of the rock in the impact area. If the joint pattern and
hardness of the rock is uniform throughout the area of impact and does not change with
depth, the pattern will be generally symmetric about the centerline of the impacting jet. The
length of the eroded hole will be from 3 to 5 times the depth of erosion and the width of the
hole will be from 2.0 to 2.5 times the width of the spillway jet.

If discontinuities in the rock exist, the potential scour depth and pattern may be strongly
affected. For example, if part of the rock is much harder and more durable than the average,
erosion will be uneven and strong lateral currents can be created in the basin. If lower layers
of the rock are more sound than the upper layers, both the lateral and longitudinal extent of
erosion will be increased significantly as a result of the high velocity of the jet being de-
flected toward the edges of the hole.

11-2.4.3.3 Erosion Resulting from Overtopping of an Arch Dam

Erosion downstream from an arch dam due to overtopping is quite different than that which
occurs as a result of operation of an overflow spillway. The depth of erosion in either case
is very much dependent upon the duration of the flow. Deep erosion downstream from a
spillway is to be expected and the overflow spillway is designed so that the jet from the
spillway impacts as far from the toe of the dam as possible. However, during overtopping
water would fall from the top of the dam and impact very near its toe. Depths of erosion
would be very much dependent upon the depth of overtopping; large depths of overtopping
would result in greater energy impacting the rock producing deeper erosion. By contrast,
depths of overtopping of a foot or less should not produce significant erosion of rock at an
arch dam site. If deep erosion were to occur near the toe of the dam the safety of the dam
could be compromised. Erosion of the abutments can be a serious concern because material
in the abutments is often less sound and more erodible than the rock at the toe of the highest
portion of the dam.

Spills are a relatively common occurrence and, depending upon the magnitude of the passing
flood, often occur with long durations. As a result, the depth and extent of erosion down-
stream from spillways have been documented for many cases as was shown in Section 11-
2.4.3.1. By contrast, overtopping of a dam, which is an infrequent event of relatively short
duration, has received less attention in the technical literature. A number of arch dams that
have been overtopped are listed in the technical literature. Unfortunately, little has been
published on any erosion resulting from the overtopping. Gibson Dam on the Sun River in

11-36
Montana is an exception; it was overtopped by depths of up to 6.5 feet for 20 hours in June
1964. Although some loose rocks were plucked out during the overtopping, the resulting
erosion was minor and did not have any effect on the safety of the dam.

Equations 11-2.7 and 11-2.8 were developed from data on long term operation of spillways.
They are not applicable for estimating the maximum depth of erosion which might occur as
a result of a single overtopping, since overtopping is a rare event of relatively short duration.
As the overtopping of Gibson Dam has shown, there is little reason to be concerned about
erosion from overtopping if the rock which is impacted by the falling jet is sound even for
relatively large depths of overtopping. However, if the rock is jointed, and analysis shows
that the dam could be overtopped by significant depths during passage of a flood, there may
be reasons to take precautions against damage by erosion, particularly on the abutments.
Such precautions could include removal of the rock in question and replacing it with dental
concrete, covering the rock in question with a layer of shotcrete, or placement of a reinforced
concrete slab along the toe of the dam in the area of concern.

1-2.4.3.4 Examples of Calculations for Maximum Scour Depth

The following table presents data and the results of calculating maximum scour depth for
several cases using both Equations 11-2.9 and 11-2.10. The cases were selected from Table
11-2.11 for those cases where all information needed for the calculation was available. For
the use in his equation (11-2.10), Mason used a rock particle size of 0.8 feet, which predicted
scour depths in close agreement with prototype experience for cases where the rock size was
not available.

Comparing the field-measured scour depths in Table 11-2.11 with the predictions of values
computed using the modified Veronese (Eq. 11-2.7) and Mason (Eq. 11-2.8) equations shows
that the two equations predict some situations well but miss others significantly. The modi-
fied Veronese Equation appears to give relatively good agreement, while Mason’s equation
appears to over-predict the maximum scour depth. Part of this may be due to the assumed
d = 0.8 ft for all of the cases where information on the fractured size of the rock was not
available. If the actual rock size were greater than 0.8 ft, the predicted scour depth would be
less. The results shown in Table 11-2.11 indicate that the modified Veronese Equation (Eq.
11-2.7) gives a reasonable estimate of scour depth and can be used for evaluation purposes.

Table 11-2.11 Example Calculations of Maximum Scour Depths

11-37
 Name q h2 H Field Eq. 11-2.7 Eq. 11-2.8
Θ
(cfs) (ft) (ft) (ft) (ft) (ft)
o
Elmali 337 40 13.1 88.6 49 54 81
Kondopoga 150 30o 3.3 39.4 21 23 39
o
Tarbela 80 50 50.8 323.1 161 142 185
o
Picote 1252 39 68.9 213.2 118 131 239
o
Ukai 851 43 13.1 154.2 95 107 145

1-2.4.3.5 Developing Water-Surface Profiles

In order to check potential scour conditions downstream from an arch dam, it is necessary
to calculate a point of impact of the spillway jet during passage of the flood. Computation
of the nappe trajectory provides the necessary parameters for use of the equations for estima-
tion of maximum scour depth, which are listed earlier in this section. This computation
involves the computation of the trajectory of the nappe leaving the spillway lip. Procedures
for estimating the nappe trajectory are given in the USBR’s “Design of Arch Dams”.

11-38
11-3 CONCRETE MATERIAL PARAMETERS
To perform a safety evaluation of a concrete arch dam as described in this chapter, the
properties of the dam concrete must be known. In the case of an existing dam, infor-
mation on the concrete properties may be available from the previous studies or con-
struction documents. If such information is adequate, only a review and interpretation of
the data may be necessary. Where little or no information, additional concrete investiga-
tions may be required. The scope of new investigations will depend on the amount and
quality of the available data and also on the existing conditions of the dam concrete.

The purpose of concrete investigation is to establish the quality of concrete and to obtain
its necessary properties for use in the structural analysis. It usually consists of the fol-
lowing field investigations and laboratory test activities:

11-3.1 Visual Inspection of the Concrete

A visual inspection of the dam is required to establish the extent of concrete deteriora-
tion, conditions of construction joints, location and size of cracks, areas of distressed
concrete, and locations for core drilling. All accessible concrete of the dam should be
visually inspected. Similarly, concrete cores taken for laboratory tests should be visually
inspected for voids, debris, joints, deterioration, and other defects.

Careful visual inspection is the most important investigation that can be done. For exam-
ple, it is usually the case that a crack in the concrete has far more effect on the behavior
of the dam than does a variation in ultimate compressive strength or the modulus of elas-
ticity. At times, elaborate and expensive coring and testing programs have been under-
taken, which in the end, reveal far less than an astute visual observation of the site.

11-3.2 Ultrasonic Pulse Velocity Test

The purpose of in-situ ultrasonic pulse velocity or UPV testing is to evaluate the overall
quality of concrete in existing concrete dams. Suitable equipment and standard proce-
dure for pulse velocity tests are described in ASTM C-597. The method is based on the
principle that the velocity of an ultrasonic pulse through a material is related to dynamic
modulus of elasticity, density, and Poisson's ratio of the material. Any changes in the
modulus caused by deterioration, cracks, poor compaction, voids, joints, etc., would af-
fect the velocity of ultrasonic pulses. Such defects and variations in the concrete in-
crease the ultrasonic transit time, and thus results in a slower velocity.

11-39
The UPV testing equipment consists of a transmitter and receiver transducer coupled to a
signal generator and a recording device displaying transit time. The piezoelectric trans-
ducers are placed against the dam to be examined in various strategic configurations. The
signal generator produces electric pulses which cause the transmitter to vibrate at its natu-
ral resonant frequency, producing sound waves that pass through the concrete, reaching
the receiver which will be detected by the recording device which will then display the
transit time. The velocity is calculated from the time elapsed for the pulse to travel a pre-
determined length through the dam or a test specimen

Path Length
UPV =
Transit Time

Unusually low velocities (less than 10,000 ft/sec) or a wide range of measured velocities,
generally indicate deteriorated or poor quality concrete. Table 11-3.1 shows the general
classification used in correlating the UPV with the quality of concrete. As shown in the
table, normal quality concrete typically produce velocities in the range of 12,000 to
15,000 ft/sec.

Attempts to correlate pulse velocity data with concrete strength parameters have not gen-
erally been successful. Although UPV does not directly measure strength, it does indi-
cate general condition and uniformity and can be used to complement and correlate with
the information obtained by visual inspection and core sampling. The UPV works effec-
tively, if both surfaces of the concrete dam are accessible and sound waves pass through
minimum number of joints. Because with each pass through a joint, the wave front is
dispersed and losses 50 to 70 percent of its amplitude.

Table 11-3.1 (Leslie and Cheesman 1949)

Ultrasonic Pulse Velocity, ft/sec General Condition of Concrete

Above 15,000 Excellent
12,000 - 15,000 Generally good
10,000 - 12,000 Questionable
7,000 - 10,000 Generally poor
Below 7,000 Very poor

11-40
Sonic Coring Tests.
Sonic coring (sonic logging) tests work the same way as the UPV tests, except that the
transmitter and receiver are sealed and placed in adjacent vertical core holes to check the
quality and uniformity of concrete. During the tests, coring holes are filled with water to
create acoustic coupling for transmission of ultrasonic pulses between transducers and the
concrete. This method can be used in certain situations, such as underwater regions of
the dam or when a more controlled spot check for establishing the quality of concrete is
required.

11-3.3 Concrete Coring and Specimen Parameters

Concrete coring should be performed in conformance with ASTM C42. The purpose of
coring is twofold. First, a random coring and testing program can be used to determine
the uniformity of the concrete, and to locate problem areas. Second, once potential
problem areas are discovered, coring can be concentrated in these areas to better define
properties. While average values of strength and elastic modulus are of some value for
structural analysis, investigations should focus on "weak links" since these problem areas
are more likely to govern the safety of the dam, than the average properties.

Normally, concrete cores are extracted from the downstream face of the dam by drilling
horizontally or from inside the dam galleries. When vertical drilling is done to extract
samples, care should be taken to obtain samples with intact bond between construction
lifts so that the strength of lift joints can be determined.

The condition of the entire core should be accurately logged during drilling. The impres-
sion of an experienced engineer during a visual inspection of the extracted core is very
important in ensuring that the test results are indicative of the condition of the dam.

Concrete cores extracted from different locations generally show different strength de-
pending on the batches of concrete placed at that location. They are also influenced by
the aggregate sizes within a particular specimen and the local deterioration of the mass
concrete. The material parameters from a testing program should therefore be based on
the overall condition of all cores and deterioration of the dam concrete and not just on a
selected “best” core samples.

11-3.4 Petrographic Examination of Concrete

Where there is evidence of concrete deterioration, a petrographic examination of concrete
specimens should be carried out to ascertain the presence of any deleterious chemical re-

11-41
actions such as alkali-aggregate reactivity. This examination should be conducted in ac-
cordance with ASTM C856, Petrographic Examination of Hardened Concrete.

11-3.5 Elastic Properties

An estimate of the elastic modulus of concrete is necessary for calculating stresses in-
duced in the dam by strains associated with loading. Poisson's ratio, which relates the
lateral strain to axial strain within the elastic range, is also needed for arch dams.

The static modulus of elasticity (chord modulus) and Poisson's ratio should be deter-
mined in accordance with the standard test method described by ASTM C469. The
modulus of elasticity and Poisson's ratio of concrete typically are measured in a compres-
sive strength test of specimen that is loaded to failure in a period of about 2 to 3 minutes,
and the modulus obtained in such a test may be called the short-term or instantaneous
value. However, concrete is a material that exhibits considerable creep under prolonged
loading, so the short-term modulus is not suitable for evaluating the deformations of a
dam subjected to continuing static loads. The effects of creep resulting from such loads
can be presented by using a reduced modulus of elasticity in the static displacement cal-
culations. Typically the short-term modulus is reduced by 25 to 30 percent to obtain a
sustained modulus value in prolonged loading. The elastic modulus is also affected by the
rapid seismic loading, which is discussed in Section 11-3.8.

11-3.6 Thermal Properties

The basic properties required for performing a thermal stress analysis include coefficient
of thermal expansion, specific heat, thermal conductivity, and thermal diffusivity.

Coefficient of Thermal Expansion. In general, concrete expands with heating and con-
tracts with cooling. The strain associated with the change of temperature depends on the
coefficient of thermal expansion and on the degrees by which temperature rises or drops.
The coefficient of thermal expansion for concrete varies directly with the coefficient of
thermal expansion of the aggregates, and typically the values range from 3.5 to 7 × 10-6
in./in./ °F. In the absence of measured data, an average value of 5 × 10-6 in./in./ °F may
be used.

Specific Heat. Specific heat is storing heat capacity per unit temperature. Compared to a
specific heat of 1.0 for water, specific heat of mass concrete typically varies between 0.20
and 0.25 Btu/lb-°F. In the absence of measured data, an average value of 0.22 Btu/lb-°F
should be used.

11-42
Thermal Conductivity. Thermal conductivity is a measure of the ability of a material to
direct heat flow. Typical values of thermal conductivity for mass concrete ranges from
13 to 24 Btu-in./hr-ft2-°F.

Thermal Diffusivity. Thermal diffusivity is the rate of heat flow through a unit area di-
vided by the product of the specific heat times the density times the gradient.
Rate of heat flow through unit area ( BTU ) ( FT 2 ∗HR ) 2)
= Thermal diffusivity ( FT ( HR )
( Specific heat ( BTU ) (#∗ DEG ) ) * ( Density (#) ( FT 3 )) * (Temp. gradient ( DEG ) ( FT ))

For mass concrete, it varies in the range of 0.02 to 0.06 ft2/hr. In the absence of meas-
ured data, an average value of 0.04 ft2/hr may be used.

11-3.7 Strengths of Concrete

11-3.7.1 Compressive Strength

The compressive strength of concrete shall be determined in accordance with ASTM 39.

11-3.7.2 Tensile Strength

While there are many methods of determining the tensile strength of concrete, this pa-
rameter has little significance in the performance of an arch dam. Core samples can be
obtained and tested in a variety of ways, but the tensile strength of an arch dam is usually
limited by the ability of horizontal lift joints, vertical contraction joints, and pre-existing
cracks to resist tension. For this reason, the accurate determination of the tensile strength
of the intact concrete may not be necessary.

Often, linear elastic analysis of arch dams will predict tensile stresses, and it can be help-
ful for the analyst and reviewer to have a tensile stress value that serves as a flag for per-
formance evaluation of the dam. This flag has no real physical significance, but it can
alert the analyst or reviewer that predicted tensile stresses are too high, and that the effect
of tensile cracking should be incorporated into the model. For the linear-elastic analysis,
helpful flag values are the apparent tensile strength proposed by Raphael (1984) for the
static and dynamic loading conditions, as shown in Fig. 11-3.2.

11-43
 Fig. 11-3.1 Apparent tensile strength (From Raphael 1984).

Fig. 11-3.2 Design chart for tensile strength (From Raphael 1984).

Several researchers, notably Raphael (1984), have recognized that the measured tensile
strength of concrete should be augmented by a factor for comparison with results of the
linear finite-element analysis. It is evident in the stress-strain curves from laboratory
tests (Fig. 11-3.1) that the stress is not proportional to strain throughout the test. Thus, if
the finite element model assumes a linear stress strain relationship, the apparent tensile
strength which is the linear tensile stress at the failure value of tensile strain is considered
to be more appropriate than the measured tensile strength. The modulus of rupture which
is obtained on the principle of linear behavior provides an experimental method for
measuring the apparent tensile strength. If only splitting tension tests are conducted, ap-
parent tensile strength can be obtained from the measured tensile strength by multiplying

11-44
by 2.3/1.7 = 1.35 (two lowest plots in Fig. 11-3.2). The concept of apparent tensile
strength also applies to seismic loading, which is discussed next.

It should be realized that the apparent tensile strength is not an "allowable" tensile stress
value. If predicted tensile stresses are below this value but extend over large areas of the
dam, the analyst should still suspect the results and re-run the analysis to account for re-
distribution of tensile stresses due to joint and crack opening.

11-3.7.3 Shear Strength

Although arch dams are designed to resist load by compressive arch stresses, shear stress
can be a problem on certain planes within the dam, especially near the foundation. The
simplest criterion for failure for concrete under multiaxial stresses is based on the Mohr-
Coulomb theory. The Mohr-Coulomb diagram shown in Fig. 11-3.3 represents a proce-
dure for determining the failure under combined stress states from which an estimate of
the shear strength can be obtained. In this figure, the point at which the failure envelope
intersects the vertical axis represents the strength of concrete in pure shear, τo. Using this
method the shear strength of the concrete has been found to be approximately 20% of the
uniaxial compressive strength.

Mohr rupture
envelope
Shear

Compression-
tension

Simple uniaxial Simple uniaxial

compression tension

το
g
Compression f e d c b a Tension

Fig. 11-3.3 Typical Mohr rupture diagram for concrete.

(From S. Mindess and J.F. Young, Concrete, 1981)

11-45
11-3.8 Dynamic Material Properties
During earthquake excitation, the rate of loading is much greater than in a short-term
compression test. The strains in a typical concrete dam earthquake response are devel-
oped at frequencies of 2.5 to 25 Hz, which corresponds to times from zero to peak load of
10 to 100 milliseconds. Tests performed at such rapid rates of loading demonstrate that
the dynamic modulus of elasticity is about 25 percent greater than that observed in short
term tests, and this increased modulus should be used in the dynamic response analyses.
Tests performed at loading rates typical of earthquake response, such as those mentioned
above with regard to the dynamic modulus of elasticity, have shown that on average ten-
sile strength is increased by about 50 percent at these high strain rates. Comparison of
the lowest and highest plots in Fig. 11-3.2 shows that the apparent tensile strength of
concrete under seismic loading is twice its splitting tensile strength under short-term
loading. Similarly, the rapid rate of seismic loading increases the compressive strength
but this increase is reported to be 30 percent as opposed to 50 percent for the tensile
strength.

Since this measure of tensile strength has been developed without regard for any specific
weakness in the mass concrete, such as the lift joints, and because the tensile strength
across such joints may be 15 to 20 percent less than in the homogenous material, it would
be judicious to assume the tensile strength for the lift joints is somewhat less than that for
the homogenous concrete. In fact, the actual tensile strength across the poorly con-
structed lift joints of some older dams could be even drastically lower than that for the
homogeneous concrete. Thus it is important that such weaknesses in the mass concrete
are accounted for in the seismic safety evaluation, and that the actual reduced strength at
lift joints is determined by material testing.

11-46
11-4 LOADING

Arch dams are subjected to various loads. Loads can be categorized into 2 basic types,
static and dynamic. Static loads are sustained loads that do not change, or change very
slowly compared to the natural periods of vibration of the structure. A dam’s response to
static loads is governed by its stiffness. Examples of static loads include dead load, hy-
draulic load from normal or flood conditions, forces from flowing water changing direc-
tion, uplift, forces from ice expansion, and internal stresses caused by temperature
changes. Dynamic loads are transitory in nature. They are typically seconds or less in
duration. Because of the speed at which they act, the inertial and damping characteristics
of the dam as well as its stiffness affect the dam's behavior. Examples of dynamic loads
include earthquake-induced forces, blast-induced forces, fluttering nappe forces, or
forces caused by the impact of ice, debris, or boats.

11-4.1 Dead Load

Dead load in arch dams is the weight of the concrete plus appurtenant structures such as
gates, bridges, and outlet works. The unit weight of the concrete is based on the labora-
tory test results of the mix design and/or physical measurements of concrete cores. How-
ever, mass concrete containing natural sand and gravel or crushed-rock aggregates gen-
erally weighs about 150 pounds per cubic foot (pcf). In the absence of measured data,
this unit weight can be assumed for the concrete. Dead load is normally imposed on can-
tilever monoliths prior to the grouting of the contraction joints. This should be taken into
account when analyzing then dam. (See section 11-5.2.2.)

Compared to the dam itself, the weight of appurtenances is typically negligible and may
be ignored in the stress analysis. Massive outlet works and overflow-ogee-weir spill-
ways, however, may have noticeable effects on the static and dynamic stresses and their
weight should be considered.

11-4.2 Hydraulic Loading

11-4.2.1 Normal Water Loads

Normal water loads include hydrostatic pressures on the dam faces resulting from the
reservoir and tailwater during the normal operation of the hydroelectric project. The res-
ervoir levels should correspond to the maximum normal-high-water-level (NHWL),
which is usually the level of the spillway crest for ungated spillways and the top of the
spillway gates for gated spillways. Normal tailwater can be obtained from historical op-
eration records.

11-47
For computation of normal water loads, headwater and tailwater pressures are considered
to vary linearly with depth and to act normal to the dam surfaces. In addition to the dam
surface, headwater pressures should also be applied on the foundation rock at the valley
floor and flanks since reservoir water load causes foundation rock to deform and produce
additional deformations and stresses in the dam. In computation of water loads, a con-
stant unit weight of 62.4 pcf may be assumed for fresh water.

Acting in the opposite direction, tailwater generally produces stresses with opposite signs
of those induced by the headwater. The effect of tailwater, therefore, is to reduce both
tensile and compressive stresses below the tailwater elevations. This effect diminishes
when the tailwater depth is less than 20% of the dam height. For these reasons, it is gen-
erally considered conservative to ignore tailwater loads in an arch dam stress analysis,
and may be omitted for simplicity. If, however, tailwater effects uplift pressure on a fail-
ure plane on which sliding stability is being analyzed, uplift should be considered.

11-4.2.2 Flood Loads

The basic flood loads include hydrostatic pressures on the dam faces resulting from the
reservoir and tailwater elevations which occur during the passage of the inflow design
flood (IDF). Chapters 2 and 8 describe the methods to be used for defining the IDF and
the probable maximum flood.

The water pressures due to flood are also assumed to vary linearly with depth and to act
normal to the dam surfaces and to valley floor and flanks of the foundation rock.

11-4.2.3 Uplift
Uplift or pore water pressures develop when water enters the interstitial spaces within the
body of an arch dam as well as in the foundation joints, cracks, and seams. Under static
loading conditions, the effect of pore water pressure is to reduce normal compressive
stresses within the concrete and to increase the corresponding normal tensile stresses
should they exist (NRC 1990). A computer analysis of these effects on Morrow Point
Dam, a 465-ft-high thin arch dam, and on a thick arch dam of similar height and crest
length, showed a stress change of less than 20 psi -- i.e. about less than 5% of the tensile
strength of the concrete. Because of this minor change in stress, the effects of pore water
pressure on stresses within an arch dam may be ignored in the absence of any cracks.

When the stress results and field conditions for a gravity (thick) arch dam indicates that
tensile cracking will develop at the dam-foundation interface, uplift should be considered
and applied as external loads on both faces of the crack. Uplift does not need to be con-
sidered in the stress analysis for thin arch dams.

11-48
Uplift should always be considered in the sliding stability analysis of the potential failure
planes within the foundation (Section 11-5.4) or along the dam-foundation interface
(Section 11-5.6.1).

When required, distribution of uplift pressure at the contact surface between an arch dam
and the foundation rock can be determined using the field data from piezometer readings
or by performing seepage analysis. In general, the distribution of uplift pressure is influ-
enced by the geological conditions of the foundation rock near the base of the dam, by
location and length of drain, and by the crack length. When the field data and seepage
analysis are not available, uplift on the average can be represented by the conventional
linear or bi-linear approximation described in Chapter 3. When analysis or inspection
indicate cracking, full uplift should be assumed to exist over the entire crack when it is
exposed to the reservoir.

11-4.2.4 Silt Load

Existing arch dams are usually subjected to silt pressure due to sedimentary materials de-
posited in the reservoir over many years. However, the significance of silt pressure as an
additional static load depends on the sediment depth. For U-shaped and broad base arch
dams, sediment depth of less than 1/4 of the dam height produces negligible deformations
(Fig. 11-4.1) and stresses (10 to 15 psi), and thus their effects may be ignored. For V-
shaped dams the effects of silt pressure are even less and may be ignored if the depth of
sediment is less than 1/3 of the dam height (Fig. 11-4.2). Both of these figures also show
that the maximum deformations due to silt loading occur at locations below the silt level
identified by dashed lines.

500
Silt depth as a percentage
450 of dam height

400 50%
DISTANCE FROM BASE (FT)

35%
350
20%
25%
300
Silt Level
250

200

150

100

50

0
-0.01 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
DISPLACEMENT (IN.)

Fig. 11-4.1 Effect of silt on displacement (U shaped valley)

11-49
When silt depth is significant, silt load is treated as an equivalent fluid exerting hy-
drostatically varying pressures on the upstream face of the dam and on the valley floor.
The equivalent fluid density for the saturated silt is assumed to be 85 pounds per cubic
foot (pcf); that is 22.5 pcf in addition to 62.4 pcf for the water.

300
Silt depth as a %
DISTANCE FROM BASE (FT)

of dam height
250
50%
35%
200 25%
20%
150 Silt Level

100

50

0
-0.01 0 0.01 0.02 0.03
DISPLACEMENT (IN.)

Fig. 11-4.2 Effect of silt on displacement (V shaped valley)

11-4.2.5 Ice Load

Ice can produce significant loads against the face of an arch dam. For this reason, ice load
must be considered where reservoir freezing can be expected. Ice loads can be catego-
rized into 2 different types; static loads, produced by the ice in contact with the dam
when the reservoir is completely frozen, and dynamic loads, caused by the impact of
large floating sheets of ice colliding with the dam.

Static load

This type of ice load is caused by the thermal expansion of the ice or by the wind and
current drag. Pressures generated by the thermal expansion depend on the temperature
rise and the ice properties. Wind drag depends on properties of the exposed surface and
on the direction and velocity of the wind.

11-50
The magnitude of ice loading depends on the thickness of the ice cover. When actual
measurements of ice pressure are not available, ice loading may be taken as 5 kips per
square foot along the contact surface with the dam. For example, a 2-foot thick layer of
ice would apply a 10 kip per linear foot load along the axis of the dam. The method of
Monfore and Taylor (1954) can also be used to estimate ice pressure.

The radial distribution of ice pressure is of some concern, especially for thin arch dams.
Arch dam design assumes that loads will be radially uniform. If this is not the case, large
bending stress in the arch direction could result. Radial variation of the ice load could be
caused by un-even heating, differences in thickness, or the absence of ice over part of the
arch due to powerhouse intakes. In addition to the possibility of non-uniform loading,
there is the fact that the ice itself interacts structurally with the dam, complicating the
determination of the arch's response. The applied ice load must be representative of the
site specific conditions.

Ice Impact

Another possible source of ice loading is ice impact. In many northern rivers, large ice
sheets, sometimes weighing many tons, can float down river under the influence of high
spring discharges. The force of these impacts can be roughly calculated by equating the
kinetic energy of the moving ice sheet and the energy dissipated in crushing ice against
the object that it impacts. Refer to the U.S. Army Corps of Engineers EM-1110-2-1612
"Ice Engineering" for additional guidance.

11-4.2.6 Hydraulic Loading of Spillways

The hydraulic loading induced by operation of the spillway is only of concern when the
spillway is located on the dam. Forces produced by discharge are usually not significant
and are typically ignored in the analysis of arch dams. However, if it is determined that
hydrodyanmic forces could effect dam stability, methods for determining spillway pres-
sures are outlined in Chapter 3 of this guideline, and in Corps of Engineers EM 1110-2-
1602.

In rare instances arch dams with crest overflow spillways can be subject to forces pro-
duced by a "fluttering nappe". Nappe flutter is caused by resonance between air trapped
in the cavity between the nappe and the downstream face of the dam. Vibrations induced
by such a fluttering nappe could be of importance to the safety of tall and thin arch dams.
In addition, if flow over the spillway is controlled by a flap or bascule gate, the fluttering

11-51
of the nappe can excite the gate to vibrate; the gate in turn could transfer an important
dynamic load to the top of the dam. The phenomenon, and methods to prevent such vi-
brations, is thoroughly described in ICOLD Bulletin 102, Vibrations of Hydraulic
Equipment for Dams (ICOLD 1996).

11-4.3 Thermal Loading

Temperature loads in arch dams result from the differences between the closure tempera-
ture when construction joints between cantilever monoliths are grouted or filled by con-
crete to bind them together, and the concrete temperatures during the operation of the
dam. The closure or stress-free temperature is a design parameter selected such that to
minimize thermally induced tensile stresses in the dam. In the case of an existing dam,
the actual value(s) of closure temperature can usually be found in the construction or de-
sign records. If such records are not available, it can be assumed to be equal to either the
mean annual concrete temperature or the mean annual air temperature that exists at the
dam site. The concrete temperatures are determined from either computation of heat flow
through the dam due to the air and water temperatures adjacent to the dam surfaces and
exposure to the solar radiation, or from embedded instruments.

The following subsections provide a brief description of procedures and guidance for de-
termining thermal loading for arch dams. Additional details are contained in Chapter 8 of
EM 1110-2-2201 (1994).

11-4.3.1 Temperature Distribution

Variation of temperatures through the dam thickness primarily depends on the thickness
of the dam. For relatively thin arch dams, a linear temperature distribution from the res-
ervoir temperature on the upstream to the air temperature on the downstream face pro-
vides a reasonable approximation. The linear temperature distributions can be obtained
by a simplified method (Section 11-4.3.5) or by the FEM (Section 11-4.3.5). Dams with
relatively thick sections exhibit a nonlinear temperature distribution. In these cases, con-
crete temperatures near the dam faces respond quickly to the air and water temperatures,
whereas temperatures in the center of the section remain near the closure temperature
with minor fluctuations. The nonlinear temperature distributions for thick dams can be
determined using the FEM (Section 11-4.3.5). For dams with reliable embedded tem-
perature monitoring instruments, actual temperature should be used. An example of re-
corded temperatures in the interior of an arch dam is shown in Figure 11-4.3.

11-52
 Fig. 11-4.3 Example of interior temperature distribution

11-4.3.2 Air Temperature

Estimates of air temperatures at a dam site are based on the past air temperatures meas-
ured at the dam site or at nearby locations. If not measured at the dam site, air tempera-
tures near the dam site can be obtained from the US Weather Bureau, which collects
weather data at many stations. Data from the nearby station should be adjusted for the
differences in elevation and latitude that may exist between the station and the dam site.
The actual air temperature data for a period of 5 years or longer is required to assemble a
chart showing various mean temperatures as well as the maximum and minimum re-
corded temperatures (Fig. 11-4.4).

11-53
 Fig. 11-4.4 Hungry Horse Dam – climatic and mean concrete temperatures
(USBR, Townsend 1965)

11-4.3.3 Reservoir Water Temperature

The reservoir water temperatures vary with depth and season. Estimates of temperatures
for the impounded water are obtained by measuring temperatures directly at and below
the water surface at locations near the dam. Such data usually include one recording per
month for several months each year and for a period of several years. When no direct
data are available, the best estimate of the expected reservoir water temperature can be
obtained from water temperatures recorded at nearby lakes and reservoirs of similar depth
and with similar inflow and outflow conditions. These data are then used to estimate a
range of reservoir water temperatures showing the mean annual high and the mean annual
low temperatures for the impounded water (Fig. 11-4.4).

11-54
11-4.3.4 Solar Radiation
Solar radiation on the exposed faces of a dam increases the temperature of the structure.
The solar radiation, therefore, has the net effect of reducing temperature loads for the
winter conditions and increasing for the summer conditions. The mean concrete tem-
peratures discussed in Section 11-4.3.5 should be adjusted for the effect of solar radiation
on the downstream face and on the portion of upstream face not covered by reservoir
water. The amount of temperature rise due to solar radiation depends on the slope and
orientation of the exposed surface as well as the latitude. Knowing the slope and orienta-
tion of a point on the dam and the latitude, the solar temperature rise for that point can be
obtained from a set of charts developed by USBR (Townsend, 1965). Since the sun's
rays strike different parts of an arch dam at varying angles, values of the solar tempera-
ture rise should be evaluated at the quarter points.

11-4.3.5 Concrete Temperatures

The range or amplitudes of concrete temperatures arising from exposure to air and water
can be determined by a simplified method or the finite-element method. In the simplified
method assumed external sinusoidal temperature variations are applied to the edges of a
theoretical flat slab, whereas in FEM they are applied to the faces of a finite-element
model of the dam using a conductive boundary condition.
Simplified Method. Described fully by USBR (Townsend, 1965), this method is based on
computation of heat flow through a flat slab of uniform thickness exposed to sinusoidal
temperature variations on both faces. The method has been simplified by reducing the
heat flow computation into a curve showing the ratio of the variation of the mean tem-
perature of the slab to the variation of the external temperature as a function of an "effec-
tive" slab thickness. The simplified method can be used in the trail load method as well as
the FEM.

Finite Element Method. Amplitudes of concrete temperatures can also be determined

using the finite-element method. The FEM is especially suitable for dams with relatively
thick sections that exhibit a nonlinear temperature distribution through the dam thickness.
Since very little heat is transmitted along the axis of the dam, the finite-element heat-flow
analysis of arch dams may be conducted on the basis of a 2-D model without significant
loss of accuracy. Two-dimensional heat-flow model of an arch dam should be developed
as a vertical section through the crown section with three or more elements through the
thickness. For dams with varying arch thickness along the dam axis, an additional verti-
cal section taken near the quarter point of the dam axis is recommended. Three-
dimensional heat-flow analysis may be performed using the same 3-D model developed
for the static stress analysis, provided that for relatively thick arch dams three or more

11-55
elements are included through the dam thickness. The water and air temperature cycles
discussed previously are applied to the boundaries of the dam model and the foundation
is either subjected to the mean annual air temperature or assumed to be adiabatic. Most
computer programs with heat flow capabilities permit for both steady-state and periodic
analyses. When performing these analyses, the periodic solution using sinusoidal tem-
perature cycles should be employed. The closure temperature or the mean annual air tem-
perature should be used as the initial temperature. The important factor in these analyses
is to let the solution run long enough for the cycle to settle down to a final stable value.
The effects of solar radiation, if important, can be estimated from the USBR charts
(Townsend, 1965) and superimposed to the FE results. The temperature distributions ob-
tained in this manner can then be applied directly to nodal points of the 3-D stress model
(Section 11-5.2.2).

11-4.4 Earthquake Loading

Earthquake loading described in this section is required for arch dams located in seismic
zone 3 and higher. Arch dams in seismic zone 2 may also require analysis for earthquake
loading on a case by case basis.

11-4.4.1 Safety Evaluation Earthquakes and Associated Ground Motions

The safety evaluation earthquake for analysis of existing arch dams requiring earthquake
loading is the maximum credible earthquake (MCE). The MCE is defined in Chapter 12.

11-4.4.2 Response Spectrum Earthquake Input

Site-specific response spectra of earthquake ground motions are required. (Fig. 11-4.5)
The response spectrum must be smoothly enveloped to avoid the possibility of low en-
ergy notches in the response spectrum coinciding with the natural frequencies of the dam.
Response spectra should be developed for both horizontal and vertical ground motions.
Spectra should be developed consistent with the Guidelines contained in Chapter 12, Sec-
tion 3. The spectra should be developed for 5% damping. In addition, relationships or
factors should be provided to obtain response spectra for higher damping ratios (as high
as 10%) if required for the analysis of the dam. These relationships or factors may be
based on a documented site-specific study; alternatively, the relationships presented by
Newmark and Hall (1982) may be used.

11-4.4.3 Acceleration Time History Earthquake Input

Acceleration time histories of ground motions should be developed consistent with the
guidelines contained in Chapter 12, Section 3.3. Acceleration time histories should be

11-56
Fig. 4-5 Comparison of smooth response spectrum with spectrum modified acceleration
time history, 5% damping.

11-57
developed for three components of motion (two horizontal and one vertical). Time histo-
ries may be either (a) recorded or simulated-recorded time histories or (b) response-
spectrum matched time histories (Figs. 11-4.5 and 11-4.6), as described in Chapter 12,
Section 3.3. For recorded or simulated-recorded time histories, three recordings should
be used. Whereas for the response spectrum matched time histories, one set would be
adequate for linear dynamic analysis.

11-4.4.4 Spatial Variation of Ground Motion

Recorded earthquake ground motions at Pacoima Dam during the 1994 Northridge earth-
quake (CDMG, CSMIP, 1994) indicated that the seismic input for arch dams should vary
along the dam foundation interface. However, the recorded abutment motions at Pacoima
Dam included contributions from both the canyon topography and the dam-foundation
interaction. These motions, therefore, are not free-field accelerograms that would have
been recorded if the dam were not there. Except for one recording at an arch dam in
Taiwan, which included free-field motions at locations on the canyon slopes and valley
floor away from the dam, other cases of measured free-field ground motions across the
canyon suitable for analysis of arch dams have not been reported. At present time scar-
city of data prevents a realistic definition of non-uniform free-field motions for arch
dams, even though procedures for handling non-uniform input have been developed. In
view of these difficulties, the use of standard uniform seismic input , while not as realistic
as one may desire, will continue to be acceptable.

11-58
Fig. 11-4.6 Spectrum matched acceleration time histories foe excitation in upstream,
vertical, and cross-stream directions.

11-59
11-4.5 Load Combinations

Arch dams should be evaluated for all appropriate load combinations using the safety
factors prescribed in Section 11-1.4. Depending on their probabilities of occurrence, three
basic loading combinations, Usual, Unusual, and Extreme should be considered. The
usual loading combination considers the effects of all loads that may exist during the
normal operation of the dam. The unusual loading combination refers to the loads acting
on the dam during the flood stage. The extreme loading combination includes any of the
usual loading combinations plus the effects of the Maximum Credible Earthquake de-
scribed in Section 11-4.4. Rare loading conditions which have a remote probability of
occurrence at any given time, have a negligible probability of simultaneous occurrence
and should not be combined. When a very low water level or empty reservoir may be
expected, its effects should be considered by a special loading combination described in
Section 11-4.5.2. The loading combinations to be considered are as follows:

11-4.5.1 Usual Loading Combinations

Two usual loading combinations representative of the summer and winter temperature
conditions should be considered. The reservoir water level is assumed to be at the normal
high water level (NHWL) defined in Section 11-4.2.1 unless the most probable water
level at the time of respective mean concrete temperatures can be established.

a. Summer Condition:
• Maximum mean concrete temperatures
• Normal high water level (NHWL), or the most probable water level occurring at
the time of maximum mean temperature
• Dead load
• Silt load (if applicable)
• Tailwater (if applicable)

b. Winter Condition:
• Minimum mean concrete temperatures
• Normal high water level (NHWL), or the most probable water level occurring at
the time of minimum mean temperature
• Dead load
• Silt load (if applicable)

11-60
 • Ice load (if applicable)
• Tailwater (if applicable)

11-4.5.2 Unusual Loading Combinations (IDF)

Depending on the time of flooding, one or both of the following unusual loading combi-
nations should be considered. The maximum mean concrete temperatures should be used
with the summer flooding and the minimum mean concrete temperatures should be em-
ployed with the winter flooding, unless the most probable mean concrete temperatures at
the time of respective flooding can be established.

a. Summer Flooding:
• Flood water level
• Maximum mean concrete temperatures, or mean concrete temperature occurring at
the time of flood
• Dead load
• Silt load (if applicable)
• Tailwater (if applicable)

b. Winter Flooding:
• Flood water level
• Minimum mean concrete temperatures, or mean concrete temperature occurring at
the time of flood
• Dead load
• Silt load (if applicable)
• Ice impact (if applicable)
• Tailwater (if applicable)

c. Special Loading Combination:

Special loading combinations correspond to the seasonal minimum water level (NLWL)
or a complete reservoir drawdown condition. They are considered as a safeguard against
possible instability conditions due to the reduced or lack of water pressures.
• Minimum (NLWL) or no headwater, whichever applicable
• Most probable mean concrete temperatures at that time
• Dead load
• Silt load (if applicable)

11-61
 • Tailwater (if applicable)

11-4.5.3 Extreme Loading Combinations (MCE)

Extreme loading combinations include any of the usual loading combinations plus the
effects of the maximum credible earthquake.

• Summer usual loading combination + MCE

• Winter usual loading combination + MCE

When more than one MCE ground motions governs, the effects of each MCE should be
combined with each of the usual loading combinations described above.

11-62
11-5 STATIC ANALYSIS

11-5.1 Overview

This section describes analysis and evaluation procedures required for assessing the structural
stability of arch dams and their abutment foundation under static loads. The acceptable methods
of analysis for computing deflections and stresses developed in the dam include three dimen-
sional finite element (FE) and in certain cases continuum solution procedures, as applicable. The
FE stress analysis should be conducted by developing an accurate three-dimensional model of
the dam-foundation system. The manner by which various static loads are applied should be de-
scribed. The results of analyses should be presented appropriately in order to facilitate examina-
tion, interpretation, and evaluation of the findings.

11-5.2 Finite Element Analysis

The finite element procedure is the numerical method most often used for the structural analysis
of arch dams. This guideline assumes that the reader is already familiar with the general theory
of finite element analysis of elastic solids (Zienkiewics, 1971; Bathe and Wilson, 1976). The
following remarks are intended only to point out some special considerations in the application
of this technique to arch dam analysis.

11-5.2.1 Structural Modeling Assumptions

The finite element analysis of arch dams is based on the same assumptions that underlie all finite
element analyses. This being the case, the basic principles that govern element formulation,
mesh construction, and load application are as valid in the analysis of arch dams as they are
anywhere in structural mechanics. There are, however, certain special considerations in the use
of the finite element in arch dam analysis:

1. The body of the dam is typically assumed to be bonded to the foundation rock throughout
its contact with the canyon. However, the validity of this modeling assumption is often
what the analysis is seeking to determine. If this assumption results in excessive shear or
tensile stresses on the foundation contact, this modeling assumption may require modifi-
cation.

2. The dam is typically assumed to be a monolithic structure with linear elastic and isotropic
material properties. In reality, the typical arch dam is divided by construction joints, con-

11-63
 traction joints, and pre-existing cracks. In addition, concrete by its nature is not isotropic
because its compressive strength is typically 10 times its tensile strength.

3. The foundation rock is assumed to be monolithic with linear elastic and isotropic material
properties, when in reality it is jointed with non -linear characteristics. The use of a "de-
formation modulus" instead of the actual Young's modulus is an attempt to deal with the
complex character of the typical foundation.

11-5.2.1.1 Dam Model

The basic geometry data for developing a 3D finite element mesh for the dam can be obtained
from the construction drawings. In some situations, however, it may be necessary to confirm the
accuracy of such data by visual inspection, and possibly by field surveys, to ensure that the ex-
isting conditions of the dam matches the as-built drawings. For example, a severely deteriorated
layer of concrete near the dam surface may have lost its strength, suggesting that a reduced dam
thickness or a reduced effective modulus of elasticity might better represent the actual condi-
tions. In other situations, structural modifications may have increased both stiffness and mass of
the dam. Critical gravity abutment thrust blocks that may exist at one or both ends of an arch
dam should be included in the dam model. Smaller and less important thrust blocks may be con-
sidered as part of the foundation rock, and not modeled separately. The FE model developed for
the dam should closely match the dam geometry and be suitable for application of the various
loads.

Meshing

The type of FE mesh employed is highly dependent on the geometry of the dam and the
ability of the displacement field of the element to capture the displacement and stress
fields that one is attempting to model. For this reason, there is no "magic number" of
elements that constitute a good FE mesh. In general, higher order elements such as 16
node shell and the thick shell (Ghanaat, 1993a), and the general 3D element (Bathe and
Wilson, 1974) (See Fig. 11-5.1) can have relatively coarse meshes. The linear 8 node
solid element requires finer meshing to accommodate the same displacements.

The size of elements sometimes may be dictated by the foundation profile. Highly ir-
regular foundation profiles in general require smaller elements to match the dam geome-
try. In general, as elements get smaller, they become increasingly sensitive to geometric
discontinuities, such as re- entrant corners found at the dam / foundation interface. The
result can be large stress concentrations that are fictitious because of the formation of

11-64
cracks in the foundation material. The size of elements also affects how accurately the
dynamic characteristics, and thus response of the dam to earthquake loading is being
evaluated. Large thin arch dams such as Morrow Point Dam may have numerous vibra-
tion modes (Fig. 11-6.2). To account for the contribution of all significant modes, suffi-
cient number of elements must be included in the finite element mesh. As a general rule,
an FE mesh of an arch dam should include 5 or more element rows along the dam height
with sufficient number of elements along the dam axis.

η η
6 2
9 1
5 ζ
10 6
Y 13
2 X 5
7 12
14
ξ
1
3 4 16
8 11
3
ξ 7
ζ 4
15 8

(a) 8-Node Solid Element (b) 16-Node 3D Shell Element

Z
Y
2
1
5 X 18 9
ζ 9 2 1
13 10 6
8 η
ζ 13 17
6 η
ξ 10 5
16 12
ξ =1 14
3
7 21 ξ
4 14
3 19
16
7 11
12 15
η =-1 11 4

15 20
8
(c) Thick Shell Element
(d) General 3D Element with
Variable No. of Nodes

Fig. 11-5.1 Typical finite elements used in solid modeling.

11-65
 Rf ELEVATION (FT)
7165

7090

7015

6940

6875

6790 Shell Element

6730
6700
8-Node Solid
=ElFixed tNodes

(a) Perspective view of dam-foundation (b) Mesh Layout

Fig. 11-5.2 Dam modeled using single layer of shell elements with circular foundation
mesh of 8 node solids.

ELEVATION (FT)

3715
3660

3570

3480

3390

3300

3210

3120 8-Node Solid Element

8-Node Solid Element 3025

= Fixed Nodes

(b) Mesh Layout

(a) Perspective view of dam-foundation model

Fig. 11-5.3 Dam modeled using 2 layers of 8 node solids with stepped foundation.

11-66
The number of elements along the dam axis are selected such that the elements have as-
pect ratios of less than 2. The Morrow Point Dam model shown in Figure 11-5.2 is an
example of a fine mesh for the quadratic shell elements. The Glen Canyon Dam model in
Fig. 11-5.3, shows an example of a fine mesh for the 8-node solid elements.

11-5.2.1.2 Foundation Model

An appropriate volume of the foundation rock should be included in the dam model to
account for the effects of foundation flexibility on the static deflections and stresses of
the dam. The foundation model should extend to a large enough distance beyond which
its effects on deflections and stresses of the dam become negligible. Although finite ele-
ment mesh for the foundation rock can be developed to match the site topography, such
an elaborate model is not required in practice. Instead, a prismatic foundation mesh con-
structed on semi-circular planes (Fig. 11-5.2), or developed by simple projection along
the global coordinate axes (Fig. 11-5.3) may be employed. Typically, smaller elements
are employed near the dam-foundation contact region where the largest deformations and
stresses occur, whereas larger elements are used away from the dam, where the interac-
tion with the dam is reduced.

The size of the foundation model should be determined based on the ratio of the founda-
tion deformation modulus to the concrete modulus of elasticity (Ef /Ec). For a competent
foundation rock with Ef /Ec equal or greater than 1, a foundation mesh extending one
dam height in the upstream, downstream, and downward directions should suffice. For a
very flexible foundation rock with Ef /Ec in the range of ½ to ¼ , the foundation model
should extend twice the dam height in all directions and include more number of ele-
ments.

11-5.2.2 Application of Loads

In FE analysis, internal and external static loads are computed for each individual finite
element and are applied as equivalent forces at nodal points of the dam model.

Dead Load

As described in Section 11-4.1, dead loads in arch dams are the weight of the concrete
plus appurtenant structures. The application of dead load should consider the manner in
which the dam was constructed. For example, arch dams are often constructed as inde-
pendent cantilever blocks separated by vertical joints. Since these joints are not capable

11-67
of transferring dead load horizontally until they are grouted, dead loads should be applied
to individual cantilevers to simulate this condition. This may be accomplished by per-
forming dead load analysis in two steps. First, dead loads are applied to alternate cantile-
vers (Set-1) by assuming a zero modulus of elasticity for the remaining cantilevers
(Set-2). In the second analysis, the modulus of elasticity is switched on for the Set- 2
cantilevers, and set to zero for the Set-1. If the dead load is applied to the dam all at
once, without taking into account the fact that horizontal load transfer can not occur be-
fore the dam is complete, fictitious stresses will be indicated.

The sequence of construction for some dams may significantly differ from the independ-
ent cantilever blocks. It may involve staged construction and sequence of grouting and
partial reservoir filling that can have a major effect on the distribution of dead loads. In
these situations the alternate cantilever loading discussed above is not appropriate, in-
stead the actual sequence of construction should be considered in the application of dead
load.

Water Load

Water loads due to hydrostatic pressures of the normal water level or flood, are external
forces acting on the surfaces of all finite elements in contact with the impounded water.
These include dam elements having faces coincident with the upstream face of the dam
and foundations having surfaces at the reservoir floor and sides.

Fig. 11-5.4 Tributary area and consistent lumping of nodal forces due to a uniform sur-
face pressure acting on linear and quadratic 3D elements.

11-68
Water loads are lumped at discrete nodal points as equivalent nodal forces. They are ap-
plied to the monolithic arch structure after the contraction joints are grouted. The hydro-
static pressures are lumped into equivalent nodal forces using the "consistent" or ap-
proximate tributary area lumping process. For linear elements, the equivalent nodal
forces computed using either the tributary area or consistent formulation will be the same,
as shown in Fig. 11-5.4. For higher order elements, however, the consistent nodal forces
completely differ from those obtained on the basis of the tributary area. As shown in the
lower graphs of Fig. 11-5.4, the consistent nodal forces are even negative for the corner
nodes, a fact not so obvious. It should be noted that the tributary area lumping of nodal
forces would converge to accurate results for a very fine mesh. However, since the use of
higher-order elements implies a relatively coarse mesh, this kind of error can be very sig-
nificant and should be avoided.

Temperature Load

Temperature loads arise from the differences between the closure (stress free condition)
temperature and concrete temperatures expected during the operation of the dam. Tem-
perature loading should be applied in accordance with section 11-4.3

Silt Load

Similar to the impounded water, silt load is applied to the monolithic structure as an
equivalent fluid exerting hydrostatically varying pressures on the upstream face of the
dam and on the valley floor.

Ice Load

Ice loads are applied radially (normal to the upstream face) to the monolithic structure.
Ice loads in the form of distributed surface pressures may be applied on the upstream
faces of a horizontal layer of elements included in the model at the ice level. Loading
should be in conformance with the guidance given in section 11-4.2.5

Uplift load

Ideally uplift in FE analysis should be introduced as pore water pressures at element

nodes throughout out the dam-foundation model, if the program used has this capability.
In the absence of such capability, uplift pressures at the base of thick arch dams may ap-
proximately be applied as distributed pressures or equivalent nodal forces on the faces of
a thin pervious layer placed between the dam and the foundation rock (Fig. 11-5.5). In

11-69
this approach the pressures or nodal forces due to uplift are transferred to the dam and to
the foundation one row of elements away from the interface, so that the equal magnitude
uplift pressures along the uncracked portion of the interface are not canceled out. The
resulting stress output from these elements will include the effect of uplift pressure. This
procedure can also be used to model uplift in an abutment rock joint.

Fig. 11-5.5 Procedure for application of uplift.

11-5.2.3 Presentation of Results

The basic results of a finite element analysis include nodal displacements and element
stresses. As a minimum, nodal displacements and surface stresses should be presented for
the static loading combinations described in Section 11-4.5 in clear graphical form. Sur-
face stresses should be presented in the local arch and cantilever directions. Additionally,
since nodal loads can be obtained from finite element analyses on an element by element
basis, dam thrust needed for the rock wedge stability analysis (Section 11-5.4) can be
determined from loads acting on elements having a common surface or a common edge
with the dam/foundation contact surface.

Although displacements are not directly used in evaluation of dam safety, they provide a
visual means by which acceptability of the analysis results can be assessed. Nodal dis-
placements may be displayed as simple deflected shapes across selected arch and cantile-
ver sections or presented in the form of 3D plots for the entire dam structure. Fig. 11-5.6
is an example of the static deflections displayed across the crest and crown cantilever

11-70
sections. As expected, hydrostatic loads push the dam in the downstream direction; con-
traction due to temperature drop below the grouting temperature produces downstream
deflections; and expansion of the concrete due to temperature rise above the grouting
temperature results in the upstream deflections of the dam.

-400 -300 -200 -100 0 100 200 300 400

(a) Crest Deflection

500

Hydrostatic
Linear Temp. Drop
400 Uniform Temp. Drop
Uniform Temp. Rise
Linear Temp. Rise
Crown Section
300 Mid-surface Line

200

100

0
(b) Crown Deflection

Fig. 11-5.6 Static deflection due to various load conditions.

Maximum tensile and compressive stresses in an arch dam usually occur at the faces of
the dam, therefore evaluation of stresses on the faces of the dam is required. The surface

11-71
stresses resolved into arch and cantilever stresses are usually presented in the form of
stress contours on each face of the dam, while surface principal stresses are displayed in
the form of vector plots, as illustrated in Fig. 11-5.7.

+ = Tension - = Compression

Fig. 11-5.7 Arch and cantilever stress contour and principle stress vector plots.

In addition to the arch and cantilever stresses, the magnitudes of the shear stresses caused
by the bending and twisting moments should be examined, especially for very thin arch
dams and those with cracked sections. These include radial cantilever shear stresses act-
ing radially on a horizontal plane and radial arch shear stresses acting radially on a verti-
cal plane. Also, excessive tangential shear stress acting on the foundation can be a cause
for concern. (See 11-5.6.1)

11-72
11-5.2.4 Evaluation of Stress Results

Evaluation of the computed stresses should begin with validation of numerical results by
careful examination of the deflected shapes and stress distributions due to individual
loads. Such data should be inspected for unusual deflected shapes, exceptional high or
low displacement and stress magnitudes, and unexpected stress distributions that differ
significantly from the results of other arch dams and cannot be explained by intuition.
Force equilibrium should also be verified by comparing the sum of reaction forces to the
sum of applied loads. Problems usually arise from the input data and modeling errors and
should be corrected. After the numerical results have been validated for accuracy, the
stress results are used to evaluate the dam performance for each static loading combina-
tion in accordance with the criteria set forth in Section 11-1.4.

A concrete arch dam under static loading conditions is considered to be safe from over
stressing failure if the allowable stresses are not exceeded in any extensive area. Allow-
able stresses of concrete are obtained by dividing the strength capacities (Section 11-3.7)
by the appropriate safety factors given in Table 11-1.1. This requirement is easily satis-
fied for a well designed arch dam which resists the loads by developing essentially com-
pressive stresses with very little tension (Fig. 11-5.7). In other cases compressive stresses
usually meet the criteria but tensile stresses caused by temperature loads, or other unfa-
vorable situations may be significant. When significant tensile stresses are indicated,
sections of the arches and cantilevers subjected to excessive tension are assumed to be
cracked. This cracking will result in the re-distribution of stresses and loads. For exam-
ple, localized loss of cantilever action caused by cracking at the base of the dam can be
compensated by increased arch action. If cracking appears to be significant, non-linear
analysis or linear analysis of the "as cracked" model may be required.

In general, areas of high indicated tensile stress should correspond to observable cracks in
the dam, however this is not always the case. It is not unheard of to observe cracking in
the dam in an area which the FEM model indicates is in compression. This may indicate
an error in the FEM model. It may also be a result of some past loading condition, such
as high thermal loads during curing. If the crack is not an indication of a modeling error,
it is not necessarily important to determine the cause of the crack if it can be shown that
the dam is stable without requiring tensile strength across the crack.

Judgement is required in deciding when tensions indicated by a linear elastic model are
in-significant enough to be overlooked, and when re-analysis must be done. Tensile
stresses on the upstream face normal to the foundation are often indicated. If tensile
normal stresses at the base of the dam are confined to a small region of the dam surface

11-73
area, and all other stresses are low and well within the allowable values, then it can be
assumed any cracks formed under this condition would be shallow cracks limited to the
dam-foundation interface with no serious consequences. However, if high tensile stresses
persist over a large area, then an analysis considering the effects of cracks is required to
demonstrate that the stress redistribution, as described above, would take place safely.

As cracks form at the base of the dam, the modulus of elasticity for the elements in the
cracked portion of the interface must be reduced in the finite element method so that the
indicated tensile stresses can be effectively eliminated. If a crack occurs at the upstream
face of a cantilever below the water surface, the uplift pressure as described in Section
11-4.2.3 should be considered in the analysis.

Shear stresses in excess of the shear capacity of the concrete/foundation interface can
also be of concern. Massive shear failures of the type treated in section 11-5.6 are rare,
however local shear over stress is more common. Situations of shear over stress should
be evaluated in a manner similar to indicated tensile overstress. If stress can be relieved
by re-distribution without overstressing other areas, local shear failure can be tolerated.
Elements may have to be allowed to "slide" by reducing their elastic modulus or dis-
connecting them. Many FE programs also have non-linear gap friction elements which
can be used.

11-5.3 Alternative Continuum Models

The finite element method outlined in the previous sections may not be required if the
dam has a simple geometry and its performance only due to the static loads is to be
evaluated. In these cases, if the dam geometry can be represented by one-, two-, or three-
centered layouts and if uniform material properties can be assumed for the concrete and
for the foundation-rock, the dam may be analyzed using the trial load method.

A complete description of the trial load method and its computerized version known as
Arch Dam Stress Analysis System (ADSAS) is given by the USBR (1977). Following iss
a brief overview of the method and the conditions under which it can be applied to analy-
sis of existing arch dams.

11-5.3.1 Trial Load Method

The trial load method is based on the assumption that an arch dam is made of two sys-
tems of structural members: horizontal arch units and vertical beams or cantilever units

11-74
(Fig. 11-5.8); that the waterload is divided between the arch and cantilever units in such a
way that the resulting arch and cantilever deflections and rotations at any point in the
dam are equal (Fig. 11-5.9). The preceding agreement is accomplished by subjecting arch
and cantilever units to a succession of self-balancing trial-load patterns and solving the
simultaneous equations involved. The solution is normally obtained by computers using
a trial load program such as ADSAS developed by the US Bureau of Reclamation. The
resulting load distributions required to achieve geometric continuity are then used to
compute stresses in the dam.

(a) PLAN

CANTILEVER UNIT ARCH UNIT

(b) ELEVATION PROFILE

Fig. 11-5.8 Arch and cantilever units in Trial Load Method

A1
A 2

ΘV

(a) DEFLECTED ARCH

A 1

A
ΘH

A2

(b) DEFLECTED CANTILEVER

Fig. 11-5.9 Translations and rotations of arch and cantilever units.

Progressing from the simplest to the most comprehensive, a trial load analysis may con-
sist of crown-cantilever adjustment, radial deflection adjustment, or the complete adjust-
ment, which includes adjustments for the radial and tangential translations as well as ro-

11-75
tations. The crown-cantilever and radial deflection analyses are usually used for the pre-
liminary and feasibility studies of new dams. For safety evaluation of existing arch dams
only the complete trial load analysis should be attempted.

Many comparisons with measurements from actual dams and scale models as well as
with 3D fintite element analyses have shown that ADSAS gives reliable results for one-,
two- or three-centered dam layouts, subjected to standard static loads. It has been used
successfully in the design of new dams over many decades, but its use in the evaluation
of existing dams is limited to the geometry configurations just described and to static
loading only. Complex geometry and material property variation, the effects of openings
within the body of the dam, and nonradial abutments cannot be analyzed by ADSAS.
Unlike FEM, ADSAS does not permit analysis of the effects of rapid changes in the dam
geometry where detailed stress information may be required. Analysis of the effects of
unusual loads, special boundary conditions, and seismic loading is not possible.

The use of trial load method and ADSAS should be limited to the geometry configura-
tions described above so far as the computed static stresses are not excessive. Existing
dams located in seismic regions requiring dynamic analysis should be evaluated by the
finite element method.

11-5.3.2 Other Methods

Other mathematical formulations and approaches can also be applied to the analysis of
arch dams. Finite difference solutions and global variational energy methods, etc. may
all be acceptable provided force equilibrium is satisfied and realistic constitutive relation-
ships are enforced.

Small single-centered (circular) arch dams located in seismic Zone 1, where seismic
loading does not control, may be analyzed using the cylinder theory. In the cylinder-
theory analysis of arch dams, the stresses at each elevation are assumed to be the same as
in a cylinder of equal outside radius. That is, water loads are resisted entirely by individ-
ual arch sections acting independently. If average stresses from this approximate solution
provide large allowance for stress uncertainties and meet the criteria, then no further so-
phisticated analysis will be required.

11-5.4 Rock Wedge Stability

Stability of the abutments of an arch dam is crucial to the safety of the dam. The abut-
ments are required to resist the majority of the thrust imposed upon the dam by the im-
pounded reservoir. Design of a new arch dam or inspection of an existing arch dam must

11-76
incorporate a careful evaluation of the stability of the abutments. Since rock failures al-
most always occur along preexisting discontinuities in the rock mass, abutment stability
analysis must focus on an evaluation of any wedges of the abutment foundation which
could fail under the loads applied naturally and by the dam and reservoir. As described
in Section 11-8.2.1, the first failure of a thin arch dam occurred in 1959 with Malpasset
Dam in France and resulted from the displacement of a large wedge of rock in the left
abutment. Extensive post failure investigations and analysis led to the conclusion that the
wedge was formed by the interaction of a downstream fault with the gneissic foliation in
the rock and that it was displaced by arch thrust assisted by abnormally high seepage up-
lift forces. The primary lesson learned from this experience was the importance of per-
forming a careful geologic investigation of the abutments of an arch dam followed by
stability analysis of any kinematically capable wedges of rock identified.

11-5.4.1 Identification of Kinematically Capable Potential Failure Planes and

Wedges

The first step in an abutment analysis is the identification of any wedges of rock that have
the possibility of sliding under the loads expected to develop during construction and op-
eration of the dam. This is initiated by a geologic investigation of the abutments de-
signed to identify those discontinuity patterns that could contribute to the development of
kinematically capable failure wedges. Discontinuities include such features as joints,
faults, shears, foliation, schistocity, bedding planes, clay seams, coal beds, shale partings
and any other planar weaknesses that may be present in the rock mass.

This is accomplished by first mapping the discontinuity pattern existing in each abut-
ment. Some of the structural geologic features are mapped individually such as faults,
shears, clay seams, and coal beds while others which are more ubiquitous such as joints
are mapped as sets and are assumed to be capable of occurring anywhere in the rock
mass. Bedding, schistocity, and foliation may be ubiquitous, but their orientation may
vary from one part of an abutment to another due to folding and this must be identified
during the mapping.

The joint system in a rock mass normally will consist of from two to six individual joint
sets. A joint set is defined as a series of essentially parallel planar fractures which have
not experienced translational movement. Joint sets are usually identified by plotting the
results of a large number of field joint orientation measurements on a stereonet and con-
touring the density of the pole positions to locate the orientation of the largest concentra-
tions. Detailed instructions for performing a joint system analysis are contained in most

11-77
structural geology texts, such as Billings (1954). Software programs are available which
facilitate stereographic analysis by use of computers.

The geological investigation then uses the assembled data to locate any and all kinemati-
cally capable wedges of rock. This means the location of any rock mass which can fail
by sliding on one or more preexisting fractures with out the necessity of shearing through
a large mass of intact rock. The intact shear strength of most rock is sufficient to prevent
sliding failure except on daylighting discontinuities. Such a daylighting discontinuity
may be a single plane subparallel to a slope with a slightly lower dip angle than the slope,
or it may be a combination of two or more intersecting discontinuities whose trend of in-
tersection daylights on the slope at an angle of dip slightly lower than that of the slope.

11-5.4.2 Analysis by Stereographic Projection Procedures

The step by step procedures for performing a graphical slope stability analysis by use of
stereonets is presented in Chapter Four of Hendron, Cording and Aiyer (1971) or Chap-
ters 7 and 8 of Hoek and Bray (1981).

11-5.4.3 Vector Analysis

Vector analysis procedures have been developed for three dimensional analysis of po-
tentially unstable wedges of rock in both natural and cut slopes. The failure of Malpasset
Dam in 1959 provided much motivation to develop this procedure. Early researchers
were Wittke and Londe. More recent developments have been documented by Hendron,
Cording, and Aiyer (1971), and Hoek and Bray (1981). Procedures for performing this
analysis are outlined and explained in great detail in each of these last two references.
These publications are recommended as acceptable guides for performing the analysis.
Computer programs have also been developed for performing the rigid block three-
dimensional vectorial analysis. The Bureau of Reclamation publication by Scott and Von
Thun (1993) describes one such program RIGID, which performs rigid block limit equi-
librium analyses utilizing three dimensional vector procedures. Hoek, Carvalho, &
Kochen (1995) have developed a rock wedge stability software program called
SWEDGE.

11-78
Fig. 11-5.10 Three dimensional representation of an arch dam acting on an abutment
showing the planes of three discontinuities forming a rock wedge or tetra-
hedron and the vectors of forces acting upon the rock.

A vector representation of a tetrahedral rock wedge with the kinematic capability of fail-
ing under loads imposed by its own weight and the thrust and water pressure forces im-
posed by an arch dam is illustrated in Fig. 11-5.10 after Londe (1993). This illustration
shows a three dimensional picture of the left abutment of an arch dam underlain by a
large rock wedge formed by the intersection of three rock discontinuities or fractures.
The vectors representing the water pressure on each of the fractures, the weight of the
rock mass, and the thrust of the dam are shown. Forces resisting failure of the wedge of
rock may include both friction and cohesion. The analytical procedures described in the
above references provide for inclusion of cohesion in addition to friction where it is ap-
propriate.

11-5.4.4 Loads to be Considered

Refer to section 11-4.5, Load Combinations, for detailed discussion of the loading com-
binations to be evaluated. These include summer and winter conditions for each of the
following loading combinations:

11-79
 1. Usual (normal operating conditions)

2. Unusual (flood condition)

3. Extreme (seismic)

The loads that must be considered in the analysis of abutment rock wedges include the
weight of the rock wedge, thrust from the dam, uplift from hydrostatic forces applied on
each plane which defines the wedge, and dynamic forces applied by the design earth-
quakes. The weight of the rock wedge is obtained by calculating the volume of the
wedge times the unit weight of rock. The thrust from the dam is obtained from the
structural analysis either by direct computation of nodal forces acting on the dam-
foundation interface or from integration of element stresses in contact with the founda-
tion. In determining the dam thrust, the effects of load redistribution due to possible joint
opening and tension cracks should be considered. The following assumptions are made
in calculating uplift on the various wedge defining planes in the absence of actual pie-
zometric measurements:

(a) Fractures are open over 100 percent of the wedge area and are completely hydrau-
lically connected to the surface.

(b) Head varies linearly from maximum at backplane to zero at daylight.

(c) Planes acted upon directly by the reservoir receive full hydrostatic force.

Dynamic loads are obtained from the design earthquake studies that specify the ground
motions anticipated from the Maximum Credible Earthquake (MCE). Approximate dy-
namic loads may be incorporated into stability analysis of individual rock wedges by use
of acceleration values represented by an equivalent static force as described by Hoek and
Bray (1981) or by a seismic coefficient as described by Hendron, Cording, and Aiyer
(1971) for the Extreme Case of load combinations.

Rock wedge stability can also be modeled directly by including element layers aligned
along shear planes in the foundation. Refer to Sections 11-4.4 and 11-6 for further guid-
ance on earthquake loading and finite element analysis.

11-80
11-5.4.5 Appropriate Factors of Safety

There is considerable debate among arch dam design professionals with regard to how
safety factors should be applied. For instance, Londe (1993) discusses the safety factor in
the light of applying a unique safety factor individually to each different property in the
analysis depending upon the degree of uncertainty associated with that material rather
than a single safety factor applied to the final computation. He also develops the case for
the use of the probabilistic approach rather than the deterministic approach required by
this guideline.

The safety factors to be used on projects under the jurisdiction of the FERC are based on
a deterministic approach and the safety factors are applied to the final computation rather
than to individual properties in the stability analysis. The assumption is made that the
strength parameters and uplift forces assumed for the analysis will be conservatively se-
lected. They must be based upon a comprehensive field investigation and testing pro-
gram as described in Section 11-2, Foundation Considerations, which provides a high de-
gree of confidence in the definition of geologic conditions and shear strength parameters.
In the absence of such detailed foundation knowledge, much more conservative safety
factors must be employed. These will be established for individual projects in consulta-
tion with and as approved by the FERC.

Following are the safety factors to be used where foundation conditions are well defined:

(a) Worst Static Case -------------------- FS = 1.5

(b) Extreme (Seismic) -------------------- FS = 1.1

It is important to understand that meeting an acceptable safety factor, as part of a stability

analysis does not necessarily assure a safe structure. Assurance only comes when sound
engineering and geologic judgement are applied at every step of the investigation, analy-
sis and review to preclude the existence of undiscovered conditions which could defeat
the stability analysis. An example again was the failure of Malpasset Dam in 1959 which
was designed by one of the world’s preeminent arch dam designers.

11-81
11-5.5 Parameter Sensitivity

11-5.5.1 Effects of foundation modulus on dam stresses

Considering that an arch dam resists a large portion of water pressures and other loads by
transmitting them through arch action to the abutments, foundation deformations caused
by the dam loads are likely to significantly affect stress distributions within the dam.
Proper evaluation of stresses within the dam, therefore, requires an adequate determina-
tion of the foundation deformation characteristics. As described in Section 11-2, deform-
ability of the foundation rock is characterized by the modulus of deformation. Depending
on the type of material and discontinuities present in a foundation, values of the modulus
of deformation may vary significantly from abutment to abutment, or with the elevation
(depth). It is also possible that the differences are small or rock masses are of such a high
quality that a uniform average modulus of deformation can be assumed for the entire
foundation contact region.

A comprehensive field investigation to provide extensive definition of the variation of

modulus of deformation is costly and may not be necessary, for example when the de-
formation modulus of foundation is much higher than the modulus of the concrete. In-
stead, the minimum field investigations and laboratory tests described in Section 11-2
should be supplemented by parameter sensitivity studies to account for the uncertainties
and extrapolation to untested areas when necessary. The results of a parametric study of
Morrow Point Dam in Figs. 11-5.12 to 11-5.22, are provided to demonstrate the relative
importance of the foundation modulus on the dam response. The cases presented include:

a) Er / Ec > 1
b) Er / Ec < 1
c) Variable foundation modulus with abutments
d) Variable foundation modulus with elevation

where Er is an effective modulus of deformation for the foundation and Ec is modulus of

elasticity of the mass concrete. The results presented are for water pressures only, ap-
plied to upstream face of the dam as well as to the floor and flanks of the flexible foun-
dation. Hydrostatic deflections along the crown section for various uniform rock to con-
crete modulus ratios (Cases a and b) are displayed in Fig. 11-5.11, whereas the variation
of crest and heel deflections as a function of Er / Ec are given in Fig. 11-5.12. These
graphs demonstrate that the dam deflections are more sensitive to modulus ratios of less
than 1, and that they especially increase dramatically when Er / Ec becomes less than 0.5.
While dam deflections for stiffer foundation rock (Er / Ec > 1) change only slightly from

11-82
those for a rigid foundation. The effects of foundation flexibility on the dam response
essentially diminish for Er / Ec > 2.
500
Er/Ec = Rigid 2 1 1/2 1/4
450

400
DISTANCE FROM BASE (FT)

350

300

250

200

150

100

50

0
0.0 0.5 1.0 1.5 2.0 2.5
DISPLACEMENT (IN.)

Fig. 11-5.11 Hydrostatic deflections of dam crown section for Er / Ec = rigid, 2, 1, 1/2,
and 1/4.

3.5

2.5
DISPLACEMENT (IN.)

1.5

DAM CREST
1

0.5

DAM BASE
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Er/Ec

Fig. 11-5.12 Hydrostatic deflections of dam crest and dam base as a function of Er / Ec.

Similarly, the values of arch and cantilever stresses are more sensitive to modulus ratios
of less than 1, and only slightly differ from those for the rigid foundation when Er / Ec
exceeds 2 (Figs. 11-5.13 and 11-5.14). For Morrow Point Dam subjected to hydrostatic
loading, foundation flexibility increases arch stresses mostly in the central part of the up-

11-83
stream and on the lower part of the downstream face of the dam (Fig. 11-5.14), while
cantilever stresses are primarily increased in the lower 1/3 portion of the dam, as shown
in Fig. 11-5.14.
450

Er/Ec value
400

Rigid
350
DISTANCE FROM BASE (FT)

Er/Ec = 2
300 Er/Ec = 1
Er/Ec = 1/2
250
Er/Ec = 1/4
UPSTREAM
200

150
DOWNSTREAM
100

50

0
-1000 -800 -600 -400 -200 0 200 400 600
ARCH STRESS (PSI)

Fig. 11-5.13 Hydrostatic arch stresses at dam crown section for Er / Ec = rigid, 2, 1, 1/2,
and 1/4.
450
Er/Ec values
400
Rigid
350 Er/Ec = 2
DISTANCE FROM BASE (FT)

300 Er/Ec = 1

Er/Ec = 1/2
250
Er/Ec = 1/4
200

150
UPSTREAM DOWNSTREAM
100

50

0
-800 -600 -400 -200 0 200 400 600 800
CANTILEVER STRESS (PSI)

Fig. 11-5.14 Hydrostatic cantilever stresses at dam crown section for Erf/Ec = rigid, 2, 1,
1/2, and 1/4.

11-84
Variable Foundation Modulus with Abutments. Variation of foundation modulus with
abutment tends to skew dam deflections and stress distributions towards the weaker
abutment, increase stresses within the body of the dam, and cause stress concentrations
near the base of the dam (Figs. 11-5.16 and 11-5.17). The amount of distortion in de-
flections and stress distributions and increase in stress values depends on the modulus
ratios between the abutments and between the abutments and the concrete. Softer foun-
dation rock with greater differences between the abutment moduli, produce larger distor-
tions and higher stresses. Whereas compared to the uniform foundation with Er / Ec = 1,
stiffer foundation rock with both abutment moduli higher than that of the concrete, cause
negligible distortions with slightly more stress concentrations, even if the differences be-
tween the abutment moduli are significant.
500

400
Undeformed
Deformed
300

200

Erl/Ec = 1 Err/Ec = 1/4

100

-100
-400 -300 -200 -100 0 100 200 300 400

200
Undeformed
Deformed
100

Erl/Ec = 1 Err/Ec = 1/4

0

-100
-400 -300 -200 -100 0 100 200 300 400

Fig. 11-5.15 Hydrostatic deflections of Dam/foundation contact for Err/Erl = ¼.

Sensitivity analyses of Morrow Point Dam with assumed foundation-to-concrete modulus

ratios of 1 for the left abutment and 1, 1/2, and 1/4 for the right abutment clearly demon-
strates the behavior discussed above, as shown in Figs. 11-5.16 to 11-5.18. The results

11-85
indicate that while hydrostatic cantilever stresses increase on the weaker abutment side
(right), their values on the left side only slightly differ from those obtained for the uni-
form foundation (Figs. 11-5.17 and 11-5.18). Cantilever stresses also increase at the base
of the dam due to abrupt change of the foundation modulus at this location, with the
stress concentration shifted toward the stiffer abutment side.

Erl/Ec = 1 Erl/Ec = 1
Err/Ec = 1 Err/Ec = 1/4

Fig. 11-5.16 Downstream hydrostatic cantilever stresses for uniform and varying foun-
dation moduli with abutments.

450

400
U/S D/S
350
DISTANCE FROM BASE (FT)

1/2
300 1/4 1/2 1
1/4
1 1/2 1/4
250

200

150

100
Right 1/4-span
50
Left 1/4-span
0
-150 -100 -50 0 50 100 150
CANTILEVER STRESS (PSI)

Fig. 11-5.17 Hydrostatic cantilever stresses at 1/4-span locations for Erl/Ec = 1 and
Err/Ec= 1, 1/2, and 1/4.

11-86
 450

400
Right 1/4-span
350
DISTANCE FROM BASE (FT)

Left 1/4-span
300

250

200

150
.
1/4 1/2 1 1/2 1/4 1/4 1/2 1 1/2 1/4
100

50
DOWNSTREAM UPSTREAM
0
-800 -700 -600 -500 -400 -300 -200 -100 0 100
ARCH STRESS (PSI)

Fig. 11-5.18 Hydrostatic arch stresses at 1/4-span locations for Erl/Ec = 1 and Err/Ec =
1, 1/2, and 1/4.

Hydrostatic arch stresses for the case of uniform foundation modulus are generally com-
pressive in the entire dam, except for minor tensile stresses near the base. Compared to
the uniform foundation case, compressive arch stresses in the weaker abutment side tend
to increase on the upstream and decrease on the downstream face of the dam (Fig. 11-
5.18). The variation of foundation modulus with abutment also increases tensile arch
stresses near the base of the dam, with larger values on the upstream face concentrating
on the stiffer abutment side and those of the downstream face on the weaker abutment
side.

The effects of variation of foundation modulus with abutment are most significant when
the modulus of the weaker abutment is substantially (factor of 2 or more) less than the
modulus of the concrete, and should be considered in the analysis. As the modulus of the
weaker abutment approaches the modulus of the concrete, its effects on the dam response
diminishes and may be ignored. Variation of the foundation modulus with abutment may
also be ignored when foundation moduli for both abutments are greater than that for the
concrete. In these situations, if the variation of foundation modulus is ignored, the
modulus of the weaker abutment should be applied to the entire foundation.

Variable foundation modulus with elevation. At some dam sites, the foundation modulus
may vary significantly with the elevation. A situation in which the modulus of the highly

11-87
weathered and fractured upper abutment regions could be much lower than that of the
lower foundation regions. The effects of weaker upper abutments on the dam response
are most significant when they extend to large depths and their moduli are substantially
less than the modulus of the concrete. Sensitivity analyses of Morrow Point Dam with
the foundation rock overlain by one or two weaker zones, indicate that the variation of
foundation modulus with elevation generally increases dam deflections and stresses.
Figs. 11-5.19 to 11-5.21 show that, the larger the weaker abutment regions and the
smaller their moduli, the greater would be the dam deflections and stresses. The increase
in deflections and stresses for this case, however, is much less than those cases where one
abutment or the entire foundation consist of low modulus materials as discussed previ-
ously. Nevertheless, the effects of variable modulus with elevation should be considered
when the area the weaker zone is substantial and its modulus is less than the modulus of
concrete. In situations where the moduli of the entire foundation including the weaker
zones are higher than the modulus of the concrete, the effect of variation of modulus with
abutment may be ignored and the smallest modulus employed for the entire foundation.

500
Er/Ec values
450
Upper-half =1; Lower-half & Base = 2

400
Uniform foundation; Er/Ec =1
DISTANCE FROM BASE (FT)

350
Upper-half =1/2; Lower-half & base =1

300
Upper-third =1/4; Middle-half =1/2; Base & lower =1

250
Upper-half =1/4; Lower-half =1/2; Base & lower =1

200

150

100

50

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
DISPLACEMENT (IN.)

Fig. 11-5.19 Crown-section hydrostatic deflections for varying foundation

modulus with elevation.

11-88
 450

400

350 Er/Ec values

UPSTREAM
DISTANCE FROM BASE (FT)

DOWNSTREAM

300 Upper-half =1/4; Lower-half =1/2; Base =1

Upper-third =1/4; Middle-half =1/2; Base =1

250
Upper-half =1/2; Lower-half and base =1
200
Uniform Er/Ec =1
150
Upper-half = 1; Lower-half =2

100

50

0
-1000 -800 -600 -400 -200 0 200
ARCH STRESS (PSI)

Fig. 11-5.20 Crown-section hydrostatic arch stresses for varying foundation modulus
with elevation.

450
Er/Ec values
400 Upper-half = 1/4; Lower-half = 1/2; Base = 1

Upper-third = 1/4; Middle-half = 1/2; Lower and

350
base = 1
DISTANCE FROM BASE (FT)

Upper-half = 1/2; Lower-half = 1

300
Uniform foundation Er/Ec =1

250
Upper-half = 1; Lower-half = 2

200

150 DOWNSTREAM
UPSTREAM

100

50

0
-1000 -800 -600 -400 -200 0 200 400 600 800 1000
CANTILEVER STRESS (PSI)

Fig. 11-5.21 Crown-section hydrostatic cantilever stresses for varying foundation

modulus with elevation.

11-89
11-5.6 Limit State Analysis

In addition to failures resulting from exceeding the load-resisting capacity of the concrete
and potential foundation instability, other possible types of failure should also be consid-
ered in the evaluation of arch dams for static loads. The analyses presented below are
examples of the process of ruling out failure mechanisms. The purpose of this section is
to call attention to the fact that computation of the stress within the body of the dam may
not be the sole indicator of the dam’s stability. Although the failure modes considered
here are highly unlikely for majority of exiting arch dams, they cannot be completely
ruled out for dams with unusually gentle abutment slopes and for very thin dams.

11-5.6.1 Sliding on the Abutment Contact

Usually, sliding stability along the dam-foundation contact of a concrete arch dam is un-
likely because of the wedging produced by arch action. However, arch dams with rela-
tively flat abutment slopes such as Dike “F”, FERC project # 1759, shown in Fig. 11-
5.22, or in cases where the concrete is not thoroughly bonded to the foundation rock and
adequate drainage is not provided, the benefit of wedging will be reduced. In these
situations the possibility that a portion of the dam might slide along the dam-foundation
contact should be evaluated. A failure of this type did occur at Plum dam, located in
southeast China. (See 11-8.3.2 )

1
5

3
10

Fig. 11-5.22 Arch dam with relatively flat abutment slopes.

The potential for sliding can be evaluated by comparing computed shear forces with the
shear resistance along the dam-foundation contact surface. An example of limit state
analysis is shown below. In this example, non-linear finite element analysis first allows
for the elimination of tensile stress normal to the dam/foundation contact through crack-
ing. Then, the shear strength of the contact is reduced by allowing local sliding until the

11-90
non-linear finite element solution fails to converge. The direction of impending failure is
depicted in figure 11-5.23

Fig 11-5.23 Direction of Impending Motion with Base Shear Failure.

Fig. 11-5.24 Angle “φ” between forces applied to the foundation and outward directed
normal to foundation under various modeling assumptions.

Figure 11-5.24 depicts the friction angle mobilized at the dam foundation contact under
various modeling assumptions. The linear model indicates that shear strength require-

11-91
ments are highest near the center of the dam (Sta. 0). When tensile stress normal to the
dam/foundation contact is disallowed, the shear strength required on the portion of the
dam/foundation contact that is not cracked increases. The shear strength of the portion of
the dam foundation contact that is cracked is assumed to be 0. Finally, shear stress re-
distribution is allowed and load is transferred up the abutments. When the shear strength
is reduced to 31° and no cohesion, the non-linear finite element solution fails to achieve
force equilibrium.

11-5.6.2 Buckling Failure Modes

Over and above the determination of stresses and displacements in arch dams, under
some extreme dam geometries such as thin, single curvature dams with large radii, the
question of buckling stability of an arch dam structure may arise. Figure 11-5.25 and the
corresponding equation describe the buckling mechanism of circular arches subject to
uniform compressive load, qcr

Fig. 11-5.25 Snap through buckling of a simply supported circular arch under uniform
load.

EI  π 2 
qcr = 3  2 − 1
R α 
The critical uniform load in the above equation can be related to the average compressive
arch stress σcr given by

EI π 2 
σ cr =  2 − 1 (11-5.1)
R2 α 

11-92
General formulas for different types of arches with different boundary and load condi-
tions can be found in “The Theory of Elastic Stability” by Timoshenko and Gere.
While this arch analogy provides some insight into how arch dams could fail in buckling,
it is overly simplistic since it treats the problem in only two dimensions. It also ignores
the cantilever resistance of an arch dam and thus is very conservative. If the dam has
double curvature, equation 11-5.1 is even more conservative. The equation is of some
value however, since if the average computed compressive arch stress is less than σcr
predicted by the equation, buckling failure can be ruled out.
Rigorous determination of the buckling stability of a general shell under variable load
such as a double curvature arch dam requires the use of non-linear finite element analy-
sis. If buckling instability is a concern, simple conservative analysis techniques such as
that discussed above should be used if possible.

11-93
11-6 DYNAMIC ANALYSIS

11-6.1 Overview
All dams in seismic zone 3 and higher should be evaluated using dynamic analysis tech-
niques. Dams in zone 2 may also require dynamic analysis on a case by case basis. Cur-
rently, three-dimensional linear-elastic finite-element analysis is the most common tech-
nique used for dynamic analysis.

A linear-elastic dynamic analysis of arch dams typically consists of the following four
basic steps:

1. Determination of design or evaluation earthquakes and the associated ground mo-

tions;
2. Development of appropriate three-dimensional finite-element models including
dam-foundation and dam-water interaction effects;
3. Specification of dynamic material properties, damping, and reservoir-bottom ab-
sorption, if applicable; and
4. Computation of the earthquake response and presentation, interpretation, and
evaluation of the results.

The requirements for development of design or evaluation earthquakes and earthquake

ground motions are discussed in Section 11-4.4. The design or evaluation earthquake for
arch dams is the maximum credible earthquake (MCE). The earthquake ground motions
include the horizontal and vertical response spectra, or three components of accelera-
tion time histories. They are applied uniformly at the fixed boundaries of the foundation
model.

Three-dimensional finite-element models for the arch dam and the foundation rock are
essentially identical to those described for the static analysis (Section 11-5.2), except that
the dam-water interaction effects should also be represented by the added hydrodynamic
mass models or by the frequency-dependent hydrodynamic terms, as appropriate (Section
11-6.5.1).

A seismic safety evaluation of an arch dam should be based on the dynamic material
properties of the dam concrete, foundation rock, and the energy loss at the reservoir bot-
tom, if applicable. Dynamic modulus of elasticity and dynamic strength of the concrete
for earthquake excitation are determined as described in Section 11-3. Damping associ-
ated with dissipation of energy in the concrete arch structure and the foundation rock
must be consistent with the level of ground shaking, amount of non-linear responses de-
veloped in the dam and foundation, and the properties of the foundation rock. For the
purpose of safety evaluation, a damping value of 5% or 10% should be used. A 5%
damping should be applied to stress and sliding stability analysis of all dams. The in-

11-94
crease to a 10% is acceptable for stress analysis of those dams showing energy dissipa-
tion through joint opening and tension cracking. The sliding analysis of thrust blocks and
abutment wedges, however, should always be conducted using a 5% damping.

The linear-elastic response of arch dams to earthquake loading is computed using the re-
sponse-spectrum mode-superposition and the time-history method. The response-
spectrum method of analysis gives only the maximum displacements and stresses. The
response-spectrum method may be sufficient in most cases, especially when the dam re-
sponse remains mainly within the linear elastic range of behavior. Linear time-history
analysis is recommended when tensile stresses significantly exceed tensile strength of
the concrete and consideration of time-dependent nature of the dynamic response is es-
sential to assessment of dam safety, or when the frequency-dependent dam-water and
dam-foundation interaction models are employed.

Evaluation of earthquake performance of an arch dam should consider many modeling,

material properties, and seismic parameter assumptions commonly made in dynamic
analysis of the dam. Whether the time-history or response-spectrum method of analysis is
employed, the results are still influenced by such assumptions and cannot always be
viewed as conclusive in itself. Rather than just checking stress contour plots to see if the
apparent dynamic tensile strength of the concrete has been exceeded, evaluation of a dy-
namic analysis should also be aimed at answering the following questions:

1) How do the dynamic stresses compare in magnitude with the static stresses that the
dam is currently bearing? Does the dynamic load represent a substantial increase
over the static load?

2) Where do the major natural periods of the dam fall on the earthquake response-
spectrum curve? As the earthquake shaking progresses, contraction joints may
open and tension cracks develop. The joint opening and cracks increase the periods
of vibration of the dam, possibly shifting the periods into different region of re-
sponse spectrum, and hence changing the maximum response. If the natural periods
of the dam fall in the ascending region, the dynamic response of the dam will in-
crease as joint opening and cracks develop, as long as the modified periods still re-
main in this region. If the natural periods of the dam are in the descending region,
the dynamic response of the dam will decrease as joint opening and cracks develop.
This kind of evaluation can therefore indicate whether the assumption of an un-
cracked linear-elastic structure is conservative or unconservative with respect to the
dynamic loading. If the primary natural periods fall far from the peak of the earth-
quake response spectra , the analyst must make sure that a change in the value of
uncertain parameters will not greatly increase the response.

11-95
 SPECTRAL
ACCELERATION

ASCENDING DESCENDING
REGION REGION

PERIOD

3) Contraction joints are known to have very little or no tensile strength, and the ten-
sile strength across the lift joints, especially in older dams, is expected to be less
than that for the intact concrete. Consequently, the contraction joins may open and
close repeatedly and some of the lift joints crack regardless of the tensile strength of
the “intact concrete” being exceeded or not. The question to be answered, then, is
whether or not such joint opening and cracks will lead to failure mechanisms.

11-6.2 Finite-Element Response Spectrum Analysis

The response-spectrum analysis uses earthquake response spectra as the seismic input to
compute the maximum response of an arch dam due to earthquake loading. The seismic
input includes the two horizontal and the vertical response spectra of the MCE earth-
quake ground motions discussed in Section 11-4.4. The finite-element model of the dam
system, including the dam-foundation and the dam-water interaction effects, is developed
as described in the following section. Dynamic material properties are used for the dam
concrete and the foundation rock, and a constant damping ratio as specified above is em-
ployed for all modes of vibration. The maximum nodal displacements and element
stresses are computed separately for each individual mode of vibration, and are then
combined to obtain total maximum response values due to all significant modes and all
three components of the ground motion.

11-6.2.1 Structural Models

The finite-element model of the dam system for the response-spectrum mode-superposi-
tion method is essentially identical to that developed for the static analysis, except that
the effects of the dam-water interaction must be also included.

11-96
The arch dam is modeled as a monolithic structure represented by a single or more layers
of finite elements of appropriate types and sizes in accordance with the guidelines pro-
vided in Section 11-5.2. In addition, sufficiently small finite elements must be used, so
that the contribution of all significant vibration modes with higher harmonic deflected
shapes are represented accurately. The interaction effects of the power intakes and other
appurtenant structures attached to the dam, if significant, must be considered by a lumped
mass representation or by direct modeling of the appurtenant structure as part of the dam
model, as appropriate.

Dam-foundation interaction effects are represented by a foundation model as described in

Section 11-6.5.2.

Dam-water interaction effects are represented in accordance with the guidelines given in
Section 11-6.5.1.

11-6.2.2 General Principles

The linear-elastic dynamic response of arch dams to earthquakes can be obtained by the
mode-superposition response-spectrum method. This method provides estimates of the
maximum response directly from the earthquake response spectrum.

The modal or mode-superposition method is based on the fact that for certain forms of
damping, the response in each natural mode of vibration can be computed separately, and
then combined to obtain the total response. Each mode responds with its own particular
pattern of deformation or mode shape φn; with its own natural frequency of vibration
ωn; and with its own modal damping ratio ξn. Considering that the response of struc-
tures including dams to earthquakes is essentially due to the lower modes of vibration,
only the response in the first few modes need be considered.

The response of the nth mode of vibration for an idealized arch dam structure can be ob-
tained from the analysis of a single-degree-of-freedom (SDOF) system expressed as

Ln
y n + 2ξn ω n y! n + ω n2 y n = −
!! u!! (t ) 11-6.1
Mn g
with natural vibration frequency ωn and damping ratio ξn, excited to the degree Ln/Mn
(participation factor) by the ground acceleration u!!g (t ) . The earthquake excitation factor
Ln and the modal mass Mn are given by:
N M

Ln = ∑ m j φ jn + ∑ mia φin 11.6.2

j =1 i =1

11-97
 N M

M n = ∑ m j φ + ∑ mia φ na2
2
jn 11.6.3
j =1 i =1

where mj is the dam nodal mass, N is total number of degrees-of-freedom, mia is the water
added-mass defined at the dam-water interface nodes, and M is number of the dam-water
interface degrees-of-freedom.

In the response spectrum method, the maximum response for the nth mode expressed by
the above equation is obtained directly from the earthquake response spectrum. As illus-
trated in Fig. 11-6.1, for a natural period Tn, and modal damping ratio ξn the maximum
modal displacement due to response spectrum of the k component of earthquake ground
motion is:

S (T , ξ )
Ln
Ynk,max = 11-6.4
M n ω n2 an n n
where San is the spectral acceleration, and is related to pseudo-velocity, Sv, and relative
displacement, Sd, by

S an = ωS v = ω 2 S d 11-6.5

Alternatively, the maximum modal displacement may be computed by using any of these
response spectrum ordinates.

SPECTRAL
ACCELERATION

San ( T n , ξ n )

ξ n

PERIOD
Tn

Fig. 11-6.1 Acceleration response spectrum illustrating how spectral acceleration

for at a given period and damping ratio is obtained.

11-98
Finally, the maximum displacement vector for the dam nodal points is

unk,max = φ nYn ,max 11-6.6

Knowing the maximum nodal displacements, the maximum element stresses for each
mode are obtained from the displacement-stress relationship.

Number of modes required.

In general, only a few lower modes of vibration are needed in the response spectrum
analysis of arch dams. The actual number of modes to be included in the analysis de-
pends on the natural periods and natural vibration modes of the dam, on the response
spectrum ordinates, and on the response quantity of interest. The first three to five modes
generally contribute the most, but for dams with numerous closely spaced modes with
periods longer than 0.1 sec, the contribution of modes beyond the third may be signifi-
cant. Some response quantities such as nodal displacements and arch stresses normally
require fewer modes than cantilever and shear stresses to achieve more than 90% of their
"exact" values.

As a general rule, the response spectrum analysis should include sufficient number of
modes until the computed response quantities are at least 90% of their "exact" values.
Since the "exact" response values are not known in advance, a trial and error procedure
may be adapted, in which successive analyses are repeated with additional modes until
the addition of modes does not affect the result more than a few percent. Alternatively,
the analysis may be carried out by including a minimum of five modes or by considering
all modes having natural periods longer than 0.1 sec (frequencies < 10 Hz), whichever
gives more number of modes.

Combination of modal response.

The maximum displacements, element stresses, and reaction forces (arch thrusts) com-
puted for each significant mode of vibration must be combined appropriately to obtain
the total response of the dam due to each component of the earthquake ground motion.
Since the maximum modal responses do not occur at the same time during the earthquake
excitation, the complete quadratic combination (CQC) method or the square-root-of-the-
sum-of-the-squares (SRSS) method may be used to obtain an approximate estimate of the
total response.

The CQC modal combination method (Wilson, Der Kiureghian and Bayo 1981) is based
on the random vibration theory and can be used if the duration of the strong motion por-
tion of the earthquake record is several times longer than the fundamental period of the
dam and if the response spectrum ordinates vary slowly over the dominant period range
of the dam. Both of these conditions are easily satisfied for most arch dams using smooth
response spectra with 5% damping or more. The CQC formula for the maximum total
k
displacements umax due to the k component of the earthquake ground motion is given by

11-99
 p p
k
u max = ∑∑u k
m,max ρmn unk,max 11-6.7
m =1 n =1

where umk,max and unk,max are the maximum modal displacements for modes m and n, re-
spectively, and p is the number of modes considered; and ρmn is the cross-modal coeffi-
cient which for the constant modal damping ξ is defined by:

3
8ξ 2 (1 + r )r 2
ρ mn = 11-6.8
(1 − r 2 ) + 4ξ 2 r(1 + r )
2 2

in which r is the ratio of natural periods Tm/Tn. For well separated modes (r < 0.75) and a
5% damping, ρmn is less than 0.1, indicating negligible cross-correlation between the
modes. Whereas for closely spaced modes (r > 0.75) and especially for higher damping,
cross-modal coefficient increases showing significant interaction between the modes.

For dams with well separated modes, cross-modal coefficient, ρmn, approaches zero (for
m ≠ n), and the CQC method converts into the standard SRSS method:

∑ (u )
p
2
k
umax = k
n ,max 11-6.9
n =1

While the SRSS method leads to conservative estimates for the well-separated modes, it
may provide conservative or unconservative results when the modes are closely spaced.

Combining for multi-earthquake components.

The total maximum response of the dam due to each component of the earthquake ground
motion is obtained using one of the modal combination methods described above. The
maximum response due to all three components (two horizontal and vertical) of the earth-
quake ground motion should be obtained using the SRSS method:

∑ (u )
3
2
u max = k
max 11-6.10
k =1

11-6.2.3 Presentation and Interpretation of Results

The basic results of a response-spectrum analysis include natural frequencies and mode
shapes, maximum nodal displacements, and maximum element stresses. Maximum nodal

11-100
displacements and maximum element stresses are due to the earthquake loading only.
They must be combined with the effects of static loading in order to obtain the total
maximum response values.

Natural frequencies and modes of vibration

In a response spectrum analysis, the modal properties including the natural frequencies
and natural modes of vibration for the dam system must be known and thus are computed
first. T he natural frequencies and modes of vibration provide important information on
the dynamic characteristics of the dam, its level of interaction with the impounded water,
and its level of response to earthquake loading characterized by a response spectrum. In
order to acquire an advance knowledge of the dynamic behavior of the dam and also to
examine the accuracy of the results, the computed natural modes of vibration should be
presented in the form of deflected shapes shown in Fig. 11-6.2, or in other appropriate
forms.

Fig. 11-6.2 Ten lowest vibration modes of Marrow Point Dam with full reservoir and
flexible massless foundation.

11-101
Dynamic response quantities
The maximum nodal displacements and element stresses are obtained by combining re-
sponses from the individual modes and multi-component earthquake input. The resulting
dynamic displacements and stresses estimated in this manner are all positive but for
evaluation they must be assumed to be positive or negative. The positive dynamic
stresses are interpreted as the maximum tensile stresses, whereas the negative dynamic
stresses serve as the minimum compressive stresses. The response spectrum stresses
generally occur at different of times, thus static equilibrium checks cannot be performed
to validate the results.

Total maximum stresses

For evaluation of the earthquake performance of arch dams, the response-spectrum esti-
mates of the dynamic stresses (σd) must be combined with the static stresses (σs) due to
the usual loading combination (Section 11-4.5.3). Since response-spectrum dynamic
stresses can be either positive or negative, this combination leads to the maximum total
tensile stresses and the minimum total compressive stresses:

σ max,min = σ s ± σ d 11-6.11

Only similarly oriented stresses such as the static and dynamic arch stresses, static and
dynamic cantilever stresses, and the static and dynamic shear stresses can be combined in
this manner. The basic stress results needed for evaluation include arch and cantilever
stresses on the upstream and downstream faces of the dam. The resulting total stresses
may be displayed in the form of stress contours shown in Figs. 11-6.3 and 11-6.4. These
results were computed for Morrow Point Dam due to the static loads discussed previously
in Section 11-5.2.3 and the example response spectra described in Section 11-4.4.

The envelope of maximum total tensile stresses in Fig. 11-6.3, show the computed peak
tensile arch and cantilever stresses on the upstream and downstream faces of the dam that
generally occur at different times during the earthquake excitation. Similarly, the enve-
lope of minimum total compressive stresses in Fig. 11-6.4 exhibits the non-concurrent
peak compressive arch and cantilever stresses on the upstream and downstream faces of
the dam. The dam performance is considered satisfactory if the evaluation criteria set
forth in Section 11-1.4.4 is met. In cases where high tensile stresses indicate significant
joint opening and tension cracking will occur, consideration of stress histories, concur-
rent stresses, and the number and duration of significant tensile stress excursions com-
puted using a time-history analysis provide additional information for evaluation of the
earthquake performance of the dam.

11-102
 max = +729 max = +610
min = -508 min = -344

+ = Tension
max = +716 max = +443
min = - 391 - = Compression min = -605

Fig. 11-6.3 Envelope of maximum total (static + SRSS) tensile stresses (in psi) on up-
stream and downstream faces of Morrow Point Dam on flexible foundation
with full reservoir due to upstream, vertical, and cross stream response
spectra .

max = +128 max = +0

min = -1798 min = -707

+ = Tension
max = +158 max = +0
min = -1398 - = Compression min = -1290

Fig. 11-6.4 Envelope of minimum total (static SRSS) compressive (in psi) on upstream
and downstream of Marrow Point Dam on flexible foundation with full reser-
voir due to vertical, upstream, and cross valley response spectrum.

11-103
11-6.3 Finite-Element Time-history Analysis
The response of arch dams to earthquakes can also be evaluated by the time-history
method of analysis. Although time-history method is applicable to both linear and non-
linear response behavior of dams, the linear time-history analysis is most commonly used
for seismic safety evaluations of arch dams. In most cases, linear time history analysis
coupled with engineering judgement is sufficient to evaluate the safety of an arch dam
under seismic loading. If severe damage is indicated, nonlinear dynamic analysis can be
done and will be accepted by the FERC. Currently however, the FERC has no accep-
tance criteria for the allowable level of damage . No matter how accurate, nonlinear
analyses are subject to numerous assumptions that would still require engineering judge-
ment .
The linear time-history response analysis is advantageous because of its ability to ana-
lyze time dependent characteristics of the dynamic response. This provides additional
information for safety evaluation of the dam such as whether or not high stresses occur
simultaneously, have short duration, and repeated many times. The seismic input for lin-
ear time-history analysis consists of three components of the earthquake ground motion
acceleration time histories developed and applied to the dam in accordance with guide-
lines of Sections 11-4.4. The finite-element model of the dam-water-foundation system
is developed as described in Section 11-6.5.1. Dynamic material properties are used for
the concrete and the foundation rock with a constant damping ratio in accordance with
Section 11-6.1. If applicable, the reservoir-boundary absorption coefficient is selected as
described in 11-6.5.1.3. The maximum and history of nodal displacements and element
stresses are computed by the time-history mode-superposition method or by the direct
integration of the equations of motion.

11-6.3.1 Structural Models

The three-dimensional finite-element models of the arch dam and the foundation rock for
the linear time-history analysis are identical to those described previously for the re-
sponse spectrum analysis. The dam-water interaction effects may be represented by an
added hydrodynamic mass model or by the frequency- dependent hydrodynamic terms
discussed in 11-6.5.1.

The added hydrodynamic mass models are applicable to arch dams for which the funda-
mental frequency of the reservoir water is at least 2 times greater than the fundamental
frequency of the dam. The finite-element incompressible added-mass model is the pre-
ferred model for the added-mass representation. However, the generalized added-mass
model is also acceptable when it gives conservative results.

11-104
11-6.3.2 General Principles
Linear time-history analysis uses acceleration time-histories as the seismic input, and
computes complete response histories of the dam for the entire duration of the earthquake
ground shaking. A time-history response analysis provides not only the maximum stress
values, but also the sequencing, spatial extent and number of excursions beyond any
specified stress value, all of which together with the rational interpretation are essential
parts of a dam safety evaluation.

The complete response history of the dam structure to earthquakes is obtained by the so-
lution of the equations of motion. The equations of motion assembled for an idealized
dam-water-foundation system generally contain several hundreds to several thousands of
ordinary differential equations. In general, these equations are coupled and can be solved
simultaneously by direct integration method, or separately by the mode-superposition
method. Both methods are applicable to the linear dynamic response analysis, but only
direct method is usually employed in the non-linear dynamic analysis.

11-6.3.2.1 Direct Integration Methods

In direct integration the coupled system equations of motion is solved by a numerical
step-by-step integration procedure. The term “direct” simply means that prior to the nu-
merical integration, the equations are not transferred into a different form. A direct inte-
gration method is based on two basic principles: 1) the equations of motion are satisfied
at discrete time intervals ∆t apart, and 2) a variation of displacements, velocities, and ac-
celerations must be assumed within each time interval ∆t. It is, in fact, the form of the
assumption of variation of displacements, velocities, and acceleration that determines the
stability, accuracy, and efficiency of the integration procedure. The direct integration
method is not restricted to the linear analysis as is the modal-superposition method dis-
cussed in 11-6.3.2.2. Therefore, it can be used when the contraction joint opening and
material or geometric non-linearity are considered.

A numerical integration procedure is either called explicit or implicit. When the solution
at time t+∆t is based on the equilibrium conditions at the previous time step t, the inte-
gration method is called an explicit integration method. Methods that use equilibrium
conditions at time t+∆t to obtain solution at time t+∆t are called implicit integration
methods. The implicit method is normally selected for the linear earthquake response
analysis of arch dams. The main advantage of the implicit method is that it is un-
conditionally stable: this means there is no mathematical limit on the size of the time in-

11-105
terval, and ∆t can, in general, be selected much larger than that for the explicit methods.
The time interval ∆t should be small enough that response in all modes which signifi-
cantly contribute to the total structural response is calculated accurately. For example,
when the lowest p frequencies and mode shapes of the dam are considered significant, the
smallest period is Tp , and thus ∆t = Tp /10 should provide sufficient accuracy.

Some of the commonly used implicit integration methods include Houbolt integration
scheme, Newmark's integration scheme, and Wilson θ method. Both the Newmark's and
Wilson θ methods are an extension of the linear acceleration method, in which a linear
variation of acceleration from time t to t+∆t is assumed. For certain selection of the inte-
gration parameters, the Newmark's method converts to the constant-average-acceleration
and the linear acceleration schemes. The linear acceleration scheme can also be obtained
using θ = 1 in the Wilson θ method.

11-6.3.2.2 Mode Superposition Method

The time-history mode-superposition method involves similar step-by-step integration
described above, except that integration is applied to the uncoupled equations of motion.
The coupled equations of motion are first transferred into uncoupled modal equations
using the natural frequencies and natural modes of vibration. The equation of motion for
each mode, Eq. (11-6.1), is then integrated for the entire duration of the ground shaking.
The response history for each mode can be obtained by the Duhamel integral (Clough and
Penzien, 1993)

t
L 1
M n ω nD ∫0 g
Yn ( t ) = − n [ ] [ ]
u!! (τ ) exp − ξn ω n (t − τ ) sin ω nD (t − τ ) dτ (11-6.12)

or other integration schemes described above. In this equation, ω nD = ω n 1 − ξ n2 is the

damped vibration frequency, but its difference with the undamped frequency is negligible
for damping ratios of less than 20% and is normally ignored. The displacement response
history of the dam for the nth mode is given by

unk ( t ) = Yn ( t )φn (11-6.13)

which when combined algebraically for the lowest p modes and for three components of
the earthquake ground motion, the total displacement histories are obtained.

11-106
 3 p

u( t ) = ∑ ∑ unk ( t ) (11-6.14)
k =1 n =1

These operations are also performed for other response quantities. For example, the dis-
placement history in Eq. (11-6.13) and the displacement-stress relationships are used to
obtain the stress history in the nth mode, and then are combined according to Eq. (11-
6.14) to obtain the total dynamic stress histories.

11-6.3.2.3 Total Static Plus Dynamic Stresses

Total stresses required for the evaluation of the earthquake performance of the dam are
obtained by algebraic addition of the initial static stresses due to the usual loading combi-
nations and the dynamic stress histories. In a linear response analysis, static and dynamic
responses are usually computed by two separate analyses and combined afterwards to
obtain the total response.

11-6.3.3 Presentation and Interpretation of Results

The basic results of a time history analysis include response histories of the nodal dis-
placements and element stresses computed for the entire duration of the earthquake
ground shaking.

11-6.3.3.1 Mode Shapes and Nodal Displacement Histories

Mode shapes and frequencies are computed when the time-history mode-superposition
method is employed. In this case mode shapes are presented and evaluated as described
in 11-6.2.3.

The magnitude and time history of nodal displacements provide a visual means for evalu-
ation and validation of results. As a minimum, displacement time histories at several
critical locations along the dam axis such as the mid-crest and quarter-span points should
be presented and evaluated. From displacement magnitudes, for example, it is possible to
decide whether the displacements are sufficiently small to infer that the stability of the
dam is maintained. Comparison between displacement components at some strategic lo-
cations (mid-span, quarter-span, etc.) can provide insight to the dynamic response be-
havior and a means for validation of the results. Fig. 11-6.5 shows displacement re-
sponse histories for the mid-crest point of Morrow Point Dam due to earthquake input
acceleration time histories given in Fig.11-4.6. As expected, Fig. 11-6.5 indicates that
the mid-crest displacement of the symmetric Morrow Point Dam is largest in the stream
direction. It also shows that the response of Morrow Point Dam (with hydrodynamic

11-107
added-mass model) to vertical component of the ground motion is small, as evident by
the magnitude of the vertical displacements.

4
3.6
UPSTREAM COMPONENT
3
Displacement (in)

2
1
0
-1
-2
-3
-4
0 2 4 6 8 10 12 14 16 18 20
Time (sec)

4
VERTICAL COMPONENT
3
Displacement (in)

2
1 0.3

0
-1
-2
-3
-4
0 2 4 6 8 10 12 14 16 18 20
Time (sec)

4
CROSS-STREAM COMPONENT
3
Displacement (in)

2
1
0
-1 -0.6
-2
-3
-4
0 2 4 6 8 10 12 14 16 18 20
Time (sec)

Fig. 11-6.5 Displacement time histories of center of dam crest.

11-108
 max = +890 max = +631
min = -514 min = -373

+ = Tension
max = +838 - = Compression max = +405
min = -398 min = -615

Fig. 11-6.6 Envelope of maximum tensile stresses due to static plus upstream, vertical,
and cross-stream earthquake acceleration time histories

Linear continuum models typically under predict deflections because they ignore the ef-
fects of cracking and joint opening. For this reason, excessive displacement should be of
concern. Engineering judgement is required to decide when deflections should be con-
sidered excessive. In one instance, a dynamic analysis of a dam revealed that while dy-
namic stresses in the service bridge piers were within reason, the predicted deflections
indicated that the bridge would fall off the piers. What deflections can be tolerated must
be evaluated on a case by case basis.

11-6.3.3.2 Envelopes of Maximum and Minimum Total Stresses

The envelopes of maximum and minimum total stresses are the first stress results to be
presented and evaluated. They should be presented as stress contour plots of arch
stresses and cantilever stresses on each face of the dam. The magnitudes of maximum
and minimum arch and cantilever stresses at each stress point on faces of the dam, are
obtained by first adding static stresses to the dynamic stress histories and then searching
for the maximum and minimum values through their respective histories. Contours of the
resulting maximum and minimum stresses obtained in this manner represent the largest
tensile (positive) and the largest compressive (negative) stresses that occur at any location
in the dam during the entire earthquake excitation. It is evident that the maximum or
minimum stresses at different locations generally occur at different times, thus are non-
concurrent. Fig. 11-6.6 is an example of maximum tensile stress contours for Morrow
Point Dam. These contours are used to identify the regions where the excessive tensile
stresses may suggest contraction joint opening and/or cracking. The extent and severity

11-109
of damage is determined by further examination of stresses in these regions and consid-
eration of the time-dependent nature of the dynamic response, as discussed in the fol-
lowing sections.

Similarly, contours of the minimum stresses in Fig. 11-6.7 indicate the largest compres-
sive stresses that could develop in the dam during the earthquake ground shaking. They
should be compared with the allowable compressive stress to ensure that they meet the
required factors of safety given in Table 11-1.1. Usually compressive stresses are small
and within the allowable values, but may become critical if cracking significantly reduces
the area of concrete available for transmitting cantilever forces.

If envelop values show that both tensile and compressive stresses satisfy the performance
requirements for the Extreme Loading Combination, the dam will be considered safe
against overstressing and no further stress evaluation will be required. Otherwise, a more
detailed stress evaluation (discussed in the following sections) should be conducted, so
that the expected level of damage and the stability conditions of the dam can be deter-
mined. After the numerical results have been evaluated for reasonableness, the stress re-
sults should be evaluated with respect to the guidance laid out in 11-1.4.4

max = +139 max = 0

min = -1975 min = -721

max = +174 + = Tension max = +6

min = -1514 - = Compression min = -1293

Fig. 11-6.7 Envelope of highest compressive stresses due to static, plus upstream, verti-
cal and cross valley earthquake acceleration time histories.

11-110
11-6.3.3.3 Simultaneous or Concurrent Critical Stresses
The envelopes of maximum and minimum stresses discussed in 11-6.3.3.2 indicate the
largest tensile and compressive stresses that are developed on faces of the dam at any
time during the earthquake excitation. They serve to identify the overstressed regions,
the location of largest maximum stress points (called the critical stress points), and the
time-steps at which critical stresses reach their peaks. This information is used to obtain
the concurrent (or simultaneous) stresses corresponding to the time-steps at which the
critical arch and critical cantilever stresses reach their peak values. Fig. 11-6.9 shows
concurrent stress contours at the time of maximum critical arch stress occurring at loca-
tion SP-1, and Fig. 11-6.11 presents concurrent stresses at the time of maximum critical
cantilever stress associated with location SP-2. In this example, time history of the criti-
cal arch (Fig. 11-6.8) and critical cantilever stresses (Fig. 11-6.10) are compared with an
assumed tensile strength of the intact concrete as an indicator of the severity of tensile
stresses. The earthquake performance of the dam should be carried out in view of the
criteria stated in 11-1.4.4.

Concurrent stresses obtained in this manner can be interpreted as snap shots of the critical
stress distributions. They serve to identify the location and extent of overstressed regions
at critical time steps. In addition, since concurrent stresses are given for both faces of the
dam, the severity of possible joint opening and cracking can be estimated from the com-
parison of stresses on opposite faces of the dam.

11-6.3.3.4 History of Critical Maximum Stresses

When maximum and concurrent stresses show significant tensile stresses, the time histo-
ries of the peak tensile stresses should be presented and evaluated. Time histories of the
most critical arch and cantilever stresses are examined to estimate the amount of damage
that could be expected. If the tensile strength of the concrete is exceeded many times by
large amounts, more damage is to be expected than if stress excursions above the con-
crete strength are few and of short duration.
The critical arch stress history shown in Fig. 11-6-8, shows several significant tensile
stress cycles that can momentarily open the contraction joint at the crown with no serious
consequences, as discussed in the next section. The critical cantilever stress history
shown in Fig. 11-6.10 indicates that only one stress peak exceeds the tensile strength of
the intact concrete by 5%. This localized stress peak together with several lower tensile
stress cycles is partly fictitious and may be interpreted to have been caused by the as-
sumption of linear-elastic behavior for the foundation rock, a material which is frag-
mented by joints and fissures and can resist limited tension. This again has no significant
effect on the stability of the dam.

11-111
 1000
Tensile Strength of Concrete = 600
500

Arch Stress (psi)

0

-500

-1000
Initial Static Stress

-1500

-2000
0 2 4 6 8 10 12 14 16 18 20

Fig. 11-6.8 Arch stress time history at point SP1 (Shown in Fig. 11-6.9) Point of maxi-
mum instantaneous tensile arch stress.

SP-1

+ = Tension

- = Compression

Fig. 11-6.9 Maximum tensile arch stress contours @ t=12.23 sec.

11-112
 700
Tensile Strength of Concrete = 600
600

Cantilever Stress (psi)

500 Initial Static Stress
400
300
200
100
0
-100
-200
-300
0 2 4 6 8 10 12 14 16 18 20
Time (sec)

Fig. 11-6.10 Cantilever stress time history at location SP-2 (shown below), where the
largest tensile cantilever stress occurs.

SP-2

+ = Tension
- = Compression

Fig. 11-6.11 Instantaneous stresses at the time of maximum tensile cantilever stress (t =
7.38 sec) due to static plus upstream, vertical, and cross-stream earthquake
ground acceleration time histories.

11-6.3.3.5 Final Evaluation of Stress Results

Evaluation of Arch Stresses. Considering that the vertical contraction joints in arch dams
can resist little or no tension, the arch tensile stress distribution through the dam thickness

11-113
can be interpreted as brief joint opening as the dam vibrates in response to earthquake
ground motion. The joints, however, are usually designed to limit upstream-downstream
shearing translation.
An accurate characterization of the behavior of the joint opening and closing mechanism
and its effects on redistribution of stresses can probably be captured only by a non-linear
dynamic analysis (Fenves et. al. 1989). However, severity of joint opening may be
roughly estimated from contour plots of the average arch stresses to ensure that the joint
openings are small and that the cantilevers bounded by the partial joint openings are not
overloaded. Fig. 11-6.12 illustrates an approximate procedure for estimation of the ex-
tent of joint opening. Assuming that the vertical joints cannot resist tension, even a mo-
mentary net tensile force across the joints might indicate a joint opening. The vertical
extent of joint opening may be approximated by the extent of the region showing net ten-
sile arch stresses across the joint. A joint opening presumably releases arch stresses
across the joint and transfers overturning forces to the cantilevers. The cantilever blocks
must have sufficient tensile strength to resist the additional overturning to remain stable.
The basic concept, therefore, is first to identify possible cantilever monoliths bounded by
partial joint openings, and then to assess their stability, considering that they must resist
additional forces resulting from momentary loss of arch action.

After the critical tensile arch stresses have been established, time histories of the up-
stream and downstream arch stresses at location of the largest tensile arch stress are aver-
aged to obtain the time history of the average arch stresses at that location. Each positive
stress pulse in the resulting averaged time history is indicative of net tensile arch stresses
across the dam thickness. For example, the middle graph in Fig. 11-6.12 suggests that
the contraction joints at location of the largest arch tensile stress are subjected to 16 cy-
cles of net tensile arch stresses, each of which might momentarily open the joint for hun-
dredth of a second. The largest joint opening is expected to occur at t = 12.23 sec, when
the net tensile stresses reach their maximum value. Next the concurrent stresses corre-
sponding to the time of maximum net tensile stress (t = 12.23 sec) are retrieved and aver-
aged separately for the arch and cantilever stresses. Tension region in the resulting aver-
age arch stress contours (bottom graph in Fig. 11-6.12) indicates the location and ap-
proximate depth of joint opening. The corresponding average cantilever stress contours
show the magnitudes of cantilever stresses and hence the reserve capacity available to
resist additional cantilever stresses should the contraction joints open. For small and
moderate joint openings, the partially free cantilevers bounded by opened joints may re-
main stable through interlocking (wedging) with adjacent blocks. The extent of inter-
locking depends on the depth and type of shear keys, and on the amount of opening to be
expected. If the results indicate excessive joint separation with little or no wedging ac-

11-114
tion, then the most critical "free" cantilever bounded by two opened joints should be
identified and analyzed for overturning stability.
1000

500
Arch Stress (psi)

-500

-1000
Upstream Stress
-1500
Downstream Stress
-2000
0 2 4 6 8 10 12 14 16 18 20

500
Arch Stress (psi)

Maximum net tensile arch stress

-500

-1000
0 2 4 6 8 10 12 14 16 18 20
Time (sec)

Figure 11-6.12 Procedure for identification of potential free cantilevers.

Stability Analysis of Free Cantilevers. When contraction joint opening occurs, what
should be shown is that a free cantilever bounded by the temporarily opened joints will
not fail. In order to insure the free cantilever remains stable, the total vertical force re-

11-115
sultant across the lift joint in question must be compressive and fall within the structure.
Even if the resultant force across the lift joint falls temporarily outside the base, it is usu-
ally the case that the overturning forces are of in-sufficient duration to topple the free
cantilever.

Fig. 11-6.13 shows the possible failure mode for a free cantilever block defined by two
fully opened contraction joints and a horizontal lift joint. For the amount of time that ver-
tical contraction joints are open, the free cantilever ‘s behavior is governed by differen-
tial equation 11-6.15

Fig. 11-6.13 Free body diagram of a cantilever with associated forces.

 δ 2φ 
−M + A S WR sin( α + φ ) − WR cos( α + φ ) − M =  2  ρ I P
 δt
HD HS

(11-6.15)
Where:
MHD = Moment due to hydrodynamic pressure (function of time)
AS = Acceleration of the structure at the location of the block (function of time)
MHS = Moment due to hydrostatic pressure
W = Weight of free cantilever block
R = Distance from pivot point to block center of mass
ρ = Mass Density of block
IP = Polar moment of inertia about pivot point

11-116
Equation 11-6.15 can be solved numerically. AS and MHS can be obtained directly from
a linear elastic time history analysis, or approximated by combining significant modes of
vibration from a spectral analysis. Equation 11-6.15 assumes that the horizontal lift joint
upon which the block sits is de-bonded, and therefore it is a very conservative treatment
of the problem. In most cases, the results of this type of conservative block rocking
analysis show that even if a free cantilever can form, it does not displace very much.

Fig. 11-6.14 Example of response of free cantilever block to sinusoidal acceleration of

2g @ 3Hz.

In the example depicted in Fig 11-6.14, equation 11-6.15 results in a peak rotation of
0.4°. It also indicates that the cantilever block should fall back into place as acceleration
switches sign. In this example, even if free cantilevers do form they can not topple. It
should also be noted that shear keys often prohibit cantilevers from being fully dis-
engaged unless joint opening exceed the key depth or the keys have failed.

The toppling of blocks under seismic excitation is further treated by George Housner in
his1963 article in the Bulletin of the Seismological Society of America.

11-117
Evaluation of Cantilever Stresses. At the dam-rock interface, the excessive tensile canti-
lever stresses can be interpreted as openings of the rock-concrete joints or of the joints
within the rock below. The magnitudes of these stresses usually decrease when the prob-
able weakness of the fragmented foundation rock in resisting tension is accounted for in
the analysis. When subjected to tensile loading, the foundation region adjacent to the
base of the dam "softens" as rock joints open and additional micro-cracks develop, but
the magnitudes of stresses should not be such that this region would develop major
cracking or would expand beyond the base of the dam. In severe cases, the influence of
the fragmented rock on the stress results and crack propagation can be accounted for by
assuming a reduced modulus (smeared crack) for the foundation-rock elements attached
to the dam which would limit tension-resistance capability of the dam-rock interface.

For locations within the body of the dam away from the foundation boundary, tensile
cantilever stresses exceeding the tensile strength of lift joints can be expected to cause
cracking. Cracking should be assumed to occur along the lift joints or in the direction
normal to the major principal stresses, if the directions of cantilever and major principal
stresses differ significantly. The tensile cracking should be assumed to propagate
through the dam thickness, if tensile cantilever stresses are also developed on the oppo-
site face of the dam. In general, tensile cantilever stresses result from bending of the can-
tilevers which exhibit compression on the opposite face. In these cases, cracking usually
occur only on a portion of the dam section experiencing tensile stresses in excess of the
tensile strength of the concrete. However, so long as the excessive tensile stresses occur
over a very limited area (less than 5% of the dam surface area) and do not exceed the ten-
sile strength more than 5 cycles, it can reasonably be concluded that the cracking does not
necessarily indicate unacceptable performance for the MCE event. Nevertheless, the ex-
tent of such cracking should be estimated and the computed stress distributions should
indicate that the adequate compressive capacity exits to accommodate subsequent load
redistributions. In addition, the vector plots of principal stresses at critical instance of
time should be evaluated to confirm that the indicated regions of tensile cracking are not
likely to join together to form surfaces along which partial sliding failures might occur.

11-6.3.4 Time-History Stability Analysis

The safety against sliding along the dam-foundation interface or along the potential foun-
dation failure planes or wedges is determined on the basis of shear-friction factors of
safety. The shear-friction factor of safety is defined as the ratio of the resisting to “driv-
ing” forces along a potential failure surface. The resisting forces are obtained from shear
strength of the dam-foundation interface (11-2.3.1) or that of the potential foundation

11-118
failure planes or wedges (11-2.3.2). The driving forces acting on the dam-foundation in-
terface are obtained from integration of the interface stresses or interface nodal forces
computed as part of the finite-element analysis described previously. The driving forces
for stability analysis of a rock wedge consist of the dam thrust and the static and seismic
inertia forces acting on the wedge.

In the time-history earthquake analysis, factors of safety are obtained by combining the
"driving" force histories due to the ground motion with "driving" forces due to the usual
static loads including the uplift. At each time step, the seismic and static driving forces
are combined and then transformed into a resultant force having components normal and
tangential to the failure surface. Similar to the static analysis, the resisting forces are rep-
resented by shear strength of the foundation-dam interface or the potential foundation
failure planes or wedges, and the driving forces are taken as the resultant of shear forces
acting on the failure surface. The time history or instantaneous factor of safety is then
computed from the ratio of the resisting to driving forces at each time step.

The time-history factor of safety starts with a value equal to the static factor of safety and
then oscillates as the dam or the rock wedge responds to the ground shaking. Under
earthquake excitation, the stability is maintained and sliding does not occur if the factor
of safety is greater than one (Table 11-1.1). A factor of safety of less than one indicates
that sliding may occur. In this situation, the sliding may be permitted, if it can be shown
that the permanent displacement is small and that the safety of the dam will not be jeop-
ardized. Even small movement of an abutment rock wedge or gravity thrust block could
have drastic effects on the stress distribution with the dam. In such a case, it must be
shown that the dam can continue to retain the reservoir even with support from the abut-
ment wedge or thrust block reduced or eliminated. A rough estimate of the permanent
sliding displacement can be obtained using the Newmark’s rigid block model (Newmark,
1965). According to Newmark’s concept, the sliding takes place during a short period of
time when the ground acceleration cycles exceed a value known as critical acceleration,
and diminishes when the ground acceleration falls below this value and the velocity
reaches zero. The critical acceleration is the acceleration at which the sliding initiates
and is obtained by equating resisting with driving forces.

11-6.4 Alternative Analysis Techniques

The procedures discussed in this section are not intended to rule out the use of alternative
techniques. As the state of the art advances, new techniques which more accurately treat
all aspects of the dynamic behavior of the arch dam/foundation/reservoir system will be-

11-119
come more common. The FERC will always favor the simplest way to evaluate dam
safety.

11-6.5 Reservoir and Foundation Effects

Dynamic response of arch dams to earthquake ground motion is affected by interaction
between the dam and impounded water, interaction between the dam and foundation
rock, damping, and the characteristics of earthquake ground motion. A realistic appraisal
of the seismic safety of the dam is achieved, if the effects of these factors are understood
and properly represented in the analysis.

11-6.5.1 Dam-Water Interaction

Interaction of an arch dam with the impounded water leads to an increase in the dam vi-
bration periods. This is because the dam cannot move without displacing the water in
contact with it. The fact that water moves with the dam increases the total mass that is in
motion. This added mass increases the natural periods of the dam, which in turn affects
the response spectrum ordinates and hence the effective earthquake inertia forces. It can
also cause an increase in damping due to partial absorption of pressure waves at the res-
ervoir boundary and radiation towards the upstream. These effects tend to change the
earthquake response of the dam with respect to that for the dam with empty reservoir,
with the net result depending on the characteristics and component of earthquake ground
motion and on the dam-water interaction model used.

11-6.5.1.1 Generalized Westergaard Added-Mass

The added-mass representation of dam-water interaction during earthquake ground

shaking was first introduced by Westergaard (1933). In his analysis of a rigid, 2D, grav-
ity dam with a vertical upstream face, Westergaard showed that the hydrodynamic pres-
sures exerted on the face of the dam due to the earthquake ground motion is equivalent to
the inertia forces of a body of water attached to the dam and moving back and forth with
the dam while the rest of reservoir water remains inactive. He suggested a parabolic
shape for this body of water with a base width equal to 7/8 of the height, as shown in Fig.
11-6.15.

A general form of the Westergaard added-mass concept which accounts for the 3D ge-
ometry (Clough 1977; Kuo 1982) can be applied to the earthquake analysis of arch dams.
The general formulation is based on the same parabolic pressure distribution with depth
used by Westergaard (Fig. 11-6.15), except that it makes use of the fact that the normal

11-120
hydrodynamic pressure Pn at any point on the curved surface of the dam is proportional
unt :
to the total normal acceleration, !!

Pn = α !!
unt (11-6.16)

7
α= ρ H( H − Z ) (11-6.17)
8 w
where ρw is mass density of water, α is Westergaard pressure coefficient, and H and Z
are defined in Fig. 11-6.15. The normal pressure Pn at each point is then converted to an
equivalent normal hydrodynamic force by multiplying by the tributary area associated
with that point. Finally, the normal hydrodynamic force is resolved to its Cartesian com-
ponents, from which a full 3x3 added-mass matrix at each nodal point on the upstream
face of the dam is obtained (Kuo 1982):

ma = α A λT λ (11-6.18)

where A is the tributary surface area and λT is a vector of normal direction cosines for
each point. Note that while the added-mass terms are coupled with respect to the nodal
degrees-of-freedom, they are uncoupled with respect to individual nodes. Such a 3x3 full
nodal added-mass matrix can easily be incorporated in a computer program using consis-
tent mass matrix (non-diagonal), but it should be diagonalized for those programs that
employ diagonal mass matrix.

The basic assumptions of the generalized Westergaard method are: 1) pressure at any
point on the face of the dam is expressed by the Westergaard parabolic shape, and 2) the
same parabolic shape is used for all three components of the earthquake ground motion.
It should be noted that there is no rational basis for these assumptions, because they do
not meet the conditions imposed in the original Westergaard analysis which included a
rigid dam with vertical upstream face being subjected to the upstream component of
ground motion only.

The Westergaard method usually gives the largest added-mass values, as evident by its
increasing the vibration periods the most. However, this does not automatically give the
largest stresses, because response of the dam also depends on the characteristics of the
earthquake ground motion. If vibration periods of the dam fall in descending region of
the response spectrum, the larger Westergaard added-mass will shift the periods further
into region of smaller effective earthquake forces and thus smaller stresses. The Wester-
gaard added-mass model may be used in the preliminary analysis and also in the final

11-121
evaluation provided that the results show that the dam is safe with an adequate margin of
safety with little or no damage.

Fig. 11-6.15 Generalized Westergaard added hydrodynamic mass model for arch dams.

11-6.5.1.2 Incompressible Finite-Element Added-Mass

The added-mass representation of the impounded water can be obtained more accurately
by a finite-element solution of the pressure wave equation, which fully accounts for the
complex geometry of the dam and the reservoir (Kuo, 1982; Ghanaat, 1993b). The im-
pounded water represented by the wave equation 11-6.19 is discretized using a finite-
element mesh of incompressible liquid elements, with nodal pressures as the unknowns
(Fig. 11-6.16a). The solution is obtained by numerical procedures with the following
boundary conditions:

δ 2P δ 2P δ 2P
+ + =0
δx 2 δy 2 δz 2 11-6.19

1. The hydrodynamic pressures at the water - free surface are assumed to be zero,
that is the effects of surface waves are neglected.
2. The reservoir bottom and sides, as well as a vertical plane at the upstream end of
the reservoir model, are assumed to be rigid (Fig. 11-6.16a). For rigid boundaries
the normal pressure gradients or the total normal accelerations are zero.

11-122
 3. The normal pressure gradients at the dam-water interface are proportional to the
total normal accelerations of the fluid.

The finite-element mesh of the incompressible water can be developed to match the res-
ervoir topography, but in most cases a prismatic model constructed by projecting the
dam nodal points in the upstream direction would suffice. It should at least extend three
times the water depth in the upstream direction and include three or more layers of ele-
ments in that direction, with distances between successive sections increasing with dis-
tance from the dam. The computed pressures for the nodal points on the upstream face of
the dam are then converted into equivalent nodal forces, from which an added-mass ma-
trix representing the inertial effects of the incompressible water is obtained. The result-
ing added-mass matrix is a symmetric full matrix coupling all nodal degrees-of-freedom
on the upstream face of the dam. This matrix can be directly used in computer programs
with the consistent mass (non-diagonal) capabilities, otherwise it should be appropriately
diagonalized. The diagonalization of the mass matrix should be performed such that the
procedure gives the correct total hydrodynamic forces for rigid body acceleration of the
dam.

Fig. 11-6.16 Finite element models of fluid domain with and without water compressi-
bility.

11-123
11-6.5.1.3 Compressible Water with Absorptive Reservoir Boundary

The added-mass representation of the impounded water described above ignores the ef-
fects of water compressibility and reservoir boundary absorption. A refined dam-water
interaction analysis including these factors (Fok and Chopra, 1985) indicates that water
compressibility and reservoir boundary absorption can significantly affect the hydrody-
namic pressures and hence response of arch dams to earthquakes. Rigorous formulation
of dam-water interaction in analyzing arch dams, introduces frequency-dependent hydro-
dynamic terms in the equations of motion that can be interpreted as an added mass, an
added force, and an added damping (Chopra, 1988). The added-mass component of the
dam-water interaction reduces vibration frequencies of the dam, especially for the vibra-
tion frequencies corresponding to the fundamental mode. The reductions in vibration fre-
quencies in turn affect the response spectrum ordinates and hence the effective earth-
quake inertia forces. The added force component of the dam-water interaction increases
the dam response due to the upstream and vertical ground motions, but decreases that due
to the cross-stream ground motion, because the added force for the cross-stream excita-
tion is out-of-phase with the effective earthquake inertia force. The added damping com-
ponent of the dam-water interaction arising from the energy absorption at the reservoir
boundary reduces the fundamental resonant response due to the upstream and vertical
components of ground motion, but slightly increases the response due to the cross-stream
component. The net effect, however, is an overall reduction of the dam response. At
higher excitation frequencies, the added-damping which is dominated by the radiation
energy of hydrodynamic pressure waves propagating in the upstream direction, reduces
the dam response below that for the dam with empty reservoir, for all components of
ground motion.

Finite-element model
Procedures for earthquake response analysis of arch dams including dam-water interac-
tion, water compressibility, and reservoir boundary effects, have been developed (Fok,
Hall, and Chopra, 1986). The finite-element models of the dam and the foundation rock
for such analyses are identical to those described previously. The finite-element model of
the impounded water, however, consists of a finite-region adjacent to the dam and a uni-
form channel of infinite length (Fig. 11-6.16b). The finite-region is idealized as an as-
semblage of fluid elements and can be developed to match the reservoir topography. As
a minimum, it should include three element layers and extend upstream a distance equal
to the water depth. The infinite region is discretized using infinitely long subchannels
coupled to the finite region at a transmitting boundary.

11-124
Reservoir boundary absorption

A hydrodynamic pressure wave impinging on the reservoir boundary is partly reflected

into the water, and partly refracted (absorbed) into the boundary materials. The partial
absorption at the reservoir boundary is approximately represented by a reflection coeffi-
cient known as "α", which is the ratio of reflected to incident wave amplitudes. The re-
flection coefficient α varies between 1 and -1, where α = 1 represents a nonabsorptive
(rigid) boundary with 100% reflection, α = 0 corresponds to a complete absorption with
no reflection, and α = -1 characterizes 100% reflection from a surface with an attendant
phase reversal.

Parametric studies (Fok and Chopra, 1985) of the dam-water interaction indicate that the
earthquake response of arch dams, particularly due to vertical and cross-stream compo-
nents of ground motion are sensitive to the values of α. If the reservoir boundary materi-
als are relatively soft (small α's), an important fraction of the reservoir water energy can
be absorbed, leading to a major reduction in the dynamic response of the dam. Therefore,
the values of α for the design and safety evaluation of dams subjected to earthquake
loading should be measured or selected conservatively. Experimental procedures for
measuring α in-place (Ghanaat and Redpath, 1995), have resulted in values of α ranging
from -0.55 to 0.66 at seven dam sites (Table 11-6-1). The results indicate that the reflec-
tion coefficient of the bottom materials could vary significantly for different sites. An-
other important finding of the study was that α values for the alluvium, silt, and other
sedimentary material at the reservoir bottom could markedly differ from those for the
reservoir side walls consisting of rock. In this situation, two values of α should be used,
one for the sediment materials at the bottom and another for the rock on the side walls.

Because the value of α can effect the dam’s response, a parameter sensitivity study will
be required when α is assumed to be lower than 0.8. Conservative values of α shall be
assumed.

11-125
 Table 11-6.1 Measured reflection coefficient for bottom sediments and rock at seven
concrete dam sites (Ghanaat, Redpath 1995)

Dam Name Bottom Material α

Folsom Bottom sediments with trapped gas, such as that from organic -0.55
matter

Pine Flat Bottom sediments with trapped gas, such as that from organic -0.45
matter

Hoover Bottom sediments with trapped gas, such as that from organic -0.05
matter

Glen Canyon Sediments 0.15

Monticello Sediments 0.44
Glen Canyon Rock - Jurassic Navajo sandstone 0.49
Crystal Rock - Precambrian metamorphic rocks 0.53
Morrow Point Rock - biotite schist, mica schist, micaceous quartzite, and 0.55
quartzite

Monticello Rock - sandstones interbedded with sandy shale and pebbly 0.66
conglomerate

Hoover Rock - Canyon walls 0.77

11-6.5.2 Dam-Foundation Interaction

Interaction of the dam with the foundation rock leads to an increase in vibration periods,
primarily due to flexibility of the foundation rock. Dam-foundation interaction also de-
creases the dam response if damping arising from material damping in the foundation
rock and the radiation damping associated with wave propagation away from the dam are
considered in the analysis. These effects of dam-foundation interaction depend on the
foundation flexibility (Chopra and Tan 1996). As the foundation rock becomes more
flexible, radiation damping increases and vibration periods elongate further.

In practice, dam-foundation interaction effects are typically represented by a "standard"

massless foundation model, in which only flexibility of the foundation rock is considered
but its inertia and damping are ignored. A rigorous treatment of dam-foundation interac-
tion is represented by the foundation impedance matrix (i.e. frequency-dependent stiff-

11-126
ness matrix), which is computed for an unbounded homogeneous foundation rock region
assuming to contain an infinitely long canyon of uniform cross-section. A parameter
study of Morrow Point Dam with the standard and refined dam-foundation interaction
models (Chopra and Tan 1996), indicates a substantial reduction in dynamic stresses if
inertia and damping effects of the foundation rock are included. The results of this study,
when combined for all three components of the earthquake ground motion, show that the
standard foundation model overestimates maximum dynamic stresses by 6%-to-10% for
Ef /Ec = 2, by 12%-to-22% for Ef /Ec = 1, by 29%-to-46% for Ef /Ec = 1/2, and by 45%-
to-78% for Ef /Ec = 1/4. It should be noted that Ef /Ec = 1/2 and 1/4 represent extreme
cases of highly fractured rock, which even if existed at a site, do not normally extend
beyond shallow depths. The refined dam-foundation interaction model ignores this con-
dition and assumes the same low modulus ratio for the entire unbounded foundation rock
region. Another factor is that the results may depend on characteristics of the earthquake
ground motion and on the assumption of uniform earthquake input. These are some of
the important issues that require further studies, before this unconservative dam-
foundation interaction can be used in practice. The FERC will welcome any attempt to
account for dam foundation interaction more accurately, however it must be realized that
whenever more parameters are required for an analysis, more uncertainty is introduced.
This uncertainty must be covered by higher safety factors or parameter sensitivity studies.
The uncertainties discussed above tend to make the continued use of standard massless
foundation model attractive.

11-6.5.2.1 Size of Standard Foundation Rock Model

The standard foundation rock model for dynamic analysis is identical to that for the static
analysis (Section 11-5.2.1.2). It is developed on semi-circular or rectangular planes cut
into the canyon walls at the dam-foundation contact surface, with each plane oriented in
the upstream-downstream direction as moving from the base to the dam crest. In static
analysis, size of the foundation model is selected on the basis of static deflections and
stresses, whereas in dynamic analysis it is selected such that the vibration frequencies and
mode shapes of the dam are computed more accurately. Vibration frequencies of the dam
decrease as the size of flexible foundation model increases (Clough et al. 1985; Fok and
Chopra 1985), but the reduction diminishes beyond a certain size depending on the
modulus ratio of the foundation to the concrete (Ef /Ec). Similar to the static analysis, a
foundation model extending one dam height in the upstream, downstream, and downward
directions (Rf = dam height, for semi-circular foundation model) is also adequate for dy-
namic analysis when Ef /Ec is equal or greater than 1. For more flexible foundation rock

11-127
with Ef /Ec ratios equal to 1/2 and 1/4, the foundation model should extend 1.5 and 2
times the dam height in all directions, respectively.

11-6.5.2.2 Effects of Foundation Modulus

Dam-foundation interaction reduces natural frequencies of the dam and affects mode
shapes primarily due to foundation flexibility. The effects of foundation flexibility on the
natural frequencies and mode shapes increases as the modulus ratio Ef /Ec decreases. The
variation of the natural frequencies for the five lowest modes with the modulus ratio is
shown in Fig. 11-6.17. The natural frequencies are normalized with respect to their val-
ues for a rigid foundation. This figure shows that the natural frequencies are more sensi-
tive to modulus ratios of less than 1, and that they reduce significantly below their values
for a rigid foundation as Ef /Ec becomes less than 1/2; this trend is also true for the higher
natural frequencies.

Foundation modulus affects higher mode shapes more than the fundamental symmetric
and anti-symmetric mode shapes. The change in mode shapes increases as the modulus
ratio Ef /Ec decreases; modulus ratios of 1/2 and smaller may even produce mode shapes
that are considerably different from those for a rigid foundation. Thus, for a more flexi-
ble foundation rock with the modulus ratio less than 1, it is important that the foundation
modulus be determined more accurately by conducting field investigations or performing
sensitivity studies that considers the lowest possible modulus values.
1.00

0.95

0.90
Frequency Ratio

Mode-1
0.85 Mode-2
Mode-3
Mode-4
0.80
Mode-5

0.75

0.70
0 1 2 3 4 5
Modulus Ratio Ef/Ec

Fig. 11-6.17 Variation of five lowest frequencies of dam-foundation rock system with
modulus ratio of foundation to concrete.

11-128
Variation of foundation modulus with abutment also affects natural frequencies and espe-
cially mode shapes of the dam. These effects are most significant when the modulus ratio
of the weaker abutment to concrete is less than 1/2, but can be ignored, as modulus ratio
of the weaker abutment approaches unity, or when modulus ratios for both abutment are
greater than unity. In these cases, variation of foundation modulus with abutment can be
ignored. Only the modulus of the weaker abutment may be used and applied to the entire
foundation rock model.

Variation of foundation modulus with elevation affects the natural frequencies and mode
shapes in the same manner discussed for static deflections and stresses in Section 11-
5.5.1. More flexible upper abutments tend to reduce natural frequencies and affect mode
shapes, especially for the higher modes. These effects, however, are much less than for
those cases where one abutment or the entire foundation region comprises of low
modulus materials as discussed previously. The effects of variation of foundation
modulus on frequencies and mode shapes and thus on dynamic response of the dam
should be considered when the region of weaker zones are substantial and their moduli
are less than the modulus of the concrete. When the modulus of the foundation rock in-
cluding weak zones are higher than the modulus of the concrete, the effects of variation
of modulus with elevation can be ignored and the smallest modulus may be used for the
entire foundation.

11-6.5.2.3 Foundation Boundary Condition

The standard massless foundation model contains a volume of foundation rock region
that extents a distance equal to one or two dam heights in the upstream, downstream, and
downward directions. Since wave propagation is ignored in such a model, all nodes on
exterior surfaces of the foundation mesh are fixed in space, as shown in Figs. 11-5.2a and
11-5.3a.

11-6.5.3 Direction of Ground Motions

The seismic input for time-history analysis of arch dams consists of three acceleration
time histories that are applied simultaneously at the fixed boundaries of the foundation
model; in the stream, vertical, and cross-stream directions. Although the vertical ground
motion is always applied in the up-down direction, the direction for the two horizontal
components is not obvious and should be selected carefully. Even if the peak ground ac-
celerations for the two horizontal components are the same, their sustained duration of

11-129
maximum shaking which can affect the number of critical peak stresses may be different.
Fig. 11-6.18 demonstrates that, an example arch dam subjected to three components of
ground motion develops higher number of stress peaks when the more energetic hori-
zontal component of ground motion is applied in the stream direction. Since this is gen-
erally true in most cases, the more energetic component of ground motion should be ap-
plied in the stream direction.
Cantilever stress (psi)

1500
13 s tres s peaks exceed 1000 ps i
1000

500

0
-500
0 2 4 6 8 10 12 14 16 18 20
Time (sec)

a) Energetic horizontal component applied in stream direction.

Cantilever stress (psi)

1500
4 s tres s peaks exceed 1000 ps i
1000

500

0
-500
0 2 4 6 8 10 12 14 16 18 20
Time (sec)

b) Energetic horizontal component applied in cross-stream direction.

Fig. 11-6.18 Time histories of cantilever stress on upstream face of dam

11-6.6 Post-Earthquake Safety Evaluation

A post-earthquake safety evaluation is required to assure the safety of the dam if: 1) a
maximum credible earthquake should occur near the dam site, and 2) the predicted per-
formance of the dam due to a postulated MCE event should indicate substantial damage.

Should an intense earthquake occur near any dam, a very detailed examination should be
made as soon as possible to evaluate the extent of damages that may have resulted. Dur-

11-130
ing the post-earthquake inspection, the galleries and downstream face of the dam should
be examined for evidence of new cracks and of working of existing cracks and monolith
joints. Other visible damages to look for include vertical offsets or settlement, horizontal
movement of the crest, and joint separation within the body of the dam and at the inter-
face with thrust block(s). The abutments should be carefully examined for evidence of
cracking, rockfalls, and rock mass sliding. Also, detailed inspections should be made for
damage to appurtenances such as gates and their lift mechanisms, intake and elevator
towers, and transmission towers and substation equipment at the site. In addition to the
detailed inspections of the dam system and its abutments, measurements from the posi-
tion indicating system (if installed) should be carefully studied for possible evidence of
permanent displacements caused by earthquake, because such permanent changes may be
the best indicators of any significant internal damage done to a dam or its foundation.
The change in leakage past the gates, or into the galleries and shafts, and a change in the
seepage through the abutments should be noted and examined.

In the case of safety evaluation of an existing dam, even though substantial damage is
permitted under the MCE excitation, the inflicted damage should not weaken the dam to
a point where its capacity to resists static loads and aftershocks is threatened. The proce-
dures for post-earthquake evaluation for static loads and aftershock events are discussed
below.

11-6.6.1 Evaluation for Static Loads

If post-earthquake inspection or seismic safety evaluation due to a postulated maximum
earthquake event indicates that the dam has or might suffer substantial damage, the per-
formance of the dam in its damaged condition should be evaluated to assure that the
safety of the dam to resist static loads has not been threatened. This is because the dam-
age inflicted upon the dam might have weakened its capacity to resist sustained static
loads, which require a more demanding performance criteria than the transient seismic
loads.

In these cases, a post-earthquake static analysis considering the effects of cracks is re-
quired. A possible simple approach to account for the effects of cracking is to perform a
FE linear elastic static analysis, as described in Section 11-5.2, with reduced modulus for
the elements affected by cracks. In such analysis, the effects of opened contraction joints
indicating loss of arch action may be simulated by introducing double nodes at the loca-
tion of joint openings. The location and extent of cracks and joint openings should be
determined from the earthquake safety analysis of the dam, or from the post-earthquake
inspections should an actual earthquake excite the dam.

11-131
11-6.6.2 Evaluation for Aftershock Events
Aftershock earthquakes invariably occur after any major quake, but the rate of aftershock
activity decreases rapidly with time after the main shock. Although the aftershocks are
not as intense as the main shock, in some cases they could be severe enough to produce
additional damage and possibly limit the use of the dam. After a major earthquake,
safety assessments of the damaged dams should be a major priority for the dam owners
and the government agencies. Such assessments are critical to determination of the level
of hazard from an aftershock event. Depending on the severity of damage, the conditions
of the dam may be described by a classification system comprising of three categories:
Safe, Limited Usage, and Unsafe. These categories are defined as follows:

• Safe: No apparent hazard found, although repairs may be required. Cracking is

primarily surficial and confined to small regions of the dam and abutments. Con-
traction joints may open and close during the earthquake excitation, but they are
close with minimal offsets after the shaking stops. Original load-resisting capacity
of the dam is not significantly reduced; thus no restriction of water level is re-
quired.

• Limited Usage: Hazardous condition believed to be present due to damage result-

ing from relatively significant cracking and permanent joint separation of the dam,
or possible minor sliding of the thrust blocks or abutment rock masses resisting the
dam thrusts. If necessary, reservoir water level should be lowered to reduce the
possible major aftershock hazard. The continuous usage should be confined to low
water levels until the necessary repairs and strengthening measures have been ac-
complished.

• Unsafe: Extreme hazard condition with imminent danger of collapse from an after-
shock. Such a condition has not occurred in the past, mainly because no dam with
full reservoir has ever been subjected to a maximum credible earthquake. Should
such an extreme case occur, the damage is expected to be widespread with major
cracking, joint opening with offsets, and possible sliding originating along the
cracked sections within the dam, at the dam-foundation interface, or along the
planes of weakness within the abutments. The reservoir water should be lowered
and warning issued for immediate evacuation of the population living downstream.
When a seismic safety evaluation of an existing dam predicts substantial damage due to a
postulated maximum credible earthquake, the performance of the dam to withstand the
major aftershocks should also be evaluated. Any seismic response analysis of a dam for
aftershocks should consider the effects of cracking and other structural damage that might
have weakened the dam during the application of the main shock. When subjected to a

11-132
strong aftershock, the dam may suffer additional damage but its capacity to retain the res-
ervoir water should not be jeopardized. The magnitude of the aftershock selected for the
safety evaluation should be consistent with those observed in the past earthquakes. Table
11-6.2 shows the main shock and the largest aftershock sequence of some recent earth-
quakes.

Table 11-6.2 Main shock and largest aftershock of some recent earthquakes.

Main Largest Time of

Earthquake Shock Aftershock Aftershock
1983 Idaho ML= 7.3 ML= 5.8 6 hours later
1984 Morgan Hill, CA ML= 6.2 ML= 4.5 9 days later
1985 Chile Ms = 7.8
Mb = 6.9 mb = 6.5 52 min. later

1987 Wittier Narrows, CA ML= 5.9 5.3 3 days later

1988 Armenia Mb= 6.3 5.9 5 min. later
1989 Loma Prieta, CA Mw = 6.9 5.0 33 hours later
1994 Northridge, CA Mw = 6.7 5.9 1 min. later
ML= Local Magnitude
Ms = Surface-wave Magnitude
mb = Body-wave Magnitude

11-133
11-7 INSTRUMENTATION

11-7.1 Purpose and Need for Instrumentation

The observed historical behavior of a dam provides the best indication of the dam's future
performance. As such it is essential that existing arch dams be provided with adequate
instrumentation for evaluation of dam safety and for the verification of the analytical models
used to predict their performance under unusual and extreme loading conditions.

The most common instrumentation for arch dams includes measurements of movements,
reservoir and tailwater levels, seepage, uplift pressures, and ambient and internal temperatures.
In regions of high seismicity, it is also important to install earthquake instrumentation system to
record both the earthquake ground motions that excite the dam and the dynamic motions of the
structure that result from this excitation. A complete discussion of the types of instruments
typically used at arch dams is contained in Chapter 9. The following paragraphs are intended to
help staff understand the special requirements in evaluating the instrumentation layout,
frequency of measurements, and evaluation of the collected data at arch dams.

11-7.2 Special Instrumentation Considerations for Arch Dams

The layout for monitoring seepage and uplift at arch dams is the same as other concrete dams. In
general, seepage should be monitored at various elevations along abutment contacts in order to
determine if areas of seepage are influenced by reservoir elevation, loading on the foundation,
ground water, or a combination of these. While uplift may not be an important factor in the
stresses in the body of the dam, it can be extremely important in the stability of the abutments,
and instrumentation should be considered for any areas where potential failure wedges may
form.

Temperature can play a significant role in the behavior of an arch dam. As discussed in Section
11-4.5, the evaluation of arch dams includes either "winter" or "summer" mean concrete
temperatures in each load combination. These concrete temperatures can be measured directly
by embedding instruments in the concrete or can be approximated by using reservoir and air
temperatures. Thermometers, thermocouples, or thermistors embedded at various depths in the
concrete provide the best means of determining the magnitude and timing of the temperature
loads. Embedded instruments should be installed at various elevations near the center of the dam
cross-section and near each face to obtain a reliable approximation of the maximum and
minimum concrete temperatures. The primary advantage to direct measurement of concrete
temperatures is that it eliminates the uncertainty in estimating concrete temperatures based on

11-134
reservoir and air temperature data. The primary disadvantage to embedded instruments is the
cost of installing the instruments and taking regular measurements.

When using reservoir and air temperatures to estimate the concrete temperature, the procedures
discussed in section 11-4.3 should be followed. It should be clearly understood that the timing of
the peak average concrete temperatures would not coincide with the timing of the peak ambient
conditions. It will usually take the mass concrete one or two months to respond to the ambient
conditions depending upon, among other factors, the thickness of the dam and the thermal
properties of the concrete. For example, if minimum air temperatures typically occur in mid-
January, the minimum mean concrete temperature could occur sometime between the first of
February to the end of March. In other words, the "winter" temperature loading could be the
concrete temperatures that occur in early spring. This fact can be extremely important when
establishing the timing of measurements or when determining reservoir levels occurring at usual
minimum and maximum concrete temperatures.

In measuring deflections in arch dams, the instrumentation points should be established at the
crown cantilever and near the quarter points as seen in Figure 11-7.1. If the dam is slightly non-
symmetrical, then the points should be set at the crown cantilever and at points mid-way between
the crown cantilever and the abutment contacts. For highly non-symmetrical dams, additional
points should be added at equal intervals along the longest side of the crest.

11-7.3 Frequency of Measurements

The frequency of measurements should generally agree with the monitoring schedules in Chapter
9. As noted in Section 9-6, even though existing dams have generally reached equilibrium with
imposed loads, baseline data must be obtained to compare with subsequent measurements. In
any instrumentation program it is especially important that the range of extremes be captured.
As such, it is preferable to take monthly measurements until the adequacy of the equipment,
personnel, and procedures are known and all patterns can be clearly identified. This may take 2
to 3 years of monthly measurements before the frequency can be reduced to quarterly.

Once the timing of the measured extremes has been clearly defined, quarterly measurements can
be scheduled which will capture those extremes each year. The quarterly measurements of
deflections will usually correspond to the times of minimum and maximum concrete
temperatures and/or minimum and maximum normal reservoir elevations. If the reservoir
variation does not sufficiently affect the deflections of the dam, then the frequency of the
measurement could be extended to semiannual at the times of minimum and maximum seasonal
extremes, with additional measurements taken whenever an unusual reservoir level occurs (for
example, very low water levels during droughts).

11-135
11-7.4 Presentation of Data and Interpretation of Readings

Deflection data for arch dams should always be converted to radial and tangential deflections as
shown in Figure 11-7.1. Terms such as “North and South”, “Upstream and Downstream", etc.
are not appropriate for a curved structure such as an arch dam and should not be used.

Instrumentation data should be plotted with respect to time. Reservoir elevation should be
included on the plot so that linkage between reservoir elevation and instrument output can be
observed. Examples are shown in Figure 11-7.2 and 11-7.3. Since temperature plays a major
role in dam behavior, it may also be appropriate to include temperature data on the plot. The
radial deflections should be seasonal with maximum upstream deflection occurring at the time of
maximum mean concrete temperatures and the maximum downstream deflection occurring at the
time of minimum mean concrete temperatures. Deflection data that doesn't match these trends
could indicate that the accuracy or precision of the instrumentation program is not adequate.

When plotting instrumentation data the scale used in the plots is extremely important. Trends
and relationships can be lost if the scale not appropriate for the data shown.

11-7.5 Comparison of Predicted and Measured Deflections

Once staff has confidence that the deflection measurements are of sufficient accuracy and
precision and are representative of the actual behavior of the dam, the next step is to compare the
measured deflections with those predicted by the analytical methods discussed earlier in this
chapter. The purpose of these comparisons is to gain confidence that the mathematical model is
an accurate predictor of structural behavior. If there is a credible comparison of predicted and
measured deflections under normal loading conditions then there is additional confidence that the
mathematical model may be able to predict the behavior of the dam under flood and seismic load
conditions.

However, caution should be exercised in attempting to require an exact comparison between the
mathematical model and the measured data. An exact comparison is not always realistic. Among
other problems, there are usually limits to the accuracy and precision of the measurement
techniques. There can also be problems associated with gaps in data or changes in equipment or
personnel. In addition, while the structure response is nonlinear the mathematical models
usually assume linear elastic behavior. Unless there are numerous thermometers throughout the
dam and unless temperature and deflection measurements are made daily, it is doubtful that the
exact loading conditions will apply to both the measured and predicted deflections.

11-136
These limitations do not preclude making a comparison that can be reasonable and useful to the
dam safety engineer. For example, an arch dam that has survey points at the crest of the dam
but has no embedded thermometers to measure concrete temperature. Assume that the historical
data collected indicates that the instrumentation procedures are of sufficient accuracy and
precision and all trends are consistent with expected behavior. Also assume these points are
measured in the late winter (February) and late summer (August) and the range of annual
deflection is typically in the order of 0.9 inches. If a finite element model of this dam predicts a
winter-to-summer deflection of 1.2 inches it could be concluded that the model is a reasonable
predictor of the dam's behavior. However, if the finite model predicts deflections of 2.0-inches,
then the material or loading assumptions used in the model may need to be re-evaluated.

11-7.6 Long Term Instrumentation Performance

The life span of embedded electronic instruments is finite. It is to be expected that the output
from imbedded temperature sensors, strain gauges, and piezometers will become unreliable over
time, and instruments will eventually have to be abandoned. For this reason, instrumentation
plans should include redundancy to attempt to offset the attrition that will occur. It is also
important to obtain sufficient information early, while the majority of instruments are still
functioning.

7.7 Interpretation of Data

The gathering and recording of instrumentation data is of little value if it is not reviewed and
interpreted by personnel who understand the implications of the data. This review should be
timely, so that instrument readings that are unexpected can be quickly double checked. Of
prime importance is the discernment of trends in the data over time. The absolute values of the
readings are of lesser value. For this reason, it is important to avoid changes is reading and
recording procedures. If readings from an instrument change from what is expected, this should
be noted.

The readings from various instruments should be correlated. For example, does an increase in a
piezometer reading correspond to a movement? Does temperature effect gallery drain flow? It
is only through the careful interpretation of the readings from all instruments that the behavior of
an arch dam is properly understood.

11-137
 Fig. 11-7.1 Typical layout for a symmetric arch dam

Fig 11-7.2 Deflection plot with reservoir elevation included.

11-138
Fig. 11-7.3 Piezometer data

11-139
11-8 HISTORIC DAM INCIDENTS

11-8.1 Overview
There is no comprehensive listing of arch-dam failures. However, available listings are
contained in more than one publication. Babb and Mermel (1968) listed 600 dam
incidents (including failures) of which only seven involved arch dams, 2 involved
multiple arch dams, and 2 involved gravity-arch dams. The dams they list are as follows:
Arequipa (Peru, 1965); Matilija (U.S.A., 1965); Vajont (Italy, 1963); Malpasset (France,
1959); Moyie River (U.S.A., 1938); Alla Sella Zerbino; Allessandria (Italy, 1935); Lake
Lanier (U.S.A., 1926); Gleno (Italy, 1923); Lake Hodges (U.S.A., 1918); Manitou
(U.S.A., 1917); and Tolla (France, 1892).

The International Commission on Large Dams (ICOLD) Bulletin 99 (1995) lists seven
arch dam failures, three of which are also listed by Babb and Mermel (Malpasset, Gleno,
and Moyie River). The other four arch dam failures listed by ICOLD are: Leguaseca
(Spain, 1987), Meihua also called Plum (China, 1981), Idbar (Yugoslavia, 1960); and
Vaughn Creek (U.S.A., 1926).

In addition, in a paper entitled “Lesson from Serious Incidents at Seven Arch Dams”
presented at the 1997 annual conference of the Association of State Dam Safety Officials,
G.S. Sarkaria described incidents at Le Gage, El Fryle, Koelenbrein, and Zeuzier dams.

Following is a summary of the historical incidents and failures involving arch dams:

Arequipa, Peru. This thin-arch concrete dam failed in 1965 as a result of fractures
caused by a vibrating penstock which passed through the dam. Inflow was normal at the
time.

Matilija, California. The concrete dam was completed in 1949 and was a combination of
gravity and arch structure. It was 163 feet high with a crest length of 620 feet. In 1965
the dam was judged to be unsafe as a result of deterioration of the concrete due to
expansive aggregate. Foundation conditions were also judged to be poor. The reservoir
was drained, the dam was eventually demolished, and the site was submerged by a new
dam downstream.

Vajont, Switzerland. This thin-arch concrete dam, which is 905 feet high, was
overtopped by a huge landslide-generated wave. Inflow to the reservoir was normal at
the time. The resulting overtopping was estimated to be as much as 300 feet. The dam

11-140
itself suffered little damage, but the reservoir was a total loss. The resulting flood caused
great loss of life in the downstream areas. A detailed description of failure of Vajont
Dam is provided in Section 11-8.2.1.

Malpasset, France This thin-arch concrete dam, which was 218 feet tall, failed due to a
movement of the left abutment in December 1959. The movement was thought to be
due to sliding on a rock wedge formed by intersection of a fault with gneissic foliation in
the rock of the left abutment. The principle cause of the failure was not directly due to
the passage of a flood in that the dam was never overtopped. However, a very large flood
was being passed when the failure occurred. The official death toll was 396 people killed
in the ensuing flood, which suddenly struck the village of Frejus. The dam was a
complete loss and is further discussed in Section 11-8.3.

Le Gage, France This Very thin 150’ high arch dam developed extensive cracking on
both faces of the dam after first filling of the reservoir in 1955. Cracking continued to
worsen for the next 6 years. After the failure of Malpasset dam, Le Gage was abandon
and a new thicker arch dam was constructed upstream.

El Fryle Dam, Peru This 200’ high arch dam experienced a major slide on one of the
abutments during filling. The dam did not collapse. A concrete thrust block abutment
was constructed and the dam was saved.

Moyie River, Idaho This 53-foot high concrete arch dam, located on the Moyie River,
was approximately 64 feet thick at the base and 24 feet wide at the crest. During passage
of a major flood in 1926, the spillway, which was located on one abutment, was
undermined. The erosion completely washed out one of the abutments. The abutment
was replaced and the dam is still in use.

Alla Sella Zerbino, Allessandria, Italy This concrete arch-gravity structure was only 39
feet high with a crest length of 262 feet and a reservoir capacity of 14,000 AF. It failed
on August 13, 1935, as a result of overtopping and sliding on its foundation. One
hundred lives were lost.

Lake Lanier, North Carolina This constant-radius concrete arch dam was constructed in
1925. It had a thickness of 12-1/2 feet at the base and 1 foot at the top. It was 62' high
with a crest length of 236 feet. One of the abutments (cyclopean masonry) washed out as
a result of the failure of soft rock in the abutment on January 21, 1926. The remainder of
the dam was unharmed.

11-141
Gleno, Italy This multiple-arch concrete dam contained 25 arches for a total length of
250 feet. Concrete gravity sections made up the ends of the dam. Total crest length was
863 feet. The dam was 143 feet high. It was completed in 1923 and failed on December
1, 1923 only 30 days after filling. Nine arches fell due to a poor masonry base. Some
600 persons lost their lives.

Lake Hodges, California This multiple-arch concrete dam was completed in 1918 by the
City of San Diego. It was 136 feet high with a crest length of 616 feet and a reservoir
capacity of 33,550 AF. The dam was damaged by cracked piers in 1918 but did not
completely fail.

Manitou, Colorado This concrete arch dam was 50 feet tall with a crest length of 300
feet. A portion of the dam failed in 1924 due to deterioration of the concrete.

Tolla, France This very thin arch dam was 295' high with a crest length of 435 feet.
Owned by Electricite DeFrance, the dam experienced severe cracking. It was buttressed
in response. Cracking may have been the result of large temperature stresses.

Koelnbrein, Austria Cracks and substantial leakage appeared in the lowest foundation
gallery when the reservoir was 80% full two years after first filling. Full uplift pressure
was observed over the entire base in the central portion of the dam. Major repair was
undertaken between 1989 and 1994.

Zeuzier, Switzerland The dam behaved satisfactorily for 20 years, then began to deflect
upstream due to riverward movement of the left abutment.

Leguaseca, Spain This concrete multiple-arch dam, which was 66 feet high with a crest
length of 230 feet and held a reservoir volume of only 16 AF, was constructed in 1958
and failed in 1987. The dam body failed structurally, apparently due to deterioration due
to both aging and the effects of freezing and thawing. No details of the failure are given
in the literature.

Meihua (Plum), China This experimental masonry arch dam was 72 feet high and had a
crest length of 211 feet. It was completed in 1981 held a reservoir of only 93 AF. It
failed shortly after filling in 1981. The dam failed as a result of structural failure due to
excessive uplift movement along a peripheral joint as described in Section 11-8.3.2. The
scheme was abandoned after failure.

11-142
Idbar, Yugoslavia This concrete arch dam was 125' high with a crest length of 354 feet.
It was completed in 1959 and failed in 1960. Failure was during first filling and resulted
from piping and erosion of the foundation.

Vaughn Creek, USA This concrete arch dam was 62' high with a crest length of 312
feet. The dam was completed in 1926 and failed during first filling. Seepage and poor
materials in the dam caused failure.

The above incidents indicate that safety of an arch dam can be threatened by overtopping
due to major floods and landslides, abutment sliding, erosion of foundation-abutment
rock, and the deterioration or poor construction materials. They also show that small,
thin arch dams could be susceptible to vibrating appurtenant structures such as a penstock
passing through the dam, and that thermal cracks caused by nonuniform temperature
distribution could result in failure of thin arch dams. Significant landslide and abutment
failure cases are described in Section 11-8.2 and 11-8.3, respectively. A notable case of
severe overtopping and damage to stilling basin caused by the passage of extreme floods
is provided in Section 11-8.4. Although no arch dam has ever reported to fail from
earthquake ground shaking, performance of several arch dams, and in particular Pacoima
Dam, under significant earthquake excitation are described in Section 11-8.5. The effects
of alkali-aggregate reactions (AAR) and a summary of case histories of concrete arch
dams with AAR problems are given in Section 11-8.6.

11-8.2 Landslide Case

11-8.2.1 Vajont Dam

The Vajont Dam, constructed between 1957 and 1960 is located on the Vajont River in
northern Italy near the towns of Longarone, Pirago, Casso and Erto. The dam is a 276
meter (905 ft.) high, double curvature, thin arch dam. On 9 October 1963 during reservoir
filling a catastrophic landslide movement occurred suddenly over a 2 km (1.2 mile) reach
of the southern or left bank of the reservoir. According to Muller (1987) the slide mass
consisted of a volume of 275 million cubic meters (360 million cubic yards), which
generated a wave 260 meters (853 feet) high. Hendron and Patton (1985) describe a
wave which crested 100 meters (328 feet) above the top of the dam and had a height of
70 meters (230 feet) downstream at the confluence of the Vajont and Piave Rivers. More
than 2,000 people lost their lives in this catastrophe. Longarone and Pirago were the
towns most severely affected. The dam structure itself survived the overtopping by the
wave of water and the impact of the load of earth placed against it by the landslide, thus

11-143
providing an excellent example of the structural strength of an arch dam.

11-8.2.1.1 Geologic Conditions Contributing to Failure

The dam is located in the Dolomite Region of the Italian Alps in a narrow valley with
steep side slopes. Jurassic and Cretaceous age limestone are the predominant rock types
represented in the valley walls and in the landslide mass. Many have studied the
geology. Hendron and Patton (1985) include as Appendix G "Synthesis of Geological
Studies of the Vajont Landslide" by Edoardo Semenza (1966-67) which gives a clear
discussion of the details of the geology.

The dip of the formations generally controls the configuration of the landslide surface,
which forms a somewhat chair-like structure described by Muller (1987) with the upper
portion dipping rather steeply toward the valley while the lower or seat portion flattens
into a more nearly horizontal configuration. Semenza (1966-67) offers evidence that the
landslide mass had moved previously in geologic time and that it had previously blocked
the valley. During an investigation in 1959 he and F. Giudici discovered an uncemented
mylonitic zone extending 1.5 km (0.93 miles) along the left side of the valley. They also
concluded that a rock mass on the right side of the valley was a remnant of a prehistoric
landslide. Semenza thinks that the gorge that existed when the dam was constructed had
been eroded through the old landslide mass by the Vajont River. Hendron and Patton
(1985) also concluded that the 1963 slide was a result of the reactivation of an old slide.

Semenza (1966-67) described the rock mass as being intensely fractured and faulted and
containing solution cavities and sinkholes. These features would likely provide easy
access for water to infiltrate the rock mass. Hendron and Patton (1985) found during
their investigation that interbeds of clay existed in the formations adjacent to the slide
planes. These clay beds could be the source of the low resistance to shear that was
needed to explain the initiation of the landslide.

11-8.2.1.2 Stability Analysis

Several attempts were made to explain the landslide by using back analysis as a means of
developing data on the many factors involved in the landslide development. Hendron and
Patton (1985) provide a very plausible analysis and description. Their work on the clay
beds that were discovered adjacent to the failure surface provides strong evidence that
the low shearing resistance of the clay may have provided a path for development of the
landslide shear planes. They concluded that the peak strength of the failure plane

11-144
materials was lost during prehistoric slide movements and that the residual strength of
the materials was appropriate for use in back analysis.

Hendron and Patton (1985) also compared the rate of measured movement that occurred
in the slide mass during the years immediately preceding the failure with precipitation,
reservoir level, and water in piezometers. This analysis lead them to the conclusion that
rainfall rate was a major factor in the activity of the slide. Where others had concluded
that reservoir filling levels governed the rate of movement, they demonstrated that rate of
precipitation was more closely associated with the rate of movement measured in the
slide. The sinkholes mentioned above, as well as the intensely fractured condition of the
rock, both could provide ready access for rain water to enter the rock mass.

11-8.2.1.3 Response of the Arch Dam Structure

According to Muller (1987) the thin arch dam resisted the forces imposed by the
landslide failure and suffered only minor damage. The crest of the dam was the only
portion damaged and that was not severe. In summarizing the August 1985 Purdue
University Workshop on Dam Failures, Dr. G.A. Leonards (1987) stated that it had been
reported that the Vajont Dam withstood a load eight times greater than it designed to
withstand. In this same workshop Dr. J. Laginha Serafim (1987) noted that the dam
survived the impact and overtopping of an asymmetrical wave of water and stones that
subjected the structure to more than double the design load. This provides an excellent
demonstration of the strength of a well-designed arch dam.

11-8.2.1.4 Lessons Learned

This catastrophic landslide failure demonstrated to the dam building profession the
importance of performing detailed geologic investigations of the rim of narrow steep
walled valleys which are planned as the reservoir for large dams. These studies should
focus on locating and analyzing existing or potential landslides that could slide with
sufficient volume and velocity to displace a wave of water that could endanger lives or
property.

The failure mechanism of a large landslide mass such as this is very complex and
difficult to evaluate as the failure is progressing. Even the leading experts may fail to
reach precisely correct conclusions if they do not fully understand all of the factors
affecting the landslide. In this case, the failure to realize that rainfall was more a factor
than reservoir level in the rate of movement of the slide mass led the decision makers to
think that the landslide could be controlled by adjusting the level of the reservoir and the

11-145
rate of filling. When dealing with a slow moving landslide it is imperative that all
significant factors affecting its movement be identified and considered in all decisions
regarding stabilization or control.

11-8.3 Abutment Failure Cases

11-8.3.1 Malpasset Dam

The Malpasset Dam located on the Reyran River in France, was a double curvature, thin
arch dam with a maximum height of about 197 ft. (60 m) and a crest length of about 732
ft. (223 m) according to Londe (1987). Its thickness varied from 22.25 ft. (6.78 m) at the
base to 4.92 ft. (1.5 m) at the crest. The dam failed in 1959 after a slow initial filling
period which Londe says took about 5 years and was within about one ft. (30 cm) of the
spillway crest at the time of failure.

The failure occurred due to movement within the left abutment foundation which caused
the dam to rotate about the right abutment and ultimately collapse. According to Londe
(1987) the failure occurred as a sudden movement. He says that 30 minutes before the
failure a workman was painting the crest and observed nothing unusual to warn of the
impending failure.

According to Serafim (1987) and Londe (1987), Malpasset was the first failure of a thin
arch dam. Its failure resulted in the destruction of the city of Frejus, France, with much
loss of life and property. The design of the dam was under the supervision of Andre
Coyne, who was considered one of the most eminent dam designers in the world.

11-8.3.1.1 Geologic Conditions Contributing to Failure

The following description of geologic conditions is a synopsis based upon the description
presented by Londe (1987).

The dam site is located in a banded gneiss formation in the Tanneron Massif. The dip of
the foliation is between 30 and 50 degrees in a downstream direction and toward the right
abutment. The left abutment, which failed, tends to be schistose, as does the lower part
of the right abutment. The foliation/schistocity is due to layers of micaceous minerals
and fine networks of microscopic openings near the layers of mica. The joint system is
composed of three different patterns or sets. The spacing is very close, varying from 2-3
cm to as much as 20-50 cm. They are very tight and irregular in shape There are
numerous shears and faults and they tend to be oriented parallel to the three joint sets.

11-146
They are characterized by breccia and mylonite.

The left abutment failure wedge was formed on its downstream side by a fault dipping 45
degrees upstream and striking east-west. The strike of the fault is parallel to the chord of
the arch and is symmetrical with regard to each abutment. The upstream portion of the
wedge was formed by a stepped pattern of sheared planes that followed the foliation
planes in the gneiss. The foliation planes dip toward the river on the left abutment and
are not symmetrical on each abutment. The existence of the fault was not known at the
time of design. It was located from 49 ft. (15 m) to 131 ft. (40 m) below the foundation
and exited at the surface beneath overburden some 65 ft. (20 m) downstream.

The rock was very impervious and when pressure tested with water the average take was
less than one Lugeon unit. It was consequently not thought to require pressure grouting
except in the interface zone between the concrete and the foundation. After the dam
failed, detailed laboratory tests performed on the rock revealed that its permeability
decreased markedly under compressive stress. This became a major factor in the theory
developed to explain the failure.

In-situ tests of the rock mass modulus of deformation revealed a foundation with a low
elastic modulus which averaged 180,000 psi (1260 MPa). Intact samples tested in the
laboratory had much higher values, ranging from 1,214,300 psi (8500 MPa) to 8,571,000
(60,000 MPa). This demonstrates the importance of either determining the modulus of
deformation by in-situ tests or reducing the laboratory values from testing intact samples
by an appropriately determined reduction factor.

Unconfined compression tests performed on intact laboratory samples gave average

values of 8285 psi (58 MPa) for dry specimens and 6070 psi (42.5 MPa) for saturated
specimens. The test results, however, were very widely scattered.

After the failure occurred, a laboratory testing program was performed to evaluate the
extent of stress concentration at depth in the foundation taking into consideration the
discontinuities in the rock mass. Photoelastic modeling techniques were used. This
confirmed that the load applied by the dam could induce high stress at considerable depth
in the foundation concentrated along discontinuities in the rock mass. This was
significant in providing one of the factors necessary to the development of the theory of
failure.

11-147
11-8.3.1.2 Theory of Cause of Failure

Londe (1987) provides a theory of failure that satisfies all of the facts revealed during the
extensive investigations that followed the event. Following is a synopsis of the
explanation of failure provided in the referenced paper.

A large wedge of rock was removed in the left abutment by the failure. As stated above,
this wedge was bounded on the downstream side by a fault striking parallel to the chord
of the arch and dipping upstream at about 45 degrees. The upstream side of the wedge
was bounded by shears parallel to the gneissic foliation of the rock. There seems to be
little doubt that the failure occurred due to sliding of the rock in this wedge. The difficult
question to answer was how the forces involved could be resolved to explain the
movement that had occurred within this wedge.

The thrust from the arch on the left abutment was almost parallel to the foliation in the
rock. Because of this parallel alignment, the stress distribution and dissipation with
increased depth normally expected in design was unable to develop in the left abutment.
This left it concentrated in a segment of the rock mass and essentially undiminished with
depth which allowed it to concentrate at the downstream fault. Because this parallel
alignment with the foliation did not exist on the right abutment, stress in that rock mass
was able to dissipate with depth as is normally expected.

As the reservoir filled, the stress in the rock mass in the left abutment increased
abnormally and the unusual sensitivity of the gneiss to decreased permeability with
increased stress came into play. This resulted in very high uplift pressures that may have
approached full reservoir head. This uplift pressure provided an upward force which
contributed significantly to movement of the rock wedge that ultimately failed. As the
rock wedge began to move under the combined forces of the arch and the uplift, the
upstream foundation crack frequently associated with arch dams began to enlarge and
increase the uplift pressure downstream. At some point the rock in the wedge was no
longer able to resist the forces and was lifted out of the foundation. As this block of rock
began to move, the thrust of the dam was translated up the abutment and had to be carried
by the abutment thrust block and wing wall. The loads were so great that they could not
restrain them. The left abutment failed and the dam rotated around the right abutment
which had not failed. This resulted in sudden collapse of the dam and release of the water
in the lake.

Wittke and Leonards (1987) report on a finite element analysis of the interaction of the
dam and its foundation which supports the mechanism of failure reported by Londe

11-148
(1987) and described above, but provides some additional insight into the rock mechanics
logic involved in the explanation. Their analysis indicates that a large stress reduction
takes place perpendicular to the foliation on the upstream side of the wedge due to water
forces on the upstream face of the arch dam and to the water seepage forces under and
around the structure. This results in increased permeability parallel to the schistocity and
high seepage pressures in the very adverse upward direction along the upstream side of
the failure wedge. Ejection of the rock wedge would occur as explained by Londe (1987)
but this additional analysis indicates that the friction angle resisting movement on the
fault may have been as low as 15 to 20 degrees where Londe thought it to be about 30
degrees.

11-8.3.1.3 Lessons Learned

Abutment analysis must include 3 dimensional analysis of any rock wedges that may be
capable of failure under the loads applied by the dam thrust and under uplift forces
created by seepage from the reservoir.

Detailed geological investigations must be conducted in the abutments to identify all

adversely oriented rock discontinuities.

Very complex and unique conditions may exist in the rock that forms the abutments that
may not be readily evident. For example the unexpected reduction in rock mass
permeability under increasing load that resulted in the development of an underground
dam in the left abutment at Malpasset and resulting high uplift downstream of the dam.

Foundations of arch dams should be drained to reduce uplift forces.

Instruments should be provided in the abutments to monitor uplift pressure and

deflections in the foundation.

The existence of an upstream crack near the heel of an arch dam is now generally
accepted.

11-8.3.2 Experimental Plum Dam

Plum Dam, located in Fujian Province in south-east China, was an experimental

cylindrical arch dam with a height of 72 ft (22 m) and a crest length of about 238 ft (72.6
m). The dam failed in September 1981 shortly after it was completed in May of the same
year. Since the dam was an experimental structure built at a coastline site, its failure
caused negligible property damage and no loss of life.

11-149
Field investigations indicated that the failure occurred possibly due to the upward and
downstream sliding of the dam along a peripheral joint. This mode of failure and the dam
geometry is further described in Section 11-5.6.1 and Figs. 11-5-23 and 11-5.24. The dam
was built as a masonry structure composed of granite blocks in the main body of the dam
and included a peripheral joint between the dam and its artificial concrete abutment. The
joint surfaces were coated with bitumen and polyvinyl chloride was used to seal the joint.

CRACK

JOINT

Fig. 11-8.1 Upward sliding of Plum Arch Dam along its artificial flat abutment.

Completed in May 1981, full storage of the reservoir was reached in June 1981, and the
dam was overtopped by 1 foot over the crest on July 20. The dam was overtopped again
on September11 and 12, but no damage or unusual behavior was observed. On the
morning of September 18, the dam was inspected and nothing unusual was noticed. At
1:25 pm on the same day a local person had walked across the dam, but 10 minutes later
the dam ruptured spectacularly without any warning.

Field investigations of the failure resulted in the following observations:

• The artificial concrete abutment was found to be intact with no sign of cracking or
movements.
• The dam body was totally destroyed with smaller debris from the right and central
portions of the dam found long distances downstream. Large trapezoid blocks were
found near the left abutment.

11-150
 • The top 5.6 ft (1.7 m) of the dam, which had no peripheral joint, was sheared off at
both abutments along the masonry placement joints. At the right abutment, debris
showed signs of upward movement of 2 to 3 inches (5 to 8 cm). At the left
abutment some debris were found upstream, confirming that the rotation of the dam
must have occurred with respect to this abutment.
• Peripheral joint surface showed two sets of frictional traces, one parallel to the dam
axis and another inclined toward the downstream at 30°. The traces parallel to the
dam axis were light and those toward the downstream were deep scratches at the
upper elevations and shallow traces at lower levels.

Based on the above observations and detailed inspection of the failed dam, the following
scenario was offered as the most probable mode of failure

• The body of the dam moved up along the peripheral joint, producing the first set of
frictional traces parallel to the axis of the dam.
• The upward movement in turn caused widening of the horizontal arch spans,
stressing the crown to the point of rupture.
• The asymmetric topography and the site geology caused the rupture of the dam to
be offset toward the right side.
• The sudden failure of the structure was triggered by shearing of the top portion
which did not include any joint.
• The sudden release of pressure caused the water to rush through the rupture and
allowing the material to fall back onto the peripheral joint, thus producing the
second set of traces on the joint surface.
• The material from the right side was washed downstream, but the left side rotated
about the left abutment causing some debris to be pushed upstream.

11-8.4 High-Discharge Induced Incidents

11-8.4.1 Failure of Arch Dams

One of the chief causes of dam failure, considering all types of dams, is overtopping or
inadequate spillway capacity. However, the number of arch dams that have been
damaged or have failed while passing extreme floods is quite small. The International
Commission on Large Dams (ICOLD) Bulletin 99 (1995) lists only one arch dam that
failed due to overtopping (Alla Sella Zerbino Dam, see 11-8.1). One possible

11-151
explanation for the fact that overtopping is not as significant in arch dams as in other
types of dams could be that arch dams are typically built on sound rock foundations.
The chief danger from overtopping an arch dam would be erosion of the abutments or the
foundation. Thus, a minor amount of overtopping of an arch dam may not be dangerous
as long as the abutments and foundation are sound and the depth of overtopping is not
great and does not occur with a long duration. A description of the experience with
Gibson dam is notable and is presented here because the dam, although severely
overtopped, did not fail.

Gibson Dam. Gibson dam is an concrete arch-gravity dam on the Sun River in Montana
which was completed in 1929. It is protected by a “morning-glory spillway” near the left
abutment which is controlled by six 34-ft by 12 ft radial gates which were added in 1938
to increase the reservoir capacity. Although the dam did not fail, it was overtopped for
20 hours at a depth of approximately 6.5 feet during passage of a record flood in June of
1964. Although some rock was plucked from the abutments, neither the dam nor the
abutments suffered significant damage and stability of the dam was never threatened. It
is of interest to note that the spillway gates were only partially open at the time the
overtopping began. Since the caretaker’s house was on the opposite side of the dam, the
controls for the gate were not accessible after overtopping began.

11-8.4.2 Damage to Stilling Basins and Plunge Pools

Plunge Pools. As the previous section states, only a few arch dams have failed or been
damaged by the passage of extreme floods. However, serious threats to arch dams could
arise as a result of erosion of the downstream plunge pool if the erosion were to occur
near enough to the dam. Kariba dam in Zimbabwe provides an example of such a threat.
Kariba, completed in 1962, is a 426-foot high arch dam which impounds the world’s
largest man-made lake. The spillway consists of six orifices (28 x 30 feet) which
discharge through the dam and impact in the downstream plunge pool. The rock is
generally regarded to be sound gneiss. When the dam was designed in 1955, two power
plants were envisioned and it was estimated that the spillway would discharge only about
once in five years. However, the second power plant was never built; as a result the
spillway has operated more frequently than planned with spill durations of several
months. By 1967 the maximum scour depth in the plunge pool had reached 160 feet and
a total volume of more than 500,000 cubic yards of rock was removed and carried
downstream by the flow. By 1981 the scour depth had reached almost 200 feet and there
was considerable concern about the potential instability of the dam foundation. As a
result, some repairs have been made to the plunge pool and erosion seems to have been
abated. However, a prolonged drought in the drainage basin above Kariba has greatly
reduced the frequency of spills for the past several years. Currently, the plunge pool is

11-152
being monitored annually and necessary repairs are being made including rock bolting
and placement of concrete in critical areas.

A side issue at Kariba is erosion of the abutments by runoff produced by spray. As much
as four inches of water per day falls on some parts of the abutments when the spillway is
operating. Gravel has been placed on the abutments in critical areas to prevent serious
erosion.

11-8.5 Earthquake Induced Damage

Concrete arch dams have an excellent record of performance with respect to earthquake
motion. No failure has ever resulted from earthquake damage to an arch dam. It must be
realized however, that very few major earthquakes have occurred close to an arch dam.
Major earthquakes on the order of the maximum credible earthquake are very rare events,
and in most cases the MCE for a given dam site represents an unprecedented loading
condition.

Among some 43 arch dams in 14 countries that are known to have been subjected to
significant earthquake excitation (Serafim, 1987), only four have experienced a
maximum or a near-maximum earthquake shaking with epicenter close to the dam site.
The four arch dams are Pacoima, Lower Crystal Springs, and Gibraltar dams in the
United States, and Ambiesta Dam in Italy. Except for Pacoima Dam, which suffered
damage during two recent earthquakes, all other 42 dams experienced very little or no
damage. Following is a description of the performance of Pacoima Dam and other three
dams for their historical or design significance:

11-8.5.1 Pacoima Dam

The Pacoima Dam constructed in 1929 is located across Pacoima Creek in the San
Gabriel Mountains about 5 miles north of San Fernando in southern California. The dam
is a 372 ft high, moderately thick concrete arch dam, with a crest length of 640 ft. The
dam is made of ten 50-foot-wide cantilevers separated by vertical contraction joints. The
left-most joint (looking downstream) is the interface between the dam and a low-gravity
thrust block that is built into the left abutment. Pacoima Dam was designed for static
loads only, with no consideration for earthquake loading. The design was checked in
1928 by stress analysis using the crown cantilever trial-load method (Section 11-5.3.1).
The principal functions of Pacoima Dam are flood control and water conservation.

There are seven significant faults within a 3.8-mile radius of the dam, with the San
Andreas fault located 20 miles northeast of the dam. The two largest earthquakes in the

11-153
dam site region so far this century are the 1971 San Fernando (moment magnitude Mw
6.7) and the 1994 Northridge (Mw 6.7) earthquakes. The 1971 and 1994 earthquakes
occurred on different fault systems and differ primarily in the dip of the faults, with a
north-northeast-dipping fault plane in 1971 and a south-southeast-dipping fault plane in
1994 (EERI, 1995). The 1994 sequence started deeper and was more damaging because
it produced stronger ground motions. Although severely shaken, Pacoima Dam survived
both earthquakes with no noticeable cracking of the dam in 1971 and with some visible
cracks and block offsets in 1994, which attests to strength reserve inherent in arch dams.

11-8-5.1.1 Performance during the San Fernando Earthquake

On February 9, 1971, Pacoima Dam was subjected to a Mw 6.7 earthquake originating on
the part of Sierra Madre fault system which passes beneath the dam at a depth of 3 miles.
The focus of the main shock was 4 miles north of the dam, at a depth of about 8 miles
where the faulting initiated. The fracture then propagated up and to the south, past under
the Pacoima Dam, and intercepted the surface after reaching a length of about 10 miles.
Pacoima Dam survived the severe ground motion of this major event without any
observed cracking in the arch.

An accelerograph located on the left abutment 52 ft above the dam crest recorded, a then
unprecedented value of 1.25g on both horizontal components and maximum value of
0.70g on the vertical direction. Post-earthquake studies indicated that base-rock peak
acceleration might have been in the range of 0.6 to 0.8g, thus relating the high peak
accelerations to the canyon topography effects and possible shattered condition of the
bedrock. In addition, the dam included a Wilmot Seismoscope at the crest which was
seriously damaged during the first few seconds of the crest motion, so that no record of
the dam response was obtained.

The Pacoima Dam was designed with no consideration for earthquake loading, but it did
not develop structural cracks or experience relative movements between adjacent blocks
as a result of the 1971 earthquake. The main damage was a separation of the arch dam
from the left thrust block of 3/8 in. at the crest and extending downward about 50 ft. The
left abutment, however, experienced extensive cracking of the gneissic granite-diorite
rock. Many cracks penetrated through the gunite coating into the rock below, and slight
movement of large blocks of rock mass occurred. Following the earthquake, stability of
the dam was evaluated using the three-dimensional finite-element analysis and the left
abutment supporting the thrust block was subsequently strengthened with 35 post-
tensioned tendons to increase stability against future earthquakes. The crack in the thrust
block and the open joint between the dam and the thrust block were also repaired.

11-154
According to Hansen and Roehm (1979), the exceptional performance of Pacoima Dam
during the 1971 event can be related in part to the following :

• the tensile strength of the concrete is significantly increased when subjected to rapid
strains;
• the rock foundation absorbs a portion of the earthquake's energy;
• inelastic behavior of the concrete contributes to an ultimate capacity for the dam
considerably greater than that predicted by elastic analysis; and
• the grouted vertical contraction joints with low tensile capacity provide planes across
which large horizontal tensile stresses are relieved.

In addition, it should be noted that at the time of the earthquake, the water level for the
flood control Pacoima Dam was 147 ft below the crest. A higher water level can be
expected to cause more damage than that observed during the 1971 event. The duration
of strong ground motion was also approximately 7.5 seconds. A magnitude 8+
earthquake of longer duration on the San Andreas Fault would probably be a more severe
test of the dam.

11-8-5.1.2 Performance During the Northridge Earthquake

On January 17, 1994, Pacoima Dam once again was subjected to a damaging earthquake
of Mw 6.7. The epicenter was approximately 11 miles southwest of the dam. The
earthquake occurred on a deep concealed thrust fault beneath the San Fernando Valley.
The rupture began at 12 miles, and terminated at a depth of about 4 miles, leaving no
obvious surficial expression. Although the Northridge earthquake was the same size as
the 1971 San Fernando earthquake, it produced stronger ground shaking. The dam
withstood the larger shaking, but experienced more damage in the abutments and in the
body of the dam than occurred during the 1971 event.

Strong Motion Records.

Following the 1971 San Fernando earthquake, Pacoima Dam was extensively instru-
mented with additional sensors, so that strong motion response of the dam and its abut-
ments during future earthquakes can be recorded. The accelerometer on the upper left
abutment recorded accelerations of 1.47g and 1.70g on the horizontal components and
1.36g on the vertical component, as opposed to the peak accelerations of 1.25g horizontal
and 0.7g vertical measured previously at the same location in the 1971 event. The
downstream records obtained approximately 430 feet downstream from the base of the

11-155
dam, showed peak accelerations of 0.44g and 0.20g on the horizontal and vertical
components, respectively. These records clearly illustrate the effect of canyon
topography on ground motion, as evident from differences in acceleration amplitudes and
waveforms of the upper abutment and the downstream motions. At the base of the dam,
peak acceleration was 0.54g in the radial direction and 0.43g in the vertical direction.
The radial peak accelerations at the left quarter-point of the crest and at 80% of the dam
height reached 2.3g. Peak accelerations exceeding 2g were recorded at the center of the
crest, but the corresponding strong motion records could not be digitized because the
traces exceeded the range of instruments and became intertwined. The dam strong
motion records showed higher frequency components than the downstream and base
records, possibly a result of higher mode contributions and/or impact caused by closing
of the contraction joints.
Observed Damage
During the Northridge earthquake, the severely shaken Pacoima Dam suffered damage
both in the left abutment and within the dam body. The damage observed at the site was
consistent with the strong shaking indicated by the accelerograms, and was more severe
than occurred during the 1971 San Fernando earthquake. Again, the water surface at
Pacoima Dam was low and stood at 131 feet below the crest.

Damage visible at the site (County of Los Angeles, 1994) included rock slides, extensive
cracking of the shotcrete cover of the left abutment, opened contraction joint between the
dam and thrust block on the left abutment, and several cracks on the downstream face of
the dam. The contraction joint between the dam and thrust block opened and remained
open 2 inches at the crest, decreasing to 1/4 of an inch at the bottom of the joint (60 feet
below the crest), at which point a large diagonal crack extended down the thrust block to
meet the foundation rock (Fig. 11-8.2). Apparently, the diagonal crack and contraction
joint opening were caused by movements of two rock masses on the left abutment: Rock
Mass A, and Rock Mass B. Rock Mass A and its adjacent Rock Mass B which is
supporting the thrust block, are underlain by a slip plane (Plane 1 in Fig. 11-8.2) and are
known to have marginal factors of safety against sliding. Survey measurements made
after the earthquake indicated that Rock Mass B slipped about 2-3 inches horizontally
and 2 inches down, thereby accounting for opening in the contraction joint between the
dam and thrust block, while Rock Mass A moved 16-19 inches horizontally and up to 14
inches down. During the 1971 event, Rock Mass A moved 50% less and the movement of
Rock Mass B was slight but enough to open up the same joint by 3/8 of an inch. The 35
tendons installed to secure the thrust block after the 1971 earthquake may have played a
significant role in limiting the movement of Rock Mass B during the Northridge
earthquake.

11-156
Over-stressing of the dam was indicated by permanent block offsets and by several
cracks visible on the downstream face of the left-most block between elevations 1,925
feet and 1,967 feet. Two lift joints at elevations 1,967 feet and 1,978 feet showed
evidence of broken bond and remained open slightly. Several cracks were found on the
crest, each ran diagonally from the recessed corner of a shear key to either the upstream
or downstream face, thus indicating working of monolith joints during the closing.
Opening of the contraction joints was evident by their clean appearance after the
earthquake. Permanent vertical offsets were observed along most joints, with the right
block always remaining lower than the left block. The post-earthquake measurements
with laser plumb line showed a permanent upstream displacement of the dam crest of 2
inches.

Fig. 11-8.2 Downstream elevation of the left side of the Pacoima Dam arch and the
thrust block showing cracking.

Possible failure modes

From the damage observed during the Northridge earthquake, earthquake induced failure
would probably involve the upper 65 feet of the dam. Such failure could originate from
loss of the thrust block caused by a sliding failure of rock masses A and B, or through
cracking and opened contraction joints in the dam itself, which could lead to unstable

11-157
concrete blocks. In either case, a failure of the upper part of the dam more likely would
not release the reservoir water, because the intake to the tunnel spillway is 65 feet below
the crest. Only concurrent flood and a damaging earthquake might possibly result in a
sudden release of water or leakage through opened joints, but more likely the lower part
of the dam would remain intact.

11-8.5.2 Other Significant Cases

11-8.5.2.1 Lower Crystal Springs Dam

One of the earliest cases of a concrete dam excited by a major earthquake was Lower
Crystal Springs Dam located in San Mateo, California, about 20 miles south of San
Francisco. Completed in 1890, the 154-ft-high, 600-ft long, curved gravity dam
withstood the 1906 San Francisco earthquake (estimated magnitude of 8.3) without a
single crack, even though it was located within 1,100 ft of the San Andreas fault rupture
and earthquake loading had not been considered in its design. The dam was shaken again
during the Loma Prieta earthquake of October 17, 1989, a magnitude 7.1 earthquake that
occurred 40 miles south of the dam. The ground motion at the dam site was more
moderate this time, and again, the dam was not affected.

A recent seismic safety evaluation of the dam showed that Lower Crystal Springs Dam
could be expected to resist the forces of a magnitude 8.5 earthquake with no serious
structural damage.

The excellent performance of the dam was attributed to a high reserve capacity arising
from:

• The truncated cross section was designed as a gravity dam, but was curved in plan for
additional load resistance by the arch action.
• The meticulous design included the use of interlocking blocks staggered so that there
were no continuos horizontal or vertical joints through the dam.
• The foundation rock was highly fractured with visible fracture surfaces which
absorbed much of the earthquake's strain energy.

11-8.5.2.2 Gibraltar Dam

Gibraltar Dam, a 169-foot-high arch dam built in 1920 near Santa Barbara, California,
suffered no damage during the 1925 Santa Barbara earthquake of magnitude 6.3 which
occurred beneath the dam. I has been reported "the dam was so severely shaken that a
watchman who was on the dam at the time had difficulty in standing up." This and earlier

11-158
experience at Lower Crystal Springs dams provided the early evidence that properly built
arch dams could be expected to perform satisfactory under earthquakes, even though the
earthquake forces were not considered in the design of the dam.

11-8.5.2.3 Ambiesta Dam

Ambiesta Dam, built in an area of high seismicity near Gemona in northern Italy, is an
arch structure specifically designed to withstand earthquakes. The 194-foot-high dam
with a crest length of 475 ft was designed as a symmetric arch with striking downstream
overhang, which the designers believed would resist the exceptional overloads that the
dam might ever experience (Hansen and Roehm, 1979).

An actual test of the design occurred during the 1976 Gemona-Friuli earthquake of
magnitude 6.5, which generated a maximum acceleration of 0.33g at the right abutment
of the dam located about 14 miles away from the epicenter of the quake. The dam was
also subjected to a foreshock of magnitude 4.5 about one minute before the main event,
and to 4 major aftershocks of magnitude 5.1, 5.5, 5.9, and 6.0 that struck the region in a
period of more than 4 months.

Hansen and Roehm (1979) reported that neither the Ambiesta dam nor 13 other concrete
arch dams in the region suffered any damage from these events. In particular, the
experience at Ambiesta Dam confirmed the results of model studies that indicated much
larger acceleration of 0.95g in the transverse direction, or an acceleration of 0.76g in the
vertical direction would be required to induce damage in the dam.

11-8.6 Detrimental Chemical Reactions

Detrimental chemical reactions such as alkali-aggregate reactions (AAR), also known as
alkali-silica reaction (ASR) or alkali-carbonate reaction (ACR), have been found in a
large number of concrete dams and hydroelectric plants around the world. The reactions
occur between certain aggregates and alkalis in the cement, leading to the formation of
gels which then absorb water and expand, causing increased stress, micro-cracking of the
concrete and structural deformations. Three basic requirements for expansive AAR to
occur are: 1) presence of deleteriously reactive aggregates, 2) sufficient concentration of
alkali, and 3) adequate levels of moisture in the concrete. Hobbs (1990) notes that where
alkalis come primarily from the cement, AAR expansion has been observed to be largely
complete in 8 to 15 years. However, there are some increasing cases which suggest that
the supplementary alkalis provided by the coarse aggregates might be significant, thus
creating "auto-generation" conditions with an indefinite period of expansion..

11-159
Charlwood and Solymar (1995) have reported 104 worldwide known cases of AAR in
hydraulic structures, of these 32 cases are concrete arch dams listed in Table 11-8.1. For
each dam, the table includes name, height, location, date of construction, a brief
description of damage, and the repair performed on the structure. Most arch dams with
AAR were built prior to discovery of AAR in California (Stanton 1940), or shortly after
when clues on the AAR processes were still emerging. However, a few cases of AAR
reported in South Africa and Mozambique occurred in the 1960's and even as recent as
1974. A large number of arch dams subjected to AAR have continued to function
adequately for many years. Well-known cases include Gene Wash and Copper Basin and
Parker dams, where substantial strains occurred within the first 20 years after the
construction but expansion appear to have ceased. In cases such as Stewart Mountain,
Churchil, and Gmued dams, strengthening measures were required; in Matilija, the upper
40 ft of the dam was replaced due to severe deterioration of the concrete; and in Drum
Afterbay the reaction was continuing at such a high rate that the dam was taken out of
service and replaced by a new arch structure immediately downstream.

The primary adverse effects of expansive AAR in arch dams include micro-cracking and
structural deformations. The micro-cracking typically extends to shallow depths and can
lead to loss of tensile and shear strength of the concrete, although the compressive
strength may be retained. The deformations typically cause upstream and upward
movements, which can lead to additional cracking and reduced strength in the dam
concrete. They can also result in overstressing and cracking of lift joints, thus reducing
shear strength along the lift lines. In fact, loss of joint shear strength has been of concern
in static and particularly in seismic safety assessments when combined arch-cantilever
action is necessary (e.g. Stewart Mountain where anchors were installed). Other adverse
effects may include leakage, leaching, loss of gate clearances, and problems at the
abutments or spillway openings where significant structural discontinuities exist.

The case histories described in the subsequent sections illustrate that the effects of
expansive AAR can range from minor cracking to major abnormal movements; the
reaction may have ceased or the expansion rates are slow and ongoing. Nevertheless, as
the condition of the dam worsens the potential for the subsequent failure due to the
imposed loads increases. Therefore, it is prudent that appropriate performance parameters
for safety evaluation of arch dams affected by AAR are developed. One such
performance parameter based on the visual and instrumentation monitoring of the dam
has been developed and implemented by USBR as part of their dam safety program
development (Veesaert and LaBoon 1995).

11-160
Table 11-8.1 Summary of case histories of concrete arch dams with AAR problems
(adopted from Charlwood and Solymar, 1995)

Dam Height Country Year Damage Repair or

(m) Built Replacement
Alto Ceira 37 Portugal 1949 movements, cracking Under study
Bartlett 87 Arizona, USA 1936 some strength loss lithium hydroxide tests
(multiple arch)
Bimont 87 France 1952 expansion, cracking grouting, epoxy coating
on affected parts
Cahora Bassa 170 Mozambique 1974 swilling but no cracking Under study
Churchill 39 South Africa 1943 horizontal cracking Strengthening/extension
of buttress with
reinforced concrete
Coolidge 77 Arizona, USA 1929 deterioration concrete, spillway gate
reinforcement
Copper Basin 64 Calif., USA 1938 Reaction completed
Dinas 14 Wales, UK 1957 serious cracking epoxy sealing, cut
exp. joint, grouting
Drum Afterbay 25 Calif., USA 1924 Replaced in the 1960's
Gene Wash 48 Calif., USA 1937 strength loss Reaction completed
Gmuend 30 Austria 1945 Strengthened by adding
gravity section
Gibraltar 50 Calif., USA 1920
Horse Mesa 93 Arizona, USA 1927 surface deterioration, Surface repair
some joint separation
Kariba 126 Zambia/Zimbabwe 1955 cracks in spillway
Keerom 36 South Africa 1954 cracking Grouting
Matilija 58 Calif., USA 1947 spillway gate jams, Cutting a 85 m long and
cracking, joint opening 9 m deep notch in 1965,
replacem. of upper 12 m
Maury 72 France 1947 cracks
N'Zilo/ 75 Zaire 1952 Extensive cracking
Delcommune
Owyhee 127 Oregon, USA 1932 cracking after 5 yrs and
deterioration after 11 yrs
Parker 96 Calif., USA 1938 surface cracking and Expansion assumed
strength loss ceased by 1965
Pathfinder 65 Wyoming, USA 1909 deterioration of parapets Surface repair
and power house walls
Paul Sauer 82 South Africa 1969 cracking lift joints,
/Kougha movement
Peti 46 Brazil 1946
Pietersfontain 29 South Africa 1966 none but low tensile
strength
Poortjlieskloof 36 South Africa 1955 horizontal cracking Minor
Roode Elsburg 72 South Africa 1968 none
San Esteban 115 Spain 1955 U/S deterioration, U/S water proffing,
seepage grouting of joints
Santa Luzia 76 Portugal 1943 movements Possible seal of u/s
Santeetlah 61 N. Carolina, USA 1928 cracking Slot cut, epoxy
grouting, extra concrete
Stewart Mountain 63 Arizona, USA 1930 expansion. movement, Grouting, structural
deterioration modification, expansion
ceased
Stolsvatn 18 Norway 1970? surface cracking, freeze Epoxy coat, Foundation
(multiple arch) thaw damage anchors
Stompdrift 49 South Africa 1965 cracking, opening joints Grouting of
(multiple arch) construction joints

11-161
11-8.6.1 Kouga Dam, South Africa
Kouga Dam (formerly Paul Sauer Dam), is a 78m (256 ft) high, 317 m (1040 ft) long,
double curvature arch dam completed in 1969. In 1981, an evaluation of the monitoring
program revealed that the dam crest was rising slowly with the center of the crest also
moving slowly upstream (Elges et al. 1995). The movements, which amounted to 8 mm
vertical movement at the time, were suspected to be the results of AAR. A further
investigation, which included laboratory testing of concrete cores, concluded the
expansion measured in the dam is caused by the alkali-aggregate reaction. The cement
used for the construction of the dam is now is regarded as high-alkali cement.

The geodetic measurements taken twice a year from 1972 to date, indicate a continuous
expansion trend since 1976. The expansion now amounts to a total upward displacement
of 22 mm. A crack on the upstream side indicates that the spillway has suffered more
severe AAR, and that the bulging of the spillway sill has raised its level between 25 and
45 mm. Since 1984 expansion appears to be decreasing, but it is not clear whether AAR
is decreasing or the successive and prolonged periods of low water levels are responsible
for the reduction.

11-8.6.2 Santa Luzia, Portugal

This thin cylindrical arch dam with a maximum height of 76 m (250 ft) and a crest length
of 115 m (377 ft) was completed in 1943. The expansion due to AAR was detected from
the continuous upstream and upward movements of the crest (Ramos et al., 1995). Since
the first filling of the reservoir, geodetic and alignment measurements have been used to
monitor such movements. The maximum upward displacement of the crest accumulated
over a period of 40 years is reported to be about 50 mm. During the same period the
maximum upstream movement has reached 30 mm. The mineralogical and petrographic
analyses conducted on the dam concrete samples showed that AAR has occurred from the
reaction between the reactive silica provided by cataclastic quartz, and the alkalis
supplied by feldspars aggregates ("auto-reaction"). Apart from providing the necessary
alkaline environment, the low-alkali-content cement has little value to the reaction.
Ongoing deformation monitoring, periodic ultrasound pulse velocity tests, and structural
analysis are being considered for better characterization of AAR and its effects on the
long-term performance of the dam.

11-8.6.3 Alto-Ceira Dam, Portugal

Completed in 1949, Alto-Ceira is a thin arch dam with a maximum height of 37 m (121-
ft) and a crest length of 120 m (394-ft). Shortly after the first filling of the reservoir, the
dam began experiencing increasing movements in the upstream and upward directions,

11-162
accompanied with extensive cracking both in the upstream and downstream faces of the
dam (Ramos et al., 1995). Numerous studies including visual inspections, mapping of
cracks, core testing, and petrographic analysis of concrete samples were conducted to
identify the source and effects of such expansion on the operation and safety of the dam.
The studies concluded that alkali-silica reactions are responsible for the expansion. It
was determined that the reactions occur among quartz and metapelite aggregates
containing reactive silica, feldspar aggregates as a supplementary source of alkalis, with
the cement providing the alkaline environment necessary for the reaction. In such cases,
the presence of alkalis in feldspar aggregates creates "auto-reaction" conditions (Ramos
et al. 1995), in which the expansion may continue indefinitely. The measured
deformations and 3D finite-element analysis of the dam indicate that expansion has now
advanced to the entire dam. Recent ultrasound pulse velocity tests showed significant
deterioration of the concrete; and a survey of cracks indicated major cracks have reached
a depth of about 60 cm (2-ft). Ongoing expansion continues at approximately constant
rates, which if not reduced, inevitably might adversely affect the operation and possibly
the safety of the dam.

11-8.6.4 Cahora-Basa Dam, Mozambique

The 170m-high Chora-Basa Dam with a crest length of 300 m, is a double curvature arch
dam constructed between November 1971 and December 1974. Since the first filling of
the reservoir in January 1975, the dam has been showing continuous movements in the
upstream and upward directions. The expansion of the concrete was first detected by
continuous rising of the strains measured by a group of 50 extensometers embedded in
the concrete, and later was confirmed by the alignment measurements and petrographic
analysis of concrete samples. The reaction is also suspected to have developed by "auto-
reaction" caused by the presence of alkalis in the aggregates. Expansion at the moment is
rather moderate, showing a maximum upstream bulging of about 11 mm accumulated
between 1977 and 1994. The expansion has not induced any visible cracking, but a more
elaborate testing and analysis program is underway to better characterize its development
and effects on the safety of the dam.

11-8.6.5 Gene Wash and Copper Basin, California

Located close together in San Bernardino County, California, Gene Wash and Copper
Basin were constructed in 1937 and 1938 using the same materials. Gene Wash is a 131-
foot-high arch structure with a gravity thrust block on the right abutment, and Copper
Basin is a taller arch with a maximum height of 187 feet.

11-163
Within the first 20 years after their construction, both dams experienced active
expansions and extensive pattern cracking. During this period the total deformation of
the dam reached 11 to 12 cm in the upstream and about 9 cm in the upward direction, and
then decreased essentially to zero for the past 40 to 50 years. A comprehensive concrete
testing program indicated that the expansion was caused by autogenuous growth due to
AAR. While the effects of this distress are still evident, it appears that the dams have
stopped deteriorating, and that the AAR is in a dormant state.

11-8.6.6 Horse Mesa, Arizona

The 93 m (305 ft) high Horse Mesa Dam with a crest length, including the abutment
spillways, of 201 m (660 ft), is a thin arch dam completed in 1927. Located about 65
miles northeast of Phoenix, the dam impounds Apache Lake on the Salt River in central
Arizona.

After the discovery in 1946 of AAR problems at Stewart Mountain Dam, cores were
taken from the Horse Mesa Dam. Tests showed AAR has occurred, but to an
insignificant degree. Additional cores taken in 1968 indicated no deterioration of the
concrete and produced satisfactory compressive strength. The reaction has caused typical
surface cracking and minor expansions in the dam concrete with permanent deformations
of the arch without affecting the integrity of the dam. Also, there appears to be
disbonded lift joints as evident by seepage passing along the downstream face of the dam.
At the present time, the dam crest is deforming approximately 0.25 mm/year (0.01
inch/year) upward, and between 0.25 and 0.75 mm/year (0.01 and 0.03 inch/year)
upstream (Veesaert and LaBoon 1995).

11-8.6.7 Owyhee, Oregon

Constructed between 1928 and 1932, Owyhee Dam is a concrete thick-arch structure with
a maximum height of 127 m (417 ft) and a crest length of 183 m (600 ft). Deterioration
and cracking of concrete, which first appeared in 1948, is continuing and is thought to be
caused by AAR (Veesaert and LaBoon 1995). An investigation conducted in 1988
concluded that AAR is occurring at varying degrees in the dam. It is strong in the upper
and outer portions of the dam, and only mild to nonexistent in the lower and deep interior
portions.
According to Veesaert and LaBoon (1995), a monitoring program began in 1985 shows
an upstream movement at an average rate of 9 mm/year (0.37 inch/year) near the center
of the dam, with much slower upward movements averaging less than 2 mm/year.
Ongoing visual and instrumentation monitoring indicate dramatic increases in seepage
when the reservoir water level is near or above elevation 2650 ft (crest El. 2675 ft),

11-164
probably because of significant cracking in the upper 50 to 75 feet of the dam which has
occurred as a result of AAR. Although AAR cracks and seepage through the concrete
reduce sliding resistance, the computed large factors of safety indicate that the relatively
thick gravity-type sections of Owyhee Dam tend to preclude this type of failure.

11-8.6.8 N'Zilo, Zaire

Completed in 1952, N'Zilo (formerly Delcommune) Dam in Zaire is a thin double
curvature arch dam about 246 feet high, which suffered an expansion of approximately
1000 micro-strain in 30 years, resulting in extensive cracking.

11-165
REFERENCES

American Society of Testing Materials (1984a), “Standard Test Method for Determining
the In Situ Modulus of Deformation of Rock Mass Using the Rigid Plate Loading
Method,” ASTM-D4394-84, Annual Book of ASTM Standards, Vol. 04.08, Philadelphia,
PA.

American Society of Testing Materials (1984b), “Standard Test Method for Determining
the In Situ Modulus of Deformation of Rock Mass Using the Flexible Plate Loading
Method,” ASTM-D4395-84, Annual Book of ASTM Standards, Vol. 04.08, Philadelphia,
PA.

Annandale, G. W. (1995), “Erodibility,” Journal of Hydraulic Research, Vol. 33, No. 4,

International Association for Hydraulic Research, Delft, The Netherlands.

Babb, A.O. And T.W. Mermel (1968), Catalog of Dam Disasters - Failures and
Accidents, PB 179 243, U.S. Bureau of Reclamation, Washington, D.C.

Barnes, H.H. (1977), Roughness Characteristics of Natural Channels, U.S. Geological

Survey Water-Supply Paper 1849, Arlington, Virginia.

Barton, N., Lien,R., and Lunde, J. (1974), “Engineering Classification of Rock Masses
for Design of Tunnel Support,” Rock Mechanics Vol. 6, No. 4.

Barton, N., “The Shear Strength of Rock and Rock Joints (1976),” Int. J. Rock Mech.
Min. Sci. & Geomech. Abstr. Vol. 13 pp. 255-279, Pergamon Press, Great Britain.

Barton, N. And Choubey, V. (1977), “The Shear Strength of Rock Joints in Theory and
Practice” Rock Mechanics, Vol. 10 pp. 1-54, Springer-Verlag, Wien, Austria.

Barton, N. (1988), “Rock Mass Classification and Tunnel Reinforcement Selection Using
the Q-System,” in Rock Classification Systems for Engineering Purposes: ASTM STP-
984, Kirkaldie, L. (Ed.), American Society of Testing and Materials, Philadelphia, PA.

Bathe, K.J. and Wilson, E.L., (1974). "Thick Shell Structures," Proceedings International
Symposium on Structural Mechanics Software, University of Maryland, College Park,
Maryland.

Bathe, K.J., and Wilson, E.L. (1976). "Numerical Methods in Finite Element Analysis,"
Prentice-Hall, Englewood Cliffs, NJ.

Bieniawski, Z.T. (1979), “The Geomechanics Classification in Rock Engineering

Application”, Proceedings of 4th International Congress on Rock Mechanics, ISRM,
Montreux, Balkema, Rotterdam, Vol. 2.

11-166
Bieniawski, Z.T. (1990), “Tunnel Design by Rock Mass Classifications,” Technical
Report GL-79-19, U.S. Army Corps of Engineers, Waterways Experiment Station,
Vicksburg, MS.

Billings, M.P. (1954), “Structural Geology,” Prentice Hall, Inc., Englewood Cliffs, NJ.

Bradley, J.N. (1952), “Discharge Coefficients for Irregular Overfall Spillways,”

Engineering Monograph No. 9, U.S. Bureau of Reclamation, Denver Federal Center,
Denver, Colorado.

Burgi, P.H. (1988), “Cavitation Damage and Repair of Glen Canyon Spillway Tunnels,”
Advance Dam Engineering, for Design and Construction, (R.B. Jansen, Editor), Van
Nostrand Reinhold, New York, New York.

Cameron, C.P., Patrick, D.M., May, J.H., Palmerton, J.B., McAneny, C.C. (1986, 1988a,
1988b, 1989), “Geotechnical Aspects of Rock Erosion in Emergency Spillway
Channels,” Tech. Report REMR-GT-3, U.S. Army Corps of Engineers, Waterways
Experiment Station, Vicksburg, MS.

Cassidy, J.J. (1970), “Designing Spillway Crests for Negative Pressures,” ASCE, Journal
of the Hydraulics Division, Vol. 96, HY3. March.

Charlwood, R. G., and Solymar Z. V. (1995)."Long-term management of AAR-affected

structures - an international perspective," USCOLD 2nd International Conference on
Alkali-Aggregate Reactions in Hydraulic Plants and Dams, Chattanooga, Tennessee,
October 22-27.
Chopra, A. K. (1988), "Earthquake analysis of concrete dams," Chapter 15 in Advanced
Dam Engineering for Design, Construction, and Rehabilitation, edited by Robert B.
Jansen, Van Nostrand.

Chopra, A.K., and Tan H. (1996), "Foundation modeling in earthquake analysis of arch
dams," 16th annual USCOLD lecture series, Los Angeles, California, July 22-26, 1996.

Chow, V.T. (1959), Open Channel Hydraulics, McGraw-Hill Book Company, New York,
New York.

Clough, R. W. 1977 (Feb.). Lecture Notes, unpublished.

Clough, R.W., Chang, K.T., Chen, H.Q., and Ghanaat, Y. (1985). "Dynamic Interaction
Effects in Arch Dams," Report No. UCB/EERC-85/11, University of California,
Earthquake Engineering Research Center, Berkeley, CA.

County of Los Angeles, (1994), Report on Initial assessment of the effects of the January
17, 1994, Northridge/San Fernando earthquake on Pacoima Dam, Phase 1. Prepared by
Morrison Knudsen Corporation, April.

11-167
CSMIP (1994), "Strong Motion Records from the Northridge, California Earthquake of
January 17, 1994," Report No. OSMS 94-07, California Department of Conservation,
Strong Motion Instrumentation Program.

Davis, C.V. (1969), Handbook of Applied Hydraulics, McGraw-Hill Book Company,

New York, New York.

Deere, D.U. and D.W. Deere (1988), “The Rock Quality Designation (RQD) Index in
Practice,” Rock Classification Systems for Engineering Purposes, ASTM STP-984, (L.
Kirkaldie, Editor), American Society for Testing and Materials, Philadelphia,
Pennsylvania.

EERI (1995), "Northridge Earthquake of January 17, 1994 Reconnaissance Report,"

Supplement C to Volume 11 Earthquake Spectra, April.

Elges H., Geertsema A., Lecocq P., and Oosthuizen C. (1995), "Detection, monitoring
and modeling of Alkali-Agregate Reaction in Kouga Dam (South Africa)," USCOLD 2nd
International Conference on Alkali-Aggregate Reactions in Hydraulic Plants and Dams,
Chattanooga, Tennessee, October 22-27.

Fenves, G. L., Mojtahedi, S., and Reimer, R. B., (1989),"ADAP-88: A Computer

Program For Nonlinear Earthquake Analysis of Concrete Arch Dams," Report No.
UCB/EERC-89/12, University of California, Earthquake Engineering Research Center,
Berkeley, CA.

Fok, K. L., and Chopra, A. K. (1985), "Earthquake Analysis of and Response of Concrete
Dams," Report No. UCB/EERC-85/07, University of California, Earthquake Engineering
Research Center, Berkeley, CA.

Fok, K.-L., Hall J. F., and Chopra, A. K. (1986), "EACD-3D: A Computer Program for
Three-dimensional Earthquake Analysis of Concrete Dams," Report No. UCB/EERC-
86/09, Earthquake Engineering Research Center, University of California, Berkeley.
Ghanaat, Y., (1993a), "Theoretical Manual for Analysis of Arch Dams," Technical Report
ITL-93-1, US Army Corps of Engineers Waterways Experiment Station, July 1993.

Ghanaat, Y., (1993b), "User's Manual - GDAP: Graphics-Based Dam Analysis

Program,” Technical Report ITL-93-3, US Army Corps of Engineers Waterways
Experiment Station, Vicksburg, MS.

Ghanaat, Y. and Redpath, B.B. (1995). "Measurements of reservoir-bottom

reflection coefficient at seven concrete dam sites," QUEST Report No.
QS95-01 issued to the US Army Corps of Engineers, Waterways Experiment
Station and Bureau of Reclamation.

11-168
Hall, J.F. (1988), "The Dynamic and earthquake behavior of concrete dams: review of
experimental behavior and observational evidence," Journal of Soil Dynamics and
Earthquake Engineering, Vol. 7, No. 2, pp. 58-121.

Hall, J. F. (1988), "The dynamic and earthquake behavior of concrete dams: review of
experimental behavior and observational evidence," Soil Dynamic and Earthquake
Engineering, Vol. 7, No. 2.

Hansen, K.D., and Roehm, R.H. (1979), "The response of concrete dams to earthquakes,"
Water Power & Dam Construction, Vol. 31, No. 4, pp. 28-31.

Henderson, F.M. (1966), Open Channel Flow, The McMillan Company, New York, New
York.

Hendron, A.J. (1968), “Mechanical Properties of Rock,” Pp 21 to 41, Rock Mechanics in

Engineering Practice, Stagg, K.G., and Zienkiewicz, O.C., (Ed.), John Wiley & Sons,
N.Y.

Hendron, A.J., Cording, E.J., and Aiyer, A.K. (1971), “Analytical and Graphical Methods
for the Analysis of Slopes in Rock Masses,” NCG Technical Report No. 36, U.S. Army
Corps of Engineers, Waterways Experiment Station, Vicksburg, MS.

Hendron, A.J., and Patton, F.D. (1985), “The Vajont Slide, a Geotechnical Analysis
Based on New Geologic Observations of the Failure Surface,” U.S. Army Corps of
Engineers, Waterways Experiment Station, Vicksburg, MS.

Hill, C. J. (1995), "Gene Wash and Copper Basin Dams are surviving alkali-aggregate
reaction," USCOLD 2nd International Conference on Alkali-Aggregate Reactions in
Hydraulic Plants and Dams, Chattanooga, Tennessee, October 22-27.

Hobbs, D. W. (1990), "Cracking and expansion due to alkali-silica reaction: it's effects on
concrete," Structural Engineering Review, pp. 65-79.

Hoek, E. And Bray, J.W. (1981), “Rock Slope Engineering,” Institute of Mining and
Metallurgy, Elsevier Applied Science, London & N.Y.

Hoek, E., Carvalho, J. & Kochen, R. (1995), “SWEDGE, Analysis of the Geometery and
Stability of Surface Wedges”, Rock Engineering Group, Univ. of Toronto.

International Commission on Large Dams (1989), “Monitoring of Dams and Their

Foundations,” Bulletin 68, Paris.

International Commission on Large Dams (1992), “Improvement of Existing Dam

Monitoring,” Bulletin 87, Paris.

11-169
ICOLD (1987), Spillways for Dams, Bulletin 58, International Commission on Large
Dams, Paris, France.

ICOLD (1995), Dam Failures, Statistical Analysis, Bulletin 99, International Commission
on Large Dams, Paris, France.

ICOLD (1996), Vibrarions of Hydraulic Equipment for Dams, Bulletin 102, International
Commission on Large Dams, Paris, France.

Kirsten, H.A.D. (1982), “A Classification System for Excavation in Natural Materials,”

The Civil Engineer in South Africa, July.

Kuo, J. 1982. "Fluid-structure Interactions: Added Mass Computations for

Incompressible Fluid," Report No. UCB/EERC-82/09, University of California
Earthquake Engineering Research Center, Berkeley.

Leonards, G.A. (1987), “Overview and Personal Commentary,” Dam Failures,

Proceedings of the International Workshop on Dam Failures, Leonards G. A. (Ed.),
Purdue University, West Lafayette, Indiana, Aug. 6-8, 1985, Elsevier, N.Y.

Leslie, J.R., and Cheesman, W. J. (1949), "An Ultrasonic Method of Studying

Deterioration and Cracking of Concrete Structures," ACI Journal, Proceedings V. 46, No.
1, September 1949, pp. 17-36.

Londe, P. (1973), “Rock Mechanics and Dam Foundations,” Bulletin of the International
Commission on Large Dams, Paris.

Londe, P. (1987), “The Malpasset Dam Failure” Dam Failures, Proceedings of the
International Workshop on Dam Failures, Leonards, G.A. (Ed.) Purdue University, West
Lafayette, Indiana, Aug. 6-8, 1985, Elsevier, N.Y.

Londe, P. (1993), “Rock Foundations for Dams,” Bulletin 88 of the International

Commission on Large Dams, Paris.

Monfore, G.E., and Taylor, F. W. (1954),"The problem of an expanding ice sheet,"

Bureau of Reclamation Memorandum, March 18, 1954 (unpublished).

Mason, P.J. (1984), “Erosion of Plunge Pools Downstream of Dams Due to the Action of
Free-Trajectory Jets, Paper 8734, Proceedings of Inst. Civ. Engrs, Part 1, Pp 523-537,
London.

Mason, P.J. (1985),”Free Jet Scour Below Dams and Flip Buckets,” ASCE Journal of
Hydraulic Engineering, Vol. 111, No. 2, February, New York, New York.

Muller-Salzburg, L. (1987), “The Vajont Catastrophe - A Personal Review,” Dam

Failures, Proceedings of the International Workshop on Dam Failures, Leonards, G.A.
(Ed.), Perdue University, West Lafayette, Indiana, Aug. 6-8, 1985, Elsevier, N.Y.

11-170
National Research Concil, (1990). "Earthquake Engineering for Concrete Dams: Design,
Performance, and Research Needs."

Newmark, N.M. and Hall, W.J. (1982) "Earthquake Spectra and Design; Engineering
monographs on earthquake criteria, structural design, and strong motion records," Vol.
3: Earthquake Engineering Research Institute, University of California, Berkeley, CA

Patton, F.D. (1966), “Multiple Modes of Shear Failure in Rock,” Proceedings of 1st Int.
Cong. Rock Mechanics, Vol. 1, No.3.47, Pp. 509-514.

Patton, F.D. (1966), “Multiple Modes of Shear Failure in Rock and Related Materials,”
Ph.D. Thesis, Univ. of Ill.

Patton, F.D. (1968), “The Determination of Shear Strength of Rock Masses,”

Terrametrics Short Course on Measurement Systems for Control of Construction and
Mining, Denver.

“Portugues Dam Foundation Investigation,” Design Memorandum No. 22, an

unpublished internal document, U.S. Army Corps of Engineers, Jacksonville District,
Jacksonville, FL., (1988).

Raphael, J.M. "Tensile Strength of Concrete" , ACI Journal, March-April 1984

Ramos J.M., Btista A.L., Oliveira S.B., de Castro A.T., Silva H.S., and de Pinho J.S.
(1995), "Reliability of arch dams subjetc to concrete swelling," USCOLD 2nd
International Conference on Alkali-Aggregate Reactions in Hydraulic Plants and Dams,
Chattanooga, Tennessee, October 22-27.

Roberson, J.A., J.J. Cassidy, and M.H. Chaudry (1998), Hydraulic Engineering, John
Wiley and Sons, Inc. New York, New York.

Roberts, D.F. (1977), “Energy Dissipation by Dam Crest Splitters,” Transactions of the
South African Institute of Civil Engineers, Johannesburg, South Africa, November.

“Rock Testing Handbook,” U.S. Army Corps of Engineers, Waterways Experiment

Station, Vicksburg, MS., (1990).

Sarkaria, G.S. (1997) “Lessons from Serious Incidents at Seven Arch Dams” Proceedings
of the 1997 Annual Conference of the Association of State Dam Safety Officials.

Scott, G.A. and Von Thun, J.L. (1993), “Interim Guidelines Geotechnical Studies for
Concrete Dams (Draft),” U.S. Bureau of Reclamation, Denver.
Semenza, C., DiBrai, L., and Capra, U. (1977), "Ambiesta Dam", Sixth International
Congress of Large Dams, Question 22.

11-171
Semenza, E. (1985), “Synthesis of Geological Studies of the Vajont Landslide from 1959
to 1964,” App. G, Technical Report GL-85-5, The Vaiont Slide, A Geotechnical Analysis
Based on New Geologic Observations of the Failure Surface, by Hendron, A.J. and
Patton, F.D., U.S. Army Corps of Engineers, Waterways Experiment Station, Vicksburg,
MS.

Serafim, J.L., and Pereira, J.P. (1983), “Considerations of the Geomechanics

Classification of Bieniawski,” Proceedings of the International Symposium on
Engineering Geology in Underground Construction, Laboratorio Nacional Engenharia
Civil, Lisbon.

Serafim, J.L. (1987), “Malpasset Dam Discussion - Remembrance of Failures of Dams,”

Dam Failures, Proceedings of the International Workshop on Dam Failures, Leonards,
G.A. (Ed.), Perdue University, West Lafayeete, Indiana, Aug. 6-8, 1985, Elsevier, N.Y.

Serafim, J. L. (1987), "A Note on the Earthquake Performance of Arch Dams,"

Proceedings of China-US Workshop on Earthquake Behavior of Arch Dams, Beijing,
China.

Stanton, T. E. (1940),"Expansion of Concrete Through Reaction Between Cement and

Aggregate," Proceedings, American Society of Civil Engineers, December.

Tan, H. and Chopra, A. K. (1995a). "Earthquake Analysis of Arch Dams Including Dam-
Water-Foundation Rock Interaction," Earthquake Engineering and Structural Dynamics,
vol. 24 (11), pp. 1453-1474.
Townsend, (1965,. "Control of Cracking in Mass Concrete Structures," Engineering
Monograph No. 34, U.S. Department of the Interior, US Bureau of Reclamation, Denver,
CO.

U.S. Army Corps of Engineers (1990), “Hydraulic Design of Spillways”, Engineer

Manual EM 111-2-1603, US Army Corps of Engineers, Washington DC

U.S. Army Corps of Engineers (1996) , Ice Engineering, EM-1110-2-1612, Headquarters,

Department of the Army, Office of the Chief of Engineers, Washington D.C.

U.S. Army Corps of Engineers (1964), Structural Design of Spillways and Outlet Works,
Engineer Manual EM 1110-2-2400, Headquarters, Department of the Army, Office of the
Chief of Engineers, Washington D.C.

U.S. Army Corps of Engineers (1990), HEC-2, Water Surface Profiles, Hydrologic
Engineering Center, Davis, California.
U.S. Army Corps of Engineers (1984), "Geotechnical Investigations," Engineer Manual
EM 1110-1-1804, US Army Corps of Engineers, Washington, D.C.

11-172
U.S. Army Corps of Engineers (1994), " Arch Dam Design," Engineer Manual EM
1110-2-2201, US Army Corps of Engineers, Washington, D.C.
U.S. Army Corps of Engineers (1994), “Rock Foundations," Engineer Manual EM
1110-2-2908, US Army Corps of Engineers, Washington, D.C.

U.S. Army Corps of Engineers , "Instrumentation for Concrete Structures," Engineer

Manual EM 1110-2-4300, US Army Corps of Engineers, Washington, D.C

U.S. Bureau of Reclamation (1938), Model Studies of Spillways, Boulder Canyon

Project Final Design Reports, United States Department of the Interior, Denver,
Colorado.

U.S. Bureau of Reclamation (1977). " Design of Arch Dams," Design Manual for
Concrete Arch Dams, US Department of the Interior, Denver, CO.

U.S. Bureau of Reclamation (1987). "Concrete Dam Instrumentation," US Department

of Interior, Denver, CO.

U.S. Bureau of Reclamation (1993), Dam Foundation Erosion, Survey of the Literature,
Hydraulics Branch, Research & Laboratory Services Division, Denver Research Center,
Denver, Colorado.

Veesaert C.J., and LaBoon J.H. (1995), "The use of performance parameters in
monitoring the safety of dams experiencing alkali-aggregate reaction," USCOLD 2nd
International Conference on Alkali-Aggregate Reactions in Hydraulic Plants and Dams,
Chattanooga, Tennessee, October 22-27.

Veronese, A. (1937), “Erosioni de Fondo a Valle de uno Scarico,” Annali dei Lavori
Publicci, Vol. 75, N0. 9, September, Italy.

Warnock, J.E. (1950), “Hydraulic Similitude,” Chapter II of Engineering Hydraulics,

(H. Rouse, Editor), John Wiley and Sons, Inc., New York, New York.

Westergaard, H. M. (1933), "Water pressures on Dams During Earthquakes,"

Transactions, American Society of Civil Engineers, Vol 98.

Wilson, E.L. Der Kiureghian, A. Bayo, E.R. "A Replacement for SRSS in Seismic
Analysis", Earthquake Engineering, and Structural Dynamics, Vol. 9, pp 187-192 (1981)

Xuehai Luo, Z. Wei, H. Wanchun, and D. Jinglin, (1987), "Field Vibration on Banxia
Arch Dam by Rocket Exciter," Proceedings of China-US Workshop on Earthquake
Behavior of Arch Dams, Beijing.

Zhang, L.-P. and Chopra, A.K. (1991), "Impedance Function for Three-dimensional
Foundations Supported on an Infinite Long Canyon of Uniform Cross-Section in a

11-173
Homogeneous Half-Space," Earthquake Engineering and Structural Dynamics, vol. 20
(11), pp. 1011-1027.

Zienkiewicz, O.C., (1971), "The Finite Element Method in Engineering Science," 2nd
ed., McGraw-Hill, New York.

11-174
 CHAPTER 14

DAM SAFETY
PERFORMANCE MONITORING PROGRAM

Revision 0 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

TABLE OF CONTENTS

14.1 INTRODUCTION ............................................................................................... 1

14.2 INSPECTION PROCESS and COORDINATION .......................................... 2
14.2.1 Scope and Purpose ................................................................................................ 2
14.2.2 Description and Interrelationship of Dam Safety Program Elements Using a
Potential Failure Mode Analysis Approach .......................................................... 2
14.2.3 Process flowchart .................................................................................................. 4
14.3 POTENTIAL FAILURE MODE ANALYSIS .................................................. 5
14.3.1 Introduction........................................................................................................... 5
14.3.2 Description............................................................................................................ 5
14.3.3 Key Goals and Typical Outcomes ........................................................................ 6
14.3.4 Conduct of the “Initial” Potential Failure Mode Analysis .................................... 8
14.3.5 Use of the Potential Failure Mode Analysis report as a support document to the
conduct of the FERC Part 12 Dam Safety Inspection......................................... 21
14.3.6 Updating the Potential Failure Mode Analysis ................................................... 22
14.4 PERFORMANCE MONITORING PROGRAM ........................................... 23
14.4.1 Principles and Methods of Performance Monitoring.......................................... 23
14.4.2 Performance Monitoring Procedures and Guidelines ......................................... 26
14.5 EMERGENCY PREPAREDNESS .................................................................. 36
Appendix A ..................................................................................................................... 14-A-1
Example Potential Failure Mode Descriptions ..................................................................... 1
Appendix B ..................................................................................................................... 14-B-1
Potential Failure Mode Identification Cover Letter and Questionnaire................................ 1
Appendix C ..................................................................................................................... 14-C-1
A Typical Potential Failure Mode Analysis Session ............................................................ 1
Appendix D .....................................................................................................................14-D-1
Suggested General Format for Potential Failure Mode Analysis Reports ............................ 1
Appendix E ..................................................................................................................... 14-E-1
Major Findings and Understandings - Example Write Up ................................................... 1
Appendix F ......................................................................................................................14-F-1
Estimated Time Requirements.............................................................................................. 1
Appendix G .....................................................................................................................14-G-1
Potential Failure Modes Analysis (PFMA) Dam Safety Evaluation Process ....................... 1
Appendix H .....................................................................................................................14-H-1
Part 12D Safety Inspection Report Outline .......................................................................... 1
Appendix I ....................................................................................................................... 14-I-1
Guidelines for Supporting Technical Information................................................................ 1

Revision 0 i April 11, 2003

Chapter 14 Monitoring the Performance of Dams

14.1 INTRODUCTION

The guidelines presented in this chapter provide recommended procedures and criteria to
develop a Performance Monitoring Program based upon "failure mode thinking" to assist in
reviewing and evaluating the safety and performance of water retaining project works
regulated by FERC. The procedure includes:
• A Potential Failure Mode Analysis (PFMA); and
• Development of a Performance Monitoring Program (PMP).
The Potential Failure Mode Analysis is conducted jointly by the licensee, Independent
Consultant and FERC staff. For the most part the PFMA is a one-time exercise. Guidance
on conducting a PFMA is provided in Section 14.3
Based upon the results of the PFMA, the Performance Monitoring Program is developed.
The PMP defines the appropriate monitoring for the water retaining project works based
upon the PFMA. An integral part of the PMP is the integration of the licensee’s operation,
maintenance and inspection programs.
In addition, the Part 12D Independent Consultant’s inspection and report and the FERC’s
inspection program will also be focused using the PFMA and the PMP.
The integration of a Potential Failure Mode Analysis with a Performance Monitoring
Program, results in a more efficient and effective dam safety program. With the knowledge,
vision, and understanding gained from a PFMA, the PMP will be highly effective. The
added value to dam safety includes:
• Uncovering data and information that corrects, clarifies, or supplements the
understanding of potential failure modes and scenarios;
• Identifying the most significant potential failure modes;
• Identifying risk reduction opportunities;
• Focusing instrumentation, monitoring and inspection programs to provide
information on the potential failure modes that present the greatest risk to the safety
of the dam; and
• Developing operating procedures to assure that there are no weak links that could
lead to mis-operation failures.
Although the traditional emphasis of Part 12D inspections has been on project dams,
18CFR12.32 specifically states that all project works with the exception of transmission and
transformation facilities and generating equipment are to be included in the inspection by the
independent consultant. In addition, certain other water retaining structures such as canals,
flumes, tunnels and penstocks may impact public safety if they were to fail. Accordingly,
these types of project works may also warrant a PFMA. In this document dam and project
works may be used interchangeably to designate those licensed project works that could
impact public safety in the event of a failure.

Revision 0 14-1 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

14.2 INSPECTION PROCESS and COORDINATION

14.2.1 Scope and Purpose

To define the roles, responsibilities and coordination of the Licensee, Independent Consultant
and FERC and to develop a process flowchart which links together all of the inspection,
analysis, evaluation and emergency action planning elements of the FERC’s dam safety
program using a potential failure mode analysis approach.

14.2.2 Description and Interrelationship of Dam Safety Program Elements Using a

Potential Failure Mode Analysis Approach
1. Daily routine inspections / observations - Persons performing the routine inspections
or observations should be provided with background information on the potential
failure modes identified for the site along with a performance monitoring and visual
surveillance plan for each potential failure mode. The licensee is responsible for
performing these inspections and for coordinating with the FERC resolution of any
issues discovered during the inspections. After a discussion with FERC, a decision
will be made whether any action such as analysis, repairs or monitoring needs to be
implemented.
2. Licensee operation and maintenance inspection and training programs - Those
persons performing the inspections or observations should be provided with
background information on the potential failure modes identified for the site along
with perfo rmance monitoring and visual surveillance plan for each potential failure
mode. The licensee is responsible for ensuring that its personnel are properly trained
and remain current in the knowledge of proper operation and maintenance of the
project. Any deficiencies in these matters need to be coordinated with FERC.
3. FERC operation inspection - FERC will schedule with the licensee in advance and
perform this inspection. After the inspection FERC will discuss with Licensee any
concerns found during the inspection. The discussion will also include various items
relating to the project, such as the operation and maintenance of the project, any
instrumentation and monitoring currently at the project and the emergency action plan
that is in place at the project site.
If during the FERC operation inspection a new potential failure mode is identified,
the FERC will provide this information to the licensee in the Operation Inspection
follow-up letter. If the potential failure mode needs to be evaluated prior to the next
Part 12D inspection, a schedule will be established to accomplish this. If it is
determined that evaluation of the potential failure mode may be delayed until the next
Part 12D Inspection, the FERC will include the request in its one year reminder letter
to the licensee.
FERC will document this inspection.

Revision 0 14-2 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

4. Joint Part12D and FERC Operation Inspection - Every 5 years a joint inspection will
be made by the Independent Consultant and FERC with proper coordination and
support of the licensee.
The FERC Operation Inspection will be done at the same time that the Part 12D
Inspection is done.
The Consultant will be provided the current Potential Failure Mode Analysis (initial
plus any updates). The first ½ to 1 day will be devoted to a meeting between the
necessary licensee representatives and the consultant to review the project history
including any past or current deficiencies, completed remediation, special
investigations previously completed, instrumentation, etc. The group will discuss the
development of performance parameters and potential failure modes.
The FERC’s operation inspection and the consultant ’s Part 12D inspection, though
conducted concurrently, will take place and be done independently. It is intended that
the inspections allow opportunities for discussions of any problem areas and other
important items that might come up.
Upon completion of the inspections, the group will meet to discuss any additional
thoughts concerning the performance parameters and the potential failure modes to be
developed.
The performance parameters and the potential failure modes will be prepared by the
Independent Consultant and included as appendices to the Part 12D report.
5. FERC Construction and Special Inspections - FERC will be responsible for
performing and documentation of these inspections on as needed bases with proper
coordination with the Licensee.
6. Licensee Initiated and FERC Directed Analyses and Evaluations - If during the
operation or inspection of the project a concern or issue is raised that requires
additional studies it is the responsibility of the party identifying the concern or issue
to initiate a discussion with all parties involved. If policies change as to the design
standards, FERC may direct further analyses and evaluation to determine if a
deficiency exists.
7. Recommended Action
Performance monitoring - If after the PFMA a concern or issue is thought to require
monitoring to determine if the dam's performance is at risk, it will be the
responsibility of the Licensee to install, monitor, and evaluate monitoring
instrumentation with the coordination of the Consultant and FERC.
Modification - If after the PFMA a modification is required it will be the
responsibility of the Licensee to design and make the necessary modifications, with
the coordination of the Consultant and FERC.

Revision 0 14-3 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

14.2.3 Process flowchart

CONSULTANT LICENSEE FERC

Daily Routine
Daily Routine

Inspections
Inspections

Coordination if Coordination if
Issues
Required Required

Operation,
O&M and Training

Maintenance &
Training

Coordination if Coordination if
Issues
Required Required
FERC Op Inspection

Coordination / Operation
Support Inspections

Coordination if
Coordination Issues
Required

Part 12D Coordination / FERC 5-Yr

Joint Inspection - Part 12D
& FERC 5-Yr Operation

Inspection Support Operation

Potential Failure
Team Support Team Support
Modes Analysis

Background Information Inspection Reports Instrumentation Data

Design Drawings Foundation Investigations
Construction Reports EAP

Revision 0 14-4 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

14.3 POTENTIAL FAILURE MODE ANALYSIS

14.3.1 Introduction
Potential Failure Modes Analysis is intended to be a tool utilized in the context of the
existing Part 12D program of dam and project works safety evaluation. Traditional dam and
project works safety evaluations have tended to focus on a limited number of “standards
based” concerns such as hydraulic capacity of spillways and stability of structures under a set
of pre-defined load conditions. PFMAs are intended to broaden the scope of the safety
evaluations to include potential failure scenarios that may have been overlooked in past
investigations. By definition, a Potential Failure Mode Analysis is an exercise to identify all
potential failure modes under static loading as well as all external loading conditions for
water retaining structures and to assess those potential failure modes of enough significanc e
to warrant continued awareness and attention to visual observation, monitoring and
remediation as appropriate.
This section provides the following:
• A brief description of a Potential Failure Mode Analysis;
• A listing of the key goals and outcomes anticipated from a PFMA;
• Guidance for the conduct of a PFMA is given in two ways:
o A brief statement of the expectations and requirements for a PFMA
o Detailed, step by step guidance for the conduct and documentation of a PFMA
modeled after a procedure that has been successfully used for conducting PFMAs
on a large number of dams.
These descriptions are intended to serve as guidance for the conduct of the “initial”
Potential Failure Mode Analyses to be carried out on all FERC regulated projects
subject to Part 12, Subpart D Safety Inspections
• A description of the intended application of the results of the PFMA as a support
document for conducting the FERC Part 12 Dam Safety Examination with specific
emphasis on the development of the Performance Monitoring Program for the project;
and
• A description of the process for “updating of the PFMA” by future Part 12
independent consultants.

14.3.2 Description
A Potential Failure Mode Analysis is an informal examination of “potential” failure modes
for an existing dam or other project work(s) by a team of persons who are qualified either by
experience or education to evaluate a particular structure. It is based on a review of existing
data and information, first hand input from field and operational personnel, site inspection,
completed engineering analyses, identification of potential failure modes, failure causes and
failure development and an understanding of the consequences of failure. The PFMA is
intended to provide enhanced understanding and insight on the risk exposure associated with

Revision 0 14-5 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

the dam. This is accomplished by including and going beyond the traditional means for
assessing the safety of project works and by intentionally seeking input from the diverse team
of individuals who have information on the performance and operation of the dam. A PFMA
includes and uses all of the available data and information from a standard engineering
analysis of an existing dam. A PFMA should be viewed as a supplement to the traditional
process in which a dam’s safety is judged upon its ability to pass standards-based criteria for
stability and other conditions.
Utilizing an intensive team inquiry beginning from a basis of no preconceived notions, the
potential failure mode examination process has the ability to:
• Enhance the dam safety inspection process by helping to focus on the most critical
areas of concern unique to the dam under consideration
• Identify operational related potential failure modes and structural related potential
failure modes (e.g. piping) not covered by the commonly used analytical methods
(e.g. slope stability, seismic analysis)
• Enhance and focus the visual surveillance and instrumented monitoring program
• Identify shortcomings or oversights in data, information or analyses necessary to
evaluate dam safety and each potential failure mode
• Help identify the most effective dam safety risk reduction measures.
• If the study is documented and used for guidance on future dam safety inspections
and is updated (as a living document) then the benefit (of increased understanding
and insight) lives on.

14.3.3 Key Goals and Typical Outcomes

The primary product and the main focus of a Potential Failure Mode Analysis is identifying
and obtaining a clear understanding of each dam’s – site specific - potential failure modes.
The potential failure mode “identification” is intended to go beyond a simple generic
statement of the potential problem (e.g. – operations, piping, slope instability, foundation,
overtopping, liquefaction, etc.). The potential failure mode identification, examination and
description provides background information on the loadings and structure conditions,
circumstances and events at each site that identify why this potential failure mode is being
considered for this site. Also the significance of this potential failure mode for the site in
terms of the need for awareness, for monitoring and surveillance, for analyses and
investigation or for making operational changes or structural repairs is discussed. Example
descriptions of potential failure modes that have come from actual potential failure mode
analyses are provided in Appendix A for a potential operational type potential failure mode
and for a potential piping type failure.
The Potential Failure Mode Analysis (PFMA) process is not a substitute for but rather a
guide to help focus periodic, comprehensive, dam safety inspections. Both activities require
and benefit from a comprehensive review and discussion of all available information
(historic records and photos, engineering analyses, previous inspection reports, etc.). Hence,
the detailed reviews commonly done prior to a periodic inspection, especially if an
Independent Consultant is not familiar with a project, are still necessary. Linking the

Revision 0 14-6 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

accomplishments of the PFMA and periodic inspections is efficient and effective because it
allows others, not often in the direct safety evaluation loop, to participate and contribute
importantly to the outcome.
Every organization / dam owner / dam regulator / A-E firm currently carries out or
participates in dam safety inspections for dams under their charge. It is suggested that the
inspections in the future should incorporate “potential failure mode thinking”, and that
integration of potential failure mode analyses and dam safety inspections can be adapted to
meet the needs and resources of all dam owners.
Although potential failure mode identification is the focus product from the process there are
other outcomes that result from carrying out a PFMA in the manner described in this
guidance document.
• The process of searching out all the information about the dam for the specific
purpose of identifying potential failure modes (plus the involvement of a diverse
group of people in the PFMA process), typically results in uncovering data and
information that most personnel currently involved in the dam’s safety evaluation had
not been aware of. Frequently this information plays an important role in identifying
a potential failure mode.
• The most significant potential failure modes and failure scenarios will be identified
and documented for use and consideration by future Independent Consultants and
inspection teams.
• Certain problems, issues and concerns that have been associated with the dam may be
found to be of lesser significance than previously perceived from the standpoint of
consequence, remoteness or physical possibility.
• Enhancements to the monitoring and visual inspection programs are recognized and
readily developed. Monitoring efforts can become more focused on the important
issues.
• A wide range of persons (from the dam tender to the owner’s dam safety program
manager), become aware of the dam’s most significant vulnerabilities and the
relationship of the surveillance and monitoring programs to these vulnerabilities.
• Gaps in data, information or analyses that prevent characterizing the significance of a
potential failure mode are recognized and identified for consideration / action by the
owner.
• Non-structural risk reduction opportunities applicable to the Dam Safety Performance
Monitoring Program, operations, structure response or emergency preparedness are
recognized and identified for consideration by the owner.
• Provides the opportunity to easily and effectively educate all who are concerned with
the dam – (dam tender – owner –regulator- periodic reviewers – inspectors –
designers and others) about:
1. The potential failure modes for this dam

Revision 0 14-7 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

2. How monitoring, including use of specific instrumentation and visual surveillance

is used to look for specific symptoms, behaviors or evidence that might warn of a
developing failure for the identified potential failure modes and,
3. How “general health” monitoring (e.g. – crest monitoring, piezometers) is used as
basic data to help watch for conditions that were not identified as potential failure
modes.
4. How operations (i.e., regulated, normal, unusual) of this dam and others upstream
may influence dam safety.
5. Emergency actions that may be more commonly encountered

14.3.4 Conduct of the “Initial” Potential Failure Mode Analysis

A Potential Failure Mode Analysis is to be conducted for all FERC regulated dams that are
required to undergo Independent Consultant safety inspections as defined in 18 CFR Part 12,
Subpart D unless granted an exemption.
Specific steps and actions for carrying out a PFMA for a dam are enumerated below and
these steps are recommended, as a minimum, for a PFMA to be comprehensive, consistent,
and complete. However, in completing these specific steps it is very important that the
principles of the process be understood and followed in order for the full value of the process
to be achieved. These principles include:
• Diligence in searching for all the background information.
• An open – investigative attitude toward identifying and understanding potential
failure modes and failure scenarios.
• Dedication of the assigned persons to the reviewing / reading of all the background
information on the dam prior to the PFMA session.
• Diversity in input to the process – field personnel, operations personnel, technical
personnel, management personnel and others all contribute to the pool of information.
There is no monopoly on good ideas and key information.
• Documentation is the key to capturing the insight and ideas resulting from the
process.
• Willingness of all parties to set aside their normal hats and focus on what the data,
information, and experience / knowledge of individuals can teach us about the dam.

The FERC in association with Dam Owners and the Independent Consultants who perform
the Part 12 Dam Safety Evaluations have developed these procedures for use as a focal point
within the Part12 Examination Process. Specifically, they combine plans to improve and
focus the Dam Safety Performance Monitoring Programs for FERC regulated dams, and also
provide a fundamental enhancement to the inspection process by focusing on site-specific
factors of greatest importance at each project. The Potential Failure Mode Analysis, as
outlined below, will serve as the focal point and linking feature within the Part 12 Inspection.
Guidance for the conduct of a PFMA is given below in two ways :

Revision 0 14-8 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

1. A statement of what needs to be done – in terms of expectations and requirements.

2. Step by step procedural guidance for carrying out a PFMA that ties directly to the
statement of expectations and requirements. This is provided for those who desire
more detailed guidance.

Overall Guidance – Potential Failure Mode Analysis Expectations / Requirements

1. Collect all data, studies and information on the investigation, design, construction,
analysis, performance and operation of the project. All studies and investigation
reports existing that relate to the ongoing safety of the dam must be included and
reviewed and evaluated. A listing should be made of the data available for review and
considered in the Potential Failure Mode Analysis and included in the PFMA report
documentation.
2. Visit the project site with an eye out for potential failure modes, structural and
geologic conditions, review operations, and interview owners/operators for their
input on potential failure modes
A core team of at least 3 persons experienced in dam safety evaluation (familiar with
dam failure mechanisms) are to review all the background information for general
understanding and with these specific questions in mind:
• How could this dam fail? (Site-specific consideration of loadings, structure
condition, and project operations )
• What happens if the dam fails?
• Are the identified potential failure modes recognized and being appropriately
monitored by visual surveillance or instrumental monitoring?
• What actions (immediate or long term) can be taken to reduce dam failure
likelihood or to mitigate failure consequences? These actions could include any
of the following: data collection, analysis or investigations, operational changes,
communication enhancement, monitoring enhancement and structural remediation
measures.
3. Brainstorm potential failure modes and failure scenarios with a team of persons most
familiar with design, analysis, performance, and operation of the dam. Record the
identified potential failure modes, the reasons why each potential failure mode is
favorable / less likely and adverse / more likely to occur and identify any possible
actions related to each that could help reduce risk (i.e. monitoring enhancement,
investigation, analysis, and/or remediation).
4. Specifically identify possible performance monitoring enhancements for each
potential failure mode for consideration of the owner and the Independent Consultant
in the Part 12D report.
5. Document the analysis, including immediately recording the major findings and
understandings from the brainstorming session.

Revision 0 14-9 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Procedural Guidance – Potential Failure Mode Analysis Step by Step Guidance

Step 1 Designation of the Potential Failure Mode Analysis Participants
Step 2 Collection of Background Data on the Dam for Review by the Core Team
Step 3 Site visit, interviews with key owner personnel at the Project and
comprehensive review of the Background Data on the Dam by the Core Team
Step 4 Conduct the PFMA Session
Step 5 Consideration of Performance Monitoring Opportunities for Identified
Potential Failure Modes - (Note the Performance Monitoring Plan for
Identified Potential Failure Modes is provided to the owner by the
Independent Consultant in the Part 12D report)
Step 6 Documentation of the PFMA and Performance Monitoring Requirements

The following sections describe each step in detail:

Step 1 – Designation of the Potential Failure Mode Analysis Participants

The potential failure mode analysis participants (team members) consist of all those who will
participate in the brainstorming session in which potential failure modes are identified,
defined, discussed and categorized. Fundamentally these are persons who have experience
with the design, construction, analyses, performance and operation of the dam. A dam-
experienced engineering geologist should be a part of the team and should be included in the
site visit. The primary advantage of having a variety of people participate in the potential
failure mode identification process (and it is a very significant advantage) is that more ideas
and more questions are put forward, more knowledge and more information is available and
a greater diversity of opinion is input to the process.
Some of the team members have specific roles and responsibilities and need to have the
requisite experience and capability to fulfill these roles. These roles and requirements are
given below:

Team Leader - The dam owner would designate one of the participants as the team
leader, responsible for coordination activities – including coordination of the collection of
background information.

Core Team - At least three of the participants are designated as the “core team members.
They are the designated “readers of all background material”. The core team members
are each assigned the responsibility of reviewing / reading all the background information
collected for the dam. One of the core team members will facilitate the PFMA session
and one will be responsible for documentation of the Potential Failure Mode Analysis
report. The team leader is not necessarily one of the designated “readers of the material”
because the coordination / logistic activities often will divert the Team Leader’s attention
away from the reading and study requirements. Exceptions to this general guidance may
be made if there is no other practical alternative and the team leader is judged to be
ideally suited for the core team (see criteria given below).

Revision 0 14-10 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

The team will generally consist of the following four persons:

• The Independent Consultant(s) who will do the current Part 12D inspection
• Representative(s) of the Owners Staff (i.e., engineer, field operations person)
• The FERC inspector for the dam
• The Facilitator designated for the Potential Failure Mode Analysis session

The following criteria should be considered when selecting the core team members:
• The core team members should have knowledge and experience related to dam
safety evaluations. It is especially helpful to have persons who have interest and
knowledge related to dam failures and who have an inquisitive / investigative
personality (they think like coroners or detectives).
• The facilitator would, in general, be new with respect to examining the dam’s
operation and history. This is considered an advantageous situation with respect
to providing a fresh and vigorous look at the structure.
• Dam owner representatives who have the knowledge, skill and interest and who
gain the requisite experience to serve as facilitators are encouraged to do so via an
exchange program with other dam owners. Dam owners facilitating the PFMA on
their own structures would not in general be considered appropriate.
• The Independent Consultant may or may not be new to the facility, but like the
facilitator must have extensive experience in dams and an open mind relative to
identification of potential failure modes. In accordance with current regulations
the Independent Consultant must still meet FERC requirements and be approved
by FERC.

Facilitator Requirements - The Potential Failure Mode Analysis (PFMA) facilitator

should be a civil engineer with a broad background and experience in dam safety
engineering and experience in performing a PFMA similar to that described in this
guidance. A basic recommended qualification for the facilitator is that the proposed
facilitator for a project should have been involved in an actual PFMA of the nature
described in these guidelines.

Qualifying experience is participation as a core team member of a PFMA or actually

facilitating a PFMA. This ensures that the person leading the PFMA process knows not
only how the process is carried out, but also is aware of what can be accomplished. This
is especially critical if the other core team members have not been through a PFMA
which may often be the case. As an alternative to actual experience participating or
facilitating a PFMA, the proposed facilitator should have attended an FERC sponsored
Dam Safety Performance Monitoring Program Training Workshop. FERC will
periodically provide training opportunities to help develop facilitators, especially during
the implementation phase of this new program.

Revision 0 14-11 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

It is important to understand that if the facilitator does not accomplish the goals of the
PFMA, which is identifying and obtaining a clear understanding of each dams site’s
specific potential failure modes, the PFMA may be required to be supplemented or
redone entirely.

The facilitator is to serve as the peer reviewer of the PFMA report/documentation of

findings prepared by the Independent Consultant / documentation of findings. The
facilitator is to complete the Peer Review of the Major Findings and Understandings
within a 5 day period and the PFMA report within a 10 day period after the Independent
Consultant submits them to him.

Supplemental Resources - In addition to the team participants there are other people
who have specific technical knowledge or experience that may be useful to the team.
These people would be notified and asked to be available on call on the day of the PFMA
session. This would include such persons as seismo-tectonic specialists, hydrologists,
structural engineers, electrical engineers, mechanical engineers, geotechnical engineers,
field personnel, inspectors, instrumentation personnel, emergency preparedness
personnel, etc.
In formulating the team it is important to include those individuals with intimate
knowledge of the project operations and structures, especially the senior dam tenders and
those responsible for collecting monitoring data. The benefits from conducting this
exercise include not only bringing focus to the most likely modes of failure based on
engineering judgment but also through increasing the general awareness of dam safety
issues by sharing knowledge at all levels. Experience has shown that it is very helpful and
valuable to include senior (experienced) field personnel in the actual PFMA session
because all information has not been written down and in certain cases assumptions in
written reports differ from what is actually done in practice.

Step 2 - Collection and Review of the Background Data on the Dam

1. Preparation / Input by the Dam Owner’s Team Leader and FERC
The Team Leader, working in conjunction with the FERC inspector, would collect
and gather for review, all background information on the project (investigation and
design reports, boring logs, core reports, construction photos, inspection reports,
instrumentation and surveillance data, incident reports, repair plans and
specifications, etc). This data and information would be collected in a centralized
location for reading by the core team members and would also need to be available
during the PFMA Session. The types of material which should be collected (if
available) include:
• Any FERC or state agency construction inspection reports (these have been found
to be extremely useful )
• Current or most recent dam safety engineering analyses, including stability and
stress analyses

Revision 0 14-12 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

• The most recent monitoring and instrumentation data along with the historic
records of monitoring data.
• Current flood routings and any hazard / consequence analysis
• The current Emergency Action Plan
• The most up-to-date aerial photographs of the downstream areas that could
potentially be impacted by failure of the project struc tures.
• The most recent surveys for each of the project structures (i.e. horizontal and
vertical survey data). This should preferably be the survey that was conducted as
part of the current Part 12D inspection.
• The most recent underwater inspection report. This should preferably be the
underwater inspection that was conducted as part of the current Part 12D
inspection.
• Recent meteorological and pertinent river gage records
(http://waterdata.usgs.gov/nwis).
• The most recent seismic loading parameters that have been prepared for the site
and print records of recent seismic activity (http://neic.usgs.gov/).
(Note: Basic demographic, seismic, meteorological and/or stream flow data should
be reviewed to ensure that previous findings or assumptions related to potential
failure mode hazards or consequences are up to date. Hence, recent data and
information should be brought to the session or generated at the session as necessary.
This will ensure that the PFMA report is an accurate representation of the likely
potential failure modes and consequences based on the best information that was
available on the date the exercise was conducted.)
A listing of the data available for review and considered in the Potential Failure Mode
Analysis should be included in the PFMA report documentation.
An advance review package on the dam would be prepared for all participants – this
package would consist of material already prepared that provides an overview of the
dam and its performance. The purpose of an advance package is twofold: to give the
facilitator familiarity with the dam prior to the site review and to refresh knowledge
of the dam and stimulate “potential failure mode thinking” by all participants prior to
the PFMA session. The previous Part 12 D Inspection report is a good “advance
package document” to provide to the facilitator and the core team (and any other
proposed participants) for familiarization with the project prior to the site review.
The owner should establish a means to retain / archive all the information collected
for the PFMA
2. Core team members are to review all of the above information searching for site
specific conditions or situations that would lead to uncontrolled release of the
reservoir or other incidents, conditions or situations that would have an adverse
impact. This review of materials is scheduled to occur following the site visit and
discussion with project personnel.

Revision 0 14-13 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

3. A questionnaire on potential failure mode identification and performance monitoring

is to be sent by the Team Leader to all PFMA participants and support personnel
(Appendix B provides an example questionnaire along with a draft note to be sent
explaining the request for information). Note that prior to the PFMA session, team
participants, other than the core team members, are only required to complete this
questionnaire, review their own files (re-acquaint themselves relative to the work
within their area of expertise), and bring their historic knowledge of the project to the
session. Only the core team members are responsible for reading all the historical
and technical documents related to the project.

Step 3 - Site Review of the Dam and Project

1. The detailed Part 12 Inspection of the project will be performed and the
accompanying report prepared by the Independent Consultant following the Potential
Failure Mode Analysis. However prior to the initial PFMA session, a review of the
site, “thinking” potential failure modes, is carried out with the owners personnel and
includes the facilitator, the independent consultant, the FERC representative, the
owner’s core team representative, (these 4 comprise the core team), and an
appropriate geologist for the project. Owner’s may find it valuable to include all or
most of the employees that they plan to have participate in the PFMA also participate
in the site review session.
2. The advance review package should be sent to site review participants prior to their
travel to the site. Typically the site review performed in association with the
Potential Failure Mode Analysis should be scheduled just prior to the PFMA just
before the core team members review the background materials. Such a schedule
takes greatest advantage of the interaction between potential failure mode analysis
and site visitation.
3. The site review should include the opportunity to visit with field maintenance
personnel and plant operators, including but not limited to those who will be team
participants.
4. The comprehensive review of background data and information on the dam by the
core team is scheduled to occur following the above site visit and discussion with
project personnel.

Step 4 - Conduct of the Potential Failure Mode Analysis Session

A brief description of the Potential Failure Mode Analysis Session is given below – a more
comprehensive example of a typical session is given in Appendix C. It is important for the
facilitator to involve all participants in the discussions and give everyone an opportunity to
provide their knowledge, understanding and views on the potential failure modes,
consequences and possible risk reduction actions / measures.

1. Consider the possibilities for failure, loading by loading condition (static reservoir,
hydrologic, seismic, ice, debris impact and any other loading relevant to the site) for
each component of the project (main dam, spillway, gates, dikes, outlet works, power
plant, etc.). Consider how an uncontrolled release of the reservoir or a dam breach

Revision 0 14-14 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

could occur. Also consider total system operation aspects (communication and
response [i.e., personnel, remote telemetry], facility access, weather conditions,
equipment) with respect to the possibility of their contribution to development of a
potential failure mode/failure scenario.
2. Team participants are asked to identify “candidate” potential failure modes. A
candidate potential failure mode is discussed until a clear characterization of a
potential failure mode and failure scenario is developed. However, sometimes the
initial suggestion may lead to two or more separate or related potential failure modes,
which need to be developed separately. Or sometimes the idea brought up as a
‘candidate” and discussed is not developed as potential failure mode. Such ideas are
termed “other considerations” and should be noted and documented as part of the
PFMA. (see Appendix D and the documentation section below)
3. Once a candidate potential failure mode has been characterized / described such that
there is a common understanding of the potential failure mode, (See Appendix A –
Part 1 for example potential failure mode descriptions.) The potential failure mode
description is noted on a flip chart by the facilitator and should be recorded in detail
by the Independent consultant at that time; then the potential failure mode and failure
sequence is discussed. The nature of the breach (or other failure condition) is defined
and the potential consequences of failure are discussed. All the data, information,
factors and conditions that suggest the ways that the potential failure mode is more
likely or less likely to occur (adverse factors and positive factors) are noted down.
(See appendix A – part 2 for an example) Also during this discussion possible actions
to be taken may be suggested:
• opportunities for risk reduction,
• possible investigations or analyses,
• means for monitoring/inspecting for the development of potential failure modes

All of this information is noted (in brief) on a flip chart to facilitate documenting the
suggestions.
The consequences of failure and the circumstances surrounding a failure (advance
warning, detection possibilities, impact of the failure, etc.) should be discussed for
each potential failure mode during the discussion of the potential failure mode since
these factors play a role in assessing how significant the potential failure mode is.
However, experience has shown that it is necessary, valuable and instructive to
specifically raise the topic of “consequences” as part of the PFMA and brainstorm
site-specific factors and potential failure mode consequence related factors (in the
event they have been overlooked during the technical discussion of the potential
failure mode)
4. Occasionally a candidate potential failure mode is dismissed as a significant potential
failure mode without carrying out number 3 and 4 above. In such cases the PFMA
report will include its introduction under the heading of “other considerations” and
identify why the team did not discuss it in further detail.

Revision 0 14-15 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

5. When each site-specific potential failure mode is identified, the nature of the breach /
uncontrolled release that may occur is discussed and the range of failure scenarios
and consequences that may result are identified. The emergency action plan
response to potential failure scenarios is examined and any concerns with the plan are
identified.
6. After a potential failure mode has been identified, described and discussed, each
potential failure mode is classified / categorized according to the classification system
given in Table 1 below or according to a comparable system developed to meet FERC
/ dam owner needs. After all potential failure modes have been discussed the
classifications made are reviewed and discussed.
The Potential Failure Mode Analysis process incorporates a qualitative likelihood
estimate for the identified potential failure modes through the process of putting the
potential failure modes into categories. Category I Potential Failure Modes are those
considered most credible and most important to be brought to the attention of the dam
owner, dam operators, personnel performing the monitoring and personnel
performing routine and periodic inspections. Category II Potential Failure Modes are
also considered credible, in that they are physically possible, but are not highlighted
for one or more reasons such as – no direct or indirect evidence of any indication of
problem development, extremely remote loading required to initiate the potential
adverse response, etc. Category III Potential Failure Modes are those where more
information or analyses are needed in order to be classified. Category IV PFMs are
those that have been ruled out. Attention to monitoring and surveillance relates to
Category II and III potential failure modes just as it does for Category I modes.
(When the additional information/analyses required to resolve a Category III PFMA
are completed, that potential failure mode should be categorized.) Two categories of
viable potential failure modes are provided to allow the use of judgment by the team
and to provide an easy differentiation of relative importance for the owner. The
Catego ries are described as in Table 1.

7. At the close of the session, each participant takes a few minutes to note what
information or understanding was most significant to them. The facilitator then
records these major findings and understandings achieved as a result of the Potential
Failure Mode Analysis on the flip chart.

Revision 0 14-16 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Table 1 - Categories of Identified Potential Failure Modes

Category I - Highlighted Potential Failure Modes - Those potential failure modes

of greatest significance considering need for awareness, potential for
occurrence, magnitude of consequence and likelihood of adverse
response (physical possibility is evident, fundamental flaw or
weakness is identified and conditions and events leading to failure
seemed reasonable and credible) are highlighted.

Category II - Potential Failure Modes Considered but not Highlighted - These are
judged to be of lesser significance and likelihood. Note that even
though these potential failure modes are considered less significant
than Category I they are all also described and included with reasons
for and against the occurrence of the potential failure mode. The
reason for the lesser significance is noted and summarized in the
documentation report or notes.

Category III - More Information or Analyses are Needed in order to Classify These
potential failure modes to some degree lacked information to allow a
confident judgment of significance and thus a dam safety investigative
action or analyses can be recommended. Because action is required
before resolution the need for this action may also be highlighted.

Category IV - Potential Failure Mode Ruled Out Potential failure modes may be
ruled out because the physical possibility does not exist, information
came to light which eliminated the concern that had generated the
development of the potential failure mode, or the potential failure
mode is clearly so remote as to be non-credible or not reasonable to
postulate.

It is important to note that the Potential Failure Modes are placed into categories by
judgment. The basic purpose is to help the dam owner’s personnel and the current and future
inspectors dealing with the dam to understand what the evaluation team considered were the
most significant potential failure modes, so that they can consider / prioritize for action a
smaller number of items rather than the total array of potential failure modes considered.
The breakdown may also help with prioritization of actions to be taken. It is quite common
in the PFMA for a monitoring or visual inspection action to be identified, often that is easy to
implement, to be made for a Category II potential failure mode that is “considered but not
highlighted”.

Step 5 – Evaluation of Performance Monitoring Requirements

As a part of the Part 12D report the Independent Consultant will be required to present a
Dam Safety Performance Monitoring Program for the dam / project. The Dam Safety
Performance Monitoring Program will include a “Dam Safety Performance Monitoring Plan”

Revision 0 14-17 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

for each Category I Potential Failure Mode. Dam Safety Performance Monitoring Plans will
also be included for selected Category II and III Potential Failure Modes which the
Independent Consultant believes are warranted. In the Part 12D report the Independent
Consultant must explain why Performance Monitoring is not warranted for any specific
Category II or III potential failure modes. In addition any requirements for “General Health
Monitoring” independent of an identified potential failure mode will be defined. The plan
presented should consider the items enumerated below. To facilitate development of Dam
Safety Performance Monitoring Plans the Potential Failure Mode Analysis Team should
include comment and discussion on these items as appropriate for each potential failure mode
identified.

1. The type and frequency of inspections (visual surveillance requirements) should be

evaluated to address the identified potential failure modes. This item may include the
recommendation of developing customized checklists for the dam. (The nature and
content of the checklist, if recommended, is developed by the Independent Consultant
in consultation with the owner. The checklist should identify specific visual cues that
may indicate a suspected potential failure mode has activated, and the checklist
should provide instructions as to what step(s) should be taken once a cue is
observed).
2. The current instrumentation and visual surveillance program should be critiqued. In
some cases, instruments may be obsolete and serve no purpose in monitoring for the
development of a potential failure mode. In other cases additional instrumentation or
visual surveillance may be needed to monitor for a potential failure mode
development
3. Reporting requirements should be reviewed. Action limits may need to be
established for some of the instruments and procedures developed for reporting
variations in instrumentation readings. As a minimum, annual engineering review,
evaluation and reporting of the instrumentation data is required.
4. In some cases additional analyses or investigations may be required to fully evaluate
a potential failure mode prior to establishing a performance monitoring plan for it.
The PFMA team should identify what information is needed. The Part 12D
Independent Consultant would recommend what and how to obtain this information.
5. If enhancements to the monitoring or visual surveillance are identified by the
PFMA/Part 12D process then priorities for improvement in the Dam Safety
Performance Monitoring Program should be discussed within the Dam Safety
Performance Monitoring Program presentation and appropriate recommendations and
schedules provided in the Recommendations Section of the Part 12 D Report.

Revision 0 14-18 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Step 6 - Documentation of the Potential Failure Mode Analysis

1. For the knowledge gained, information obtained and results achieved in the Potential
Failure Mode Analysis to be effectively used for the current Part 12 D and for future
dam safety the Part12D inspections the documentation of the work must:
• be done promptly
• be definitive in describing the identified potential failure modes
• be complete in recording factors considered relative to the viability of each
potential failure mode considered
• discuss possible risk reduction actions identified relative to each credible
potential fa ilure mode– performance monitoring – investigations – remediation
activities
• clearly relay the major findings and understandings achieved as a result of the
process
It was specifically noted during the review of pilot study draft reports that greater
attention needs to be paid to fully stating the sequence of conditions and events that
constitute the potential failure mode and failure scenario. (See Appendix A for an
example potential failure mode description)
2. The Independent Consultant writes up the “major findings and understandings”
immediately after the session. (Within 15 days of the PFMA session.) The items
noted during the session are typically abbreviated and the major findings and
understandings should flesh out the implication of the finding or understanding
relative to the associated potential failure mode. The write up of the major findings
and understandings is then sent to the facilitator for peer review and to the other core
team members for input. (The facilitator peer review and input from the core team
should be completed within 5 days of receipt of the write up from the independent
consultant). Appendix E provides an example of a write up of major findings and
understandings resulting from a potential failure mode analysis.
3. The Independent Consultant prepares the draft Potential Failure Mode Analysis
Report (within 30 days of the PFMA) , describing each potential failure mode
considered and referencing key adverse/likely and positive/not likely factors,
identifying any suggested visual surveillance or instrumental monitoring, describing
consequences of potential failure and site-specific conditions or factors related to
consequences and noting any potential actions identified (information inquiries,
investigations, analyses or risk reduc tion opportunities). The write up should include
a brief statement as to the adequacy of the project documentation and overall quality
of the data that formed the basis of the PFMA. If prepared technical presentations of
new material, not contained in the record documents, were made by consultants
during the course of the PFMA their presentation should be documented in, or
appended to the PFMA report. Appendix D provides an example outline for the
documentation of the analysis. This outline is designed to take advantage of the
information collected on flip charts during the potential failure mode analysis session
in order to make the documentation process simple, fast and effective. The facilitator

Revision 0 14-19 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

peer reviews the draft report on behalf of the Owner/Independent Consultant (within
10 days). The peer reviewed draft report is then sent by the Independent Consultant
to each participant of the PFMA session for review and comment.
4. All reference material available and used by the team in the Potential Failure Mode
Analysis is recorded and key items of data and information (that led to important
findings or conclusions – see discussion under point 5 below) are included in an
appendix to the PFMA report for ready reference. Photos of past conditions or
photos of current conditions, elucidating key information about a potential failure
mode, are highly recommended for inclusion in the body or appendix of the PFMA
report. The PFMA appendix should be concise and not duplicate parts of the STI or
Part 12D report.
5. Preparation of a listing of the documents gathered by the owner for review, in
advance of the review, has been found to serve as a valuable tool for the reviewers to
use during their review to assure that they have seen all the materials collected and
should be included in the PFMA report.
6. The PFMA report will then become Section 1 in the Supporting Technical
Information (STI) document and the findings of the PFMA report will be discussed
and summarized in the Part 12 D report. It is not the intent of the PFMA appendix to
include the reports and documents that comprise the “background material” that was
read and used in the discussions. However, often a key paragraph, photograph, test
results or other documentation is found in a document that elucidates whether or not a
potential failure mode is more or less likely and it is valuable to include that specific
information in the PFMA appendix. (e.g. photographs may show planar joints, or
gunite treatment of the foundation, or shear keys ; statements might be made by the
consulting review board about the condition of the filter material, tests results might
provide definitive information that counters what has been stated in opinions /
observations in construction reports; erosion or the lack of it may have been
documented following a flood). These specific pages, photos, quotations or data that
provide direct support to the “likely” or “not likely” aspects of a potential failure
mode should be reproduced and included in the appendix to the PFMA report.
7. The report should state whether the findings are a consensus of the team. If not a
consensus, the reasons for differences of opinion should be documented in the report
findings.
8. Other Considerations – thoughts / ideas / concepts / future changes that were
considered related to possible potential failure modes that were brought up and
discussed but not developed by the PFMA team as a potential failure mode should be
documented in the section “Other Considerations” such that future teams will know
what items were considered and why they were not carried forward as a potential
failure mode at that time.
9. The report should include an assessment of the overall adequacy, completeness and
relevance of background data that was furnished for the Potential Failure Mode
Analysis, identify any discrepancies, inaccuracies, or deficiencies in the records, and
determine if adequate information was provided to conduct the PFMA. The report

Revision 0 14-20 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

should document any potential shortcomings in the PFMA due to lack of sufficient
data for consideration of specific potential failure modes.

Appendix G provides the PFMA process in a task by task table format for dam owners as
a supplement to the above discussion format for their convenience if desired.

14.3.5 Use of the Potential Failure Mode Analysis report as a support document to the
conduct of the FERC Part 12 Dam Safety Inspection
Appropriate sections of the Part 12D inspection and report should provide commentary
and/or information that relates to and addresses the potential failure modes identified in the
PFMA.
The manner in which that is intended to be accomplished is outlined in general terms below:
1. General Observations – The Part 12, Subpart D Independent Consultant should make
observations of project features independent of the PFMA. It is important that the
consultant keeps an open mind during the Part 12D Dam Safety inspection and be
alert for any unusual conditions that may not have been identified in the PFMA. The
purpose of the Part 12D inspection is not to only inspect for those conditions that may
develop as described in the PFMA but to document the actual condition of the project
structures. However, in addition to making these necessary observations, the
consultant will also now need to pay special attention to those issues that were
identified in the PFMA.
2. Inspection - The inspection report would include a discussion of the observations
relative to each of the identified potential failure modes as well as the Independent
Consultant’s own assessment on the significance of the identified potential failure
modes and on whether any other potential failure modes exist, or conditions may have
changed that would impact previous conclusions regarding potential failure modes
3. Historic and current performance indicators - Any relevant comments relating these
factors to identified potential failure modes are provided.
4. Performance Monitoring - Each potential failure mode identified shall be reviewed
to determine whether visual surveillance or instrument monitoring is adequate to
detect the onset of the potential failure mode or the onset of conditions which may
contribute to or “allow” development of the potential failure mode
5. Recommendations – Actions that could be taken with regard to information inquiry,
investigations, analyses, or structural or non-structural actions shall be discussed in
terms of the identified potential failure modes. Recommendations may pertain to
changes in operations or maintenance required in order to maintain the status quo.
6. Emergency Preparedness – discussion related to identified Potential Failure Modes
7. Independent Consultant’s Commentary on the Potential Failure Modes Identified in
the Potential Failure Mode Analysis - This section of the Part 12D report is provided
to allow the Independent Consultant the discretion to place emphasis on or to de-
emphasize any of the “team findings” presented in the Potential Failure Mode
Analysis report. It also allows for incorporation of any new information, results of

Revision 0 14-21 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

analyses, or other findings that come to light during the Independent Consultant’s
inspection and report.

14.3.6 Updating the Potential Failure Mode Analysis

The comprehensive “initial” Potential Failure Mode Analysis and the resulting section in the
STI appended to the Part12D report described above is intended to be performed only once
for each project (or at extended intervals (e.g. - 15-20 years), but it should be regarded as a
living document to be appended as conditions at the site change or as new information is
obtained at any time following the initial PFMA or discovered during subsequent Part 12,
Subpart D inspections.
If the initial Potential Failure Modes Analysis is successfully performed, then that report will
serve as a key document and foundation for the Independent Consultant Inspection in
subsequent Part 12 inspections. (Availability of this document should make the Independent
Consultant ’s work easier, more focused and effective and less costly) If as a result of the
detailed inspection, the Independent Consultant finds new or varying information or has a
professional opinion that necessitates revision of the findings of the original PFMA, the
Independent Consultant would append such revisions to the existing report, and the appended
/ updated PFMA document would be incorporated in the STI which accompanies each Part
12D inspection report. That “updated” PFMA would then again be the foundation for next
Part 12D Independent Consultant inspection report 5 years later.
It is also possible that new informa tion would come to light in the interim between the Part
12 inspections – the owner and FERC would append that information to the original PFMA
Report in a similar fashion. In this way, the Potential Failure Mode Analysis report as
maintained in Section 1 of the STI is a living document that will document the progression
and variety of analyses and professional opinions that went into the current updated /
appended PFMA report findings.
It is important to retain the original PFMA report as prepared so that the findings and
discussions so that the thought processes at that time are retained for future evaluations.

Revision 0 14-22 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

14.4 PERFORMANCE MONITORING PROGRAM

Monitoring the performance of the dam / project to assure that possible dam failures are
avoided or adequate warning time is provided is an essential part of a dam safety program.
The procedures outlined in Section 14.3, above, provide guidance on developing a Potential
Failure Modes Analysis for a dam.
These guidelines:
• Provide a discussion of the various performance monitoring principles and methods
used to aid in evaluation of a structure; and
• Present performance monitoring procedures and principles for a number of common
adverse responses or conditions that typically are indicators or contributors to
potential failure modes. These basic principals and procedures provide general
guidance that is then made specific for an individual dam for the potential failure
modes identified as part of the PFMA process.
As part of the Part 12D report the Independent Consultant shall assess the PMP for the dam /
project. The PMP will include a “Performance Monitoring Plan” for each Category I
Potential Failure Mode. Performance Monitoring Plans will also be included for each
selected Category II Potential Failure Mode that the Independent Consultant believes is
warranted. In addition, any requirements for “General Health Monitoring” independent of an
identified potential failure mode should be identified. Chapter 9 of the Engineering
Guidelines provides guidance on the level of instrumentation necessary for monitoring the
general health of a dam.
The adverse responses and conditions and the companion monitoring procedures and
principles described in this guideline should not be considered as complete, as each dam will
have its own characteristics. ALL combinations of failure, and particularly operating
conditions that may present more complex potential failure modes and failure scenarios, must
be developed and the appropriate means for monitoring these unique or complex modes
established.

14.4.1 Principles and Methods of Performance Monitoring

This section describes fundamental principles and methods used to aid in the evaluation of
the performance of a dam. Performance is assessed through evaluation of the visual
observations and instrument data relative to design response expectations and subsequent
observations of structural behavior.
1. Visual observation
Visual observation is an important surveillance activity. Many dams were constructed
without the benefit of instrumentation and thus visual observation offers a first
impression to evaluating integrity, movement and loads. Visual observation at regular
intervals by trained personnel will often detect unusual conditions, such as increased
seepage, or cloudy seepage, or movements and is the dam owner’s primary defense
against serious problems However, visual observations are judgmental rather than

Revision 0 14-23 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

quantifiable. Instrumentation may be needed to provide information to enhance our

ability to analyze the condition of the structures.

2. Instrumentation systems
The types of instruments used for investigating a certain behavior are generally outlined
in Chapter 9. Each instrument should be reviewed for its location/depth, suitability to
provide the desired information and confidence that the instrument is providing valid
readings. The overall number and types of instrumentation should be reviewed to
determine if they are sufficient to assess the total structure. The critical sections of the
structure should be defined and the location of instruments relative to the critical section
reviewed. The frequency/regularity of reading and timing of readings should be
reviewed. The occurrence of taking the reading should be logically related to the date and
the corresponding reservoir levels.
The personnel taking the readings should be queried for the procedures used to acquire
the readings and their awareness of certain threshold levels. The procedure for
processing the raw data should be reviewed for correctness and timeliness. If data are not
being processed and evaluated in a timely and correct manner, personnel involved in the
instrumentation and monitoring program should be reminded, and further trained if
necessary, in the importance of each phase of the program and the potential impacts with
respect to dam safety. The type of presentation graphs should be reviewed for the data
included and the use of proper scales and format for the ability to interpret data (refer to
Chapter 9 of the Engineering Guidelines). Often great clutter is apparent because graphs
are presented monochromatically using only minute symbols to differentiate the lines.
Project plan drawings should be prepared that clearly show the locations of all
instruments at the development site should. Details of the instrument installation should
also be available.

3. Comparison of instrument readings to predicted and required action levels

Threshold limits should be developed and the criteria used to develop them should be
documented. Then threshold limits should be established based on the specific
circumstances. In some cases, they can be based on theoretical or analytical studies (e.g.
uplift pressure readings above which stability guidelines are no longer met). In other
cases, they may need to be developed based on measured behavior (e.g. seepage from an
embankment dam). Sometimes they may be used to identify unusual readings, readings
outside the limits of the instruments, or readings which, in the judgment of the
responsible engineer, demand evaluation. Both magnitude and rate of change limits may
need to be established. If trends or inter-relationships between data are not clear, it may
be appropriate to take more frequent measurements or collect additional complementary
data
All data should be compared with design assumptions. For example, measured pore
pressures and uplift pressures should be compared against those used in stability
analyses. If data are available for unusual load cases, such as rapid drawdown and
floods, it should be compared with assumed pressures.

Revision 0 14-24 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

More than one phreatic surface may exist where there are impervious strata in the
foundation. Piezometric data should be evaluated with geologic data to identify multiple
phreatic surfaces. If the phreatic surface for any strata is above the ground surface, the
stability of the dam should be evaluated using the elevated phreatic surface.
All data will follow trends, such as decreasing with time or depth, increasing with time or
depth, seasonal fluctuation, direct variatio n with reservoir or tailwater level, direct
variation with temperature, or a combination of such trends. The trends are usually
evident in the plotted data. Statistical analysis of data may be useful in evaluating trends
that are obscured by scatter. However, such analyses are no substitute for judgment
based on experience and common sense. Data inconsistent with established trends should
be investigated. Readings deviating from established trends should be verified by more
frequent readings. Erroneous readings should be so noted on the original data sheets and
should be removed from summary tables and plots.
If no unusual behavior or evidence of problems is detected, the data should be filed for
future reference. If data deviates from expected behavior or design assumptions, action
should be taken. The action to be taken depends on the nature of the problem, and should
be determined on a case-by-case basis. Possible actions include:
• performing detailed visual inspection;
• repeating measurements to confirm behavior;
• verify that instruments and reading devices are working properly;
• reevaluating stability using new data;
• changing frequency of measurements;
• installing additional instrumentation;
• designing and constructing remedial measures;
• operating the reservoir at a lower level; and
• emergency lowering of the reservoir.
Guidance on methods for establishing threshold parameters is presented for the various
types of instrumentation described. Threshold parameters are defined as the
measurement parameters that trigger need for further investigation, deliberate action or
emergency action. The Independent Consultant in consultation with the licensee should
establish threshold parameters.

4. Consider a way to flag instruments that are trending in an adverse manner and
what additional focus should be placed on those instruments
Instruments that do not appear to be functioning properly should be further investigated.
For example, data should be checked against redundant data to determine whether or not
trends and magnitudes are the same. Calibration of the instruments should be checked
(this is paramount). Often, tests can be devised to evaluate proper functioning.

Revision 0 14-25 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

5. Additions/deletions/duration (how long an instrument should be read)

Instrumentation, in addition to the minimum recommended, should be required wherever
there is a concern regarding a condition that may affect dam safety or other critical water
retaining structures. Typical reasons to require additional instrumentation are: to check
design assumptions; to provide data to evaluate specific problems such as continuing
movement, excessive cracking, or increased seepage; to provide data to support design of
remedial modifications; and to provide data to evaluate effectiveness of remedial work.
Note that continually progressive conditions may require immediate action rather than
belated installation of extra instruments.
Instruments should be reviewed for their life expectancy. Readings from advanced age
instruments should also be evaluated with respect to whether the instrument readings can
be trusted. A failed instrument should be removed to avoid obtaining erroneous data later.

6. Redundancy
There is no such thing as a redundant instrument. All instruments should have real value,
if not they should be eliminated. The only redundancy would be to use different
instruments to measure the same feature.

7. Summary
Instrumentation and visual surveillance provide the means for helping to develop the
understanding or verify the performance of a dam.
The purpose of instrumentation and monitoring is to help evaluate whether the dam is
performing as expected and to provide a warning of developing or changing conditions
that could endanger the safety of the dam. This information and data are used to maintain
and improve dam safety.
If there is a discrepancy between the measured and expected behavior of the dam, it may
indicate that the dam is not performing satisfactorily and that failure is developing or
occurring or it may be that the data or observations do not adequately represent the
behavior of the dam, or that conditions exist that were not accounted for in the expected
behavior. In either case it is often useful to perform field investigations and install
additional instrumentation to evaluate the behavior. Note again that rational judgment
must be used to take action rather than do further investigation. If what is going on is
serious enough you could put in more instruments just to see the dam fail.

14.4.2 Performance Monitoring Procedures and Guidelines

From these guidelines, necessary performance monitoring techniques and devices and
threshold parameters to be employed at a specific dam can be developed. The existing
performance monitoring systems in place at that specific dam can then be reviewed by the
licensee, the Independent Consultant and the FERC Inspector together and supplemental
performance monitoring systems agreed upon as appropriate. Additional information on the
details of performance monitoring instrumentation is presented in Chapter IX,
Instrumentation, of the Engineering Guidelines.

Revision 0 14-26 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

This section has been designed to acquaint you with some of the adverse responses of dams
and the associated performance monitoring systems and suggested method to develop
threshold parameters. Many dams will share the commonality of a potential failure mode but
the PFMA must be customized for each structure. Some of the types of dams are:
• Concrete Arch Dams (including multiple arches)
• Concrete Gravity Dams (including cyclopean and RCC)
• Masonry Dams
• Earthfill Dams (homogenous dams, zoned dams, asphalt core or faced dams, and
concrete or membrane faced dams.)
• Rockfill Dams (earth core dams, asphalt core or faced dams, and concrete or
membrane faced dams.)
• Concrete Slab and Buttress Dams
• Timber Crib Dams
• Rubber Dams
Some typical adverse responses and conditions related to potential failure modes and
scenarios are:
• Abutment or Foundation Movement
• Abutment and Foundation Seepage
• Structure Movements and Stresses
• Overtopping Washout of Abutments or Foundations
• Deterioration of Concrete
• Operations Procedures

1. Performance Monitoring Guidelines for Abutment or Foundation Movement

a. Visual Observation
The first line of defense for monitoring almost all potential failure modes is visual
observation. While visual observation of gross movement of a dam or foundation
would indicate that a very serious condition is occurring or developing, more
subtle indications of movement can be observed. Cracking, new areas of leakage
through the dam or foundation, and displaced foundation material, are all visual
clues of possible movement. Visual observation is beneficial in that it may
readily identify changed conditions and it has the advantage of complete coverage
(as opposed to instruments that often only monitor point locations. For concrete
dams pins can be established at the crest or in galleries along contraction joints to
determine whether differential movement is indicated or has taken place. These
pins can readily be observed during routine site visitations and after significant
loading events as well as during regular inspections rather than depending only on
annual surveys for an indication of movement.

Revision 0 14-27 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

b. Precise Movement Surveys – Horizontal and Vertical

Precision surveys of permanent monuments on the dam and adjacent foundation is
a periodic monitoring requirement. Typically, movement monuments are placed
at several points along the crest of the dam where they are line-of-site visible from
benchmarks established some distance away from the dam abutments.
Monuments may need to be located at foundation contact locations where
abutment instability is a potential failure mode. Annual measurements of the
location of such monuments provide data for detecting movements of the dam or
adjacent foundation. To avo id seasonal influences on the readings, it is helpful to
take the readings at the same time of each year.
c. Movement Monitoring Devices (Inclinometers, Deformeters, Tiltmeters,
extensometers, optical surveys)
Devices for more frequent monitoring of small movements of structures and
foundations include inclinometers (generally used to define planes of movement
in soil), extensometers (measure change in distance between two fixed points),
tiltmeters (measures vertical or horizontal offsets) and embedded cross-arm
settlement devices for internal embankment movement. These devices are used to
take frequent readings, generally quarterly, monthly or weekly, to obtain
information on specific small movements, generally related to ongoing
investigations or to establish movement history with regard to changing reservoir
or foundation water levels or in regard to special concerns triggered by other
observations.
d. Establishing Threshold Parameters for Movements
Once a series of movements over some period of time has been developed, and
confirmed by stress analysis as being appropriate, threshold parameters can be
established that would require further investigation or action. Before initiating
action however, measurements falling outside of a threshold parameter should be
carefully checked and confirmed.
Other threshold parameters can be defined relative to assumed parameters used in
the stability analyses.

2. Performance Monitoring Guidelines for Dam Structure, Abutment and

Foundation Seepage
Seepage through a dam or through the foundations or abutments of dams is a normal
condition. However, increases in historically observed amounts of seepage, in the
elevation of the phreatic surface in the dam, or abutments, in the uplift/seepage
pressures beneath the dam or the appearance of transported material in the seeping
water may be symptoms of a developing potential seepage related problems. The
appearance of transported material in the seeping water of an embankment or soil
foundation may indicate piping or seepage erosion which could lead to a failure.

Revision 0 14-28 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

a. Visual Observation and Leakage Weirs

The visual observation of new seepage or an increase in volume of seepage
requires action be taken to quantify the problem and to watch for the presence of
material being transported in the seepage. For example, if rapid increase in the
seepage rate is observed, it may be a strong indication of a developing failure
situation and emergency action must be taken. Visual observations of depressions
or sinkholes in an embankment or upstream abutments or foundations are strong
indications of piping occurrence.
To accurately monitor any seepage, it must be collected and passed through a weir
for periodic measurement. A weir is a superior way for monitoring for the
possibility of material movement for several reasons. It provides a continuous
means for settling and trapping particles that may be piping or eroding as a result
of the seepage flow. Episodic material discharge has been observed in several
instances, thus a periodic check of seepage flow for material may not reveal
whether the seep is actually moving material. Weirs also allow the material
collected over a period of time to be measured and weighed. If weirs are used in
an area where fines may be blown into the weir a cover is necessary. Also the
weir should be routinely cleaned after each periodic measurement so the amount
of new material between collections can be accurately assessed.
Flumes allow for accurate measurement of seepage rate but do not provide a
means for collecting material. Regardless of the method used to measure the
seepage rate (weir, flume or bucket and stop watch), a sample of the seepage
water should be collected and allowed to settle out, at least overnight, to check for
the presence of any suspended material (fines) being transported (piping or
seepage erosion). Drainage pipes within a downstream embankment provide a
convenient method for collecting and measuring seepage. Care must be taken that
such pipes are properly filtered to prevent piping and if not it is even more
important that the flow from the drain pipe(s) be routed through a weir to allow
capture of any material being moved.
When new wet areas are observed on the downstream face of an earthfill dam, a
determination needs to be made as to whether this water is emanating from a
perched, more pervious zone in the embankment lying above a less pervious
layer, or is indicative of a high phreatic surface. Wet areas and points of seepage
exit should be marked on the dam face by large stakes so that any change with
time / season can readily be assessed. Monitoring of vegetation (big roots can
initiate piping) and rodent holes is critical. The determination of the nature of
such seepage can usually only be confirmed by the installation of piezometers.
The flow rate of such seeps should be monitored by weirs and checks made on
transported material in the flow.
Although increases in seepage are generally considered to be more of an indicator
of a potential problem, decreases in seepage (particularly in a concrete dam
foundation but also within embankment dam foundations or drainage elements)
may indicate that flow paths / drains are being blocked within or near their exit
from the dam or foundation resulting in an increase in pressures. Thus when

Revision 0 14-29 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

seepage decreases occur, checks should be made on piezometer or uplift gauges

(if this instrumentation is present) and on the cleanliness of the drain elements.
The onset of significant increases in seepage may correlate with reservoir
elevation reaching particular levels and this possibility should be reviewed at sites
with significant seepage.
b. Piezometers and Observation Wells
A great benefit to understanding the potential for failure mode development
related to seepage from a dam /foundation system is to develop an understanding
of the relative pore pressures and direction of flow within and through the dam. If
the pressures in the foundation (below the core) exceed those in the dam then the
direction of flow indicated from the foundation to the dam and the possibility of
piping of material from the dam to the foundation is remote. Conversely if the
direction of flow indicated is from the dam base into the foundation then the
physical possibility of piping from the dam through the foundation is indicated.
The best way to determine this flow regime is to review the piezometric,
observation well and seepage data.
Whenever there is concern for stability that may be sensitive to the phreatic
surface or seepage forces in the abutment, foundation or embankment (such as in
a rockfill dam with a wide central core), periodic measurement of water levels
must be made. The measurement of seepage forces in abutments and foundations
and particularly in a dam embankment is usually made by piezometers sealed to
determine the water pressure in specific strata or zones. The phreatic surface in
the abutments or foundation can be measured by observation wells, usually open
tube pipes with long sensing zones and with only the top of the tube sealed to
prevent surface water infiltration. If stratification exists in the abutment
consideration for different piezometers sealed in the various soil horizons should
be given.
c. Monitoring for Movement of Material (Piping)
Whenever seepage is observed emanating from / through the dam, foundation or
abutments, periodic checks on movement of material should be made. A sample
of the seepage water should be collected and allowed to settle out, at least
overnight, to check for transported material (piping). Crest settlement surveys are
an important source of early warning of piping failures, particularly in earth and
rock-fill dams. Specific inspections to look for depressions or sinkholes,
particularly in upstream areas, should frequently be made.
Increases in phreatic surface or seepage pressures in the foundation or abutments
may also indicate that movement of material is occurring.
Where the foundation materials may be susceptible to solution from water of
certain chemical properties, frequent checks on groundwater and reservoir water
chemistry and on the chemical composition of seepage water should be made.
Evidence of solution of foundation materials or strata requires prompt
intervention.

Revision 0 14-30 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

d. Establishing Threshold Parameters for Seepage

Seepage through a concrete dam is usually monitored by observation and
mapping (see Visual Observation). If seepage appears to be spreading or
increasing in volume, then an investigation and action to reduce the seepage may
be necessary. It is generally difficult to accurately determine the effects of
through seepage on concrete dams.
Seepage through an embankment dam is usually monitored by observation, timed
flows discharging from drainage pipes and weirs or other flow measurement
devices. If seepage flow is increasing, then an investigation and action to
evaluate the situation and determine if remedial action is necessary. Once a
history of variation in seepage flow has been established with respect to season
and reservoir level, then corresponding threshold parameter levels can be
established that will trigger the need for further investigation and remedial action.
The location of the phreatic surface in the embankment or the seepage pressures
at specific strata is determined by piezometers. Seepage through the abutments or
foundation is similarly monitored. A steadily rising phreatic surface or increasing
seepage forces should trigger a prompt review and, if necessary, remedial action.
Periodically, the measured phreatic surface or seepage pressures must be
reviewed against those surfaces or pressures that were used for the most recent
stability analyses. If the actual phreatic surfaces or seepage pressures exceed
those used in the stability analyses, then a special engineering review must be
initiated and remedial action may be required.
Seepage through abutments or foundation however can usually be collected and
measured by weirs or other flow measurement devices. Once a history of
variation in seepage flow has been established with respect to season and
reservoir level, then threshold parameter data related to seasons and reservoir
levels can be established that will trigger further investigation and remedial
action.

3. Performance Monitoring Guidelines for Structure Movements and Stresses –

Static and Seismic Loading
When possible distress of the dam structure itself is suspected as a result of
observation of cracking, new leakage, movement monument measurements, or
updated stress analyses, more detailed measurements of dam structural performance
are required. Slab and buttress dam designs have typically been designed for in plane
loading only. They are often inadequately reinforced and are incapable of resisting
cross canyon earthquake accelerations.
It is difficult to determine stresses directly on an existing dam unless stress or strain
meters or load cells were installed during the initial construction. Therefore, most
performance monitoring is aimed at determining strains under varied loadings to
calibrate stress analyses.

Revision 0 14-31 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

a. Precise movement surveys (surface) – horizontal and vertical.

Additional surface monuments can be quickly installed and more frequent
measurements made to obtain additional data.
b. Plumb lines
Plumb- lines are very difficult to install on existing concrete structures. They also
require vertical alignment that cannot be achieved in double-curvature arch dams
unless galleries have been specifically placed to accommodate installation.
c. PMS - Tilt meters
Measurements by tilt- meter are also useful. Tiltmeters can be installed on
existing concrete structures and readings can be obtained quickly after
installation. Tiltmeters are sometimes used instead of plumb- lines because they
are easier to install and require little maintenance.
d. Load cells
For direct measurement of loads in the dam, load cells must be installed during
construction. While they can provide meaningful data, if they are not in the area
of highest stress, they have limited usefulness. If post-tensioned anchors are used
to improve stability, either in part of the dam (such as abutment blocks) or in the
foundation, some anchors should have load cells installed to monitor their loss of
tension so that retensioning can be performed as necessary.
e. Strain Gages
Stain gages can provide valuable direct stress data if installed during construction.
However, they are subject to the same limitations as load cells.
f. Seismographs
Seismographs provide a valuable research tool when they are mounted on dams
and triggered by significant earthquakes. They can provide response data for
improving seismic stress models of the structure or for deformation models of an
embankment. However, they are not useful for monitoring performance of dams
but can confirm the response of the dam to an earthquake, e.g. crest amplification
g. Additional piezometers can be installed and more frequent measurement made to
obtain additional data.

4. Performance Monitoring Guidelines for Erosion of Abutments or Foundations

Performance monitoring of embankment dams relative to washout is pertinent only
with respect to ensuring the condition and levelness of the crest. Low spots, rutting
or “built in” unevenness in the crest can exacerbate the potential for overtopping
failure of an embankment dam. Spillway adequacy must be adequate to prevent this
potential failure mode. In the case of embankment dams, the rule is to prevent
overtopping, because it can lead to catastrophic failure.

Revision 0 14-32 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

a. Observation and Measurement of Deterioration of Abutments, Foundations,

and/or Spillway Outfall/Energy Dissipater Areas
In order to monitor the deterioration of dam abutments, foundations, and/or
spillway outfall/energy dissipater areas to assess the potential for washout failure,
it is necessary to have data on the potential for flow over the abutments, including
volume and frequency relationships, and specific data on large flows that would
impacting the foundation.
The foundation, abutments, and spillway outfall areas should be surveyed and a
profile of the foundation impact area of overflows, abutments and/or spillway
outfall/energy dissipater areas made. Because survey markers in such situations
will probably be lost in flow situations, the survey should be at precise station
points along the abutments within the flow zone and on the downstream
foundation offset from the crest of the dam. Such surveys should be repeated
after major flows have occurred and the changes in the profile plotted and
reviewed and the erosion potential quantitatively estimated. Utilizing the flow
volume and frequency relationships, an assessment of potential failure due to
washout can then be made.
b. Periodic Assessment of Geologic Conditions and Deterioration
In addition to the survey and assessment above, close geologic inspection of the
foundation, abutments and/or spillway outfall/energy dissipater areas should be
made including mapping of joints that could permit loss of foundation or
abutment rock material. Such inspections should be repeated after major flows
and the potential for washout failure made.

5. Performance Monitoring Guidelines for Leakage through Dam Joints or Cracks,

Along Penetrations, Conduits and Structures
a. Periodic Visual Mapping of Leaks/Wet Areas on Downstream Dam Face
Using a downstream profile map of the dam, showing any visible vertical joints
and horizontal joint or lift lines, make a periodic map of all cracks, leakage
locations and seepage areas. The mapping should be supplemented by detailed
photographs. If there are significant leakage locations or areas, such maps and
photographs should be made at least semi-annually, at coolest and warmest times
of the year, and regularly compared. Particular care must be taken to note areas
of increasing leakage flow or extension of cracks.
b. Measurement of Seepage Quantity by Weirs, Flow Meters or Other Devices
Where it is difficult to determine if the quantity of dam seepage is increasing,
various devices are available to measure flow. Unfortunately, they may be
physically difficult to install on the vertical or overhanging face of an arch dam.
They would generally only be utilized in cases of significant concern.
c. Periodic Visual Mapping of Leaks/Wet Areas on Downstream Dam Face
Using a downstream profile map of the dam, showing any visible vertical joints
and horizontal joint or lift lines, make a periodic map of all cracks, leakage

Revision 0 14-33 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

locations and seepage areas. The mapping should be supplemented by detailed

photographs. If there are significant leakage locations or areas, such maps and
photographs should be made at least semi-annually, at coolest and warmest times
of the year, and regularly compared. Particular care must be taken to note areas
of increasing leakage flow or extension of cracks.

6. Performance Monitoring Guidelines for Deterioration of Concrete

Concrete in dams sometimes deteriorates. Usually such deterioration is due to poor
quality concrete having been used for construction. On older dams, alkali aggregate
reaction is not uncommon. For dams at high elevations and in northern areas, freeze-
thaw deterioration is a concern.
Concrete deterioration is more critical in Ambersen slab and buttress type dams
where design stresses in the water retaining slabs and reinforced concrete beams are
about 50% of the ultimate capacity utilized under normal loading. Concrete
deterioration not only reduces the cross-section properties but also exposes the
reinforcing allowing it to corrode and reduce cross-section.
a. Periodic Visual Inspection and Mapping of Deteriorated Areas.
Make a periodic map of all areas of deterioration whether caused by freeze thaw,
alkali-aggregate reaction, or other mechanisms. The map should show any visible
vertical joints, horizontal joints, lift lines, any loss of masonry elements or mortar
from the joints. The mapping should be supplemented by detailed photographs. If
there are significant deteriorated locations or areas, such maps and photographs
should be made at least semi-annually.
b. Periodic Measurement of Deteriorated Areas
Measurements should be periodically made to assess the changes in the structure.
In the case of alkali-aggregate reaction, periodic surveys of dam crest elevation
should be made to ascertain the amount of swelling of the concrete. Also, cores
of the expanded concrete should be taken at intervals and tested for compression
and tension (if subject to tensile stresses.) The strength data obtained from the
tests should be compared with the results of a stress analysis, to determine the
adequacy of the structure.
In the case of freeze thaw or other deterioration mechanisms, the depth of
deterioration should be periodically determined. If the area is quite localized, the
depth can be determined by probing and measurements from the surface of
unaffected areas. One technique is to install reference markers set in the deeper
undisturbed concrete and periodically measure the distance from the end of the
marker to the sound concrete.

7. Operations Procedures Common to all Dams

Mis-operation of a dam, either through equipment malfunction or human error, is
often a viable potential failure mode. This section identifies some adverse conditions
and associated defensive measures.

Revision 0 14-34 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

a. Human error
This includes all the site specific scenarios of mis-operation or failure to act. For
example, if gate operation is required to pass a flood, but nobody raises the gates,
the dam could fail.
Defensive measures could include additional specific training of personnel and
emergency procedures exercises that are tailored to specific identified potential
failure modes.
b. Equipment malfunction
Power failure - Anytime electrical equip ment such as gate hoists, sensors,
communications, etc is required for the safe operation of the dam, power failure
can lead to a dam failure.
Defensive measures could include having standby power available at the site, and
having manual overrides on critical equipment
Sensor/Telemetry malfunction - If the site is remotely operated, a sensor error or
telemetry failure can lead to mis-operation or failure to operate.
Defensive measures may include performing regular telemetry testing, having
redundant sensors for critical instrumentation, and having redundant
communication systems.
c. Access failure
If personnel must be dispatched to a site to operate it in an emergency, they have
to be able to get there. Large flood events are typically accompanied by severe
weather that may make roads impassible and helicopter travel impossible.
Defensive measures may include having identified alternative routes to access the
site or stationing personnel at the site prior to failure of access roads.

Revision 0 14-35 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

14.5 EMERGENCY PREPAREDNESS

The Independent Consultant should consider the various potential failure modes determined
and review pre-planned procedures for dealing with potential emergencies. If any pre-
planned measures need to be put in place, the Independent Consultant will make the
necessary recommendations. These measures include necessary equipment, materials, etc.

Revision 0 14-36 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

APPENDICIES

Revision 0 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix A
Example Potential Failure Mode Descriptions

Operational Related Potential Failure Mode

The design flood was routed through Dam A by the hydrologic engineering consulting firm
using the traditional means and assumptions and the capacities for the facilities provided by
the owner. The dam was found to pass a sizeable portion of the probable maximum flood
using the Main Spillway gates and the emergency spillways. Thus there was concern for the
hydrologic deficiency but not great concern. However, examination at the site for potential
failure modes revealed a significant potential for an Overtopping Failure mode due to the
following factors:
• The emergency spillway bays were fronted by arch rings designed to be blasted away
if the emergency spillway was needed. Discussions with the owner revealed that use
of the emergency spillway in such a manner was highly improbable. This was due to
the potential liability from such an action (a sizeable town is located just a mile or so
downstream) and also due to the physical arrangement of the dynamite ports on the
top of the spillway bays (it was likely that these would be underwater by the time a
decision to use them was made). Further there were no plans or procedures in place
to do the blasting.
• The first location for overtopping of the structure was immediately above the
transformer yard. Overflow at this location would have resulted in loss of capability
to pass flow through the turbines and while this flow was not relied on in the routing,
the early shut off would have exacerbated the overtopping situation.
• Drawings were located for the secondary emergency spillway, which was referred to
as a “fuse plug” spillway but this fuse plug actually had to be excavated by a dozer
before it was functional. Operators at the site did not know the location of the
spillway limits and had no procedures or equipment to initiate this spillway.
• Design crest elevations indicated that the concrete structures would be overtopped
prior to the embankment structures. However, examination of survey data, settlement
records and settlement projections (along with the physical location of the
monuments relative to the crests) revealed that the low point for the project was
currently an earthen saddle dam.

Piping Related Potential Failure Mode

The following potential failure mode was highlighted because the specific conditions at Dike
1 and 2 are such that this potential failure mode is physically possible and is one of the most
significant potential failure modes definable at this site. Failure of the Dikes poses a high
hazard, and diligence in monitoring for development of this potential failure mode is
warranted.

Revision 0 14-A-1 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Potential Failure Mode 1 - Dikes 1 and 2 - Seepage Erosion and Piping

During site investigation the foundation of these dikes was found to contain joints much
more open than anticipated, based on pre-construction investigations. These joints provide a
potential path for subsurface erosion of the Zone 1 material leading to an unprotected exit
downstream of the dam. Although grouting was performed following construction (during
the first filling of the reservoir) and the seepage levels were reduced, the fundamental
potential failure mode remains). The presence of 4 to 5 ft3 /s of seepage, which occurred
during first filling, from a dike of moderate height and length attests to the possibility of open
joints in the foundation capable of carrying adequate flow to result in erosion, and transport
of eroded material downstream. The specific potential failure mode paths and the factors
relative to the likelihood for the development of this potential failure mode are as follows:
Potential failure mode paths - there are two primary potential paths for seepage
erosion/piping to take place through the foundation jointing and two of lesser likelihood.
These are:
• Flow through the dike embankment across the Zone 1/ foundation interface. This
could result in the Zone 1 materials eroding and being carried through the open joints
to an unprotected exit downstream. (Failure would result if backward erosion
(piping) through the Zone 1 materials reached the reservoir source. An ever
increasing flow potential could then progressively enlarge the flow channel
downstream of the point of erosion initiation in the core to an extent large enough to
carry continually increasing flows).
• Flow under the foundation attacking the base of the Zone 1 material and removing it
by seepage erosion through the foundation jointing
The other two potential flow paths leading to a seepage erosion/piping failure are (1) piping
of the Zone 1 through the foundation alluvium and (2) seepage erosion of the foundation
alluvium exiting through the open joints in the rock. These are considered to be of
significantly lesser likelihood

Part 2 - Example Record of Likely and Not Likely Factors Identified for a Specific
Potential Failure Mode
After each potential failure mode is identified and clearly stated and recorded the team
brainstorms and discusses what conditions make this potential failure mode more adverse /
more likely and what conditions about this potential failure mode are positive making it less
likely or less significant. An example of the factors identified for the potential failure mode
previously described above is presented below.
Factors increasing the likelihood of this potential failure mode developing include:
• the observation of very open joints in the foundation (greater than 2" wide)
• surface treatment was not provided to exposed bedrock
• grouting procedures used likely resulted in some of the most open joints remaining
open due to the presence of the reservoir produced seepage flows during the grouting
(this is most likely in the higher head, lower elevation portion of the dikes)

Revision 0 14-A-2 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

• grouting near the surface was likely not very effective, considering the method used
• large seepage flows are occurring, which can dilute and potentially mask observation
of particles being carried by the flows. If an attack begins, large flows can erode
large amounts of material relatively quietly.
Factors indicating less likelihood of this potential failure mode developing include:
• There has been no observation of any material being carried by seepage flows at these
dikes.
• The Zone 1 appears to be clayey, very impervious, and not easily erodible.
• The placement of the Zone 1 cutoff- wall well upstream of the dike centerline creates
a closer source for the reservoir’s water, but a large portion of the dike would remain
if an erosion path developed at the base of the cutoff.
• The lack of water in the toe drains is a likely indicator that the dikes have not
saturated and that the foundation rock behaves as a drain keeping water away from or
at low pressures at the dam/foundation contact.
• Seepage flows started downstream of the dikes prior to the water reaching the
upstream toe of the dikes. This indicates the pervious nature of the foundation and
the likelihood that a large portion of the seepage water passes beneath the dikes
(relatively independent of it).

Revision 0 14-A-3 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix B
Potential Failure Mode Identification Cover Letter and
Questionnaire

POTENTIAL FAILURE MODE IDENTIFICATION

The PFMA concept provides that the identification of potential failure modes with potential

effects and consequences be prepared with input from all persons with data and information

relevant to the design and performance of the dam. Your input is requested. This memo

tells you all about how it will work and what your role is.

*** Individual Input to Potential Failure Mode Analysis ***

How is this special effort going to work?

The idea is to produce the potential failure mode analysis efficiently without sacrificing the

specific knowledge, information and opinions of people who have worked on the dams in the

past or are working on them at present. The information and knowledge you have will be

collected in three ways as enumerated below. Please note that the potential failure modes

developed as part of this concentrated effort will be based on available data and information.

We do not expect you to do any additional studies. However, you may suggest what

additional study may be valuable as part of a Safety of Dams Review for this dam.

How will we get your input?

1. Through a questio nnaire completed individually and designed specifically to obtain
the information needed to help develop the potential failure mode analysis.

2. Through core team review of all work that has been prepared for the dam.

3. Through a round table discussio n with available team members following completion
of 1 and 2 above.

The concentrated effort will allow efficiency in preparation and consistency in the product.

Revision 0 14-B-1 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Please complete each portion of the attached questionnaire for which you feel you can make

a contribution. Feel free to review any relevant portion of your past work on the project to

refresh your memory prior to completing the questionnaire. When you have completed the

questionnaire send the completed document to ------------

Thanks for your he lp.

Revision 0 14-B-2 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

POTENTIAL FAILURE MODE QUESTIONNAIRE

for _______________________ Dam

Name ___________________________________________

Team Members Experience / Role _____________________________________

Phone _________________

Date Due _______

Part 1 - Potential failure modes

"Failure" is considered to be something which causes uncontrolled release of the

reservoir (or a portion thereof).

I. What potential failure modes do you consider are of specific concern

at this project? Please provide specific description of the potential failure mode, including

location of the area of concern.

Potential failure mode 1.

Potential failure mode 2.

Potential failure mode 3.

Revision 0 14-B-3 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Part 2 - Performance Observation - For each potential failure mode identified in Part 1

II A. Please describe as specifically as possible what may be

observed or measured to verify that performance is okay relative to the potential failure

mode, or that conditions that are conducive to initiation of the potential failure mode have

become present or that "activation" of the potential failure mode appears to be taking place.

Please suggest how that may be observed or monitored.

II B. Is the monitoring or observation you suggested currently in use at the

dam?

III. If you have visited the site either as part of an official inspection or a

general site visit, please identify any conditions that are noteworthy from the standpoint of

performance of the dam.

Revision 0 14-B-4 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix C
A Typical Potential Failure Mode Analysis Session

Step by Step Description of a Typical Work Session

• Identifying Key Technical Back up Information Needed
• Typical Sequence of Brainstorming Activities
• Key information to document during the process
The intent of this appendix is to describe what is done in a typ ical PFMA session. This is
done so participants will know what to expect and so that all the right information and data
will be on hand and the people needed are there or on call.

Owner / organization prerequisite work: - Gather all background materials for review
prior to the session and have available at the session

Required individual advanced preparation activities:

1. Core team members have read all background materials
2. All participants have read a general background package (inspection report and / or
standards based engineering dam safety report) to become familiar with or to recall
the project elements and issues.
3. All participants have completed / considered questionnaire on identification of
potential failure modes.
4. Inspector or instrumentation group has instrumentation and surveillance data updated
and ready for review by core team prior to meeting
5. Project leader has references available in meeting room – this includes:
• all engineering reports and key feature drawings (large scale)
• construction photos and construction / design history data ,
• flood frequency data and routings and earthquake loading data
• data on consequences and emergency preparedness
• inspection reports and instrumentation and surveillance data.

Revision 0 14-C-1 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Begin Session

I. Adequacy of Project Documentation - Discuss adequacy of documentation provided

for the exercise and determine if any deficiencies exist for specific potential failure
modes. Determine if sufficient informatio n exists to adequately perform the Potential
Failure Modes Analysis for the project. Document your findings regarding adequacy or
deficiencies in project documentation.
II. Potential Failure Mode Identification – Go loading by loading and feature by feature
– Have drawings or sketches of features - Typically start with flood loading since it is
the easiest for all to understand and sets out the reservoir loading conditions ----
Flood loading:
• go over all the key data - size and frequency of floods analyzed, routing data
• show crest elevations of key features and amount of overtopping / freeboard for
each flood routed – discuss slope protection, crest condition , discharge locations
• Examine potential effects / potential failure modes / adverse conditions for dam
structures and spillways / outlets. Note: go over each structure to see if any
potential failure mode is evident.
• Candidate (suggested) potential failure modes are called for and means /steps to
failure are discussed – decision is made whether or not to consider as a potential
failure mode.
• If a suggested potential failure mode is considered, the potential failure mode is
clearly described – then the reasons why the candidate potential failure mode is
more or less likely to develop (adverse and positive factors) are listed on flip chart.
Based on this discussion the team classifies the potential failure mode (highlighted,
considered but not highlighted, etc.)
• Breach formation character and rate of failure are discussed. Consequences of a
flood related failure and of operational spillway discharges are discussed.
• Failure scenarios (exposure conditions and warning aspects– detection, decision to
warn – dissemination of warning - evacuation, etc) are discussed. Anticipated
response is reviewed with field / site personnel – in reality this exercise is looking
for “potential failure modes in the preparedness arena”.)
• Opportunities to achieve risk reduction – structural or non – structural and ways to
improve detection via instrumentation or surveillance are identified and listed
• Data / information needs are discussed and identified.

Revision 0 14-C-2 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Reservoir loading / Static load:

Repeat above process – key changes or additions or points of emphasis in examining
this loading are noted below:
• Discuss annual and historic pattern of reservoir loading
• Discuss performance history of each feature
• Discuss any instrumentation clues to potential failure mode
• Discuss Geologic / foundation rock and soils relationship to structures – examine
for potential failure modes – for concrete dams discern whether or not a foundation
analysis based on adequate engineering geologic has been completed
• Trace (from a projection) on a white board a sketch of the cross section of each
structure to be evaluated in turn. Sketch each potential failure mode as suggested to
enable brainstorming / developing / understanding. Plan or profile sometimes
needed as well.
• A list of the reasons why and why not these potential failure modes are likely to
occur are discussed and recorded – This discussion is very important in evaluating
significance / category of potential failure modes suggested (which is the next step)
• Based on this discussion the team classifies the potential failure mode (highlighted,
considered but not highlighted, etc.)
• Evaluate failure scenarios for each significant potential failure mode – (these
scenarios do not have the same warning associated with flood or earthquake
loading)

Earthquake loading
Repeat above process – key changes or additions or points of emphasis in examining
this loading are noted below: (Note –dynamic loading follows static loading because
some of the potential failure modes are similar – can examine the degree additional
loading may impact static condition.)
• go over all key magnitude and frequency data for site (historic and tectonic study
data) and site attenuation or amplification data to get sense of the loading
likelihood.
• review any dynamic analyses
• examine what it takes for failure to occur – i.e. does damage result in failure
• compare structures at site to case histories of earthquake related failure

Consider any other loading relevant to site:

Ice, avalanche, landslide, etc. – Repeat process used for other loadings

Revision 0 14-C-3 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

III. Make final team assessment

1. Significant Potential Failure Modes (Highlighted) in each loading category and
Summarize / Rank
2. Potential failure modes considered but not highlighted
3. Potential failure modes considered but lacking key data or information to allow
designation of significance – identify data needs
4. Failure Modes ruled out

IV. Review possible risk reduction / instrumentation / surveillance opportunities

identified for all potential failure modes considered
Place the identified into two categories:
1. Possible alternative mitigation actions to investigate, and
2. Actions to be considered by the Independent Consultant for implementation by the
owner.

V. Identify and Record– “Major Findings and Understandings”

The major findings and understandings achieved as a result of the session (give team
members a few minutes to think about these before listing) along with the description
of each potential failure mode considered should be written up and distributed to
participants shortly after the PFMA session.

Revision 0 14-C-4 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix D
Suggested General Format for Potential
Failure Mode Analysis Reports

(For further guidance and example see a typical report developed from this outline – these
reports may be obtained from any FERC Inspector)

I. Introduction and Background

Purpose / description of study

II. Brief Description of Dam and other Key features of project

III. Major Findings and Understandings from study

IV. Potential Failure Modes Identified

(For each potential failure mode identified there would be:
• A Detailed Description of the Potential Failure Mode / Adverse Consequence
• A listing of adverse / likely factors and a listing of and positive/ not -likely factors
• The likelihood category selected by the team for that potential failure mode
• A description of rationale used for selecting that category (i.e. – the factors with the
greatest weight)
The presentation of potential failure modes would be grouped according to the 4
likelihood / importance categories.
1. Highlighted
2. Considered but not highlighted
3. Information needed to allow classification
4. Ruled out-not physically possible/extremely remote

Other Considerations – thoughts / ideas / concepts / future changes that were

considered related to potential failure modes that were brought up and discussed but
not developed by the PFMA team as a potential failure mode (That is - there was no
detailed description of the Potential Failure Mode / Adverse Consequence and a
listing of adverse / likely factors and a listing of and positive/ not - likely factors was
not developed) should be documented in this section such that future teams will
know what items were considered and why they were not carried forward at this
time.

Revision 0 14-D-1 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

V. Likely Consequences of Each Potential Failure Mode

• Flood related potential failure modes
• Normal Operating related potential failure modes
• Earthquake related potential failure modes
• Other condition potential failure modes

VI. Potential Risk Reduction Actions Identified

Note: This is often effectively included as a part of section IV under each potential
failure mode.
• Surveillance and monitoring enhancements
• Risk Reduction measures to evaluate
• Investigations / analyses needed
• Data and information needed to collect / prepare for decisions on prioritization of
dam safety actions

VII. Other considerations related to study

VIII. Summary of potential actions identified in the PFMA with respect to Performance
Monitoring (Instrumentation and Visual Surveillance)

IX. Summary and Conclusions

Appendix to Report
Key supporting data and information and references, Figures, Sketches, Photos made
during field review showing key elements of dam and auxiliary features should be
included along with any photos that show conditions leading to potential failure modes,

Revision 0 14-D-2 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix E
Major Findings and Understandings - Example Write Up

Given below is an example write up of the major findings and understandings gained from a
Potential Failure Mode Ana lysis for a project consisting of a Main Concrete Dam
incorporating a Power Station and Two Auxiliary Embankment Dams. Although this was an
actual study and the actual findings, the names of the dams and the river in the example are
not the real names.
• Currently in the event of a very large flood on the Blue River, approaching the PMF,
overtopping failure of Auxiliary Dam 1 is the main point of vulnerability at the
project. This is because the crest of Auxiliary Dam 1 it is at a lower elevation than is
the crest of Auxiliary Dam 2. In the event of Auxiliary Dam 1 failure, peak
discharges downstream would nearly triple (from about 1900 m3 /s at failure to 5700
m3 /s at breach) and the consequences of failure of Auxiliary Dam 1 would be high
(life loss potential and large economic losses). On the other hand if Auxiliary Dam 2
were to be established at a lower elevation than Auxiliary Dam 1 and thus allowed to
fail from overtopping the effects and consequences of overtopping failure would be
significantly less. Auxiliary Dam 2 failure peak discharges downstream are estimated
to only be slightly larger than flows resulting from the PMF (from about 2100 m3 /s at
PMF to 2400 m3 /s at breach). Several measures to achieve overtopping failure risk
reduction are identified in the report and the best alternative should be selected after
an appropriate risk management evaluation. However, the Potential Failure Mode
Analysis team emphatically concluded that it is essential that as long as the potential
for overtopping failure of the earthfill dams exists, Auxiliary Dam 2 should be
established at a lower elevation than Auxiliary Dam 1.
• Dam failure as a result of piping is a physically possibility at Auxiliary Dam 1 as the
result of one or more potential flow paths. Although there is no unequivocal physical
evidence that piping has occurred or will occur in the future, the nature and
relationship of the materials in the dam and foundation, the water level and
piezometric observations, and the performance of the structure (observation of
surface seepage and a depression at the toe) allow for this possibility. Further the
surveillance and instrumentation have not been extensive enough to rule out the
possibility that piping episodes (turbid water or particle transport) have occurred, and
even if they had been transport of material could occur subsurface and thus not be
amenable to observation. The consequences of a “sunny day” piping failure of
Auxiliary Dam 1 would be high with a greater life loss potential due to the possible
lack of advance warning. From the standpoint of the Potential Failure Mode Analysis
Team, awareness of this potential piping condition is a key finding of the study as this
potential failure mode is the most significant structural vulnerability found at the
project. Several risk reduction measures, both structural and non-structural were
identified and should be considered in the risk management evaluation of the project.

Revision 0 14-E-1 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

• A potential foundation failure mode identified for the Main Dam was the only PFM
of significance identified for this structure. Although, this foundation potential
failure mode is considered physically possible it is highly probable that a foundation
stability analysis would show that the factor of safety against failure is quite high and
thus risk reduction measures would not be required. Thus, the Potential Failure Mode
Analysis team considers that analysis rather than consideration of remedial work is
the appropriate initial course of action relative to this potential failure mode.
• The Potential Failure Mode Analysis team considered that failure of the Main Dam
by overtopping was not a realistic potential failure mode and should not be
highlighted. However, the main dam would be the first structure to be overtopped
during a major flood condition, well before either of the earthfill dams. This would
occur for the PMF but also for floods significantly less than the PMF. It was
determined in the Potential Failure Mode Analysis that overtopping of the main dam
would result in significant damage to the power facility and that much if not all of this
damage is preventable. For example the cable trays are at crest level on the
downstream side of the dam and would be destroyed with mild overtopping. Also
water flowing over the top of the dam would currently spill directly on the power
plant and likely knock that plant out of service and cause considerable damage. This
flow could be directed away from the plant area. Thus, main dam overtopping, while
not a potential failure mode, would be a significant incident for the Owner and the
Potential Failure Mode Analysis team considers that risk reduction measures to
reduce the impact of the overtopping of the main dam should be evaluated.
This example was chosen for brevity – only the most significant of the findings are recorded
– examples which provide significant greater detail are available from a FERC inspector)

Revision 0 14-E-2 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix F
Estimated Time Requirements
(PFMA items only – not subsequent Part 12D activities)

The typical time requirements for the major activities in the set up and completion of an
initial PFMA are as follows:

Task 1 - Advance preparation

Facilitator – Consult with and provide information/guidance to owner – ½ day
Owner – Assemble Materials, notify participants / send out advance packet of material,
setup facility and make arrangements - 3 days

Task 2 ----Visit site, review materials, carry out PFMA session

Owners core team representative – travel to and visit site with core team – review
materials and participate in session – 3 days
Independent Consultant – Facilitator – FERC representative – as above - 3 days each

Task 3 – Documentation of study

Independent Consultant- prepare draft of Major Findings and Understandings,
distribute for review and comment. Prepare draft of PFMA report distribute for
review and comment and finalize - 5 days
Facilitator – peer review and edit of Major Findings and Understandings, PFMA draft
and final report - 1 ½ days
Owner’s representative – review and comment on report – ½ day

Revision 0 14-F-1 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix G
Potential Failure Modes Analysis (PFMA)
Dam Safety Evaluation Process
Task Description IC Responsibility
No. Report
1 Data Assembly: Assemble all available background Chap 2, Owner, IC
data for the dam and appurtenant works for the Chap 8
project. These data should include investigation, &
design, and construction reports, construction photos App B
(and all other project photos), construction and as
built drawings, geologic reports, construction
inspection reports by the owner / designer and those
by state and federal agencies, seismic studies, most
recent flood studies, stability and/or stress analyses,
lab test results on rock, soil and concrete,
instrumentation monitoring data and visual inspection
reports, periodic dam safety reports including all Part
12 D reports, Emergenc y Action Plans, photographs
of key elements and features showing present
condition and any remedial work.
Prepare summary informational packet describing the
dam / project (e.g. – the most recent Part 12D)
2 Initiate PFM A: Establish “Core Team” to carry out Chap 3 Owner, FERC,
PFMA - Experienced PFMA Facilitator, Independent IC, Facilitator
Consultant (IC), Owners representative/PFMA
Coordinator, FERC Inspector. Also determine who
the individuals will be that will support / attend the
PFMA, dam engineering geologist, mechanical-
electrical specialists (if needed for gate/valve issues),
consulting engineering staff, operating staff
responsible for surveillance and maintenance of each
specific dam, project manager, etc.
Send out Potential Failure Modes Analysis
Questionnaire (Ref: Appendix B – Forthcoming
PFMA Guidance Document) to all participants in
advance of the PFMA to get all participants “thinking
potential failure modes” and gathering relevant
materials and information that may be helpful to the
session.
Send out a summary informational packet to the core
team to familiarize them with the project before their
site review and to familiarize / provide review to all

Revision 0 14-G-1 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

other participants before they come to the PFMA

session
3 PFMA Field Review: Physically inspect all aspects Chap 3 Core Team
of the dam structures and appurtenances (relevant to Geologist,
dam safety). Try to observe / discuss potential Operating
circumstances / conditions that could lead to a Staff,
potential failure (structural or operational). Discuss Specialists
operations with plant operators / site personnel. The
PFMA field inspection is to familiarize the core team
with the project features and conditions so that they
can effective carry out the PFMA. It is not a
substitution for the Part 12 D field inspection by the
IC, which is to determine the condition of all safety
aspects of the dam and power facilities. (Note: Allow
approximately one-half day per dam. May be more
depending on the complexity of the project.)
4 PFMA Data Review: Assemble and review data on Chap 3 Core Team
the dam. Should be at convenient location
considering location of site/data/PFMA. Assemble in
a group setting for efficiency in sharing the collected
data and to provide a “captive” condition to ensure
that the material is reviewed by all the core team
members. Also being together allows for
collaboration on items that may need clarification for
the entire core group. The team geologist should
ensure that the relevant geologic data is available for
core team review and the geologist should also
review this material personally. (Note: Allow one
day for the review plus evening of site review and
evening before PFMA as necessary. This review is
essential to an effective PFMA.)
5 Facilitated Potential Failure Mode Analysis: Chaps Core Team,
Discussion lead by Facilitator of candidate potential 3, 4 & 6 Geologist,
failure modes (PFMs) for Flood, Earthquake and Operating
Normal Operating conditions for each of the Staff,
structures. Develop (describe events and conditions Specialists as
from initiation to breach or other adverse outcome) needed
each Potential Failure Mode considered realistic and
credible by the PFMA team then brainstorm and list
all the positive factors that make the potential failure
mode less likely to occur and all the adverse factors
that make the potential failure mode more likely to
occur.
Discuss consequences of each PFM (and / or at end
of PFMA discuss consequences and response to

Revision 0 14-G-2 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

potential failure).
After discussion of each PFM, categorize the
potential failure mode (see Table 1 PFMA Guidance
Document). Review the categorizations at the
conclusion of the PFMA.
During each PFM discussion, identify possible risk
reduction actions including – monitoring,
surveillance, investigations, analyses, remediation
(structural or non-structural) and operational
procedures & maintenance programs.
For each Category I-III potential failure mode a
performance monitoring program (PMP) must be
identified. Performance monitoring can vary from
periodic visual inspections, to continuous recording
instrumentation and may include monitoring of
weather forecasts and monitoring of earthquake
activity.
To complete the PFMA solicit individual input on the
Major Findings and Understandings (MFU) reached
during the PFMA process (all the key things learned
or more fully understood during the 2-3 day effort).
All the input should be recorded on flip charts (hand
written) and may also be recorded electronically
(laptop). Ensure that all participants have their
input/concerns listed. Photograph hand written
charts. End PFMA meeting.
6 Draft MFU: IC prepares (amplifies/fleshes out) Chaps Facilitator and
items identified as Major Findings and 3, 4 & 6 IC
Understandings (MFU) of the PFMA, emails it to
Facilitator for peer review (edits, additions and
comments). IC then modifies MFU to include
Facilitator comments.
7 MFU Review: IC then emails MFU to core team Chaps Core Team,
and geologist. (Note: Owner may distribute to staff 3, 4 & 6 Geologist
for further input). All review and comment /send
email to IC. IC modifies MFU to include comments.
8 Draft PFMA Report: IC prepares draft PFMA Chaps Core Team,
report using FERC report outline and emails draft to 3, 4 & 6 Geologist
facilitator for peer Review. IC modifies PFMA report
to include facilitator comments. IC sends peer
reviewed draft PFMA report to Core Team and
Geologist. All review and email comments to IC. IC
modifies PFMA report to include all comments
(some discussion between IC and others is usually

Revision 0 14-G-3 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

needed to achieve final agreement). The final report

must specify the PMPs for all Category 1 and 2
PFMs.
9 Finalize PFMA Report : IC prepares final PFMA Chaps IC, Owner,
report as agreed with Owner. Facilitator Peer 3, 4 & 6 Facilitator
reviews final report. Owner forwards final report to
FERC with copies to Core Team. Review and
comments on the MFU and PMFA drafts typically
average about one day for the facilitator and ½ day or
less for each Core Team member and the geologist.

Revision 0 14-G-4 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix H
Part 12D Safety Inspection Report Outline

Table of Contents
The Table of Contents must show the inclusive page numbers for each section and
subsection. If any subsection is not applicable, include the subsection with a statement of
“Not Applicable” and an explanation of the reason(s) why.
For licensed projects that include multiple independent dam and powerhouse developments,
separate Part 12D reports should be published for each development.

1. Executive Summary
2. Project Description
3. Discussion of Potential Failure Mode Analysis Report
4. Performance Monitoring and Visual Surveillance with Respect to Potential
Failure Modes
5. Field Inspection
6. Operations and Maintenance Programs Relative to Potential Failure Modes
7. Assessment of Supporting Technical Information Document
8. Recommendations

List of Tables (with location)

List of Figures (with location)
List of References

Appendices for Part 12D Inspection Report

A. FERC Letter Requiring Part 12D Inspection
B. FERC Letter Approving Part 12D Consultant - Include date of current report
outline provided by FERC
C. Project Figures
Only provide general overview drawings necessary to understand the project and
items discussed in the report. If figures are placed in Section 2, provide a statement
that figures may be found in Section 2. Optionally, if the STI is bound with the Part
12D report provide a statement that figures may be found in the STI document;
duplicate drawings from the STI do not need to be included in the Part 12D report
proper.
Additional detailed drawings will be included in the Supporting Technical
Information document.

Revision 0 14-H-1 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

D. Instrumentation Monitoring Data Plots

List each figure and drawing included in the report. Optionally, instrumentation plots
may be placed in Section 4 of the report and a statement included in Appendix D that
the plots may be found in Section 4.
E. Inspection Photographs
Optionally, some or all of the photographs may be included in the appropriate
sections of the report. If photographs are included within the report, provide a list of
the photographs and the corresponding page number in Appendix E.
F. Inspection Checklists (Optional)
G. Operation and Maintenance Documentation (If required)

Revision 0 14-H-2 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

1.0 Executive Summary

The executive summary is intended to be a concise summary of the Part 12D
Independent Consultant’s findings, assessments, conclusions and recommendations.
The information can be provided in narrative and/or bulleted format.

1.1 General (include brief project description)

1.2 Summary Assessment of the FMA report
1.3 Summary of Field Inspection Findings
1.4 Summary of O&M status
1.5 Summary Assessment of “Supporting Technical Information” document
Note: Specifically identify any new calculations conducted by the Part 12D
Independent Consultant for this report.

1.6 Conclusions
Provide a summary of conclusions from each section of the report listed by
section.

1.7 Summary of Recommendations

Provide a summary of conclusions from section 8 of this report.

1.8 Certification
Note: By signing this document, the Part 12D Independent Consultant is stating
that the entire report has been developed by and under the direction of the
undersigned. The Part 12D Independent Consultant shall make a clear statement
that he/she generally concurs with the assumptions, methods of analyses, and
results of all studies documented in the report. The Part 12D Independent
Consultant is thus taking responsibility for the Part 12D report contents as a
Professional Engineer. It is not required to repeat this statement in each section or
sub-section of the report.
1.8.1 List of all field inspection participants
1.8.2 Reference to FERC Order 122 dated March 1, 1981 and paragraph 12.37
(c) (7).
1.8.3 Signature(s) of Part 12D Independent Consultant(s)
1.8.4 PE Stamp

See Appendix A: FERC Letter Requiring Part 12D Inspection

See Appendix B: FERC Letter Approving Part 12D Consultant - (Include date of
current report outline provided by FERC)

Revision 0 14-H-3 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

2.0 Project Description

2.1 Brief Project Description

For each major element and ancillary structure, provide a brief description of the
type of structure, general dimensions, etc. The detailed project description will be
in the “Supporting Technical Information” document.
For multi-project or development licenses, include a brief outline of how this site
fits with the other projects.

2.2 Hazard Potential Classification.

Based on views from the dam, other project works inspected and discussion with
the licensee, document any changes in upstream or downstream conditions that
might affect the Hazard Potential Classification. Review with the licensee the
methods and assumptions used to develop the IDF. If the IDF is less than the
PMF, the IC should confirm that the IDF is still valid based on an assessment of
the downstream conditions as noted above.

2.3 Summary of Standard Operating Procedures

2.3.1 Purpose of Project (Run of river, storage, flow augmentation, flood
surcharge storage, control reserve, pumped storage, etc.)
2.3.2 Reservoir rule curves by season (include operating restrictions if any)
2.3.4 Standard gate operation procedures (lead and following gates, emergency
power systems, etc.)

2.4 Modifications Conducted for Project Safety

Document any modifications to project works since the last Part 12D inspection
that have been done to improve project safety. (i.e.: spillway gates reinforced,
seepage drain, berm added, crest raised, post-tensioned anchors installed,
foundation drains or relief wells cleaned, etc.). In the next Part 12D Safety
Inspection Report, these items will become part of Section 2.1. This information
should be fully described in the updated “Supporting Technical Information”
document submitted with the Part 12D report.
Do not include routine maintenance such as unit overhaul, gate painting, etc.
Note, that generators, transformers, and transmission facilities are excluded from
the Part 12D program under 18CFR subsection 12.35.

2.5 Flood History

2.5.1 Flood of Record, PMF, IDF
2.5.2 Zero freeboard spillway capacity
2.5.3 Peak spillway discharge during last five year period
2.5.4 Peak reservoir elevation during last five year period

Revision 0 14-H-4 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

2.6 Conclusions (If any)

See Appendix C: Project Figures (Note: If the STI is bound with this report, do not
duplicate figures)

Revision 0 14-H-5 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

3.0 Discussion of Potential Failure Mode Analysis Report

Do not include security issues in the Part 12D report. For licensed projects that include
multiple independent dam and powerhouse developments, separate PFMA studies and
reports should be made each development.

3.1 General
Identify the team members, and their affiliations, who developed the
comprehensive Potential Failure Modes Analysis (PFMA) or its update. Note that
the process was in accordance with FERC Engineering Guideline Chapter 14.

3.2 Assessment of Potential Failure Mode Analysis Report

3.2.1 General
List the viable potential failure modes identified in the PFMA report.
These would generally be Category 1 through Category 3 events. Provide
an assessment of the reasonableness and completeness of the failure mode
scenario and whether the PFMs identified have a real possibility of
occurrence. Potential Failure modes should be listed in order of
importance.
3.2.2 Potential Failure Mode Scenarios
Each realistic potential failure mode description should inc lude the
sequence of conditions and events that would lead to the failure mode.
• PFM 1. (i.e. Internal erosion, piping)
• PFM 2. (i.e. Seismic induced deformation)
• Etc.
3.2.3 Assessment of Mitigation Actions for Each Potential Failure Mode
For each potential failure mode, assess the actions that can be taken to
mitigate the developing potential failure. (This would come from the
PFMA report).
• PFM 1
• PFM 2
• Etc

Revision 0 14-H-6 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

3.2.4 Assessment of Monitoring Program for Each Potential Failure Mode

For each potential failure mode, assess the monitoring (visual and
instrumentation) that exists or is recommended in the PFMA report that
will warn of development of the potential failure mode, of adverse
performance, or of an impending failure condition.
• PFM 1
• PFM 2
• Etc
3.2.5 Are there other potential failure modes that have been identified and
addressed in this report or that should be assessed?

3.3 Supporting Documentation

Has the Licensee archived all materials and documents that were utilized in the
PFMA session?

3.4 Conclusions

See “Supporting Technical Information” document: Potential Failure Mode Analysis

Study Report (Update as appropriate)

Revision 0 14-H-7 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

4.0 Performance Monitoring and Visual Surveillance with Respect to Potential

Failure Modes
Note: Review and Assessment of performance monitoring programs must always be
done from the point of view of potential failure modes. Although the primary
assessment is with respect to the potential failure modes identified in the PFMA study,
the Independent Consultant must determine if there are potential failure modes not
previously addressed or not adequately considered.
For the purposes of this section, a Threshold value is the value used in the analysis or
design, or is established from the historic record. An Action Level is the instrument
reading that triggers increased surveillance or an emergency action.

4.1 Operator’s Surveillance Program

Daily/weekly operator’s inspections and reports.

4.2 Active Instrumentation

This will vary by project. Discuss only the instruments actually at the project. Is
instrumentation in accordance with Chapter 9 of the Engineering Guidelines?
4.2.1 Piezometers
4.2.2 Weirs
4.2.3 Settlement/alignment monuments
4.2.4 Crack gages
4.2.5 Upstream river and/or rain gage stations
4.2.6 Headwater/tailwater (alarm systems)
4.2.7 etc.

4.3 Threshold and Action levels

For each instrument, or group of instruments as appropriate, provide a table of
Threshold and Action levels as defined above. If the information is included in
the STI provide an assessment of the informatio n.

4.4 Reading procedures/frequency

4.4.1 Data acquisition and evaluation (manual/automated)
4.4.2 Data evaluated in a timely manner by a qualified engineer
4.4.3 Spurious readings (are spurious readings confirmed or explanations
provided)
4.4.4 Readings compared to Threshold and Action levels defined for each
instrument

4.5 Assessment of Instrumentation Data and Performance Monitoring Programs

Relative to Potential Failure Modes
Include newly identified potential failure modes

Revision 0 14-H-8 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

4.6 Conclusions
Instrumentation plots may be presented in either Appendix D or Section 4 of the
Part 12D report. If the plots are included in Section 4, include a statement in
Appendix D that the instrumentation plots may be found in Section 4.

Revision 0 14-H-9 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

5.0 Field Inspection

5.1 Field Inspection Observations

For each element of the project (i.e.: spillway, earthfill embankment, gravity
section, intake, powerhouse, conveyance system, etc.), observe and report visual
observations of the following issues as appropriate. Include photographs of
significant project features and observations. If an inspection checklist is used,
include a copy of the checklist in this section. A site specific inspection checklist
should be formatted to include specific visual surveillance items identified in the
PFMA.
The intent of this section is to highlight changed conditions for the report
reviewer, not to document unimportant or minor details.
The report should be in text format by structure or element addressed
individually. For each structure or element of the project, The Part 12D
Independent Consultant should consider the following items:
• Settlement
• Movement – including abutments (cracks or other signs of distress or change)
• Erosion
• Seepage/Leakage
• Cracking
• Deterioration
• Spillway gate Operation/Standby Power (At a minimum, the Part 12D
Independent Consultant needs to review the licensee’s annual certificates of
spillway gate operation and interview project operating staff to assure that
emergency backup systems work and that operating personnel know how to
use them.)
• Outlet/Sluice Gate Operation
• Water conveyance systems (canals / flumes / penstocks / tunnels / surge
chambers, emergency bypass or closure systems, etc.)
• Foundation Drain/Relief Well Operation
• Evidence of high artesian or uplift pressures (structures / foundations /
abutments)
• Observations of sediment transport (piping evidence)
• Observations of seeps, wet areas, springs, green grass
• Other Pertinent Observations

5.2 Status of Response(s) to Recommendation(s) in Last Part 12D Report.

Revision 0 14-H-10 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

5.3 Field Observations with Respect to Potential Failure Modes

Document field observations pertinent to each potential failure mode noted in
Section 3

5.4 Adequacy/Operation of Public Alert Systems

Note: Are upstream spillway warning buoys, and downstream sirens and lights
operable?

5.5 Conclusions

See Appendix E: Inspection Photographs (Optionally, some or all of the photographs

may be included in the appropriate sections of the report. If photographs are included
within the report, provide in Appendix E a list of the photographs and the
corresponding page number)

See Appendix F: Inspection Check List (optional)

Revision 0 14-H-11 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

6.0 Operation and Maintenance Programs Relative to Potential Failure Modes

Do not include security issues in the Part 12D inspection report. If observations of
significant O&M issues are made, include in report for possible new potential failure
mode analysis.

6.1 Summary PFMA identified O&M issues (from PFMA report)

6.2 Operation and Maintenance Procedures

6.2.1 Communication/Response
Address adequacy and reliability of remote monitoring, communication
and control systems (Operations / Instrumentation / Telemetry – Do the
systems provide adequate reliability and redundancy? Can a specific gate,
valve or other project component be operated remotely on demand?)
6.2.2 Electrical/Mechanical Systems
• Spillway Gate Motors (line/line voltage, amperage draw)
• Standby and Redundant Power Sources
• Manual/Remote/Automatic Operation of Gates and Valves
• Gate Operation Sequence
• Icing protection (heaters/bubblers/reservoir level restriction)
6.2.3 Human Factors
• Adequate Staff for Emergency Response (Multiple Sites)
• Reliable Access Routes (winter/storm conditions)
• Training
• Electricians/Mechanics/Laborers
• Adequate Time to Respond
• Call Out Systems (time for crew to reach site after call out)

6.3 Assessment of O&M Procedures Relative to Potential Failure Modes

6.4 Conclusions

See Appendix G: Operation and Maintenance Documentation

Revision 0 14-H-12 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

7.0 Assessment of Supporting Technical Information Document

The purpose of this section is for the Part 12D Independent Consultant to assess the
contents of the “Supporting Technical Information” document compiled by the
licensee.

7.1 Potential Failure Mode Analysis Study Report (Include a statement referring to
Section 3 for a discussion of the Potential Failure Mode Analysis)
7.2 Project Description and Drawings
7.3 Construction History
7.4 Standard Operating Procedures
7.5 Assessment of Geology and Seismicity
7.6 Assessment of Hydrology and Hydraulics
7.7 Assessment of Instrumentation
7.8 Assessment of Stability and Stress Analyses of Project Structures
7.9 Assessment of Spillway Gates
7.10 Pertinent Correspondence Related to Safety of Project Works
7.11 Status of Studies in Process and Outstanding Issues
7.12 References
7.13 Conclusions

The Licensee is responsible for compiling the “Supporting Technical Information”

(STI) document.
The initial STI should be provided to the Independent Consultant and three hard copies
and two digital copies shall be submitted to the FERC. As new information is obtained,
or modifications are made to the project, the licensee will update this document as
required. Updates to this document shall be provided to the current FERC Part 12D
Independent Consultant for review, to the FERC and to other document holders.
Document holders should be requested to insert the updated pages in the STI, and add
the revision to the revision notice log in the front of the STI.
Except for the initial submittal of an STI document, if no significant changes have been
made to the STI since the prior Part 12D Inspection report, either a digital copy of the
most current STI in *.pdf, *.jpg, *.tif, or other acceptable formats (check with the
FERC for acceptability of alternative formats prior to submittal) or a hard copy of the
STI shall be included with the Part 12D report. For small projects, the STI document
may be bound with the Part 12D report.

Revision 0 14-H-13 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

8.0 Recommendations
Each Recommendation should be identified as a maintenance or dam safety item and
include a schedule for completion/implementation.
Each section of the report should be included for completeness. If there are no
recommendations pertinent to a given section of the report include that section with a
“None” comment.

8.1 Project Description

8.2 Potential Failure Mode Analysis Report
8.3 Performance Monitoring and Visual Assessment with Respect to Potential Failure
Modes
8.4 Field Inspection
8.5 Operations and Maintenance Programs Relative to Potential Failure Modes
8.6 Assessment of Supporting Technical Information Document

Revision 0 14-H-14 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

APPENDICES

List of Tables (with location)

List of Figures (with location)
List of References

A. FERC Letter Requiring Part 12D Inspection

Note: May include specific FERC concerns to be addressed by Part 12D Independent
Consultant.

B. FERC Letter Approving Consultant

Note: Include date of report outline provided by FERC.

C. Project Figures
This Appendix should include the following figures as appropriate. All Figures should
be consecutively numbered. Figures should be general without excessive detail so as to
be clearly legible. Figures should include documentation of significant changes since
last Part 12D report. If STI document to be directly bound in this report, do not
duplicate the figures. FERC Exhibit and relicensing drawings can be used.
• Location map with project facilities located including conveyance systems and
access routes from main roads and nearest town
• Plans of project facilities
• Typical sections and profiles of key project features (dams, spillways, powerhouses,
intakes, emergency/fuse plug spillways, chute profiles, etc.)
• Profiles and typical sections of water conveyance systems (canals, tunnels,
penstocks, flumes, surge chambers, etc)
• Satellite or aerial photo of project and downstream area
• Spillway and tailwater rating curves

D. Instrumentation Monitoring Data Plots

Note: Plans and cross-sections with locations of each instrument, including design
phreatic surface or uplift pressure profile, and tabulated data for each instrument are
included in the “Supporting Technical Information” document only. See Chapter 9,
Instrumentation and Monitoring, of the Engineering Guidelines for additional
information. Only time versus reading graphs are included here as NEW information.
If data plots are included in Section 4 of the Part 12D report, a statement should be
provided here directing the reader to Section 4 for the information.
• Time versus Reading data plots

Revision 0 14-H-15 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

• Plot all data to date, not just last five years (alternative is to plot last 15 years and
note historic range for each unit)
• Do not put too many instruments on one plot
• Try to put all instruments from one section or profile on the same plot
• Mark tip elevation, unscreened length, ground elevation and top of piezometer
elevation for each unit on the data plot
• Use symbols for each unit, not just colors (colors do not reproduce in black and
white and some people are color blind)
• Include headwater and tailwater levels on each plot
• Force all time scales to show full year cycles from January through December
• For multiple plots for the same project, force vertical and horizontal scales on all
plots of the same type to have the same scale or total range so plots can be directly
overlaid
• Mark threshold values
• Show monthly precipitation on one sheet
• Mark action levels requiring emergency response

E. Inspection Photographs

F. Inspection Checklist (optional)

G. Operation and Maintenance Documentation (if required)

Revision 0 14-H-16 April 11, 20030

Chapter 14 Monitoring the Performance of Dams

Appendix I
Guidelines for Supporting Technical Information

Purpose:
The purpose of the Supporting Technical Information document (STI) is to summarize those
project elements and details that do not change significantly between quinquennial FERC
Part 12D Independent Consultant Safety Inspection Reports. The licensee will create and
maintain this document for use by themselves, the Part 12D Consultant and the FERC.
The STI should include sufficient information to understand the design and engineering
analyses for the project such as:
• A complete copy of the Potential Failure Mode Analysis report
• A detailed description of the project and project works
• A summary of the construction history of the project
• Summaries of Standard Operating Procedures
• A summary of geologic conditions affecting the project works
• A summary of hydrologic and hydraulic information
• Summaries of instrumentation and surveillance for the project and collected data
• Summaries of stability and stress analyses for the project works
• Pertinent correspondence from the FERC and state dam safety organizations related
to dam safety

The STI should use tables, figures, and drawings in preference to text and should not include
complete copies of the original documents except for the “Potential Failure Mode Analysis”
study report. Only key paragraphs of the original reports should be included in this document
for clarity.
The STI is a “living” document, in that as new data or analyses become available they are
appended to the initial STI and outdated material is removed. The document should be
bound in a three ring binder to facilitate updating the STI as necessary.
The Licensee should coordinate this document with the Part 12D inspection report outline to
be sure the Independent Consultant will have all the information necessary for review of the
project. The initial STI should be provided to the Independent Consultant and three hard
copies and two digital copies shall be submitted to the FERC. Updates to this document shall
be provided to the current FERC Part 12D Independent Consultant for review, to the FERC
and to other document holders. Document holders should be requested to insert the updated
pages in the STI, and add the revision to the revision notice log in the front of the STI.
Except for the initial submittal of an STI document, if no significant changes have been made
to the STI since the prior Part 12D Inspection report, either a digital copy of the most current
STI in *.pdf, *.jpg, *.tif, or other acceptable formats (check with the FERC for acceptability
of alternative formats prior to submittal) or a hard copy of the STI shall be included with the

Revision 0 14-I-1 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Part 12D report. For small projects, the STI document may be bound with the Part 12D
report.
The complete STI should be reviewed and reprinted at least every 15 years and hard copies
submitted with the Part 12D report.
The “Supporting Technical Information” document must include a revision sheet and contain
the following sections:

Section Title
Table of Contents
1. Failure Mode Analysis Study Report
2. Description of Project Structures
3. Construction History
4. Standard Operation Procedures
5. Geology
6. Hydrology / Hydraulics
7. Instrumentation
8. Stability / Stress Analysis of Project Structures
9. Spillway Radial Gates
10. Pertinent Correspondence Related to Safety of Project Works
11. References

The information to be included in each section is described below.

Revision 0 14-I-2 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

SUPPORTING TECHNICAL INFORMATION

Revision Log
Table of Contents

1.0 Potential Failure Mode Analysis Study Report

Include a complete copy of the latest “Failure Modes Analysis” study report with all
attachments. All updates shall be included in this Section of the STI.

2.0 Description of Project Works and Project Drawings

This is a detailed description of the project and project works that is part of the Part
12D Independent Consultant review. In general, this information will come directly
from existing sources such as prior Part 12D Inspection Reports, licensing or
relicensing documents or company brochures. The detailed descriptions would include
the following elements as appropriate:
2.1 General project description including project name and owner
2.2 Project location including nearest town(s), river system, etc.
2.3 Purpose of Project
2.4 Main dam and any auxiliary dams
2.5 Spillway(s) including stilling basins
2.6 Non-overflow water retaining structures such as powerhouses
2.7 Intakes
2.8 Conveyance systems (penstocks, tunnels, surge chambers, flumes, canals,
inverted siphons, including control, regulating, and pressure relief devices, etc.)
2.9 Powerhouse(s)
2.10 Low level outlets including minimum flow devices
The following drawings shall be included
2.11 USGS Quad map or other location map with project facilities located including
conveyance system alignment
2.12 Plan of licensed project facilities and project boundaries
2.13 Typical sections and profiles of key project works (dams, spillways, powerhouses,
intakes, canals, tunnels, penstocks, flumes, surge chambers, inverted siphons, etc.)
2.14 Satellite or aerial photo of project and downstream area if available

Revision 0 14-I-3 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

3.0 Construction History

In general, this information will be copied directly from existing sources such as prior
Part 12D reports, construction reports or company brochures. Include a summary of the
project construction history based on the following sources of information:
3.1 Design reports and pertinent memoranda from licensing and permitting
documents
3.2 Laboratory investigations and construction testing reports
3.3 Field and lab geotechnical investigations
3.4 Construction reports and photographs
3.5 Specification documents
3.6 Reports of major modifications conducted for dam safety since last Part 12D
inspection
3.7 Construction chronology that includes all a summary of original construction and
all significant work completed related to project safety. Do not include routine
maintenance items such as gate painting, unit overhauls, etc.

4.0 Standard Operation Procedures

Include a statement summarizing the standard operating procedures for the project. This
section should include seasonal minimum flow requirements, lead and follow gate
sequence, reservoir level restrictions by season, etc.

5.0 Geology and Seismicity

In general, this section should be copied from existing reports and company brochures.
Include summaries of applicable information in the following sections:
5.1 Geology
5.1.1 Regional geology
5.1.2 Site geology and local foundation conditions including geologic maps,
cross-sections and profiles under the dam(s) and pertinent project works.
5.1.3 Potential landslides, loose rock formations or adverse bedding orientations
that could affect project works
5.1.4 Potential sinkhole, karst, solutioning, basalt flow issues, etc. that could
impact project works
5.1.5 Potential weak seams such as bentonite or soluble gypsum layers
5.1.6 Geologic artesian sources (geothermal, high abutments, etc.). Do not
include artesian pressures due to normal dam seepage.

Revision 0 14-I-4 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

5.2 Seismicity
5.2.1 Map of fault traces that effect project
5.2.2 Table of fault, distances, depths, magnitude at fault, PGA at site, etc.
5.2.3 Site MCE and DBE development
5.2.4 Time history of adopted earthquakes
5.2.5 Floating earthquake magnitude, PGA, and distance
5.2.6 Historic earthquake centers map

The USGS website (http://neic.usgs.gov) includes information on seismicity and

may be a useful reference.

6.0 Hydrology and Hydraulics

Provide supporting information to document the development of the Probable
Maximum Flood (PMF) and the routing of the PMF through the reservoir and project
spillways. In general, this information will come directly from existing sources such as
prior Part 12D inspection reports or company reports. The following information that
should be included as applicable:
6.1 Hydrology
6.1.1 Hydrometeorology report used
6.1.2 Probable Maximum Precipitation for general and local storms
6.1.3 Drainage basin description including drainage area
6.1.4 Antecedent conditions
6.1.5 Loss rates
6.1.6 Basin and sub-basin precipitation/runoff models
6.1.7 Unit Hydrograph
6.1.8 Reservoir inflow and outflow hydrographs for the PMF event
6.1.9 Floods of record including highest flood flows and reservoir elevations
6.2 Hydraulics – Dams
6.2.1 Project discharge-rating curves (For multiple gate spillways, outlet
structures, powerhouse units, and emergency/fuse plug spillways, include
the contribution of each component as well as the total capacity. Include
the equations used to develop the curves including overtopping and orifice
flow where appropriate).
6.2.2 Tailwater rating curve (Compare to dam break studies)
6.2.3 Normal and flood freeboard without wave action
6.2.4 Zero freeboard flood capacity (without wave allowance)
6.2.5 Inflow Design Flood (based on dam break)
6.2.6 Reservoir Probable Maximum and Inflow Design Flood outflow
hydrographs and corresponding reservoir levels
6.2.7 Freeboard for general and thunderstorm events

Revision 0 14-I-5 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

6.2.8 Stilling basin or plunge pool design flood flow

6.2.9 Operating rule curve (if storage reservoir)
6.3 Hydraulics – Water Conveyance Systems
6.3.1 Hydraulic capacity of water conveyance system(s)
6.3.2 Normal operating freeboard
6.3.3 Spillway discharge rating curve(s)
6.3.4 Summary of transie nt analysis

7.0 Instrumentation
This section is to include drawings and/or sketches showing the location of each active
instrument. Include cross-sections of project structures showing instrument tip
elevation, ground elevation and readout point location. See Chapter 9, Instrumentation
and Monitoring, of the engineering guidelines for additional discussion.
Note: time versus reading graphs for each instrument will be included only in the Part
12D inspection report.
7.1 Plans, sections, and details of active or useful reference instrumentation
If a unit has been abandoned or replaced, but the historic data is still being used
for safety evaluations, include the appropriate information for the record
7.2 Reading frequency for each instrument (reading procedures should not be
included in this document)
7.3 Procedures for resolving spurious readings
7.4 Tabulated Data for each instrument
7.5 Type of instrument (pneumatic/vibrating wire piezometer, Parshall flume, gape
gage, inclinometer, etc)
7.6 Predicted value for each instrument (threshold values are values used in design or
analysis of project structures)
7.7 Historic range of readings for each instrument
7.8 Threshold and Action level for each instrument

8.0 Stability and Stress Analyses of Project Structures

Because every dam and hydroelectric project is unique, it is not possible to list here all
the various items that are required to adequately detail stability or stress assessments of
the project water retaining structures. It will be the responsibility of the Licensee to
include all information necessary for the reader to understand the assumptions, methods
of analysis, and load cases assessed for each project structure. Stability and stress
analyses for each structure shall be summarized graphically for ease of understanding
and communication and to create a permanent record storage.
The following types of information should be provided:
8.1 General

Revision 0 14-I-6 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

8.1.1 Listing of credible load cases analyzed (including water levels for dam,
canal and flume analyses or pressure for penstock and flowline analyses)
8.1.2 Statement of the method of analysis used and the computer program
adopted.
8.1.3 Properties of materials based on site specific tests or assumptions (state
which). Include representative test data and summary sheets.
8.2 For each gravity structure and load case:
8.2.1 Graphic free body diagram (cross-section) of each structure showing:
• the assumed self weight of the cross section
• all applicable loads including, as appropriate:
• assumed uplift pressure distribution
• silt loads
• headwater and tailwater loads
• point loads
• ice load
8.2.2 Key elevations
8.2.3 Key lateral dimensions
8.2.4 Piezometer and drain locations
8.2.5 Foundation shear strength parameters
8.2.6 Minimum cohesion to meet stability criteria
8.2.7 Negative crest pressures
8.2.8 Concrete unconfined and splitting tensile strength test results
8.2.9 AAR potential or evidence, failure planes investigated, etc.

8.3 For each embankment structure and load case:

8.3.1 Graphic cross-section showing
• embankment zoning
• phreatic surface
• critical failure surfaces
• key elevations
• key lateral dimensions
• slopes
• headwater and tailwater elevations
8.3.2 Potential for uncontrolled seepage at toe
8.3.3 Summary of liquefaction analyses
8.3.4 Summary of deformation analyses

Revision 0 14-I-7 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

8.3.5 Procedures used to determine soil properties, etc.

• Soil Classification
• Atterberg limits
• etc.

8.3.6 Procedures used to determine soil strengths

• Triaxial Tests (type and loadings)
• Standard Penetration Tests
• Cone Penetration Tests
• Becker Hammer Tests
• etc.

8.4 For each arch dam load case:

8.4.1 Finite element mesh
8.4.2 Stress contours
8.4.3 Vector diagrams
8.4.4 Thrust block stability and joint sterionets
8.4.5 Pulsating load potential, etc.

8.5 For each water conveyance system that has a highlighted PFM
8.5.1 Stress and stability analyses

8.6 Summary table of factors of safety for each structure and load case, with required
value.
For embankment structures and overburden foundations, the material strengths
used in the stability analyses should be properly identified i.e. effective stress or
total stress. The methods used to determine pore pressures should also be
described.
For gravity structures, it is useful to provide a spreadsheet of the key numbers
from the analysis.

9.0 Spillway Gates

For each spillway gate type, include the following information:
9.1 Table of material properties (steel type, trunnion bearing type and friction
properties, etc).
9.2 A summary of the stress analysis computations
9.2.1 Graphic of gate model used for stress analysis
9.2.2 Table of critical stresses in each member for each load condition.
9.3 Trunnion, wheel, or other lubrication procedures, schedule, etc.
9.4 Summary of gate hoist motor load tests to date (line- line voltage, amperage draw,
reservoir level, and initial draw if available)

Revision 0 14-I-8 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

9.5 Spillway gate detailed inspection report

10.0 Pertinent Correspondence Related to Safety of Project Works

Include FERC annual operation inspection reports for the five years preceding the
current Part 12D inspection report. Include any major correspondence from FERC or
State Dam Safety Agencies related to outstanding dam safety issues for the project.

11.0 References
List of references available for review of dam safety issues and that were used to
assemble this document.

Revision 0 14-I-9 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Attachments
1) Example of Detail Expected in Supporting Technical Information Document
2) Example of Summary of Embankment Stability Analysis
3) Example of Summary of Structural Stability Analysis
4) Example of Summary of Hydrologic and Hydraulic Information
5) Example of Summary of Instrumentation and Surveillance Information
6) Example of Document Control Log Sheets

Revision 0 14-I-10 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Attachment 1
Example of Detail Expected
in
Supporting Technical Information Document

Revision 0 14-I-11 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

BASIC DATA AND ANALYSES

Spillway Adequacy maximum reservoir water surface Elevation

A. Previous Studies
The original design flood study for
Big Hole Dam was presented in "A Study
of the Maximum Probable Floods for the
Middle Fork of the Big River Project",
dated October 1961 and revised in August
1962 by ABC. The design flood developed
by that flood study, when routed through
the reservoir and spillway, resulted in a
peak discharge of 58,800 cfs at reservoir
water surface Elevation 4,643.2, leaving a
freeboard of 6.8 feet to the nominal dam
crest.
In February 1965, ABC updated
the flood study by deriving a probable
maximum flood based on U.S. Weather
Service Hydrometeorological Report No.
36 and U.S. Corps
of Engineers
reductions. This JANUARY - FEBRUARY GENERAL PMF HYDROGRAPHS
1965 study resulted
80,000
in a maximum water
surface Elevation 70,000
4,646.7, leaving a Peak Outflow 69,300 cfs
Flow - Cubic Feet per Second
4,644, leaving 6.0 feet of freeboard.
B. Methodology to Determine PMF
In 1982, FERC directed Big County
Water Agency to provide additional
information on flood hydrology for the
Middle Fork American River Project. The
Agency retained Hydrotech of Bigville to
perform that study. Hydrotech's report
"Probable Maximum Flood Study for Big
Hole, Interbay, and Little Hole Afterbay
Dams", dated October 1982, was reviewed
and excerpts presented as part of the 1986
Five Year Dam Safety Inspection report.
Excerpts from that report describing the
methodology used to determine the PMF
for Big Hole Dam follow.
“The probable maximum floods
HELL HOLE DAM
Big Hole Dam
4,670
Peak Inflow 71,600 cfs
4,665

freeboard of 3.3 feet 60,000 4,660

to the nominal dam Reservoir Elevation - Feet
crest elevation. 50,000 4,655

In 1973, T. J. 40,000 Peak Reservoir 4,650

Corwin reviewed the Elevation 4644.8
flood studies, made 30,000 4,645
some independent
20,000 4,640
evaluations, and
concluded that the 10,000 4,635
Spillway Crest Elevation - 4630
maximum inflow
flood would be 0 4,630
66,000 cfs. When 0 6 12 18 24 30 36 42 48 54 60 66 72 78 84
routed through the Hours
reservoir and
spillway, the peak
discharge would be 55,000 cfs with a
Figure 1
Revision 0 14-I-12 April 11, 2003
Chapter 14 Monitoring the Performance of Dams

are based upon a probable maximum
storm that could occur during January or
February. The storm is estimated using
procedures presented in Hydro-
meteorological Report No. 36, entitled
“Interim Report, Probable maximum
Precipitation in California”, which was
originally issued in October 1961, and was
revised in October 1969.”
“The probable maximum storm
was distributed into hourly precipitatio0n
amounts, hourly melting of snowpack was
calculated using probable maximum storm
temperature and winds, immediate surface
runoff losses were subtracted from the
hourly precipitation and snowmelt, and the
remaining excess amounts of precipitation
and snowmelt along with spill from any
upstream reservoirs were routed through
the basin stream channels to the basin
discharge point…”
The Hydrotech report is organized
into three sections describing the basin
characteristics, the flood analysis, and the
probable
maximum HELL HOLE DAM - SPILLWAY RATING CURVE
4,650 Big Hole Dam Spillway Rating Curve
flood
calculation. 4,648
The Clark 4,646
Reservoir Elevation - Feet
snowmelt associated with the probable
maximum storm, applying losses based
upon losses during historic storms, and
routing the sub-basin runoff to the
appropriate reservoir.” “Each reservoir
was assumed to be full at the start of the
probable maximum flood.”
The flood routing from the
Hydrotech study is presented on Figure 1.
Peak flood inflow was found to be 71,600
cfs and peak outflow was 69,300 cfs at a
maximum reservoir stage of 4,644.8.
It is considered that the Hydrotech
flood study was carried out in accordance
with previously appropriate standards and
procedures. However, in May 1996 and
January 1997, large rainfall floods
occurred which could affect the unit
hydrographs and loss rates used to develop
the probable maximum flood (PMF) from
the probable maximum precipitation
(PMP). In 1999, the National Oceanic and
Atmospheric Administration and the Army
Corps of Engineers issued

Unit Graph 4,644

method (C. O.
4,642
Clark,
“Storage and 4,640
the Unit 4,638
Graph” ASCE 4,636
Transactions,
4,634
1945.)
4,632
“Proba
4,630
ble maximum
0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000
flood inflows
Discharge - Cubic Feet per Second
to Big Hole
Reservoir …
have been calculated using estimated Figure 2
probable maximum precipitation,
Hydrometeorological Report (HMR) No.
Revision 0 14-I-13 April 11, 2003
Chapter 14
59, which superseded HMR No. 36 on
which the precipitation developed in the
Hydrotech 1982 study was based. While
the precipitation, loss rates and unit
hydrographs used in the Hydrotech study
were appropriate and conservative for that
time, the precipitation is no longer current
data and the loss rates and unit
hydrographs should be reviewed as a
result of the 1996 and 1997 storms.
The maximum reservoir stage for
the routed flood left 5.2 feet of freeboard to
the nominal dam crest elevation. That
freeboard is judged to be adequate for a
dam the height, configuration and material
of Big Hole Dam but a new flood study
must be prepared for Big Hole Dam and the
freeboard adequacy must be reviewed when
the results of that study are available.
C. Spillway Rating Curve
The spillway rating curve from the
Hydrotech 1982 report is presented on
Figure 2. That rating curve has again been
reviewed and found to be conservative and
appropriate for the spillway at Big Hole
Dam.
Structural Stability
A. Previous Studies
The original design analysis was
carried out by AB Engineers (ABE) as
reported in their 1982 Engineering Data
Report. Harlan Miller Tait Associates
(HMTA) updated the Stability Analysis for
the dam as a supplement, dated April 4,
1984, to the fourth Five Year Dam Safety
Inspection Report. In the fifth Five Year
Dam Safety Inspection Report (HMTA,
1986), the stability analysis was again
updated using additional seismicity data
and a more complete analysis.
Revision 0
Monitoring the Performance of Dams
To comply with FERC criteria, a
Simplified Displacement Analysis (SDA)
was performed as part of the 1991 Five
Year Dam Safety Inspection by HTA.
Data and analysis descriptions from those
HTA reports are summarized in this
Appendix.
B. Method of Analysis
1. Static Analysis
To analyze the static loading
conditions, the static stability analysis
computer program TSLOPE (by TAGA
Engineering Software Services of San
Ramon, California) was used.
The Spencer's method program option
was selected to determine the factor of
safety of a slope using noncircular failure
surfaces selected by the investigator. The
sliding mass is divided into slices, and all
interslice side forces are parallel to each
other. Spencer's method satisfies
equilibrium conditions fo r overall moment,
individual slice moment, and vertical and
horizontal forces.
2. Earthquake Deformation
Analysis
For the earthquake deformation
analysis, a simplified displacement
analysis (SDA) was carried out. The
controlling causative fault for the Big Hole
Dam site is the undefined “local
earthquake” that is capable of an MCE of
M=5 at a distance of 5 km that could result
in a peak ground acceleration (PGA) at
Big Hole Dam of up to 0.19g (Section IV,
E. Seismicity) (Sadigh, et.al., 1997). A
conservative seismicity of 0.2g was used
in the 1991 analysis.
14-I-14 April 11, 2003
Chapter 14
The SDA was performed in three
steps:
1. A yield acceleration was
determined using static analysis for
failure surfaces that would intersect the
crest and adversely affect the freeboard.
2. A time-history of accelerations in
the dam caused by the controlling
earthquake was calculated for the
selected yield acceleration surface.
3. The time-history accelerations that
exceeded the yield acceleration were
used to estimate incremental
displacements, which were averaged to
calculate the permanent displacement
along the failure surface.
TSLOPE was used to determine the
yield acceleration for the selected failure
surfaces. The yield acceleration is defined
as the pseudostatic seismic coefficient
necessary to reduce the static factor of
safety for the selected failure surface to 1.0,
the point of incipient static failure. The
Spencer's method program option was
selected to determine the factor of safety of
a slope using failure surfaces selected by
the investigator, as described under Static
Analysis, above. Various failure surfaces,
which were selected to intersect the crest of
the dam and to adversely affect the
freeboard, were tried until several surfaces
with lower yield accelerations were found.
The program SHAKE, developed at
the University of California at Berkeley by
Per Schnabel, John Lysmer, and H. Bolton
Seed, was used to model the dam's dynamic
response to site-modified earthquake
records and to estimate the time-history of
acceleration on the selected yield
acceleration surfaces. SHAKE is used to
compute the dynamic response of a one-
dimensional system of infinitely long,
homogenous, visco-elastic layers subjected
to vertically traveling shear waves. The
Revision 0
Monitoring the Performance of Dams
program is based on the continuous
solution to the wave-equation (Kanai,
1951) adapted for use with transient
motions through the use of the Fast Fourier
Transform Algorithm (Cooley and Tukey,
1965). The input to the program is discrete,
consisting of acceleration values spaced at
a constant time interval. The maximum
absolute acceleration value and the time
interval between the acceleration values are
varied so that the acceleration and
predominant period of the record matches
those expected at the site. The earthquake
records chosen to model the site were the
1954 Taft earthquake and a synthetic
motion created by H. B. Seed and I. M.
Idriss, both scaled to a PGA of 0.20g. The
output from SHAKE is also in discrete
format, consisting of acceleration values at
constant time intervals that represent the
acceleration time- history curve of points
within the soil profile chosen by the
investigator.
The dynamic response of the dam
embankment due to an earthquake with a
PGA of 0.20g was computed on the
maximum section at several vertical
locations in profiles located near the axis of
the dam, near mid-slope, and near the
embankment toe (see Figures D-3 and D-4)
using the program SHAKE. In a trial and
error process, the material stiffness
parameters were softened at the failure
surface until the average acceleration above
the failure surface was approximately equal
to the yield acceleration. This procedure
was required to keep the system in static
equilibrium. The acceleration time history
at the failure surfaces at the various profiles
was then computed, and the maximum
value is presented on the Figures.
Typically, the Taft motion produced greater
accelerations at the failure surfaces.
The program DISPLMT, developed
at Arizona State University by William
14-I-15 April 11, 2003
Chapter 14
Houston and Sandra Houston, was used to
compute expected permanent
displacements in the dam embankment
using the acceleration time-history on the
selected yield failure surfaces. The
program estimates permanent
displacements by double integrating
accelerations above the yield acceleration
produced at the failure surface. The
program utilizes the Newmark numerical
method to perform the integration,
determining the area under the portion of
the acceleration time-history curve that
exceeds the minimum yield acceleration.
The DISPLMT program calculates
the movements along the failure surface
using the acceleration time-history
generated by SHAKE as the response or
input motion. The yield acceleration is
assumed constant with respect to time and
displacement. The calculated displacement
varies from the near-axis, middle, and toe
profiles of the embankment slope.
C. Material Properties
ABE data (1962 and 1968) indicate
that the foundation was to be on
unweathered rock. As-built drawings show
the foundation to be hard, massive granitic
rock. For the purpose of the stability
analysis, it is considered that the foundation
is substantially stronger than the
embankment.
Materials obtained from the borrow
areas and used in the core of the
embankment were tested by ABE for their
strength characteristics. These tests are
summarized in their report (ABE, 1962),
and the design parameters adopted are
given below. The material strength
parameters discussed in that report were
judged conservative and appropriate;
therefore, those same parameters were used
in the 1986 and 1991 analysis and are as
follows:
Revision 0
Monitoring the Performance of Dams
Material Properties Used in Analyses
Parameter Zone Zone
1 2-7
Moist Unit Weight, γm (pcf) 127 124
Saturated Unit Weight, γs 130 140.4
(pcf)
Friction Angle, Ø (degrees) 33 40
Cohesion, c (psf) 200 0
D. Phreatic Surface Assumptions and
Seepage Pressure Distribution
The phreatic surfaces and
hydrostatic forces are based on the water
levels shown on the stability analysis
drawings (Figures D-3 and D-4). Uniform
head loss was used through the core, and it
was considered that the transition, drain
material, and rockfill are free draining with
no head loss in the upstream rockfill and
transition. In 1984, FERC questioned the
assumption that the upstream rockfill
would be free-draining with respect to rapid
drawdown stability conditions. An analysis
of the permeability of the upstream shell
with respect to maximum rate of rapid
drawdown was made and presented in the
1986 Five Year Dam Safety Inspection
report. The assumption that the upstream
rockfill was free-draining with respect to
rapid drawdown was affirmed.
E. Stability Conditions, Minimum
Criteria, and Calculated Factors of
Safety
The stability conditions, minimum
criteria, and the calculated factors of safety
are presented in the following table. The
minimum criteria factors of safety are from:
“Engineering and Design Stability of Earth
and Rockfill Dams,” EM1110-2-1902, by
the Corps of Engineers, April 1, 1970.
14-I-16 April 11, 2003
Chapter 14
For the earthquake deformation
analysis, the criterion adopted is that the
displacement must not be great enough to
lower the crest of the dam below the
maximum normal storage elevation in the
case of occurrence of the MCE.
Stability analyses were performed
for the maximum dam section with the
crest at elevation 4660 to allow for the
original camber. Pool elevations used in
the analysis are elevation 4640, 10 feet
above normal maximum operating level,
PMF pool elevation 4646, and partial pool
elevations 4340, 4465, and 4565.
Analyses were performed for full
suites of possible failure surfaces for each
of the loading conditions listed above. For
each static loading case, we found a surface
with a lowest factor of safety and bounded
above and below with more stable slip
surfaces. To determine the yield
accelerations for the earthquake
deformation analysis, the same procedure
was used. Earthquake time-histories and
displacement analyses were performed on
only the most critical surfaces (based on the
results of the yield acceleration analyses)
that, based on their location relative to the
crest, adversely affect the freeboard of the
dam.
F. Summary of Results - Embankment
The detailed results of the analyses
were presented on Figures VII-1 and VII-2
in the HTA 1991 report, and those figures
are reproduced as Figures D-3 and D-4 in
this Append ix. A Summary of the results is
presented in the table following.
An SDA was performed on the
downstream failure surface with the lowest
yield acceleration (Figure D-3, Case 3,
Surface G). The maximum expected
displacement was calculated to be 0.01 feet.
Revision 0
Monitoring the Performance of Dams
On the upstream slope, the failure
surface with the lowest yield acceleration
(Figure D-4, Case 8, Surface E) was judged
to not adversely affect the freeboard of the
dam if displacement occurred. Therefore,
an SDA was performed on the failure
surface with next lowest yield acceleration,
Surface D with water at Elevation 4465.
The maximum expected displacement was
calculated to be 0.01 feet.
The results of these stability and
displacement analyses were reviewed as
part of this (2001) Inspection. The results
indicate that the dam has satisfactory
factors of safety for all static loading
conditions and that the deformations under
earthquake loading are expected to be small
and will not reduce the freeboard. The
seismicity used for the displacement
analysis is judged to be conservative and
appropriate.
References
Cooley, J. W., and Tukey, J. W., 1965,
“An Algorithm for the Machine
Calculation of Complex Fourier Series,”
Mathematics of Computation, Vol. 19,
No. 90, pp. 297-301.
Harlan Miller Tait Associates, 1984,
Stability Investigation, Big Hole Dam.
Harlan Miller Tait Associates, 1986, Five-
Year Dam Safety Inspection Report, Big
Hole Dam.
Harlan Tait Associates, 1991, Five-Year
Dam Safety Inspection Report, Big Hole
Dam.
McCreary Koretsky Engineers, 1962, “Big
Hole Dam Engineering Data,” dated
March, 1962.
McCreary Koretsky Engineers, 1968,
“Placer County Water Agency, Middle
Fork American River Project, Big Hole
Dam, Sections and Details,” dated January,
1968.
14-I-17 April 11, 2003
Chapter 14 Monitoring the Performance of Dams

Sadigh, K.;Chang; Egan; Makdisi; Youngs

(Geomatrix) Attenuation Relationships for
Shallow Crustal Earthquakes Based on
California Strong Motion Data.
Seismological Research Letters, Volume
68, Number 1; January/February 1997,
Seismological Society of America

Revision 0 14-I-18 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Case Slope/Condition Min. Criteria F.S. Min. Computed F.S.

1 Downstream - Pool Elev. 4640, 10 1.5 1.47
feet above Normal Maximum
Operating level
2 Downstream/PMF 1.4 1.47
Pool Elev. 4646
3 Downstream - Pool Elev. 4640 with N. A. Displacement 0.01 ft.
Seismic Yield Accel. 0.22
4 Upstream - Pool Elev. 4640 1.5 2.33
5 Upstream - PMF 1.5 2.34
Pool Elev. 4646
6 Upstream Pool Elev. 4640 with N.A. Max. Displacement 0.01 ft.-
Seismic Minimum Yield Accel. 0.30
7 Upstream: Pool Elev. 4340 1.5 1.98
Pool Elev. 4465 1.88
Pool Elev. 4565 2.10
8 Upstream w/ Seismic: Pool El. 4340 N. A. Min. Yield Accel's.0.34
Pool Elev. 4465 0.25
Pool Elev. 4565 0.24
9* Upstream/Rapid Drawdown from 1.2 1.95
Normal Maximum Operating Pool to
Elev. 4340
*Results of Analysis taken from 1986 Five Year Dam Safety Inspection Report
(HMTA, 1986).

Revision 0 14-I-19 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Attachment 2
Example of Summary
of
Embankment Stability Analysis

Revision 0 14-I-20 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

TABLE 5.2.1 - MATERIAL PROPERTIES

Stability Analysis of Embankments
SAMPLE Project

Material Description Ysat (pcf) Ydry (pcf) Φ ' (deg.) c' (tsf)
Lower Reservoir
Random Fill 147 135 40 0
Compacted Rockfill 144 130 45 0
Filter 141 125 35 0
Clay Core 134 113 25 0
Silt Core 140 — 35 0
Sand/Gravel Foundation 141 125 35 0
Red Silt Foundation 140 — 40 0
Berm Fill 132 110 30 0
Upper Reservoir Dike

Random Fill 147 135 40 0

Core/Random Fill 147 135 40 0
Filter 141 125 35 0
Sand/Gravel Foundation 141 125 35 0

Ysat = saturated (total) unit weight (pcf)

Ydry = dry unit weight (pcf)
Φ' = effective stress friction angle (degrees)
c' = effective stress cohesion intercept (tsf)

Revision 0 14-I-21 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

TABLE 5.2.3 - FACTORS OF SAFETY

Lower Reservoir Sta. 12+00
Sample Project

LOWER RESERVOIR DAM EMBANKMENT1

STA.12+00
Factor of Safety
Load Description Reservoir Downstream Slope Upstream Slope
Case Elevation
(feet) Calc. Req'd Calc. Req'd
I Normal maximum pool with 900.5 1.74 1.5 1.95 1.5
steady seepage 2.442 2.132
II Flood surcharge (PMF) 908.5 1.66 1.4 N/A N/A
III Rapid drawdown 860.0 N/A N/A 1.51 1.1
IVa Normal maximum pool 900.5 1.25 1.0 N/A N/A
w/earthquake3
IVb Rapid drawdown 860.0 N/A N/A 1.03 1.0
w/earthquake3

1. From: Second Safety Inspection Report, Supplement 2, dated February 1988.

2. Based on infinite slope.
3. Pseudo-static earthquake coefficient, ah = 0.10g.

Revision 0 14-I-22 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-23 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

TABLE D.5.2.4 - Factors of Safety

Upper Reservoir Sta. 21 + 50
Sample Project

UPPER RESERVOIR DIKE1 STA. 21+50

Load Case Description Reservoir Factor of Safety

Elevation
(feet) Downstream Slope Upstream Slope
Calc. Req'd Calc. Req'd
I Normal maximum pool with 2003 2.31 1.5 2.97 1.5
steady seepage 2.002 1.242
III Rapid drawdown 1940 N/A N/A 1.84 1.1
IIIa Normal maximum pool 2003 1.56 1.0 N/A N/A
w/earthquake3
IIIb Rapid drawdown 1940 N/A N/A 1.24 1.0
w/earthquake3

1. From: Second Safety Inspection Report, Supplement 2, dated February 1988.

2. Based on infinite slope.
3. Pseudo-static earthquake coefficient, ah = 0.10g.

Revision 0 14-I-24 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-25 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Attachment 3
Example of Summary
of
Structural Stability Analysis

Revision 0 14-I-26 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

VALUES AND ASSUMPTIONS

STABILITY ANALYSIS
CONCRETE SECTIONS

1. Nomenclature:
Effective Length = Uncracked Portion of Base
Σ FH = Summation of Horizontal Forces - Kips
Σ FV = Summation of Vertical Forces - Kips
Σ MR = Summation of Resisting Moments - Kip-Ft.
Σ MO = Summation of Overturning Moments - Kip-Ft.
MR = Factor of Safety Against Overturning
MO
FH = Coefficient of Sliding
FV

2. Unit Weight of Concrete: 150 lbs./cu. ft.

3. Unit Weight of Water: 62.4 lbs./cu. ft.

4. Uplift Pressure:
The base pressure was assumed to vary linearly from full head-water pressure at the
upstream side to full tailwater pressure at the downstream side taken over 100% of the
base area for each case analyzed.
For analyses which included a reduction in uplift due to foundation drainage, the
drains were assumed 50% effective.

Uplift pressure at drain = TW = 0.5(HW - TW), where HW and TO are the headwater
and tailwater pressures, respectively.
Full headwater pressure over 100% of the area was assumed to extend into the concrete
bedrock contact during any case where an assumed crack formed due to the existence of
tension stresses in the section foundation. The pressure was assumed to vary linearly
from full headwater pressure at the upstream end of the uncracked effective base length
to full downstream tailwater pressure at the downstream face.
Due to the transient or short-term nature of earthquake loading, the uplift is not changed
from the pre-earthquake condition due to further propagation of a tensile crack.
In the event of a tensile crack extending from the heel to the drain', the foundation drains
were assumed of greater capacity than the crack. This will result in an uplift pressure
distribution equal to that without the crack (full headwater at heel and TW + 0.5(HW-
TW) at drain).

Revision 0 14-I-27 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

5. Lateral Water Pressure:

Headwater pressures were computed using the full heights of water to headwater
elevations over the projected height of the structures. Tailwater pressures were
computed using full heights of water Co tailwater elevations for nonoverflow
sections and at 60% of full value for cases where deep flow occurs over the ogee
spillway, in accordance with U.S. Army Corps of Engineers EM 1110-2-2200
"Gravity Dam Design."
Tailwater pressures were computed at 100% full value when deep flow occurs over the
ogee spillway such that the structure becomes completely submerged, in accordance with
data from U.S. Bureau of Reclamation presented in Open Channel Hydraulics by Chow,
Ven Te (1959). Figures 14-17 and 14-18.

6. Ice Load:
5 kips per linear foot at normal water level. If the normal water level is
maintained by pin flashboards, water level and ice load are assumed to be at the
top of the concrete ogee.

7. Earthquake:
An acceleration of 0.10 g was applied in a horizontal direction.
The hydrodynamic force was determined using a method presented in Design of
Small Dams, USBR, pages 336-338.

8. Resistance to Sliding:
Where the ratio of FH/FV is greater than 0.75, the shearing resistance of the foundation
to horizontal movement must be investigated using the Shear Friction Formula.
The factor of safety against sliding is determined by the Shear Friction Formula
as:
Sg- f = f V + CA
H
where:
f = Coefficient of the angle of internal friction of foundation material (Tan Φ = 0.75)
ΣV = Summation of vertical forces
c = Unit shearing strength at zero normal load on foundation material (0.192 ksi)
A = Area of potential failure plane (area of base in compression)
ΣH = Summation of horizontal forces

According to U.S. Army Corps of Engineers, Engineering Technical Letter No. 1110-2-256,
dated June 24, 1981, which is intended to supersede portions of EM 1110-2-2200 "Gravity
Dam Design" criteria, the minimum allowable Ss-f for static loading conditions is 2.0, and for
seismic loading conditions, 1.3. Typical values of "f" and "c" were taken from "The Sliding
Stability of Dams" by Harald Link in Water Power Magazine, March, April & May, 1969.

Revision 0 14-I-28 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

9. Bearing Pressure:
Maximum bearing stress = 20 tsf on bedrock (278 psi).

Revision 0 14-I-29 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

SAMPLE PROJECT
CASES USED IN STABILITY ANALYSIS OF CONCRETE STRUCTURES

CASE I NORMAL OPERATING WATER LEVELS

H.W.L = 242.0
T.W.L = 157.0

CASE II NORMAL OPERATING LEVELS WITH ICE

H.W.L = 242.0
T.W.L = 157.0
ICE LOADS 5 kips/ft

CASE III NORMAL OPERATING WATER LEVELS WITH EARTHQUAKE

Water Levels same as CASE I
Horizontal acceleration due to earthquake is 0.10g

CASE IV PROJECT DESIGN FLOOD

H.W.L = 249.0
T.W.L = 157.0

Revision 0 14-I-30 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-31 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-32 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-33 April 11, 2003

 Chapter 14 Monitoring the Performance of Dams

SAMPLE PROJECT
STABILITY SUMMARY
CONDITION BASE FH FV FH S s-f Resultant from MR MO MR BASE STRESS (psi)
TOT LEN CRK LEN EFF LEN (kips) (kips) FV Downstream MO Upstream Downstream
Spillway

CASE I 87.0 0.0 87.0 15170 30880 0.49 10.94 42.24 2112000 807300 2.62 37.5 44.7

CASE II 87. 0 0.0 87.0 15170 30880 0.50 10.73 41. 38 2112000 834000 2.53 35.1 47.1

CASE III 87.0 0.0 87.0 20440 30880 0.66 8.12 36.37 2112000 988600 2.14 20. 9 61.3

CASE IV 87.o 0.0 87.0 17610 29870 0.59 9.38 38.12 2122000 987000 2.15 25.3 54.7

North Bulkhead

CASE I 95.0 0.0 95.0 18600 45210 0.41 10.21 37.86 3147000 1209000 2.60 38.9 71.2

CASE II 95.0 0.0 95.0 18910 45210 0.42 10.04 42.20 3147000 1239000 2.54 36.6 73.5

CASE III 95.0 0.0 95.0 26010 45210 0.58 7.3 36.33 3147000 1505000 2.09 16.2 94.0

CASE IV 95.0 0.0 95.0 21320 43980 0.48 8.86 38.76 3147000 1430000 2.20 25.2 62.1

Non-Overflow
Section
CASE I 75.6 0.0 75.6 18590 37710 0.49 8.20 32.36 2249000 1028000 2.19 32.8 82.6

CASE II 75.6 0.0 75.6 18890 37710 0.50 8.07 31.55 2249000 1058000 2.12 29.1 86.3

CASE III 75.6 1.2 74.4 25100 37710 0.67 5.99 24.80 2249000 1513000 1.71 0.0 117.3

CASE IV 75.6 0.0 75.6 21310 38670 0.55 7.18 28.80 2263000 1147000 1.97 17.0 101.1

Revision 0 14-I-34 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Attachment 4
Example of Summary
of
Hydrologic and Hydraulic Information

Revision 0 14-I-35 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

SAMPLE PROJECT

Revision 0 14-I-36 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

SAMPLE PROJECT
SPILLWAY RATING CURVE

Revision 0 14-I-37 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

SAMPLE PROJECT
TAILWATER RATING CURVE

Revision 0 14-I-38 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Attachment 5
Example of Summary
of
Instrumentation and Surveillance Information

Revision 0 14-I-39 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-40 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-41 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

TABLE 1 - PIEZOMETER READINGS

PIEZO PIEZO
PRESS ELEV PRESS ELEV
PIEZO DATE (psi) (ft) PIEZO DATE (psi) (ft)
P2-3 1/30/1997 0.0 1685.0 P4-4 8/30/1997 37.0 1770.5
P2-3 2/28/1997 0.0 1685.0 P4-4 9/30/1997 38.0 1772.8
P2-3 3/30/1997 0.0 1685.0 P4-4 10/30/1997 36.0 1768.2
P2-3 4/30/1997 0.0 1685.0 P4-4 11/30/1997 36.0 1768.2
P2-3 5/30/1997 0.0 1685.0 P4-4 12/30/1997 36.0 1768.2
P2-3 6/30/1997 0.0 1685.0 P4-4 1/30/1998 36.0 1768.2
P2-3 7/30/1997 0.0 1685.0 P4-4 2/28/1998 35.0 1765.9
P2-3 8/30/1997 0.0 1685.0 P4-4 3/30/1998 35.0 1765.9
P2-3 9/30/1997 0.0 1685.0 P4-4 4/30/1998 36.0 1768.2
P2-3 10/30/1997 0.0 1685.0 P4-4 5/30/1998 36.0 1768.2
P2-3 11/30/1997 0.0 1685.0 P4-4 6/30/1998 36.0 1768.2
P2-3 12/30/1997 0.0 1685.0 P4-4 7/30/1998 36.0 1768.2
P2-3 1/30/1998 0.0 1685.0 P4-4 8/30/1998 36.0 1768.2
P2-3 2/28/1998 0.0 1685.0 P4-4 9/30/1998 36.0 1768.2
P2-3 3/30/1998 0.0 1685.0 P4-4 10/30/1998 36.0 1768.2
P2-3 4/30/1998 0.0 1685.0 P4-4 11/30/1998 36.0 1768.2
P2-3 5/30/1998 0.0 1685.0 P4-4 12/30/1998 35.0 1765.9
P2-3 6/30/1998 0.0 1685.0 P4-4 1/30/1999 35.0 1765.9
P2-3 7/30/1998 0.0 1685.0 P4-4 2/28/1999 36.0 1768.2
P2-3 8/30/1998 0.0 1685.0 P4-4 3/30/1999 35.0 1765.9
P2-3 9/30/1998 0.0 1685.0 P4-4 4/30/1999 35.0 1765.9
P2-3 10/30/1998 0.0 1685.0 P4-4 5/30/1999 35.0 1765.9
P2-3 11/30/1998 0.0 1685.0 P4-4 6/30/1999 36.0 1768.2
P2-3 12/30/1998 0.0 1685.0 P4-4 7/30/1999 36.0 1768.2
P2-3 1/30/1999 0.0 1685.0 P4-4 8/30/1999 36.0 1768.2
P2-3 2/28/1999 0.0 1685.0 P4-4 9/30/1999 36.0 1768.2
P2-3 3/30/1999 0.0 1685.0 P4-4 10/30/1999 36.0 1768.2
P2-3 4/30/1999 0.0 1685.0 P4-4 11/30/1999 36.0 1768.2
P2-3 5/30/1999 0.0 1685.0 P4-4 12/30/1999 35.0 1765.9
P2-3 6/30/1999 0.0 1685.0 P4-4 1/27/2000 36.0 1768.2
P2-3 7/30/1999 0.0 1685.0 P4-4 2/29/2000 36.0 1768.2
P2-3 8/30/1999 0.0 1685.0 P4-4 3/31/2000 36.0 1768.2
P2-3 9/30/1999 0.0 1685.0 P4-4 4/27/2000 36.0 1768.2
P2-3 10/30/1999 0.0 1685.0 P4-4 5/30/2000 36.0 1768.2
P2-3 11/30/1999 0.0 1685.0 P4-4 6/27/2000 36.0 1768.2
P2-3 12/30/1999 0.0 1685.0 P4-4 7/31/2000 36.0 1768.2
P2-3 1/27/2000 0.0 1685.0 P4-4 8/30/2000 36.0 1768.2
P2-3 2/29/2000 0.0 1685.0 P4-4 9/29/2000 37.0 1770.5
P2-3 3/31/2000 0.0 1685.0 P4-4 10/30/2000 36.0 1768.2
P2-3 4/27/2000 0.0 1685.0 P4-4 11/29/2000 36.0 1768.2
P2-3 5/30/2000 0.0 1685.0 P4-4 12/27/2000 35.0 1765.9
P2-3 6/27/2000 0.0 1685.0 P4-4 2/1/2001 35.0 1765.9
P2-3 7/31/2000 0.0 1685.0 P4-4 2/28/2001 35.0 1765.9

Revision 0 14-I-42 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Plot B-5 - Dam Embankment Piezometers in Core

2000
1975
1950
1925

1900

1875
1850
Elevation (FT)

1825
1800
1775

1750
1725

1700
1675
1650
1625

1600

Nov-01
Jan-92

Jun-97

Jul-98
Feb-93

Mar-94

Apr-95

May-96

Sep-99

Dec-02
Oct-00
Date

P2-3 P4-4 RESRNDBT

PLOT of PIEZOMETER READINGS

Revision 0 14-I-43 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-44 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

SAMPLE PROJECT
Summary of Vertical Movements (Spring 1993 to Spring 1996)

Point To F93 95% S94 95% F94 95%

A=> Conf. Region A=> Conf. Region A=^ Conf. Region
From (mm) (mm) (mm) (mm) (mm) (mm)

SM1 S93 -0.278 1.134 -0.426 1.159 -0.212 1.143

Incremental -0.278 1.134 -0.444 1.133 0.511 1.142
SM2 S93 -0.641 1.130 -0.502 1.157 -0.330 1.138
Incremental -0.641 1.130 -0.159 1.130 0.470 1.138
SM3 S93 -0.672 1.130 -0.609 1.158 -0.105 1.136
Incremental -0.672 1.130 -0.234 1.128 0.801 1.135
SM4 S93 -0.94 1.140 -0.946 1.185 -0.364 1.145
Incremental -0.94 1.140 -0.359 1.140 0.936 1.145
SM5 S93 -1.388 1.138 -0.853 1.185 -0.399 1.143
Incremental -1.388 1.138 -0.183 1.138 0.807 1.143
SM6 S93 -0.887 1.136 -0.548 1.184 -0.278 1.140
Incremental -0.887 1.136 -0.014 1.137 0.623 1.141
SM7 S93 -0.893 1.135 -0.119 1.184 -0.196 1.138
Incremental -0.893 1.135 0.421 1.135 0.276 1.138
WP1 S93 0.210 0.843 -0.209 0.750 -0.111 0.853
Incremental 0.210 0.843 -0.790 0.843 0.468 0.853
WP2 S93 0.057 0.845 1.224 1.184 0.390 0.848
Incremental 0.057 0.845 0.815 0.845 -0.482 0.848
MM6 S93 -0.236 0.777 0.188 0.672 -0.228 0.781
Incremental -0.236 0.777 0.023 0.777 -0.015 0.782
MM7 S93 -0.031 0.124 0.022 0.078 -0.051 0.125
Incremental -0.031 0.124 -0.048 0.124 0.028 0.125

Revision 0 14-I-45 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-46 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

SAMPLE PROJECT
Summary of Horizontal Movements (Spring 1993 to Spring 1996)

Point To F93 95% S94 95% F94 95%

A=> Direction Conf. Region A=> Direction Conf. Region A=^ Direction Conf. Region
From (mm) (degrees) (mm) (mm) (degrees) (mm) (mm) (degrees) (mm)

SM1 S93 1.955 76 1.85 0.846 329 1.751 0.935 352 1.582
Incremental 1.955 76 1.85 2.813 290 2.073 0.752 149 1.746
SM2 S93 2.463 39 1.752 1.227 356 1.742 1.389 0 1.606
Incremental 2.463 39 1.752 1.933 268 2.182 0.627 165 1.833
SM3 S93 2.277 35 1.793 1.593 357 1.782 0.981 13 1.658
Incremental 2.277 35 1.793 1.677 279 2.264 1.353 165 1.922
SM4 S93 1.797 32 1.802 1.286 12 1.749 1.028 331 1.742
Incremental 1.797 32 1.802 0.943 280 2.157 1.204 218 1.682
SM5 S93 1.004 36 1.798 0.739 96 1.982 0.776 351 1.723
Incremental 1.004 36 1.798 0.555 187 2.071 0.904 291 2.11
SM6 S93 1.487 84 1.447 0.935 64 1.376 1.776 356 1.705
Incremental 1.487 84 1.447 0.947 300 1.868 1.361 314 1.719
SM7 S93 2.942 57 1.779 0.288 124 1.875 2.003 2 1.625
Incremental 2.942 57 1.779 2.896 235 1.867 1.877 354 1.762
WP1 S93 2.355 56 1.821 0.63 299 1.318 1.089 33 1.197
Incremental 2.355 56 1.821 2.839 265 1.96 1.268 102 1.375
WP2 S93 3.648 133 1.781 1.946 214 1.645 2.515 277 1.591
Incremental 3.648 133 1.781 3.953 282 1.848 2.291 320 1.734
MM6 S93 0.557 215 1.308 0.84 299 1.308 0.236 231 1.204
Incremental 0.557 215 1.308 0.988 312 1.447 0.906 131 1.337
MM7 S93 1.053 289 1.426 0.997 158 1.374 0.708 211 1.115
Incremental 1.053 289 1.426 1.152 147 1.73 0.312 234 1.29
MM8 S93 0.346 249 0.961 4.087 242 3.494 0.395 262 0.916
Incremental 0.346 249 0.961 0.605 245 1.052 0.732 52 0.907
MM9 S93 0.18 63 1.458 -0.218 159 0.651 0.357 107 1.511
Incremental 0.18 63 1.458 1.468 87 1.808 1.315 255 1.498

Revision 0 14-I-47 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-48 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-49 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Revision 0 14-I-50 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

Attachment 6
Example of Document Control Log Sheets

Revision 0 14-I-51 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

THIS
SUPPORTING TECHNICAL INFORMATION
IS THE PROPERTY OF

LITTLE POWER COMPANY

111 MAIN STREET

ANYTOWN, USA

(If this STI is lost,

finder please return to
the above address)

ISSUED TO:

___________________________________

___________________________________

STI NO. ______________________

The person or organization to whom this manual

is issued, is responsible for its safekeeping and
its being kept up to date.

Revision 0 14-I-52 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

LITTLE POWER COMPANY

Big Power Project: FERC No. XYZ

Supporting Technical Information Change No. ____

Section Instructions Summary of Changes

Table of Contents Remove previously issued Updates revision numbers
Table of Contents (6/18/2002) and effective dates
and replace with
Table of Contents dated
1/22/2003 rev. 1
Section 1 Insert Addenda 1, Addends Failure Modes
Analysis Report to include
Failure Modes Analysis Failure Mode 7
new Failure Mode 7
dated 1/2/2003

Revision 0 14-I-53 April 11, 2003

Chapter 14 Monitoring the Performance of Dams

SUPPORTING TECHNICAL INFORMATION

REVISION LOG NOTICE
BIG POWER PROJECT

Revised Item Date Entered Initials Date Read Initials

Revision 0 14-I-54 April 11, 2003