CONCRETE FACE ROCKFILL DAMS NEW CHALENGES FOR

MONITORING AND ANALYSIS

Anna Szostak-Chrzanowski
1
, Michel Massira
2
, Nianwu Deng
1,3

Canadian Centre for Geodetic Engineering, University of New Brunswick, Canada
1

Facult dingnierie (gnie civil), Universit de Moncton, Canada
2
State Key Laboratory of Water Resources and Hydropower Engineering Science,
Wuhan University, P.R. China
3



ABSTRACT

Concrete face rockfill dams (CFRDs) are gaining a worldwide recognition as the most
economical type of dams to be constructed in extreme northern and sub-Antarctic
regions. Safety of CFRDs depends on the proper design, construction, and monitoring of
actual behaviour during the construction and during the operation of the structure. The
main concern for the safety of CFRDs is the deformation of the concrete face. During
the reservoir filling, the load of water and deformations of the dam rockfill force
upstream concrete slab to deform. The displacements of the concrete face during the
reservoir filling should not exceed the maximum allowed values in order to maintain the
structural integrity of the concrete face. Due to uncertainty of the model parameters,
careful monitoring of the dams and their surroundings are required in order to verify
and enhance the model. This paper presents a study of the behavior of the upstream
concrete face slabs and internal displacements that develop in the rockfill during
construction and during filling the reservoir. The research is based on two examples: the
Toulnustouc Dam (75 m) located in Northern Quebec, Canada, and the Shuibuya Dam
in P. R. China, the tallest (233 m) concrete face rockfill dam in the world. The results of
the study serve also as a basis for designing deformation monitoring schemes for the
concrete face rockfill dams.

Key words: concrete face rockfill dam, deformation analysis, finite element method,
monitoring.

1. INTRODUCTION

Concrete face rockfill dams (CFRDs) are gaining a worldwide recognition as the most
economical type of dams to be constructed in extreme northern and sub-Antarctic
regions. Use of the rockfill material, which is not sensitive to the frost action and the
construction technology allow lengthening the construction season. The total duration of
the construction of CFRDs with regard to the total duration of construction of earth
dams is on average reduced by one year. The reduced construction time reduces the
costs of construction and makes hydroelectric projects more economic. Cooke (1984)
indicated that the use of this type of dams seems to be inevitable in the regions of the
world, which have extreme climates. Canada has built CFRDs of a height up to 100 m.
382

China has built more than 170 CFRDs with 40 of them taller than 100 m. Examples are
Shuibuya dam of 233 m height, Jiangpinghe of 221 m, Sanbanxi of 186 m, Hongjiadu of
179.5 m, Tianshengqiao I of 178 m, and Tankeng of 162 m height.
Deformations of CFRD dams start occurring during the construction. These
deformations are caused by the increase of effective stresses during the construction by
the consecutive layers of earth material and also by effects of creep of the material.
Deformations may also be influenced by deformations of the foundation, by transfer of
stresses between the various zones of the dam and by other factors. After the
construction is completed, the considerable movements of the crest and of the body of
the dam can develop due to pressure of water during the first filling of the reservoir.
The load of water and deformations of the rockfill of the dam, force concrete slab to
deform. The concrete slab acts as an impervious membrane and any development of
cracks in the slab would allow for water to penetrate the rockfill of the dam and cause
the structure to weaken or even loose stability. According to the working state, force
distribution and hydraulic features of CFRD, proper zoning of dam filling material is
carried out to take a full utilization of the material from structure excavation and to
reduce the investment under the condition that the safety of operation is ensured. After
filling of the reservoir, the rate of movement in the dam and in the concrete face
generally diminishes with time, except for variations associated with periodic raising
and lowering the level of the reservoir. In classic CFRDs where the concrete face is
constructed after the end of construction of the rockfill embankment, it is very
important to estimate the displacements of the concrete face during the filling of the
reservoir and to verify whether these displacements are lower than displacements
compatible with the structural integrity of the concrete face. Most of the constructed
CFRDs rest on the bedrock. However, there are some CFRDs constructed on soil
foundations. Foundation conditions of the planned constructions call for studies to
determine the range of possible movements of the concrete face slab during the
construction of the dam and especially, during the filling of the reservoir.

The main concern for the safety of CFRDs is the deformation of the concrete face. The
most important reason for observing the deformation of the dams is to assure that the
deformation of the concrete face dam does not exceed the critical value. Too large or
unexpected deformations can be an indication of potential problems of the dam or its
foundation. Another reason for observing the deformations of dams, of less immediate
concern but of potentially great long-range significance to engineering profession, is the
need for better understanding of basic design concepts, stress-deformation
characteristics, and geotechnical characteristics of soil and rockfill. The development of
prediction methods, which allow for a determination of expected deformations and
stress distribution and a comparison of predicted values with observed deformations,
constitute very valid tools for control safety control.

The study presented in this paper shows that fusion of the monitoring results and FEM
analysis in real time would give warning that the deformations of the concrete face have
reached critical values before the reservoir would reach the maximum level. This
information could trigger the proper action from engineering team and lead to a
prevention of cracking of the concrete face. The in real-time detection and evaluation of
differences between monitoring results and results obtained from a prediction analysis
could result in a redesign of the concrete slab and prevent its cracking. Examples of two
CFRDs, Toulnustouc Dam in Quebec, Canada, and the Shuibuya Dam in P. R. China,
are discussed below.
383

2. DESIGN OF TOULNUSTOUC DAM

The Toulnustouc dam is located north of the city of Baie-Comeau on the Toulnustouc
River in Northern Quebec, Canada. The existing dam is 75 m high and 575 m long
(Figure 1). The dam is built on bedrock foundation. The concrete face slab is of a
constant 0.3 m thickness (ICOLD, 2006).





Figure 1: CFRD Toulnustouc dam in Canada.

The dam is composed of 4 zones (Figure 2). Zones 7B and 8A consist of crushed stone.
Zones 8B and 8C are built with compacted rockfill. Geotechnical instrumentation
installed in the structure included 16 survey markers, 22 crack meters, 13 submersible
tilt meters, two strong motion accelerometers and one weir (Hammamji et al. 2005).

Concrete Face Slab
Zone 7B
Zone 8A
Zone 8B
Zone 8C


Figure 2: CFRD Toulnustouc of 75 m of height resting on bed rock.


3. DESIGN OF SHUIBUYA DAM

Shuibuya CFRD is located in P.R. China. It is presently the highest of its kind in the
world. Shuibuya dam is 233 m high and 608 m long (Figure 3). The crest width is 12 m
384

with a 5.2 m high "L" shape parapet at the crest. The concrete slab thickness varies with
the height of the dam. The slab is 0.3 m thick at the top and 1.1 m thick at the bottom.
The slab is divided into three construction stages with two construction joints. The dam
zoning is shown in Figure 4. Geotechnical monitoring with wire extensometers to
monitor interior displacements of points located in the rockfill was started at the
beginning of the filling up of the reservoir.



Figure 3: CFRD Shuibuya dam.

The dam construction encountered certain difficulties because cracks started developing
in the concrete face during filling up the reservoir. Total of 255 cracks developed and
they were observed before the water level in the reservoir reached 200 m.
(file://E:\LecturesCM\Shuibuya Water Conservancy Project.htm). The development of
the cracks of the concrete slab indicated that the actual deformations of the slab were
larger than predicted and exceeded the maximum allowed deformations. The designed
deformation of the slab of Shuibuya dam was evaluated as being too small in relation to
its 233 m height and 403 m length of the concrete slab.
(http://www.biztrans.cn/CN/Samplefiles/SbyKeytech.htm) The deformation of the
concrete slab was following the deformation field of rockfill of the dam. Most likely one
of the reasons for the larger than expected deformations of the rockfill was that the in-
situ values of geotechnical parameters were different from those assumed during the
design of the dam.


4. MONITORING INSTRUMENTATION

The purpose of installing geotechnical or geodetic instrumentation is to observe the
behavior of dams. Type, number, and distribution of instrumentation depend on
characteristics of the site of a dam (narrow valley with steep banks, rough variation of
the geometry of foundations, more or less permeable deposits in the riverbed or on the
abutments, etc.). The number and the distribution of monitoring instruments depend on
specific problems foreseen in the design stage.
385


To have an overview of the deformations which occur in the body of a dam it is
necessary to have, at least, a distribution of measuring instruments in two main cross-
sections. The cross-section, where a plane strain state may be assumed, generally,
coincides with the transverse section where the dam is the highest. The second cross-
section is the plane corresponding to the axis of the dam. Monitoring data allows to
make comparisons with bi-dimensional analyses and to verify the behavior of the dam.

ASCE Task Committee (2000) presented in detail the various measuring instruments
used to evaluate the behavior of dams. CFRDs contain typically two layers of crushed
stone of transition of at least 3 m of width at once under the concrete face slab. These
layers of transition are generally in crushed stone 0-80 mm for the first layer, and in
crushed stone 0-200 mm for the second layer of transition. It is possible to install, in the
first layer, inclinometers with telescopic joints parallel to the concrete face slab.
Furthermore in the surface of the concrete face slab, movements in three directions can
be measured punctually with submersible tilt meters connected with a reading post.

In the CFRDs, there are crack meters across the vertical joints in the slab, close to the
abutments, to measure any opening of the construction joints. In most of the CFRDs,
they are installed at the top of the dam to measure the movements of the joint in the base
of the rail. It is also recommended to use different types of instruments for the same type
of deformation. For example, settlements can be measured with tilt meters and using
inclinometer with telescopic joints.

The geodetic monitoring should be combined with geotechnical monitoring. There are
new fully automated techniques for monitoring structural stability (Duffy et al. 2001) of
the dams. The techniques combine the use of Robotic Total Stations (RTS) and Global
Positioning System (GPS). The monitoring systems are supported by ALERT DDS
software (Szostak-Chrzanowski et al. 2007) developed at the Canadian Center for
Geodetic Engineering. ALERT DDS provides fully automated data collection, data
processing, and graphical presentation of displacements. The fully automatic systems
may also include geotechnical instrumentation. In the case of CFRDs, the monitoring
scheme must be capable of monitoring the concrete face slab, which after completing the
reservoir filling will be submerged under water. One should note that the maximum
displacements are expected to be below the crest of dams. Therefore new monitoring
techniques will have to be developed to detect the expected displacements at various
elevations of the submerged concrete face slab.


5. ANALYSIS OF DEFORMATIONS OF TOULNUSTUC AND SHUIBUYA DAMS

The FEM analyses of the behaviour of two dams were performed using first the design
geotechnical parameters, followed by the analyses with geotechnical parameters verified
(calibrated) by monitoring results. The analyses of the behaviour of the Toulnustouc and
Shuibuya CFRDs were performed for two stages: end of construction and during filling
the reservoir. The finite element analysis (FEM) was performed using GeoStudio
software (Krahn, 2004) with hyperbolic model of the behaviour of the rockfill material
(Duncan and Chang, 1970).

386


Figure 4: Shuibuya Dam, FEM mesh and dam zoning.

The finite element model of the Toulnustouc dam is shown in Figure 2. The geotechnical
parameters used in the analysis were accepted as given by Massira et al. (2005). The
finite element model of Shuibuya dam is shown in figure 4. The FEM analysis was using
values of geotechnical parameters given by Szostak-Chrzanowski et al. (2008).

The calculated vertical displacements (settlements) for the Toulnustouc dam and
Shuibuya dam at the end of construction stage are shown in Figures 5 and 6
respectively. The vertical displacements at the end of the construction are much larger
in case of Shibuya dam. They are equal to -0.8 m for Toulnustouc dam and -2.2 m for
Shibuya dam.

The calculated vertical displacements (settlements) for the Toulnustouc dam and
Shuibuya dam at the end of filling the reservoir are shown in Figures 7 and 8
respectively. The calculated for Toulnustouc dam maximum vertical displacement of -
0.20 m is at the middle height of the upstream face. The calculated for Shuibuya dam
maximum vertical displacement is on the upstream face of the dam and is -0.30 m and is
at one third of the height.



Figure 5: Calculated settlements (m) at the end of construction for Toulnustouc CFRD.


-0.8
-0.6
-0.4
-0
.2

387



Figure 6: Calculated settlements (m) at the end of construction for Shuibuya CFRD.



Figure 7: Calculated settlements (m) at the end of the filling of the reservoir
for Toulnustouc CFRD.



Figure 8: Calculated settlements (m) at the end of the filling of the reservoir
for Shibuya CFRD.

The calculated total displacements for the Toulnustouc dam and Shuibuya dam at the
end of filling the reservoir for different heights of water in reservoir are shown in
Figures 9 and 10 respectively. The calculated for Toulnustouc dam maximum total
displacement of -0.23 m is at the middle height of the upstream face.
-0
.2
8

-0
.2
2
-0
.1
6
-0.12
-0.08
-0.06
-0.04
-0.02
-
0
.2



-
0
.
1
6




-
0
.
1
2




-
0
.
0
8




-
0
.
0
4


388


The calculated for Shuibuya dam maximum total displacement is on the upstream face
of the dam and is -1.18 m and is at one third of the height. Figure 10 shows the total
calculated displacement for the maximum water level using designed geotechnical
parameters. It shows that when the level of water in the reservoir reached 150 m, the
maximum modelled displacement of the dam with verified geotechnical parameters was
larger than the maximum predicted displacement.




Figure 9: Calculated total displacements (m) of the concrete face slab during
the filling of the reservoir of the CFRD.



wat er l evel s ( m)
0
50
100
150
200
250
0. 00 0. 20 0. 40 0. 60 0. 80 1. 00 1. 20 1. 40
Total displacement (mm)
E
l
e
v
a
t
i
o
n

(
m
)
0
25
50
75
100
125
150
175
200
215
215( desi gn
par amet er )



Figure 10: Calculated total displacements (m) of the Shuibuya concrete face
slab during the filling of the reservoir.
389

5. CONCLUSIONS

The study shows that the in real time fusion of the monitoring results and FEM analysis
would generate information that the deformations of the concrete face reached critical
values before the reservoir level reached the maximum level. The fusion process of the
monitoring results and FEM analysis should work in real time using data sets for
measurements results and FEM results interconnected using specific correlation
criteria. The result of real time fusion would trigger the proper action from engineering
team what would lead to prevention of the cracking of the concrete face.

In case of CFRD dams, the maximum displacements are expected to occur on upstream
face of the dam, where classical geodetic surveys cannot be implemented. Thus, in this
case permanently installed geotechnical instruments (e.g., inclinometers, extensometers)
should be used on the upstream face while geodetic surveys could be utilized on the crest
and on the downstream face. Here one should note that the maximum displacements of
the upstream (concrete) face are expected to take place below the crest. This is very
often overlooked by geodetic engineers, who tend to install their points along the crest
due to the easiest access.The maximum displacements of the concrete face slab are also a
function of the height of the dam during the filling of the reservoir.

REFERENCES

ASCE Task Committee, 2000. Guidelines for instrumentation and measurements for
monitoring dam performance. ASCE Task Committee on Instrumentation and
Dam Performance.
Cooke, J.B. 1984. Progress in rockfill dams: Journal of Geotechnical Engineering,
ASCE, 110(10): 1383-1414.
Duffy, M., Hill, C., Whitaker, C., Chrzanowski, A., Lutes, J. and Bastin, G. 2001. An
automated and integrated monitoring program for Diamond Valley Lake in
California. Proceedings of the 10
th
FIG International Symposium on Deformation
Measurements, Orange, California, pp. K-1 to K-23, (available at:
http://ccge.unb.ca).
Duncan, J.M., Chang, C.-Y., 1970. Nonlinear analysis of stress and strain in soils:
Journal of the Soil Mechanics and Foundation Division, ASCE, 96 (SM5): 1629-
1653.
Hammamji, Y., Beausjour, N., Massira, M. Vautour, J., Landry, L.-M. 2005.
Toulnustouc CFRD main dam : stress-deformation predictions and behaviour
during reservoir filling. Proceedings, CDA 2005 Annual Conference, Calgary, 3-6
October, CD Rom, Technical Session 5, 12 p.
ICOLD, 2006. Concrete face rockfill dams. Concepts for design and construction.
ICOLD Committee on Materials for Fill Dams. Bulletin No. 141 (rough).
Krahn, J. 2004. Stress and deformation modeling with SIGMA/W, an engineering
methodology. GEO-SLOPE International Ltd., Calgary, Alberta.
Massira, M., Szostak-Chrzanowski, A., Vautour, J. and Hammamji, Y. (2005),
Deformations of Concrete Face Rockfill Dams (CFRDs) Resting on Soil
Foundation, Technical Sciences Journal, No. 8, UWM, Olsztyn, pp. 65-78.
Szostak-Chrzanowski A., Nianwu Deng, and Massira, M. 2008. Monitoring and
Deformation Aspects of Large Concrete Face Dams, Proceedings, 4
th
IAG
Symposium on Geodesy for Geotechnical and Structural Engineering and 13-th
390

FIG Symposium on Deformation Measurements, Lisbon, Portugal, 12-15 May,
CD-Rom.
Szostak-Chrzanowski A., Chrzanowski, A., Massira, M., Bazanowski, M., Whitaker,
C., 2007. Monitoring and Analysis of Long-term Behaviour of Large Earth Dams,
Proceedings, The 100th CIG Annual Conference, The 3rd International
Symposium on Geo-information for Disaster Management, Toronto, Ontario,
Canada, May 23-25, CD- Rom.
Key Technical Issues of Shuibuya CFRD
http://www.biztrans.cn/CN/Samplefiles/SbyKeytech.htm