Contents

PREFACE xiii

PART ONE CYCLE THE HYDROLOGIC 1 CHAPTER lntroduction

1.1 t.2

r.3
t.4 1.5 1.6 I.7 1.8

3 Hydrology Defined 3 A Brief History 5 The Hydrologic Cycle 5 The Hydrologic Budget 11 HydrologicModels 11 HydrologicData 12 Common Units of Measurement Problems to Environmental Application of Hydrology

t2

2 CHAPTER Precipitation 1 5
2.1 2.2 2.3 2,4 2.5 2.6 2.7
15 Water Vapor 17 Precipitation Distribution of the Precipitation Input 27 Point Precipitation 29 Areal Precipitation 34 Precipitation ProbableMaximum 36 Grossand Net PreciPitation

vi

coNTENTS

CHAPTER 3 Interception and Depression Storage

3.1 3.2 3.3 Interception 40 Throughfall 44 Depression Storage 45

40

CHAPTER 4 Infiltration 52
4.I 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 Measuring Infiltration 53 Calculation of Infiltration 53 Horton's Infiltration Model 57 Green-AMPT Model 64 Huggins-MonkeModel 67 Holtan Model 68 Recoveryof Infiltration Capacity 69 Temporal and Spatial Variability of Infiltration Capacity SCS Runoff Curve Number Procedure 73 76 @Index

70

CHAPTER 5 Evaporation and Transportation 82 "

5.1 5.2 5.3 5.4 5.5 5.6 5.7 Evaporation 86 EstimatingEvaporation 86 EvaporationControl 95 Transpiration 95 TranspirationControl 100 Evapotranspiration 100 EstimatingEvapotranspiration

103

CHAPTER 6 Streamflow 111

6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 DrainageBasinEffects 111 The Hydrograph 11,2 Units of Measurement for Streamflow 113 Measuringand RecordingStreamflow 113 Measurements of Depth and Cross-Sectional Area II4 Measurement of Velocity lI4 RelatingPoint Velocity to Cross-Sectional Flow Velocity The Slope-AreaMethod for DeterminingDischarge II7

115

oONTENTS Vii

PART TWO AND MONITORING 121 fT'IEASUREMENTS HYDROLOGIG 7 CHAPTER DataSources Hydrologic
7.1 7.2 7.3 7.4

123

I23 GeneralClimatologicalData 123 Precipitation Data 124 StreamflowData Evaporationand TranspirationData

I24

8 CHAPTER 126 fnstrumentation

8.1 8.2 8.3 8.4 126 Introduction 127 HYdrologicInstruments 135 TelemetrySYstems 135 RemoteSensing

9 CHAPTER Networks 144 Monitoring The Purposeof Monitoring 9.r

9.2 9.3 9.4

144

I45 SpecialConsiderations I47 in Monitoring Uie of ComPuters 147 Networks Hydrological-Meteorlogical

PART THREE WATERHYDROLOGY151 SURFACE 1O CHAPTER andthe Catchnient Runoff

10.1 tO.2 10.3 IO.4 10.5

153

153 and DrainageBasins Watersheds, Catchments, 155 Affecting Runoff Basin Characteristics RudimentaryPrecipitation-RunoffRelationships 164 166 StreamflowFrequencyAnalysis 168 Forecasting StreamflQw

11 CHAPTER Hydrographs 171

11.1 Il.2 11.3 HYdrograPhs 171 Streamflow FactorsAffecting HydrographShape ComPonents 174 HydrograPh 172

viii

ooNTENTS lI.4 11.5 11.6 Il.7 BaseFlow Separation I77 HydrographTime Relationships Time of Concentration I82 BasinLae Time I82 181

CHAPTER 12 UnitHydrographs 188

l2.I 12.2 12.3 12.4 12.5 12.6 Unit Hydrograph Definition 188 Derivation of Unit Hydrographsfrom StreamflowData 190 Unit HydrographApplications by Lagging Methods I94 S-Hydrograph Method 198 The Instantaneous Unit Hydrograph 201 SyntheticUnit Hydrographs 205

CHAPTER 13 Hydrograph Routing

13.1 13.2 13.3

234

HydrologicRiver Routing 235 HydrologicReservoirRouting 245 Hydraulic River Routing 248

CHAPTER 14 SnowHydrology
I4.l I4.2 I4.3 I4.4 14.5 14.6

265

Introduction 265 Snow Accumulation and Runoff 267 Snow Measurements and Surveys 268 Point and Areal Snow Characteristics 269 The SnowmeltProcess 271. SnowmeltRunoff Determinations 284

CHAPTER 15 Urbanand SmallWatershed Hydrology 309

15.1 15.2 i5.3 15.4 Introduction 309 PeakFlow Formulasfor Urban Watersheds 311 PeakFlow Formulasfor Small Rural Watersheds 33I Runoff Effects of Urbanization 344

CHAPTER 16 Hydrologic Design

16.l 16.2

359

Hydrologic DesignProcedures 360 Data for HydrologicDesign 363

CONTENTS

IX

16.3 16.4 16.5 t6.6 16.7 16.8

HydrologicDesign-Frequency Criteria DesignStorms 373 Critical EventMethods 391 Airport DrainageDesign 400 Designof Urban Storm Drain Systems FloodplainAnalysis 409

365

402

PART FOUR GROUNDWATER HYDROLOGY 425 CHAPTER 17 Groundwater, Soils,and Geology

l7.l I7.2 I7.3 I7.4 I7.5 I7.6

427

Introduction 427 Groundwater Flow-General Properties 429 Subsurface Distribution of Water 429 GeologicConsiderations 430 Fluctuationsin GroundwaterLevel 433 Groundwater-Surface Water Relations 433

CHAPTER 18 Mechanics of Flow

18.1 t8.2 18.3 18.4 18.5 18.6

435

Hydrostatics 435 GroundwaterFlow 436 Darcy's Law 436 Permeability 438 Velocity Potential 440 HydrodynamicEquations 441 r8.7 ' Flowlines and EquipotentialLines 18.8 BoundaryConditions 447 18.9 Flow Nets 449 1 8 . 1 0 VariableHydraulic Conductivity 1 8 . 1 1 Anisotropy 452 18.t2 Dupuit's Theory 453

444

451

CHAPTER 19 Wellsand Collection Devices

19.1 19.2 19.3 19.4 19.5

460

Flow to Wells 460 SteadyUnconfinedRadial Flow Toward a Well 461 SteadyConfined Radial Flow Toward a Well 462 Well in A Uniform Flow Field 463 Well Fields 465

CONTENTS

19.6 I9.7 19.8 I9.9 19.10 19.ll 19.12

The Method of Images 466 UnsteadyFlow 467 Leaky Aquifers 4'13 Partially PenetratingWells 473 Flow to an Infiltration Gallery 473 Saltwater Intrusion 474 GroundwaterBasin Development 475

CHAPTER 20 Modeling Regional Groundwater Systems

20.I 20.2 20.3 20.4 20.5 RegionalGroundwaterModels 481 Finite-DifferenceMethods 484 Finite-ElementMethods 493 Model Applications 494 GroundwaterQuality Models 500

481

PART FIVE HYDROLOGIC MODELING

505 5O7

CHAPTER 21 Introduction to Hydrologic Modeling

2l.I 2t.2 21.3 21.4

HydrologicSimulation 508 Groundwater Simulation 509 Hydrologic Simulation Protocol 524 Corps of EngineersSimulation Models

526

CHAPTER 22 Synthetic Streamflows 535

22.I 22.2 SyntheticHydrology 536 Serially DependentTime SeriesAnalysis 539

CHAPTER 23 Continuous Simulation Models

23.1 23.2

548

Continuous Streamflow SimulationModels 549 ContinuousSimulation Model Studies 570

CHAPTER 24 Single-Event Simulation Models

24.1 24.2 24.3

594
597

StormEventSimulation 594 Models FederalAgency Single-Event Storm SurgeModeling 625

CONTENTS Xi

CHAPTER 25 Urban Runotf Models Simulation

25.1 25.2 25.3

630

Urban StormwaterSystemModels 63I Urban Runoff Models Compared 659 Vendor-DevelopedUrbanStormwaterSoftware 663

PART SIX METHODS STATISTICAL

669 671

26 CHAPTER Probability and Statistics

26.1 26.2 26.3 26.4 26.5 26.6 26.7 26.8 26.9

2 6 .r0

RandomVariablesand StatisticalAnalysis 672 Concepts of Probability 673 ProbabilityDistributions 676 Moments of Distributions 681 Distribution Characteristics 682 Types of Probability Distribution Functions 685 ContinuousProbabilityDistributionFunctions 685 Bivariate Linear Regression and Correlation 690 Fitting Regression Equations 692 Regression 697 and Correlation Applications

CHAPTER 27 Frequency Analysis

27.1 27.2 27.3 27.4 27.5 27.6 27.7

708

FrequencyAnalysis 708 GraphicalFrequencyAnalysis 709 FrequencyAnalysis Using FrequencyFactors RegionalFrequencyAnalysis 7I9 Reliability of FrequencyStudies 730 FrequencyAnalysis of Partial Duration Series Flow Duration Analysis 737

7Il

734

APPENDICES 751 INDEX 757

Preface

New federalthrusts,the growis taking on new dimensions. Watermanagement environmentalprotecregarding public sentiment ing list of global iisues,and strong forces. driving tion havebeen the principal and managedevelopment In the early yearsof the 20th century,waterresources Today,these control' flood and supply water on ment were focuied almostexclusivd drinking safe ensuring environment, the protecting but issuesare still important, -and for equally compete experiences recriatioinal water, and providing aesthetic for pressing public is conscious environmentally an attentionand funds.Furthermore, compostructural fewer with practices, management greaterreliance on improved to ients, to solve this nuiion'. water problems.The notion of continually striving natural precious this husbanding of one by replaced provide more water has been resource. There is a growing constituencyfor allocatingwater for the-benefitof fish and wildlife, for protection-of marshesand estuary areas' and for other natural system uses.But estimatingthe quantitiesof water neededfor environmentalprotectionand is difficult, and there are still many for maintainingand/or restoringnatural systems of the complexinteracunderstanding our and ,putt", are unknowns.Scilntific data this is a critical issue' Indeed, rudimentary. is scales of an tions inherentin ecosystems can be substantial protection in environmental involved sincethe quantitiesof water The nationsof keen' is users water traditional from and competitionfor thesewaters that are systems-decisions natural regarding decisions the world are facing major associurgency is_an there Thus impacts. social and ladenwith significantectnomic hydrologic of their and systems of ecologic a betterunderstanding atedwith developing components. Water policies of the future must thereforetake on broader dimensions'More and regionalinstitumustbe placedon regionalplanningand management, emphasis be practicedat, must management Water devised. be this muJt tions to accommodate planning mustbe more use water and use Land government. andbetween,all levelsof tightly coordinatedas well'

XIV

PREFACE

Water scientistsand engineersof tomorrow must be equipped to addressa diversity of issuessuch as: the design and operation of data retrieval and storage systems; forecasting;developingalternativewater use futures; estimatingwater requirementsfor natural systems; exploringthe impactsof climate change;developing more efficient systems for applyingwaterin all water-usingsectors; and analyzingand designingwater managementsystemsincorporating technical, economic, environmental, social, legal, and political elements. A knowledgeof hydrologicprinciplesis a requisitefor dealingwith suchibsues. This fourth edition hasbeendesigned to meet the contemporaryneedsof water scientistsand engineers. It is organizedto accommodate studentsand practitioners who are concernedwith the development,management,and protection of water resources. The format of the book follows that of its predecessor, providing material for both an introductory and a more advanced course. Parts One through Four provide the basicsfor a beginninglevel course,while Parts Five and Six may be used for a more advancedcourseon hydrologicmodeling. This fourth edition has been updated throughout, and many solved examples havebeen added.In addition, new computer approaches have been introduced and problem-solvingtechniques includethe use of spreadsheets as appropriate.New featuresof eachchapterincludean introductory statement of contentsand,at the conclusion of the chapter,a summaryof key points. Many sourceshave been drawn upon to provide subject matter for this book, and the authorshope that suitable acknowledgment has been given to them. Colleagues and students are recognizedfor their helpful commentsand reviews,particularly the following reviewers. Gert Aron, ThePennsylvania StateUniversity JohnW. Bird, Universityof Nevada-Reno IstvanBogardi, Universityof Nebraska RonaldA. Chadderton, VillanovaUniversity RichardN. Downer,Universityof Vermont Bruce E. Larock, Universityof Califurnia-Davis Frank D. Masch,Universityof Texas-San Antonio Philip L. Thompson, FederalHighwayAdministration A specialnote of thanks is due to Dr. John W. Knapp, President of the Virginia Military Institute,coauthorof previouseditionsof this book, for his pastcontributions andvaluableguidance. WarrenViessman. Jr. Gary L. Lewis

PARTONE

CYCLE THE HYDROLOGIC

L.

Chapter1

lntroduction

I Prologue
The purPoseof this chaPteris to: . , . . Define hydrology. . earth -,.1Lscience' Give a brief niJiory of the evolution of this important Statethe fundamentalequationofhydrology' decision trow ffiofogic principle, "urib" appliedto supplement Demonstrate for water and environmentalmanagement' support systems

DEFINED 1.1 HYDROLOGY

Hydrologyisanearthscience'Itencompassestheoccuffence'distribution,moveA knowledgeof hydrologyis fundamenr, and propertiesof the watersof the earth. mentaltodecisionmutingp,o.",,e,*he,ewaterisu"ompon"nto.f.th.esystemof inextricably linked' and it is important concern.water and environmentalissuesare toclear$understandhowwaterisaffectedbyandhowwateraffectsecosystem maniPulations'

1.2 A BRIEFHISTORY
focusedtheir i Ancient philosophers flows Production of surfacewater oc"ur."n"e of water in variousstag to t from the seato the atmosPhere faulty'l early speculationwas often reservoirsth of large subterranean is interestingto note, however'tha
*suoeriornumbersindicatereferencesattheendofthechapter.

CHAPTER1

INTRODUCTION

Greek aqueducts on both conveyance crosssectionand velocity.This knowledgewas lost to the Romans,and the proper relation betweenarea,velocity, and rate of flow remainedunknownuntil Leonardoda Vinci rediscovered it duringihe Italian Renaissance. During the first century s.c. Marcus Vitruvius, in Volume 8 of his treatise De Architectura Libri Decem (the engineer'schief handbookduring the Middle Ages), setforth a theory generallyconsidered to be the predecessor of modernnotionsof the hydrologiccycle. He hypothesized that rain und ,no* falling in mountainousareas infiltrated the earth's surface and later appearedin the lowlands as streamsand springs. In spiteof the inaccurate theoriesproposed in ancienttimes,it is only fair to state that practical applicationof varioustry-orotogic principleswas often carried out with considerable success. For example,about4000 s.c. u du- was constructed acrossthe Nile to permit reclamation of previously barren lands for agricultural production. Severalthousandyears later a canal to convey fresh water from Cairo io Suezwas built. Mesopotamian towns were protecteduguinrt floodsby high earthenwalls. The Greek and Roman aqueducts and early Chineseirrigation and flood control works were also significantprojects. Nearthe endof the fifteenth century the trend towarda more scientific approach to hydrology based on the observationof hydrologic phenomena becameevident. Leonardo da Vinci and Bernard Palissyindepende-ntly reachedan accurateunderstandingof the watercycle.They apparentlybised theii theoriesmore on our"*iion than on purely philosophical reasoning. Nevertheless, until the seventeenth century it seemsevident that little if any effort was directed toward obtaining quantitative measurements of hydrologicvariables. The adventof what might be calledthe "modern" science of hydrologyis usually considered to beginwith the studiesof suchpioneersasPerrault,Mariotte, and Halley in the seventeenth century.r'a Perraultobtainedmeasurements of rainfall in the Seine River drainagebasin over a period of 3 years. Using these and measurements of runoff, and knowing,thedrainage areasize,he showeJthat rainfall was adequate in quantity to accountfor river flows. He also mademeasurements of evaporati,on and capillarity. Mariotte gaugedthe velocity of flow of the River Seine. Recordedvelocities were translatedinto termsof dischirgeby introducingmeasurements of the river crosssection'The English astronomer Halley measured the rate of evaporation of the Mediterranean Seaand concludedthat the amountof water evaporated was sufficient to accountfor the outflow of rivers tributary to the sea.Measurements suchas these, althoughcrude,permitted reliable conclusions to be drawn reggrdingthe hydrologic phenomena being studied. brth numerousadvances in hydraulic theory zometer,the Pitot tube, Bernoulli's theorem, ples.8 perimental hydrology flourished.Significant ydrology and in the measurement of surface water. Such significantcontributionsas Hagen-Poiseuille'scapillary flow equation, Darcy's law of flow in porous media, und th" Dupuit-Thiem well formula were evolved'e-lrThe beginningof systematicstream guoling can also be traced to this period' Although the basis for modern hydrology wui tirrr established in the nine,)

BUDGET 1.4 THE HYDROLOGIC

of teenth century, much of the effort was empirical in nature. The fundamentals early the In recognized. or widely established yet well been not physicalhydtotogyhad of many earlier empirical formulayears of tle twJntieth ""niury the inadequacies began to governmentalagencies interested result, As a known. tions becamewell rational 1950, to 1930 about From research' of hydrologic programs their own develop Horton's hydrograph, unit Sherman's empiricism.3 ieplace analysis began to to well hydraulicsare outinfiltration theory, und Th"it's nonequilibrium-approach made'r2-'o progress great the of standingexamples Since 1930 a theoreiical approachto hydrologicproblemshas largely replaced methods of ttre past. Advancesin scientific knowledgepermit a less sophisticated ' ofthe physicaibasisofhydrologic relations,and the adventand better understanding in both of high-speeddigital computershavemadepossible, continueddevelopnient would that manipulations mathematical extensive iense, a practical and an economic havebeen overwhelmingin the past. For a more compiehensivi historical treatment, the reader is referred to the and their co-workers'1'2'4'5'15 works of Meinzer,Jonls, Biswas,

CYCLE 1 . 3THE HYDROLOGIC

the The hydrologiccycle is a continuousprocessby which water is transportedfrom The exist' subcycles Many sea. the to back landind to the oceansto the atmosphere precipitation over land beforereturnevaporationof inlan-dwater and its subsequent for the global watertransportsystem force driving The example. ingio the oceanis one Note that requiredfor evaporation' energy the furnishes is providedby the sun,which sea example, for cycle; the passage through during changes the water quality also evaporation' through water fresh water is convertedto The completewater cycle is global in nature. world water problems require Practical studieson regional,national,internitional, continental,and global scales.16 earth is the to available water fresh of supply total the that significanceof the fact has limited and very small compared with ihe salt water content of the oceans at the receivedlittle attention.Thus watersflowing in one country cannotbe available u's' same time for use in other regions of the world. Raymond L' Nace of the are a global problem with resources thatoowater GeologicalSurvey has aptly sta=ted with problems requiring cope to obligated are hydrologists local roots."tu Mtdern In addition, developing difference. magnitude of oider of definition in varying scales since climatological attention, careful receive must weather techniquesto contiol therefore the water and hydrology the affect profoundly can changesin one area of other regions. resources

BUDGET 1 . 4 THE HYDROLOGIC

the total quantity of water availableto the earth is finite and indestructible, Because subsysthe global hydrolojic ,yrt"* may be lookedupon as closed.Open hydrologic system' For any and theseare usuallythe type analyzed' temsare abundantlhowever, ' components' hydrologic the for to account developed be can a water budget

CHAPTER1

INTRODUCTION

FiguresI'I,I.2, and 1.3 showa hydrologic budgetfor the coterminous United States, a conceptualized hydrologiccycle,andthe distributionof a precipitationinput, respectively. Thesefiguresillustrate the components of the water cycle with which a hydrologist is concerned. In a practicalsense, somehydrologicregionis dealtwith and a budgetfor that region is established. Suchregionsmay be topographicallydefined (watersheds and river basinsare examples), politically specified(e.g- couniy or city limits), or chosenon some other grounds. Watersheds or drainagi tasins are the easiest to deal with sincethey sharply define surfacewater boundaries. Thesetopographically determinedareasare drainedby a river/streamor systemof connecting rivers/streamssuch that all outflow is discharged through a single outlet. Unfortunately,it is often necessary to deal with regions that are not well suitedto tracking hydrologiccomponents. For theseareas, the hydrologist will find hydrologicbudgeting somewhatof a challenge. The primary input in a hydrologicbudgetis precipitation.Figures 1.1-1.3 illustrate this. Someof the precipitation (e.g.,rain, snow,hail) may be interceptedby trees,grass,other vegetation, and structuralobjectsand will eventuallyreturn to the atmosphere by evaporation. Onceprecipitationreaches the ground,someofit may fill (becomedepression depressions storage), part may penetralethe ground (infiltraie) to replenishsoil moisture and groundwaterreservoirs,and some may become surface runoff-that is, flow over the earth's surfaceto a definedchannelsuchas a stream. Figure 1'3 showsthe dispositionofinfiltration, depression storage, and surface runoff.

and vegetation

'1.'

*r-d{i;

Consumptive use 100bgd

bgd = billion gallons per day

Figure 1.1 Hydrologic budgetof cotermiriousunited States.(U.S. Geologicalsurvey.)

BUDGET 7 1.4 THEHYDROLOGIC

Clouds and water vaPor Clouds and water vaPor

" 1 ) t t l l ,
p\--

ffi

--1vtivv
P P P P P P

t t
E' evaporation;P' Figure 1.2 The hydrologic 'surfac-e cycle: ?, transpiration; flow; and I' groundwater runoff; G, R, p.Sqipi*i"tt inflltration.

t Precipitation inPut (hyetogaph)

'l,r\
t SSeamflow (hyclrograPh)

input' of precipitation Figure 1.3 Distribution

CHAPTER1

INTRODUCTION

water enteringthe ground may take severalpaths.Somemay be directly evaptransferfrom the soil to the surfaceis maintained.This can easily oratedif adequate groundwater table (free water surface) is within the limits of a high occur where usingsoil moistureor groundwaVegetation to the ground surface. transport capillary by a processknown atmosphere to the water infiltrated transmit can also tei directly deficiencies moisture soil replenish may likewise water transpiration.Infiltrated as dry turn maintain in which reservoirs, groundwater provided in enter storage and so that flowing usually groundwater are of bodies Important streamflow. weather inflltrated water reachingthe saturated,on" muy be transportedfor considerable' Groundwatermovement is subject, of course,to distancesbefore it is discharged. physicaland geologicalconstraints. will eventually evaporateor infiltrate the ground I Water storedin depressions minor channels(gullies, rivulets, and the : reaches otti-ut"ty runoff surface.Surface an ocean.Along the course finally reaches rivers, and and major streams like), flowsto occur. also can infiltration and evaporation of a stream, The foregoing discussion suggeststhat the hydrologic cycle, while simple in complex.Pathstaken by particlesof water precipiconcept,is actually exceedingly and varied before the seais reached.The time scale numerous are tated in any arca days,or years. minutes, seconds, order of may be on the basedon theprocessesillusdeveloped be can general equation hydrologic A hydrologicvariablesP, E, T, the it, In Fig. 1.4. 1.3. Consider 1.2 and trated in Figs. g are introduced to denote and s Subscripts in Fig. 1.2. defined l are as R, G, and R, For example, respectivd. surface, the earth's below and above vectorsoriginating

["
Earth's surface Surface channels

R2

Level of plastic rock . (no water below this level)

Figure 1.4 Regionalhydrologiccycle.

BUDGET 1.4 THE HYDROLOGIC

signifies groundwaterflow that is effluent to a surface streamoand E, represents areas.Letter S stands evaporationfrom surfacewaterbodiesor other surfacestorage lower boundarybelow A a as has specified consideration region under The for storage. surface. Vertical earth's is the boundary found. The upper not be which water will Remembering periphery the region. of projections of the as set boundsare arbitrarily in,storage, that the water budgetis a balancebetweeninflows, outflows,and changes all values where statements, mathematical into the following be translated Fig. 1.4can per unit time: given of volume in units are 1. Hydrologicbudgetabovethe surface P+R1 -RrIRr-E"-7,2. Hydrologicbudgetbelow the surface I + Gt- G2- Rr- E, - 4: AS, (1.2)

1:AS"

(1.1)

3. Hydrologicbudgetfor the region (sum of Eqs' 1.2 and 1.3) p - (Rr- R,) - (E" + E) - (r" + Tr) - (Gr- G,) : a(S, + ss), (1.3) refer If the subscripts are droppedfrom Eq. 1.3 sothat letterswithout subscripts flow, evaporation, flow, underground surface values of to total precipitation and net the hydrologicbudgetfor a regioncan be written simply as transpiration,and storage; p_R-G_E_T:LS (1.4)

whereterms For a simplifiedhydrologicsystem This is the basicequationof hydrology. G, E, and Z do not apply, Eq. 1.4 reducesto p-R:AS (1.5)

of any degreeof complexity and is therefore Equation 1.4 is applicableto exercises problems. all hydrologic basic to the solution of The difficulty in solvingpractical problemslies mainly in the inability to measure or estimateproperly the various hydrologic equation terms. For local studies, often are made,but on a global scaleqqantificationis usuallycrude. reliableestimates locatedthroughoutan area.Surface by rain or snowgauges Precipitationis measured such as weirs, flumes,velocity meters, devices various using flows can be measured the area.Under goodconditions of and streams gauges in rivers located the and depth but large floods cannot be percent accufate, or more these measurementsare 95 Soil are sorelyneeded. events such data on and methods directly by current measured infiltration gravimetric methods; probes and using neutron moisturecan be measured can be deterrnined locally by infiltrometers or estimated through the use of of soil moistureand infiltration are generprecipitation-runoff data. Areal estimates areusually rate of movementof groundwater and The extent illy very crude,however. groundwater are quantities of on data adequate determine, and difficult to exceedingly groundwater for essential is geology aregion of of the Knowledge not alwaysavailable. estimatesif they are to be more than just rough guides.The determinationof the

1O

CHAPTER1

INTRODUCTION

quantities of water evaporatedand transpired is also extremely difflcult under the are of evapotranspiration Most estimates the science. presentstateof developmentiof empirmethods, or mass transfer pans,energybudgets, obtainedby usingevaporation ical relations.A predicamentinherent in the analysisof large drainagebasinsis the fact that rates of evaporation,transpiration,and groundwatermovementare often to be highly heterogeneous. assumed that it can The hydrologicequationis a usefultool; the readershouldunderstand of distribution time and magnitude the be employed in various ways to estimate found will be others given here, and is hydrologicvariables.An introductory example throushout the book. EXAMPLE 1.1 The average 20 in. of precipitation. wabrshedreceived In a given year,a 10,000-mi2 (cubicfeet cfs 700 found to be was area the river draining measured in the rate of flow and of the combinedamountsof waterevaporated per second). Make a rough estimate year record. of during the the region transpiredfrom Solution. Beginningwith the basic hydrologicequation P - R - G - E - Z : A S can be combined, and transpiration and sinceevaporation ET:P-ft-G-AS (1.6)

(r.4)

{4
t:

.+ v' a'E-;.

The term EZ is the unknown to be evaluatedand P and R are specifled.The equation thus has flve variables and three unknowns and cannot tre solved without additional information. In order to get a solution, two assumptionsare made. First, since the in hundredsof squaremiles),a presumpdrainageareais quite large (measured divide is probadivide (boundary)follows the surface tion that the groundwater In this casethe G componentmay be consideredzero. The bly reasonable. vector R, existsbut is included in R. The foregoingassumptionis usually not valid forsmall areasand mustthereforebe usedcarefully.It is alsopresupposed that AS : 0, thus implying that the groundwaterreservoir volume has not changedduring the year. For such short periods this assumptioncan be very inaccurate,evenfor well-wateredregionswith balancedwithdrawalsand good is beingmined (AS consispotentials.In arid areaswheregroundwater recharge Neverin many cases. supposition it would be an unreasonable tently negative), qualified by is made here for illustrative purposesand the assumption theless, approximate sayingthat pastrecordsof waterlevelsin the areahaverevealedan Hydrology is not an exact science,and reaconstancyin groundwaterstorage. are required if practical problemsare to be sonablewell-foundedassumptions solved. Using the simplificationsjust outlined, the working relation reducesto ET:P_R

'I1 1.6 HYDROLOGIC DATA which canbe solveddirectly.First, change R into inchesper yearsothat the units are compatible:
_ ft3
r \ t

1
. ,

sec

sec

area (m n-l

yt

II

^ :R,in.

R _

7 0 0 x 8 6 ; 4 0 0 x 3 6 5 x 1: 2
104x (5280)'

0.95in.

Therefore, ET : 20 - 0.95 : 19.05in./yr. The amountof evapotranspiration for the year in questionis estimatedto be 19.05in. This is admittedlya crudeapproximation but could serveasa useful guide for water resources planning. ll

1.5 HYDROLOGIC MODELS

Hydrologic systemsare generally analyzedby using mathematicalmodels.'These modelsmay be empirical, statistical,or foundedon known physicallaws.They may be usedfor suchsimplepurposes as determiningthe rate of flow that a roadwaygrate mustbe designed to handle,or they may guidedecisions aboutthe bestway to develop a river basinfor a rnultiplicity of objectives. The choiceof the modelshouldbe tailored to the purposefor which it is to be used.In general,'thesimplestmodel capableof producinginformation adequate to deal with the issueshouldbe chosen. Unfortunately,most waterresources systems of practicalconcernhavephysical, social,political, environmental, andtheir interactions cannotbe andlegaldirnensions; in mathematicalterms. Furthermore,the historical data necessary exactly described for meaningful hydrologic analysesare often lacking or unreliable. And when one considersthat hydrologic systems are generallyprobabilistic in nature, it is easyto understand that the modeler'stask is not a simpleone.In fact, it is often the casethat of the the best that can be hoped for from a model is an enhancedunderstanding systembeing analyzed.But this in itself can be of great value, leading,for example, to the implementationof datacollectionprogramsthat canultimately supportreliable modelingefforts. For the most part, mathematicalmodels are designedto describethe way a system's elements respondto sometype of stimulus(input). For example,a model of 'a groundwater system might be developed to demonstrate the effectson groundwater storageof various schemes for pumping. Equations 1.1 and L2 are mathematical modelsof the hydrologicbudget,and Figure 1.3 can be considered a pictorial model of the rainfall-runoff process. In later chapters, a variety of hydrologicmodelswill be presented and discussed. Thesemodelsprovidethe basisfor informed watermanagement decisions.

DATA 1.6 HYDROLOGIC

evaporation;soil Hydrologic dataarc neededto describeprecipitation; streamflows; moisture; snow fields; sedimentation;transpiration;infiltration; water quality; air, s9i!, and water temperatures; and other variablesor componentsof hydrologicsys-

12

CHAPTER 1

INTRODUCTION

tems. Sources of data are numerous,with the U.S. Geological Survey being the primary one for streamflow and groundwaterfacts. The National Weather Service (NOAA or National Oceanicand AtmosphericAdministration)is the major collector and other organizaof meterologicdata.Many other federal,state,and local agencies tions also compile hydrologicdata. For a completelisting of theseorganizationssee Refs.3 and 17.

OF MEASUREMENT 1 . 7 COMMON UNITS

cubic Streamandriver flowsare usuallyrecordedascubic metersper second(m3/sec), (sec-ft);groundwater flowsand watersupplyflo\i/s feetper second(cfs),or second-feet are commonly measuredin gallons per minute, hour, or day (gpm, gph, gpd), or millions of gallonsper day (mgd); flowsusedin agricultureor relatedto water storage are often expressedas acre-feet (acre-ft), acre-feet per unit time, inches (in.) or per hour (acre-in./hr). centimeters(cm) depth per unit time, or acre-inches Volumesare often given as gallons,cubic feet, cubic meters,acre-feet,secondfoot-days,and inchesor centimeters.An acre-footis equivalentto a volume of water (cfs-day,sfd) is the 1 ft deep over 1 acre of land (43,560 ft3). A second-foot-day accuinulated volumeproducedby a flow of 1 cfs in a24-hr period.A second-foot-hour (cfs-hr) is the accumulatedvolume produced by a flow of 1 cfs in t hr. Inches or centimeters of depthrelate to a volume equivalentto that many inchesor centimeters useful it is sometimes of water over the areaof concern.In hydrologicmassbalances, to note that 1 cfs-day : 2 acre-feetwith sufficient accuracyfor most calculations. Rainfall depthsare usually recordedin inchesor centimeterswhereasrainfall rates are given in inches or centimetersper hour. Evaporation,transpiration, and infiltration rates are usually given as inchesor centimetersdepthper unit time. Some usefulconstantsand tabulatedvaluesof severalof the physicalpropertiesof water are given in Appendix A at the end of the book.

'

PROBLEMS 1.8 APPLICATIONOF HYDROLOGYTO ENVIRONMENTAL

It is true that humanscannot exist without water; it is also true that water, mismanaged,or during times of deficiency(droughts),or times of surplus(floods),can be life threatening.Furthermore,there is no aspectof environmentalconcernthat doesnot relate in someway to water. Land, air, and water are all interrelatedas are water and all life forms. Accordingly, the spectrum of issuesrequiring an understandingof is almost unlimited. hydrologicprocesses the need As waterbecomes more scarceand as competition for its useexpands, for improved water managementwill grow. And to provide water for the world's food production, recreational expandingpopulation, new industrial developments, and other purdemands,and for the preservationand protection of natural systems poses, it will becomeincreasinglyimportant for us to achievea thoroughunderstandwith which we must contend.This is the ing of the underlyinghydrologicprocesses policymakers,lawyers, planners, engineers, challengeto hydrologists, waterresources economists,and others who must strive to see that future allocationsof water are sufficient to meet the needsof human and natural svstems.

PROBLEMS

13

r summary
the occurrence,distribution' moveHydrology is the scienceof water. It embraces an accountsense, -"nt, urii propertiesof the watersof the earth. In a mathematical a history that of a regionso ing may be madeof the inputs,outputs,and waterStofages of water movementfor the region can be estimated' the hydrologic After reading this chapter you should be able to understand shouldalso You budgetand make a simpleu".ouniing of water transportin a region' to be used facilitate have gained an undersiandingof trow hydrologic analysescan for water resourcessystems' processes designand management

PROBLEMS
1.1.. One-half inch of runoff resultsfrom a stormon a drainagearea
of 50 mi2.Convertthis amount to acre-feetand cubic meters. surfacearea of t.2. Assume you afe dealing with a vertical walled reservoir having a will it take to hours many How occurs: m3/sec 1.0 of 500,000 m' and that anlnflow raise the reservoirlevel bY 30 cm? time is 15 acre-ft and 1.3. consider that the storageexistingin a river reachat a reference from the reachis outflow the and cfs 500 is reach tie at the sametime the inflow to 650cfs.onehourlater,theinflowis550cfsandtheoutflowis630cfs.Findthe meters' changein storageduring the hour in acre-feetand in cubic walled reservoir was t.4. During a24-hr time period, the inflow to a 500-acre vertical a rise or fall in there in. was 1 was evaporation interval, same the 100 cfs. During centimeters' and in inches How muchwasit? Give the answer waterelevation? surface areais 3000 acres' 1.5. The annualevaporationfrom a lake is 50 in. If the lake's surface centimeters? in and acre-feet in rate evaporation daity what would beiire time requiredto raise 1.6. A flow of 10 cfs entersa 1-mi2vertical walledreservoir.Find the levelbY 6 in. the reservoir Iftheaverageannualrunoffis5l02cfsand areaof4511mi2. t.7. Adrainagebasinhasan for the areain losses the averalerainfall is 42.5 in.,estimatethe evaportranspiration is? estimate this you think 1 year. Iiow reliable do the storage in a reachof a river is 16.0acre-ft at a given time. Determine 1.8. The storage hour are the during outflow and inflow of rates (u"r"-f""tj t hr later if the average 700 and 650 cfs, resPectivelY. areafor 3 days' (a) L.9. Rain falls atataverage irrtensity of 0.4 in./hr over a 600-acfe the 3-day (b) determine per second; feet rate ofrainfau in cubic Determinethe average volumeofrainfallinacre-feet;and(c)determinethe3-dayvolumeofrainfallininches of equivalentdepth over the 600-acrearea' 100 acre-ft/day'Deter1.10. The evaporationrate from the surfaceof a 3650-acrelake is year ifthe inflow to the lake a 365-day during lake (feet) in the mine the depthchange is25.2cfs.1s the changein lake depth an increaseor a decrease? if the drainage to how many acre-feet 1.11. One and one-half inchesof runoff areequivalent : ft"') 43,560 areais 25-mi2?lNote: I acte rate of how many cubic feet L.12. one-half inch of rain per day is equivalentto an average per second? meters many p". ,".ona if the areais 500 acrei?How

14

cHAPTER 1 INTRoDUCTIoN

REFERENCES
1. P. B. Jones,G. D. Walker,R. W. Harden,and L. L. McDaniels,"The Development of the Scienceof Hydrology," Circ. No. 60-03, TexasWater Commission,Apr. 1963. 2. W. D. Mead, Noteson Hydrology. Chicago:D. W. Mead, 1904. 3. Ven Te Chow (ed)., Handbook of Applied Hydrology. New York: McGraw-Hill,1964. 4. O. E. Meinzer,Hydrology,Vol. 9 of Physicsof the Earth.New York: McGraw-Hlll, 1942. Reprintedby Dover, New York, 1949. 5. P. D. Krynine, "On the Antiquity of Sedimentationand Hydrology," Bull. Geol. Soc. Am. 70. l7 2I - l7 26(1960\. 6. RaphaelG. Kazmann,Modern Hydrology.New York: Harper & Row, 1965. 7. H. Pazwoshand G. Mavrigian, "A Historical Jewelpiece-Discovery of the Millennium Bull. 16(6), 1094-1096(Dec. 1980), Hydrologic Works of Karaji," WaterResources 8. Hunter Rouseand Simon Ince, History of Hydraulics, Iowa Institute of Hydraulic Re- :. search,State University of Iowa, 1957. in engen cylindrischenRohren," 9. G. H. L. Hagen, "Ueber die Bewegungdes Wassers gendorf Pog s Ann. Phys. Chem.16, 423- 442(1839). V. Dalmont, 1856. 10. Henri Darcy, Les fontaines publiques de la ville de Dijon. Paris:. des eaux dans les canauxs 11. J. Dupuit, Etudesthdoriqueset practiques sur le mouvement ed. Paris:Dunod, 1863. dtcouvertset d travers les terrainspermdables,2nd 12. L. K. Sherman,"Stream Flow from Rainfall by the Unit-Graph Method," Eng. NewsRec.108(1932). 13. R. E. Horton, "The Role of Infiltration in the Hydrologic Cycle," Trans.Am. Geophys. Union 14, 446- 460(1933). l 14. C. V. Theis,"The RelationBetweenthe Lowering of the PiezometricSurfaceand the Rate Union 16, and Duration of a Well Using Ground WaterRecharge,"Trans.Am. Geophys. 519-524(1935\. 15. Asit K. Biswas,"Hydrologic EngineeringPrior to 600 s.c.," Proc. ASCE J. Hyd. Div., Proc. Paper5431,Vol. 93, No. HY5 (Sept.1967). 16. RaymondL. Nace,"WaterResources: A GlobalProblemwith Local Roots,"Environ. Sci. Technol. 1(7) (July i967). 17. D. K. Todd (ed.), The WaterEncylopedia.New York: Water Information Center, 1970.

Chapter 2

Precipitation

r Prologue
The purposeof this chapteris to: ' Define precipitation, discussits forms, and describeits spatial and temporal attributes. ' Illustrate techniques for estimatingareal precipitation amountsfor specific storm eventsand for maximum precipitation-generating conditions. Precipitation replenishes surfacewater bodies,rbnewssoil moisture for plants, and recharges aquifers.Its principal forms are rain and snow.The relative importanceof theseforms is determinedby ttre climate of the area under consideration. In many parts of the westernUnited States,the extentof the snowpackis a determiningfactor relative to the amountof waterthat will be availablefor the summergrowing season. In more humid localities, the timing and distribution of rainfall are of principal concern. Precipitatedwaterfollows the pathsshownin Figs. r.2 and,1.3. some of it may be intercepted,evaporated, infiltrated, and becomesurfaceflow. The actual disposition depends on the amountof rainfall, soil moistureconditions,topography, vegetal cover soil type, and other factors Hydrologic modeling and water resourcesassessments dependupon a knowledgeof the form and amountof precipitation occurringin a region of concernover a time period of interest.

2.1 WATERVAPOR
The fraction of watervapor in the atmosphere is very small comparedto quantitiesof other gases present,but it is exceedingly important to our way of life. Precipitationis derived from this atmosphericwater. The moisture centent of the air is also a significantfactor in local evaporation processes. Thus it is necessary for a hydrologist to be acquainted with waysfor evaluatingthe atmospheric watervapor contentand to understand the thermodynamiceffects of atmospheric moisture.l

16

2 CHAPTER

PRECIPITATION

Under most conditions of practical interest (modest ranges of pressureand point is excluded),water vapor essentemperature,provided that the condensation tialiy obeys the gas laws. Atmospheric moisture is derived from evaporationand transpiration,the principal sourcebeing evaporationfrom the oceans.Precipitation the watervaporbeing overthe United Statescomeslargelyfrom oceanicevaporation, system. circulation primary atmospheric over the continentby the transporated to condirelated basically are humidity of watervapor or atmospheric Measures pure water. of surface a level over occurring tions of evaporationand condensation and air of water volumes equal Considera ilosed systemcontaining approximately evaporais dry, the air of condition initial If the maintainedat the sametemperature. A measurement tion takesplace and the quantity of water vapor in the air increases. pressurein the proceeds, of pressurein the airspacewill reveal that as evaporation vapor (vapor pressure the water of partial in of an increase because airipaceincreases equalsthe air pressure the overlying of preJsure). Evaporationcontinuesuntil vapor (evaporatleaving molecules water of of the excess surfacevapor pressure [a measure ceases, point, evaporation At this ing from) the water surfaceover thosereturning]. to be is said airspace equal, the are of the air spaceand water and if the temperatures not would equilibrium the of closed, saturated.If the containerhad beenopeninstead comSome have evaporated. havebeenreached,and all the water would eventually absomoistureor humidity are vapor pressure, of atmospheric monly usedmeasures point temdew and humidity, lute humidity, specifichumidity, mixing ratio, relative Perature.

Water Amount of Precipitable

. Estimatesof the amount of precipitation that might occur over a given region with favorable conditions are often useful. These may be obtained by calculating the extendingup from the earth's amountof water containedin a column of atmosphere althoughit cannotall be 14{ precipitable water the as quantity is known This surface. Precipitable water is usually by natural processes. removed from the atmosphere in centimetersor inches. expressed An equationfor computingthe amountof precipitablewater in the atmosphere can be derived as follows. Considera column of air having a squarebase 1 cm on a side.The total water masscontainedin this column betweenelevationzero and some height z would be

W:

J^

r p*dz

(2.1)

humidity and IVis the depthof precipitablewaterin centimewherep. : the absolute into ters. The integral can be evaluatedgraphically or by dividing the atmosphere uniform specifichumidities,solvingfortheseindividually, and layersof approximately amountof precipitablewater for the then summing.Figure 2.1 illustratesthe average continentalUnited Statesup to an elevationof 8 km.2

Variations and Temporal Geographic

water vapor varieswith location and time. ThesevariaThe quantity of atmospheric tions may be attributed mainly to temperatureand sourceof supply considerations. The greatestconcentrationscan be found near the ocean surfacein the tropics, the

2.2 PRECIPITATION

17

Sault Ste.Mtrie Portled

0;7
VCT -NJ 0.8

0.9 1.0

0.8 b.z 1'o o.d 1.1 r.2

t.J

Bromsville

Figure 2.L Mean precipitablewater, in inches,to an elevationof 8 km. (U.S. WeatherBureau.)2

generallydecreasing concentrations with latitude, altitude, and distanceinland from coastalareas. About half the atmospheric moisturecanbe found within the first mile abovethe earth's surface.This is becausethe vertical transport of vapor is mainly through convectiveaction,which is slight at higher altitudes.It is also of interestthat there is not necessarily water vapor over a any relation betweenthe amount of atmospheric regionand the resultingprecipitation.The amountof water vapor containedover dry that over considerablymore areasof the Southwest,for example,at times exceeds precipitation while humid northern regions,eventhough the latter areasexperience the former do not.

2.2 PRECIPITATION
Precipitation is the primary input vector of the hydrologiccycle. Its forms are rain, snow,andhail andvariationsof thesesuchasdrizzle and sleet.Precipitationis derived from atmospheric water, its form and quantity thus being influencedby the action of Atmoother climatic factors such as wind, temperature,and atmosphericpressure. sphericmoistureis a necessary but not sufficient condition for precipitation. Continental air masses are usuallyvery dry sothat mostprecipitationis derivedfrom moist maritime air that originates overthe oceans. In North America about50 percentof the evaporated water is taken up by continental air and movesback againto the sea.

18

2 CHAPTER

PRECIPITATION

Formationof PreciPitation
to be capableof supportingthe growth of dropletsof are considered Two processes (droplets about 500 to 4000 p'min diameter)to overcomeair from sufficient mass fall and consequently to the earth asprecipitation.Theseare known as the resistance process' ice crystalprocessand the coalescence their the small cloud dropletsincrease which by process is one The c^oalescence may be droplets Water collision. through droplets other with size due to contact resistance air and gravitational to both subjected are that bodies falling as considered effects. Fall velocities at equilibrium (terminal velocities) are proportional to the more quickly squareof the radius of the droplet; thus the larger dropletswill descend by larger overtaken are often droplets smaller result, As a ones. than the smaller increasingly producing the drops, to unite tend collisions resulting droplets,and the largir particles.Very large drops (order of 7 mm in diameter)break up into smallof a chain effect. processand producesomewhat dropletsthat repeatitre coalescence significant generate to produced be may raindrops large In this *unn"r, sufficiently in tropical important particularly be to ionsidered process is precipitation. This regionsor in warm clouds. An important type of growth is known to occurif ice crystalsand waterdroplets down to about -40'C- Under temperatures are found toexist togetherat subfreezing theseconditions,certain particles t saltsserveasfreezingnucleisothat theseconditionsis higher over the t occurson the surface condensation particle sizedistributionsde' uneven particles. This is considet with other mechanism. The artificial inducementof precipitation has been studied extensively,and nuclei supthat condensation thesestudiesare continuing.It has been demonstrated producthe to ensure humans of ability The precipitation. induce plied to cloudscan yet been not has timing or location geographic its control or to iion of precipitation attained,however' with the prospects problemsare associated Many legal as well as technological ..rain-makiig" that estimates hydrologic on impact is the here interest processes. Of of Many have. might precipitation artificial partially controlled uncontrolled oi onty as statisticalvariatesthat are naturally occurringLydrologicvariablesare considered component.If the distriburandom a with distributed or either randomty distrlUuted asto the frequencyof inference an modeled, can be variable the tion or time seiiesof (suchas precipitagiven magnitude a of events hydrologic occurrenceof significant effectsof these the if used and are controls artificial however, If, tion) can be made. unreliable totally prove to be may analyses frequency predicted, cannot be reliably tools.

TyPes Precipitation
and is responsible of condensation Dynamic or adiabaticcoolingis the primary cause air massesis a of transport for most rainfall. Thus it can be seen that vertical to the condiaccording be classified requirementfor precipitation.Precipitationmay

'1 2.2 PRECIPITATIOI'|9 of vertical air motion. In this respect,the three major categories tions that generate precipitation type are convective, orographic, and cyclonic. precipitation is typical of the tropicsand is Convective Precipitation Convective brought about by heatingof the air at the interfacewith the ground. This heatedair quantities expands with a resultantreductionin weight.During this period,increasing of water vapor are taken up; the warm moisture-ladenair becomesunstable;and pronouncedvertical currents are developed. Dynamic cooling takes place, causing condensation and precipitation.Convectiveprecipitationmay be in the form of light are a typical example). showersor stormsof extremelyhigh intensity (thunderstorms precipitation resultsfrom the mechanical Orographic Precipitation Orographic lifting of moist horizontal air currentsover natural barriers suchas mountainranges. This type of precipitation is very common on the West Coast of the United States where moistureladen air from the Pacific Oceanis interceptedby coastalhills and mountains.Factorsthat are important in this processinclude land elevation,local slope,orientation of land slope,and distancefrom the moisture source. In dealingwith orographicprecipitation,it is commonto divide the regionunder study into zonesfor which influencesasidefrom elevationare believedto be reasonably constant.For eachof thesezones,a relation betweenrainfall and elevationis maps(seeSection2,5). for usein producingisohyetal developed of is associated with the movement Cyclonic Precipitation Cyclonicprecipitation differregions.Thesepressure regionsto low-pressure from high-pressure air masses encesare createdby the unequalheatingof the earth's surface. Cyclonicprecipitationmay be classifiedasfrontal or nonfrontal. Any barometric low canproducenonfrontal precipitationasair is lifted throughhorizontalconvergenceof the inflow into a low-pressure area. Frontal precipitation results from the lifting of warm air over cold air at the contact zone between air masseshaving colder If the air masses are moving so that warm air replaces different characteristics. warm air, the front is known asawarmfront; if , on the otherhand,cold air displaces air, the front is saidto be cold.If the front is not in motion,it is saidto be a stationary front. Figure 2.2 illustratesa vertical sectionthrough a frontal surface.

through a frontal surface. Figure 2.2 Vertical cross-section

20

CHAPTER 2 PRECIPITATION

Thunderstorms
convectivestorms,which are Many areasof the United Statesare subjected to severe generallyidentifiedasthunderstorms of their electricalnature.Thesestorms, because although usually very local in nature, are often productive of very intenserainfalls that are highly significantwhen local and urban drainageworks are considered. with intense associated Thunderstorm cells develop from vertical air movements in the life history surface heatingor orographiceffects.Thereare threeprimary stages Theseare the cumulusstage,the mature stage,andthe dissipating of a thunderstorm. stage.Figure 2.3 illustrateseachof thesestages. All thunderstorms beginascumulusclouds,althoughfew suchcloudseverreach the stage of developmentneededto produce such a storm. The cumulus stageis characterized by strongupdrafts that often reachaltitudesof over 25,000ft. Vertical wind speeds at upperlevelsare often as greatas 35 mph. As indicatedinFig.2.3a, there is considerable horizontal inflow of air (entrainment)during the cumulusstage. of the storm, as additional moisture This is an important elementin the development is provided.Air temperatures inside the cell are greaterthart thoseoutside,as indicatedby the convexity of the isothermsviewed from above.The number and size of The duration ofthe cumulusstage the water dropletsincreaseasthe stageprogresses. is approximately i0-15 min. and the The strong updrafts and entrainmentsupport increasedcondensation in development of waterdropletsand ice crystals.Firrally, whenthe particlesincrease size and number so that surfaceprecipitation opcurs,the storm is said to be in the mature stage.In this stage strong downdrafts are created as falling rain and ice crystals cool the air below. Updraft velocities at the higher altitudes reach up to of over20 mph are Downdraft speeds 70 mptrin the early periodsof the maturestage.

El

o o F

(a)

(b)

(c)

cell. (Departstages of a thunderstorm Figure 2.3 Cumulus,mature,and dissipating ment of the Army.)

2.2 PRECIPITATION 21

usual aboveabout 5000 ft in elevation.At lower levels,frictional resistance tendsto decrease the downdraft velocity. Gusty surfacewinds move outwardfrom the region of rainfall. Heavyprecipitationis often derivedduring this preiod, which is usuallyon the order of 15-30 min. In the final or dissipating stage, the downdraftbecomes predominant until all the air within the cell is descending and being dynamically heated.Since the updraft ceases, the mechanismfor condensation ends and precipitation tails off and ends.

PrecipitationData
Considerabledata on precipitation are available in publications of the National WeatherService.a's Other sources includevariousstateand federal agencies engaged in water resourceswork. For critical regional studiesit is recommendedthat all possibledata be compiled; often the establishmentof a gauging network will be (seealso Chapters necessary 7-9).

pitation Variabi Preci Iity

Precipitationvariesgeographically, temporally, and seasonally. Figure 2.4 indrcates the mean annualprecipitation for the continentalUnited States,while Fig. 2.5 gives an exampleof seasonal differences.It should be understoodthat both regional and temporal variationsin precipitation are very important in water resources planning and hydrologicstudies. For example,it may be very important to know that the cycle of minimum precipitationcoincides with the peakgrowing season in a particular atea, or that the periodofheaviestrainfall shouldbe avoided in scheduling certainconstruction activities. Precipitation amounts sometimesvary considerablywithin short distances. Recordshaveshowndifferences of 20 percentor more in the catchof rain gauges less that2Oft apart.Precipitationis usuallymeasured with a rain gaugeplacedin the open so that no obstacle projects within the inverted conical surfacehavingthe top of the gaugeas its apexand a slopeof45'. The catchofa gauge is influenced by the wind, which usually causes low readings.Variousdevicessuchas Nipher and Alter shields havebeendesigned to minimize this error in measurement. Precipitationgauges may be of the recording or nonrecordingtype. The former are requiredif the time distribution of precipitation is to be known. Information about the featuresof gaugesis readily available.3 Because precipitationvariesspatially,it is usuallynecessary to usethe datafrom severalgaugesto estimatethe average precipitation for an area and to evaluateits reliability (seeChapter27). This is especiallyimportant in forestedareaswhere the variation tendsto be large. Time variations in rainfall intensity are extremely important in the rainfallrunoff process, particularly in urban areas(see Fi g. 2.6a). The arealdistributionis also significantandhighly correlatedwith the time history of outflow (seeFig. 2.6b).These considerations are discussed in greaterdetail in following chapters.

c) o U)
H

()

o E t r 9

.n

EF

C)
d

.v)
C) k

h0
A

C)

()

n"i
c.)

()
6

s
o E o

e<

(.)
o
H 6

U
F o

t^
0,) $ 6l

bo

,t,
,Ff

/i E

.;

.t

.o
k P)

t z

c)
H
'ir @h{ 6d F O

* \ - r (
E 1

H 0)

e
>r

\
F Fi
j;i.
oooooo aFhs {6d^O ./

/
,,-C)

t:
F o 5 > 6 k

ts

EU)
q . -

$,-,\ qi g( J:
F

O r +
! H

5 >
!2-

i < F "

. E
-

( ,E ; \ * \ i \
F h i -

,/

'r
"

9999qa9
h

\ --II--LItr--I-----LIL,]

lh '' LH : ,,l..15 r+: H= 5

R F =-Y

H S

au)

: R

t J

oooo t s

h+6Nio o
ooooo *odio

Ntr
o L ?1 ^

CHAPTER2

PRECIPITATION

- 3

'

sec) Time(x 102 (a)

-/'?:;"
E [u Y.s.

Isohyets-lines of equal rainfall depth

-"@l5
Ashd 1.56

I o R.H.R. i 3.50

t.e+ in.a'1t :.zs

1.55 upl.

^ 1<

'k '& &.1 *'uffi

Dsn. 1.08
(b)
F,,**

Figure 2.6 (a) Rainfall distribution in a convective storm June 1960, , Baltimore, Maryland. (b) Isohyetal pattern, storm of SeptemberL0, 195'1 Baltimore, Maryland. O, recording rain gauge.

INPUT 2.3 DISTRIBUTION OF THE PRECIPITATION

25

2.3 DISTRIBUTION OF THE PRECIPITATION INPUT

Total precipitation is distributedin numerousways.That interceptedby vegetation and treesmay be equivalent to the total precipitationinput for relatively small storms. Once interception storageis fllled. raindrops begin falling from leavesand grass, where water storedon thesesurfaces eventuallybecomesdepletedthrough evaporation. Precipitationthat reaches the ground may take severalpaths. Some water will fill depressions and eventually evaporate;some will infiltrate the soil. Part of the infiltrated watermay strike relatively imperviousstratanearthe soil surfaceand flow parallel to it as interflow until an outlet is reached. approximately Other portions may replenishsoil moisturein the upper soil zone, and someinfiltrated water may reach reservoirthat sustains The component of the the groundwater dry weatherstreamflow. precipitation input that exceeds the local infiltration rate will developa film of water Detention depths on the surface(surfacedetention)until overlandflow commences. varying from I to 1j in. for various conditions of slope and surfacetype havebeen reported.3 Overland flow ultimately reaches defined channels and becomes streamflow. Figure 2.7 ilhxtrates in a generalway the dispositionof a uniform storm input in nature,the to a natural drainage basin.Although suchan input is not to be expected indicated relations are representativeof actual conditions. Modifications resulting from nonuniform stormswill be discussed as they arise. In Fig. 2.7anote that the storm input is distributeduniformly over time /o at a rateequalto i (dimensionally intocomponents equalto LT '). This inputis dissected I, through lo, the sum of which is equal to I at any time r. Figure 2.7b illustratesthe manner in which infiltrated water is further subdividedinto interflow, groundwater, and soil moisture. Figure 2.7c showsthe transition from overland flow supply into The mechanics strearhflow. of theseprocesses will be treatedin detail in later sections. The nature of the curvespresenteddepictsthe generalrunoff process.It should be realized,however,that actual graphsof infiltration and/or other factorsversustime might appear quite different in form and relative magnitudewhen comparedwith of the effects of nonuniform storm patterns,antecedent theseillustrations becatrse conditions,and other factors. The rate and areal distribution of runoff from a drainagebasin are determined by a combination of physiographic and climatiOfactors. Important climatic factors includethe form of precipitation (rain, snow,hail), the type of precipitation (convective, orographic,cyclonic),the quantity and time distributionof the precipitation,the characterischaracterof the regionalvegetativecover,prevailingevapotranspiration factorsof significance tics, and the statusofthe soil moisturereservoir.Physiographic soil type, includegeometricpropertiesof the drainagebasin,land-usecharacteristics, geologicstructure,and characteristics ofdrainage channels(geometry,slope,roughness,and storagecapacity). Large drainagebasinsoften react differently from smalleroneswhen subjected to a precipitationinput. This can be explainedin part by suchfactorsas geologicage, characterrelativeimpact ofland-use practices,sizedifferential,variationsin storage istics, and other causes. Chow definesa small watershedas a drainagebasin whose

CHAPTER2

PRECIPITATION

(a/

\
I I

Soil moisture \a Int".flot

Mechanics of surface runoff

(c)

Figure 2.7 The runoff process:(a) dispositionof precipitation, (b) componentsof infiltration, and (c) dispositionof overland flow supply.

27 PRECIPITATION 2.4 POINT ofhigh-intensity, shortdo not filter out (1) fluctuationscharacteristic characteristics practices.6 On this basis,small (2) management the effectsof land duration storms;or is one in which basin A large mi2. 100 basinsmay vary from lessthan an acre up to precipitation imposed of frequencies high channelstorageeffectively filters out the practices. and effectsof land-use

2.4 POINTPRECIPITATION
at specificlocations.The resulting data Precipitation eventsare recordedby gauges p"rmit determinationof the frequency and charactei of precipitation eventsin the vicinity of the site. Point precipitation data are used collectively to estimateareal designstorm variability ofrain and snowand are alsousedindividually for developing in Design storms are discussed for small urban er other watersheds. characteristics detail in Chapter 16. Point rainfall data are usedto deriveintensity-duration-frequencycurvessuch as those shown in Fig. 2.8. Such curves are used in the rational method for urban in Chapter27 'ln storm drainagedesign(Chapter 25); thek constructionis discussed the average applyingthe rational method,a rainfall intensityis usedwhich represents duration.The frequencychosen intensity of a storm of given frequencyfor a selected should reflect the economics of flood damage reduction. Frequenciesof up to 100 yearsare commonlyusedwhereresidentialareasare to be protected.For highervaluedistricts and critical facilities,up to 500 yearsor higherreturn periodsare often Local conditionsand practicenormally dictatethe selectionof thesedesign selected. criteria. (ExecutiveOrder 11988,Floodplain Management,I97 7).

\
00-yr frequency 50-yr frequency I -20-yr frequency f, 10-yrfrequency
7 , 5-Yr,frequencl

\ \

t\
B.
a

\ \

\1

tl=.t

120
Duration (min)

curves Figure 2.8 Typical intensity-duration-frequency for Baltimore, Maryland, and vicinity.

28

CHAPTER2

PRECIPITATION

Figure 2.9 Four quadrants surrounding precipitation stationA.

It is occasionallynecessary to estimatepoint rainfall at a given location from recordedvaluesat surroundingsites.This can be doneto completemissingrecordsor precipitation to be used at the point of interest. The to determinea representative National WeatherServicehasdeveloped a procedurefor this which hasbeenverified on both theoreticaland empiricalbases.T Considerthat rainfall is to be calculatedfor point A in Fig. 2.9. Establisha set of axes running through A and determinethe absolutecoordinatesof the nearest surroundingpoints B, C, D, E, and F. These are recorded in columns 3 and 4 of precipitationat A is determined as a weightedaverage of the Table 2.L The estimated points. of AX and AY; other five The weightsarereciprocalsof the sumsof the squares that is, D2 : LX2 + LYz, and W : llDt. The estimatedrainfall at the point of interestis given by I (P x W)/> I{. In the specialcasewhere rainrall is known in only two adjacentquadrants(e.g.,I and II), the estimateis given asI (p x lV). This has the effect of reducing estimatesto zero as the points move from an area of
TABLE 2.1 DETERMINATION OF POINT RAINFALLFROM DATA AT NEARBY GAUGES

t1)
Point

(2) Rainfall
(in.)

(3) AX

(4) AY

(5) (D')

(6) wx103

(7) PxWx103

B C D E F Sums

r.60 1.80 1.50 2.00 r .

4 1 3 3 7 0

2 6 2 3 2

20 37 1,3 18 8

50 27.O 76.9 55.6 125.0

,TT3

80.0 48.6 115.4 111,.2 2t2.5 567.7

*Note.'Estimatedprecipitation(P) at A = 567.7 /334.5; P = 1.70 in.

_l

29 PRECIPITATION 2.5 AREAL to be the most logical proceprecipitation to one with no records.This is considered result will alwaysbe lessthan the The estimated dure for handlingthis unusualcase.7 greatestand greaterthan the smallestsurroundingprecipitation. For specialeffects suchas mountain influences,an adjustmentprocedurecan be applied.

2.5 AREALPRECIPITATION
it is important to know the areal distributionof precipFor most hydrologicanalyses, itation. Usually, averagedepths for representativeportions of the watershedare determinedand used for this pwpose. The most direct approachis to use the arithare unimetic average of gaugedquantities.This procedureis satisfactoryif gauges is flat. Other commonly usedmethodsare the formily distributedand the topography at one method.The reliability of rainfall measured isohyetalmethodand the Thiessen gaugein representing depth over a surroundingareais a function of (1) the average area,(2) the sizeof the the distancefrom the gaugeto the centerof the representative (e.g., (4) the natureofthe rainfall ofconcern stormeventversus area,(3) topography, For more information on meanmonthly), and (5) local stormpatterncharacteristics.8 errorsof estimation,the readershouldconsultRefs.7 and 8. Chapter27 alsocontains a discussion of areal variability of precipitation. Figures 2.10 and 2.11 illustrate how the measuredrainfall at a single gauge in ( 1) the relativeposition with change rainfall over a watershed relatesto the average of the gaugein the watershedand (2) the time period over which the averageis In the first caseit is clearthat the more centralthe gaugelocation,the more calculated. area, providing for a representative will match the average closely its observations that areal averages not surprisingly, that the regionis not too large.Figure 2.11 shows,

o bo o >
J

o > o

2, *,
d 9 1

b 9
d 9 n

3.=

B.E

/
k l O

,.1:i'j
1 2 3

(n

Storm rainfall at one gauge in inches (a)

Stormrainfall at one gaugein inches (b)

average Figure 2.L0 Errors resultingfrom useof a singlegaugeto estimatewatershed area is 0.75 mi2 and (giuge location effect, Soil ConservationService),(a) Watershed areais 0.75 mi2 and gaugeis 4 mi outsidethe gaugeis near the center. (b) Watershed watershed boundary.

30

CHAPTER2

PRECIPITATION
d bo
6 o 2 o

o, bI) o >

EE 2
5 . =

b 9

a
d

220

'a
d r l

,/
10
U)

v)
1 ) {

Storm rainfall at one gauge in inches

Stormrainfall at one gaugein inches

(a)

(b)

average estimatewatershed Figure 2.L1 Errors resultingfrom use of a singlegaugeto and the mi2 5'45 is area (aiWatershed Service). ConserVation (time period effect, Soil gaugeisontheboundary.(b)Watershedareais5.45In|zandthegaugeisonthe boundarv. to conform more closely over long time periods, in this case one year, may be expected event' This suggests storm to a single guog" uu"ruge than those for an individual space and time both with tempered that the Oerlgn-of guuging networks should be considerations.

lsohyetalMethod
of rainfall are the isohyetal The two principal methodsfor determiningareal averages basedon interpolation method and the Thiessenmethod. The isohyetal method is in surveyingand between gauges.It closely resemblesthe calculation of contours plot the^rain gauge to -upptng."Tf" first step in developingan isohye?1..-up is Next' an (Fig' 2'I2)' locationson a suitablemap and to reJord the rainfall amounts increselected at interpolation betweengaugesis performed and rainfall amounts to form connected then mentsare plotted.tdenticaldepthsfrom eachinterpolationare of average weighted the is isohyets(lines of equal rainfall depth)' The areafaverage isohyetal The isohyets' depthsbetweenisohyets,that is, the meanvaluebetweenthe precipitation over an method is the most accurateapproachfor determiningaverage to topographic attention area,but its proper oserequir#a skilled analystand careful the represenillustrates and other tactori that impact on areal variability. Figure 2. 13 tationofamajorstormeventinNorthCarolinabyanisohyetalmap.

ThiessenMethod
In this method' is theThiessen arealrainfallaverages of calculating method Another centers' as raingauges using subareas intopolygonal is subdivided thearea procedure Thiessen depth' average watershed the estimiting in asweights areused Thesubareas for suitable is not in Fig. i.t+. fnis procedure as shown are constructed diagrams

2.5 AREAL PRECIPITATION

31

Average

precipitation
for area A4 is 4.25 in.

----x------*
A2

-*!fl1
--^ 3 in'
Average preciPitation = 2 A i P i \ i for entire basin

(c)

and map:(a)locateraingauges of an isohyetal Figure 2.12 Construction (c) plot isohyets' and gauges; between (b) interpolate pdt values; of orographicinfluences'The Thiessennetwork is fixed mountainousareasbecause are if any gauges mustbe reconstructed io, u glu"o gaugeconfiguation, and polygons relocated.

AccuracY
locatior\of the Irrespectiveof the methodusedfor estimatingareal precipitation,the of the estimate application of point tothe guu* orra in derivingthe estimaterelative may be distances vertical lbcattieso In mountainous mustbe takeninto consideration. are spacings horizontal gentle landscapbs, For -ft" irnpottant than horizontal ones.

32

CHAPTER 2 PRECIPITATION
a . !.1 a.l - o !

oo>
I
I

I ti-

a ( h w t

o 'r-r

r'.

-a)

!?+

? z

+ r Q

r E
I f.\

o ( g 9 - 's
o ( )

c R o >

\
s

sz' g
d . a

c..l &

c]

(f)x
!-l +i

N i 4

e-H

TH

+t Figure 2.14 Constructionofa Thiessen diagram:(a) connectrain gaugelocations;(b) draw perpendicularbisectors;and (c) calculate Thiessenweights l,er, ,qr, A3). (d) A completed network. the most important. when a precipitation gauging network is to be developed, both spacing and arrangement of gaugei must be considered.

Average depth overentire = watershed

EXAMPLE 2.1 $yen 1rrcdrainagearea of Fig. 2.r5 and the rainfall data displayedin column 3 of Table2.2, calculatethe average rainfall over the areausing 1aj tne arithmetic mean, and (b) the Thiessenpolygonweighting system.

Figure 2.15 Thiessendiagram for Example 2.1.

34

CHAPTER 2

PRECIPITATION

TABLE 2.2 DATA AND THIESSEN POLYGON FOR EXAMPLE 2,1. CALCULATION (1) GaugeNo.

(2) "/" Area

(3)
Precip.-in 1.56

(4)

(2) x (3) 0.08 0.12 0.10

0.44

2 3

2.95
3.44

3 t5
ll

2.91
4.17 4.21

0.46 0.80
0.11 0.17 0.81

6
7 8 9

19
4

2.'| 2.45
3.88 3.98

21
6 5

10
ll

0.24 0.13
3.45

2.51

Total

100

Solution. falling within the areaboundary.They includegauges a. Identify thosegauges yields an 1, 4 through 6, 8, and 9. Averagingthe valuesfor thesesix gauges estimatedmean areal rainfall of 3.20 inches. b. Followine the Thiessen method as described in Section 2.5, construct polygonsusingtrianglesto connectgaugepoints. Thesepolygonsare shown with each on Fig. 2.15. Calculatethe percentof the total area associated weightedaverage gaugeandrecord asin column2 of Table2.2.The Thiessen is obtainedby multiplying the valuesin column 2by the yaluesin column 3. The Thiessenaverage is computedas 3.45 inchesof rainfall. The use of a (Table 2.2) facilitates computations and aids in organizing spreadsheet data. lr

MAXIMUM PRECIPITATION 2.6 PROBABLE

rainThe probablemaximum precipitation (PMP) is the critical depth-duration-area which would result from a storm containing fall relation for a given areaand season probable.e Such storm events conditionsconsidered the most critical meteorological are used in flood flow estimatesby the U.S. Corps of Engineersand other water of conditions are basedon analyses The critical meteorological resourcesagencies. air-mass properties(effectiveprecipitablewater,depthof inffow layer,wind, temperature, and other factors),synoptic situationsduring recorded stormsin the region, and locationofthe area.The rainfall derivedistermedprobable topography, season, theory and maximumprecipitation sinceit is subjectto limitations of meteorological data and is basedon the most e_ffectiwcqmbination of factors controlling rainfall

) {)
o q)
k C)
H

9.'T\):

. o >

E
o

(.)
k

U
o
o L*

- N r

C)

,r)
ri s c'l

N k

o o
H

H g

\o ol I
b0 It

q6

,/

) {)
o q)
k C)
H

9.'T\):

. o >

E
o

(.)
k

U
o
o L*

- N r

C)

,r)
ri s c'l

N k

o o
H

H g

\o ol I
b0 It

q6

,/

36

CHAPTER2

PRECIPITATION 1000 800 600 400

$ zoo
E
9?

s 99 t r o u
4 4 0 20 10 l0

1oo

20 30 40 50 6Q 70 80

90 100 110 120 130 140 150

Percentageof 200 miz, 24-hr values

Figure 2.17 Seasonal variation, depth-area-duration relations; percentage to be applied to 200 ni2-24 hr probable maximum precipitation values for August in Zone 6. (U.S. Department of Commerce,National WeatherService.)

intensity.e An earlierdesignation of "maximumpossible precipitation" is synonymous. Additional information on PMP is given in Chapter 16. The seasonalvariation of PMP is important ip the designand operation of multipurposestructures and in floodingconsiderations that may occurin combination with snowmelt. In both of these cases,annual probable maximums might be less important than seasonal maximums.Figures2.16 and,2.r7 display 24-hr pMp for the eastern half of the United Statesfor 200-ffi2 watersheds during the month of August (similar figuresare availablefrom the National WeatherService).

2.7 GROSS AND NETPRECIPITATION

The net (excess) precipitation that contributesdirectly to surfacerunoff is equivalent to the gross precipitation minus lossesto interception, storm period evaporation, depression storage, and infiltration. The relation betweenexcess precipitationP" and grossprecipitation P is thus P":P.-Ilosses (2.2)

where the lossesinclude all deductions from the grossstorm input. The paths that water precipitatedover an areamay take can be represented by flow diagramsof the type given in Fig. 1.3 and by equationsof the form ot&q. 2.i. Modelssuchastheseare the basisfor most hydrologicinvestigations, and muchbf the contentof this book is devotedto the conceptualization of individual components of the various hydrologicprocesses and to synthesizing thesecomponentsinto holistic representations of hydrologicevents.

PROBLEMS 37

r Summary
for the planetEarth. Too much precipitation is the sourceof fresh waterreplenishment In betweenthese prosperityand disaster. or too little canmeanthe differencebetween with a frequency extremesare the normal precipitation eventsthat are experienced features' topographic and position geographic to mainly related and intensity After reading this chaptei you should understandthat both the timing and considamountof precipitition occuiring over anareaareimportant and that thereis areal erable g"olrapttic variability in precipitation. You should be able to estimate process simplehydrologic precipiLtiin amountsfrom gaugedata and conceptualize for a region shed ofprecipitation values average recognizedihat be It should models. uses, some light on the quantiiy of water that might be made availablefor various of tne time-aistribution and time-dispositionof precipitation are while a i<nowledge and shortage' plans for periods of excess management requisitesfor developing PROBLEMS 2.1.

"r)

estimatethe amount of rainfall for gaugeX' average, artdc compute the rainfall for gaugeX in Problem 2.1 if the storm readingsat A, B, 'werc respectively' 3.7,4.I, and4'8 in., in the following figure using 2.3. compute the meanannualprecipitationfor the watershed and the isohyetalmethod. The method, the arithmetic mean, the'itri"*"n polygon

38

CHAPTER2 PRECIPITATION gaugereadings for gauges A-K, respectively, are: 29.79,34.97,25.6,24.2i,24.6, 42.61,42.35,15.51,39.99, 43.04,and28.41. 2.4. Computethe meanannualprecipitationfor the watershed in the figurefor Problem2.3 using the arithmetic mean and the Thiessen polygonmethod.The gaugereadingsfor gauges A-K, respectively, are: 28.1, 33.7, 25.6, 23.9, 24.6, 40.1, 41.3, 37.2, 38.7, 41.1,and29.3. 2.5. The chart from a rain gaugeshown in the sketchrepresentsa record that you must interpret.Find the average rainfall intensity (rate)between6 e.r'1. and noon on August 10. Find also the total precipitation on August 10 and August 11.
0.30 o.25 0.20 0.15 0.10 0.05 0
Midnight 6 a.m.

Aus. l 0

Ais..I7

6 p.m. Midnight 6 a.m. Time

Noon

6 p.m. Midnight

2.6. Refer to the chart of Problem 2.5. Calculate the rainfall intensity for the period between6.q..1u. and noon on August 11. Would you considerthis to be a period of intenserainfall? 2.7. Use the map of Fig. 2.6 andfrom it constructa setof Thiessen polygons. Using these, estimatethe mean rainfall for the region. 2.8. A meandraft of 100mgd is producedfrom a drainageareaof 200 mi2.At the flow line the reservoiris estimated to cover 4000 acres.The annualrainfall is 37 in., the mean annualrunoff is l0 in., and the mean annual lake evaporationis 30 in. Find the net gain or lossin storage.Computethe volume of water evaporated. How significantis this amount? 2.9. A meandraft of 380,000m3/dayis producedfrom a drainageareaof 330 km2.At the flow line, the reservoiris estimatedto cover about 1600hectares. The annualrainfall is 96.5 cm, the meanannualrunoff is 22.8 cm, and the meanannuallake evaporation is 77.1 cm. Find the net gain or loss in storageand compute the volume of water evaporated. Calculatevolumes in m3. 2.10. Drainage areas within each of the isohyetal lines for a storm are tabulated for a watershed.Use the isohyetal method to determinethe averageprecipitation depth within the basin for the storm. Make a conceptualsketch.

lsohyetalinterval(in.)

Area (acres) 2700 1900 1000 0

0-2
a i

4-6 6-8

2.11. ReworkProblem2.10 if the valuesin the secondcolumn of the table are2,500, 2,100 1,200, and 300, respectively,

REFERENCES 39 2.12. Discusshow you would go about collectingdata for analysisof the water budgetof a region.What agencies would you contact?What other sourcesof information would you seekout? 2.13. For an areaof your choice,plot the meanmonthly precipitation versustime. Explain how this fits the pattern of seasonal water usesfor the area.Will the form of precipitation be an important consideration?

REFERENCES
I

2. 3.
A

5. 6.
,1

8. 9.

Tennessee Valley Authority, "Heat and Mass TransferBetweena Water Surfaceand the Atmosphere,"Lab. Rep. No, 14, TVA EngineeringLab. Noiris, TN, Apr. 1972. A. L. Shands,"Mean PrecipitableWater in the United States," U.S. WeatherBureau, Tech.PaperNo. 10, 1949. R. K. Linsley, Jr., M. A. Kohler, and J. L. H. Paulhus,Applied Hydrology. New York, McGraw-Hill, 1949. D. W. Miller, J. J. Geraghty,and R. S. Collins, WaterAtlas of the United States.Port Washington, NY Water Information Center, 1963. U.S. WeatherBureau,Tech. Papers1-33. Washington, DC: U.S. GovernmentPrinting Office. Ven Te Chow (ed.), Handbook of Applied Hydrology. New York, McGraw-Hill,l9&. Staff, Hydrologic Research Laboratory, "National WeatherServiceRiver ForecastSystem ForecastProcedures,"NOAA Tech. Mem. NWS HYDRO 14, National Weather Service,SilverSpring,MD,Dec. 1972. V. Mockus, Sec.4, in SCSNational EngineeringHandbookon Hydrology,Washington, DC: Soil ConservationService, A:ug.1972. J. T. Riedel, J. F. Appleby, and R. W. Schloemer,"SeasonalVariation of the Probable Maximum Precipitation East of the 105th Meridian for Areas from 10 to 1000 Square Miles andDurations of 6,12,24,and48 Hours,"Hydrometeorological Rept.No. 33, U.S. WeatherBureau,Washington, D.C., 1967.

Chapter3

Interception and Depression Storage

r Prologue
The purposeof this chapteris to: . Define interception and depression storqge. ' Define the roles theseabstracting play in affectingthe amountof mechanisms precipitatedwater ultimately availablefor other distribution. ' Providesomeapproaches to estimatingthe quantitiesof waterinterceptedand storedin depressions during precipitation events. Figure 1.3indicatesthe pathsthat precipitatedwatermay take asit reaches the earth. The first encountersare with interceptingsurfacessuch as trees,plants, grass,and structures. Waterin excess of interceptioncapacitythen beginsto filI surfacedepressions.A film of water alsobuilds up overthe ground surface. This is known as surface detention.Once this film is of sufficientdepth, surfaceflow toward definedchannels commences, providing that the rate at which water seeps into the ground is lessthan the rate of surfacesupply.This chapterdealswith the first two mechanisms by which the grossprecipitation input becomestransformedinto net precipitation.

3.1 INTERCEPTION
Part ofthe stormprecipitationthat occursisintercepted by vegetation andotherforms ofcover on the drainage area.Interception canbe defined asthat segment ofthe gross precipitationinput which wets and adheres to aboveground objectsuntil it is returned to the atmospherethrough evaporation.Precipitation striking vegetationmay be retained on leavesor blades of grass,flow down the stemsof plants and become stemflow,or fall off the leaves to becomepart of the throughfall. The modifying effect that a forest canopycan haveon rainfall intensity at the ground (the throughfall) can be put !o practical use in watershed management schemes. The amountof waterintercepted is a function of ( 1) the stormcharac ter,(2) the species, age,and density ofprevailingplantsandtrees,and (3) the season ofthe year. Usually about 10-20 percentof the precipitationthat falls during the growing season

41 3.1 INTERCEPTION is interceptedand returned to the hydrologiccycle by evaporation.Water lossesby closedforest stands-as much as interceptionare especiallypronouncedunderdense hasreportedthat the average 25 percentof the total annualprecipitation.Schomaker is annualinterceptionlossby Douglasfir standsin westernOregonand Washington about24 percent.l A lO-year-oldloblolly pine plantation in the South showedlosses pine forestsin Califoron a yearlybasisof approximately14 percent,while Ponderosa nia were found to intercept abort 12 percent of the annual precipitation. Mean interception lossesof approximately 13 percent of gross summer rainfall were reported for hardwoodstandsin the White Mountainsof New Hampshire.Additional information given in Table 3.1 includes some data on interception measurements obtainedin Maine from a mature spruce-firstand, a moderatelywell-stockedwhite and gray birch stand,and an improved pasture.r Lull indicatesthat oak or aspenleavesmay retain as much as 100 drops of For a well-developed tree, interception storageon the order of 0.06 in' of wdtter.z retentionof about precipitationcould thereforebe expected on the basisof an average CATCH OFSTANDARD PRECIPITATION TABLE 3,1 WEEKLYAVERAGE IN LOCATED RAIN GAUGES WEATHER BUREAU.WPE U,S.
A SPRUCE-FIR STAND, A HARDWOOD STAND, AND A PASTUREDURINGTHE WINTER OF 1965_1966 Weekly average precipitationcatch (in. ot equivalent rain) Percent interception by forest cover

Measuring date'

Spruce-fir 0.24 1.01 1.01 1.41 0.55 0.66 0.20 0.36 Trace 0.25 1.38 0.05 0.29, 0.76 0.17 0.86 0.76 0 0.73 10.69

Birch

Pasture 0.39 L45 r.36 t.79 0.87 1.08 0.26 0.61 Trace 0.59 1.96 0.06 Trace 0.98 0.22 t.45 0.97 0 1.27 15.31

Spruce-fir 38 30 26 2l 37 39 23 4l 58 30 l7 22
ZJ

Birch l5 l4 10 8 7 l2
A

tl/9t65 11tL6/65 ru23t65 12tr0t65" 12n7 t65 12/30t65 lt4t66 1tr2t66 U18t66 U25t66 2tr/66 2t8t66 2n1t66 2n5t66 2t21t66 3t2t66 3t7t66 3t15t66 3t29t66 Total

0.33 r.25 r.23 1.65

0.81 0.95 0.25 0.55 Trace 0.58 l.9l 0.07 0.02 0.81 0.22
l -za

10 2
J

t6 t7 0 I6 13 1l 9.5

0.84 0 1.13 13.83

4l 22 43 30.2

" The period betweenmeasuringdatesis 7 days,exceptwhen precipitation occurred on the seventhday. In this event, measurgmentwas postponeduntil precipitation ceased. bMeasurementswere delayeduntil a method was devisedto melt frozen precipitation on the site. 'This measurementin the spruce stand was the result of foliage drip during a thaw from previously intercepted snow. "The Effect of Forest and Pastureon the Disposition of Precipitation," Source: Aftet C, E. Schomaker, Maine Farm Res.(July 1966).

42

oHAPTER s rNrER;;-;;rr;-.r,o*

"ro*ou=

20 drops per leaf. For light showers(where gross precipitation P < 0.01 in.) 10t percentinterceptionmight occur, whereas for showers whereP > 0.04 in., losses in the rangeof 10-40 percentare realistic.3 Figure 3.1 illustratesthe generaltime distribution pattern of interceptionloss intensity.Most interceptionlossdevelops during the initial stormperiod and the rate of interception rapidly approacheszero thereafter.l-6Potentialitorm interception losses can be estimated bv usins2'3,6 Zi:,S+KEt

(3.1)

where . L, : the volume of water intercepted(in.) S : the interceptionstorage that will be retainedon the foliage againstthe forcesof wind and gravity (usuallyvariesbetween 0.01 and 0.05 in.) K : the ratio of surfaceareaof interceptingleaves to horizontal projection:, of this area E : the amountof waterevaporated per hour duringthe precipitationperiod ' (in.) r : time (hr)

i= it+i2+4+i4

Figure 3.1 Disposition of rainfall input in terms of interception, depression storage,infiltration, and overlandflow.

3.1 INTERCEPTION 43 Equation3.1 is basedon the assumption that rainfall is sufficientto fully satisfy the storageterm S. The following equationwas designed to accountfor the rainfall amountT-e
L;:S(1 -e-P/s)+KEt

(3.2)

whereP : rainfall and e is the baseof natural logarithms.Note in Eqs. 3.1 and 3.2 that the storm time duration t is given in hours, while ,L,, S, and E are commonly measured in in. or mm. It is important to recognizethat forms of vegetationother than trees can aiso interceptlarge quantitiesof water. Grasses, crops,and shrubsoften haveleaf-areato ground-arearatios that are similar to thosefor forests.Table3.2 summarizes some observations that havebeen madeon crops during growing seasons and on a variety of grasses. Interceptedamountsare aboutthe sameasthosefor forests,but sincesome of thesetypes of vegetationexist only until harvest,their annualimpact on interception is generallylessthan that of forested areas. Precipitationtype, rainfall intensityand duration,wind, and atmospheric conditions affecting evaporationare factors that serve to determineinterception losses. Snow interception,while highly visible,usually is not a major loss sincemuch of the interceptedsnowfall is eventuallytransmittedto the ground by wind action and melt. Interceptionduring rainfall eventsis commonly greaterthan for snowfall events.In both cases, wind velocity is an important factor. The importanceof interceptionin hydrologicmodelingis tied to the purposeof the model. Estimates of loss to gross precipitation through interception can be significantin annual or long-term models,but for heavy rainfalls during individual storm events,accountingfor interceptionmay be unnecessary. It is important for the modeler to assess carefully both the time frame of the model and the volume of precipitation with which one must deal. TABLE3.2 OBSERVED PERCENTAGES OF INTERCEPTION BY
VARIOUSCROPSAND GRASSES'

Vegetation type
Crops Alfalfa Corn Soybeans Oats Grassesb Little bluestem Big bluestem Tall panic grass Bindweed Buffalo grass Blue grass Mixed species Natural grasses

(%) Intercepted

Comments

36 16 15 7 50-60-.1

s7 l s 7 f 17 l )
31 17
)A

Water applied at rate of ] in. in 30 min

Pdor to harvest

t4-t9

"Valuesroundedto nearest percent.Data for table were obtainedfrom Refs.2,4, and 5, 'Grass heightsvary up to about 36 in.

44

CHAPTER3

INTERCEPTION AND DEPRESSION STORAGE

but for Equations3.1 and 3.2 canbe usedto estimatetotal interceptionlosses, to dealwith the areal variability ofindividual storins,it is necessary detailedanalyses however. are not available, for estimatingsuchlosses Generalequations of suchlosses. Most researchhas been related to particular speciesor experimentalplots strongly associated with a given locality. In addition, the lossfunction varieswith the storm's experimentaldata are available,the nature of the varianceof character.If adequate interceptionversustime might be inferred. Otherwise,common priictice is to deduct volumeentirelyfrom the initial period of the storm(initial abstraction). the estimated EXAMPLE 3.1 by Horton6for interceptionby ash and oak Using the following equationsdeveloped trees, estimatethe interception loss beneaththesetrees for a storm having a total precipitation of 1.5 in. Solution 1. For ashtrees.

L ; : 0 . 0 1 5+ 0 . 2 3 P : 0.36 : 0.015 + 0.23(1.5) in.

2. For oak trees,

L;:0.03+0.22P : 0.36 : 0.03+ 0.22(1.5) in. rl

3.2 THROUGHFALL
for estimatingthroughfall for a variety of foresttypeshave A numberof relationships tt Deiermining factors for throughfall quantities include canopy been developed.n wind velocity, and total leaf area,numberand type of layersof vegetation, coverage, rainfall intensity.The arealvariability ofthesefactorsresultsin little or no throughfall in somelocationsand considerable throughfall in others.In general,prediction equaareaand coverasprime of canopysurface tions for throughfall mustincludemeasures variables.An example of a throughfall relationship for an easternUnited States hardwoodforest follows.l2 For the growing season - 0.O3ln Tn: 0.901P For the dormant season T n : 0 . 9 I 4 P- 0 . 0 1 5 n where ?1,: throughfall (in.) P : total precipitation (in.) n : number of storms

(3.3) (3.4)

3.3

DEPRESSION STORAGE

45

3.3 DEPRESSION STORAGE

Precipitationthat reaches the ground may infiltrate, flow over the surface, or become trappedin numeroussmall depressions from which the only escape is evaporationor infiltration. The natureof depressions, aswell astheir size,is largely a funition of the original land form and local land-usepractices.Because of extremevariability in the nature of depressions and the paucity of sufficient measurements, no generalized relation with enoughspecifiedparameters for all cases is feasible.A rational model can, however,be suggested. Figure 3.1 illustratesthe dispositionof a precipitationinput. A studyof it shows that the tate at which depression storageis filled rapidly declinesafter the initiation of a precipitationevent.Ultimately, the amountof precipitation goinginto depression storage will approachzero,given that thereis alargeenoughvolume of precipitation to exceedother lossesto surfacestoragesuch as inflltration and evaporation.Ultimately, all the water stored in depressions will either evaporateor seep into the ground.Finally, it shouldbe understood that the geometryof a land surface is usually complex and thus depressions vary. widely in size, degreeof interconnection,and contributingdrainagearea.In general,depressions may be looked upon as miniature reservoirsand as suchthey are subjectto similar analytical techniques. According to Linsley et a1.13 the volume of water storedby surfacedepressions at any given time can be approximated using
Y:Sd(l -e-kP")

r? 5)

where V : the volume actually in storageat sometime of interest S, : the maximum storagecapacityof the depressions P" : the rainfall excess (grossrainfall minus evaporation,interception,and infiltration) k : a constantequivalentto l/So The valueof the constantcan be determined by considering that if P" : 0, essentially all the water will fill depressions and dv/dp" will equal one. This requires that k : r/Sa. Estimatesof s, may be securedby making samplefleld measurements of the areaunder study.Combiningsuchdata with estimates of P" permits a determination of V. The mannerin which Vvarieswith time must still be estimated if depression storagelossesare to be abstracted from the grossrainfall input. One assumptionregarding dVldt is that all depressions must be full before overlandflow supply begins.Actually, this would not agreewith reality unlessthe locationsof depressions were gradedwith the largestonesoccurring downstream. If the depression storagewere abstracted in this manner, the total volume would be deducted from the initial storm period suchas shownby the shadedareain Fig.3.2. Such postulateshave been used with satisfactory results under special circumstances.ra Depression storage intensitycan alsobe estimated usingEq. 3.5. If the overland flow supplyrale oplus depression storage intensityequali - /, wherei is the rainfall intensity reachingthe ground and/is the infiltration rate, then the ratio of overland flow supply to overlandflow plus depression storagesupply can be proved equal to

i - f

(3.6)

46

CHAPTER 3 INTERCEPTION ANDDEPRESSION STORAGE

o b0 o

!t o
d

1
Time (min)

16

20

Figure3.2 Simple depression storage abstraction scheme. This expression can be derivedby adjudging c : i - f - o (3.7) i - f i - f and noting that o is equal to the derivativeof Eq. 3.5 with respectto time. Then )

o:fiso1t_e-kP")

(3.8)

u : (Soke-kg#
It was shown that k : 1/S, so that

(3.e)

u : ,-o'"d!"
dt

(3.10)

precipitationP, equalsthe grossrainfall minus infiltrated water, and since The excess the derivativewith respectto time canbe replacedby the equivalent intensity(i * f), the intensity of depression storagebecomes
o:e-or.(i-f)

(3.11)

i - f

(i-f)*G-f)e-."" i - f

(3.r2)

o.125

o.25

0.315

0.50(turf)

Mass overlandflow and depression storagesupply ( P - F)

0.0625
All depressions filled before overland flow supply begins -=
:\ on

0.0938

0.125(pavements) 1.00

----l

9)E o " 7

RO

0.80
bl I

l-

Exponential relationship a

o.7o I
-\ I ,

F i 6 0 6
o

-lP-F)tS, " = I - e I

i,t

tl

OGEE sumrnation of the probability curve standard

n5n a

o b!

B
o E

{
.: 40

9 d J U

o o0

s2
o

6zo t
!)

I
E

50

100

150

200

Mean depth as a percentage of overall depth of depression storage

Figure 3.3 Depth distributioncurve ofdepressionstorage. Enter graph from top, readdown to selected curve, and project right or left as desired.(After Tholin and Kiefer.r5) and o f)(I - e-kP")

i * f

:(i

i - f
| g-kP"

(3.13) (3.r4)

Figure 3.3 illustratesa plot of this function versusthe massoverlandflow and depression storage supply(P - F), whereF is the accumulated massinfiltrationl5 and

48

3 CHAPTER

AND DEPRESSION STORAGE INTERCEPTION

P is the grossprecipitation.In the plot meandepthsof 0.25 in. for turf and 0.0625in. Maximum depthswere 0.50 and0.125in., respectively. were assumed. for pavements The figure also depictsthe effect on estimatedoverlandflow supplyrate, which model. Three modelsare shown storage is derivedfrom the choiceof the depression are full before overlandflow all depressions that in the figure: the flrst one assumes a mean depth of 0.25 in., the figure with begins.For a turf area having depressions is no overlandflow supply,while in., there less than 0.25 F values showsthat for P flow supply is equal to i - f . greater overland in., the F values than 0.25 for P will be greaterthanzero. Tholin For the exponentialmodel (Model 2), c always between thosepreviouslymentionedis a relation that and Kiefer haverecommended A cumulative normal urban areas.15 of fully developed likely more representative in Fig. 3.3 described and is also probability curve was selected for this representation (Model 3). are usually madefrom the first part of the storm storage deductions Depression as illustrated in Fig. 3.2. The amount to be deductedis a function of topography, this loss During major storms, groundcover,and extentand type of land development. storage guidelines for estimating depression to be negligible.Some is often considered other watersheds. and of experimental studies basedon losseshavebeen developed Values for depressionstoragelossesfrom intense storms reported by Hicks are Tholin andKieferhaveused 0.15in. for loam,and0.10in. for c1ay.16 0.20 in. for sand, Studies of in. forpavements.ls 0.0625 and valuesof 0.25 in. inpervious urban areas yielded in information shown the by Viessman four small imperviousdrainageareas This is slope. correlated with loss is highly storage Fig.3.4, where mean depression

0.15

I
Ei U.IU

2 Slope(70)

Figure 3.4 Depression storage loss versus slope for four ra) impervious drainageareas.(From Viessman.

PROBLEMS

49

oo

Antecedent rainfall during preceding 30 min

o

Time (min)

Figure3.5 Depression storage intensity versus timefor animpervious area.(After Turner.l7)

easilyunderstood, sincea given depression will hold its maximum volumeif horizontally oriented. Using very limited data from a small, paved-streetsection, Turner devised the curvesshownin Fig. 3.5.17 Other sources of datarelatedto surface storase are available in the literature.2,18,1e

r Summary
Accountingfor the dispositionof precipitation is an important part of the hydrologic modelingprocess.Two abstractions from the precipitation input, intercepiion, and depression storagewere coveredin this chapter. Interception lossesduring the courseof a year may be substantial, but during intensestorms,they may be sufficiently small to neglect.Precipitationtype, rainfall intensity and duration, wind, and atmospheric conditions affeiting evaporationare factorsthat serveto determineinterceptionlosses for a given foresi standor ground coverconfiguration.Interceptionduring rainfall eventsis commonly greaterihan for snowfall events. Depressionstoragedeductions occur early in a storm sequence and they are a function of topography,ground cover, and extent and type of land development. During major storms,this loss is often considered to be negligible.

PROBI.EMS
3.1. UsingFig. 3.2, estimatethe volume of depression storage for a 3-acrepaveddrainage area' Statethe volume in cubic feet and cubic meters.Convert it to equivalentdep;h over the area in in. and cm.

50

CHAPTER3

STORAGE AND DEPRESSION INTERCEPTION

of the total volume of rainfall that is indicatedas depression 3.2. Estimatethe percentage in Fig. 3.2. storage annual precipitationfor your state,estimatethe annual amountof 3.3. Using the average interceptionloss. 3.4. Refer to Fig. 2.4 and estimate the annual interception lossesin lllinois, Florida, California, and New Mexico. How good do you think theseestimatesare?In which Why? In which of thesestateswould the do you havethe most confldence? estimates water budgetbe most affectedby interception? of rainfall that would be lost to depression 3.5. Using Fig. 3.4, estimatethe percentage for a l0-acre parking lot havinga mean slopeof 1 percent.Repeatfor a slope storage of 3 percent.Using the total rainfall volume determinedin Problem 3.2, estimatethe storageloss for both slopes.Stater equivalentdepth over the area of the depression in. in mm and depths 3.6. Refer to Fig. 3.3 and estimatethe ratio of overlandflow supplyto overlandflow and storage supplyif the areais turf, the OGEE summationcurve is the model, depression storageis (a) 75 percent and (b) 125 percent. of depression mean depth and the 3,7. Explain how a relation such as that given in Fig. 3.3 could be used in a simulation model of the rainfall-runoff process' 3.8. Using Eqs. 3.3 and 3.4, estimatethe throughfall in in. for 28 in. of rainfall during the growing season(21 events), and 17 in. of rainfall during the dormant season(13 events). 3.9. Using Horton's equationsgiven in Example 3.1, estimatethe interceptionlossesby ash and oak trees for a storm having a total precipitation of 1.33 in'

REFERENCES
"The Effect of Forestand Pastureon the Dispositionof Precipitation," 1. C. E. Schomaker, Maine Farm Res.(July 1966). 2. ven Te chow (ed.), Handbook of Apptied Hydrology. New York: McGraw-Hill,1964. New York: McGraw-Hill' 1948. 3 . JosephKittredge, ForestInfluences. "Interception of Rainfall by HerbaceousVegetation,"Science86(2243), i + . O. n. Ctark, "Resultsof the Mountain Home Rainfall Interceptionand Inflltration Project 5. J. S. Beard, on Black Wattle, 1953-1954," J. S. Afr. Foresty Assoc'27,72-85(1956)' "Rainfall Interception," Monthly WeatherRev.47,603-623(L9I9)' 6. R. E. Horton, "A Note on the InterceptionLoss Equation," J. Geoplrys. Res.65' 38507. R. A. Meriam, 1 ( 1 9 6 0 ) . 385 Council, on the Principles of Hydrology.National Research 8. D. M. Gray (ed.),Hand.book WaterInformation Center,Inc., 1973. Canada,Port Washington: and J. L. Thames,Hydrology and the 9. K. N. Brooks, P. F. Folliott, H. M. Gregersen, University Press/Ames,1991. State Iowa IA: Ames, of Watersheds. Management "The InterceptionProcess."In Prediction in Catchment Hydrology,National 10. G. J. Blake, Symposiumon Hydrology,eds.T. G. Chapmanand F. X. Dunin, MelbourneAust. Acad. S c i . , 5 9 - 8 11 , 975. "Throughfall in PlantedStandsof Fourth SouthernPines 11. F. A. Roth, II, and M. Chang, Bulletin 17' 880-885(1981) in EastTexas,"WaterResources Species

59r-s92(r937).

REFERENCES 51 "canopy and Litter Interceptionby Hardwoodsof Eastern J. D. Helvey and J. H. Patric, Res.l, 193-206(1965)' Resour. United States,"Water New York: R. K. Linsley, Jr., M. A' Kohler, and J. L. H. Paulhus,Apptied Hydrology' McGraw-Hill, 1949. "A Linear Model for synthesizingHydrographs for Small Drainage warren viessman,Jr., Geophysical American the of Meeting Annual Forty-eighth the presented at paper Areas," D.C., APr. 1967. Union, Washington, "The Hydrology of Urban Runoff," Trans. ASCE 125, A. L. Tholin una C. J. Kiefer, ( 960). 1 3 0 8 -1 3 7 91 ..A Method of Computing Urban Runoff ,', Trans' ASCE |09, I2L,7 W. I. Hicks,

Aue. 1966.

Chapter4

lnfiltration

Prologue
The purposeof this chapteris to: . Defineinfiltration. ' Indicatethe role infiltration playsin affectingrunoffquantities and in replenishing soil moistureand groundwaterstorages. ' Presentmodelsfor estimatinginflltration and provide examples of how they can be used. Infiltration is that processby which precipitation movesdownwardthrough the surface of the earth and replenishes soil moisture, rechargesaquifers,and ultimately supportsstreamflows during dry periods.Along with interception,depression storage, and stormperiodevaporation, it determines the availability,if any,of the precipitation input for generatingoverland flows (Fig. 1.3). Furthermore,infiltration rates influence the timing of overland flow inputs to channelizedsystems. Accordingly, infiltration is an important componentof any hydrologicmodel. The ratef at which infiltration occursis influencedby suchfactorsas the type and extent of vegetalcover, the condition of the surfacecrust, temperature,rainfall intensity,physicalpropertiesof the soil, and water quality. The rate at which wateris transmittedthrough the surface layeris highly dependent on the condition of the surface.For example,inwashof fine materialsmay seal the surfaceso that infiltration ratesare low evenwhen the underlyingsoils are highly permeable. After water crosses the surfaceinterface,its rate of downwardmovement is controlled by the transmissioncharacteristics of the underlying soil profile. The volume of storageavailablebelow ground is also a factor affecting infiltration rates. Considerable research on infiltration hastakenplace,but considering the infinite combinationsof soil and other factors existing in nature, no perfectly quantified generalrelation exists.

OF INFILTRATION 53 4.2 CALCULATION

4.1 MEASURING INFILTRATION Commonly usedmethodsfor determininginfiltration capacityare hydrographanalyasrainfall simulators Infiltrometersareusuallyclassified sesandinfiltrometer studies. a small test plot over simulated is arlificalrainfall former, In the or flooding devices. with considerrunoff, and ofrainfall observations from calculated and the inhltration are infiltrometers Flooding detention.l surface and storage ation given to depression at a maintained and applied is Water ground. in the inserted iubes usually rings or required' replenishment of rate the of made observations constantlevel and over havethe advantage Estimatesof infiltration basedon hydrographanalyses and precipitation of conditions prevailing to directly more relating infiltrometers of are runoff and rainfall which with precision the than no better are field. However,they rainfall. of variability areal is the studies such in importance partitular Of measured. Several meth;ds have been developedand are in use. Reference1 gives a good descriptionof thesemethods.

OF INFILTRATION 4.2 CALCULATION

from the applicationof reported averInfiltration calculationsvary in sophistication to the useof differential equations covers vegetal and types ageratesfor specificsoil media.For small urban areasthat porous unsaturated g6verningthe flow of wateiin warranted'On are sometimes precise methods more input, iespondiapidly to storm or average prolonged storms, from production peak flow to large waterlhedssubject adequate. may be values representative The infiltration pto"".r is -omplicated at best. Even under ideal conditions in (uniform soil propertiei andknown fluid properties),conditionsrarely encountered considerbeen has there Accordingly, to characterize. difflcult is practice,the process Most of theseeffortshaverelatedto the developabtestudyof the infiltration process. and (2) the solution of on field observations based equations ment of i1) empirical porous media.l'2 in flow saturated of mechanics equationsbasedon the are discussed. models infiltration used commonly several Later in this chapter, follows' process infiltration the of description brief a discussion, As a prefaceto that of the out some points and infiltration affecting factors principal It reviews the modelers. hydrologic by problemsencountered with an idealcase,onein which the soil is homogeneous Webeginour discussion the pores are directly interconnectedby capillary all profile and throughout the that the rainfall is uniformly distributedovef the assumed it is purru!"r. Furtheimore, may be chatactetized the infiltration process conditions, these ur"u of "on"ern. Under soil type and therefore are factors influencing major the as one dimensional and moisturecontent.3 through which the sizeand numberof the passages The soil type characterizes and relative potential capillary the sets content moisture the the watermustflow while capillary to due head hydraulic the is potential Capillary conductivity of the soil. sign. opposite with potential but capillary as same is the forces. capillary suction

54

CHAPTER4

INFILTRATION

9 aoo
Ei o

. 300 -.
d

2oo

0.1

0.2

0.3

0.4

0.5

Moisture content, 0 (vol/vol)

conducFigure 4.1 Typical capillary suction-relative (AfterMein andLarson.e.) tivity-moisture content relation. Capillary conductivity is the volume rate of flow of water through the soil under a gradient of unity (dependent on soil moisture content). Relative conductivity is the capillary conductivity for a specifiedmoisturecontent divided by the saturatedconductivity. Figure 4.1 illustratesthe relations amongthesevariables.Note that at low moisturecontents,capillary suctionis high while relative conductivityis low. At high moisture contentsthe reverseis true. With this background,an infiltration event can be examined.Consider that rainfall is occurringon an initially dry soil. As shownin Fig. 4.l,the relative conductivity is low at the outsetdue to the low soil moistureconditions.Thus, for the water to move downwardthrough the soil, a higher moisture level is needed.As moisture builds up, a wetting front forms with the moisturecontentbehindthe front beinghigh (essentially saturated) and that aheadof the front being low. At the wetting front, the capillary suctionis high due to the low moisture content aheadof the front. At the beginning of a rainfall event, the potential gradient that drives soil moisture movementis high because the wetting front is virtually at the soil surface. Initially, the infiltration capacity is higher than the rainfall rate and thus the and more water infiltration rate cannot exceedthe rainfall rate. As time advances entersthe soil, the wetting zone dimensionincreasesand the potential gradient is reduced.Infiltration capacitydecreases until it equalsthe rainfall rate. This occursat Figures4.2 and4.3 illustrate the time the soil at the land surfacebecomessaturated.

4.2

OF INFILTRATION CALCULATION

55

Moisturecontent.d

I
proflledevelopment with a constant rainfall Figure4.2 Typicalmoisture rate. these conditions. Figure 4.2 showshow a moisture profile might developwhen a rainstormofconstant intensityoccurs.In the diagramthe soil moistureat the surface value at the top is shownto rangefrom its initial value at the top left to its saturated right. Thus in moving downwardon the left-handside of the diagram,one can trace the downward progressionof the wetting front for varying levels of soil moisture

92

::

Time, r Figure 4.3 Infiltration rate versustime for a given rainfall (After Mein and Larson.e) intensity.

56

CHAPTER4

INFILTRATION

at the contentat the land surface. Figure 4.3 indicatesthat until saturationis reached surface, the infiltration rate is constantand equalto the rainfall applicationrate at the surface. At Point 4, apoint that coriesponds to the time at which saturationoccursat the surface,the infiltration rate beginsto proceedat its capacity rate, the maximum rate at which the soil can transmit water acrossits surface.As time goes on, the infiltration capacity continues to decline until it becomesequal to the saturated conductivity of the soil, the capillary conductivity when the soil is saturated.This ultimate infiltration rate is shown by the dashedline to the right of K" in Fig. 4.3. Of particular interest is the determinationof Point 4 on the curve of Fig. 4.3. This is the point at which runoff would beginfor the conditions specifiedabove.It is also the point at which the actual infiltration rate/becomes equal to the infiltration capacityratefo ratherthan the rainfall intensityrate i. The time of occurrenceof this point depends, for a given soil type, on the initial moisture content and the rainfall rate. The shapeof the infiltration curve after this point in time is also influencedby thesefactors. Another factor that must be reckonedwith in the infiltration processis that of hysteresis. In Fig. 4.1 it can be seenthat the plot of capillary suctionversussoil moistureis a loop. The curve is not the samefor wetting and drying of the soil. The curves shown on the figure are the boundary wetting and boundary drying curves, curves applicableunder conditions of continuouswetting or drying. Betweenthese curves, an infinite number of possiblepaths exist that dependon the wetting and problemhavebeen to the hysteresis drying history of the soil. A numberof approaches reportedin the literature.3 was basedon an ideal soil. The illustration of the infiltration processpresented Natural soils are Unfortunately,suchconditionsare not replicatedin natural systems. highly variable in composition within regions and soil cover conditions are also far-ranging. Becauseof this, no simpleinfiltration model can accuratelyportray all the conditionsencountered in the fleld. The searchhas thus beenfor modelsthat can of the rates at which infiltration occurs be called upon to give acceptable estimates durine rainfall events. Mein and Larson have describedthree generalcasesof infiltration associated with rainfall.3The first caseis one in which the rainfall rate is lessthan the saturated conductivity of the soil. Under this condition, shownas (4) in Fig. 4.4, runoff never this condition occurs since all the rainfall infiltrates the soil surface.Nevertheless, sincethe level of soil moisture mustbe recognized in continuous simulationprocesses is affectedeven though runoff doesnot occur. The secondcaseis one in which the rainfall rate exceeds the saturated conductivity but is lessthan the infiltration capacity. Curves(I), (2), and (3) of Fig. 4.4 illustratethis condition.It shouldbe observed that the period from the beginningof rainfall to the time of surfacesaturationvaries with the rainfall intensity.The final caseis one in which the rainfall intensityexceeds the infiltration capacity.This condition is illustratedby the infiltration capacitycurve of Fig. 4.5 andthoseportionsof infiltration curves(l), (2), and (3) of Fig. 4.4 that are in their declining stages. Only under this condition can runoff occur. All three to hydrologicmodeling,particularly when it is continuousover caseshaverelevance time.

4.3

MODEL HORTON'SINFILTRATION

57

tsatt

tsal, Time, t

tsat,

Figure 4.4 Inflltration curves for several rainfall intensities.(After Mein and Larson.e)

fo: f"+ Ao-f")"u'

Infiltration capacitY curve

f, 0

,r*"

Figure 4.5 Horton's infiltration curve and hyetograph.

MODEL INFILTRATION .I3 HORTON'S

An The inflltration processwas thoroughly studiedby Horton in the early 1930s.o for relation following the outgrowth of his work, shown graphically in Fig. 4.1, was determininginfiltration capacity: (+.r; fo: f, + ("fr f")e'n'

58

CHAPTER4

INFILTRATION

where fo : k : f" : ,fo :

the infiltration capacity(depth/time) at sometime / inf capacity the rate of decrease a constantrepresenting d final or equilibrium capacity the initial infiltration capacity

It indicatesthat if the rainfall supply exceedsthe infiltration capacity,infiltration in an exponentialmanner.Although simplein form, difficulties in tendsto decrease determininguseful valuesfor/. and fr restrict the useofthis equation.The areaunder the depth of water infiltrated during that the curve for any time interval represents interval. The infiltration rate is usually given in inches per hour and the time r in minutes,althoughother time incrementsare usedand the coefficientk is determined accordingly. By observingthe variation of inflltration with time and developingplots of / / asshownin Fig. 4.5,we canestimate versus fsandft. Two setsof/and / are selected havingtwo unknownsare thus from the curve and enteredin Eq. 4.1. Two equations forfi and k. approximations obtained;they can be solvedby successive Typical infiltration ratesat the end of t hr ( f) areshownin Table4.1. A typical relation betweenf, and the infiltration rate throughout a rainfall period is shown graphically in Fig. 4.6a; Fig.4.6h showsan infiltration capacity curve for normal conditionson turf. The data given in Table4.1 are for aturf areaand must antecedent A range be multiplied by a suitablecover factor for other types of cover complexes. of cover factorsis listed inTable 4.2. Total volumes of inflltration and other abstractionsfrom a given recorded (plot of the streamflow rate versus hydrograph rainfall are obtainablefrom a discharge time) if one is available.Separationof the base flow (dry weather flow) from the for hydrograph resultsin a direct runoff hydrograph(DRH), which accounts discharge Direct surfacerunoff or the direct surfacerunoff. that is. rainfall less abstractions. can readily be precipitation excess in inchesuniformly distributedover a watershed at equal time incrementsthrough the calculatedby picking valuesof DRH discharge hydrograph and applyingthe formulas P": where P" : 4r : A : r?7:

(0.0371e)() q')
Ano

( a) \

(in.) precipitation excess DRH ordinatesat equal time intervals (cfs) area(mi2) drainage hurrber of time intervalsin a 24-hr period

For most cases the differencebetweenthe original rainfall and the direct runoff can be consideredas infiltrated water. Exceptionsmay occur in areasof excessive drainageor tracts of intensiveinterceptionpotential.The calculatedvalue subsurface asdistributedaccordingto an equationof the form of infiltration can then be assumed of Eq. 4.1 or it may be uniformly spreadover the stormperiod. Choiceof the method on the accuracyrequirementsand size of the watershed. employeddepends To circumvent some of the problems associatedwith the use of Horton's Fig. 4.5. Notethat where Consider canbe made.6 infiltration model,someadjustments the actual rate of infiltration the infiltration capacitycurve is abovethe hyetograph,

MODEL INFILTRATION 4.3 HORTON'S

59

o - 1

1.0 Time (hr)

(a)

3.0 2.8
2.O

-TT
-

2.4 2.2 2.0 1.8 1.6 t.4 7.2 1.0 0.8 0.6 0.4 0.2 0

u.)9( l l, = U.UUUUJTI

,4 f=0.53+2.4

0691ty

l
e

l
fle
rOa"

9l rras!
o\

r\$4"

-iro$dts"
Infiltration capacity curve (/.

10

70 80

100

t20

140

160

t80

Time (min) from beginning of infiltration capacity cuve' t/ (b)

Figure 4.6 (a) Typical infiltration curve. (b) Infiltration capacity and mass conditionsof turf areas.fAfter A. L. Tholin and curvesfor normal antecedent "The Hydrology of Urban Runoff," Proc. ASCE J. Sanitary Clint J. Kiefer, Ens. Div. S4(SA2),56 (Mar. 1959).1

60

CHAPTER4

INFILTRATION

TABLE 4.1 ryPICAL f, VALUES

Soilgroup
High (sandysoils) Intermediate(loaps, clay, silt) Low (clays,clay loam)

f, (in./hr)

f' (mm/h)

0.50-1.00 0.10-0.50 0.01-0.10

t2.50-25.00 2.so-r2.50 0.25-2.50

Source: After ASCE Manual of Engineering Practice,No.28.

TABLE4,2 COVERFACTORS Cover

Permanentforest and grass Cover fac{or

Good(1 in. humus) Medium(f-1 in. humus; Poor(< j in. humus)

Good Medium Poor Good Medium Poor

Close-growingcrops

Row crops

3.0-7.5 2.0-3.0 1.2-1.4 2.5-3.0 t.6-2.0 1.11 - .3 1.31.5 1.1-1 3 1.01.1

Source: After ASCE Manual of Engineering Practice, No.2t.

is equal to that of the rainfall intensity, adjustedfor interception,evaporation,and other losses. Consequently, the actuafinfiltration is given by

f(t) : minlfof), i(t)l

(4.3)

where/(r) is the actual infiltration into the soil and i(l) is the rainfall intensity.Thus the infiltration rate at any time is equal to the lesserof the infiltration capacity, f,(t) or the rainfall intensity. Commonly,the typical valuesof foandf" are greaterthan the prevailingrainfall intensitiesduring a storm. Thus, when Eq.4.l is solvedforS as a function of time alone, it shpwsa decrease in infiltration capacity even when rainfall intensitiesare much lessthanfo. Accordingly,a reductionin infiltration capacityis maderegardless of the amount of water that entersthe soil. To adjust for this deficiency,the integratedform of Horton's equationmay be used,

: F(tp)

lr'' ,

o, : f,tp +

(t - a' *,,1

(4.4)

whereF is the cumulativeinfiltration at time to, as shownin Fig. 4.7.rnthe figure, it is assumed that the actual infiltration has been equal to$. As previouslynoted, this is not usually the case,and the tnie cumulativeinfiltration must be determined. This

MODEL INFILTRATION 4.3 HORTON'S

61

0
0

tptpt tl Equivalent time

infiltration. Figure4.7 Cumulative can be done using

F\t) :

o, ,u, l,'

(4.s)

usingF,q.4.3. where/(r) is determined Equations4.4 and4.5 may be us6djointly to calculatethe time t, that is, the equivalent time for the actual infiltrated volume to equal the volume under the infiltration capacity curve (Fig. 4.7). The actual accumulatedinfiltration given by Eq. 4.5 is equatedto the area under the Horton curve, F,q. 4.4, and the resulting is solvedfor r' This equation, expression
F:fJrt (l-e-k'n)

(4.6)

It should explicitly for to,but an iterativesolutioncan be obtained. cannotbe solved time r. Thus that the time to is lessthan or equal to the actual elapsed be understood that given as shownin Fig. 4.7 is equalto or exceeds infiltration capacity the available a functionof the actual described,fbecomes by Eq. 4.1. By makingthe adjustments in the amount of water infiltrated and not just a variable with time as is assumed original Horton equation. In selectinga model for use in inflltration calculations,it is important to know shortcomings; a model can be adjustedto accommodate its limitations. In somecases are not realisticfor the natureof the useproposed, if its assumptions in othercases, the model shouldbe discardedin favor of anotherthat better fits the situation. Part One of this book deals with the principal componentsof the hydrologic cycle. In later chapters,the emphasisis on putting these componentstogetherin

62

CHAPTER4

INFILTRATION

for continuWhen thesemodelsare designed varioushydrolqgicmodelingprocesses. of the hydrois to calculatethe appropriatecomponents ous simulation,the approach of how infiltration could overtime. A discussion logic equation,Eq. 1.4, continuously be incorporatedinto a simulation model follows. It exemplifiesthe use of Horton's model (SWMM)." equationin a storm water management First, an initial value of /o is determined.Then, consideringthat the value of$ depends on the actualamountof infiltration that hasoccurredup to that time, a value infiltration capacity,fo, availableover the next time stepis calculated of the average using

'o f o : * 1 " = ' *fo ' ':lW

-tp

(4.7)

rate of infiltration, /. Equation 4.3 is then usedto find the average

'v - [f, tt

iti >1, iri <f,

(4.8)

rainfall intensity over the time step. where i is the average Following this, infiltration is incrementedusing the expression F(t + Lt) : F(t) * AF : F(t) + f Lt

(4.e)

infiltration (Fig.4.7). whereAF : f Lt is the addedcumulative The next stepis to find a new valueof ro.This is doneusingEq. 4.6.If LF : Fig. 4.7),Eq. 4.6 must f o Lt, tp : to t Ar. But if the new /oris lessthanto + A/ (see using the be solvedby iteration for the new value of lo. This can be accomplished procedure.6 Newton-Raphson horizontal and When the valueof tp > I6f k, the Horton curve is approximately : iteration since need for is no further point reached, there this has been Once f". fo on F. dependent longer and equal to and no is constant f" f
EXAMPLE 4.I

ft of 0.29hr-', Givenan initial inflltrationcapacityfiof 3.0 in./hr anda time constant capacity is infiltration if the ultimate derive an infiltration capacity vs. time curve in infiltrated of water total volume 0.55 in./hr. For the first ten hours, estimatethe inchesover the watershed. Solution. Using Horton's equation(4.1), valuesof infiltration can be computed for various times. The equationis as follows: f : f" + Lfo + f")e-k, Substitutingthe appropriatevaluesinto the equationyields ,f = 0.55 + (3.0 0.55)s-o'zo' Table4.3, valuesof f atecomputedand Then for the times shownin spreadsheet graphics package,the curve of enteredinto the table. Using the spreadsheet Fig. 4.8 is derived,

MODEL 4,3 HORTON'SINFILTRATION FOR EXAMPLE4.1 TABLE 4.3 CALCULATIONS

63

Time (hr) 0.00 0.10 0.25 0.50 1.00 2.00 3.00 4.00

lnfiltration (in./hr) 3.00 2.93 2.83 2.67 2.38 t.92

1.58

Time (hr) 5.00 6.00 7.00 8.00 9.00 10.00 15.00 20.00

lnfiltration (in./h0 t.l2

0.98

0.87 0.79 0.73

0.68

0.58 0.56

r.32

4.1 can To find the volumeof waterinfiltrated duringthe flrst 10hours,Eq' be integratedover the range of 0-10 v : Y: I J [0.5s + (3.0 0.55)e-o2e' ldt

[0.55/ + (2.45I-0.29)e-o2s\o : 12.47in' V ll

is thus 12'47 in' The volume in inchesover the watershed

3.0

\ ;
l? 6 .:

2.0

\ -\-

15

1.0

"'-0 0.5

L
8

10

12

t4

18

Time (hr) Figure 4.8 GraPh for ExamPle 4'1.

64 4.4

CHAPTER4

INFILTRATION

MODEL GREEN-AMPT
in 1911,has had a resurThe Green-Ampt infiltration model,originallyproposed gence on Darcy's law (seeChapter18).In its This approach is based of inte1gs1.3'6-1t of original form, it was intendedfor use where infiltration resultedfrom an excess water at the ground surfaceat all times. In 1973, Mein and Larson presenteda methodologyfor applying the Green-Ampt model to a steadyrainfall input.eThey the valueof the capillary suctionparamealsodeveloped a procedurefor determining the applicability of the model for ter usedin the model. In 1978,Chu demonstrated useunder conditionsof unsteadyrainfall.lo As a result of theseand other efforts, the Green-Ampt model is now employedas an option in such widely used continuous simulation modelsas SWMM.6 that the soil surfacewas The original formulation by Greenand Ampt assumed coveredby ponded water of negligible depth and that the water infiltrated a deep homogenous soil with a uniform initial watercontent(seeFig. 4.9). Wateris assumed to enter the soil so as to define sharply a wetting front separatingthe wetted and unwettedregions as shown in the figure. If the conductivity in the wetted zone is definedas K", applicationof Darcy's law yields the equation Io:

"

K"(r + s)
r

(4.10)

where I is the distancefrom the ground surfaceto the wetting front and S is the capillary suction at the wetting front. Referringto Fig. 4.9, it can be seenthat the cumulativeinfiltration F is equivalentto the product of the depth to the wetting front L and the initial moisture deficit, 0, - 0, : IMD. Making these substitutionsin

Ponded depth considerednegligible

I
Ho

I
I

I I
Figure 4.9 Definition sketch for GreenAmpt model.

MODEL 4.4 GREEN-AMPT

65

we obtain Eq. 4.10 and rearranging,

K.( - , \ -, f- : rp

t * jt' )
F /

(4.11)

Considering thatfp : dFldt. we can state

#: *,(t.'#)
: 0 at t : 0' we obtain Integratingand substitutingthe conditionsthat F F - S x I M D X l o g " ({ IMD

(4.r2)

+ I MD X {) : K.t
X

(4.r3)

for usein watershed This form of the Green-Ampt equationis more convenient infiltrationto the cumulative the processes than Eq. 4.l}beciuse it relates modeling a ponded assumes equation of this time at which infiltration began.The derivation at all capacity infiltration to the equal surfaceso that the actualrate of infiltration is a time, any at infiltration cumulative the times.Using Eq. 4.13, we can determine equation the in parameters the All modeling. featuredesiiablefor continuoussystems The determinaare physicalpropertiesof the soil-water systemand are measurable. particuhowever, difficult, is often s tion of suitablevaluesfor the capillary suction It can be 4.10. Fig. in soil larly for relationssuchas that ihown for a clay-type capillary of variation wide is a there observedfrom the figure that for this curve content.3 with soil moisture suction The Mein-Larson formulation using the Green-Ampt model incorporatestwo The first stagedealswith prediction of the volume of water that infiltrates stages.3,6 before the surfacebecomessaturated.The secondstageis one in which infiltration usingthe Green-Ampt equation.In the widely usedstormwater capacityis calculated of the ,nunug"*"nt model, the irodif,ed Green-Ampt model of infiltration is one using made are Computations optiois that can be employedto estimateinfiltration.6

v)

Moisture content,0

Figure 4.10 Capillary suction versus moisture contentcurves.

66

CHAPTER4

INFILTRATION

the following equations:for F ( F,(f : i),

F. :7:

IM?

i/K" - |

rori > K"

\4.r4)

and thereis no calculationof F"for i < K"; for F > F,(f : fi):

*\ ' f,: x,(t

where f : fo: I : F : { :

t * j") F

(4.11)

acttal infiltration rate (ftlsec) infiltration capacity(ftlsec) rainfall intensity (ftlsec) cumulativeinfiltration volume in the event (ft) cumulative infiltration volume required to causesurfacesaturation (f0 S : average capillary suctionat the wetting front (ft of water) IMD : initial moisture deficit for the event (ftlft) K" : saturatedhydraulic conductivity of soil (ftlsec)

Equation 4.10 showsthat the volume of rainfall neededto saturatethe surface is a function of the rainfall intensity.In the modelingprocess,for eachtime stepfor which I ) K", the value of d is computedand comparedwith the volumeof rainfall infiltrated to that time. If F equalsor exceeds saturates and calculations {, the surface for infiltrationthenproceed usingEq.4.I4. Notethat by substituting/for i in Eq.4.l4 and rearranging, the equationtakesthe sameform as Eq. 4.11. For rainfall intensitieslessthan or equal to K", all the rainfall infiltrates and its amount is used olly to update the initial moisture deficit, IMD.6 The cumulative infiltration volume {" is not altered. After saturation is achieved at the surface,Bq.4.l1 showsthat the infiltration capacityis a function of the infiltrated volume,and thus of the infiltration ratesduring previous time steps.To avoid making numerical errors over long time steps,the integratedform of the Green-Ampt equation(Eq. .B) is used.This equationtakes the following form as it is usedin SWMM: (4.15) &Gr- tr): Fz- Cln (F2+ C) - Fr* Cln(F1+ C) where C : IMD X .t (ft of water) / : times (sec) 1,2 : subscripts indicating the starting and ending of the time steps. Equation4.15 mustbe solvediteratively for F2,the cumulativeinfiltration at the end of the time step.A Newton-Raphsonroutine is used.6 In the SWMM model, infiltration during time step tz - tt is equal to (t, - tr)i if the surfaceis not saturatedand is equal to F, - F, if saturationhas previously occurred and there is a sufficient water supply at the surface.If saturationoccurs during an interval, the infiltrated volumesover each stageof the processwithin the time stepsare computedand summed.When the rainfall endsor becomeslessthan

MODEL 4.5 HUGGINS_MONK

67

theinfiltrationcapacity,anypondedwaterisallowedtoinfiltrateandisaddedtothe cumulative inflltration volume'

MODEL 4.5 HUGGINS-MONKE by introducing problem thetimedependency circumvented have investigators Several by proposed equation the following as the depenO"niuutluUfe.2'10-13 soilmoisture andMonkeis an examPle:2 Huggins

f : f,*A(
where AandP:

(4.16)

coefficients layer (Q ,,orug" potential of a soil overlying the impeding moisture) minus antecedent F _ the total volume of water that infiltrates impeding stratum T o = the total porosity of soil lying over the The using data from sprinkling infiltrometer studies' The coefficientsare determined At the process' for eachtime incrementin the iteration variableF mustbe catculated

s - itr"

exceedsthe inflltration capacity,the rate zone," which determinesthe soil moistu (1) wherethe moist asfollows:10 evaluated the field capacity(amountof waterheld in z drained),the drainagerate is considered s is soil to the infiltration rate when the constan| and (3) if the watercontentis be drainagerate is comPutedas

rate: ,"(t - 2)' drainage

(4.r7)

pore volume where P, : the unsaturated G:maximumgravitationalwater,thatis,thetotalporosityminusthefield caPacitY Datafromsprinklinginfiltrometerstudiesofvariouswatershedsofinterestare in Eq' 4'16'' the coefficients usedto estimate

5.6

EVAPOTRANSPIRATION

101

of a masstransferequationthat has often been employedfor this purpose.However, Linsley and co-workers indicate that there is some question as to the adequate The equation is losses.27 verification of this model to estimate evapotranspiration expressed as ^ E : @ 833x2(e1 er)(V, V,)

(s.2s)

E : K: 1,2 : Vr, Vz : Z =

(in./hr) evaporation (0.4) von K6rm6n's constant (in. vaporpressures Hg) (mph) wind speeds the mean temperature("F) of the layer betweenthe lower level zt and the upper level z2

is adiabatic and the wind speedand in Eq. 5.25 that the atmosphere It is assumed moistureare distributedlogarithmically in a vertical direction. In view of the sm4ll to be expectedat two levelsso closely differencesbetweenwind and vapor pressure evaporation, and sincethesegradientsare directly relatedto the sought-after spaced, highly exactinginstrumentationis required to get reliable results.

PotentialEvapotranspiration
as"the waterlosswhich will occur Thornthwaitedeflnedpotential evapotranspiration if at no time there is a deficiencyof water in the soil for the use of vegetation."In a that potential evapotranhave assumed practical sense, however,most investigators spiration is equal to lake evaporationas determinedfrom National WeatherService the albedo(amountof ClassA pan records.This is not theoreticallycorrect because of vegetatedareas and soils incoming radiation reflectedback to the atmosphere) should be As a result, potential evapotranspiration rangesas high as 45 percent.28 somewhatlessthan free water surfaceevaporation.Errors in estimatingfree water from pan recordsare such,however,asto make an adjustmentfor evapotranspiration value. potential evapotranspiration of questionable by the Agrideveloped An equationfor estimatingpotential evapotranspiration Service (ARS) illustratesefforts to include vegetalcharacteristics cultural Research potentialfor any given and soil moisturein sucha calculation.The evapotranspiration day is determinedas follows:2e

E T : G I x ku x, " ( # l
where

(s.26)

potential (in./day) ET : evapotranspiration : of maturity growth index of crop in percentage GI usually 1.0-I.2 for short grasses' K : ratio of G1 to pan evaporation, height,and 1'6-2.0 for forest 1,.2-L6 for cropsup to shoulder (in./day) Eo: pan evaporation ,S: total porosity SA : availableporosity (unfilled by water) AWC : porosity drainableonly by evapotranspiration x : AWC/G (G : moisture freely drainedby gravity)

102

CHAPTER5

EVAPORATION AND TRANSPIRATION

0.2 0.1 ,.} 0.0

ti

>

n ,

0.1

Months (b)

Figure 5.5 Averagedaily consumptionof water: (a) for year 1953 by corn, followed by winter wheat under irrigation; (b) for year 1955, with irrigated first-year freadow of alfalfa, red clover, and timothy. Both measurements taken on lysimeterY 102 C at the Soil and Water ConservationResearchStation. Coshocton.Ohio. (After Holtan et al.2e)

l x F rrllS
N I

]l{

tl rH

'

Weeks

Figure 5.6 Growth index GI = ETfET^,, from lysimeter records, irrigated corn,andhayfor 1955,from Coshocton, Ohio.(AfterHoltanet al.2e)

5.7

ESTIMATINGEVAPOTRANSPIRATION 103

OF SOIL TEXTURECLASSES CAPACITIES TABLE 5.5 HYDROLOGIC

class Texture
Coarsesand Coarsesandy loam Sand Loamy sand Loamy fine sand Sandyloam Fine sandy loam Very fine sandy loam Loam Silt loam Sandy clay loam Clay loam Silty clay loam Sandy clay Silty clay Clay

(v")
24.4 24.5 32.3 37.0 32.6 30.9 36.6 32.7 30.0
JI,J

sa

GD

(Y") t'7.7 15.8 19.0 26.9 27.2 18.6 23.5 21.0 14.4 11.4 13.4 13.0 8.4 rl.6
o 1

AWC" ('/") 6.7 8.7 13.3 10.1 5.4 t2.3 13.1 11.7 15.6 19.9 tr.9 12.7 t4.9 7.8 12.3 11 . 5

x AWC/G 0.38 0.55 0.70 0.38 0.20 0.66 0.56 0.56 1.08 1.74 0.89 0.98 r.77 0.6'l r.34 r.58

25.3 25.7
z5-J

19.4
z l.+

18.8

7.3

aS = total porosity - 15 bar moisture 7o' bG : total porosity - 0.3 bar moisture 7o. "AWC: S - G. "Land Capability: A Hydrologic Response Unit in Source: Adaptedfrom C. B. England, U.SlDepartmentof Agriculture,ARS 41-172' Sept' 1970' Asiicultural'Watersheds," Aiter H. N. Holtan et a1.2e

The GI curveshave been develoPed for severalcrops (Fig. : evapotranspiration (Fig. Equation5.26is used 5.6). daily rate hydrolq model of watershed USDAHL-74 Represental evapotranspiration. late daily Table5.5.

EVAPOTRANSPIRATION 5.7 ESTIMATING

areas, Transpirationis an important componentin the hydrologicbudgetof vegetated on phytological but it is a difficult quantity to measurebecauseof its dependence and soil moisture soil plants, of types and number of the is a function It variables. previously, temperature,and averageannual precipitation. As noted type, qeason, in their combinedevapotrantianspiration are commonly estimated and evaporation form. spiration ' If the precipitation and net runoff for an atea ate known, and estimatesof of ET canbe had using groundwateiflow and storagecan be made,rough estimatesby approachdeveloped ihe basic hydrologicequatiJn,Eq. 1.1. A more sophisticated peaman foilows.t3It iJ representative of the methodsmost often used'

104

CHAPTER 5 EVAPORATION ANDTRANSPIRATION

The PenmanMethod
Both the energybudgetand masstransportmethodsfor estimatingevapotranspiration (ET)have limitations dueto the difficulties encountered in estimatingparameters and in making other required assumptions. To circumventsomeof theseproblems,penman developed a method to combinethe masstransport and energybudgettheories. This widely usedmethodis one of the more reliableapproaches to estimatingETrates 13'rs'23,30 usingclimaticdata. The Penmanequationis of the form of Eq. 5.18; it is theoretically basedand showsthat EZ is directly related to the quantity of radiative energy gained by the exposedsurface.In its simplified form, the Penmanequationisls
t s t : -

LH + 0.27E L + 0.21

(s.27)

where A : the slopeof the saturatedvapor pressurecurve of air at absolute temperature(mm Hg/'F) H : the daily heatbudgetat the surface (estimate of net radiation) (mm/day) E : daily evapoiation(mm) ET : the evapotranspiration or consumptive usefor a given period (mm/day) The variablesE and.Fl are calculatedusing the following equations:

( s .28) 0.0098ar) : where eo the saturationvapor pressureat mean ak temperature(mm Hg) e6 : the saturationvapor pressure at meandew point (actualvapor pressure in the air) (mm Hg) u2 : the mean wind speedat 2 m abovethe ground (mi/day)
The equationusedto determinethe daily heat budget at the surface,11,is 11 : R(1 - r)(0.18 + 0.55.t)- 8(0.56 * 0.092e2s)(0.10 (5.29) + 0.905) where R : the mean monthly extraterrestrialradiation (mm HrO evaporated per dav) : the estimatedpercentage of reflecting surface B = a temperature-dependent coefficient s : the estimatedratio of actual duration of brisht sunshineto maximum possibleduration of bright sunshine. The empirical reflectivecoefficientr is a function of the time of year,the calmness of the water surface,wind velocity, and water quality. Typical rangesfor r are 0.05 to 0.12.31 valuesof e" andA can be obtained from Figs.5.7 and 5.8, thosefor R and B can be obtainedfrom Tables5.6 and 5.7. The use of Penman's equationrequiresa knowledge of vaporpressures, sunshine duration,net radiation,wind speed, and mean temperature. Unfortunately,regular measurements of theseparameters are often unavailableat sites of concern and they must be estimated.Another complication is making a reductionin the valueof EZwhen the calculationsare for vegetated surfaces. While results of experimentsto quantify reduction factors have not completelyresolved the problem, there is evidencethat the annual reduction factor is close to

E : 0.35(e"- e)(l'+

5.7
"C 60 50
I

EVAPOTRANSPIRATION ESTIMATING
.F
111

105

.F t40 't22 104 86 68 50 10 20 30 40 50 60 70 80 90 100

!?

"C 50 40

r04 86
68

313 =
ts

: o
o B t ^

303
ro? z6J | | 0.8 | I I I t273

ct
!d 6 6 F.

+ o
o

o F

t r 1 n

- 1 0
14

20 10

s0l. zzV |
0.2

0.4 0 6

1.0 1.2 1.4 1.6 1.8

satuated vapor pressure' ea (mm Hg)

Figure 5.7 Relation betweentemperatureand saturatedvapor Pressure.

of A (mmHg/'F) Value Figure 5.8 TemperatureversusA relation for use *i?rt ,n" PenmanLquation. (After Criddle'23)

that using to supportanothervalue'it appears thereis evidence Thus,unless unity.32-34 surfaces for results satisfactory a value of 1 for the reduction coefficient may give evaporation water free of having varied vegetal covers.Accordingly, aty estimate by an appropriatereduction could be used ro estimaieEZ, providin[ it is modified coefficient.

EXAMPLE 5.4

to 5.29, estimateET, giventhe following data: ': using the PenmanMethod, Eqs.5.21 c' c, temperatureof 1l : 30 degrees 20 degrees temperatureat warer r"tru"" is month : (48 mi/day)' the relative humidity : +O p"r""nt, wind-velocity i mph 0'75' north, r is given ut 0'07' and S is found to be Juneat latitude 30 degrees Solution l.Giventhedatafortemperature,thevaluesofeoandeacanbedetermined. UsingFig.5.TorAppendixTableA.2,thesaturatedVaporpressuresare 31'83'andfor foundto be l7.53unO-:t'Sl mm Hg respectively' ThI'"-: : : 12'73' X 0'4 a relativehumidity of 40 percent,e,t 31'83 Then,usingEq. 5.28' + 0.0098x 48) .83 - 12.73)(1 E : 0.35(31 E:9.83 mm/daY 2.ThevalueofAisfoundusingFig.5.8;forthegivenlatitudeandmonth,R isobtainedfiomTable5.6;andBisgottenfromTable5.'Tforatemperature : I7 '01' found are A : 1'0' R : 16'5' andB of 30"C.fne vatues Then,usingEq.5.29, + 0's5 x 0'75) H: 16'5(I - 0'07)(0'18 - 17.01(0.5 0.092x 12j30)(0.10 + 0'90 x 0'75) 6 H = 6.04 mm/daY

O \ O h O \ 6 \ t o O \ \ O r ) O o r) F. O N + r) \O F-

t--

l--

F-

\O F- :+ f.-' r) O\ O F- * C.l * d \ + \ O O \ i o * r ) \ O \ O \ O \ O n

Ol

o
LL 6

ft

o
IL

trJ

t- * co o o\ * o co cn oo oo n tF-O\dcl\fh!nv1<.cqolO

F-

f U)
J

tr

oN catf,r +s+coNooo\o
: : * * * d d i * -

nv?qnnqqnv.t-nv.tol

z o o
I
tf

F.

O\

cn ci

e.l O

c.i ri + r; ri ri {

cd -.j oi F ri c.i

\n

\O h

O\

I = o
E.

r.)

\O \O \O r.)
* * i

-.: dl .l \

!f
*

oq v? q g oq 9 q cl
c.l r O f\ \A tri d
d

(r o q)
lJ-

r F- tr - \r ) $\O\O**6 \o \o \o \o r) * c7) d o\ F- r) N o

o 6 z
F

9nqq\qqn\\-q*:
$ h n \ o r ) h c n N o \ o $ *
* i i i i * i * d

uJ

z
J T

-j 6i di + ri ri + ri 6i ci *, <j +

F-

O\ oo a]

a-

t1

oo oo \O o

o
4 Z tIO

\ O O \ : c . l o $ n h * o a l ( ) H * * * * r *

e-:q\qoq.l-9qqvln

>tr
u O olu 6 Z ui< f > 1 z
t=P

ds

n q q n r t

q q \ g o q q - \

O r} oO O al cn r) r) \O r) r)

<+ C.l

S IIJ 2O-

cn\oor)66r)ooooo6*\o .j d; \o od o 6i + r; \ci i-- r- r- \c; i : :

ft+
) z
muJ 4 ,Ct) F ] q lo UJ 6

s
z

\Or ) *6al -

- 6l o+r ) \O
E V

E:
a

> 6

SUMMARY

107

TABLE5.7 VALUES OF TEMPERATURE-DEPENDENT B FORUSEINTHEPENMAN COEFFICIENT EQUATION

Tu (K)

B (mmHrO/day)

T" fF)
Jf

B (mm HrO/day)

270 275 280 285 290 295 300 305 310 315 320 325

r0.73
11.51 12.40 13.20 14.26 15.30 16.34 r7.46 18.60 19.85 21.15 22.50

40
A<

50 55 60 65 70 75 80 85 90 95 100

11.48 I 1.96 12.45 12.94 13.45 13.96 14.52 15.10 15.65 16.25 16.85 t'7.46 18.10 18.80

Source: Afler Criddle.23Note that B = oT1 where o is the Boltzmann c o n s t a n t . 2 . 0x l l0-e mm/day.

3. UsingEq. 5.27, Er : 0.0 x 6.04 + 0.27 x 9.83)/(1+ 0.27) ET : 6.85 mm/day is 6.85 mm/day. ll Thus the estimated evapotranspiration

Simulating Evapotranspiration
The volume of water evaporatedor transpiredfrom a watershedover time can be shouldincorpoAccordingly, continuoushydrologicmodelingprocesses substantial. rate an EZcomponent.The modelsgiven in this chaptertypify suchan approach(see aboundon this subjes1.28'2e'rs':o':z alsothe flow chart of Fig. 1.3).References

Summary
Figure 1.1 and Table5.1 showthe overall importanceof ET in the hydrologicbudget. annual precipitation by a In many regionsof the United States,annual ET exceeds and use must development significantamount.As a result, plans for water resources incorporate estimatesof ET losses.Where irrigated agriculture is practiced, these are especiallyimportant. estimates They generally to estimatingEThavebeendeveloped. A numberof approaches analyttheoretical,basedon the physicsofthe process; fall into the following classes: Equaical, basedon energyor water budgets;and empirical, basedon observations. are usually of the type illustrated by Eqs. 5.1, tions usedin making ET calculations 5.8,5.10,5.19,5.22,and5.26.

1oB

cHAp+R 5 EVApoRATtoN ANDTRANsptRAloN

PROBLEMS
5.1. An 8000-mi2watershed received20 in. of precipitationin a 1-yearperiod. The annual streamflowwas recorded as 5000 cfs. Roughly estimatethe combined amounts of water evaporated and transpired.Qualify your answer. 5.2. Find the daily evaporation from a lake during which the following datawere obtained: air temperature 90oF,water temperature60oF,wind speed 20 mph, and relative humidity 30 percent. 5.3. Find the monthly consumptiveuse of an alfalfa crop when the mean temperatureis 70"F, the average percentageof daytime hours for the year is 10, and the monthly consumptiveuse coefficientis 0.87. 5.4. During a given month a lake having a surfacearea of 350 acreshas an inflow of 20 cfs, an outflow of 18 cfs, and a total seepage lossof 1 in. The total monthly precipitation is 1.5 in. and the evaporationloss is 4.0 in. Estimatethe changein storage. 5.5. What are two filethodsthat might be usedto reduceevaporationfrom a small pond? 5.6. Computethe daily evaporationfrom a ClassA pan if the amountsof water required to bring the level to the fixed point are as follows:

Day Rainfall (in.) Water added(in.) Evaporation

1 0 0.29

2 0.65 0.55

3 0.12 0.07

4 0 0.28

5 0.01 0.10

5.7. For Problem 5.6, the pan coefficientis 0.70. What is the lake evaporation(in inches) for the 5-day period for a lake with a 250-aqe surfacearea? 5.8. The pan coefficientfor a ClassA evaporationpan locatednear a lake is 0.7. A total of 0.50 in. of rain fell during a given day. Determinethe depth of evaporationfrom the lake during the sameday if 0.3 in. of water had to be addedto the pan at the end of the day in order to restorethe waterlevel to its original valueat the beginningof the day. 5.9. A 2500 mi2drainagebasinreceives25 in.lyr rainfall. The discharge of the river at the basinoutlet is measured at an average of 650 cfs. Assumingthat the change in storage for the system is essentiallyzero, estimatethe EZlossesfor the areain inchesand cm for the year. Stateyour assumptions. 5.10. Determine the daily evaporationfrom a lake for a day during which the following mean values were obtained: air temperature78'F; water temperature62oF; wind speed,8 mph; and relative humidity, 45 percent. 5.11. Usingthe Meyer and Dunne equations, find the daily evaporation rate for a lake given that the mean value for air temperaturewas 80T, for water temperature60'F, the ziverage wind speedwas 10 mph, and the relative humidity was 25 percent.Refer to Appendix Table A.2 for vapor pressrire values. 5.12. Determinethe seasonal consumptiveuse of truck crops grown in Pennsylvania if the meanmonthly temperatures for May, June,July, and August are 62,71,16, and 75'F respectivelyandthepercentdaylighthoursforthegivenmonths arc10.02,10.1,10.3, and 9.6 as percent of the year respectively.

REFERENCES

109

5.13. Using the PenmanMethod, Eqs.5.27 to 5.29, estimateET, giventhe following data: temperatureat water surface: 20 degreesC, temperatureof air : 32 degrees C, relative humidity : 45 percent,wind velocity : 3 mph, the month is Juneat latitude 30 degrees north, r is given as 0.08, and S is found to be 0.73.

REFERENCES
1 . WaterResources Council,TheNation's WaterResources: 1975-2000,U.S. Govt. Print. Off., Washington, D.C., 1978. 2 . U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National WeatherService,NOAA TechnicalReport 33,EvaporationAtlas of the Contiguous 48 UnitedStates, Washington, D.C., June1982. 3 . U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National WeatherService,NOAA TechnicalReport NWS 34, "Mean, Monthly Seasonal, and Annual Pan Evaporationfor the United States,"Washington, D.C., June 1982. 4. "Water-LossInvestigations,"Vol. 1, Lake Hefner Studies,U.S. GeologicSurveyProfessionalPaperNo. 269 (1954). (Reprintof U.S. Geological SurveyCitc.229, 1952.) 5 . N. N. Gunaji, "EvaporationInvestigations at ElephantButte Reservoir, New Mexico," ^lnl. Assoc.Sci.Hydrol. Pub. 18, 308-325(1968). 6. I. S. Bowen, "The Ratio of Heat Lossesby Conduction and by Evaporationfrom Any Water Surface,"Phys.Rev. 27, 779-7 81(1926). 7 . E. R. Anderson,L. J. Anderson,andJ. J. Marciano, "A Reviewof Evaporation Theory and Development of Instrumentation,'o Lake Mead WaterLoss Investigation;Interim Report, Navy Electronics Lab. Rept.No. 159 (Feb. 1950). 8 . O. G. Sutton, "The Application to Micrometeorologyof the Theory of Turbulent Flow over RoughSurfaces," R. Meteor,Soc.Q. "I. 75(No. 236),335-350(Oct. 1949). 9 . A. F. Meyer, "Evaporationfrom Lakes and Reservoirs,"MinnesotaResources Commission,St. Paul,June1944. 1 0 . T. Dunne and L. B. Leopold, Waterin EnvironmentalPlanning, San Francisco:Freeman and Co., 1978. 1 1 . V. M. Ponce,EngineeringHydrology: Principles and Practices.EnglewoodCliffs. New Jersey: PrenticeHall, 1989. 1 2 . M. A. Kohler, T. J. Nordenson, and W E. Fox. "Evaporationfrom Pansand Lakes," U.S. Department of Commerce, Weather Bureau, Res.PaperNo. 38, Washington, D.C., 1955. 1 3 . H. T. Haan, H. P. Johnson,and D. L. Brakensiek(eds.),Hydrologic Modeling of Small Watersheds, ASAE MonographNo. 5. St. Joseph, MI: American Societyof Agricuitural Engineers,1982. 14. R. K. Linsley,M. A. Kohler, and J. L. H. Paulhus, Hydrologyfor Engineers,3rded. New York: McGraw-Hill, 1982. 1 5 . H. L. Penman,"Natural Evaporationfrom Open Water,Bare Soil, and Grass,"Proc. R. Soc.LondonSer.A 193(1032),120-145(Apr.1948). 16. F. G. Millar, "Evaporationfrom FreeWaterSurfaces,l' CanadaDepartmentof Transport, Division of MeteorologicalServices,Can, Meteor. Mem. vol. l, No. 2, 1937. 1 7 . J. B. Franzini, "Evaporation Suppression Research,"Water and SewageWorks (May 1961). 1 8 . Victor K. La Mer, "The Case for Evaporation Suppression," Eng.(June10, 1963). Chem. 1 9 . D. R. Maidment(ed.),Handbookof Hydrology. New York: McGraw-Hill ,1993. 20. H. F. Blaney,"Water and Our Crops," inWater, the Yearbook of Agriculture. Washington, D.C.: U.S. Department of Agriculture,1955.

110

ANDTRANSPIRATION 5 EVAPORATION CHAPTER "Monthly Consumptive Use Requirementsfor Irrigated Crops," Proc. ZI. H. F. Blaney, J. Irrigation DrainageDiv. 85(IR1), 1-12(Mar' 1959)' ASCE, 22. yen Te Chow (ed.), Handbook of Applied Hydrology. New York: McGraw-Hill ' 1964. "Methods of Computing ConsumptiveUse of Water." Proc. ASCE J. 23. W. D; Criddle, Irrigation DrainageDiv. 84(IR1), 1-27(Jan' 1958)' ,;Estimating Evaporation," Trans. Am. Geoplrys. union 37(l), 4324. H. L. penman, "Reductionof Transpiration," Res.66(10), 3309-3312(Oct. J. Geophy. 25. N. J. Roberts, 1961). "Research Proc. Ninth Annual water conon control of Phreatophytes," 26. E. H. Hughes, ference,New Mexico StateUniversity,Las Cruces,Mat. 1964' New York: Hydrology for Engineers. 21. R. K. Linsley,Jr., M. A. Kohler, and J. L. H. Paulhus, McGraw-Hill, 1958. "National WeatherServiceRiver ForecastSysLaboratory, 28. Staff, Hydrologic Research tem ForecastProcedures,"NWS HYDRO 14. U.S. Departmentof Commerce,Washington. D.C..Dec. 1972. "usDAHL-74 Revised 29. H. N. Holtan, G. J. Stiltner,w. H. Henson,and N. C. Lopez, ARS Tech.Bull. No. 1518.U.S.DepartmentofAgriculModelof WatershedHydrology," D.C., 1975. ture, Washington, "Evapotranspiration-Review of ReH. E. Hart, and K. w. Brown, N. J. Rosenberg, Lincoln, 1968' search,"MP 20, Agricultural ExperimentStation,University of Nebraska, Prentice Jersey: New cliffs, Englewood Design. and Analysis Hydrologic H. McCuen R. 31. , Hall, 1989. "Correlation of ClimatologicalData with WaterR'equire3 2 . W. O. Pruitt and F. J. Lourence, California Water ScienceEng. Paper9001. Davis, June of University ments of Crops." 1968. "Potential Evaporation:The cornbination concept and Its Experi33. C. H. M. Van Bavel, Res.2' 455-467(1966)' mental Verification," WaterResources "Discussionof Paperby H. L. Penman,'Estimating Evaporation,"' Trans' 34. H. F. Blaney, Union37,46-48(Feb. 1956). Am. Geoplrys. "Digital Simulationin Hydrology:StanfordWater35. N. H. Criwford and R. K. Linsley,Jr., shedModel IV," Tech. Rept. 39, Departmentof Civil Engineering,Stanford University' July 1966. 36. W. C. Huber, J. P. Heaney,S. J' Nix, R. E. Dickinson, and D"J' Polmann' Storm Water ManagementModel (lsei's Manual, Version /11, EPA-60012-84-109a(NTIS PB84198423).Cincinnati, oH: EnvironmentalProtection Agency, Nov. 1981. Model User's R. P. Shubinski,and J. A. Aldrich, StormWaterManagement 37. L.A. Roesner, (NTIS PB84-198431). I, Extran, EPA-600/2-84-109b Manual, versionIII: Addendum Cincinnati, OH: EnvironmentalProtection Agency,Nov' 1981'

5o(1es6).

6 Chapter

Streamflow

r Prologue
The purposeof this chapteris to: . Introduce the conceptof streamflow. . Describethe characteristics of a hydrograph. ' Presentapproaches to measuringstreamflow. The amountof water flowing in surfacewater coursesat any instant of time is small importanceto those in termsof the earth's total waterbudget,but it is of considerable A knowledge management. and supply, development, resources water with concerned industrial, for: municipal, requisite is a quality of streamflows quantity and of the and design reservoir control; flood endeavors; supply water and other agriculiural, fish navigation; recreation; water-based generation; power hydroelectric operation; as such natural systems of the management drainage; management; und .ildlit" treatment. wastewater water and wetlands;and by precipitationduring stormeventsand by groundwaStreamflowis generated During dry periods, streamflowsare sustainedby surface channels. ter entering reservoirsare below streamchannelsWhere groundwater groundwater discharges. to flow during protractedprecipitationcease regions-streams in arid often the case precipitation and streamflow are complex, being between periods. Relations free chapters.As a result, many foregoing in the factors discussed the influJnced by havebeen developed.l-3 variables hydrologic important these relating to approaches the text' of Three in Part in detail are discussed Severalof them use of flow-measuring on the are based of streamflow Field measurements channelcross-sections of measurement the on flumes and and weirs devicessuchas (see 8). Chapter velocities along with streamflow

BASINEFFECTS DRAINAGE
in a drainagebasin are affectedby The quality and quantity of streamflowgenerated Accordingly,it is important features.a-e climatic and physical, vegetative, the bisin's rocks,plants,topography, soils, ofthe good understanding a have hydrologist that the

J12

CHAPTER 6 STREAMFLOW of events that influencethe sequence land-usepatterns,and otherbasincharacteristics that while pointed however, out, be precipitation It should runoff. and separating process, land-use in the runoff important elements are very natural basin features parking lots, agricultural ieatures createdby humans (e.g., housingdevelopments, practicescan patterns)may,in somecases, be the dominantones.Land management whenthey detrimental can also be they and in retarding erosion, such as be beneficial, principal basin 10, the processes. In Chapter natural hydrologic to accelerate function are discussed. hydrologist concern to the of characteristics

6.2 THE HYDROGRAPH

This by a hydrograph. at a givenlocationon a watercourse,is represented Streamflow, normally respect to time, with properties graph of streamflow displays the continuous obtained by nteansof a continuousrecorder that shows stage(depth) versustime hydrographby applica(stagehydrograph),and is then transformedinto a discharge used herein means a as general, hydrograph the term curve. In tion of a rating hydrograph. discharge As was shownin Fig. 1.3, the hydrographproducedin a streamis the result of that occur-duringand after any precipitation event. A various hydrologicprocesses has processes is given in Chapter11. A hydrograph of these more completediscussion (3) groundwater (2) or interflow, (1) runoff, surface elements: direct four component is known portion a hydrograph of rising (4) precipitation. The channel baseflow, and as the concentrationcurve; the region in the vicinity of the peak is calledthe crest depends The shapeof a hydrograph andthefalling portion is therecession.lo segrnent; the properties. 6.1 illustrates Figure basin pattem and precipitation characteristics on presented. definitions

Stormperiod hydrograph

'{1 oa-. . I'a

\%
. Continuous hydrograph

Endof direct runoff

Time Figure 6.1 Hydrograph definition.

6.4

AND RECORDING STREAMFLOW MEASURING

113

FORSTREAMFLOW 6.3 UNITSOF MEASUREMENT

They are units of water flowing in streams. Two types of units are usedin measuring and units of volume.Discharge, or rate of flow, is the volume of water that discharge passes point in a unit of time. The basicunits usedin connection a particularreference of with streamgaugingin the United Statesare the foot and meter for measurements of time. Commonly usedunits of disdimensionand the secondfor measurements chargemeasurement are cubic feet per second(cfs) and cubic meters per second per squaremile (sec-ft/mi2), (m3/sec). in useare second-foot Other units of discharge from a drainagebasinor definedarea,and for expressing the average rate of discharge Units of million gallonsper day (mgd),commonly usedin water supplycalculations. volume used are the cubic foot, cubic meter, liter, gallon, and acre-foot (a volume equivalent to 1 ft of water over an acre, 43,560 ft2, of land). The latter unit is commonly used in irrigation practice in the westernUnited States.Irrespectiveof the standardunit of time for whetherEnglish or metric units are usedfor dimensions, is the second. streamflowobservations

AND RECORDING STREAMFLOW 6.4 MEASURING

usinggaugingdevicessuchas flumes,weirs,and Streamflowratesmay be determined of head (depth), control sections,or they may be calculated from measurements velocity, and cross-sectionalarea.t-t Usually, specific devices such as flumes are

' bI) 6 6
o bo ri

3000
Discharge (sec-ft)

Figure 6.2 Station rating curve for RaquetteRiver at Piercefield,New York. (U.S. GeologicalSurvey.)

114

CHAPTER6

STREAMFLOW

limited to small streamsbecauseof problems of scale.For large stream systems, velocity and using the cross-sectional is normally estimatedby measuring discharge devicesare Where flow-measuring into discharge. areato translatethis measurement this into translate to curve rating used,it is customaryto observethe head and use a calculations discharge made, be cannot When direct flow measurements discharge. are often facilitated by use of velocity-area relations, chemical tracers, electrical methods,and empirical equationssuchas the Manning formula' In the United States,the primary responsibilityfor gaugingmajor streamslies record of streamflowin terms of with the U.S. GeologicalSurvey,with a systematic mean daily dischargebeing the norm. Usually, a stagerecording is obtained at a by one or more of several gauging siie and this record is convertedinto discharge for this purpose(seeChap-"tttoAr. Rating curves,tables, and formulas are used The instrumentsand curve' rating ter 11).Figure 6.2 rho*r a typical stage-discharge to the adapted carefully mustbe to discharge recordings methodsusedto convertstage the to ensure as so gauging site natural or artiflcial conditions encounteredat the reliability of conversions.'

AREA AND CROSS.SECTIONAL OF DEPTH 6.5 MEASUREMENTS

devicecanbe installedin a stream,aratity for streams Unlessa direct flow-measuring areawill be needed of depth of flow and cross-sectional of any scale,measurements be taken using may to permit dischargeto be calculated. Depth measurements devices.Crosssounding weighted soundin! [nes, calibrated rods, and ultrasonic surveyingtechordinary sectionalareasat streamsectionscan be determinedusing taken below ate that or other depth measurements niquescombinedwith soundings of depth and measurement the water level at the time of the survey.Although the calibration careful require determinations simple,accurate al areaseems cross-section conditions. of instrumentsand the ability to deal with submerged

OF VELOCITY 6.6 MEASUREMENT

area,permit calculacombinedwith thoseof cross-sectional Velocity measurements, velocities can be flow Point tion of dischargeat a given stream or river location. dynamometer' tube, Pitot devicessuch as the determinedusing velocity-measuring long been a has meter and current -"i"r. In the United States,the Price current vanesto the cupped exposing by standardin streamflowgauging.This deviceoperates The velocity. wind measuring used in direction of flow, *o"h lik" the anemometer rotation of rate the and rotatesin nearproportion to flow velocity cup-vaneassembly is convertedto point velocity using a rating table or appropriateequation' , Various chemical and electrical methods are also employed in determining velocities.Commonly usedchemicalmethodsinclude salt velocity, salt dilution' and is the detectionof radioactivetracers. Of these methods,the salt velocity method into the introduced salt principle that perhapsthe most widely used.It is basedon the of the streamwill increaseits electricalconductivity.Electrodesplaceddownstream

FLOW VELOCIry TO CROSS-SECTIONAL POINT VELOCIry 6.7 RELATING

115

profile Figure6.3 Verticalvelocity

and thesecan be translatedinto velocity by conductivity changes salt injection sense track of time. Electrical methods keeping knowing the spacingof electrodesand oxygen voltagegeneration, electromagnetic includethe useof hot wire anemometers, no thus widely and vary conditions Streamflow polarography, waves.a and supersonic method The best suitable. is universally specificapproachto velocity determination of for use at a given site must be determinedon the basis of the characteristics streamflowat that site. Field observationshave shown that the mean velocity in a verticai stream of the velocitiesoccurringat depthsof by the average sectionis closelyapproximated (seeFig.6.3).11'12 depthrespectively percent section of the total 20percentand 80 readingsat about meter feet, single of 0.5 the order on Where depthsarevery shallow, measurements good Velocity yield results.t' to percent point shown havebeen the 50 permit channel estimating cross-sections channel geometric of stream definitions and made. have been measurements and depth velocity flows at locationswhere

FLOWVELOCITY TO CROSS-SECTIONAL POINT VELOCITY 6.7 RELATING

are important, what is desiredis a method to While point velocity measurements velocity, flow velocity. This average cross-sectional translatethem into the average given stream at a yields the discharge area, cross-sectional when multiplied by the vertical numerous at point measurements velocity procedure is to take section.One and horizontal positions in a crosi section,plot them, and then determinevelocity of the average contours.By calculatingthe areasbetweenthe contoursand assigning of a determination areas, these to contours confining of the two the flow velocities is easilycalculated. discharge meanvelocity can be made.Once this is accomplished, properties of stream channel geometric of the make use Other approaches usethis approach, To method. mean-section is the technique such One cross-sections. gauging location into a a at cross-section channel the stream to divide it is necessary

116

CHAPTER6 STREAMFLOW
Water surface

6.1 for Example cross-section Figure 6.4 Channel seriesof geometricshapes(seeFig. 6.4). AL eachvertical location along the crossvelocity of The average section,the meanvelocity is estimatedfrom measurements. of the average the to be equal to is considered two verticals between area flow for the vertitwo between discharge The verticals. of the bordering for each meanvelocities section. of the area the by multiplied section for the velocity average the cals is thus total flow for the are then summedto provide an estimated The individual discharges to measurements enough to have is important it Note that location. at that channel 6.1. Example in illustrated is procedure The cross-section. the characterize
EXAMPLE 6.1

at the sectiongiven in Fig. 6.4. Data from field observations Calculatethe discharge are shownin Tables6.1 and6.2.
TABLE 6.1 DATA FOR EXAMPLE6.1

6.1 TABLE6.2 DATAFOREXAMPLE

section # Vertical 0
I

Depth (ft.) 0
4

Avg. vel.

(rps)
0 2.1 2.3 2.7 2.8 2.5

2
3

'7.2
7.4

5
6 7 4.'1

2.2
0

FORDETERMINING DISCHARGE 6.8 THESLOPE-AREA METHOD TABLE6.3 CALCULATIONS FOR EXAMPLE 6.1 Area
A1

,117

Area (sq.ft.) 8.40 14.85 29.28 3't.96 26.83 30.09 13.87 t61,.27

Vel. (fps) 1.05 2.20 2.50 2,75 2.65 2.35 1.10

Flow (cfs) 8.82

A2 A3
A4

32.67 73.20 104.39 7t.09 70.71 15.25 376.t3

A5 A6 A7

Total estimateddischargeis 376.13 cfs

Solution. The first step is to calculatethe individual section areas(A1, A2, etc.). The calculatedareas(usingtriangular or trapezoidalformulas)are shown meanvelocitiesat the verticalsare on spreadsheet Table6.3. Next, the estimated (see multiplied by the section areasto obtain the individual area discharges Table 6.3). Thesedischarges are summedto yield an estimated376.13 cfs of flow being deliveredby the full channelwidth. r I

DISCHARGE METHOD FORDETERMINING 6.8 THE SLOPE.AREA

neededto deter: it is difflcult to make velocity or other measurements In somecases mine discharge. This is often the caseduring large flood events.Under suchcircumof high possible stances, it is sometimes to estimatethe flow by taking measurements areas,and channelslopesand then water lines (after the flood event),cross-sectional the flow. The applicable usingthesedatain an equationsuchasManning's to estimate Manning equationis Q : \I.49ln)APzrzgrrz where Q : n : A : R: S: discharge(cfs) coefficient Manning's roughness cross-sectional area(ft2) the hydraulic radius the headloss per unit length of channel

(6.1)

For streamflows, Manning's n valuesmay rangebetweenabout0.03 and 0.15. When reasonabledeterminationscan be made of n, A, R, and S, Eq. 6.1 can be used to estimate the streamflow that occurred during the high-water period. For a of this and other streamflowdeterminationmethods,the more completediscussion

118

CHAPTER6

STREAMFLOW

4 and 10 give an references at the end ofthe chaptershouldbe consulted.References excellent overview of techniques and include a valuable list of references on streamflow.

r Summary
Streamflowis the result of storm-periodprecipitation, snowmelt,and groundwater discharge.l3 It is a primary sourceof water for a host of instreamand offstreamuses. The graphicalrepresentation of streamflowis the hydrograph,a plot of flow versus time at a prescribedlocation alongthe water courseof interest.As illustratedby Fig. is a hydrographwhich is 1.3, the end product of many hydrologicmodelingprocesses derived from a precipitation input, modified appropriatelyby various abstractions in streamflowin the field were presented suchas infiltration. Methodsfor measuring this chapter.In later sectionsofthe book, a variety oftechniquesfor deriving hydrographs from precipitation and other hydrologic data are covered (seePart Three).

PROBLEMS
6.t. Considerthat you haveobtaineda gaugeheightreadingof4 ft at a gaugingsite on the
to be in cfs and in RaquetteRiver (Fig. 6.2). What would you estimatethe discharge be?Which of the m3/sec? If the gaugeheighthad been9 ft, what would the discharge two estimatesdo you think would be the most reliable?Why? 6.2. SolveProblem 6.1 if the gaugeheight readingswere 5 ft and 7 ft. 6.3. Consulta USGSWaterSupplypaperand plot the streamflowdatafor a drainagebasin ofinterest.Discussthe factorsthat you believeinfluencedthe shapeofthe hydrograph. 6.4. For the major surfacewater course in your locality, discussthe value of making streamflowforecasts. at 6.5. Calculatethe discharge at the sectiongiven in Fig. 6.4 if the depth measurements Give resultsin the verticalswere:0, 3.8, 5.4, 7.1,8.1,7.0,4.5, and 0 ft respectively. cfs and m3/s. Calculatethe discharge at the sectiongiven in Fig. 6.4 if the velocitieswere: 0, 2.3, and the depthsof Problem6.5 applied. 2.6,3.1,2.9,2.7,2.5, and0 fps respective$, Give resultsin cfs and m3/s.

REFERENCES
1 . N. C. Grover and A. W. Harrington, StreamF/ow. New York Wiley, 1943. 2. United StatesDepartmentof the Interior, Bureau of Reclamation,Water Measurement 3 . I. E. Houk, "Calculation of Flow in Open Channels,"Stateof Ohio, The Miami Conservancy District, Tech. Rept. Part IV, Dayton, OH, 1918. Manual. Washington,D.C.: U.S. GovernmentPrinting Office, 1967.

4. Ven Te Chow (ed.), Handbook of Applied Hydrology. New York: McGraw-Hill, 1964. 5 . American Society of Civil Engineers,"Hydrology Handbook," Manuals of Engineering
Practice. No. 28. New York: ASCE. 1957.

REFERENCES 1 19 6 . O. E. Meinzer, Hydrology.New York: Dover, 1942. 7 . A. N. Strahler, "Geology-Part II," Handbook of Applied Hydrology. New York: McGraw-Hill. 1964. 8 . R. E. Horton, "Drainage Basin Characteristics,"Trans.Am. Geoplrys.Union l3r 35O36r(1932\. 9 . W. B. Langbeinet al., "TopographicCharacteristics of DrainageBasins,"U.S. Geological Survey,Water Supply Paper,968-c, 1947. 1 0 . R. K. Linsley, Jr., M. A. Kohler, and J. L. H. Paulhus,Applied Hydrology. New York: McGrawHill, 1949. 1 1 . S. S. Butler, EngineeringHydrology.EnglewoodCliffs, New Jersey:Prentice-Hall, Inc., 1951. 12. C. H. Pierce,"Investigationof Methodsand EquipmentUsedin StreamGauging,"Water Supply Paper 868-A, U.S. GeologicalSurvey,Washington,D.C.: GovernmentPrinting Office,1941. 1 3 . D. R. Maidment (ed.),Handbookof Hydrology.New York: McGraw-Hill , 1993.

PART TWO

ENTS MEASUREM HYDROLOGIC AND MONITORING

Chapter7

DataSources Hydrologic

r Prologue
The purposeof this chaPteris to: . Describethe principal sourcesof data for hydrologicinvestigations. and modeling. forecasting, Data on hydrologicvariablesare fundamentalto analyses, agencies, federal and state of publications Such data *uy 1" found in numerous significant most ofthe Several and otheroganizations. universities, institutes, research briefly in this chapter'1-3 sourcesof hydrologicdata are described

DATA 7 . 1 GENERALCLIMATOLOGICAL
solarradiation,wind, and of data on temperature, The most readily availablesources the EnvironrnentalData by published humidity are ilimatological Data bulletins (NOAA), and Administration Service of the Nationai Oceanic and Atmospheric Climatic Data National the by publtshed Monthly Summaryof Solar RadiationData MeteorologWorld the with in cooperation Center.The EnvironmentalData Service, World- A the Data Climatic Monthly for ical Organization(WMO), also publishes entitled Administration, Service 1968 publication of the Environmental Sciences evapohumidity, temperature, wind, Climalic Atlasof the UnitedStates,summarizes these to In additionof maps. series ration, precipitation, and solar radiation on a agriculand resources, geologic, water federal sourcesof data, stateenvironmental, shouldbe consulted.Most stateuniversitiesalso publish a variety of tural agencies hydrologicdata through their researchcentersand extensionprograms.

DATA 7.2 PRECIPITATION

There are probably more records of precipitation than of most otherhydrologic The priniipal federalsourceof data on precipitationis NOAA. Climatologvariables. the ical Data, publishedmonthly and annuallyfor eachstateor combinationof states, pacific area,PuertoRico, and the Virgin Islandsby the EnvironmentalData Service,

E # = ,

124

CHAPTER7

DATA SOURCES HYDROLOGIC

departuresfrom normal, and extremes of presentsa table of monthly averages, precipitation and temperatureas well as tables of daily precipitation, temperature, Hourly Precipitawind, and soil temperatarc. snowfall,snowon ground,evaporation, tion Data is issuedmonthly and annuallyfor eachstateor combinationof statesand presentsalphabeticallyby station the hourly and daily precipitation amounts for A stationlocationmap is alsoincluded.This with recordingauges. stationsequipped publication is availablefrom the EnvironmentalData Service.Another publication, periods.Data are regionsfor 10-year by geographic is issued Records, World Weather listed by country or areaname, station name,latitude and longitude,and elevation. pressure, and tempersea-level Monthly and annualmeanvaluesof stationpressure, order.Aside aiure, and monthly and annualtotal precipitationare given in sequential and universitiespublish precipitation from NOAA, other federal and state agencies dataat varying intervals,often in a storm o1 site-specificcontext.In addition, many utilities also collect and maintain precipitamunicipalitiesand water and wastewater tion and other related data. Computeized precipitation data are availablefrom the National Climatic Data Centerin Asheville,North Carolina.

DATA 7.3 STREAMFLOW

of streamflowdatafor the United Statesare the U.S. Geological The principal sources andU.S. (SCS), U.S.ForestService, Service U.S. Soil Conservation Survey(USGS), Service(ARS). In addition,the U.S. Army Corpsof Engineers Agricultural Research (COE), the Tennessee Valley Authority (TVA), and the U.S. Bureauof Reclamation and tabulatestreamflowdata relative (USBR) make somestreamflowmeasurements also organizations and variousresearch universities, Stateagencies, to their missions. compile and publish a variety of streamflowdata. The USGS Water Supply Papers(WSf; are the benchmark for referencing data are also availablefrom the USGS. streamflowdata. Furthermore,computerized Publications of the Geological Survey, publishedevery 5 years and supplemented annually, arean excellentsourceof information on that agency'sreports. The SCS historically published data on streamflow from small watershedsand plots in its Hydrologic Bulletin series,but much of the data have been republishedby ARS. "pilot watersheds" are publishedin cooperationwith the USGS. Recordsfrom SCS U.S. Forest Service streamflow data are publishedat irregular intervals in various papers. technicalbulletins and professional

DATA AND.TRANSPIRATION 7.4 EVAPORATION

of ClimatologicalData, publishedby NOAA, includepan Monthly and annualissues and water utilities are evaporationand related data. The ARS, agricultural colleges, are often other sourcesof information. In particular, data on evapotranspiration working through their Agricultural Experiment obtained by university researchers Stations.

REFERENCES 125

r summary
Numerous in hydrologicmodelingprocesses. Climatic and other data are keystones to support model developmentand sources of data exist and may be accessed verification, statisticalanalyses, and specialstudies.

PROBLEM
7.1 Developa list ofdata sources in your stateorlocality by visiting the library or through other channels.

REFERENCES
1 . Soil ConservationService,U,S. Departmentof Agriculture, SCSNational Engineering Handbook, "Hydrology", Sec. 4. Washington,D.C.: U.S. GovernmentPrinting Offlce, Au,g.1972. J. F. Miller, "Annotated Bibliography of NOAA Publications of Hydrometeorological Interest," NOAA Tech. Mem. NWS HYDRO-22, National Oceanic and Atmospheric Administration, Washington, D.C., May 1975. , D. R. Maidment (ed.), Handbookof fudrology. New York: McGraw-Hill, 1993.

ChapterB

lnstrumentation

Prologue
The purposeof this chapteris to: ' Describeinstrumentsusedin measuring hydrologicvariables. . Indicate waysin which data are recordedand transmitted. . Presentlimitations on measurements. must be obtainedin sufficientquanThe data neededto supporthydrologicanalyses form if they are to be of value. appropriate in an frequency, and tity, with adequate the data.They are the subject transmit and to obtain are used A variety of instruments of this chapter.

8.1 INTRODUCTION
problemanalyses, research, supportsarealinvestigations, Hydrologicinstrumentation are A hostof measurements planning,and environmentalpolicymakingand analysis. neededto support efforts in water resourcesplanning, management,design, and analysis,solid wastemanageconstructionrelatedto suchsubjectsas aquifer systems water supplyavailability,water quality management, ment, flood hazardassessment, groundwaterrecharge,protection of fish and wildlife, and navigation. Historically,instrumentswere often usedto obtain cumulativerather than conFqrtinuousinformation abouthydrologicvariablessuchasrainfall and evaporation. thermore,there was often no attemptto correlatewater quality constituentloadings, many historic data havelimited for example,with ratesof water flow. Consequently, but ratherfrom using instrumentation, of lack of adequate utility, not somuchbecause availableinstrumentsto measurethe wrong thing or in too limiting a fashion.Today it is widely recognizedthat it is important not only to selectappropriateinstruments of moderntimes. but to selectthem in the contextof datanetworksthat meetthe needs More will be said about this in Chapter9, In Section8.2, instrumentsfor measuring hydrologicvariablesand waysin which they can be usedjointly to createa complete of a functioning hydrologicsystemare discussed. representation

'

INSTRUMENTS 127 8.2 HYDROLOGIC

INSTRUMENTS 8.2 HYDROLOGIC

Good sourcesof information abouthydrologicinstrumentsare the National Weather U.S. Army Corpsof Survey,U.S. Bureauof Reclamation, U.S. Geological Service, Theseagencies manufacturers, instrument and Service, Soil Conservation Engineers, variablesand hydrologic measuring of in the business and industrieshavelong been Someofthe of state-of-the-artmeasuringdevices. theycanprovidedetaileddescriptions is far from but the coverage here, are described instruments major typesof measuring references.l-3 the appropriate consult should exhaustiveand the interestedreader

Precipitation
for measuringrainfall and snowfall may be recording or nonrecording.The Gauges 8-in. gauge. Servicestandard gauge is the U.S. Weather mostcommonnonrecording gaugeis The this is daily. often interval but The gaugemay be read at any desirable rainfall equivalent shows the inserted, calibrated so that a measuringstick, when they but required, are periodic volumes depth. Such gaugesare useful when only rainfall' of cannotbe usedto indicate the time distribution the ratb of rainfall and its time of occurcontinuouslysense Recordinggauges type or the tipping are usually either ofthe weighing-recording rence.Thesegauges period 1 week, at which time of for a gauges run usually buckettype. Weighing-type 2.5 is typical of the Problem with The figure associated their chartsmustbe changed. product, and this is the time versus depth recordedoutput. A masscurve of rainfall ratios of the graph by calculating curve can be translatedinto an intensity-time gauges, bucket desired. Tipping is time step rainfall to time for whatever accumulated rainfall accumulationwhen it reachesa on the other hand, senseeach consecutive 1 mm of rain, A small calibratedbucket is prescribedamount, usually 0.01 in. or gauge. port When it fills to the 0.01-in. the of located below the rainfall entry position. Thesetwo small buckets into bucket incrementit tips over,bringing a second they fiIl. Each time a forth as and tip back are placed on a swivel and the buckets form. In this recording or other chart on a strip an indication bucketspills it produces rain gauges For (intensity) is the outcome. time way a record of rainfall depth versus these made. Usually be must modifications to record snow accumulations,some measurable into be converted can the snow involveproviding a melting agentso that water. Figure 8.la is the diagram of a self-reportingrain gaugingstation.The tipping a digital input signal whenever1 mm of rainfall drains bucketmechanismgenerates The signalfrom the gaugeis automaticallytransmitted through the funnel assembly. to a receiving station where it records the station ID number and an accumulated was amountof rainfall. The receiving stationrecordsthe time at which the message Figaccordingly' calculated periods be can received and rainfall rates for desired ure 8.1b showsa similar gaugeequippedto measuresnow.In this case,a glycometh solution is usedto melt the snow.The melt water overflowsthrough a temperaturethe by the tipping bucket,which operates and is measured mechanism compensating incorporated easily be can Fig. 8.1 in of the type shown station's transmitter.Gauges that can be used in a variety of forecastingand into real-time monitoring systems operatingmodes.

"

128

CHAPTER 8 INSTRUMENTATION

Antenna

Antenna mast

Antenna

Antenna mast

Temperatffe compensation overflow mechanism Glycometh collecting section Tipping bucket Drain holes (4)

Funnel assembly

Tipping bucket

Signalcable Lifting rope Antennacable

Signalcable Lifting rope Antennacablb

(a)

(b)

Figure 8.1 Self-reporting (a) rain and (b) snow stations. (Courtesy of SierraMisco, Inc., EnvironmentalProducts,Berkeley,CA.)

Evaporation and Transpiration

Evaporationpans have been widely used for estimatingthe amount of evaporation Devicessuchasthat depictedin Fig. 8.2 areeasyto use,but from free water surfaces. relating measurements taken from them to actual field conditionsis difficult and the -data_lhey-produce value for making areal estimates.A are often of questionable

..

INSTRUMENTS 8.2 HYDROLOGIC

129

tr'igure 8.2 U.S. Weather Bureau Class A pan.

but the U.S. WeatherBureauClassA pan variety of pan types havebeen developed havebeenusedto Panevaporationobservations is the standardin the United States.4 from well-watered and evapotranspiration both free water(lake) evaporation estimate vegetation.Field experimentshave shown a high degreeof correlation of pan data from surroundingvegetationwhen there is full cover and with evapotranspiration pan datacan be recorded gauges, goodwatersupply.a As in the caseof precipitation and transmittedcontinuouslyto a central receiving station. are often made using lysimeters.Thesedemeasurements Evapotranspiration vices are containersplaced in the field and filled with soil, on which sometype of growth is maintained.The object is to study soil-water-plant relationsin vegetative a natural surrounding.The main featureof a weighinglysimeteris a block of undisin a steel shell that is 10 ft by turbed soil, usually weighing about 50 tons, encased 10 ft by 8 ft. The lysimeteris buried so that only a plasticborder marksthe top of the containedsoil. The entire block of soil and the steelcasingare placedon an underground scale sensitiveenough to record even the movement of a rabbit over its surface.The soil is weighed at intervals, often every 30 min around the clock, to most of the in soil water level. The scalesare set to counterbalance changes measure deadweight of the soil and measureonly the active changein weight of water in the The scalescan weigh accuratelyabout 400 g (slightly under 1 1b), which is soi1.5 equivalent to 0.002 in. of water. The weight loss from the soil in the lysimeter Added water coverplus any soil evaporation. waterusedby the vegetative represents is also weighedand thus an accountingof water contentcan be kept. Crops or cover are plantedon the areasurroundingthe lysimeterto provide uniformity of conditions surroundingthe instrument.Continuousrecordsat the set weighingintervalsprovide monitoring of conditions.The data obtainedcan be transmittedto almostcontinuous can produce location for analysisand/or other use.Weighinglysimeters any desirable over short periodsof time. But they are expenvaluesof evapotranspiration accurate which are lesscostly,havealsobeenused,but sive.Nonweighingtypes of lysirneters, unless the soil moisture content can be measuredreliably by some independent method,the data obtainedfrom them cannotbe relied on exceptfor long-term meaevents.s precipitation suchas between surements

and HumiditY Wind,Temperature,

to supportmany types and humidity are needed of wind, temperature, Measurements a device anemometer, an using measured is commonly Wind of hydrologicanalyses. speed propeller whose (Fig. or 8.2) cup a such as element that has a wind-propelled vane, which a using is obtained direction Wind wind velocity. is calibratedto reflect orients itself with the direction of the wind.

130

CHAPTER 8 INSTRUMENTATIoN

Temperaturemeasurements are madeusing standardthermometers of various types,while hurnidity is measured using a psychrometer. A psychrometer consistsof two thermometers, one called a wet bulb, the other a dry bulb. Upon ventilation the thermometers measuredifferently, and this differenceis called the wet-bulb depression.By usingappropriate tables,dewpoint, vaporpressure, andrelativehumidity can , be determined.6 Figure 8.3 depictsa completeweatherstation incorporatingmeasurements of precipitation, wind, temperature,barometricpressure,and humidity. Sucha station - can automatically report weatherdata from remote siteson either an event and/or

Funnel assembly Solar panel Tipping bucket

Lifting rope Antennacable

Main housing

Signal cable

Ground level

Transmitter

Figure 8.3 Self-reporting weather station. (Courtesy of Sierra-Misco, Inc., Environmental Products, Berkeley, CA.)

131 INSTRUMENTS 8.2 HYDROLOGIC timed basisto a central site. A station suchas this can be used for marine weather of oncoming storms,determinationof wind forecasting,quantitativedeterminations effect on tidal areas,and establishinga data basefor irrigation.

OpenChannelFlow
of open channel(natural and created)flow are made using standard Measurements measuringdevicessuchas flumes and weirs, and they are also madeby calibrating of depth special control sectionsalong rivers and streamssuchthat measurements so designed are devices Flow-measuring (Jtage)of flow can be related to discharge. observation the translates that sensingsomeparametersuchas depth automatically of crossWhen a control sectionis used,observations into units of flow (discharge). must velocities flow average and sectionalareafor variousdepthsmust be obtained, In be established. curve can rating so that a section for variousstages be ascertained the Reclamation, of Bureau U.S. the United States,the U.S. GeologicalSurvey,the havedoneextensive Service,andthe U.S. Army Corpsof Engineers Soil Conservation and proceduresfor instruments flow measuringand havebeen active in developing rates of flow.2'6 ascertaining devices.When they are properly inWeirs Weirs are common water-measuring and accuratemeansfor gauging simple stalled and maintainedthey can be a very weir and the V-notch rectangular the The most often usedweir types are discharge. 0.5 ft or more in about a fall of require weir (Flg. 8.4). To be effective,weirs usually structureplaced is an overflow weir a the channelin which they are placed.Basically, steadyfree-flow with shape and size acrossan open channel.For a weir of specific can exist of water depth one only relation, stateconditionsand a proper weir-to-pool measuring by is determined rate Flow in the upstreampool for a particular discharge. from the crestof the overflowpart of the weir to the watersurface the vertical distance this recordeddepth in the upstreampool. The weir's calibrationcurvethen translates into rate of flow at the device. flow section openchannel shaped flumeis a specially Parshall Flumes A Parshall The devices. of these one depicts 8.5 that can be installedin a channelsection.Figure head small a relatively with (1) operate flume has severalmajor advantages: it can to the approachvelocity; (3) it can be usedevenundet loss;(2) it is fairly insensitive conditions;and (4) its flow velocity is usually sufficientto precludesedisubmerged by the late Ralph The Parshallflume was developed ment depositsin the structure.2 throat of the The constricted flume. L. Parshill and it is a particular form of venturi as in the case Thus, discharge. to related a differential headthat canbe flume produces the rate to determine required (head) is that is all ofdepth .ofthe weir, an observation gauging to suited generally best are of flow at the control point. Weirs and flumes weirs can be installed althoughlargebroad-crested and openchannels, small streams other measuringaprivers, major For at dam sites as part of overflow structures. sectionmust be recontrol a specified proachessuchas developingfield ratings at lied on.

132

CHAPTER8

INSTRUMENTATION

Figure 8.4 Field installation of weirs: (a) rectangular and (b) Vnotch. USDA CooperativeExtensionService,Mountain StatesArea.

Golttrol Sections where the installation of a weir, flume, or some other flowmeasuring deviceis impractical, it is sometimes possibleto developa rating curve at somelocation alonga streamby taking measurements of depth,cross-sectional area, and velocity and calculatingthe rate of flow for a particular stageat the location.By doing this for a range of depths of flow, a station rating curve can be developed. Instrumentsrequired to developsuch a curve are depth-sensing devices,surveying instruments,and velocity meters.The velocity meter is similar to an anemometer. It is placedat variouspositionsin the channeland a velocity is recorded.By doing this at a numberoflocations, a velocity profile for a given depth can be developed. From this an average flow velocity can bJcomputed, ind uy uiing that determinationand

INSTRUMENTS 133 8.2 HYDROLOGIC

Diverging sectlon Throat section
a
J

Altemate45" wing wall PLAN

t x t xf,tngle

SECTIONZ-f, Service.) flume.(U.S.Soil Conservation Figure 8.5 Parshall

can be calculatedas the product of meanvelocity area,discharge the cross-sectional can be madefor a rangeof depths,a rating observations area.If and cross-sectional of depthwill sothat only measurements control section developed for the curvecanbe on this Additional information later time. flow at some to estimate rate of be needed proceduremay be found in Refs.2 and 6.

Valve shut-offkeys Connecting band Valves

Figure 8.6 Recorderhouseand stilling well for a streamgaugingstation. (U.S. Bureau of Reclamation.)

134

CHAPTER 8 INSTRUMENTATIoN

Depth (stage) Measurements Most depth measurements are made using a float and cable arrangement in a stilling well or a bubbler gauge.In the first insiance,a stilling well connectedto the channel(Fig. 3.6) is used to housea float devicethat activatesa recorder as it movesup and down. Figure 8.7 illustratesa self-reporting stilling well liquid-level station. Data from this station can be transmitted to un! central location for analysisand/orother use.A bubbler-typeinstallation makesusl of dry air or nitrogenas a fluid for bubblingthrough an oriflce into a channelbed. As the depth of flow changes,the changein head above the bubbler orifice causesa

Antenna mast Cover with special accessscrews

Transmitter
Hydraulic damping device Level sensor Counter weight Mounting brackets

lnlet tubes

Sidecleanout port Bottomcleanout port

Figure 8.7 Self-reportingstilling well liquid-levelstation. (Courtesy of Sierra-Misco,Inc., Environmental products. Berkeley,CA.)

8.4 REMOTE SENSING 135 corresponding pressure change. This resultsin a fluid-levelchangein the manometer connected to the gassupplyand this in turn is usedto reflect stagevariationover time. The foregoingdescriptionsare of a few of the instrumentsused in hydrologic work. Both the limitations associated with their use and their reliability must be understoodif they are to be used correctly and their outputs are to be considered credible.

8.3 TELEMETRY SYSTEMS

Historically,many gauges were read periodically by an individual making the rounds of installations.This servedwell whenthe purposeof the data wasto establish a base record of somevariablesuchasrainfall. But today,undermany circumstances, it has becomenecessary to make continuous recordingsof rainfalls, streamflows, and evaporation rates and to have thesedata availablefor the real-time operation of water management systems and for forecastinghydrologicevents.Someexamples of activities requiring real-time hydrologicdataarc managingreservoirs, issuingflood warnings, allocating water for various usessuchas irrigation, monitoring streamflows to ensure that treatiesand pactsare honored,and monitoringthe quality and quantity of waterfor regulatoryand environmental purposes. Accordingly,gaugingstationscapable of electronicallytransmitting their data to a central location for immediateuse havenow becomecommon.The advantages of suchstationsincludeproviding information to usersin a time frame that meetsmanagement needs,reducingthe costsof collecting data, and providing a continuousand synchronous record of hydrologic events.Figure 8.8 showsa streamgaugereporting station using radio transmission. Figure 8.9 illustratesa satellitedata collection and transmitting operation.T-|2

8.4 REMOTE SENSING

Sincethe 1960s,remote sensing hasbecomea commonhydrologictool. Examplesof aircraft and satellite datacollectionand transmission abound.13-16 Figure8.10 illustratesthe useof aircraft and satellites in a snowsurveysystem. Other typesof surveys suchasthoserelatedto determining imperviousareas, classifyingland usesfor assessing basin'wide runoff indexes, determining lake evaporation, and groundwater prospectingcan be depictedin similar fashion.Table 8.1, which summarizes operational uses of satellite data in hydrology circa 1981, showsthe great diversity of remote sensingand data transmission options that can be exercised.l6 The principal value of remote sensing is its ability to provide regionalcoverage and at the sametime providepoint deflnition.Furthermore,satellitecommunications can be digitized and are thus compatiblewith the transferof computerized information. Following the evolution of linkages between computer and communications technology, new softwaresystems incorporatingpowerful data management systems havebeen developed. Thesesystems facilitate the storage,compaction,and random access of large data banks of information. one data management option, geographic (GIS), allows the overlayingof many setsof data (particularly information systems satellite-deriveddata) for convenientanalysis.Versatilecolor pictorial and graphic display systems are also becomingattractiveas their costshavedecreased.ra

136

CHAPTER 8 INSTRUMENTATIoN

I
L. I

Utah
Anzona

q)

B .H

Boulder City VIIF Radio Station

Mt. Hualpai Repeater Station N

20

40

60

"&*" rq*
s
Taylor'sFerry gauge

,&

Scale of miles
Lake Havasu

EXPLANATION Parker Dam

Black Point Repeater Station

fl Water level gauge

X

il VIIF radio antenna tower H @ Transmitter-ieceiver radio

Cibola gauge

Colorado nver

VHF Very high frequency radio

lmperial Dam Repeater Station

Dam
-'-:4

az*<3e -{/co '-Figure 8.8 Streamgaugereporting system Water stage using radio transmission. information is requested from the gaugingstationsby VHF radio signal.In turn, this water stage information is obtained from the stream gaugesand automatically encodedand transmitted to the Boulder City receiving station. All downstream releases from Hoover Dam are determinedand integrated with this streamflow information in controlling the flow of the lower ColoradoRiver. (U.S. Bureau of Reclamation.)

8.4

REMOTESENSING

137

Figure 8.9 Hydrologic data collection by satellite.(U.S. GeologicalSurvey.)

Low-altitude gamma-ray light.

Water 4gency forecastcenter

Figure 8.10. Satellite snow survey system.(AftenCalabreseand Thome'rs)

jo (d

E o E,

32
o

o o o o

c
k

9 o
IY

(u (E o

89'
d r i

o z

F-> 'o ri
X a ! c

s
o
IJJ

Fl

t J
J

tr|
F

a
lJ-

U) uJ a f
J

eE;
* E E
o
(E o CL

o
F

(r tu o
d ut 6

' E c c

SEE
! q o H o 9r

5 E K
.=

E E>,q F-.ts ,Y

hiEE. g b " E6
F

E R ! ' i

F
o o a

;+

o
E

t T

c o

E E a a

e o
o

E
!o

Ec,

o x

o E

o r A 0

boo

o o U' o (o

6 6

l 6 f i E

, .

bo o o (B ,o o. e z a

s
o
dt uI

Ch

(E
L . =

!t e qt a)

3 -

E
o .
O

o E a

e Ci

E O :.i c >r-!!

g o o 4 =
O F , Y o E

I E t

Ee
L

+e
d

8-:

at)

142

CHAPTER 8 INSTRUMENTATIoN With the advancement of satellite technology,the use of satellitesas remote platformshasspread.Currently available sensor can operatein a multitude of sensors electromagnetic radiation wavelengths and the information content of their signals pollutants,and other types can include Surface temperatures, radiation, atmospheric of meteorological data. As remote sensors are improved to permit the attainmentof greater radiometric and geographicresolution, and as computer image-enhancing techniques becomemore sophisticated, it is certain that this powerful watermanagement tool will seeeven sreaterand more diversifieduse.

r summary
Hydrologic data are important componentsof model design and testing and of a variety of statistical analyses. The quality of data obtained relate to attributes of measuring instrumentsand to the featuresof gaugingsites.It is important to understandthe pros and consof variousinstruments and to know how they canbestbe used.

REFERENCES
1. "Irrigation WaterMeasurement," Mountain StatesRegionalPubl. 1, revisionof Extension Circ. 132,Irrigation Water Measurement, University of Wyoming, Laramie, June 1964. 2. U.S. Bureau of Reclamation,Water Measurement Manual, 2nd ed. Washington,D.C.: U.S. GovernmentPrinting Office, 1967. Data Book, 4th ed. Beaverton,OR: 3. Leupold & Stevens,Inc. StevensWater Resources Leupoldand Stevens, Jan. 1987. 4. W. Brutsaert,Evaporationinto the Atmosphere. London: D. Reidel Publishing, 1982. 5. "TexasLab InstallsWeighingLysimeters," Irrigation J. 37(3),8-12(May/June 1987). 6. R. K. Linsley, Jr., M. A. Kohler, and J. L. H. Paulhus,Applied Hydrology. NeW York: McGraw-Hill, 1982. 7. R. J. C. Burnashand T. M. Twedt, "Event-Reporting Instrumentationfor Real:TimeFlash Flood Warning," American Meteorological Society, Preprints, Conference on Flash Floods:Hydro-meteorological Aspects, May 1978. 8. D. E. Colton and R. J. C. Burnash,"A Flash-FloodWarningSystem,"American MeteoAspects, rologicalSociety,Preprints,Conference on Flash Floods:Hydro-meteorological May 1978. "AutomatedPrecipitation Aug. 1980. 9. R. J. C. Burnash, Measurements," 10. R. J. C. Burnashand R. L. Ferral, "A Systems Approach to Real Time Runoff Analysis with a Deterministic Rainfall-Runoff Model," International Symposium on RainfallRunoff Modeling,Universityof Mississippi, May 18-21, 1981. 11. Hydrologic ServicesDivision, National WeatherService,WesternRegion, "Automated Local Evaluationin Real Time: A CooperativeFlood Warning Systemfor Your Community," Feb. 1981. Involvedin 12. R. J. C. Burnashand R. L. Ferral, "Examplesof Benefitsand the Technology Optimizing HydrosystemOperation Through Real:Time Forecasting," Conferenceon Real:TimeOperation of Hydrosystems, Waterloo,Ontario, June24-26, 1981. 13. M. Deutsch,D. R. Wiesnet, and A. Rango (eds.),Satellite Hydrology. Bethesda,MD: AmericanWaterResources Association. 1981.

REFERENCES 143

t4. J. F. Bartholic,

"Agricultural Meteorology: SystemsApproacli to Weather and Climate Thirty-Second in Proceedings, Needsfor,tgricuttrire, Forestry,and Natiiral iesources,o' ResearchInstitute, Agricultural MD: Bethesda, R'searcih'Institute. Agiicultural tuteeting, 1 9 8 3p , p.75-85. *NASA Water Resources/HydrologyRemote Sensing 1 5 . M. A. Calabreseand p. G. Thome, Programinttrelg80's,"insatettiteHydrology(M'Deutsch'D'R'Wiesnet''andA'Rango' Association' 1981' pp' 9-15' MD: American Water Resources eds). Bethesda, 16. G.K.Moore,..AnlntroductiontosatelliteHydrology,''hsate.tlite-Hydrology(M. MD: American WaterResqurces Deutsch,D. R. Wiesnet,and A. Rango,eds.).Bethesda, t7-4L. 1981,PP. Association,

ChapterI

Networks Monitoring

Prologue
The purposeof this chapteris to: . Describeelementsof systems for monitoring hydrologicvariables. . Indicate the importanceof real-time and continuousrecording of hydrologic events. for manipulating Information (data)is the requisitefoundationfor designing schemes (managing) hydrologicsystems, for evaluatingthe efficacyof actionstaken to correct problem situations,and for identifying trouble spots deservingattention. But to be useful, the data must be of the right type, in the right form, and appropriatelyrepresentativeof critical spaceand time dimensions. Modeling hydrologic systems requires an understanding of how these syster-ns requirestracking the effectsof actually function; cleaningup a toxic wastedischarge remedial actions;enforcing environmentalregulationsrequires knowledgeof what has happenedsincethe rules were implemented;and regulatingreservoirreleases to meet specifiedtargetsrequiresa continuous understanding of the stateof the system lies in the productsof carefully beingoperated. The key to meetingsuchrequirements designedand managedmonitoring networks.Developingsuchnetworksis no small task, however,as the numberof variablesthat must be observedmay be very large, and the instruments to measure them costlyto install and operate,andthe datastorage managementrequirementsextensive.Accordingly, a monitoring network's design must beginwith a thoroughunderstanding of its purposeso that the degreeof resolufor the task at hand. but not excessive, tion providedby its observations is adequate, of what within the constraints A goodrule is to keepthe networkassimpleaspossible, must be accomplished.

9.1 THEPURPOSE OF MONITORING

The purpose of monitoring is to gather information in a continuum such that the dynamics of the systemcan be ascertained.According to Dressing,objectives of of baseline monitoring for nonpoint source pollution control include development

9.2 SPECIAL CoNSIDERATIoNS 145 information, generatingdata for trend analysis,developing and/or verifying models, and investigating singleincidentsor events.z Theseobjectivesare alsovalid for hydrologic monitoring in general,but they shouldbe supplemented by the following obje"tives:planning,real-time system operating,enforcingregulatoryprograms,and environmental policymaking. In the flnal analysis, the ultimate purposeof monitoring is to enhancedecisionmaking, whetherit be for development, management, regulation, or research aims.

9.2 SPECIAL CONSIDERATIONS

Beforean acceptable monitoringplan can be devised, theremustbe a full understanding of the hydrologic systemto be monitored and of the objectivesto be met by monitoring. The costs of monitoring can be very high and thus it is essentialthat monitoring networksbe efflcient and cost effective.r-5

Timeand SpaceVariability
In general, monitoring networks are designedto have both spatial and temporal dimensions. Although monitoringa specificpoint locationmay be all that is necessary undersomecircumstances, it is more commonthat what is happening in a regional settingis of importance.The temporal aspectis similar. While a snapshotat some point in time may suffice for some purposes, the time varianceof conditions to be tracked is usually critical for effective analysesandlor decisionmaking.Both the short-termand long-term variabilities of many targefsof monitoring must be ascertained. For example,water quality in a streamcan changerapidly with time, while changes in lake levels,suchas thoseexperienced in the GreatLakes in the 1980s,are the result of long-termhydrologicvariability. Spatial variability must also be represented in a monitoring network: for example, infiltration ratesmay vary considerably within a region, rainfall intensitiesmay be quite different within even short distances, and water quality in a river might be different in upstreamand downstreamlocations.Topography,soils, vegetalcovers, and many other factors affecting the performanceof a hydrologic systemare also distributed differently in space,and these differencesmust be recognizedin the monitoring plan. The trick is to developa monitoring system that can (1) provide the neededdata,(2) recognizeregionaland temporal variabilities,and (3) keep installation, operation,and maintenance coststo a minimum. To do this requiresa comprehensiveknowledgeof the system to be monitored,an understanding of what the data obtainedby the systemwill be used for, and a knowledgeof the level of detail in collecting the data that must be exercised in spaceand time.

DataRequirements
The amountand type of datato be generated by a monitoringsystem mustbe carefully consideredin its design. Selectingappropriateinstruments,determining sampling frequency,and settingdata formats are elements that must be considered. Questions such as how much do we need to know and when do we need to know it must be answered. The form and extensiveness of data must be tightly relatedto monitoring

146

CHAPTER9

MONITORING NETWORKS

Flood warnlng Water managemeff

---:
Telephone

Lift station monitoring control

Hardwire
+ + , . .

lnfrared + + xgnr
Water quality monitoring

(Courtesy of Sierra-Misco, Inc.,EnvironFigure9.1 A telemetry monitoring system. mental Products. Berkeley. CA. insteadof the to monitor surrogates objectives.Furthermore,it might be necessary condition to be tracked.' For example,if lake eutrophicationis the issue,phosphorus If this approach is taken, flre&sureS; andchlorophyllconcentrations might be surrogate is very important and the foregoingcomments selectionof appropriatesurrogates aboutdata formats and so on are also applicable.Hydrologic,waterquality, land use and treatment, topographic,soils, vegetativecover, meteorologic,and many other in a monitoringplan. It is typesof datamaybe requiredin combinationor separately easyto seethat the amount of data required for a monitoring program can be enorgreat care must be taken to seethat the data collection effort is mous.Consequently, not in excessof the objectivesof the monitoring program. Figure 9.1, depicting a gives an indication of the variety of data that might be telemetrymonitoring system, collectedin a monitoring program.

QualityControland QualityAssurance
that the data The costsof monitoring are usually substantialand thus it is essential generatedbe of consistentlyhigh quality. Accordingly, most monitoring systems (QA/QC) elements. include quality control and quality assurance Quality control is to producea quality product(datain this case) a plannedsystem of activitiesdesigned is a plannedsystemof activities that meetsthe needsof the user. Quality assurance designed to guarantee that the quality control programis being carried out properly. A quality managementplan should be part of the overall monitoring program to ensure and should be preparedwhen the monitoring program is being developed that the data collectedwill be of a satisfactorynature for the monitoring program's objectives.3

9,4

NETWORKS HYDROLOGICAL-METEOROLOGICAL

147

To useror

Figure 9.2 Microcomputeruse in streamgauging.

IN MONITORING 9.3 USEOF COMPUTERS

miof computers,especiallyinexpensive With the rapid technologicaldevelopment crocomputers, the oppoitunities foiautomated collection of all types of hydrologic and water quality data have increased substantially.Microcomputers, used with analog-to-digital converters,pressureor liquid-level sensors,and the appropriate software can, for example, be used in hydrologic monitoring systemsas flow (Fig.9.2).4Furthermore, are highly suchsystems acquisition systems metering/data versatile and they are rblatively inexpensive.Computer systemscan be customapplicationand they are often lesscostly than for almost any dataacquisition designed for the samepurpose.Comdesigned hardwaresystems other commerciallyavailable puterscan convert raw datainto other more usefulforms, storedata for later use,and As such,they are a powerwith other computerterminalsif necessary. communicate Figure 9.3 ful and important componentof modern hydrologicmonitoring systems. illustratesthe use of computersin a real-time telemetry system.

NETWORKS ETEOROLOGICAL e.4 HYDROLOGICAL-M

Most modern hydrologic-meteorologicnetworks are designedto provide real-time releasingflows for irrigainformation for purposessuchas hydropowerscheduling, models,regulatingreservoirdischarges, and testinghydrologicsystem tion, developing allocating water from multiple sources,streamflow forecasting,tracking pollutant transport, and enforcing environmental regulations. Hydrological-meteorological

148

CHAPTER9

MONITORING NETWORKS

Water level

Weather station

Computer

,/: :
Modem Data Collection

(Courtesy Figure9.3 Computer usein a real-time telemetry system. of Sierra-Misco, Inc., Environmental Products, Berkeley, CA.)

networks may be designedto monitor physiographic, climatic, hydrologic,biologic, and chemicalfeatures, or combinations of these,in a region or river basin.They must have gaugedensitiesand distributionsthat are sufficientto permit interpolationbetweengaugesitesin a mannerpermitting valid conclusions to be drawn for the entire areacoveredby the network. Typically, measureinents are madeof suchvariablesas precipitation, solar radiation, temperature,relative humidity, barometric pressure, snow depth, soil moisture,wind speed,streamflow,and water quality. In any event, special basin or regional climatic factors must be given due consideration.Each hydrological-meteorological network is different in its purposeand setting and thus its designmust reflectboth the spatialand temporally varying featuresat the locality to be monitored along with the objectivesof the monitoring program.s

REFERENCES 149

r Summary
Monitoring of hydrologicsystems is essential to better understanding interof system actions and to the designand testing of hydrologicmodels.It is also the meansby which a determinationcan be made of the effectiveness of measures taken to alter performance. watershed

REFERENCES
t . S. J. Nix and P. E. Black (eds.),Proceedings of the Symposiury. on Monitoring, Modeling,
and Mediating Water Quality. Bethesda, MD: American Water Resources Association, 1987. 2. S. A. Dressing, "Nonpoint Source Monitoring and Evaluation Guide," in Ref. 1, pp. 69-78. J. J. Lawrenceand A. S. Y. Chau, "Quality Assurancefor EnvironmentalMonitoring," in Ref. I, pp. 165-176. 4. H. E. Postand T, J. Grizzard, "The Monitoring of StreamHydrology and Quality Using Microcomputers," in Ref. 1, pp. 199-208. 5 . P. J. Gabrielsen and A. J. Carmeli, "Operation of a Hydrologic-MeteorologicMonitoring Networkin a Severe Winter Environment," in Ref, l, pp. lI3-122.

PARTTHREE .

WATER SURFACE HYDROLOGY

C h a p t e r1 0

Runoffand the Catchment

r Prologue
The purposeof this chaPteris to: . ' . . . therunoff process. Expandon definitionsof termsfrequentlyusedin describing Presentconceptsof surfacerunoff and drainagebasin discharge. Introduce elementsof drainagebasin geomorphology' characteristics. of watershed Describequantitativemeasures Familiarize the readerwith elementsof frequencyanalysis'

Surfacewater hydrologydealswith the movementof water alongthe earth's surface as a result of precipitation and snow melt. Detailed analysisof surfacewater flow is highly important to suchfieldsasmunicipaland industrialwater supply,flood control, water qualreservoirdesign,navigation,irrigation, drainage, stieamflowforecasting, management. wildlife and flsh and recreation, ity control, water-based precipitationandrunoffis influencedby variousstormand The relationbetween of thesecomplexitiesand the frequentpaucity of adeBecause basin characteristics. to relaterainfall quaterunoff data, many approximateformulashavebeendeveloped the whereas statements, empirical crude usually were andrunoff. The earliesfofihese processes. physical on based equations descriptive trend now is to develop

BASINS AND DRAINAGE WATERSHEDS 10.1 CATCHMENTS,

the land surface-some Runoff occurswhen precipitationor snowmeltmovesacross The land areaover lakes' and streams artificial or natural of which eventuailyreaches surface contributes that area land the and the catchment which rain falls is called in size few acres be a can This a watershed. is called point of interest runoff to any subwatersmaller many contain can watershed A large miles. or thousandiof ,quut" sheds. awayfrom high water and groundwater Streamsand rivers conveyboth surface The tract problems' groundwater rising and flooding surface preventing water areas,

154

CHAPTER 1O RUNOFF AND THE CATCHMENT of land (both surfaceand subsurface) drainedby a river and its tributariesis called a drainage basin. A watershedsuppliessurfacerunoff to a river or stream,whereasa drainage basinfor a given streamis the tract of land drainedof both surface runoff and groundwaterdischarge. Rain falling on a watershed in quantitiesexceeding the soil or vegetation uptake becomes surface runoff. Waterinfiltrating the soil may eventuallyreturn to a stream and combine with surfacerunoff in forming the total drainage from the basin. The network of overlandflow courses and defineddrainagechannelscomprisethe watershed.Surfacerunoff from tracts of land beginsits journey as overlandflory, often calledsheetflow,beforeit reaches a definedswaleor channel,usually beforeflowing more than a few hundredfeet. The lines separating the land surfaceinto watersheds are called divides.Thesenormally follow ridges and moundsand can be delineated using contourmaps,field surveys, pairs of aerial photographs or stereograph to identify gradient directions.

Area Contributing
In the majority of hydrologicanalyses, the magnitudeof total surface areacontributing direct runoff to somepoint of interestis needed. Because of variationsin topography, the true surface areacannotbe easilymeasured. The horizontalprojectionofland area is easilyobtainedand normally adoptedin hydrologiccalculations.This resultsin an error in actual watershedarea whereverthe projected area is less than the actual. Somesurfaceareain watersheds may not contributeto surfacerunoff, so the error in using the projectedwatershed areais somewhatoffset.

PartialArea Hydrology
For light storms,or for someflat areas,portions of the catchmentdo not contribute to runoff. Precipitationfalling on or flowing into depressed or blockedareascan exit only by seepage or evaporation,or by transpirationif vegetated. If sufficientrainfall occurs, such areasmay overflow and contribute to runoff. Thus the total area contributing to runoff varies with the intensity and duration of the storm. Methods for incorporating this phenomenonin hydrologic studies are calegorized under proceduresfor partial area lrydrology. In partial areahydrology,watershed methods areasare dividedby one of several into contributing (active) and noncontributing (passive)subareas. For infrequent (severe) storms, largerpercentages may contributeto the peak surface of the watershed flow and volume of runoff, which are the primary variables of interest to design engineers. For more frequent storms,significantly smaller portions of some waterpartial areahydrologyis seldomincorposhedsmay contribute.As a consequence, ratedin hydraulic structuredesign,and is of greaterinterestin watersupplyand water quality studies. As will be shownlater (Chapter12) unit hydrograph theory andrunoff curve numbermethodsare basedon linearity of rainfall and runoff, and assume that the full watershed contributesto runoff in all stormsand in proportional amountsat different times in the samestorm. Application of thesemethodsto watersheds that havesignificantnoncontributingzonescould, and do, introduceerror ifthe zonesare nof first delineated and the distributedeffectsproperly modeled.

10.2 BASINCHARACTERISTICS AFFECTING RUNOFF

155

Subdivisionof contributing from noncontributingareashas traditionally been subjectively accomplished from siteinspection, topographic and soilsmaps,and aerial photos.Soils having good drainageclassifications, or dark tones or colors on aerial photos, can often be consideredas passiveareas.Other signs of noncontributing areaswould includepresehce of wetlands,grassed areas,rooftops (unlessconnected to the drainage), terraces, erosioncontrol structures, stockwateringponds,and flood control dams. Boughtontdeveloped a quantitativemethod of determiningthe proportion of a watershed that contributessurfacerunoff in different storms,and at different times duringthe samestorm.by analyzingrainfall andrunoff records.His logic is asfollows: 1. Watersheds can be idealizedas a group of "surfacestorage capacity" cells, eachrepresentinga fraction of the watershedarea and eachhaving some capacityto abstractrainfall into storage, infiltration, or evapotranspiration. 2. Runoff from eachcell occurs when rain fills the surfacestoragecapacity. 3. Runoff occursfrom the cell with the smallestcapacitybeforeflowing from the cell with the next largestcapacity(this is an assumption by Boughton that has not been fully verified). 4. Using theseprinciples,storm data for the watershedare evaluatedfirst to find thosein which runoff occurs only from the area of smallestcapacity. This is done using a graphicalmethod outlined in the article that looks at slopechanges in the rainfall-runoff graph.Both the capacityof the cell and its area as a percentage of the watershed are estimated. 5. After subtracting the contributionto runoff from the smallestcapacitycell, the capacityand contributingareafor the secondsmallestcapacitycell are determinedby the sameprocedure. 6. The process is repeateduntil all the runoff is accountedfor, or until 100 percent of the watershed is contributing,whicheveroccurs first. When Boughton applied the procedureto a test watershed,r it was found that runoff occurredfrom the entire watershed on only 3 of 30 eventsin the l5-year study period.In abouttwo-thirds of the runoff events, discharge occurredonly from the cell with the smallestsurfacestoragecapacity.

10.2 BASINCHARACTERISTICS AFFECTING RUNOFF

The natureof streamflowin a regionis a function of the hydrologicinput to that region and the physical,vegetative, and climatic characteristics. As indicatedby the hydrologic equation,all the waterthat occursin an areaas a result ofprecipitation doesnot appearas streamflow. Fractionsof the grossprecipitationare divertedinto pathsthat do not terminatein the regional surfacetransport system. Precipitation striking the groundcango into storage on the surface or in the soil andinto groundwater reservoirs beneaththe surface.The characterof the soil and rocks determines to a large extent the storagesystem into which precipitatedwaterwill enter.Opportunity for evaporation and transpirationwill also be affectedby the geologicand topographic natureof the area.

156

CHAPTER1O RUNOFFAND THE CATCHMENT

StreamPatterns
Wind, ice, and water act on land surfaces to createseveraltypes of drainagepatterns seen in nature. The particular design that results is a function of several factors including slope,underlying soil and rock properties,and the historiesof hydraulic action, freeze-thaw activity, and sedimenttransport.

(a) Dendritic system

(b) Trellis system

(c) Radial system

(e) Meandering stream

(f) Braided stream

(g) Anabranching stream

(h) Reticulate stream

and individual streamshapes). Fi-gxre 10.1- -Streampatterns(combinedsystems

10.2 BASINCHARACTERISTICS AFFECTINGRUNOFF

157

Typical streampatterns are shownin Fig. 10.1.The most common,dendritic, is characterizedby numeroussmall tributariesjoining at right angles into higher-order streams,eventually forming the major rivers in the region. The smaller tributaries often occur in sufficientquantitiesthat little land surfaceareais left unintercepted by a definedchannelof someform, The maximum overlanddistancebetweenchannels in theseareasseldomexceeds a few hundredfeet. Trellis patterns are characterizedby long main streams intercepted by numerous shorterright-angletributaries.They are commonin the Appalachians (EasternUnited States)and are also seenin the Rocky Mountains along the foothills, Multi-basin patterns,also called derangedsystems, occur in low gradient swampy areaswith numeroussurfacedepressions and normally haveonly a few tributaries.Theseoccur in glaciated,windblown,and permafrostareas,and are common in plains and mountain valley regionsof the United States. Radial patternsare typically found in foothill areasor mountain areaswith more advancedsoil development. Individual streamsfavor one or more of the four patternsshown on the lower portion of Fig. 10.1.Streams are rarely straightexcepton steepslopes in homogeneous materials. Braided streams are characterizedby numerous interconnected channels flowing aroundand overislandsandbars,inundatingmostduringhigh flows. Braided streamsare generallytransportinglarge amountsof sediment,but often less than the amount supplied. They have been called incipient forms of meandering streams dueto the fact that many revert to meandering or other forms when sediment suppliesor other factorschange. Meanderingin an otherwisestraightchanneloccurs as a result of transverse currents.Thesecurrents are consideredto be the result of forcesactingon the streamparticles,includingbed and bank shearforcesand coriolis effects. In evaluatingthe effects of changesin streamflows,a relationship known as Lane's Law is often applied.It states that the productof bed slopeand waterdischarge is proportional to the product of sedimentsize and transportrate. Changingany one of thesefour terms results in the likelihood of a shift in one or more of the others. Constructinga reservoit,for example, reduces the sediment transported into the reach just downstream.By Lane's Law, either the slope or dischargemust decrease or sedimentparticle sizemust increaseto offsetthe changein sedimenttransport.Most often, the slope decreases when the sediment-hungry flows deepenthe bed in the reach.Degradation(downwardcutting) of the bed of the Missouri River, for example, has occurredbelow someof its upstreamreservoirs, isolatingboat marinasand water intake structuresin somelocations.

Geomorphology of Drainage Basins

The principal geologicfactorsthat affect surface watersare classified aslithologic and structural. Lithologic effects are associated with the composition,texture, and sequenceof the rocks, whereas structuraleffectsrelatemainly to discontinuities suchas faults and folds. A fault is a fracture that resultsin the relative displacement ofrock that was previouslycontinuous.Folds are geologicstrata that are contortedor bent. Variationsin the erodibility of the different strata can easily lead to the creation of distinctive forms of drainagesystems. Both large-scale andlocal effectson the storage and movement of surface waters existbecause of geologicactivity and structure.For example,drainagepatternsare

1 58

CHAPTER10

RUNOFFAND THE CATCHMENT

determinedto a large extentby the nature of land forms. On the other hand, flowing surfacewatersalso affect the surfacegeometrythrough the processof erosion.Thus significant land forms resulting from volcanic activity, folding, and faulting affect drainage, whereas patterns,havingbeengenerated, drainage can alsomodify the land forms by creatingvalleys,deltas,and other geomorphicfeatures. Streamsare classified asbeingyoung,mature,or old on the basisof their ability to erodechannelmaterials.Youngstreamsare highly active and usually flow rapidly so that they are continually cutting their channels.The sedimentload imposed on thesestreamsby their tributariesis transportedwithout deposition.Mature streams are those in which the channel slope has been reduced to the point where flow velocitiesarejust able to transportincoming sedimentand where the channeldepth is no longerbeingmodifiedby erosion.A streamis classified asold whenthe channels in its system havebecomeaggraded. The flow velocitiesof old streamsare low dueto gentleslopes that prevail.Wide meander belts,broad flood plains,anddeltaformation are alsocharacteristic of old streams. The lower reaches of the Mississippi, Rhine, and Nile are examples. Flows in young river basinsare often o'flashy,"*h"t"u, sluggish flows are common to older streams. The description of a drainagebasin in quantitative terms was an imponant forward stepin hydrologyand can be tracedback in largepart to the efforts of Robert E. Horton.2Strahler,Langbein,and othershaveexpanded Horton's original work.3-a To quantify the geometry of a basin, the fundamental dimensionsof length, time, and mass are used. Many drainagebasin featuresthat are important to the hydrologistcan be quantified in terms of length, length squared,orlength cubed. Examplesare elevation,streamlength, basin perimeter,drainagearea, and volume. The conceptof geometricsimilarity can be appliedto drainagebasinsjust as it is to many other systems.3 Most readerswill be awareof model-prototypestudiesof aircraft, dams,and turbomachinery. Suchstudiesinvolveconsiderations of geometricas well as dynamic similarity. In the samemannerthat inferencesas to the operationof a prototype can sometimes be drawn from a geometricallysimilar model, inferences canalsobe drawnaboutthe operationof one drainage areaon the basisof information obtained from a similar one. Perfect similarity will never be realized if natural drainagesystems are compared,but striking similarities have been observedwhich can often be put to practical use.

Measuresof DrainageBasin Characteristics

Important measures of drainagebasin characteristics include overlandflow lengths and streamlengths.The conceptof streamorder is often associated with the dimension of streamlength.

AFFECTINGRUNOFF 1O.2 BASIN CHARACTERISTICS

159

'/, "81 {,'v '\'rr

;)

Figure 10.2 Sketch indicating definition of streamorder.

The order to the streamorder, provided that a large enoughsampleis investigated. number permits comparisonsof drainage systemsthat are quite different in size quantity.Suchcomparisons shouldbe madeat the numberis a dimensionless because locations in the two systemsthat have a similar geometry; that is, second-order streams, third-order streams,and so forth. of theh projectionsonto a Streamlengthsare determinedby the measurement If horizontal plane. Topographicmaps are useful for obtaining suchmeasurements. possible Z, of order a is definedas L,, then it is the meanlenglh of a streamsegment to determine Z, using2
tr-

-2!i, L,,
N,

(10.1)

of streamorder u. whereN, is the number of streamsegments Another measurerelated to stream length is the distanceL"o from a point of the center intereston the main streamto a point on the primary channelthat is nearest of gravity of the drainage arca (center of gravity of the plane atea of the drainage basin).Studiesof basinlag (time betweenthe centersof massof effectivestorminput and the resulting runoff ) havemadeuse of this dimension. of a drainagebasin development in the physiographic Of particular significance is the overlandflow length Ls. This is the distancefrom the ridge line or drainage alongthe path of surfaceflow which is not confinedin apy defined divide, measured flow channel.If a of this flow path with an established to the intersection channel, then system, a larger drainage basic element of first order is the basin of the drainage basins. first-order for these can be determined flow length overland a representative One apprbachis to measurea numberof possibleflow paths from a map of the area these.In somecases(for example,with the rational method,Chapter and-toaverage

160

CHAPTER1O RUNOFFAND THE CATCHMENT

15), the use of the longestoverland flow length is prescribed,measuredfrom the upstreamend of the first-order stream to the most remote point of flow that will terminateat this point.

Areal Measurements
Just as linear measuresrelate to many factors of hydrologic interest, so do areal measures. For example, the quantity of discharge from any drainage basinis obviously a function of the areal extentof that basin. Correlationshavebeenobserved between the average area, Au, of basinsof order u, andthe average length of streamsegments, 2,. Thesevariablesare often relatedby an exponentialfunction. For example,studiesof sevenstreamsin the Maryland-Virginia areaby Hack haveproducedthe relationship6
L : I.4Ao6

(10.2)

where L = the streamlength measured in miles to the drainagedivide A : the drainage area(miz) Hack's observations indicatethat as the drainagebasinincreases in size,it becomes longer and narrower; thus precisegeometricsimilarity is not preserved. Drainageareahas long beenusedas a parameterin precipitation*runoff equations or in simpleequationsindexingstreamflowto area or other parameters. Many early empirical equationsare of the form3 Q: cA: where Q a measureof flow suchas mean annual runoff A : the size of the contributine drainaeeatea

(10.3)

Valuesof c and m are determined by regression analysis(seeChapter26); Fig. 10.3 illustratesa relation of this form.

r u o
8 ( ( ) 6
= a

./

10

20

30 40 5060 80 100 Area tmi2)

200

Figure 10.3 Runoff-drainage arca correlation for five Maryland streams (1933 storm

RUNOFF AFFECTING 1O.2 BASINCHARACTERISTICS

161

includedefinitionsof the basinshapeandthe densityof the Other arealmeasures drainagenetwork or drainage density, definedas the ratio of total channelsegment lengthi cumulatedfor all streamorderswithin a basin to the basin area.The stream in a drainagebasin (total segments frequency is defined as the summation of all of all orders)divided by the drainagearea. number of segments

Channeland BasinGradients
havea very strongeffecton the surface basinand its channels of a drainage The slopes profiles exhibit the characteristic channel stream runoff pro""r, of thairegion. Most Figure 10.4illustratesthis direction. downstream slopeproceedingin a of decreasing particular.trait. Also illustrated in drop dividedby the channt elevation suchthat the areasbetweenthe ave that is, A, : Azin the figure. The g to describedrainaget asparameters describe make this clear. Some mathemalicalfunctions that are used to more fully streamprofilesarelinear, exponenti ical value to representthe Primal This factor,known as th Schwartz.T a uniform channelthat is equivalen sametraveltime. This factorhasber to the peak rate of runoff ) and maximum from the centerof massof rainfall excess discharge. In additionto the slopeofthe streamchannel,the overall land slopeofthe basin factor.A quantitativerelation betweenvalley wall slopes is an important topographic A commonly usedmethod has been derivedby Strahler.3 and streamcnannetstopes by Horton'SThe method presented been has of determiningthe slopesof a basin map of the drainagearea topographic a grid over a transparent involvessuperimposing with the drainage intersections its b"t*""n in question.nacfr grlJnne is measu."d is alsoneeded' line a contour with grid line each of divide; the numberof intersections using made be then A determinationof the land slopecan nsec0, S:--t (10.4)

Distance from head of stream

Figure 10.4 Typical streamprofile'

162

oHAPTER 10 RUNoFFAND THEoATCHMENT

Drainage boundary Horizontal grid line

uContour'

Contour interval = 50 ft scaie

Figure 10.5 Determinationof meanland slope:numberof vertical intersections: 72; tumber of horizontal intersections= 120; total : 103,900ft; total length of length of vertical grid segments : 101,200 horizontalgrid segments ft. 72x50 120 x 50 : 0.035 : 0.059 ftlft Ss ftlft " : 10 t , 2 0 0 103;900 .s + (.. 0.035 + 0.059 : 0.047 Mean slope : ftlft 2

,s:

where n : the total number of contour intersections bv the horizontal and vertical grid lines t (horizontal and vertical) the total length of grid line segments h - the contour interval 0 * the anglemeasured betweena normal to the contoursand the grid line
L _

Because0 is very difficult to measureit is often neglected,and separatevaluesof average slopein the horizontal and vertical are computedand then averaged to obtain an estimate of the meanland slope.This procedure is illustratedin Fig. 10.5.

Area-ElevationRelation
How the areawithin a drainagebasinis distributedbetweencontours(Fig. 10.6)is of interestfor comparingdrainagebasinsand gaining insight into the storageand flow characteristics of the basin.For suchstudies.an areadistribution curve suchas that shown in Fig. 10.7 is used.The curve can be obtained by planimeteringthe areas

RUNOFF AFFECTING CHARACTERISTICS 10.2 BASIN

163

. ' , \: \ I . i/ z ' r '

\

l l \ ' / t r 1 l l \ / / / / r r \ f t /

,ffi_-_.w;i.i
.Figure 10.6 Topographic area map of WendyRun drainage lines. 20-,40-,and60-ftcontour showing betweenadjacentcontoursor by using a grid as in Fig. 10.5 and forming the ratio of the number of squaresbetweencontoursto the total number of squarescontained within the drainageboundaries.The mean elevationis determinedas the weighted betweenadjacentcontours.The medianelevationcan be deteraverage of elevations curvesas the elevationat 50 percent, from the area-elevation rnined

300

E 2so
o

I o
r! '

2oo l)u
100 Median elevation

10

20

30

40

50

60

70

80

90

100

of area Percentage

Figure 10.7

An area-elevation distribution curve.

1 64

CHAPTER10

RUNOFFAND THE CATCHMENT

Drainage Basin Dynamics

Geomorphology' likg hydrology,was largely qualitative in nature in its formative years'With the passing of time and the greatlr needfor reliablequantitativeinformation, the science hasprogressed to the point whererational relationsbetween variables are being developed. Theserelationsire usually intendedto quantify theinieractions betweenthe factorsthat modify the land forrnand the land iorm ifsef. In addition, equations relatingthegeomorphic propertiesto hydrologic,climatologic, or vegetative factorsare beingsought.some of the iunctional ielations of particulir significance to the hydrologistwill be discussed in the following chapters.

10.3 RUDIM ENTARY PRECI PITATION-RUNOFF RELATIONSHIPS

ritation and runoff has beento plot annual rend line, and estimatethe percentageof rtities determinedthis way, however, are :e ofreliability is higherfor drainageareas rnal or other types of variation, that is, an e procedure.The resulting equationtakes

the form

o:($)rr-Pr)
1902.
I

(10.5)

1898r

l .
a 903

^ 4 0
19 l0r
- J 6 .E

I t28 a Slope

8 3 6
. ) J +

--l-I --i-1920 1l : -1 9 1 8 i q)5-oL f--a 7921o 193 0 . 1897 90 1924 1932

X 7 ,7 23
1 926
a

0.57

trg

29_

32

o1!

1 0 1 1 t 2 t 3 ! 4

Runoff (in.)

Figure 10.8 Annual precipitation and annual runoff in the NeoshoRiver basin above Iola, Kansas.(U.S. GeologicalSurvey Data.)

165 10,3 RUDIMENTARYPRECIPITATION-RUNOFFRELATIONSHIPS

S : the slopeof the line (LPILQ) Pa : & baseprecipitation value below which Q is zero From Fig. 10.8 the relation for the examplewould be Q: 0'57(P 24) where Q and P are the annual runoff and precipitation, respectively,in inches' relation indicatesthat scatterof severaldata points from the assumed Considerable approximationsin For rough with care. used be this type of computation should however.Equahelpful, frequently are methods preliminary planning studies,such precipas antecedent parameters such if other improved Eq. 10.5are tionsresembiing relations Such included. are characteristics and storm itation, soil moisture,season, or graphicalmethods'Linsley techniques using multiple regression can be described correlafor developing methods of treatment and co-workerspresenta very complete tions involving severalvariables.e Soil moisturerelationsnormally havea soil moistureindex as the independent soil moistureare not generof actualantecedent variable,sincedirect measurements at the beginning groundwater,flow are inserted ally practical. Indexesthat havebeen values Groundwater evaporation.lo precipitation, basin and of the storm, antecedent days few within a occurring precipitation of shouldbe weightedto reflect the effects results. affect rains may previous from of the storm becausesoil moisture changes can be employedto estimatesoil moistureamounts, measurements Pan-evaporation since eviporation is related to soil moisture depletion.tl Antecedentprecipitation precipitationis readily indexes(API) haveprobably receivedthe widestusebecause basin. of the deficiency measuredand relatesdirectly to moisture precipitation index is A typical antecedent (10.6) Po: aP6'l bP, -f cP., where

p,." : po,

P.:

index (in.) precipitation the antecedent

fi"HTlJ:;fJil:rr:Hifl3*ff"-e

for and presenr vear

Coefficientsa, b, andc are found This index links annualrainfall and runoff values.r2 to producethe best correlationbetween by trial and error or other fitting techniques precipitation index. The sum of the coefficientsmust the runoff and the antecedent be 1. haveproposedthe following API for use with individual Kohler and Linsley13 storms:
+''' P o: b r P l + b 2 P 2 + btPt

(10.7)

on P refersto precipitationwhich occurredthat many daysprior wherethe r subscript to be a function b (lessthan unity) are assumed to the given storm,and the constants In daily by correlationtechniques' of t. Valuesfor the coefficientscan be determined to be related to / by evaluationof the index, b, is considered b,: K' (10.8) normallyrepoftedin the range0.85-0.98.The initial constant whereK is a recession

166

cHAprER1o RUNoFFANDTHE cAToHMENT

value of the API (P"s)is coupled to the API / days later (P",) by
Por: PagK'

(10.e)

To evaluatethe index for a particular day basedon that of the precedingone, Eq. 10.9becomes
Po, : KPox

(10.10)

because t : I. Various empirical relations for API have been proposed.Most are based on correlating two or three variablesand at best yield only rough approximations.In many casesthesewere developed physical principlesor dimenwithout considering sional homogeneity. An addedshortcomingis that many formulas fit only a specific watershedand have little generalutility. Empirical equationsdemandgreat caution and an understanding of their origin. EXAMPLE 10.1 PrecipitationdepthsP, for a l4-day period are listed in Table 10.1. The API on April 1 is 0.00. Use K : 0.9 and determinethe,API for each successive day. Solution. Equation 10.9 reducesto API,: K(API,-1)+ P,

which was applied in developingthe successive valuesof API, in Table 10.1. TABLE 10.1
(A Precipitation April 1 2
J

API,

'6 7 8 9 10
ll

t2 l3 t4

0.0 0.0 0.5 0.7 0.2 0.1 0.0 0.1 0.3 0.0 0.0 0.6 0.0 0.0

0.00 0.00 0.50 1. l5 t,24 t.22 1.10 1.09 1.28 t.l5 1.04 1.54 1,39 |.25

10.4 STREAMFLOW FREOUENCY ANALYSIS

Hydrologists estimatestreamflows. Two approaches are employed. The first is a physical processes approachin which runoff is computed on the basis of observedor precipitation.The secondis foundedon statisticalanalyses expected ofrunoffrecords .- --vvi{}out resott-to precipitation data. Such investigationsusually include frequency

ANALYSIS FREQUENCY 10.4 STREAMFLOW

167

C a o a ao

12 months

1.0 Example 1.0 in drought with R.I. = 17 yr

x.

0.1 10
RecurrenceInteryal (Yr) 100

for FiveRivdataconsolidated frequency Figure 10.9 Low-flow (AfterWhipple.r5; ers. studies(Chapter 27) to indicatethe likelihood of certain runoff eventstaking place. A knowledgeof the frequencyof runoff eventsis helpful in determiningrisks associare analyses Frequency or anticipatedoperatingschemes. atedwith proposeddesigns of miximum (flood) andminimum (drought)flows.ra'1s usuallydiiectedtowardstudies Figure 10.9 illustrates a typical drought frequency analysis.Unfortunately,many existingrunoff records are short-term; as a result they limit utility for reliable frequency analyses.Few adequaterecords are availableearlier than about 1900. In some cases,sequentialgeneratingtechniques(Chapter 22) can be used to develop syntheticrecords. Time-series analysesare particularly pertinent to the problem of estimating trends, cycles, and fluctuationsin hydrologic data. They also permit derivation of by which syntheticrecordsof runoff can be developed. generatingprocesses

Intervaland Frequency Recurrence

interval over a long period of The recurrenceinterval (R.I.) is definedas the average magnitudeof somehydrologicvariableis at least yearsduring which a corresponding met. This parameteris also called the return period, and sometimes,though less appropriately,thefrequency of the event.For the examplein Fig. 10'9, droughtsless in 8 ofthe 136yearsofrecords.The 1.0in. droughthasan average than 1 in. occurred one year of recurrenceinterval of about 17 years.Statedanotherway,on the average, 1.0 inch. most of at a drought is expectedto experience every l7-year sequence = or about 0'059, is 8/136 drought Similarly,eachyearthe probabilityof a 1.O-in. is the and or probability frequency, 6 percent. This is defined as the exceedence could drought 1.0-in. that the re-iprocal ofthe return period.It shouldbe obvious

168

CHAPTER 10 RUNoFFAND THE oAToHMENT occur in any year,or in severalconsecutive years.This type ofanalysiscannottell the investigatorwhat will happenthis year or next, and allows only an estimateof the average recurrenceinterval and the probability of occurrencein any given year. This subjectis fully developed in Chapter27.

10.5 STREAMFLOW FORECASTING

Surfacewaterhydrologyis basicto the designof many engineering works and important in waterquality management schemes. In addition, the ability to providereliable forecasts of flowsfor shortperiodsinto the future is of greatvaluein operatingstorage and other works and in planning proper actions during times of flood.e'16 A good exampleis the operationof a reservoirwith an uncontrolledinflow but with a means of regulatingthe outflow. If information on the natureof the inflow is determinable in advance, then the reservoircan be operated by somedecisionrule to minimize downstreamflood damage.Suchoperationscan be computerizedto continually improve estimates basedon incoming dataandthus offer direction on the natureof the releases to be made.For river forecaststo be reliable, adequate, dependable data on various watershedand meteorologicconditions are neededon a continuing basis.Modern monitoring stationscapableof telemetering data to computercontrol centersprovide an important supportfunction for forecasting. The methodsusedto forecastflows are basically the same ones empfoyedin design: precipitation-runoff equations,unit hydrographs, watershed mldels, and flow-routing techniques.

r Summary
Runoff is probablythe most complexyet most important hydrologicprocess to understand.It has attractedthe attention and focus of engineersand scientistsand comprisesthe greatestpercentage by far of most hydrology textbooksand publications. The conceptsintroducedhere will be more fully developed in the next six chapters, as well as in significantportions of Parts Five and Six.

PROBLEMS

10'1'ffi,',T'#:fl '#':iJf :Jfr *:l;J:ff ;J3;:1,!;";ffi :?'fi#.n.}il:H:i* Hfi

what purposesmight you use this? 10.2. Selecta rain gaugerecord of interest.Use the annual valuesas data to calculatethc coefficientsof an antecedent precipitation index of the form of Eq. 10.8. 10.3. Determinethe drainagedensityof the basin shown.Area : 6400 acres.Lengths are in miles.

REFERENCES 169

Lengths of channel segments in mi between points

Figure for Problem 10.3 from one or two other hydrologytexts, or glossaries L0.4. Using any dictionary,plus indexes find and compare definitions of the following terms: runoff, direct runoff, direct surface runoff, surface ruryoff,surface water, overlandflow, streamflow,drainage, watershed,catchruen&inage basin, subbasin,dr ainage divide.

"systematic Procedure for Evaluating Partial Areas of Watershed 1. Boughton, W. C., Runoff," Proc. ASCE J. Irrigation and Drainage Engineering116, 1 (February 1990). "Drainage Basin Characteristics,"Trans.Am. Geophys.Union 13(1932). 2. R. E. Horton, "Geology-Part II," in Handhok of Applied Hydrology. New York: 3. A. N. Strahler, McGraw-Hill, 1964. "TopographicCharacteristics of DrainageBasins,"U.S. Geological 4. W. B. L4ngbeinet al., Survey,Water Supply Paper,968-c(I941 ). "Erosional Developmentof Streamsand Their Drainage Basins:Hydro5. R. E. Horton, physicalApproach to Qualitative Morphology,"Bull. Geol. Soc.Am.56(1945). "studies of Longitudinal StreamProflles of Small Watersheds," Tech. Rept. 6. J. T. Hack, 18, Columbia University,Departmentof Geology,New York, 1959. 7. A. B. Taylor and H. E. Schwartz,"IJnitHydrographT ag and PeakFlow Relatedto Basin s. Union 33(1952). s," Trans.Am. Geoplty Characteristic "Discussionof Paper,Flood Flow characteristicsby c. S. Jarvis," ?ans. Horton, R. E. 8. ASCE 89(1926)'. 9. R. K. Linsley, Jr., M. A. Kohler, and J. L. H. Paulhus,AppliedHydrology,2nd Ed' New York: McGraw-Hill, 1975. 10. ven Te chow (ed.), Handbook of Applied Hydrology. New York: McGraw-Hill,1964. 11. R.K.Linsley,Jr.,andWC.Ackerman,"MethodofPredictingtheRunofffromRainfall," Trans.ASCE 107(1942). 12. S. S. Butler, EngineeringHydrology. EnglewoodCliffs, NJ: Prentice-Hall, 1957.

170

oHAPTER 10 RUNoFF ANDTHEoAToHMEI 13. M. A. Kohler and R. K. Linslev. , / WeatherBureau,Res.Paper34, l95L T4, Leo R. Beard, "statistical Methodsin Hydrology," U.S. Army EngineerDistrict, Sacrf mento,CA,1962. 1 5 . William W. Whipple, Jr., "RegionalDrought FiequencyAnalysis," Proc. ASCEJ. Irrigd

16.

1'1 Chapter

Hydrographs

r Prologue
The purposeof this chapteris to: . Characterizea hydrographas a time plot of the dischargeof surfacerunoff arld groundwaterfrom drainagebasins. ' Introduce the componentsof hydrographs so that the reader can rel9le them of runoff presentedin subsequdnf-chapters. to the quantltative assessments . Describethe time relationshipsmost commonly usedin hydrographanalysis. Hydrographanalysisis the most widely usedmethod of analyzingsurfacerunoff. Its presentationin most textbooks is normally confined to one chapter. Becauseof to the three chaptersare dedicated in hydrographanalyses, numerousdevelopments the conceptsintroand expands subjectin this text. Chapter 11 defineshydrographs duced in Chapter 6. The conceptsreferred to as unit hydrographtechniquesare packaged togetherin Chapter 12. Individual streamflowhydrographshapesvary as called hyflow travels downstream,and the conceptsfor analyzingthese changes, ' in Chapter 13. drographrouting methods,are presented

HYDROGRAPHS 11.1 STREAMFLOW

the rateof flow at all pointsin time duringand after hydrographprovides A streamflow or computed(synthea storm or snowmeltevent.l Hydrologistsdependon measured sized) hydrographsto provide peak flow rates so that hydraulic structurescan be designedto accommodatethe flow safely.Becausea hydrographplots volumetric flow any two rates againsttime, integration of the area beneath a hydrographbetWeen the point of interestduring the points in time givesthe total volume of water passing allow analysisof sizesof time interval. Thus, in addition to peak flows,hydrographs storage tanks, detentionponds,and other facilities that deal with volumes reservoirs, of runoff. A knowledge of the magnitude and time distribution of streamflow is and environmentalplanning. of watermanagement to many of theseaspects essential

172

CHAPTER11

HYDROGRAPHS
Storm period hydrograph
o u

q b0

Y41 oar.

Hydrograph in period of no direct runoff and where no reservoir regulation exists rcflects discharge from groundwatet Storm period hydrograph End of Beginning of

Inflection point

\?

direct runoff

direct runoff

,l

Continuous hydrograph

Figure 11.1 HydrographO"o;;.

SHAPE AFFECTING HYDROGRAPH 11.2 FACTORS

A hydrographhas four componentelements:(1) direct surfacerunoff, (2) interflow, The rising pott(rn of a (3) groundwater or baseflow, and (4) channelpregipitation.2 in the vicinity of the\eak region curye;the hydrograph is known astheconcentration The shapeof a portion is recession.3 the the falling is called the crest segment;and properties. pattern and basin precipitation characteristics hydrographdependson presented. Figure 11.1illustrates the definitions

Processes Precipitation-Streamflow
to saturatethe upper Duriirg a given rainfall, water is continually being abstracted levelsof the soil surface;however,this saturationor infiltration is only one of many Rainfall is also interceptedby trees, plants, 4nd roof continuous abstractions.4-6 surfaces,and at the same time is evaporated.Once rain falls and fulfills initial collect falling rain to form small requirementsof infiltration, natural depressions puddles,creatingdepression storage.In addition, numerouspools of water forming within the waterand impermeablesurfaces detentionstoragebuild up on permeable shed.This storedwater gathersin small rivulets, which carry the wateroriginatingas then into larger channels,and flnally as channel overlandflow into small channels, 11.2aillustratesthe distribution of a prolonged Figure to the watershed outlet. flow uniform rainfall. Although such an event is not the norm, the conceptis useful for storagewould be distributed. showingthe mannerin which detentionand depression possesses a certain amountof baseflow In general,the channelof a watershed during most of the year. This flow comesfrom groundwateror spring contributions as the normal day-to-dayflow. Dischargefrom precipitation and may be considered excess-that is, after abstractions are deducted from the original rainfallthe direct runoff hydrograph(DRH). Arrival of direct ninoff at the outlet constitutes enoughtime continues, for an initial rise in the DRH. As precipitationexcess accounts the qlapqgs for progressively distant areasto add to the outlet flow. Consequently,

11.2 FACTORSAFFECTINGHYDROGRAPTI SHNPC

179

Depressionstorage
fi

Time (b)

Figure 11.2 (a) Distribution of a uniform storm rainfall for condition of is ultimately no interceptionloss.Note that all water storedin depressions is also subjected to evaporated or infiltrated while somedetentionstorage (b) Equilibrium discharge theselosses. hydrograph.

duration of rainfall dictatesthe proportionatearea of the watershedamplifying the peak, and the intensity of rainfall during this period of time determines the resulting greatestdischarge.

Shapes Hydrograph
If the rainfall maintainsa constantintensityfor a long enoughperiod of time, a state of equilibrium dischargeis reached, as depicted by curve A in Fig. 11.2b. The inflection point on curveA often indicatesthe time at which the entire drainagearea contributes to the flow. At this time maximum storase of the watershedis only

174

CHAPTER11

HYDROGRAPHS

and partially complete.As rainfall continues,maximum storagecapacityis attained of condition The is reached. (runoff)] iqrliUri"t" finflow (rainfall) equals outflow Extended nature' in attained maximum storageand equiiibrium is seldom if ever its duration negateany rainfall lnuy o"Iu., but viriations in intensity throughout intensity' rainfall constant possibility of u IRH of the theoretical shapefor Anormalsingle.peakDRHgenerallypossessestheshapeshownbycurveBin in Fig. 11.2a.The time to peak magnitudeof this Fig. 11.2b rather tian iy the "ur',oJ onih" intensity and duration ofthe rainfall' and the size,slope, hyirograph depends for a once peak flow has beenreached capacityof the watershed. shape,and storage coming supply of its source given isolatedra'instorm,itt" ORg beginsto descend, suchas detentionand channel within the watershed largely from water accumulated storage. by visualizing processes involvedin forming the DRH can be better understood theprecipitationexcessaspartiallydisposedofimmediate$bysurfacerunoffwhile later from boundariesand is released a portion remainsheld within the watershed duration of the effects integrated Thus the shapeand timing of the DRH are storage. effect of the as well as factors and intensityofrainfall and other hydrometeorological capacity' storage upon the factorsof the watershed the physiogiaphic

COMPONENTS 11.3 HYDROGRAPH

how the hydrographcan be subdividedinto its com\It is important to understand ofprecipitation andwatershed shape andto look at the effecton hydrograph nenrpar15 purpose' Figures11.3and Il'4 areusedfor this features. graphshowingthe rate of streamflowwith respect A hydiographis a continuous strip recorderthat indicatesstage to time, ,rtr*ityoutained by -"anr of a continuous hydrograph which is then transformedto a discharge vefsustime (stale hydrograph), taken to generally is lrydrograph by applicationof a rating curve.Hereafter,the term hYdrograPh' indicate a discharge a pgriod Figure t t.gi lttustrai"* ih" hydrographof a permanentstreamduring groundbecause lrydrograph between precipitation events,known i" i bot" flow the base of modification cause water sustainsthe flow. Four general conditions of sets following the using HortonT flow hydrographshape.They aL describedby inequalities: Setl i<f F(S, Set2 i<f F)S, set3 i>f F(S, Set4 i>f F)Sr.

11.3 HYDROGRAPH COMPONENTS 175

(e)

(0

fo)

(h)

Figure 11.3 Effectsof stormandbasincharacteristics on hydrograph shape. or addedgroundwatercomponents develops. The entire effect of the storm would be to slightly reducethe soil moisturedeficiencysr. The field capacityis the amountof water held in the soil after excess gravitationalwater has drained. The conditions describedby Set 2 still do not produce direct surfacerunoff, although the comlponentsof interflow and groundwater flow are added to channel precipitation.The initial hydrograph would be modified,sincethe field capacityof the soil is exceeded. Figure 11.3b illustrates this condition. Note that deviation of the hydrographfrom the original baseflow curve is likely to be very small under these conditions. Figure 11.3c illustrates a casewhere surfacerunoff becomesa componentof flow because i > f. In this example,interflow and groundwater flow are zero,assoil

176

CHAPTER11 HYDROGRAPHS

End ofrainfall

9C

Groundwater flow

Time

of thehydrograph. Figure 11.4 Components

moisture deficiency still exists,although at a reduced level. Channel precipitation likewise constitutesa component. In the final set, Fig. 11.3d, all four componentsexist with rainfall intensity This casewould inflltration rate andthe field capacityofthe soil is reached. exceeding of a large storm event. be typical Figures 1l.je-h illustrate how hydrograph shape can be modified by areal Minor variations in rainfall and rainfall intensity and by watershedconfiguration.8 In intensity. storm in to variations linked are fluctuationsshownin thesehydrographs of section the upstream over storm to a pertinent Fig. 11.3eonly the delayingeffects 11'3g Figures condition. of this reverse the shows 11.3f Figure the areaareindicated. and h depict the comparativeeffects of basin geometry. interflow and channelprecipitation are grouped In most hydrographanalyses, Channelprecipitation begins independently. treated with surfacerunoff rather than with respectto time distribution Its the storm. with ends with inceptionof rainfall and tendsto contribution volume relative pattern. The is highly iorrelated with the storm water the rise and levels stream since proceeds, increasesomewhatas the storm and streams by area occupied of watershed fraction The areatendsto increase. surface percentage so the percent less, or of 5 order the on lakes is generallysmall, usually of rrnoff relatedto channelprecipitation is usually minor during important storms. rate up to by a slowly increasing Distribution of interflow is commonly characterized at the terminates that recession gradual a by period, followed the end of the storm 11.4 Figure hydrograph. flow base and hydrograph flow intersectionofthe surface and precipitation channel of the components of nature illustrates the approximate interflow. The baseflow componentis composedof the water that percolatesdownward asgroundreservoirand then flowsto surfacestreams the groundwater until it reaches an increase not show may or may groundwater hydrograph The water discharge.

11.4 BASE FLOW SEPARATION

177

'

during the actual storm period. Groundwateraccretion resulting from a particular storm is normally releasedover an extendedperiod, measuredin days for small watersheds and often in months or yearsfor large drainageareas. The surfacerunoff component consistsof water that flows overland until a streamchannelis reached.During large stormsit is the most significanthydrograph component. Figure 11.4illustratesthe surface runoff and groundwater components of a hydrograph. As pointed out in Fig. 11.3,the relative magnitudeof eachcomponent for a given storm is determinedby a combinationof many factors.Hydrographsare analyzed to provideknowledgeof the way precipitationand watershed characteristics interact to form them. The degreeof hydrographseparation requireddepends on the objective of the study. For most practical work, surfacerunoff and groundwater components only are required.Research projectsor more sophisticated analyses may dictate considerationof all components.When multiple storms occur within short periods,it is sometimesnecessary to separate the overlappingparts of consecutive surfacerunoff hydrographs.

11.4 BASEFLOWSEPARATION
Severaltechniques are usedto separate a hydrograph's surface and groundwater flows. Most are basedon analyses recession or depletioncurves.If there is of groundwater no addedinflow to the groundwater reservoir,and if all groundwater discharge from the upstreamareais interceptedat the stream-gauging point of interest,then groundwater discharge can be described by either e'to er: eoK' or er: eo-K' where eo : q, : K: e: a specifiedinitial discharge the discharge at any time / after flow 46 arecessionconstant baseof natural logarithms

( 11 . 1)

Time units frequently used are days for large watersheds and hours or minutes for small basins.A plot of either yields a straight line on semilogarithmicpaper by plotting / on the linear scale. groundwaterdepletioncharacteristics For most watersheds, are approximately stable,sincethey closelyfit watershed geology.Nevertheless, the recession constant varieswith seasonal effectssuchas evaporation and freezingcyclesand other factors. Becauseq, dt is equivalentto -dS, where S is the quantity of water obtainedfrom storage, integration ofEq. 11.1produces
t- 4o s-Q log" K

(rr.2)

This equation determinesthe quantity of water releasedfrom groundwaterstorage betweenthe times of occurrenceof the two discharges of interest,or it can be usedto calculatethe volume of water still in storageat a time some chosenvalue of flow occurs.To get the latter, 4, is setequalto zero and qsbecomes the reference discharge. Figure 11.5ais a plot of Eqs. 11.1and II.2 andprovides additionaldefinition.

178

CHAPTER11

HYDROGRAPHS

qo

.. ,?

K
\r3
t (a.)
?r

qo discharge at rcginningof arr interval discharge at :nd of an inten,a\ Qt

I
I

I
q d

I
/

I I

Interval = unit period in / whicht is expreissed .

qo
Qt

Time (b) flow model' Figure 11.5 Base to

t1

by variousmethodsto evaluate Groundwaterdepletioncurvescan be analyzed Data from a stream-gaugi.ng constantK. One of thesewill be described. the recession and shouldreflectrainlessperiodswith no upstreamregulastationare a prerequisite tion,'suchas a fesefvoir,to affect flow at the gaugingpoint. Otherwisean adjustment with its own enors is introduced. From the streamflow data, plot a portion of the recession hydrograph (Fig. 11.5b) to find values of dischargeat the beginning and end of selectedtime to 4e,whereasthose intervals.Flows at the beginningof eachinterval are analogous to q1.Next, selectseveraltime intervalsand plot correspondat the end are analogous values consecutive q,'s shownin Fig. 11.6.The time periodbetween ing qe'sversus curvesof of 4 shouldbe identical for eachdatum set. Figurestaken from recession times that still reflect surfacerunoff will usually fall below and to the right of a 45o with larger numbersfor line drawn on the plot. Thesevalueswill also be associated periods should approximately groundwater recession from true taken Points 4.

11,4 BASEFLOW SEPARATION 179

r.-,x. ^are points plotted from discharges for different groundwater discharge periods eo nd qt are taken for equal time

,/
o

I
A

,(
.5r"2'

/(^lI
- s l o p e o0 f A=
inEq11.1

9/ .,

fi:

/#

/
Figure 11.6 Graphical rnethod for determiningrecessionconstant K. (U.S. Departmentof Agriculture, Soil ConservationService.)

describe a straightline. Because canbefitted h : eo : 0 when4o: 0, a straightline graphicallyto the datapoints.The slopeof this line is qrfqo : K. Usingthis value,the depletioncurve plots as a straight line on semilogarithmic paper (r is the linear scale variable) or as a curveon arithmeticpaper,Fig. 11.5a.

Techniques Separation
r Severalmethodsfor baseflow separationare used when the actual amount of base flow is unknown. During large storms,the maximum rate of discharge is only slightly affectedby baseflow, and inaccuracies in separationmay not be important. The simplest baseflow separation technique is to draw a horizontalline from the point at which surfacerunoff begins,PointA in Fig. 11.7, to an intersectionwith the hydrograph recession wherethe baseflow rate is the sameasat the beginningof direct runoff as indicatedby Point B. A secondmethodprojects the initial recession curve downwardfromA to C, which lies directly below the peak rate of flow. Then point D on the hydrograph,representing N daysafter the peak, is connectedto point C by a straightline definingthe groundwater component.One estimateof N is basedon the formula3
N - Ao'2

(11.3)

where N : the time in days A = the drainageareain squaremiles A third procedureis to developa baseflow recession curve using Eq. 11.1 for data from the segmentFG, and then back-calculateall base flow to the left of Point fl

180

cHAPTER 11 HYDRocRApHS

9p F

Time

Flgure 11.7 Illustration of some hydrograph separation techniques. wherethe computedcurve beginsto deviatefrom the actualhydrograph, marking the end of direct runoff. The curve is projected backward arbitrarily to some Point E below the inflection point and its shapefrom C to E is arbitrarily assigned. A fourth widely usedmethod is to draw a line betweenA and F, and a fifth common method is to project the line AC alongthe slopeto the left of A , and then connectPointsC and

o d d

12N 6P

t21|'{{ 6A

12N 6P

tz]0|{ 6,{

I2N

6P

IzM

Figure 11.8 Illustration of base flow separation:hydrograph for the Uharie River near Trinity, North Carolina, February 25, 1939. (U.S. Departmentof Agriculture, Soil ConservationService).

TIME RELATIONSHIPS 181 11.5 HYDROGRAPH

is partly a of hydrographs B. All thesemethodsare approximatesincethe separation subjectiveprocedure. techniques to determinesurface Figure 11.8illustratestwo graphicalseparation the simpleprocedure . Line AD represents runoff and groundwaterflow components of connectingthe point of the beginningof direct runoff with the flrst point on the because overthe horizontal line technique groundwater recession curve (an advantage from the the time baseof direct runoff is much shorter).Ctrve ABCD is constructed curve. extensionof the baseflow recession

TIMERELATIONSHIPS 11.5 HYDROGRAPH

Wavetravel time is definedas the time required for direct runoff originating at the most remotepoint in the channelto reachthe outlet. The last drop of direct runoff to and reaches the outlet at the passthe outlet conceptuallytravelsoverthe watersurface the average velocity of equal to speedof a small surfacewave,rather than at a speed with channel velocity and varies flow. The wavetravel time is fasterthan the average (see 5/3 ratio is approximately shapeand other factors.For a rectangularchannel,the (Fig. is 1 1.4) of a hydrograph Section 13.1 for other wavevelocities).The time base directuntil the curve begins to be the time from which the concentration considered zero. An equationfor time basemay take the form runoff componentreaches T6:t"*t, where To : the time baseof the direct runoff hydrograph /" : the duration of runoff-producingrain rainfall releasetime t- : the excess Watershed lag time, illustrated in Fig. 11.4, is definedas the time from the penterof tt massof effectiverainfall to the centerof massof direct runoff. Other definitionsand are providedin S_eccharacteristics severalequationsrelating lag time to watershed chapters. tion 11.7 and subsequent theory, the excess-rainfallrelease Because ofits importancein unit hydrograph time is introduced.This is definedasthe time requiredfor the last, most remotedrop of to passthe outlet, signallingthe cessation ofexcessrain that fell on the watershed direct runoff. It is easilydeterminedas the time interval betweenthe end of rain and as direct runoff the end of direct runoff. Only that part of the outflow which classifles outflow nor(excess rain) is consideredin dqterminingthe releasetime. Watershed of direct runoff, in the form of interflow andbaseflow. after cessation mally continues Release time is very similar by definitionto wavetraveltime and time of concentration (Section 11.6). A foundational assumptionof unit hydrographtheoryl2 is that the watershed of the storm duration, and is related to releasetime is a constant,regardless excess release time is also The excess characteristics. basinfactorsratherthan meteorological unit hydrograph(IUH). conceptuallyidenticalwith the time baseof an instantaneous rain applieduniformly over the This is the runoff hydrographfrom 1.0 in. of excess in an instant of time (seeChapter 12). Both wavetravel time and excesswatershed with time of concentration. rainfall releasetime are often used synonymously

(rr.4)

182

CHAPTER11

HYDROGRAPHS

11.6TIMEOF CONCENTRATION
The most common definition of time of concentrationoriginatesfrom consideration the outlet of overlandflow. If a uniform rain is appliedto atract, the portions nearest contribute runoff at the outlet almost immediately.As rain continues,the depth of growsand discharge ratesincrease throughout.Runoff contribuexcess on the surface to contintions from variouspoints upstreamarrive at later times, addingthemselves uing runoff from nearerpoints, until flow eventually arrives from all points on the watershed,"concentrating" at the outlet. Thus, concentrationtime is the time required,with uniform rain, for 100percentof a tract of land to contributeto the direct runoff at the outlet.e As a secondpopular definition, the concentrationtime is often equatedwith the time for either the excess-rainfall releasetime or the wave travel time because is assumed runoff to arrive at the outlet from the most remotepoint after rain ceases to be indicative of the time required for 100 percent contribution from all points during any uniform storm having sufficient duration. The latter definition is often preferred becausefew storm durations exceedthe time of concentration,making rain recession. determinationof /. possibleonly by examiningexcess Because time of concentrationis conceptuallythe time requiredfor 100percent of the watershedto contribute, it is also often defined as the time from the end of excessrainfall to the inflection point on the hydrographrecessionlimb (e.g., see at the point Fig. 12.2).The reasoning usedin this definitionis that direct runoff ceases of inflection. For a small tract of land experiencing uniform rain, the entire areacontributes an equilibrium.This givesrise at approximately the sametime that the runoff reaches the direct to yet anotherdefinition of time of concentration.If rain abruptly ceased, releasetime t,. On the basis runoff would continueonly as long as the excess-rainfall release time and time of concentrationcan be considof the seconddeflnition.excess ered equivalent. Numerous equationsrelating time of concentrationto watershedparameters havebeen developed. severalpopular versions.Other variaTable 11.1 summarizes in Chapters12, 15,16, and25. tions are presented

11.7 BASIN LAG TIME The relative timing of rainfall and runoff must be known if drainageareashaving subbasins are to be modeledor if continuoussimulation is desired.A basic measure positionrelativeto the causative lag, which locatesthe hydrograph's of timing is basi'n stormpattern.It is most often definedas the differencein time betweenthe centerof massof effectiverainfall and the centerof massof runoff produced.Other definitions are also used.Two of theseare ( 1) the time interval from the maximum rainfall rate to the peakrate of runoff and(2) the time from the centerof massof effectiverainfall by the ratio of a flow length to a to the peak rate of flow. Time lag is characterized mean velocity of flow and is thus a property that is influencedby the shapeof the and geometry,and drainagearea,the slopeof the main channel,channelroughness thg'storm pattern.

11.7 BASIN LAGTIME TABLE11.1 SUMMARY OF TIMEOF CONCENTRATION FORMULAS

Method and date Kirpich (1940)
tc : L : S :

183

Formula for t" (min)

0.00782077S-o38s length of channel/ditch from headwater to outlet, ft average watershed slope, ftlft

Remarks
Developedfrom SCS data for sevenrural basins in Tennessee with well-definedchanneland steepslopes(37o to 1O7o); for overlandflow on concreteor asphaltsurfaces multiply t;by 0.4; for concretechannelsmultiply by 0.2; no adjustments for overland flow on bare soil or flow in roadsideditches. Essentiallythe Kirpich formula; developed from small mountainousbasinsin California (U.S. Bureauof Reclamation, 1973,pp. 67-7I).t4

USBR Design of Small Dams

t" : L : Il :

(r973)

60(lI.9L31H)o38s length of longest watercourse, ml elevation difference between divide and outlet, ft 41.025(0.0007, t c)Lozz

lzzatd (7946)ts
-c

S0.333i0.66?

I : c: Z : ,S: FederalAviation Administration ( 1970)r6

rainfall intensity,in/h retardancecoefficient length of flow path, ft slopeof flow path, ftlft

Developedin laboratory experiments by Bureau of Public Roadsfor overland flow on roadwayand turf surfaces; valuesof the retardance coefficientrange from 0.0070 for very smooth pavementto 0.012 for concretepavementto 0.06 for denseturf; solution requiresiteration; product i times Z shouldbe = 500. Developedfrom air field drainagedata assembled by the Corps of Engineers;method is intended for use on airfield drainageproblems,but has been used frequently for overlandflow in urban basins. Overland flow equation developedfrom kinematic wave analysis of surface runoff from developed method requiresiteration sinceboth i surfaces; (rainfall intensity) arrdt, are unknown; superposition of intensity-duration-frequency curve gives direct graphical solution for t". Equation developed by SCS from agricultural watersheddata; it has been adaptedto small urban basinsunder 2000 acres;found generally good where areais completelypaved;for mixod areas it tendsto overestimate; adjustmentfactorsare applied to correct for channelimprovementand imperviousarea; the equationassumes that t" : 1.67 X basinlag. Overland flow charts in Ref. 20 provide average velocity as function of watercourse slope and surfacecover.

- C)Losofso333 r" = 1.8(1.1 C : rational method runoff coefficient l, : length of overland flow, ft S : surface slope, Va O.94Lo6no6
(io 45o 3)

Kinematic WaveFormulas Morgali and Linsley (1965)'? Aron and Erborge (1973)18 SCS Lag Equation (1972)te

l, n I S

: : : :

. ._- @ ,

length of overland flow, ft Manning roughnesscoefficient rainfall intensity in/h averageoverland slope ftlft t.67 Lo8[(tooo/cN)- 9]oi

L : hydraulic length of watershed (Iongestflow path), ft CN : SCS runoffcurve number S : average watershedslope,Vo

SCS AverageVelocity Charts(1975, 1986),0

',' : l v L 60- v Z : length of flow path, ft V : ureragevelocity in feet per secondfrom Fig. 3- 1 of TR 55 for various surfaces

Sarrce.' After Ref. 13.

184

CHAPTER11

HYDROGRAPHS

Various studieshave been conductedfor the purpose of developingrelations descriptiveof time lag. Most prominent of thesewas the work by Snyderon large natural watersheds.2l His original equation has been widely used and modified in variouswaysby other investigators. Eagleson has proposedan equationfor lag time on sewered drainageareashavinga minimum sizeof 147 acres.2z An early investigation (1936) on small drainageareas(2-4 acres)was conducted by Horner in his classical work on urbandrainage in St. Louis,Missouri.23 Horner'swork wasinconclusive in that it did not yield a deflned procedure,but he did conclude that the comparativelywide rangein the lag time at eachlocation led to the inferencethat the lag was a variable,its valuebeingdetermined more by rainfall characteristics than by characteristics of the drainagearea. Snyder'sstudybasedon data from the AppalachianMountain regionproduced the following equationfor lag time:z1
t,1: c,(L""L)o3

(1 1.s)

where

/1 : the lag time (hr) betweenthe centerof massof the rainfall excess for a specifiedtype of storm and the peak rate of flow I'"o : the distancealongthe main streamfrom the baseto a point nearest the centerof gravity of the basin (mi) I : length of the main stream channel (mi) from the base outlet to the upstreamend of the streamand including the additional distanceto the watershed divide C, : & coefficientrepresenting variationsof types and locationsof str'eams

For the areastudied,the constantC, was found to vary from I.8 to 2.2, with somewhatlower valuesfor basinswith steeperslopes.The constantis considered to includethe effectsof slopeand storage. The value of 4 is assumed to be constantfor a given drainagearea,but allowanceis madefor the useof different valuesof lag for different types of storms. The relation is consideredapplicable to drainage areas rangingin sizefrom 10 to 10,000mi'. In a studyof sewered areasrangingin size from 0.22 to 7.51 mi2,Eagleson22 developed the equation
t ', " : -

L
(l.5 /n)R2/3 st/2

(11.6)

where

tr : L :

lag time, the center of mass of rainfall excess to the peak discharge (sec) the mean travel distance (ft), which is equal to the length of that portion

of the sewerwhich flows full n : the weightedManning's coefficientfor the main sewer : the weightedhydraulic radius of the main sewerflowing full { S : the weightedphysicalslopeof the main sewer Eagleson'sequationdirectly includesthe effects of channelgeometry and slope,as well as basin shape,and thus represents a refinementof the Snyderapproach. It also indirectly includesthe important effect of the storm pattern.

PROBLEMS185 of applicationof the following modified Linsley and Ackerman give examples equation.24 form of Snyder's

t,:C,+!
t t =K +

(rr.1)

where s is a weighted slope of the channel and the other variablesare as defined previously. time lag by equationsof the form haverepresented Other investigators ( 11 . 8 ) Vs

Numerous other derivations of relations for watershedlag times can be found in standardhydrologictexts and periodical literature. Others are includedwith someof in Chapter 12. the syntheticunit hydrographdiscussions

r Summary
is important to many designand water the structureof hydrographS Understanding the portion of the hydrologiccycle represents hydrograph supply applications.The for setting ratesof flow in streams to determine need in order mostoften that engineers establishing protection measures, and flood designing bridge lengthsand elevations, areal extent of flooding. Similarly, the volume of drainageinto a reservoiror past a water supply diversionis determinedfrom the areaunder the hydrograph.Accurate pipelines,and of thesevolumesare important to designof dams,reservoirs, estimates ' numerousother structures. including the time components, of hydrograph After graspingthe fundamentals preparedfor the be well should reader presented the in chapter, this relationships presented throughapplications and theory quantitativedevelopments hydrograph of Five. in Part 12 16 and out Chapters through

PROBLEMS
to separate and usetwo different techniques 11.1. Referto Fig. 11.1.Replotthis hydrograph the baseflow. for a major 11.2. Obtain streamflowdata for a water courseof interest.Plot the hydrograph flow. the base event and separate runoff 11.3. For the event of Problem I1.2, tabtlate the precipitation causingthe surfacerunoff the duration of runoff-producing.rain. Estimatethe time of concentraand determine Comparethis with I 1.4 to estimatethe time baseof the hydrograph. and use Eq. tion the time basecomputedfrom the hydrograph. rates at a cross sectionof a stream.The 11.4. Tabulatedbelow are total hourly discharge drainagearea abovethe sectionis 1.0 acre. a. Plot the hydrographon rectangular coordinate paper and label the rising limb and the recessionlimb. (concentrationcurve), the crest segment,

186

CHAPTERll HYDROGRAPHS plot of Q versus of the direct runoff usinga semilog b. Determinethe hour of cessation time. c. Use the base flow portion of your semilog plot to determine the groundwater constantK, recession d. Carefully constructand label baseflow separationcurveson the graph of Part a, using two different methods.

Time(hr) 0 1 2 3
I

Q (cfs)

Time (hr)

@ (cfs)

5 6 7

102 100 98 220 512 630 460 330

8 9 l0 11 t2
IJ

t4 15

2lo 150 105 75 60 54 48.5 43.5

11.5. On a neatsketchof a typical total runoff hydrograph,showor dimensionthe (a) storm

(b) beginningofdirectrunoff, (c) cessationtimeofdirectrunoff' (d) base hyetograph, fllw separationassumingthat additional contributions to base flow are negligible of the hydrograph' during ihe period of rise, and (e) crest segment of the watershed obtain measures your instructor, by assigned 11.6. For an urban watershed using the of concentration time the of estimates compare and slope, length, and area, in Table 11.1' Kirpich, USBR, FAA, and SCSLag equations

REFERENCES
1. American Society of Civil Engineers,Hydrology Handbook, Manuals of Engineering No. 28. New York: ASCE, 1957. Practice, "A 2. Donn G. DeCoursey, Runoff HydrographEquation," U.S. Departmentof Agricul.ture, Feb. 1966,pp.4I-116' Service, AgriculturalResearch 3. R. K' Linsley, M. A. Kohler, and J' L. H' Paulhus,Applied Hydrology' New York: McGraw-Hill, 1949. "Erosional Developmentof Streamsand Their DrainageBasins:HydroA R. E. Horton, physicalApproach to QuantitativeMorphology,"Bull. Geol' Soc'Am' 56(1945)' "An Approach Toward a PhysicalInterpretationof Infiltration Capacity," 5 . if.-n. gorton, Sci. Soc.Am. 5,399-417(1940). Soil Proc. "An Inflltration Equation with Physical Significance,"Soil Sci.77(1954). 6 . J. R. Philip, 7 . R. E: Horion";Surface Runffi Phenomena.Ann Arbor, MI: EdwardsBros., 1935. New York: Wiley' 1965' 8 . R. J. M. DeWiest,Geohydrology. ,.Hydrology,,' Sec. 4, U.s. Department of Agriculture, Soil Handbook, in Engineering 9. Service,1972. Conservation "Discussionof Analysis of Runoff characteristicsby o. H. Meyet," Trans. 1 0 . B. S. Barnes, ASCEl0s(1940). "Unit HydrographLag and PeakFlow Relatedto Basin 1 1 . A. B. Taylorand H. E. Schwartz, s. Union 33(1952). s," Trans,Am. Geophy Characteristic

REFERENCES 187 "Streamflow 12. L, K. Sherman, from Rainfall by the Unit-GraphMethod," Eng.News-Rec. 108(1932). 1 3 . D. F. Kilber, "Desk-top methods for urban stormwatercalculation," Ch. 4 in Urban Stormwater Hydrology, Water ResourcesMonograph No. 7, American Geophysical Union, Washington, D. C., 1982. 14. U.S. Bureauof Reclamation, Designof SmallDams,2nd ed., Washington,D.C., 1973. 1 5 . C.F.Izzafi,, "Hydraulicsof RunofffromDeveloped Surfaces," Proceedings,26th Annual Meetingof the HighwayResearch Boad,26, pp. 129-146, December1946. 16. FederalAviation Administration,"Circular on Airport Drainage,"ReportA/C 050-532058, Washington, D.C., 1970. n. J. R. Morgali, andR. K. Linsley,"ComputerAnalysisof OverlandFlow,"./. Hyd.Div., Am. Soc.Civ.Eng.,9l, no. HY3, May 1965. 1 8 . G. Aron, and C. E. Egborge,"A PracticalFeasibility Study of Flood PeakAbatementin Urban Areas,"U.S. Army Corpsof Engineers, Sacramento, Calif., March 1973. 1 9 . Soil ConservationService,"National EngineeringHandbook, Sec. 4, Hydrology," U.S. Dept. of Agriculture,U.S. GPO, Washington, D.C., 1972. 20. Soil ConservationService,"Urban Hydrology for Small Watersheds," TechnicalRelease 55, Washington, D.C., 1975(updated, 1986). 2 t . F. F. Snyder,"Synthetic Unit Graphs," Trans.Am. Geophys.Union 19, 447-454(1938). 22. Peter S. Eagleson,"CharacteristicsofUnit Hydrographsfor SeweredAreas," paperpresented beforethe ASCE, Los Angeles,-CA, 1959,unpublished. 23. W. W. Horner, and F. L. Flynt, "RelationBetweenRainfall and Runoff from Small Urban Areas," Trans.ASCE 62(101), 140-205(Oct 1956). "Methodof PredictingtheRunoff aA R. K. Linsley,Jr., andW. C. Ackerman, fromRainfall," Trans.ASCE 107(1942\.

C h a p t e r1 2

Unit Hydrographs

r Prologue
The purposeof this chapteris to: . Define unit hydrographsand show their utility in hydrologicstudiesand design. . Developfully the current methodsof obtaining, analyzing,and synthesizing unit hydrographs. . Presentmethodsfor converting unit hydrographsfor one storm duration to other storm durations. Waysto predict flood peak discharges from rainfall events and discharge hydrographs havebeenstudiedintensivelysincethe early 1930s.One approach receivingconsiderable use is called the unit lrydrographmethod. 12.1 UNIT HYDROGRAPH DEFINITION The concept of a unit hydrographwas first introduced by Shermant'z in 1932. He defineda unit graph as follows:2 produces given one-day drainage area, Ifa given rainfall a 1-in.depth ofrunoffoverthe thehydrograph showing at whichtherunoffoccurred canbe considered a unit therates graphfor that watershed. baseflow) Thus, a unit hydrographis the hydrographof direct runoff (excludes (the net rain total runoff after abstracfor any stormthat produces exactly 1.0inch of tions).Sucha stormwould not be expected but assumption is that to occur, Sherman's the ordinatesof a unit hydrographare t.O/P times the ordinatesof the direct runoff hydrographfor an equal-duration storm with P inchesof net rain. The term "unit" hasto do with the net rain amountof 1.0inch and doesnot mean to imply that the duration of rain that producedthe hydrographis one unit, whether an hour, day,or any other measure of time. The storm duration,X, that producedthe hasa differentunit hydrograph unithydrographmustbe specified because a watershed

DEFINITION 12.1 UNIT HYDROGRAPH

189

is defined as a direct for eachpossiblestorm duration. An X-hour unit lrydrograp,h runoff hydrograph havinga 1.0-in. volumeand resultingfrom anX-hour stormhaving would havea 1.0-in. volume a net rain rate of 1,/Xin.lhr. Az-hr unit hydrograph producedby a 2-hqstorm,and a 1-dayunit hydrograph would be producedby a storm having 1.0 in. of eicessrain uniformly producedduring a 24-hr period. The valueX is often a fraction. Figure 12.1illustratesa2-ltr,l2-hr, and24-hrwthydrograph for a given watershed.

t (b)

24hr
T (c)

)z nl

Figure 12.1 Illustration of 2-br, I2-hr, and 24-ht unit hydrographsfor the same watershed(Note: a : b : c : 1' X A).

190

CHAPTER 12 UNIT HYDROGRAPHS rainfall By Sherman'sassumption, applicationof an X-hourunit graphto design excess amountsother than 1 in. is accomplished simply by multiplying the rainfall excessamount by the unit graph ordinates,since the runoff ordinatesfor a given duration are assumedto be directly proportional to rainfall excess.A 3-hr storm producing2.0 in, of net rain would haverunoff rates2 times the valuesof the 3-hr unit hydrograph. One-half inch in 3 hr would produceflowshalf the magnitudeof the 3-hr in Section 12.3 and unit hydrograph. This principle of proportional flows is expanded appliesonly to equal duration storms. Implicit in deriving the unit hydrographis the assumptionthat rainfall is distributed in the sametemporal and spatialpattern for all storms.This is generallynot true; consequently, variationsin ordinatesfor different stormsof equal duration can be expected. first, then presentmethods This chapteris organizedto defineunit hydrographs of deriving unit hydrographs from actual rainfall and runoff records (Section 12.2). Section I2.3 preAfter familiarizing the readerwith the origin of unit hydrographs, for sentsmethodsof applyingunit hydrographs to generate direct runoff hydrographs any storm with durationsthat are multiple integersof the U.H. duration. The constructionof unit hydrographs for stormswith other than integermultiples of the derived duration is facilitated by a method known as the S-lrydrograph The procedure,as explainedin Section 12.4, developed by Morgan and Hulinghorst.3 employs a unit hydrographto form an S-hydrographresulting from a continuous appliedrainfall. The need to alter duration of a unit hydrographled to studiesof the unit rainfall. The concept of shortestpossiblestorm duration-the instantaneous instantaneous unit hydrograph(IIJH) is tracedto Clark6and can also be used(Secfor other than the derived duration. tion 12.5) is constructingunit hydrographs The previousdiscussion assumes that the analysthasrunoff and rainfall datafor deriving a unit hydrographfor the subject watershed.The application of unit hyand also drographtheory to ungauged receivedearly attentionby Snydera watersheds of the unit hydrographto waterby Taylor and Schwartz,5 who tried to relate aspects shed characteristics. As a result, a full set of synthetic unit-hydrographmethods emerged. A numberof theseare presented in Section12.6.

12.2 DERIVATION FROM OF UNITHYDROGRAPHS DATA STREAMFLOW

can Data collection preparatoryto deriving a unit hydrographfor a gaugedwatershed recordsof haveavailable many watersheds be extremelytime consuming. Fortunately, with office records of the streamflowand rainfall, and thesecan be supplemented Rainfall records pay be Water Resources Division of the U.S. GeologicalSurvey.T by the from ClimatologicalDatas publishedfor eachstatein the United States secured National Oceanicand AtmosphericAdministration (NOAA). Hourly rainfall records for for recordingrainfall stationsare publishedas a Summaryof Hourly Observations the location. Summariesare listed for approximately300 first-order situationsin the United States. it is desirable to acquireas rnanyrainfall records To developa unit hydrograph, aspossiblewithin the study areato ensurethat the amountand distributionof rainfall

12.2 DERIVATION,OF UNIT HYDROGRAPHS FROM STREAMFLOW DATA

191

over the watershedis accurate$ known. Preliminary selectionof storms to use in deriving a unit hydrographfor a watershedshouldbe restrictedto the following: 1. Stormsoccurring individually, that is, simple storm structure. 2. Storms having uniform distribution of rainfall throughout the period of rainfall excess. 3. Stormshavinguniform spatial distribution over the entire watershed. Theserestrictionsplace both upper and lower limits on size of the watershed to be employed.An upper limit of watershedsize of appro5imately1000 mi2 is overcautious, althoughgeneralstormsover suchareasare not unrealisticand somestudiesof areasup to 2000 mi2 have used the unit-hydrographtechnique.The lower limit of watershed extentdepends on numerousother factorsand cannotbe preciselydefined. A generalrule of thumb is to assume about 1000acres.Fortunately,other hydrologic techniques help resolveunit hydrographs for watersheds outsidethis range. The preliminary screeningof suitable storms for unit-hydrographformation shouldmeet more restrictive criteria before further analysis: 1. Duration of rainfall event should be approximately10-30 percent of the drainagearea lag time. 2. Direct runoff for the selected storm shouldrangefrom 0.5 to 1.75 in. 3. A suitablenumberof stormswith the sameduration shouldbe analyzedto obtain an average of the ordinates (approximately five events). Modificationsmay be madeto adjustdifferent unit hydrographs to a single duration by meansof S-hydrographs or IUH procedures. 4. Direct runoff ordinatesfor eachhydrographshouldbe reducedso that each eventrepresents1 in. of direct runoff. 5. The final unit hydrograph of a specificdurationfor the watershed is obtained by averaging ordinatesof selected eventsand adjustingthe result to obtain 1 in. of direct runoff. Constructingthe unit hydrographin this way producesthe integratedeffect of runoff resultingfrom a representative set of equal duration storms.Extremerainfall intensityis not reflectedin the determination. If intensestormsare needed, a study of recordsshouldbe madeto ascertain their influenceupon the discharge hydrograph by comparingpeaks obtainedutilizing the derived unit hydrographand actual hydrographsfrom intensestorms. Essentialstepsin developing a unit hydrographfor an isolatedstorm are: l. Analyzethe streamflowhydrographto permit separationof surfacerunoff from groundwaterflow, accomplished by the methodsdeveloped in Section l 1.4. 2. Measurethe total volume of surfacerunoff (direct runoff ) from the storm producingthe original hydrograph.This is the area under the hydrograph after groundwaterbaseflow has been removed. 3. Divide the ordinatesof the direct runoff hydrographby total direct runoff volume in inches,and plot theseresultsversustime as a unit graph for the basin. \-

192

12 UNIT HYDROGRAPHS GHAPTER 4. Finally, the effective duration of the runoff-producing rain for this unit (time history of rainfall intensity) graphmustbe found from the hyetograph of the storm eventused. Procedures other than thoselisted are requiredfor complex stormsor in developing synthetic unit graphs when data are limited. Unit hydrographscan also be An exampleillustransposed from one basinto anotherundercertain circumstances. tratesthe derivationof a unit hydrograph.

EXAMPLE I2.1 Using the total direct runoff hydrographgiven in Fig. I2.2, derive a unit hydrograph for the l7I5 ac drainage area. Solution 1. Separatethe base or groundwaterflow to get the total direct runoff hydrograph.A commonmethodis to draw a straightline AC that beginswhen

2-hr rainfall duration

I
I

*zTotal

precipitaion = 4.2 in.

o Time (hr)

1l

t2

Directrunoff. /T\ ordinate Yl

*zTotal directrunotf of l.4l5in.on1715ac 2-hr unit hydrograph of l.u rn. on I / I) ac

500
o FA

400 300 200 .100 0

Basef-tow / / \,

/^
3 4 5 6

t'-# -j
7 8 Time(hr) Dfuect runoffduration

Baseflow separatlon

Figure 12.2 Illustration of the derivation of a unit hydrograph from an isolatedstorm.

12,2 DERIVATION FROM STREAMFLOW OF UNIT HYDROGRAPHS DATA

193

the hydrograph rise and endswherethe recession curve startsan appreciable intersects the baseflow curve.The importantpoint hereis to be consistent in methodologyfrom storm to storm. 2. The depth of direct runoff over the watershedis calculatedusing > (DR x Ar) _ 2447 cfs-hr : 1 4" ''* area l7l5 ac

(r2.r)

runoff ordinateduring a chosen whereDR is the average heightof the dfuect time period Ar (in this caseA/ : 1.0 hr) . The valuesof DR determined from Fig. I2.2 are listedin Table 12.1. 3. Computeordinatesof the unit hydrographby using

8 "_ Q , 1 v,

(r2.2)

where Q, : the magnitudeof a hydrographordinate of direct runoff having a volume equal to % (in.) at someinstant of time after start of runoff Q, : the ordinateof the unit hydrographhavinga volume of 1 in. at someinstant of time In this examplethe valuesare obtainedby dividing the direct runoff ordinates of the unit-hydrograph by 1.415.Table12.1outlines the computation ordinates. 4. Determinethe duration of effectiverainfall (rainfall that actuallyproduces surfacerunoff). As statedpreviously,the unit hydrographstorm duration

TABLE 12,1 DETERMINATION HYDROGRAPH OFA 2.HR UNIT

FROM AN ISOLATEDSTORM (1) Time (hr) (2) Runoff (cfs) (3) Base flow (cfs) (4) Direct runoff, (2)-(3) (cfs) (5) 2-hr unit hydrograph (4) + 1.415 ordinate, (cfs) 0 8.5 84.8
i-l I

I 2 3 4 4.7 5 6 7 8 9 10 10.5 lt t2

110 t22 230 578 666 645 434 293 202 160 1t7 105 90 80

t10 t10 t10 110 110 110 110 110 110 110 1r0 105 90 80

0 t2 120 468 556 535 324 183 92 50 7 0 0 0

393 379 229 129 65.0 35.3 4.9 0 0 0

194

CHAPTER12

UNIT HYDROGRAPHS

should not exceedabout 25 percent of the drainage atea lag time' but violatesthis rule for the example.From Fig. 12.2, the rain duration is 2 hr. 5. Using the values from Table I2.1, plot the unit hydrograph shown in Fig. 12.2. r I

METHODS BY LAGGING APPLICATIONS 12.3 UNITHYDROGRAPH

hasbeenderivedfrom streamflowdata (or synthesized Once an X-hr unit hydrograph parameters, Section 12.6) it can be used to estimatethe direct runoff from basin hydrographshapeand duration for virtually any rain event.Applications of the X-hr usedfor stormshavingdurations UH to other stormsbeginswithlagging procedures, to stormswith fracApplications duration. that are integermultiples of the derived in are discussed procedures, IUH and tional multiples of X, known as S-hydrograph Sections12.4 and 12.5. are applicableto effective (net) rain, the processof Because unit hydrographs lossesfrom applyrngUH theory to a storm beginsby first abstractingthe watershed Any of the hyetograph' rain in an effective the precipitation hyetograph,resulting this discussion of remainder The be applied. proceduresdetailedin Chapter 4 can the storm. from losses watershed abstracted that the analysthas already assumes is'treated storm of X, the multiple integer is an If the duration of anotherstorm from eachX increment as a seriesof end-to-endX-hourstorms.First, the hydrographs are then added The ordinates hydrograph. unit X-hour from the ofrain are determined hydrograph. total the times to determine at corresponding EXAMPLE 12.2 rates for the 2-hr unit hydrographshownin Fig' I2.3 are'. Discharge

Time (hr) O (cfs)

0 0

1 100

2 250

3 200

4 100

5 50

6 0

Develophourly ordinatesof the total hydrographresultingfrom a 4-hr designstorm amounts: havingthe following excess

Hour (in.) Excess

1 Q.5

0.5

1.0

4 1.0

Solution. The 4-hr duration of the designstorm is an integermultiple of the canbe foundby addingthe duration.Thus,the total hydrograph unit hydrograph rain, asshowninFig. l2'3c. of end-to-end of two 2-hr increments contributions a unit has 1.0 in. of net rain and thus reproduces The first 2-hr stormsegment 2.0in. of netrain (in 2 hr);thus 2-hr stormsegmenthas The second hydrograph. its ordinatesare twice those of a 2-hr unit hydrograph.The total hydrograph,

12.3 F

BY LAGGINGMETHODS APPLICATIONS UNIT HYDROGRAPH

195

1.0

r>
h

u.) 0
2 3 4 . 5 6 (a) 2-hourunit storn excess

300 200

i90 zoo
1oo '6 0

1.0
ir

0.5 0 2 3 4 5 6 (c)Design stom excess 600 500

'6 400
:i

H 300 i5 200 100 0 r 2 3 4 5 6 7 (d) Contribution of each 2-hourstorm 600 500

a

t +oo ; ff:oo '$ zoo

-100 0
0 1 2 3 4 5 6 1 8 (e) Total design hydrograph

Figure 12.3 Example 12.2 deivation of total runoff hydrographusing a 2-hr unit hydrograph.

196

cHAPTER i2

uNtr HyDRocRApHS

Fig.12.3e, is found by summingthe fwo contributionsat corresponding times. Note in Fig, 12.3d that runoff from the secondstorm beginswhen the second rain begins,not at the beginningof the first storm. r I This methodof "lagging" is basedon the assumption that linear response of the watershed is not influencedby previousstorms-that is, one can superimpose hydrographs offset in time and the flows will be directly additive. The simplestway to developcompositedirect runoff hydrographs for multiple-hourstormsis in a spreadsheet.Care must be taken, however,in visually confirming, as in ExampleI2.2, that the start and end points of runoff from eachcontributingX-hr incrementof rain are properly selected. A commonerror is to lag eachadditional contributinghydrograph by Ar, the time interval betweenreadings, rather than X, the associated duration with the given unit hydrograph. Also, the multiplier for the UH ordinatesmust be the net rain occurring in X hours, not the rain occurring in the time increment A/. Example I2.3 illustrates thesepoints. EXAMPLE 12.3 Using the derived 2-hr lunit hydrograph in Table 12.1, determinethe direct runoff hydrographfor a 4-hr. storm havingthe following excess rain amounts:

Hour Excessrain. in.

2 0.7 t.2 1.2

o.7

Solution 1. Tabulatethe unit hydrograph at intervalsof the selected time interval, A/, as shown inTable 12.2.
TABLE 12.2 UNIT HYDROGRAPH APPLICATION OF EXAMPLE12.3 Effective rainfall (in.)

Time (h0 0
I

Unit hydrograph (cfs) 0 8.5 84.8

JJI

Contrib. of first 2-hrrain U HX 1 . 4 ' 0 I1.9 l19 463 531 321 181 91 49.4 6.9 0

Contrib. of second 2-hr rain

uH x 2.4'

Total outflow hydrograph (cfs)

2
J

4 5 6 7 8 9 10 1l t2

0.7 0.7 1 . 2" 1.2

379 229 129 65 35.3 4.9 0

0 20.4 203 794 910 550 310 156 84.7 11.8 0

0 11.9 119 483 734 11 1 5 1091 641 359 163 84.7 I 1.8 0

12.gUN|THYDRoGRAPHAPPL|CAT|oNSBYLAGG|NGMETHoDS197

X is 2. Determine the correct UH multiplier for eachX-hr interval. Because net a total produce storm 2 hrs for this example,the first two hours of the produce storm the of rain of 1.4 inches. Similar$, the last two hours 2.4 inchesof net rain. 3. Determinethe correct start and end times for eachof the two hydrographs and tabulatethe contributionof the l.4-inch and 2.8-inch rains at the X-hr storm startedat / : 3 hrs, the second appropriatelag times. Because : 3 hrs as shown inTable I2'2. runoff for this-stormcannotbegin until / the total runoff hydrographs 4. Add the contributionsat eachtime to determine for the 4-hr storm. and sum 5. Checkthe tabular solutionby plotting eachof the two hydrographs rr at each/, as showninFig.I2'4' the ordinates In addition to using a given X-hr UH for determiningthe runoff hydrographfor a given storm, tagging of ttt" X-hr UH can be used to developother duration unit is the sameas applyingthe X-hr UH to 1.0 in. of net rain hyirographs.The pioCedure in f nouri. As earlier, Y must be an integermultiple of X. For example,if a 1-hr unit resultingfrom a 2-hr a unit hydrograph for a given watershed, is available hydrograph siorm-is tbtained by plotting two L-hr unit hydrographs,with the secondunit hyin drographlagged t hr, adding ordinates,and dividing by 2' This is demonstrated Thus hydrograph' unit nigl ti.S, riliere the dashedline rept"sentsthe resulting2-ht the t in. of rainfall containedin the original 1-hr duration has been distributedover a Z-hr period.

Time units

hydrographfor'Example 12'3 Figure 12.4 Synthesized derivedby the unit hydrographmethod'

\_

198

CHAPTER 12 UNITHYDRoGRAPHS

l-hr unit hydrogaph 2-hIunithy&ogaph

Time

Figure12.5 Unithydrographlaggingprocedurero develop another unit hydrograph. Modificationsof the original unit-hydrograph duration can be madeso that two 1-hr unit hydrographs are usedto form a 2-hr unit hydrograph;two 2-hr unit hydrographsresult in a 4-fu diagram,and so on. Care must be taken not to mix durations in the lagging procedure, since errors are introduced; a l-hr and a}-hr unit hydrograph do not representa 3-hr unit hydrograph.Lagging procedureis therefoie restrictedto multiples of the original duration accordingto the expression D1 : nD (I2.3\ where Dl : the possibledurationsof the unit hydrographby lagging methods D : the original duration of any given unit hydrograph fl: I,2,3,..

12.4 S.HYDROGRAPH METHOD

The S-hydrograph methodovercomes restrictionsimposedby the laggingmethodand allowsconstructionof any durationunit hydrograph. By observingthe lagging system just described, it is apparentthat for a l-hr unit hydrograph,the l-in. rainfall excess hasan intensityof 1 in./hr, whereas the 2-hr unit hydrograph is producedby a rainfall intensity of 0.5 in./hr. Continuouslagging of either one of theseunit hydrographs is comparable to a continuouslyappliedrainfall at either 0.5 in./hr or 1 in./hr intensity, dependingon wfuch unit hydrographis chosen. As an example, usingthe 1-hr unit hydrograph, continouslaggingrepresents the direct runoff from a constantrainfall of 1 in./hr as shownin Fig. r2.6a. The cumulative addition of the initial unit hydrograph ordinatesat time intervalsequalto the unit storm duration resultsin an S-hydrograph (seeFig. r2.7). &aphically, construction of an s-hydrographis readily accomplished with a pair of dividers. The maximum

12.4 S-HYDROGRAPHMETHOD

199

*l

Dhr

D-hr S-hydrograph laggedt hr

Time
(a)

*l

l-

Figure 12.6 S-hydrograph method.

discharge of the S-hydrograph occursat a time equalto D hourslessthan the time base of the initial unit hydrographas shown inFig. 12.6a. To constructa pictorial 2-hr unit hydrograph,simply lag the first S-hydrograph by a secondS-hydrograph a time interval equalto the desiredduration.The difference in S-hydrograph ordinatesmust then be dividedby 2. Any duration r unit hydrograph may be obtained in the samemanner once another duration D unit hydrqgraphis known. Simply form a D-hr S-hydrograph; lag this S-hydrograph f hr, andmultiply the difference in S-hydrographordinatesby D/t. Accuracy of the graphical procbdure dependson the scaleschosento plot the hydrographs.Tabular solution of the Shydrograph methodis also employed, but hydrograph tabulationsmustbe at intervals of the original unit.hydrographduration. EXAMPLE 12.4 Given the following 2-hr unit hydrograph,use S-hydrograph procedures to construct a 3-hr unit hydrograph.
Time (hr) 0 (cfs)

o 0

| 100

2 250

3 200

4 100

5 50

6 0

Solution. The 2-hr unit hydrograph is the runoff from a 2-hr stormof 0.5 in./hr. The S-hydrograph is formedfrom a net rain rate of 0.5 in./hr lasting indefinitely as shown in Fig. 12.6a. the 2-hr Its ordinates arefoundby adding (UH) runoff ratesfrom eachcontributing unit-hydrograph 2-hr block of rain:

200

CHAPTER12

UNIT HYDROGRAPHS

/,/

90

-'{.3
u0 Figure 12.7 S-hydrograph. Time (min.)

Time (h0 0 I
z

1st2-hr 0 100 250 200 100

5n

2nd Z-hr

3rd2-hr

S-hydrograph 0 100 250 300 350 350 350 350 350

3
4

5 6 7 8

0 100 250 200 100 50 0

' 0 100 250 200 100

0 100 250

asshown To find a 3.-hrhydrograph, the S-curveis laggedby 3 hr and subtracted in Fig. 12.6b.This results in a hydrographfrom a 3-hr storm of 0.5 in./hr, or 1.5 in. total. Thus the ordinatesneed to be divided by 1.5 to producethe 3-hr unit hydrograph:

12.5 THE INSTANTANEOUS UNIT HYDROGRAPH

201

Time (hr)
0 1 2 3 4 5 6 7

S-hydrograph

Lagged S-hydrograph

Difference

3-hrUnit hydrograph 0 67 167 200 167 . 6 7 33 0

0 100 250 300 350 350 350 350

0 100 250 300 350

0 100 250 300 250 r00 '50 0

12.5 THE INSTANTANEOUS UNITHYDROGRAPH

The unit-hydrograph methodof estimatinga runoff hydrograph canbe usedfor storms of extremelyshortduration.For example, if the durationof a stormis 1 min and a unit volume of surfacerunoff occurs, the resulting hydrograph is the 1-min unit hydrograph.The hydrographofrunofffor any 1-min storm ofconstant intensitycan be computedfrom the l-min unit hydrographby multiplying the ordinatesof the 1-min unit hydrographby the appropriaterain depth.A storm lasting for many minutescan be described as a sequence of 1-min storrns(Fig. 12.8).The runoffhydrographfrom each l-min storm in this sequence can be obtainedas in the precedingexample.By superimposing the runoff hydrograph from each of the l-r.nin storms, the runoff hydrographfor the completestorm can be obtained. From the unit hydrographfor any duration ofuniform rain, the unit hydrograph for any other durationcanbe obtained.As the durationbecomes shorter,the resulting unit hydrograph approaches The instant4neous unit an instantaneous unit hydrograph. hydrograph(IUH) is the hydrographof runoff thaf would result if 1 in. of waterwere spreaduniformly over an areain an instant and then allowed to run off.e To develop an IUH, any I in.lhr S-hydrographmust first be obtained. The resulting S-curve is laggedby the interval Ar to developa Ar-hour unit hydrograph. The resulting At-hour unit graph becomesan IUH when Ar is set to 0.0 in the limit. If a continuing 1 in./hr excessstorm produces the original and lagged Sof Fig. 12.6b,the Ar-hour unit hydrographis the differencebetweenthe hydrographs two curves,divided by the amount of excess rain depth in A/ hours,or
Qo- Q" Q,(Lt-hr UH) : ILt

(r2.4)

The Qo - Q" dtfferences are dividedby I Lt to convertfrom a stormwith 1Al inches in Al hours to one with 1.0 in. in At hours,which is the definition of a Ar-hour unit graph. As Ar approaches zpro, Eq. 12.4 becomes

: !as 0,(ruH) Idt

(12.s)

202

CHAPTER12 UNIT HYDROGRAPHS

Figure 12.8 Unit-hy&ographdescriptionof the runoff process. (a) Unit hydrograph; (b) a sequence of l-min (c) superposition storms; for each of runoffhydrographs of (After Schaake.e) the l-min storms. which showsthat the flow at time I is proportional to the slopeof the S-hydrograph at time r. In applications, the slopeis approximatedbyLQ/A,I, and the IUH ordinates can be estimatedfrom pairs of closelyspacedpoints of the S-hydrograph. Ifan IUH is supplied,the aboveprocesscan be reversed,and any X-hour unit graph can be found by averaging IUH florvsat X-hr intervals, or

UH) : 1(IUH,+ IUH,_X) Q,(X-tu

{r2.6)

Use of this approximate equation is allowed for small X values and permits direct calculationof a unit graph from an IUH, bypassing the normal S-hydrograph procedure.

UNITHYDROGRAPH 12.5 THEINSTANTANEOUS

203

EXAMPLE 12.5 determinethe IUH, and then use it to Given the following 1.0 in./hr S-hydrograph, a 1-hr UH. estimate
3.5 750 4.0 800

Time (hr) S-curve (cfs)

0 0

0.5 5 0

1.0 200

1.5 450

2,0 500

2.5 650

3.0 700

Solution. The IUH is foundfrom Eq. l2.5.The slopeat time r is approximated bY (Q,*o., Q,-o)lLt

IUH = AQlAf

0 0.5 1 1.5 2 2.5 3 3.5

+

0
JU

200 450 500 650 700 750 800 800 800

0 200 400 300 200 200 100 100 50 0 0

at 1-hr intervals: from F,q.12.6,usingreadings The 1-hr uH is obtained

1-hr UH 0 200 300 150 75 25 n "

luHf

luHr_1

0 1 2 3
A

5 6

0 400 200 100 50 0 0

0 0 400 200 100 50 0

The readershouldverify that the 1-hr UH obtainedthrough use of the IUH is t hr, subtractt[e sameas that obtainedby lagging the S-hydrograph approximately 1-hr UH. to a difference the ing, and converting rainfall The ordinatesof the IUH representthe relative effect of antecedent time IUH with plotting the By time. of instant at any rate the runoff intensitieson superimposing then and (see I2.9), Fig. right the than to rather the left increasingto this plot over the rainfall hyetograph(plotted with time increasingto the right as in pig.iZ.g),the'relative weight given to antecedent rainfall intensities(asa function of the runoff rate at any time is words, In other past) observed. is easily the time into

204

12 UNITHYDROGRAPHS CHAPTER

Time-reversed image of the instantaneous unit hydrograph

Time into the past Antecedent iainfall intensities

oo=[;f@xi(t-r)dr

unit hyFigure 12.9 Calculationof runoff rates with the instantaneous antecedent the of average is a weighted The runoff rateat arrytime drograph. unit hyimageof the instantaneous The time-reversed rainfall intensities. (After Schaake'v) furtction. the weighting represents drograph of the previousrainfall intensities.Therefore,the computedas a weighted average of.the rainfall pattern computedrunoff hydrographis the weighted,moving average image of the unit hydrograph'e and itre weighting irtt"iiott is the time-reversed the runoff rate at any time is given by the convolution Statedmathematically, integral

Q(A:

- r)dr ,<,tu, l,'

(r2.7)
ll:

12.6 SYNTHETICUNIT HYDROGRAPHS

205

where

Q(t) : the surfacerunoff rate at time t f(r) : the ordinate of the IUH at time r i(t - r): the rainfall inten'sity (after abstraction of the appropriate infiltration losses) at time t - r

The variable 7 represents time into the past so that time r - r occursbefore time r. The limits on the integral allow r to vary betweena pastand presenttime (i.e., r : 0, t - r : 0)' The integral givesa continuous weightingof prbviousrainfall intensities by the ordinatesof the IUH.

12.6 SYNTHETIC UNITHYDROGRAPHS

As previously noted, the linear characteristics exhibited by unit hydrographs rf6i a watershedare a distinct advantagein constructingmore complex rto.- Oir.fruig. hydrographs. Generally,however,basicstreamflowand rainfall data arcnot available to allow construction of a unit hydrographexcept for relatively few watersheds; therefore, techniqueshave evolved that allow generationof syithetic unit lrydrographs.

GammaDistribution
The shapesof hydrographsoften closely match a two-parametergamma function, given by

f(x) :

xoe-*/B

B"+lf(d + 1)

(12.8)

where0 ( r ( m. The parametera is a dimensionless shapefactor (mustbe greater tltul - 1), and B is a positive scalefactor havingthe sameunits as x and contiolling the baselength. The product of a and B givesthe value-r corresponding to the apexl or maximum value ofl(x). For a ) 1, the distribution has a single upe* und'plot, similar to hydrograph shapes,as shown in Fig. 12.10. The dislribution mein is is p2(a + I). F@ + 1), and variance Many of the synthetic unit hydrographproceduresresult in only three to five . points on the hydrograph,through which a smooth curve must be fitted. In addition to the requirementthat the curve passes through all the points, the area under the hydrographmust equal the runoff volume from one unit of rainfall excess over the watershed. This latter requirementis often left unchecked and can result in considerablg errors in performing calculationsthrough the use of ordinatesof a hydrograph that do not representa "unit" of runoff. The mostusefulfeatureof the gammadistributionfunction (explained in greater detail later) is thit it guarantees a unit areaunder the curve. It can convenientlybe used to synthesize an entire hydrographif the calculatedpeak flow rate eo and its associated time to are known. This usesa proceduredeveloped by Aron und'White.ro If time r is substituted for x in Eq. 12.g, the time to peak tois aB. At this point, the function/(r) equalsthe peak flow rate e,, or

o,:ffi=ffrr*1

(r2.e)

206

CHAPTER12

UNIT HYDROGRAPHS

1.0

B--1

0
x

Figure 12.10 Gamma function shapesfor various shape and scale parameter values. factor A. The conversion whereC, A is the unit volumeof runoff from a basinwith arca = dimensionless. to make @(a) 1.008 is selected C, The function f(a) is shownby Aron and white to be relatedto a by1l (12.10) + 0303 a : 0.045+ 0.5d + 5.6Q2 well in the range 1 reasonably Collins showsthat this can be approximated
byt'

( a ( 8

q.:05Q+5.902 Combiningthis with F,q. 12.9 *tu"t , t o:os ffi.tn(H)'

(r2.Lr) (r2.r2)

To fit a unit graph usingEqs. I2.9 and 12.12,the peak flow rate and time must allow this. Next. @(a)is subsequ-ntly Severalof the methodsdescribed be estimated. can now foundfrom Eq.12.9, and a from Eq.12.10 ot l2.Il. The unit hydrograph alo to. Substituting by calculatingQ at any convenientmultiple, a, of be constructed : atp as for x in Eq. 12.8 gives the flow at t
Qoto: QraoeQ-o)"

(r2.r3)

which can be solvedfor all the flow rates of the hydrograph. EXAMPLE 12.6 is 1720cfs and watershed of a 36,000-acre The peak flow rate for the unit hydrograph the rest of synthesize 12.8 to Eq. Use of runoff. initiation the o".oi, 12 hr following the hydrograph.

12.6 SYNTHETICUNIT HYDROGRAPHS

207

Solution. From Fq. 12.9,

r720(12) : 0.57 6@) : 1.008(36,000)

From Eq. 12.10(and t2.It), q.:2.2 The hydrographis then found from Eq. L2.I3: a) e2'2(rQ*, : 1720a2'2 Solving for a few points, we obtain the followine values:
10.0 0 0.5 1.0 2.0 5.0 10.0
t = afo(hr)

O(ctu)
0 tt25 1720 876 9 0

0 6 l2
)4

60 120

sufficient intermediatepoints shouldbe generated to define the entire shapeof the hydrograph. tl

Snyder's Method
one- techniqueemployedby the corps of Engineersl3 and many othersis basedon methods developedby Snyderaand expandedby Taylor and 3chwartz.sIt allows computation of lag time, time base,unit_hydrograph duration, peak discharge,, and hydrograph time widths at 50 and75 perceniof peukflor. ny uring ttteseseven points, a sketchof the unit hydrograph is obtained,rig.lz.rt,and cleckel to seeif it contains 1 in. of direct runoff.

Alternate recessions I to produce 1.0 in. ofrunoff

Time,r F'igure 12.11 Snyder'ssyntheticunit hydrograph.

208

CHAPTER12

UNIT HYDROGRAPHS

that the a unit hydrographassumes Time to Peak Snyder'smethodof synthesizing from Eq. 11.5.Its locationis lag, estimated peak flow rate occursat the watershed rate are both as shownon Fig. 12.11.The lag time and peak discharge established For the lag time, the characteristics. watershed correlatedwith variousphysiographic and C, is variablesL and L"o for Eq. 11.5 are estimatedfrom map measurements, TableL2.3summaor other sources. for the locale,usingSnyder'sestimates developed rizes a variety of C, valuesfor variousregions. that lag time is a constantfor a particular watershed-that'is, It is assumed uninfluencedby variations in rainfall intensitiesor similar factors. The use of L"o shape,andC, takescare of wide variationsin topography, for the watershed accounts from plains to mountainousregions. lower valuesof C,, with extremesof 0.4 nqtec Steeperslopestend to generate in SouthernCalifornia and 8.0 alongthe Gulf of Mexico and Rocky Mountains.W{ren valuesof C' will be betweenone influencepeak discharge, snowpackaccumulations sixth to one third of Snyder'svalues. Time Base The time base of a syntheticunit hydrograph(seeFig. 12.11) by Snyder'smethod is
t , : ?
T ;

, t l 6

(r2.r4)

where t6 : the basetime of the syntheticunit hydrograph(days) t1: the lag time (hr)

FOR U.S. LOCALITIES TABLE 12.3 TYPICALSNYDER'SCOEFF|CIENTS Location

Appalachian Highlands West Iowa Southern California Ohio Eastern Gulf of Mexico Central Texas North and Mid-Atlantic states Sewered urban areas Mountainous watersheds Foothills areas Valley areas Easlern Nebraska Corps of Engineers training course Great Plains Rocky Mountains SW desert NW coast and Cascades 21 urban basins Storm sewered areas "Channel slope S

Range of C, t.8-2.2 0.2-0.6 0.6-0.8 0.4-2.3 0.2-0.5

Average Cr 2.0 o.4 0.4 0.7 8.0 1.1

Range of Co 0.4-0.8 0.7-1.0 0.6-0.7 0.3-1.2 0.1-0.6

Average Co

0.6 0.8 0.9 0.6 0.6 0.8 0.3

0.6/\4" 0.3
1.2 o;7 0.4 0.8 0.3-0.9 1.3
1,4

0.4-1.0 0.4-8.0 0.8-2.0 1.5-8.8 0;t -1.9 2.0-4.4 0.3-0.9 0.2-0.3

0.5-1.0

0.8

3.1 0.6 0.2

12.6 SYNTHETIC UNIT HYDROGRAPHS 209 Equation L2.14 gives reasonableeslimatesfor large watersheds but will produce excessively large valuesfor smaller areas.A generalrule of thumb for small areasis to use three to five times the time to peak as a base value when sketchinga unit hydrograph.In any event, the time base shouldbe adjustedas shown in rig. tz.lt until the areaunder the unit hydrographis 1.0". Duration The duration of rainfall excessfor Snyder's synthetic unit-hydrograph development is a function of lag time
l -

'

tt

5.5

(r2.1s)

where /, : duration of the unit rainfall excess (hr) t1 : the lag time from the certtroidof unit rainfall excess to the peak of thb unit hydrograph This synthetictechnique alwaysresultsin an initial unit-hydrograph duration equalto in lag time occur with changes fi/5.5.However, sincechanges in duration of tlie unit hydrograph,the following equation was developed to allow lag time and peak dischargeadjustments for other unit-hydrograph durations. tm:h+0.25(t*-t) where t1p: tt : to: t, the adjustedlag time (hr) the original lag time (hr) the desiredunit-hydrograph duration (hr) the original unirhydrograph duration : t,/5.5 (hr)

(r2.16)

Peak Discharge If one assumes that a given duration rainfall produces1 in. of direct runoff, the outflow volume is some relatively constantpercentage of inflow volume.A simplified approximationof outflow volume is r, X er, andthe equation for peak discharge can be written
Vr: 640CPA Ltn

(12.r7)

where Qp : peak discharge(cfs) C" : the coefficientaccountingfor flood waveand storageconditions;it is a function of lag time, durationof runoff ptoducingrain, effectivearea contributing to peak flow, and drainagearea A : watershedsize (mi2) tp : the lag time (hr) Thuspeakdischargetanbe calculated givenlag time andcoefficientofpeak discharge C.. Valuesfor Cp range from 0.4 to 0.8 and gengrally indicate retention or storage capacityof the watershed. Larger valuesof C" aregenerallyassociated with smaller valuesof C,. Typical valuesare tabulated inTable 12.3. Hydrograph Construction From Eqs: 1I.5, lZ,I4, I2.I5, and,12.1,7 plot three points for the unit hydrographand sketcha syntheticunit hydrograph,remembering

210

CHAPTER12

UNIT HYDROGRAPI.IS

that total direct runoff amountsto 1 in. An analysisby the Corps of Engineers(see in plotting time widths for points on the hyFig. 12.12) gives additional assistance As a generalrule ofthumb, drographlocatedat 50 and 75 percentofpeak discharge.l3 eachsideof the pegk proportioned ttr" ti-l width at l/so and IV^ ordinatesshouldbe unit-hydrograph synthetic the left of in a ratio of I:2 with the short time sideon the gives valuesfor unrealistic Eq. I2.I4 peak. As noted earlier,for smallerwatersheds, to the time total multiplying by ihe bur" time. If this occurs,a value can be estimated and amount on the based modified peak by a value of from 3 to 5. This ratio can be boundaries. watershed ii." tut" of depletionof storagewater within the The envelopecurvesin Fig. I2.I2 ate definedby Wso: 8301(Qo/A)" (Qo/A)" W15: 470f

(12.18)

(t2"]a)

The sevenpoints formed through the use of theseequationscan be plotted and a smooth curve drawn. To assurea unit hydrograph,the curve shape and ordinates shouldbe adjusteduntil the areabeneaththe curve is equivalentto one unit of direct runoff depth over the watershedarea.This can be doneby hand-fittingand planimetering or by curve-fitting. to fit a Pearson techniques regression Hudlow and Clarkla used least-squares the seven through (refer 26) to Chapter function (gamma) density probability III type similar shape and a of 1.0 area has an function points. This uoit-nyAtograph Snyaer
1000 800 400
N

,2

9 200
oo

1oo -? 8 0 E 6 0 9" 40

) \

4-l

'.,

s
*

t *tS
08,

F\

\, \ \ $ \
40 60 80100

10 8 6
0.2

I 0.4 0.60.8

810

Width of unit hYdrograPh (hr)

Figure L2.12 Unit hydrographwidth at 50 and 75 percent of peak flow' o, observedvalue of lfi5. a, observedvalue of Wso.

c uNrr HYDR.GRAPHS 211" to thator natural hydrographs. rh" ,h;:l

Q,:

,lll::i
e-G-tp')/b

nr(;

(r2.20)

where a and b are shapeand scaleparameters.Hudlow and Clark presenta trial-anderror solution to the least-squares normal equations,using Newton's method, to developestirnatesof a and b. The application of Snyder's syntheticunit-hydrographmethod to areasother than the original study area should be precededby a reevaluationof coefficientsC, and C, in Eqs; 11.5 and I2.I7 . TIns analysiscan be accomplished by the use of unit hydrographs in the region under study which havethe proper lag time-rainfall duration ratio; that is, t, : ttf 5.5. If anotherrainfall durationis selected, variations of C and C, can be expected.

SCS Method
A methoddeveloped by the Soil Conservation Servicefor constructingsyntheticunit hydrographs is basedon a dimensionless hydrograph(Fig. 12.13). This dimensionless graph is the result of an analysis of a large number of natural unit hydrographs from a wide range in size and geographii locations.The method requires only the

0.6

cSlo
b 0.5 0.4 0.3 0.2 0.1

als

t tp

Figure 12.13 Dimensionless unit hydrographand masscurve. (After Mockus.1s)

212

CHAPTER 12 UNIT HYDRoGRAPHS determination of the time to peak and the peak discharge as follovi,s:

tp:

D t+

tt

(r2.2r)

(hr) where to : the time from the beginningof rainfall to peak discharge D : the duration ofrainfall (hr) (hr) t1 : the lag time from the centroid of rainfall to peak discharge The ratios corresponding to Fig. 12.13 are listed in Table 12.4. The peak flow for the hydrographis developed by approximatingthe unit hydrographas a triangular shapewith basetime of ! t, and unit area.The readershouldverify that this produces
U r :

4844
t

(r2.22)

(cfs) where Qp: peak discharge A : drainagearea (mi2) tp : the time to peak (hr) The time base of ! ro is based on empirical values for averagerural experimental peak flow) for steepconditions watersheds and shouldbe reduced(causing increased or increased (causing decreasedpeak flow) for flat conditions. The resulting coefficientinBq.12.22 rangesfrom nearly 600 for steepmountainousconditionsto 300 for flat swampyconditions. A relation of /, to size of watershedcan be used to estimatelag time. Typical relations from two geographic regionsare
tt : t,: TABLE 12.4 COORDINATES OF SCS DIMENSIONLESS UNIT HYDROGRAPH OF FIGURE 12.13
Q/Q, Q/Qp

l.44A0'6 Texas 0.54A0'6 Ohio

(r2.23a) (r2.23b)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 t.2 1.3

0 0.015 0.075 0.16 0.28 0.43 0.60 0.77 0.89 o.9'7 1.00 0.98 0.92 0.84

1.4 1.5 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.5 4.0 5.0

0.75 0.66 0.56 0.42 0.32 0.24 0.18 0.13 0.098 0.075 0.036 0.018 0.009 0.004

HYDROGRAPHS 12.6 SYNTHETIC UNIT 213. The average lag is 0.6/",where/" is the time of concentration, definedby SCS aseither (calledthe the time for runoff to travel from the furthermostpoint in the watershed rain to the inflection of the unit upland method) or the time from the end of excess hydrograph.For the first case,
t":L1tp-D

(r2.24)

The dimensionless unit hydrograph,Fig. 1.2.1.3,has a point of inflection at approximately 1..7t,.If the lag time of 0.6t" is assumed, Eqs. 12.2I and 12.24 give D - 0.2t0 D = 0.I33t"

0).zs)
{r2,26)

A small variation in D is permissible, but it shouldnot exceed0.25t, or 0.I7t". pnce for other durdtions the 0.133r"-hourunit hydrographis developed, unit hydrographs can be developed using S-hydrograph or IUH procedures. By finding a value of t,, a syntheticunit hydrographof chosenduration D is obtained from Fig. 12.13. Atother equationusedby the SCS is . t - : "' where fi : I : Y : S: /0.'(s * 1;o.z

lgooyo's

(r2.27)

the lag time (hr) length to divide in feet average watershedslopein percent the potentialmaximumretention(in.) : (1000/CN) - 10, where.CN is a curve number described in Chapter4

or The lag from Eq. 12.27is adjustedfor imperviousness or improved watercourses, both, if the watershedis in an urban area.The multiple to be applied to thq lag is M : 1.- P(-6.8 X 10-3 + 3.4 x 10-4CN- 4.3 x 10-?CN, -2.2 x 10-8CN3) (12.28) where CN is the curve number for urbanized conditions, and P can be either the percentage that is hydraulically imperviousor the percentage of the main watercourse improved from natural conditions.If part of the area is imperviousand portions of the channelare improved,two valuesof M are determined,and both are multiplied by the lag. EXAMPLE 12.7

.\ For a drainagearea of 70 r4i2 having a lag time of 8 | hr, derive a unit hydrograph of unit hydrograph. duration 2 hr. Use the SCS dimensionless Solution 1. Using Eq.l2.21we obtain

t,:?*8*=9*t'tr

214

CHAPTER12

UNIT HYDROGRAPHS

2. From Eq.12.22 _484x70 n YP

9.5

att :9|hr Qo:3J60 cfsoccurring 3. Using Fig.12.13, we find the following: a: The peak flow occursar tfto: 1 or at t : 9+hr. b. The time baseof the hydrograph: 5toor 47.5 hr. c. The hydrographordinatesare: cfs. l . A t t / t o : 0 . 5 ,Q / Q r : 0 . 4 3 ; t h u sa t t : 4 . 7 5 h r ,Q : 1 5 3 1 2. At tlto : 2, Q/Q" - 0.32; thus at t : 19 ht, Q : 1139cfs. 3. At tft, : 3, Q/Q, : 0.07; thus at t : 28.5 hr, Q : 249 cfs. 4. Check D/to:0.21; OK. rl

Gray'sMethod
Another method of generatingsynthetic unit hydrographshas been developedby An approximateupper limit of watershed size for applicationof this method Gray.16 to the geographic areasof central Iowa, Missouri, Illinois, and Wisconsinis 94 mi2. The method is basedon dimensionalizingthe incomplete gamma distribution and graph of the form resultsin a dimensionless
O,t" : 25.0(y')n

r(q)

t 1o_7,t/pR\( \s-1

\Pol

(r2.2e)

where

Q,lPo: q and y : f : e: Pp : I :

percentflow in 0.25 PRat any given r/P^ value respectively shapeand scaleparameters, the gammafunction of q, equalto (4 - 1)!* the baseof natural logarithms the period of rise (min) time (min)

The relationfor 7' is definedzs yt : yPpandq : l.+ y'. (Fig. 12.14)allowscomputation This form of the dimensionless unit hydrograph ofthe discharge ordinatesfor the unit hydrographat times equal to I intervals of the period of rise P*, that is, the time from the beginningof rainfall to the time of peak discharge of the unit hydrograph. can be develof the watershed characteristics Correlationswith physiographic opedto get the valuesofboth Poand y'. parameAs 4qexample, the storage factorP^fy' hasbeenlinked with watershed ters Lf\/S", where L is the length of the main channel of the watershedin miles measured from the outlet to the uppermostpart of the watershed(Fig. 12.15); S. is definedas an average slopein percentobtainedby plotting the main channelprofile rrf4isnotaninreserf(q) : f(N + Z) : (N - I + z)(N- 2 + z)... (t + Z)/f(r + Z),
where N equalsthe integer.""-rl q, the function is approximated by

: qe.-al';l. r(q) & h. ;uF-

- #fu.

12.6 SYNTHETICUNIT HYDROGRAPHS

215

Muckey Creek near Mapleton, lowa

F< a-

K 1 6
F re o ' E o
6

1.0

l(Atloi -=:, penoo or nse (mrn.)

1.5 time (min.)

2.0
t PR

Figure 12.14 Dimensionlessgraph and fitted two-parameter gamma distribution for Watershed 5. (After Gray.r6)

='(ft)'
957o confidence belts

a-l> ": lH l0
e ttr x t9!
o l H

tlr

for Ppll'

Ttta
b

IL, IA,MO,WI OH NE andW. IA

(After Gray.16)

9.27 0.562 11.4

A A

al

0.531 0.498

(r = 0.92)

tgilrgljggtq. Rudo,
lchannel

-L
VS"

slope(Vo)

r.ir '

Figure 12.15 Relationof storage parameter, factor,Ppfy', and watershed (After LV 5",for watersheds in Nebraska, Iowa, Missouri, Illinois, and Wisconsin. Gray.r6) and drawing a straight line through the outlet elevationsuch that the positive and negativeareasbetweenthe streamprofile and the straightline are equal.The storage factorPofy' can also be correlatedwith the period of rise P^ as shownin Fig. 12.16. These two cerrelations allow solution of Eq. 12.29 and produce a synthetic unit hydrographof duration P^f4 for an ungauged area.

216

12 UNITHYDROGRAPHS CHAPTER

\to
. F h

ql-

"!'

0.0139 2.6761Po+

.Fl6 o t * A
'l

x t!9

40

80

120

160

200

240

280

320

360

Period of rise, Pp (min.)

factor P*fy' andperiodof risePn.(After of storage Figure12.16 Relation Gray.r6; The solutionproceedsas follows: watershed' 1. DetermineL, 5", and A for the ungauged q. and parameters P*, Determine 2. T', Ppfy'. a. With Lf\/5", useFig. I2.I5 to select Pa. Compute y' as the ratio obtain 12.16 to Fig. use With P*/y', b. Pol@*ly'). c. Substitute7' obtained in Step 2b into the equation Q : | + y', and solvefor 4. graph using Eq. 12'29' Com3. Computethe ordinatesfor the dimensionless pute the percent flow in 0.25PRfor values of tf P*: 0.125, 0.375, incrementof tf P* : 0'250 until the sum 0.625,. . . , and every succeeding of the percent flows approximates100 percent. Also compute the peak tfPR: 1. percentage by substituting 4. Computethe unit hydrograph. factorto convertthe volume of the direct runoff a. Computethe necessary over the graph to 1 in. of precipitation excess under the dimensionless entire watershed. 1. The volume of the unit hydrograph: V

V:1in.xAnizx640 Tf nu'
I

n'rnln

x 43'560 ;

.- --^ ft2

graph: Vo of the dimensionless 2. Thevolume

fiun 3. Solve fot 2 q, by equatingV and Vo, sincethey must be equal'

x 60V o : 2 q , x 0 . 2 5 x P 'R

12.6 SYNTHETIC UNIT HYDROGRAPHS

Rdinfall duration = 6 mln = Pnl 4

217

s 400 300

60 Time(min)

Figure 12.17 Derived hydrographof Example12.8usingGray'smethod. Notethat Gray's method results in a unit hydrograph for a P^/4-hr storm.

b. Convert the dimensionless graph ordinatesto the unit-hydrograph ordinates n _ percntflow in 0.25PR\ u,-T.aqi c. Translate time base of dimensionless graph to absolutetime units by multiplying t/PR x P^ for each computedpoint. Rememberthat runoff doesnot commenceuntil the centroid of rainfall, or at a time P^/8. An example problemdemonstrates the solutionof Gray'smethodfor a Missouri watershed. A plot representing the derivedhydrograph is shownin Fig. l2.r7.Dkect runoff commeirces at the centroid of rainfall. Thus it is necessary to add D/2 or Po/B to column 2,Table r2,.5,to obtain the proper times of the unit trydrograph ordinates, showsin Column 6.
EXAMPLE 12.8

For the given data, use Gray's method to constructa unit hydrographfor the Green Acrewatershed, wheredrainage area: 0.62mi2,length: 0.98mi, andS" : l.45%a. Procedure 1. a. b. c. 2. a. Figure12.15; L/t/S: 0.813 ni: Pnfy' : 8.25min. Figure 12.161, P^/^y' : 8.25 min; P*: 24.9 min. q : I I Y' : 4.02; Y' : 3.02. Tabulate percentflow in 0,25PRfortfP*: 0.250:

218

CHAPTER12

UNIT HYDROGRAPHS

TABLE 12.5 TABULATIONFOR EXAMPLE12.8

(1)
r/P

(2) Time, (min) 0 3.1 9.3 15.6 2r.8 24.9 28.0 34.2 40.4 46.7 52.9 59.3 65.5 71.7 78.0 84.2

(3) Percent flowin 0.25Pn 0 0.45 5.80 t2.70 16.35 16.85 16.25 14.20 11 . 1 0 7.97 5.55 3.56 2.28
l.4l

(4) Cumulated flow 0 0.45 6.2 18.9 35.3 5r . 5 65.7 76.8 84.8 90.4 93.9 96.2 97.6 98.5 99.0

(5) UH (cfs) 0
lt.J

(6) time Actual (min) 0 6.1 12.3 18.6 24.8 27.9 31.0 37.2 43.4 49.7 " 55.9 62.3 68.5 '74.7 81.0 87.2

0.000 0.125 0.375 0.625 0.875 1.000 r.t25 t.375 t.625 1.875 2.125 2.375 2.625 2.875 3.125 3.375

490 631 651 628 548 428 308 214 138 88.0 54.4
JJ.J

0.86 0.50

19.3

3. a. 1. V : I x 0.62 x 640 x 43,5601t2: 14.4X 10sft3' x 60 X > q,:373.52 q' 2.Vo:0.25 x 24.9 3.2 q,: 3860. b. Column 5 is tabulated by multiplying 3860 times values in Column 3 divided by 100. c. Column 2 is obtainedby multiplying24.9 times valuesin Column 1. d. Column 3 comesfrom solutionof Eq. l2'2J' rr

SyntheticUnit Hydrograph Espey10-Minute

by EspeyandAltmanlTand wasconducted of 19 urbanwatersheds A regionalanalysis points of a 10-min hyprovide seven equations.that set of regression resulted in a curve through the a smooth fitting by developed is hydrograph The entire drograph. a unit area is case, In either procedures. or curve-fitting eye-fitting points using necessary. (cfs),time base(minThe equationsfor time to peak (minutes),peak discharge rate are peak flow the percent of 75 and utes),and width at 50
* fo'1801 57 S-o'2s To : 3.ILo'23 Qp : 3L62 X I03Ao'e6T-t'o7 Ta: 125.89 x I}3AQ;o'es W5o: 16.22x I03Ao'e3Q-oe2 7eQ-o78 w 1 5: 3 . 2 4 x 1 0 3 4 0

(r2.30) (r2.3r) (r2.32) (12.33) (r2.34)

where

L : total distance (ft) along the main channel from the point being boundary to the upstreamwatershed considered

12.6 SYNTHET|C UNIT HYDROGRAPHS 219 : S main channel slope (ftlft) defined by H/0.8L, where 11 is the differencein elevationbetweenthe point on the channelbottom at a distanceof 0.2L downstream from the upstreamwatershed boundary and a point on the channelbottom at the downstreampoint being considered 1 : imperviousarea within the watershed(7o) watershed conveyance factor Q : a dimensionless A : watersheddrainagearea (mi2) T, : time of rise of the unit hydrograph (min) Q, = peak flow of the unit hydrograph (cfs) I, :,time baseof the unit hydrograph(min) ITso: width of the hydrograph at 507o of Q, @in) = width of the unit hydrographat75%oof p. (min) I4zrr The coefficients of determination (explained in Chapter 26) for the five equationg ranged from 80 to 94 percent. The watershedconveyancefactor is found from Fig. 12.18.The Wso andWrt widths are normally drawnwith one-thirdof the calculated width placedto the left of the peak and two-thirds to the right.

Clark's IUH Time-AreaMethod

A syntheticunit hydrograph that utilizes an instantaneous unit hydrograph(IUH) was developed in 1945 by clark.6It has been widely used,is often called the time-area method,and hasappeared in severalcomputerprogramsfor hydrographanalysis(see Chapters24 and25).

80

b 7 0 . 6 0

F s o
;o 4 0
?
d

6 3 0

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0..130.14 0.15 0.16 0.17
Main channel Manning z value

Figure 12.18 Watershedconveyancefactor Q as a function of percent watershed impervious cover l and weighted main channel Manning n value, for Espey method.

220

CHAPTER12

UNIT HYDROGRAPHS

at any point in time is a function of recognizes that the discharge The technique of the watershed.The translation is obthe translation and storagecharacteristics tained by estimatingthe overland and channeltravel time of runoff, which is then effectsof a watershed. by the storage combinedwith an estimateof the delaycaused mouth rainfall from its point of falling to the watershed The translationof excess This is a histogramof is accomplished using the time-area curve for the watershed. lines . The dashed as shownin Fig. 12.19 incrementalrunoff versustime, constructed in Fig. I2.I9a subdividethe basin into severalareas.Each line identifiesthe locus of "times" are drawnequal pointshavingequaltraveltimesto the outlpt. The isochrones to fully definethe time-area relation. apart, and sufficient zonesare selected The time-area graph of Fig. l2.I9b is a form of unit hydrograph.The area to 1.0 unit of rain depth over the total areaA, and it has beneaththe curve integrates \ are delineated. a translationhydrographshapeif sufflcient subareas at t : 0, the runoff from At If one unit of net rain is placedon the watershed rate of At units of runoff would passthe outlet during the first At period at an average would be At units of areatimes one unit of pei unit of time. The volume discharged rain. After all areascontribute,one unit of rainfall over the entire area would have passed the outlet.

Lt

I
I

J
^1

.l
A2

J
^3

Lt
A

As
Time interval

ft) of time-area histogramfor use Figure L2.19 Development with Clark's method: (a) isochronesspacedLt apart (shown as dashedlines) and (b) time-area histogram.

12.6 SYNTHETICUNIT HYDROGRAPHS

221

The impact of watershed storageon the translationhydrographis incorporated by routing the time-area histogramthrough a hypotheticallinear reservoirlocatedat the watershedoutlet, having a retardance coefficient K equivalentto that of the watershed. For the simplest form of reservoir,the storage S,at time f is linearly related to the outflow Q, at time /, or S,': KQ,

(r2.3s)

whereKis a constantof proportionality calledthe storage coefficient.It hastime units and is often approximated by the lag time of the watershed. From continuity,the inflow, storage, and.oq,{flow for the reservoirare relatedby

I,-Q,:#:U#
Q'

(r2.36)

If the differentialis discretized to LQ/LI, andif Q, andQ, are the flowsat t andt - l, thenEq. 12.36becomes Io,- Ao,: YQ'Lt

(12.37)

BecauseQ : (Q, + Qr)/2, the flow at the end of any Al is Qz: CoI * CtQ,

(12.38) (r2.39) (r2.40)

where
and

cr: #+T;
- Lt 2K " ,t -- z K+Lt

The IUH is found from Eq. 12.38by solvingfor Q, at the end of eachsuccessive time interval. EXAMPLE 12.9 Given the following 15-min time-area curve, find the IUH for the 1000-acrewatershed.Then determinethe 15-min syntheticunit hydrograph.The storagecoefficient K is 30 min.

Timeinterval (min) 0 - 1 5" 15-30 30-45 45-60

Areabetween isochrones (acres) 100 300 500 100

Solution. From Eqs. 12.39and 12.40,Cs : 0.4 and Ct : 0.6.Routing is most easily accomplished in a tableauas follows:

222

CHAPTER12

UNIT HYDROGRAPHS

Time (hr) 0

I (acre-in./Af )

I (cfs)

col + c1Q1

IUH (cfs)

100 0.25 300 0.50 500 0.75 100 1.00 0 1.25 0 1.50 0

400 1200 2000 400 0 0 0

160+0 480 + 96 800 + 346 160+ 688 0+509 0+305

0 80 368 861 99'7 6't9 l

due to the magnitudeof K for long recession The IUH has a characteristically 0.6 times the previous flow becomes hr, the this example.Note that after 1.25 in Section11.5, As discussed rate indefinite$. to decayat this flow andcontinues time, which is release excess-runoff the the time baseof the IUH shouldequal proproduces often method Clark's one definition of time of concentration. of this shortcoming. longedrunoff because The 15-min unit hydrographis found using Eq. 12.6, ot

UH) : lGUH, + IUH,-15) O,(l5-min

This resultsin:
Time IUH 1S-min UH (cfs)

(h0
0 0.25 0.50 0.75 1.00 t.25 1.50 :

(efs)
0 160 576 1146 848 509 :

0 80 368 861 997 679 : tl

If the waterri"O ,uU time is not available,the K value can also be estimatedby recognizingthat Q, = KdQldt when the inflow is zero in Eq. 12.36.This occurs at of Fig. t2.2, when inflow to the approximatelythe inflection point on the recession of the Kvalue is the ratio the estimate If hydrograph dataare available, channelceases. point on the hyparticular this at of the flow rate to the slope of the hydrograph drograph.

12.6 SYNTHETIC UNITHYDROGRAPHS 223

Nash'sSyntheticIUH
One of the earliest formulations of the IUH was developedby Nash.l8Instead of characteizingrunoff as translationfollowedby storage in a singlelinear reservoiras Clark did, Nash viewed the watershedas a seriesof n identical linear storagereservoirS,eachhavingthe samestoragecoefficientK. The first instantly (r : 0) receives a volume equal to a full inch of net rain from the entire watershed. This water then passes through reservoirsI,2,3, . . . , fl, with eachproviding an additional diffusion effecton the original I -in. rain. The number of reservoirs, n, is uniquely related to the reservoir storage coefficientK andthe watershed lag time. Once the IUH is developed; it can be used to synthesizeany other hydrograph by application of the convolution integral, Eq. I2.7, or from the approximate methods discussed in Section12.3. The derivation of Nash's equationfor IUH beginsfrom continuity at the first reservoir:

I,- Qu: #1,

wherc Qt, is the outflow from reservoir 1 at time t. SubstitutingS,: r ) 0 (for an IUH, I, is zero after t : 0), we obtain

(r2.4r)
KQr, at time

_Qr,: u#1,,, #1,,,:

which can be written dQ"

(r2.42)

eu:
Integrationfrom / : 0* to time t gives

' ' - kI o

(r2.43)

l n Q r , - l n Q r , l , - o :- which reducesby taking antilogarithmsto

Qt,

(r2.44)

o,l;:
BecauseQr, : S,/K and S,=o: 1 in., then

e ''-ttK

(r2.4s)

Qu:f,"-'r*

(r2.46)

wherc Qy has units of depthper unit of tirne. Equation 12.46is an exponentialdecay function having an initial value of I/K at t : 0. This monotonically decreasing outflow from reservoir 1 becomesinflow to the secondreservoir. The secondreservoiris initially empty. The continuity relation
Qu * Qz,:

K+
dt

(r2.47)

224

CHAPTER12 UNIT HYDROGRAPHS

is solved,giving | ,.. (r2.48) 9r, : V;te-"^ This equationhas a full hydrographshape,beginningwith zero flow at time zero, peaking at the maximum of the function, and eventuallyrecedingto zeto. Similarly derived,the hydrographflowing from ruthreservoirhas the form
Qn, : |n-r n-t/K

, (12.49)

which is the two-parametergammafunction,

e,,:

(12.50)

#1n-1r-t/K

rain falling by 1 in. of excess Because the outflow from the nth reservoirwas caused an IUH. Eq. 12.50describes instantaneously,

for applicationof Nash's EStimatiOn of K and n Valuesof K and n arc needed (Eq. 12.50) occurs at t : nK. IUH. By integration,the centroid of the distribution - 1). The From classicalcalculusmaximization,the peak flow occurs at t : K"(n : 0 is n(n + l)Kz. Trial combinationsof n and secondmoment of the IUH about / K can be usedto developthe IUH from Eq. 12.50, and the momentsof the plotted distribution can be estimatedto verify the productsnK and n(n * I)Kz ' If the IUH by is discretizedinto m At increments,the momentsare approximated First moment = > ttQi Lt
m

(r2.sr) (r2.s2)

and

- 2 t?Q, Lt moment Second

Another less tediousapproachis to use the definition of lag time as the time from For an IUH, this is the sameas the centroid of rain to the centroid of the hydrograph. from equationssuchas determined can be time centroidal distance.Thus, if the lag the number of reducing be established, product nK can those in Section 11.6, the trials. to basin haveattemptedto relateNash'sK andn parameters Someinvestigators relations et al.1e developed Rao techniques. usingregression and storm characteristics for urban areasgreaterthan 5 mi2:

K :
and
Lt -

po.zzz g.57 54o.zto (l + 1)o'azzPo:oa 0.83140.458D0.37r (I + 1yt'ezzPozat

(r2.s3) (r2.s4)

is the wherenK : tlandA is areain squaremiles,D is net rain duration in hours,Pn", net (effective)rain depth in inches,and / is the ratio of imperviousto total area.

12.6 SYNTHETICUNIT HYDROGRAPHS

225

ColoradoUrbanHydrograph (CUHP) Procedure

Synthetic unit hydrographscan be tailored for regional use. As an example,the Coloradourban hydrographprocedure20 provides 5-min syntheticunit hydrographs for use in the Denver metropolitan area. lt is based on Snyder's method and is considered applicableto watersheds in the size rangefrom 90 acresto 10 mi2, with "regular" shapes(length 4 times width). It was developed in the early 1980sand modifiedin 1984to reflectrefinements from earlyapplications. In its pre-1984'form, the Snyder/CUHPC, and C, valuesfor use in Eqs. 11.5 and 12.17 are
C, : 7.8I1P2378 and Co: 0.89C?'46

(r2.ss) (r2.s6)

where P" is the percent impervious.Theseregression equationswere developed for P" > 30 percent, using data for 96 storms over 19 urban watersheds. The given equations apply to normal watershed conditionsandneedto be adjusted for steep, flat, or sewered basins. If an urbanareais fully sewered, the calculated C, from Eq. 12.55 is decreased 10 percent. If sparselysewered,a 10 percent increaseis made. If the averageslope S of the lower 80 percent of the main water courseis flat (less than 0.01 ftlf| the C, value becomes C,:3.12/Po78so2 and for steepareas(,S> 0.025 ftlf$, C, : 3.75/Po78s'02

( r 2.57) ( 12.s 8)

The C, coefficient is determinedfrom Eq. 12.56 and adjustedto 10 percentup or down for fully or sparselysewered conditions,respectively. For the CUHP applications, Eqs. 12.18 and 12.19become ryro: SO}A/Qo w75:260A/Qo

(r2.se)
(12.60)

whereA andQohaveunits of squaremiles and cubic feet per second. For plotting ft., the smallerof 35 percentor 0.670is placedleft of the peak.For W15,45 percentis placedto the left, or O.424To if 0.67pwas usedfor W'r6. I is the time from beginning of runoff of the unit rainfall to the peak time. Severalinvestigatorshave suggested that Snyder's C, and slope S are correlated.a'S The original CUHP procedurewas alteredto recognizethis relation, making the adjustments in C, unnecessary. For the modifiedversion,Eqs. 12.57and 12.58are bypassed, andthetime to peakratherthanlag time is usedin Snyder's Eq. 11.5,where tp: c,(LL,o/\8100' (12.6r)

The revisedtime coefficientC, is obtainedfrom Fig. 12.20a,and the peak coefficient is found from
Co : PC,Ao'1s

(r2.62)

where

S : weightedaverage slopeof basinalongthe streamto the upstream basin boundary (ftlft)

CHAPTER12

UNIT HYDROGRAPHS
0.18

Equations of curve: Ct= aI?+ blo+ c

0.16 tro. I 00
b c -0.00371 0.163 0.146 0.000023 -0M224 ] 3 x 1 o r -8.01 x l0- 0 12(

Basic equations I LL--\0.48 'r: "\fi) 4p : 640c, --7-

0.14
r i

\r

tp = time to peak (br) L = watershed length (mi) Ic, = distance to centroid (mi) So = waterway slope (ff/ft)

E q
o

U 0.10

0.08 3q. 0.10

Ec,.2l
20

Eq.3 80
100

40 60 Percentimpervious,l, (a)

12
Equations of curve: P = aI?+ blo+ c

10

Eq.

\t<

2 | -o.t)0091 +0.228 | --T-T-l

b -0 012

+216

06

z-r ,ql.: v t Y

\ o 6

Eq. 1 0 20

84.2
40 60 Fercent impervious, l, (b) 80 100

Figure 12.20 Snyder's C, and C, coefficientsfor urban areas,for use with CUHP: (a) relation between C, and imperviousness and (b) relationbetweenpeakingparameter and imperviousness.

PROBLEMS

227

from Fig. I2.20b P : coefflcient,dependingon imperviousness, A : drainagearea (mi2) L : length of the main stream channel (mi) from the outlet to the divide L"o : length along the main channel (mi) from the outlet to a channelpoint centroid nearestthe watershed usingEqs. 12.17 canbe developed The peakrate andtime of the unit hydrograph areplotted, t2.60 12.59 and Eqs. by points defined additional After four 12.6t. and representing area provide total a to fitted be can hydrograph the 5-min rest of the 1.0 in. of direct runoff. A hand fit is applied, or the mathematical curve-fitting techniquesdescribedearly in Section 12.5 can be used if a gamma (or any other) appropriate. distribution is considered (lessthan 90 acres),the time to peak is watersheds For small
Lp

* 0 . 0 7()1, ^ : o.zg(P=l.?3,6=P: + :0- .1 4 ' ^ " Pz - 0.49P"

6P,)

02.63)

where I is the time of concentrationin minutes, and P^ is the percentimpervious.

r summary
Unit hydrographmethodsallow the hydrologistto estimaterunoff volumesand rates for virtually any storm. By far, the greatestnumber of problems in practice are Most of the current computermodelsuse procedures. usingunit hydrograph evaluated rs 23,24, and25. Thesemodelsare in Chapte procedures asdescribed unit hydrograph and convolusimply computerprogramsthat perform the unit hydrographsyntheses the origin, in this chapter.Any softwareuser shouldunderstand tion stepsdescribed method hydrograph for eachunit applicability,and parameterestimationprocedures thorough a from usesof the computermodelswill result The most successful seiected. in this chapter. described familiarity with the processes

PROBLEMS
a triangular unit hydrographfor one 12.1. Given the following storm pattern and assuming hydrograph. composite the determine time unit,

Stormpattern
Time unit Rainfall

1 1

2 1

3 4

4 2

Unit hydrographbaselength : 6 time units; time of rise : 2 time units; and maximum ordinate : I rainfall unit height. 12.2. Given a rainfall duration of 1 time unit, an effectiveprecipitation of 1.5 in., and the following hydrograph,determine(a) the unit hydrographand (b) the compositehydrographfor the given storm sequence.

228

CHAPTERl2

UNIT HYDROGRAPHS Hydrograph for 1.S in. net rain in .t time unit Time units 1 2 3 Flow (cfs) r00 98 220 4 4.5

srz 620 585 460

7 8 330 2to

9 r 0 150 105

l 1 l2 13 75 60 54

Storm sequence
Time units Precipitation (in.)
I

0.4

z l.l

2.0

1.5

12.3. Solve Problem lZ.2 if the storm sequence is as follows:

Storm sequence
Time units Precipitation (in.)

r 0.3

2 r.4

0.9

Using U.S. GeologicalSurveyrecords,or other data, selecta streamflow hydrograph for a large, preferably single-peaked runoffevent. Separate the baseflow and determine a unit hydrograph for the area. 12.5. For the unit hydrographof problem 12.1, constructan S_hydrograph. 12.6. For the unit hydrqgraphcomputedin problem 12.2, construct an S_hydrograph. 12.7. use the S-hydrograph of problem 12.6to find a 3 time-unit unit hydrograph. 12.8. Given a watershedof 100 mi2, assumethat C, = 1.g, the length of main stream channelis I 8 mi, and the length to a point nearestthe centroid is I O mi. Use Snyder,s methodto find (a) the time rag, (b) the duration of the syntheticunit hydrograpl, and (c) the peak discharge of the unit hydrograph. l2'9' Apply Snyder's method to the determinationof a synthetic unit hydrographfor a drainage area of your choice. 12'10' Use Fig. 12.13to determine a2-hr pnrthydrograph if the drainageareais 60 mi2 and Eq. 12.23ais applicable. 12.11. SolveProblem12.10using Eq.12.23b. 12.12. Assuminga Nebraskalocation, use Gray's method determine te a unit hydrograph: drainagearea = 1.0 mi2, length : 0.6 mi, S" : 1.3 percent. 12'13' A drainageareain Nebraskacontains30 mi2.The tenjtn ortne main channelis 10 mi andthe'representative watershed slopeis 2.5 percent.iJseGray's methodto determine a unit hydrograph. 12'14' Dischargerates for a flood hydrographpassingthe point of concentration for a 600acredrainagebasinare given in the tabl; belo;. The flood wasprotluced by a uniform rainfall rate of 2.15 in.rru, which started,at9 A.M., abruptry ended at il A.M: and resulted in 5.00 in. of direct surfacerunoff. The base flow (derived tiom influent seepage) prior to, during, and after the storm was 100 cfs.

l2'4'

PROBLEMS

229
6

8 a.u. Time Measured discharge 100

10

11

12

I P.M

100 300 500 700

800

600 400 300 200 100 100

cease? a. At what times did direct runoff begin and basin' the for (in'/hr) index the Determine b. @ ordinates(cfs)for eachtime.listed' c. Derive the 2-hr'unit-nydrograpir time) for the basin. release (excess concJnffation d. Estimatethe tirne of e.Atwhattimewoulddirectsurfacerunoffceaseiftherainfallof2.T5in./hrhad and had lastedfor 8 hr rather than 2? begunat 9 .q.'l\a. (in.) for a uniform urg" .ur" (cfs) and the direct runoff f. Determined;;;i;t rainfall of 2j5 it'lhr and a duration of 8 hr' rates Lz.lS. Measuredtotal hourly discharge table' The hydrr in the accompanying uniform intensity of 2'60 in'/hr startir baseflow from 8 A.M' to 3 P'v' was a ' determined as the area under the dir

Time Measured discharge

8 a.v 100

10 100 300

11 600

12 400

I P.M

2 100 100

200

a. At what time did the direct runoff begin? to the volume of the direct surface b. betermine the net rain (in') corresponding runoff of 1000 cfs-hr' the { indexfor the basin' c. Determine basin by tabulating time in hours and d. Derive " , nt ltii ttyJrog'api' for the in cfs. discharge time of the basin? release e. What is the excess the direct tt" the derived2-hr unit hydrograptr-to-determine f. For the ,"*"'t"'it, p.u. and 1 at began ralnfall ;;; ;.*, on a day when excess-(netj runoff rate (.fr) 't"i P'M' 5 at abruptly ceasing iti"tt ity of 2 in'/hr for 4 hr' continued"t " |2.|6.A5-hrunithydrographfora425}-acre.basinisshownintheaccompanyingsketch. Thegivenhyd.og.-uphactuallyappearedasadirectrunoffhydrographfromthebasin' of 5 hr,beginning

by rainf;ffi;;;;ii"'"iriv caused

in./hrfor a duration "i 0.30

att = 0. time of the basin' release a. Determinethe excess basin' the for index b. Determinethe @ 4 hr after t"r* was contributingto direct runoff JJr"g" "'i,ti" c. what n".""*"J" rain began(r = 4)? sketch'Do not scale to part c to determineQp' as shownin the d. Use your response Qp from the drawing' t = 3 andr : 5' Why did the hydrograph e. Note that rain continuedto fall between = : rathet than continue to rise during those 5, t 3 and form a ptateaut"i*""n t 2 hours?

230

CHAPTER12

UNIT HYDROGRAPHS
QO

,i: 700 = ouu I 500

e 40o

! :oo $ zoo
100

Time, t

f. Usethe given 5-hr unit hydrograph to determine the direct runoff rate (cfs)at 7 p.r'r. on a day when rain fell at an intensity of 0.60 in./hr from I p.rrr. to 11 p.vr. 12.17. The 2-hr unit hydrographfor a basin is given by the following tabie:

Time (hr) O@fs)

0 0

1 60

2 200

3 300

4 200

5 120

6 6 0

7 8 3 0 1

9 0

a. Determinethe hourly discharge values(cfs)from the basinfor a net rain of 5 in./hr and a rainfall durationof 2hr. b. Determinethe direct runoff (in.) for the storm of part a. What is the direct runoff for a net rain of 0.5 in./hr and a duration of Z Iv? c. Rain falls on the basinat arateof 4.5 in.lhr for a2-hr period and abruptly increases to a rate of 6.5 in./hr for a second2-hr period. convert theseactual intensitiesto netrain intensities usinga {index of 0.5 in./hr. Constructa tablethatproper$ lags and amplifiesthe 2-hr unit hydrograph, and determine the hourly ordinates(cfs)of direct runoff for the storm.The deriveddirect runoff hydrographshouldbegin and end with zero discharge values. 12.18. Given the following 2-hr unit hydrograph for a drainage basin, determine hourly ordinatesof the 4-hr unit hydrograph:

Time (hr) 0@f9

0 0

1 50

2 300

3 400

4 200

5 s0

6 0

12.19. Usethe following 4-hr unit hydrograph for a basinto determine the peak discharge rate (cfs)resulting from a net rain of 3.0 in./hr for a 4-hr duration foliowed immediately by 2.0 in.lhr for a 4-hr duration.

Time (hr) 0(cfO

0 0

2 200

4 300

6 100

8 50

10 0

PROBLEMS 231 12.20. Compare the time from the peak to the end of runoff for the SCS triangular unit hydrographwith the time of concentration,/". Discuss. hydrograph 12.21. Prove that the areaunderthe rising limb of the SCSbasic dimensionless equalsthat of the triangular unit hydrograph,that is, 37.5 percefi of the total. 12.15occurswhenx:aB, 12.22. Bycalculus,showthatthemaximumvalueof/(x)inEq. by taking the flrst moment aboutthe for a > 1. Also solvefor the centroidaldistance y axis. 12.23. Accordingto the rational method(seeChapter15) of estimatingpeakflow from small areasn the peak rate for a storm with uniform continuingintensity is equal to the net For what conditions,if any, would rain rate and occurs at the time of concentration. ' of the peak magnitudeand time, estimated Eqs. 12.63 and 12.17result in agreement by CUHP, with those of the rational method?Discuss. L2.24. Describetwo methodsthat could be usedto constructa 2-hr unit hydrographusing a l-hr unit hydrographfor a basin. rates (cfs)from a 2.48-n12drainagebasin are tabu12,25. Measuredtotal hourly discharge lated below.The hydrographwas producedby a rainstormhavinga uniform intensity of 2.60 in./hr starting at 9 A.M.and abruptly ending at 11 l.rvl. The baseflow from 8 ,q,.lt. to 3 p.lrl.was a constant 100 cfs.

Time Discharge(cfs)

8 e.u. 100

9 100

10 300

11 450

12 300

1 p.vr. 2 150 100

100

a. At what time did direct runoff begin? b. Determinethe grossand net rain depths(inches). c. Derive a 2-hr unit hydrograph for the basin by tabulating time in hours and in cubic feet per second. discharge d. Derive a 4-hr unit hydrographfor the basin. e. Derive a l-hr unit hydrographfor the basin. 12,26. Given below is a 3-hr unit hydrographfor a watershed.The {-index is 1.5 in./hr. 3-hr rainfall ratesof 2.5, Desiredis the DRH for an 18-hr stormhavingsix successive 3 . 5 , 1 . 5 ,4 . 0 , 6 . 5 , 2 . 5i n . l h r .

Time (hr)

10 11 12 13 14

o(ruH) 0 10 40 60 80 100 90 70 60 50 40 30 2 0 1 0 0
12.27. Use the following 2-hr unit hydrographto determinethe peak direct-runoffdischarge rate (cfs)resulting from a net rain of 2.0 in./hr for 5 hr'

Time (hr) 0(cf9

0 0

1 50

2 200

3 300

4 200

5 1s0

6 100

7 0

12.28, The ordinatefor a 5-hr unit hydrographis 300 cfs at a time 4 hr after the beginning of net rainfall. A storm with a uniform intensity of 3 in,/hr and a duration of 5 hr occursover the basin.What is the runoff rate after 4 hr if the @index is 0.5 in./hr?

232

CHAPTER 12 UNITHYDROGRAPHS 12.29. Given below is an IUH for a watershed. Use the IUH to find hourly DRH rates for a net rain of 4 in. in a 2-hr oeriod.

Time (hr)

o(ruH)

0 0

1 10

2 40

3 50

4 60 80

6 100

7 80 20

8 r0

9 10 0

12.30. A 2-fu unit hydrographfor a basin is shownin the sketch. (cfs)for a net rain of 5.00 in./hr and a duration of a. Determinethe peak discharge 2 hr. b. What is the total direct surface runoff (in inches)for the storm described in part a? c. A different storm with a net rain of 0.50 in./hr lastsfor 4 hr. What is the discharse at 8 p.vr.if the rainfall startedat 4 p.tvt.?
I I I

{, +oo
po
I I

--t

.A 2oo

T I I I

--i

I I I I I
I

I I -T I
I I

Time (hr)

12.31,. Recordedflow rates for a net rain of 1.92 inchesin 12 hours are shownin the table. ' If the baseflow is 375 cfs throughoutthe storm,determinethe 12-hrunit hydrograph, and convert it to a 6-hr unit hydrograph.Then apply the 6-hr unit hydrographto determinethe total hydrograph(including 400 cfs baseflow) for a24-hr storm having four 6-hr blocksof net rain at ratesof 0.7, 3.8, 10.8,and 1.8 in. per hour.

Time in hours

Observedflow (cfs)

0 6 12 18
aA JU

36
A'

48 54 60 66 72 78

375 825 2200 36s0 3900 3200 2375 1,725 1250 900 650 490 410 375

12.32. Starting with a triangular-shaped unit hydrographwith a baselength of 2.67toand a height of qo,deriveEq. 12.22,qo : 484A/tp. Statethe units of eachterm usedin the .- --- derivatign"

REFERENCES 233 rate 12.33. The SCS syntheticunit hydrographis derivedby computingthe peak discharge (cubicfeet per second)from qo : 484A/tp.In the derivation,it was actually assumed thatqrin.lhr:0.7sv/tp,whereVisthevolumeofdirectrunoff(inches),roisthetime to peak flow (hours),and A is the basin area(squaremiles).Derive the first equation from the second. for synthesizing a unit hydrographrequiresthe leastcompu12.34. Which of the techniques Which probably requiresthe the entire unit hydrograph? tationai effort in developing most?

REFERENCES
Division, 1948. 1. W. D. Mitchell, "Unit Hydrographsin Illinois," illinois Waterways "Stream-Flowfrom Rainfall by the Unit-GraphMethod," Eng.News-Rec. 2. L.K. Sherman, 108,501-505(Apr. 1932). "Unit Hydrographsfor Gaugedand Ungauged 3. Rand Morgan and D. W. Hulinghorst, Watersheds," U.S. EngineersOffice, Binghamton,NY, July 1939. 4. F. F. Snyder,"synthetic Unit Graphs," Trans.Am. Geophys.Union 19,447 -454(1938). 5. A. B. Taylorand H. E. Schwartz,"lJnit HydrographLag and PeakFlow Relatedto Basin s. Union 33' 235-246(19 52). Characteristic s," Trans. Am. Geoplry ASCE Trans.ll0,1419-1446(1945). 6. C. O. Clark, "storageand the Unit Hydrograph," Division. Washington, 7. Water SupplyPapers, U.S. GeologicalSurvey,Water Resources D.C.: U.S. Government PrintingOffice, 1966-1970. 8. Hourly PrecipitationData, National Oceanicand AtmosphericAdministration.WashingPrinting Office, L971. ton, D.C.: U.S. Government Jr., "Synthesisof the Inlet Hydrograph,"Tech.Rept.No. 3, Department 9. JohnC. Schaake, The JohnsHopkins University,Baltimore, of Sanitary Engineeringand WaterResources, MD, 1965. 10. G. Aron andE. White, "Fitting a GammaDistribution overa SyntheticUnit Hydrograph," Resources Bull.18(l) (Feb.1982). Water Bull. l9(2) (Apr. 1983). WaterResources 11. G. Aron and E. White, "Replyto Discussion," 12. M. Collins, "Discussion-Fitting a Gamma Distribution over a Synthetic Unit HyBull. 19(2) (Apr. 1983). drograph,"WaterResources Engin' 13. "Flood-HydrographAnalysisand Computations,"U.S. Army Corpsof Engtneets, D.C.: U.S. Government neering and Design Manuals,Emlll0-2-1405. Washington, PrintingOffice,Aug. 1959. "Hydrograph Synthesisby Digital Computer," Proc. 14. M. D. Hudlow and R. A. Clark, ASCEJ. Hyd. Div. (May 1969). in SyntheticHydrographAnalCharacteristics 15. V. Mockus, "Use of Storm and Watershed Service,1957. ysisandApplication,"U.S.Department of Agriculture,Soil Conservation for Small DrainageAreas," Proc' ASCEJ. Hyd. 16. D. M. Gray,"synthetic Unit Hydrographs Div. 87(HY4) (July 1961). for Small for Ten-minuteUnit Hydrographs 17. W. H. EspeyandD. G. Altman, "Nomographs Urban Witersheds,"EnvironmentalProtectionAgency,Rept.EPA-600/9-78-035,Washington, D.C., 1978. Unit Hydrograph," IASH Publ. No. 45' Vol. 18. J. E. Nash, "The Form of the Instantaneous ' 3. 1951. "ConceptualHydrologicModels for Urbanizing 19. R. A. Rao, J. W. Delleur, and B. Sarma, Basins,"Proc. ASCE J. Hyd. Div. (HY7) (July 1972). "First Short Courseon Urban Storm WaterModeling 20. University of Coloradoat Denver, Using ColoradoUrban HydrographProcedures,"Departmentof Civil Engineering,June 1985.

C h a p t e r1 3

Routing Hydrograph

r Prologue
The purposeof this chapteris to: . Presenttechniquesfor determiningthe effect of streamsand reservoirson hydrographshapes movedownstream through the systems. asthe hydrographs . Distinguishbetweenthe two major classifications of hydrograph routing techniques. . Familiarizethe readerwith procedures for determiningwhen to apply eachof the variousrouting methods. water Flood forecasting, reservoirdesign,watershedsimulation,and comprehensive planning generallyutilize someform of routing technique.Routingis used resources a river to predict the temporal and spatial variationsof a flood wave as it traverses reach or reservoir. Routing techniquesmay be classified into two categorieshydrologicrouting and hydraulic routing. Hydrologic routing employsthe equationof continuity with either a linear or curvilinear relation betweenstorageand dischargewithin a river or reservoir. Hydraulic routing, on the other hand, usesboth the equation of continuity and the equationof motion, customarilythe momentumequation.This particular form utilizes the partial differential equationsfor unsteadyflow in open channels.It more adequately describes the dynamicsofflow than doesthe hydrologicrouting technique. Applications of hydrologicrouting techniques to problemsof flood prediction. of the effectsof urbanization evaluations of flood control measures, and assessments are numerous.Most flood warning systems instituted by NOAA and the Corps of Engineersincorporatethis techniqueto predict flood stagesin advanceof a severe storm.It is the methodmost frequentlyusedto sizespillwaysfor small, intermediate. and large dams. Hydrologic river and reservoirrouting and hydraulic river routing techniques are presented in separate sectionsof this chapter.

RIVER ROUTING 13.1 HYDROLOGIC

235

RIVERROUTING 13.1 HYDROLOGIC

The first referenceto routing a flood hydrographfrom one river station to another was by Graeff in 1883.1 The technique wasbasedon the useof wavevelocity and a rating curve of stageversus discharge.Hydrologic river routing techniquesare all founded upon the equationof continuity dS (1 3 . 1 ) I - O : dt where 1 : the inflow rate to the reach O : the outflow rate from the reach dS/dt : the rate of changeof storagewithin the reach

Three of the most popular hydrologic river routing techniques are described in subsequent paragraphs,

Method Muskingum
Storagein a stableriver reachcan be expected to dependprimarily on the discharge of the channelsection.The into and out of a reach and on hydraulic characteristics as2 storagewithin the reach at a given time can be expressed

s:2;y7^n+(1 - X)O-n1
a

(r3.2)

of control sections at each Constants a andn reflectthe stagedischarge characteristics characteristics of the section. end of the reach,and b and m mirror the sta$e-volume The factor X defines the relative weights given to inflow and outflow for the reach. The Muskingum method assumesthat mfn - L and lets b/a : K, resulting in

S: KIXI + (1 - X)o]
where K : the storagetime constantfor the reach x : a weighting factor that variesbetween0 and 0.5.

(13.3)

closeto the wave Application of this equationhas shownthat K is usually reasonably trarel time through the reach andX averages about 0.2. Behavior of the flood wave due to changesin the value of the weighting factor X is readily apparentfrom examinationof Fig. 13.1. The resulting downstreamflood waveis commonlydescribed by the amountof translation-that is, the time lag-and As can be noted from by the amount of attenuationor reduction in peak discharge. Fig. 13.1,'thevalueX : 0.5 resultsin a pure translationof the flood wave. procedure Application of Eqs. 13.1 and 13.3 to a river reachis a straightforward if Kand X are known. The routing procedurebeginsby dividing time into a number Eq. 13.1 in finite difference form, using of equal increments,A/, and expressing subscriptsI and2 to denotethe beginningand ending times for Ar. This gives I t + 1 2_ O 1 + 0 2 _ S ' - S ' 2 2 A t

(13.4)

236

CHAPTER13

ROUTING HYDROGRAPH

99 E

Time

Figure 13.1 Effect of weighting factor. The routing time interval A/ is normally assigned any convenient value between the limits of K/3 and K. is in the river reachduringthe routing interval from Eq. 1 3 . 3 change The storage

s, - s' : Klx(I' - 1')+ (r - x)(o2- o')l

O2: CsI2+ CII. + CzOl

(13.s)
( 13.6)

and substitutingthis into Eq. 13.4 resultsin the Muskingum routing equation

in which

co

-KX + 0.5Ar K-KX+0.54/ KX + 0,5 Lt ct: K-KX+0.54t K - K X- 0 . 5 A r

cz:

(r3.7)
(13.8)

K-KX+0.54t

(13.e)

Note that K and Ar must havethe sametime units and aiso that the three coefficients sum to 1.0. if Al falls beTheoretical stability of the nunrericalmethod is accomplished the time required K is of value theoretical The X). tweenthe limits 2KX and2K(I the titne It is approximately reach. the traverse for an elemental(kinematic) waveto the wave If not, available. are peaks, if data interval betweeninflow and outflow volocity average of a function as shapes for variouschannel velocity can be estimated uniform flow canbe estiinated flow rate Q. Velocityfor steady Vfor any representative wavevelocities for differapproximate The by either thi Manning or Chzyequation: 13.1. are given in Table ent channel'shapes FOR VARIOUS VELOCITIES WAVE 13.1 KINEMATIC TABLE '
CHANNELSHAPES Channelshape Wide rectangular Triangular Wide parabolic Manningequation
t-V !v
lJrl 9 '

Ch6zyequation
3-tl 2 l

tv Zrl
6 '

ROUTING 237 RIVER 13.1 HYDROLOGIC by Since1, and Irareknown for every time increment,routing is accomplished next time for the Ol 02as each using increments time solvingEq. 13.6for successive incremeni. Example 13.1 illustratesthis row-by-row computation'

EXAMPLE 13.1 : 2 days.The Perform the flood routing for a reach of river given X : 0.2 and K : 1 day is shownin Table 13.2,column 1. Assumeequal inflow hydrographwith Ar inflow and outflow rates on the 16th. Solution. If Ar : l dayandX = 0.2 andK :2days, thenEqs.13.7toI3.9 is give Co : 0.0477,C1 : 0.428, andC2 : 0'524' Row-by-rowcomputation ll given in TabIe13.2. routing of K andX for Muskingum Determination of Muskingum K andX Values average usingK equal to the travel time in the reachand an are commonly estimated value of X : 0.2.If inflow and outflow hydrographrecords are availablefor one or more floods,the routing processis easilyreversedto provide better valuesof K and values of S versus X for the reach. To illustrate the latter method, instantaneous 13.2 TABLE
(1)
(2\

(3)
Crl't

(4) C,Q,
) )7)

Date 3-t6 17 18 19 20 21, 22

z3

lnflow

voI2

(5) Computed outflow 4,260 4,419 6,rr9 8,783 12,791 16,941 1 9 ,1 0 23,578 34903 46,705 51,469 49,109 41,514 32,67' | 34,120 39,559 43,729 42,199 37,569 29,166 22,128 t6,932 13,222 10,576 8,497

24 25 26 27 28 29 30
J I

4-l 2 4 5 6 7 8 9

4,260 7,646 11,167 16,730 21,590 20,950 26,570 46,000 59960 57,'t40 47,890 34,460 21,660 34,680 45,180 49,r40 41,290 33,830 20,5t0 t4,720 11,436
o )04

364 532 798 1,029 999 L,267 2,194 2,860

)'754 ) )24

t,823 3,2'72
4 114

7,160 9,240 8,966 tl,37l 19,688 25,662

)a1l)

2,315 3,206 4,602 6,702 8,877 10,013 12,355 18,289

)4 417

1,643 1,033 1,654 t r55 r,969 t,613 9'78 702 443

3 t J

20,496 14,748 9,270 14,843

1q??7

26,9'70
)\ 117 )'t 171 t7 1))

7,831 6,228 6,083

2t,031, 17,672 14,479 8,778 6,300 4,894 3,977

? ?{1 ) 66\

l7,879 20,729 22,914

)t 11)

29'| 290

19,686 t5,283 11,595 8,872 6,928

238

CHAPTER13

HYDROGRAPH ROUTING

XI + (l -- X)O are flrst graphed for severalselected valuesof Xas shownin Example 13.2.Because S andXI + (1 - X)O are assumed to be linearlyrelatedvia Eq. 13.3, the accepted value of X is that which gives the best linear plot (the narrowestloop). After plotting, the valuefor K is determined as the reciprocalof the slopethrough the narrowest loop, sincefrom Eq. 13.3

K :

n+0-x)o

(1 3.10)

Instantaneous valuesof S for the graphsin Example I3.2 were determinedby solvingfor S, in Eq. 13.4for successive time increinents.A valueof S, : 0 was [sed for the initial increment,but the value is arbitrary since only the slope and not the intercept of E_q.13.3 is desired.The 52 valuesare plotted against.average weighted discharges, XI + (l - X)O in Table 13.3. A preferablemethod would be to plot S, (rather than average) valuesagainstcorresponding valuesof instantaneous valuesof XIz + (l - X)Or, using recordedvaluesof inflow and outflow (not provided).

TABLE 13,3
l.ll"

Date 3- 16 t7 18 t9 20 21 22 23
z+

2 (cfs) 5,870 9,310 12,900 20,500 21,000 23,400 32,500 s5,400 62,700 52,600 43,200 25,200 22,800 41,200 s0,400 45,300 38,800 2?,000 16,200 12,400 10,200 8,080 6,010 5,050

u =

Oj+02

2 (cfs) 4,180 6,970 7,560 14,200 18,300 18,500 21,300 29,300 39,700 48,700 53,300 48,700 37,r00 35,800 35,800 35,800 42,700 44,lOO 35,400 25,200 t6,400 11,500 9,380 7,860

sz
(1O3cfs-days)

Weighted discharge (cfs) X+(1 -X)O

X: 0.1 4,350b 7,200 8,090 14,800 18,600 19,000 22,400 31,900 42,000 49,100 52,300 46,400 35,700 36,300 37,300 36.800 42,300 42,400 33,500 23,900 15,800 1r,200 | 9;040 7,300

X : 0.2 4,520 't,440 8,630 15,500 18,800 19,500 23,500 34,500 44,300 49,500 51,300 44,000 34,200 36,900 38,700 37,700 41,900 40,800 31,600 22,600 15,200 10,800 8,710 7,300

X: 0.3 4,690 7,670 9,160 16,100 19,100 20,000 24,700 37,100 46,600 50,000 50,300 41,700 32,800 37,400 40,200 38,600 41,500 39,000 29,600 2r,400 14,500 10,500 8,370 7,020

1.7 4.0 9.4 15.7 18.4

z5-J

25 3-26 27 28 29 30 3l 4-r 2
J

34.5 60.6 83.6 97.5 87.4 73.9

Jv.o

4 5 6 7 8

65.0 79.6 r 89. 85.2 68.0 48.9 36.1 29.9 26.5 23.1 20.3

"Note: ,S2 = ,Sr* f Ar - d Al fseeEq. 13.4]. ' E x a m p l e :4 3 5 0 : 0 . 1 ( 5 8 7 0 )+ ( 1 - 0 . 1 X 4 1 8 0 ) .

13.1 HYDROLOGIC RIVERROUTING

239.,

EXAMPLE 13.2 Given inflow and outflow hydrographson the Muckwamp River, determine K and X for the river reach.(SeeTable 13.3.)

o po

E e .E b0 o

ro X
I

g t\ X

,a o x 5 A

80.000 cfs-davs Storage, S

Solution. Selecting the narrowest loop gives X : 0.3; K : 80,000 cfsdays/40,000cfs : 2.0 days. These values could now be used to route other floods through the reachas in Example 13.1. Inherent in this procedureis the postulatethat the water surfacein the profile betweenupstreamand downstream reachis a uniform unbrokensurface ends of the section.Additionally, it is presupposed that K andX are constant throughoutthe range of flows. If significantdeparturesfrom theserestrictions are present,it may be necessary to work with shorterreaches of the river or to employ a more sophisticated approach. I I it is desirableto solve for a Muskingum Crest Segment Routing Sometirnes single outflow rate or route only a portion of an inflow hydrographby the Muskingum method(e.g.,the crest segment when only the peak outflow is desired).This is easily accomplishedby successively numbering the inflow rates as 11, 12, 13,. . . , In, In+t,. .. , and rewritingEq. 13.6as
O,: CsI, + Ctln-t + C2O^-l

( 1 3 .1 1)

where Onis the outflow rate at any time n. The outflow O,-1 is next eliminated from Eq. 13.11by makingthe substitution
On-t : Coln-t + ClIn-2 * C2O,-2

(r3.r2)

By repeatedsubstitutions for the right-sideoutflow tetm On-2,On-2,. . . can eachbe eliminated and On can be expressed as a function only of the flrst n inflow rates or, finally,
On : where Kr: Kz= K3: K,: K l I n + K 2 I n - 1 + K 3 I n - 2+ ' .' + K,I,

(13.13)

Co CoC2+ C1 K2C2 K,-rC.rfori>.2

240

CHAPTER13

HYDROGRAPH ROUTING

Using data from Example 13.1 to find the outflow rate on 3-26, we obtain Kr : C6: 0.0477 Kz: CoC2 + Cr : 0.0477(0.524) + 0.428:0.4530 : 0.2374 Kz : KzCz: 0.4530(0.524) Krt: KrcG : 0.0013

Thus the outflow on 3-26 is calculated as O11: 0.0477 X (47,g90)+ 0.4530 X : 51,469cfs. (57,740)+ . . . + 0.0013(4260)

SCS ConvexMethod
The U.S. Soil ConservationService (SCS) developeda coefficient channelrouting technique,similar to the Muskingum method, in their National EngineeringHanibook.3It has had widespreadapplication in planning and,designand can be used successfully evenwhenlimited storage datafor the reachare available. Until 1983,the procedurewas usedfor all streamflowhydrographrouting in TR-20, the SCS storm event simulation computer program describedin Chapter 24. Newer versions of TR-20 usethe att-kin methoddescribed in Section13.3. Analysis of Fig. 13.2 producesthe working equation for the convex routing method.Because the areasunderboth curvesare equal,and because the peak outflow

Figure 13.2 Geometricrelationsusedin the scs convexrouting method.(After U.S. Soil Conservation Service.3)

RIVERROUTING 13.1 HYDROLOGIC

241

is lessthan (and occurslater than) the peak inflow, the curvescrossat somepoint A, resultingin the fact that the value Orwill alwaysfall betweenIl andOt. At any time, the vertical distanceof 02 aboveOr (or below 01 on the right of A) is a fraction C, of the differenceIt - 01 as shown in the inset of Fig. 13.2. By proportionatevertical distances
Oz: - O) Ot + C,(11

(r3.r4)

This could be usedto route the entire inflow hydrographif C, could be established. FromEq. 13.14,

-'-

or- o, 1,.-o,

(13.1s)

Al is one limb of the triangle in the inset to Fig. 13.2, Because Lt _O"- O, Ir-O, K (13.16)

where the constantK is the horizontal time from O, to the interseptionof the line passingthrough Ol and 02. Thus C, is a function of both A/ and K, or
n r_- L w vt

(r3.r7)

routing method. is left to the Determination of K and C, Proof that K from Fig. 13.2is a constant by the K reader.It is a storageparameterwith time units and can be approximated twice the MuskingumX. from the Muskingummethod.Similarly, C, is approximately from Table 13.1) will provide The reach length divided by wavevelocity (estimated of reach travel time can be used. another estimateof K, or actual measurements by the SCS,in the Equation 13.l'l canthenbe solvedfor C,. The valuerecommended is of other estimates, absence C,:

(13.18)

steadydischarge,andV * 1.7 approxiwhere Vis the velocity for a representative travelingthrough the reach.The units wave (speed) of a kinematic matesthe celerity (fps). per second of V in Eq. 13.18are feet The after the C, valueis estimated. Routingby Eq. 13.14is easilyaccomplished to hydrograph peak inflow of the to the time less of interval Ar shouldbe one-fifth or As hydrograph. the to deflne rates outflow assurea sufficient number of calculated so that one point falls with all routing methods,the time interval shouldbe selected change. rapid of locations peak and other at or near the

242

13 CHAPTER

ROUTING HYDROGRAPH

the convexmethodequationfor O, is independent Unlike otherrouting methods, used to forecast outflow from a reach without be of 1r. Thus the proceduie can provides a method for early calculation and This tnowing the concurrent inflow. to warning be linked through microprocessors can warninf for floods.Flow recorders routing-time full least one at potentials flood systemsthat calculate downstream interval aheadof the flood. to find the inflow hydrographfor a given outflow The procedurecan be reversed masscurve of inflow to the reachinsteadof cumulative hydrograph,or it can route a itself. the hydrograPh

Method Muskingum-Cunge
the limitations of the Muskingummethodhavelot been Severalattemptsto overcome totally ,u"""riful becauseof computationalcomplexity or difficulties in physically The Muskingum parametersare best derived interpretingthe routing parameteri.a's not easilyrelatedto channelcharacteristics. are and fiom streamflowmeasurements diffusion wave method (seeSection 13.3) the of blendedthe accuracy Cunge6 resulting in one of the most recommethod, with the simplicity of the Muskingum as a hydrologicmethod,yet it gives is classified It for generaluse. mendedtechniques methods' resultscomparablewith hydraulic Cungeshowedthat ihe finite-differenceform of the Muskingum equationbefor both methodsare appropricomesthe diffusion waveequationif the parameters equationis Muskingum the I3'3, atelyrelated'From Eqs. 13.1 and

K+ln+(l-x)ol:t-o
dt-

(1 3. 1e)

obtain SubstitutingQ,for I andQi*rfot O, andrewritingin finite-differenceform' we

- xQI- (1 - x)Qi*,] + (1 - x)Q"+l fir"a':' - Oi+i : + Qi- Q',*') t(Q',*'

form of If K is set equalto Lxfc,Eq. 13.20is alsothe finite-difference

(13'20)
(t3.2r)

u# * ' H :o

which is calledthe kinematicwave equation(seeEq. 13.59)and can be derivedby The variable Ax combining the continuity and -oln"ntu- (or friction) equations.T denotesunincrerrr"ntofdistancealongthestreamaxisandcisthewavespeed' The equationto be usedfor routing is obtainedfrom Eq . 13.20by solvingfor the unknown flow rate,

+t Q i * c r Q i * , OiIi: coQ',*'c
where

(r3.22)

co

Lt/K - 2X 2(r-x)+Lt/K

(r3.23)

RIVERROUTING 13.1 HYDROLOGIC

243

cr:
4 v2 -

LtlK + 2x
2(r-x)+Lt/K 2(I-x)-cLt/Lx
" '

(r3.24) (r3.2s)

2(I-x)+Lt/K
./r .-\ L Aal?

the time for a waveto travelthe routing reachlength K : A,xfc,it represents Because that the velocity c is the celerity of a kinematic shows t. Conge at velocity Ax, moving (Table 13'1). described previously wave translation When X : 0.5 and c Lt/A,x: 1.0, the routing equationproduces attenuation : or translation no (zero length), reach 0 Ax When without attenuation. occurs. by the routing parameterc canbe extracted If previousflood data are available, obtained also be can parameters the of Estimates calculations. reversingthe routing from flow and channelmeasurements. The value of X for use in Cunge'sformulation is

x:t('-#t)
c: mV

(r3.26)

where So : channelbottom slope (dimensionless) for the peak rate per unit width (cfs/fO,normally determined 4o : discharge velocity V by The value of celerity c can be estimatedas a function of the average Q3.27) The velocity QfA, andm is aboutI for wide natural channels. where v is the average equation flow the uniform from m comes coefficient Q: bA-

(13.28)

by taking partial derivatives,to which reduces,

o Q:
A

*g:
A

*y

(r3.2e)

Substitutingthis into the continuity equation

Q*4:o
6x At

(13.30)

givesEq. I3.2t if c : mv.If dischargedata ate available,m canbe estimatedfrom are given in Table 13'1' nq. tZig. Valuesfor commonshapechannels The rout{rg can now be done using either constantm and c parameters(i.e., using a single ai'eragevelocity) or variableparameters(usingeachnew velocity v). equition I-3.2i rs solvedfor c, the valueX is derivedfrom Eq. 13.26,andE6' 13.23 to 13.25are solvedusing K : Lx/c. to assure When using this rnethod,the valuesof Ax and Ar shouldbe selected peak of inflow to time detailsare proper$ routed.Nominally' the that the flood wa-ve resoluspatial and is broken into 5 or 10 time incrementsAt. To give both temporal and length, of Ax tion, the total reachlength l, can be dividedinto severalincrements outflow from eachis treated as inflow to the next.

244

CHAPTER13

HYDROGRAPH ROUTING

EXAMPLE 13.3 Usethe Muskingum-Cungemethodto route the hydrographfrom Example13.2.Use areaat Q : 59,960is 5996 ft2, So: 0.0001,Lx : 545 mi, flow cross-sectional : : 1.0 day (asin Example13.2). 59,960is 60 ft, and Ar width at Q Solution. From the inflow, the peak rate of 59,960cfs gives

So:T

o^

59.960

Qo - 59,960: ,, V r-: T fps c : ZVo: 16.7 From Eq. 13.26

l0fps

":;1,
L* K :

1000 :0.4 (16.7)s4s (s280) 0.0001 L

I I

't

7:

545(5280) ff: 172,8 s0 e0 c C": -0.1765 Ct: 0'7647 Cz: 0'294I

and

The routing for a portion of the hydrographis as follows:

Date, t

CoQ'*L*

Ct Qti*ro*

Cz Qlutno*

Qli*.*
r9r0 (3-r7) 4440 6900 11,010 16,050 16,050 16p20 29,580 44,360 48,750 44,880 35,730 (3-29) 20,9s0

3-16 17 18 t9 20 2l 22
L3

24 25 26 27 3-28

- 1350 -r970 -2950 -3810 -3700 -4690 -8120 - 10,580 10,190 -8450 -6080 -3820 -6120

3260 5850 8540 12,790 16,510 16,020 20,320 35,180 45,850 44,150 36,620, 26,3sO 16,560

0 560 1310 2030 3240 4720 4720 4980 8700 13,050 14,340 13,200 10,510

Note that the peak outflow of 48,750 cfs on March 26 occurson the samedate as in Example I3.2 but has experiencedslightly greaterattenuationfrom the Muskingum-Cunge example. The value C, is alwayspositive,and negativevaluesof C, arenot particuin this example, this condition should larly troublesome. Although C0is negative valuesof Coare avoided be avoided in practice.As seenfrom Eq. 13.23, negative

ROUTING 245 RESERVOIR 13.2 HYDROLOGIC

when

^|,
Other Methods

,* rr

(13.31)

includingthe working havebeendeveloped, river routing procedures Other hydrolSgic method. method,Tatummethod,and multiple storage R&D method,straddle-stagger They all appear as options in HEC-I, the U.S. Army Corps of Engineer's event in Chapter24. simulation and routing model described

ROUTING RESERVOIR 13.2 HYDROLOGIC

The storageindication method of routing a hydrographthrough a reservoiris also reservoiris through a storage A flood wavepassing calledthe modified Pulsmethod.8 both delayedand attenuatedas it enters and spreadsover the pool surface.Water storedin the reservoiris gradually releasedas pipe flow through turbines or outlet spillway. or in extremefloods, over an emergency works, calledprincipal sprllways, from Flow over an ungatedemergencyspillway weir sectioncan be described by the form energy,momentum,and continuity considerations O: CYH'

(r3.32)

where O : the outflow rate (cfs) Y : the length of the spillway crest (ft)
H : deepest reservoirdepth abovethe spillway crest (ft) coefficientfor the weir or section,theoretically 3'0 C : the discharge x : exponent, theoretically J pipe is similarly described by Eq. 13.32 Flow through a free outlet discharge where' I H C .r : : : : pipe (ft') the cross-sectionalareaof the discharge head above the free outlet elevation (ft) coefficient, theoretically\/29 the pipe discharge exponent,theoreticallyj

Flow equations for other outlet conditions are availablein hydraulics textbooks. from in a reservoirare readily determined Storagevaluesfor variouspool elevations from topopool measured areas of volumesconfinedbetweenvarious computations graphic maps. Since storageand outflow both dependonly on pool elevation,the relation (Eq. 13.32) can curve and the outflow-elevation resulting storage-elivation graph. Storagein a reservoirdepends easily be combinedto form a storage-outflow on the inflow and outflow in river to the dependence only on the outflow, contrasted routing(Eq. 13.3). "surchargestorage"or the storage ,Sis often defined as the For convenience, is zero. spillwaycrest.Normally the overflowrate is zero when,S abovethe emergency line ofthe and ifthe slope relation is found to be linear, Ifthe graphedstorage-outflow

246

CHAPTER13

ROUTING HYDROGRAPH

is definedas K, then S : KO (13.33) is a and the reservoiris called a linear reservoir. Routing through a linear reservoir : 13.3' 0'0inEq. 13.l-usingr showninFig. Muskingumriverrouting specialcaseof only while the inflow exceeds rate in Fig. tg.t is increasing Note alsothat the,outfl-ow that the inflow immeassumptions with the is conJistent the outflow.This observation only outflow depends the that and overthe entirepool surface diatelygoesinto storage on this storage. by first dividing time Routing through a linear reservoiris easily accomplished : Eq. I3'4 and Ko2into s2 into a numberof equal incrementsand then substituting increment' time each for solving for oz, wtrictris the only remainingunknown through anonlinear reservoir,the storage-outflow Ro'oa To route an emergency the outflow relation and the continiity equation,Eq. L3'4,are combinedto determine as rewritten be can I3'4 and storageat the end of'eachtime inciement A/. Equation

(*-o,) r , + r n +.1

:'+

t on+t

(r3.34)

side'Pairs in which the only unknown for any time incrementis the tgt_*:n the right checked .34 and 13 Eq. satisfy that generated b" of trial valuesof S"*, andO,*1"ould procetrial this to resort than Rather confirmation. fot "uru" in the storage-outflow outflow curve are replottedas and points on the storage dure, a valrieof At is selected indication" curve shownin Fig. 13.3' This graph allowsa direct determithe ';storage a value of the ordinate Zl,*rfLt I On+rhas been nation of the outflow O,11 o11ce from Eq. f :.:4' ftre secondunknown' S,*t, can be read from the S-O calculated 13.3)orfoundfromEq'13'34' curve(whichcouldilsobeplottedonthegraphinFig.
t200 I

/ / {,
: .il< 600

0 4 0 Outflow (cfs)

O' Figure 13.3 Curveof zsl\t + O versus

RESERVOIRROUTING 19.2 HYDROLOGIC

247

using of Eq. 13.34with Fig. 13.3is illustrated integration This row-by-rownumerical : t hr in ExampleI3.4. Ar
EXAMPLE 13.4

inflow hydrographand the 2S/Lt + O curve ofFig. 13.3 Giventhe tria4gular-shaped find the outflow hydrograph for the reservoir assumingit to be completely full at the of the storm.(SeeTable 13.4.) beginning

o 90

o Time (hr)

In selecting a routing period Ar, generallyat leastfive points on the rising limb number of of the inflow hydrographare employedin the calculations.An increased since as A/ - 0 accuracy, the points on the rising limb, that is, a small A/, improves integrated, being function of the the true limit the numerical integrationapproaches in this casedSfdt. Column 3 in Table 13.4 comesfrom the given inflow hydrograph,column 4 is simply the addition of I, + In*r, andColumns5 andT are initially zero, sincein this of inflow. Therefore, problem the reservoir is assumedfull at the commencement there is no availablestorage.
TABLE 13.4 ROUTINGTABLE (1) Time (hr) 0 I z 3
+

(2)

(3)
ln

(4) I, + In+l (cfs)

-tt,

'n-o.
(5)

(6)
2Sn+1 , a -;fT vn+1

(7)
an+t

(8)
Sn+t

(cfs)

(cfs) 0 20 74 160 284 450 664 853 948 953 870 746 630

(cfs)
JU

(cfs)

(cfs-hr)

5 6 '7 8 9 t0 ll t2

1 2 3 4 5 6 7 8 9 10 ll 1 t

0 0 0 0 120, 150 180 135 9 0 45 0 2 0 3 0 3 6 9

90 150 2lo 270 330 315 225 135 0 0 0

110 224 370

))4

5 18
3Z AA

52
)6

780 979

r078 1085 998 870 746 630

63 65 65 64 62 58 54

12.5 46 96 164 250 361 458 506 510 467 404 344 288

248

ROUTING 13 HYDROGRAPH CHAPTER the sum of columns4 The starting value for n : 1 in column 6 is computedas and 5 from F,q.13.34 (L + 12)- ( * - o ' )

25" - + o2 Lt 25. 3 0 * 0 : - - + o2

Lt

column 6 gives a valuefor Enteringthe ordinateof Fig. 13.3 with the value 30 from end-of-time-interval o, of 5 cfs, which is recordedin column 7 , The corresponding 8. Moving to column in recorded and 6 and7 columns s,, is calculatedfrom storage, n be found fof : 2 usings' the secondfow, a value of the term in column 5 can now and O, from columns7 and 8' ThestepwiseprocedureusedtogetoutflowfiguresforallncanbeSumma. rized as l.Entriesincolumnsland3areknownfromthegiveninflow.hydrograph. column 3' - 2. Entries in column 4 arc the additionsof I' t I'*' in 3.Theinitialvalueofthetermincolumn5iszero,thoughitcouldalsobe 4 and 5 are based on any arbitrary starting storagevalue' and columns addedto producethe value in column 6' 4.The2SlA,t+oversusoplotisenteredwithknownvaluesotzSlLt+o to find valuesof O for column 7' 5.Columns6andlaresolvedforS,+r,whichisrecordedincolumn8. for column 5 usingthe 6. Advanceto the next row and calculatethe next value 7 and 8' valuesin the precedingrow for O and S {om columns sum in column 4 and enterthe 7. Add the value in cotuin 5 to the advanced result in column 6 for the new period under consideration' 8.ThenewoutflowforcolumnTisagainfoundfromtherelationof ZSlLt + O as in Fig. 13.3. g.Thecorrespondingnewstorageincolumn8isfoundbysolvingfrom columns6 andT' l.0.Steps6throughgarerepeateduntiltheentireoutflowhydrographisgenerated. rl

ROUTING RIVER 13.3 HYDRAULIC

ploys both the equationof continuity and utions to the complete hydrarllic routing

ffi::: ili"'nTil::ll;lffi ffffii.;fi

tions. Hydraulicfoutingtechniquesarehelpfulinsolvingriverroutingproblems,oversolution the simultaneous land flow, or sheetnow. uyaraulic routing proceedsfrom for combination the of forms general 1'ne of continoiti una-o*"ntuir, of expression, fl'ow equations' rivers are called the spatically varied unsteady

13.3 HYDRAULIC RIVER ROUTING 249 Theseequationsalso apply to sheetflow or overland flow and include terms for laterally incoming rainfall. They can be simplified and used to resolveriver routing problems.e For completeneis of presentation, a generalform of the spatially varied unsteady flow equations will be presented first.

Equationof Continuity
The equation of continuity statesthat inflow minus outflow equals the changein storage. To relatethis conceptto a river sectionundera condition ofrainfall or lateral inflow, consideran elementof length Ax and unit width into the page as shown in Fig. 13.4. The total inflow is

,,*. r)dtdx (13 35) ,(, - X+)(, - * *) * * , [.' l,'*o'

The total outflow is

+t*+)(,.#?) o' o(v

The changein storageis

(13.36)

* ofrl,"
l(x, t)

(r3.37)

,(,#+)(,l

o(v*o ?t) ( r . * + )
f
A x L (a) x

2--*l-- 2-

i(x, t)

ft) Figure 13.4 Continuity and momentum elements(where p : the density of water, V : the averagevelocity. 1l : the depth. i : the lateral inflow per elemental Ax, and S = the slopeof the river bottom).

250

CHAPTER13

HYDROGRAPH ROUTING

Consequently, continuity gives

-o(r{*tx + vfia') a, + pl A,xL, -

oX L,xL,t: o

(13.38)

where i is the average lateral inflow resulting from rainfall over Ax and Ar. The continuity equation of unsteadyflow with lateral inflow is obtained by simplifyingEq. 13.38 6V .,0y , 0y v6 -x -|v--r " 6x At :

( 13 .3e)

For otherthan a unit width, Eq. 13.39takesthe form

e d{x + v d Y * *d tx
whereA is the width times the depth,y.

l : 0

(13.40)

Momentum Equation
In accordance of momentumper unit with Newton's second law of motion, the change of time on a body is equal to the resultantof all externalforces acting on the body. The following derivationof the momentumequationof spatiallyvariedunsteadyflow (1) the flow is unidirectional and is presentedsubjectto the following assumptions: is hydrostatic;(3) the slope velocity uniform acrossthe flow section;(2) the pressure of the river bottom is relatively small; (4) the Manning formula may be used to evaluate the friction lossdueto shearat the channelwall; (5) lateral inflow entersthe stream with no velocity componentin the direction of flow; and (6) the value of I represents the spatial and time variationsof lateral inflow. Forcesactingon an elementof lengthAx andunit width are shownin Fig. i 3.4b. The forcesF1andF, represent as hydrostaticforceson the elementand are expressed

axl atvA) - ylyA f-, - -;zl F,: - - lL,!^ .f-^ - a(la)Axl rz: A- Z l

(13.41)

(r3.42)

wherey is the distancefrom the watersurface to the centroidofthe area.The resultant hydrostaticforce is F, - Fr, or
\ 1

r ,6 :

-ra(!A) *
x

(13.43)

By assuminga small slope for the river bottom, the gravitational force is given by Fr : yASA.x (13.44)

The frictional force alongthe bottom is equalto the friction slope$multiplied by the weisht of water in an elementAx. \: vASyLx

( 13 .4s )

13.3 HYDRAULICRIVER ROUTING

251

as The rate of changeof momentumin the length Ax may be expressed d(mv)

dt
: iii____ T

dV
dt

,.dm
V____

(r3.46)

dt

in which m is the massof fluid. that the incoming lateral inflow entersthe moving fluid with no If it is assumed velocity componentin the direction of flow and I representsthe spatial and time variationsof the lateral inflow, the rate of changeof momentumfor the elementcan as be expressed

dv dtuv) T:pALxi+pvi
wherc dV/dt represents

Lx

(r3.47)

dv av ,,av
dt at
T v -

Ex

(13.48)

The rate of changeof momentumis therefore

oe*(Ya,. "#) + pviL,x

(r3.4e)

Equating the rate of changeof momentumto all externalforces acting on the elementresults in

{*r{+gaFA)
A t O x A O

+L:s(s-s)
x A

(13.50)

Now for a unit width element,the relation simplifiesto

- s) : o {u, * r{u** tX+Yri : s{s

(13. s) 1

that can be expressions Equations13.50and 13.51form a set of simultaneouS solvedfor V and y subjectto the appropriateboundary conditions.

and DynamicWaves Diffusionn Kinematic,

For the casewith zerclaterclinflow, l, Eq. 13.51can be solvedfor the friction slope
c _ c _ :

Ev

VAV

IAV

(r3.s2)

Friction slope

Bed slope

Kinematic wave Diffusion wave

Water surface slope

Convective acceleration

Temporai accelelation

dynamic wave Fu11

252

ROUTING CHAPTER13 HYDROGRAPH

are classifiedas flow routing in open channels of unsteady The three typesof analysis depending kinematic,diffusion (also called noninertia), anddynamicwaveanalyses, not only by differ on which terms in F;q. 13.52 are retained. The three techniques flow regarding including different terms of Eq. 13.52 but also in the assumptions assumpthese 13.5 shows conditions for satisfyingthe momentum equation.Table tions.
HYDRAULIC IN VARIOUS USED 13.5 ASSUMPTIONS TABLE METHODS ROUTING
Method Kinematic Diffusion Full dynamic Commonflow condition
Steady Steady Unsteady

profile Watersurface
Uniform Nonuniform Nonuniform

Steadyflow is definedas flow that doesnot changewith time, and uniform flow is flow with a water surfaceparalleling the bed slope.For steadyuniform flow, the loops. Steady curve)is a singlecurve without hysteresis rating curve (stage-discharge but varying water surfaceslopesuchas that nonuniform flow has constantdischarge found at the entranceto a reservoiror at the approachof a waterfall. One way of selectingthe applicablemethod is to examinethe rating curve and whether it is the same for rising and falling stages.The choice of routing assess equation dependson whether the difference is small (kinematic), relatively large (dynamic),or somewhere in-between(diffusion). that the inertia terms of Eq. 13.52 ate The kinematic wave method assumes negligibleand that the friction slopeequalsthe bed slopeS. Momentumconservation by by assumingsteadyuniform flow, and routing is accomplished is approxi(nated combiningthe continuity equationwith any form of friction lossequation.Typically, either the Manning equationor Ch6zyequationis used.The Chdzyequationis
V : C\/RS

( 13.s 3)

and the Manning equationis

V : l'486 n O2/3St/2

( 13.s 4)

where C andn arefriction coefficients,S is the friction slope,and R is the hydraulic if area radius (areadividedby wettedperimeter).Both give velocity in feet per second are input using squarefeet and feet units. and wetted qerimeter *of these equations can be substitutedinto the kinematic portion of Either Eq. 13.52,equatingslopeof energygradeline with bed slopeto accountfor momentum. The continuity equation,Eq. 13.39,for this casereducesto dQ *64:o At Ex The Manning or Ch6zyequationhas the form Q: b'4^

( 13.5s )

(13.56)

ROUTING 253 RIVER 13.3 HYDRAULIC

is which, after taking derivatives, 6Q = 6*4*-t AA Multiplying Eq. 13.55by dQ/dA gives
: m--a : M V A

(r3.57)

, ,a Q mv -dx

+9 :o at

( 13.s 8)

or. if c - mV,the kinematic routing equationis

. Q -d x * 9 : o
At

(13.se)

" -:dQ:ldQ dA BdY

(13.60)

to the slope of the where B is the top width of the channel.Thus celerity is related c is constantand that assume Most applications rating curve,which varieswith stage.

9 * , 9 : dd{x9 dx At

(13.61)

accounts for the The left-hand side is the kinematic wave equation and the right diffusion d is diffusion effect of nonuniform water surfaceproflles. The hydraulic given by " 2 5
d : -

(r3.62)

This term reveals where4 is the flow per unit width of channel,and s is the bed slope. in small d) (resulting is valid whenbed slopesare steep why kinematiowaveanalysis the slopes, bed flat or when the channells extremelywide (resultingin small 4)' For hydraulic diffusion coefflcientis particularly important' Numerical solutionsof Eq. 13.6I ate presentedby severalinvestigators'12-ra Thesenormally involve substitutionof the relation

ao
A--

v - B3 +
dt

(13.63)

254

CHAPTER13

HYDROGRAPH ROUTING

into Eq. 13.61,whereB is the channeltop width. The Manning or Ch6zyequationis usedfor the friction slope,where

tr:#

(r3.64)

K is Q/\/Srfromeither equation.Diffusion wavesapply to in which the conveyance problems than kinematic formulations, but their use may not be a wider range of it requiresabout the sameeffort as dynamic routing. warrantedbecause for all terms inBq.13.52, includingthe accounts The third type of waveanalysis "dynamic" or It is referredto as the nonuniform,unsteady, and inertia components. "full dynamic" formulation. Dynamic wave solutionsare far more complicatedbut for analysisof flow along very flat slopes,flow into large reserare often necessary voirs, highly unsteadydam-breakflood waves,or reversing(e.g.,tidal) flows. These on coastalplains. As a generalrule, full dynamic conditions are often encountered when wave analysisbecomesnecessary

s>
where

S - bed slope T o : time to peak (sec) of the inflow hydrograph D : averageflow depth (ft) o : eravitationalacceleration
d

1 5 F 4\;

( 13 .6s )

SCSAtt-Kin TR-20Method
method(Section13.1)with themodifiedatt-kin theconvex In 1983,the SCSreplaced (attematron-kinematic)methodas the agency'spreferredchannelrouting method.ls The 1.964SCS TR-20 (Chapter 24) single-eventsimulation model used the convex modified to route by the att-kin method. method but was subsequently indication and kinematicwavemethods. The procedureis a blend of the storage Figure 13.5 showsthe two-step processof simulating attenuationfirst by meansof

o PP

Flgure 13.5 Routing principlesusedin the SCS att-kin method,

RIVER ROUTING 255 r3.3 HYDRAULIC routing and then translatingthe wavein time by the kinematic wavemethod storage in magnitudebut also to accountfor the fact that routed flow ratesnot only decrease require time to traversethe length of the routing reach.The storagerouting portion providesattenuationwith instantaneous translation,and the kinematic waverouting providestranslationand distortionbut doesnot attenuate the peak. Both are needed to produce the desired effect. The previously rhentioned full dynamic equations simultaneously accountfor both effectsbut are difficult to solve. The samevolume V1of waterflowing Figure 13.5helpsto visualizethe process. reservoirduring into the reachduring time /, would flow out of a hypotheticalstorage by kinematic interval 12.This samevolume would translateand distort downstream actibn, flowing out of the reachduring interval /r. and selectionof routing coefficients,the Through its theoretical development att-kin methodequationssatisfythe physicalpropagationand timing of the peak flow (areasunder the three hydrographs of rate first. Conservation of massis also assured Fig. 13.5are equal). thentranslates through storage, routesthe inflow hydrograph The actualprocess the peak flow rate, without attenuation,to its final time location in the outflow equal to that correhydrograph. The location in time of the peak outflow is assumed of the flood. spondingto the maximum storagein the reachduring passage Becausecelerity changeswith storage,the other flows of the storage-routed hydrographare translated,pachby a different celerity,to their respectivefinal times and values. by substitutionof the relation The storageindication routing is accomplished Q: KS^

(13.66)

into the continuity equation, Eq. 13.30, where S is the storageand K and m are coefficients.Kinematic routing solvesthe unsteadyflow equation(13.59) with Q: bA*

(r3.61)

area.If Z is the length whereb andm are input coefficients,and A is cross-sectional areathroughoutL is relativelyconstant,the ofrouiing reach,andifthe cross-sectional storageis given by S: LA (13.68)

Theseequatiohsare combined in an iterative fashion to assurethat the peak flow routing, resultingfrom the kinematic routing equalsthe peak resulting from storage and simultaneously ensuringthat the time of the peak outflow occurs at the time of maximum storagein the reach,or Qo: KS;

(13.69)

of b andm for Input to the methodrequiresselectionof a reachlength and estimates shownto be 13.27, m canbe 13.1 and Eq. for Table usein Eq. 13.59.As discussed celerity, or (under wave conditions) with bankfull velocity a factor relating average

(13.70)

256

CHAPTER13

ROUTING HYDROGRAPH

the shorterthe travel time, A value of m I 1.0 would incorThe larger ru becomes, by SCSresulted flow velocity.Studies rectly makethe celerity slowerthan the average for m values resulted errors Signif,cant in a recommendationof ! for general use. a single having sections cross for greater than2.0. Equation 13.67 is appropriate but to evaluate, diffcult more are sections ihannel with regulaishape.Complexcross stream'rs the for table la valuescan be developid from a rating attenuationof the peak flow increasesdue to As the coefficient b decreases, by in the reach.The value b canbe estimated storage reducedvelocity and increased (referlo 13.67 Eq. of form linear the fitting plotting Q andA on log-logpaper and faatei6.+1. The slope*oUa be m andtheinterceptut 4_: 1 would be b. The SCS for estimatingb and m'" nomographs has also developed As a generalguideline,the reach length L shouldbe increasedto a value that time increment,or than the selected resultsin a kinematic wavetravel time c greater
Lo> cA,t
Z*io ft LR ft 18,000

mV = wave celerity
l*ln LR

9,000 7,500 15,000

6,000 5,000

= minimum acceptablereach length = minimum recommended reach length MainTime Increment Hours

12,000 10,000 7,000 5,000 4,000 3,000 2,000 1,500 1.000

mV ft/sec

0.5 0.7

3,500 r <nn ? nnn

2
I

1,500 1,000 750

0.5
n?5

0.3
0.2'

3 /
5 7 10 T2 15

0.1 0.05

350 700 250 500

Example: mV = 4ftlsec Main time increment = 0.2 hr L*n= 1,450ft La = 2,900 ft

t00 for le ngth reach Figure 13.6 SCS nomographfor determining Service, Conservation (After Soil U.S. atl-kin method of routing. "ComputerProgramfor Project Formulation," TechnicalRelease AppendixG, 1983.) 20, Revised,

PROBLEMS 257 The length' c : mV' and Ar is the time increment' where Lo is the recommended minimumrecommendedLoisthatgivingawavetraveltimeequaltoabouthalfthe lengthy iray iesult in analytical difficulty when rime incremenr. This 1j[$;"1* inflowhydrographsorsteepStreams.areencountered.Italsoresultsinthepeak were time incrementAr. If severalreaches outflow time being ,ourJ"J"rp io the full between length reach a accumulate'Thus routed, this incrementattime^errorwould greater thal c. A/ is recommended' length a but c Lt andc Lt/2 is acceptable' Figure13.6providesth"rangeofminimumacc"eptableandminimumrecommended routing reachlengths.

ff ;'il;;'n"'uu"'i11,, "u.n ;t'1 ":".1-':^:"::::T,it?ii: ffi ff Ji,ffiffi :3:[";ffi;i;;;ffi;"'"-*:i"o1j.u""]'mJ1't^:-:T:1T':1"::i"%i"$1 presented , i. wru,iil

iir"T-l"ii,'.i,"n" ;;lrt':?
HX"J"H;il;il#il;;;;;,;;i;#

hvdraulicroutingtec|1ioue11p.p"11'i^'"Tt*::?:Yi:: ff ffi ff;;iicationsof

1"*Tl student ll :^11ti:: iil::1,:: understand can aninteresred that

of hydrologicmodeling' the structuringprocesses

Summary
ComputersoftwareforhydrographsynthesiS^androutingisavailablefromnumerous federal agelcy routines are detailed in public and private vendors. wid;ry used includesone or amflow processes Chapter24. Virtually every ( m o r e h y d r o g r a p h r o a s H E C - I ( C h a hydroloeig p t e r 2 4a1d ) , hydraulic p r o v i d e lplied severalchoices.In I the models can be found in the routing procedures' g]]:i,:Ytditg suggestions .T" At sel 1iterarure.,6,17 ln rlver or engage will The readerwho when eachof the methodsshouldbe applied' reservoirroutingisencouragedtodwelopthecomparisonrequestedinProblem13.25 before leavingthis chaPter'

PROBLEMS
techniques' betweenhydrologic and hydraulic routing 13.1. Discussthe main diff-erences 13.2.

equation' riverrouting TheMuskingum ?:,:,':'?,*

t+}i ;q":: Iit:t?;;i,Y :..ii+l ;; q :il# ffi?#t 1iT .'SX,ffff r" 191 :, -Yllli "'*ifi:'i I ilY?iBi'"".U'1-'"d;i"?';:fli!11iy":;T"':'.."t"jfl i:;#"J I"a'^#i'.h;r::;;'i;A--)-r",t'i'i""1'"'lT1,1l.o:'^"iui:.'* i:il3iil;
12' v2' .xg 9z v\v 4-_ *u,'::9,":1i: i'||w p:.'i"l' oerro.; time the the tlme or of Siif#ffi"#ru:=#;"i"e-i*ine outflow, and storage at the beginning lT"jll ;;:E;: n"t*#:J:'::3*4":::i3";ffi';

u,,t'l uutu", corresponding G :fd;#?;##ir*"p"'i"a'lndAsistl"*"1c-",'istorage'Perrorm "and ffi:Tt"'ff for Cs' C1' and C2'
A",inution the described verify ttre equations

l"3.3.IftheMuslongumKvalueis12hrforareachofariver,andiftheXvalueis0.2'what value of Ar for routing purposes? would be a reasonable

258

ROUTING, 13 HYDROGRAPH CHAPTER 13.4. A river reachhas a storagerelation given by S; : ali + boi. Derive a routing equation for O2 analogousto the Muskingum equation (13.6). Give equationsfor the coefficientsof 11,01, and 12. 13.5. List the steps(starting with a measuredinflow and outflow hydrographfor a river reach)necessaryto determinethe Muskingum K and X values. If the inflow and outflow are recordedin cubic feet per second,statethe units that would result for K and X if your list of stepsis followed. 13.6. Given the following inflow hydrograph:

'

lnflow (cfs) 6 e.v. Noon 6 p.tu. Midnight 6 A.M. Noon 6 p.Iu. Midnight

Outflow (cfs) 100

100 300 680 500 400 310 230 100

Assumethat the outflow hydrographat a section 3-mi downstreamis desired. : a. compute the outflow hydrographby the Muskingum methodusing valuesof K : 0.13. 11 hr andX b. Plot the inflow and outflow hydrographs on a single graph' c. Repeatsteps(a) and (b) usingX : 0.00. at both endsof.a30-mi river reach: discharges I3,7. Giventhe following valuesof measured a. Determinethe Muskingum K and X valuesfor this reach. b. Holding K constant (at your determined value), use the given inflow hydrograph to determineand plot three outflow hydrographsfor valuesofX equal to the computed value.0.5. and 0.0. Plot the actualoutflow and nurnericallycomparethe root mean squareof residualswhen each of the three calculatedhydrographsis compared outflow. with the measured

Time
6 e.Ira. Noon 6 p.v, Midftight 6 l.rra. Noon 6 p.tvt. Midnight 6 n.n. Noon 6 p.v. Midnight 6 A.M.

lnflow (cfs)

Outflow (cfs) 10 12.9 26.5 43.1 44.9 41.3 35.3 27.7 t9.4 1 5I. 12.7 11 . 5 10.8

l0 30 68 50 40 3l
ZJ

10 10 10 10 t0 10

PROBLEMS 259 regionthat hasrunoffrecords' 13.8. Selecta streamin your geographic X' method of routing to find K and Usethe Muskingum

l3.g.PrecipitationbeganatnoononJune14andcausedafloodhydrographinastredrn.As thehydrographpassed,thefollowingmeasuredstreamflowdataatcrosssectionsA and B were obtained:

Inflow, SectionA (cfs)

Time June14-17
6 e.v. Noon 6 p.tvt. Midnight 6 e.u. Noon 6 p.u. Midnight 6 n.u. Noon 6 p.tvt. Midnight 6 e.u. Noon

Outflow, B Section (cfs) 10 10 t3 26 43 4l 35 28 19

IJ
I J

10 10 30 10 50 40 30 20 10 l0 l0 10 10 10

1l 10

for the a. Determine the Muskingum K and X values b. DeterminethehydrographatSectionBifadifferentstormproducedthefollowing hydrographat SectionA:
lnflow (cfs)

river reach'

Time 6 e.u. Noon

6 p.tt. Midnight 6 l.rra.

Time
Noon 6 p.tvt. Midnlght 6 e.u. Noon

lnflow (cfs) 400 300 200 100 100

100 100 200 500 600

emergencyspillway of a certain 13.10. The outflow rate (cts) and storage(cfs-hr) for an units of hours'Use

: Sl3,wherethe number3 has reservoirare linearly t"fui"a Uy d - or Ltl2to determine + 02 Ltlz = I tt + sr s2 equations continuity the this and event: inflow following the itr" p""f. outflow rad fr;m the reservoir for Time (h0 0
L

I (cfs)

o
(cfs)

(cfs-hr)

4 6 8

0 400 600 200 0

260

CHAPTER13 HYDROGRAPH ROUTING ' 13.11. A simple reservoir has a linear storage-indication curve defined by the equation

" : 2* .
whereAr is equalto 10 hr. If s at 8 e.u. is 0 cfs-hr,usethe continuity equationto route the following hydrograph through the reservoir:
Time I (cf9
8 a.lra.

9 a.u. 200

10 e.Ira. 400

1l A.M.

200

Noon 0

I P.M.

13.12. For a vertical-walled reservoir with a surface area A show how the two routing equations(73.32 and 13.34) could be written to contain only o2, sz, and known, values(computedfrom or, s,, and so on). Eliminate .FIfrom all the equations.How could thesetwo equationsbe solvedfor the two unknowns? 13.13. Given: Vertical-walled reservoir, surface area: 1000 acres; emergency.spillway width : 97.l ft (ideal spillway);H : watersurfaceelevation(ft) abovethe spillway crest; and initial inflow and outflow are both 100 cfs. a. In acre-ft and cfs-days, determinethe valuesfor reservoirstorage S corresponding to the followingvaluesof I1..0, 0.5, l, 1.5,2,3, 4 ft. b. Determine the values of the emergencyspillway Q correspondingto the depths namedin part a. c. Carefully plot and label the discharge-stcirage curve (cfs versuscfs-days)and the storage-indication curve (cfsversuscfs,Fig. 13.3) on rectangularcoordinategraph paper. d. Determinethe outflow ratesoverthe spillwayat the endsof successive dayscorrespondingto the following inflow rates(instantaneous ratesat the endsof successive days):100, 400, 1200,1500, 1100, 700, 400, 300, 200, 100, 100, 100. Use a routing table similar to the one used in Example 13.4 and continue the rotating procedureuntil the outflow drops below 10 cfs. e. Plot the inflow and outflow hydrographson a single graph. Where should these curvescross? 13.14. Routethe given inflow hydrograph through the reservoir by assumingfhe initial water level is at the emergencyspillway level (1160 ft) and that the principal spillway is plugged with debris. The reservoir has a 500-ft-wide ideal emergency spillway (C = 3.0) locatedat the 1160-ft elevation.Storage-area-elevation data are Elevation (ft) 1l{0 1120 1140 I 158 I 160 1162 1164 I 166 I 168 1180
Area of pool (ft2x 106)

Storage (ft3x 106) 0 4.25 50.25 172.15 r92.75 2r5.35 239.95 266.5s 295.20 528.55

0 0.85 3.75 9.8 10.8 I 1.8 12.8 13.8 14.85 25,0

PROBLEMS dataare The inflow hYdrograPh

I (cfs)

261

Time(h0 0.0 0.5 1.0 1.5 2.0 2.5 3.0

J.)

0 3,630 10,920 to,'720

5 010

4.0
A <

1,600 460 100 10 0

table. a.Findthel5-minunithydrographbySnyder'smethod.

063 and min 30 nrst the for ofrain 18in. l i"li:l! ff;ru#iHfllSiii.?om
.. il"i:i.,i[T;iitf;,ersus

the of 15minand period a routing using o.cuwe

curvesprovloeo' outflow and storage it is full to the bottom the reservoirassuming e. Route,n" uJ#f,iu.;;;n";;"dh elevation980' of the sPillwaY height of water^inthe reservor' maxi-mum Indicate f. plu'"d to obtain 5 ft of freeboard? "f th;;;;;; g. At what "tJ;;;;iJ'n"'op Total storage
104(ft3)

Elevation (ft)

lncremental storage 1oo(ft") 40

960 970 980 990 1000

0 40 250 840 2080

zto
590 1240

the reservoirinitially empty' 13.16. RepeatProblem 13'15 with l3,IT.AfloodhydrographistoberoutedtytheMuskingummethodthroughalO-mireach ro-ml fvl,r;ach be divided in with K = Zhr.tnto how *uny ,ubr"u"h"*-**rirt" Kl3 < Lt < Kl ;;J stitt satisry ttre ,tuuitity criterid order to or" lr ]'6.i;

262

CHAPTER 13 HYDROGRAPH ROUTING 13.18. RepeatProblem 13.6aby dividing the 3-mi reach into two subreaches with equal K valuesof 5.5 hr. Comparethe results. 13.19. Discussthe problemsassociated with the useof a reservoirrouting techniquesuchas the storage-indication method in routing a flood through a river reach. 13.20. VerifyEq. 13.51. t3.21. Precipitationbeganat noon on June14 and caused a flood hydrograph in a stream.As the storm passed,the following streamflow data at cross sectionsA and B were obtained:

Time June14-17
6 .q,.N{. Noon 6 p.u. Midnight 6 .r.lr. Noon 6 p.lr. Midnight 6 a.Ira. Noon 6 p.rra. Midnight

lnflow SectionA (cfs)

Outflow B Section (cfs) 10 l0 13 26 43

+l

6 e.u.
Noon

l0 10 30 70 50 40 30 20 10 10 l0 t0 l0 10

35 28 t9 l5 l3
ll

10

a. Determinethe Muskingum K and X valuesfor the river reach. b. Determine the hydrograph at Section B if a different storm produced the following hydrograph at SectionA (continuecomputations until outflow falls below 101 cfs):

Time 6 a.u.
Noon 6 p.Ira. Midnight 6 a.Ira.

Inflow (cfs)

Time (cont.)
Noon 6 p.u. Midnight 6 a.Ira. Noon

lnflow (cfs)

100 100 200 500 600

400 300 200 100 100

13.22. If the MuskingumK valueis 12 hr for a reachof a river, and if the X valueis 0.2, what would be a reasonable value of Ar for routing purposes?

PROBLEMS 263 at both endsof a 30-mi river reach: discharges 13.23. Giventhe following valuesof measured

Time 6 n.u. Noon 6 p.rra. Midnight 61.v. Noon 6 p.Ira. Midnight 6 n.rra. Noon 6 p.tu. Midnight 6 e.lr.

lnflow (cfs)

Outflow (cfs)

10 30 68 50 40 3l 23 10 10 10 10 10 10

10 12.9 26.5 43.r 44.9 4t.3 35.3 27.7 t9.4 15.1 12.7 I 1.5 10.8

a. Determinethe Muskingum K and X valuesfor this reach' b.HoldingKconstant(atyourdeterminedvalue),usethegiveninflowhydrographto for valuesofX equalto the computed andptot tt ree outflow hydrographs determine value,0.5, and 0.0. 13.24. Given the following inflow hydrograph:

lnflow (cfs) 6 e.v. Noon 6 p.v. Midnight 6 e.l"l. Noon 6 p.v. Midnight

Outflow (cfs) 10

10 30 68 50 40 3l
ZJ

10

Assumethat the outflow hydrographat a section3-mi downstreamis desired' using values of a. Compute the outflow hydrograptrby the Muskingum method K - 1 1 h r a n dX : 0 . 1 3 ' on a singlegraph' b. Plot the inflow and outflow hydrographs : 0'00' c. RepeatSteps(a) and (b) usingX textbookfor all refert3,25. Carefully review the chapterand consult one other hydrology presented' Compile the results toipplicability ofeach ofthe routing procedures ences frequently' consulted and retained into a list or table, ihis taUleshouldbe

264

ROUTING 13 HYDROGRAPH CHAPTER

REFERENCES
1. Graeff, "Trait6 d'hydraulique." Paris,1883,pp. 438-443. 2. Y. T. Chow, Open ChannelHydraulics. New York: McGraw-Hill , 1959. 3. U. S. Soil ConservationServiie, National EngineeringHandbook, Notice NEH 4-102. PrintingOffice,August 1972. D.C.: U.S. Government Washington, PrevenDisaster Bulletin 1. Kyoto,Japan: 4. S. Hayami, On the Propagationof FloodWaves, t i o n I n s t i t u t e1 ,951. "Multiple Linearization Flow Routing Model," Proc. 5. T. N. Keefer and R. S. McQuivey, ASCEL Hyd. Div.100(HY7) (Iuly 1974). "On the Subjectof a Flood Propagation Method," J. Hyd. Res'IAHRT(2), 6. J. A. Cunge, 20s-230(1967). "KinematicFlood Routing,"Trans.ASCE L0(3) (1967). 7 . D. L. Brakensiek, "Water Studies,"Bureauof ReclamationManual, Yol. 8. U. S. Departmentof the Interior, U. S. GovernmentPrinting Offrce, 1941. D.C.: IV Sec. 6.10. Washington, "Numerical Techniques for Spatially VariedUnsteadyFlow," University 9. T. E. Harbaugh, Center,Rept. No. 3, 1967. of Missouri Water Resources 10. R. K. Price, "Comparisonof Four Numerical Methods for Flood Routing," Proc. ASCE J. Hyd. Div. 100(HY7) (July 1974). "NonlinearKinematicWaveApproximation and M. A. Stevens, 11. R. M. Li, D. B. Simons, AGU II(2) (Apr. 1975). Res. for WaterRouting,"WaterResources "WavePropagation in Rivers," HEL Ser. 8' No. 1. 12. J. A. Harder and L. V. Armacost, University of California, Engineering, of College Hydraulic Engineering Laboratory, Berkeley,June 1966. "Numerical Solution of UnsteadyFlows in Open 13. D. J. Gunaratnamand F. E. Perkins, MA, July 1970. Rept. 127,MIT, Cambridge, Hydrodynamics Channels" "Flood Proc. ASCE J. Hyd. Div. Channels," Along Modiflcation DiSilvio, Wave 14. G. 9s(HY7) (1e69). "Simplifled Dam-Breach 15. U.S. Department of Agriculture, Soil ConservationService, 66,Mar. 1979. Tech.Release RoutingProcedure," "ComparativeAnalysis of Flood Routing Methods," Research Doc. 16. Streldoff, T., et al., 24,Hydrologic EngineeringCenter,U. S. Army Corps of Engineers'Davis, CA, 1980. "Guidelinesfor Calculatingand Routinga Dam-Break l7 . D . L. Gunlachand W. A. Thomas, Center,U. S. Army Corpsof Engineers, Note 5, HydrologicEngineering Flood," Research 1977. Davis.CA.

C h a p t e r1 4

Snow Hydrology

Prologue
The purposeof this chaPteris to: in cold . Indicatethe importanceof snowmeltto water supplyand management regions. . Describemethodsfor measuringsnowfall and describingits water-producing caPabilities. ' Describethe physicsof snowmelt. . presentmodeisfor estimatingsnowmeltunder variousconditionsof temperaground cover,and snowpack. topography, ture, relativehumidity, wind speed,

14.1 INTRODUCTION
In many regions,snowis the dominantsourceof water supply.Mountainousareasin Goodellhasindicatedthat about90 percentof the year$ the Weit areprime examples. of the ColoradoRockiesis derivedfrom snowfall.' Llevations high the suppliin water likely in the Sierrasof California and numerous also proportions are high Equally t"giotrr m tne Nortfrwest.A significantbut lessershareof the annualwateryield in the Northeastand Lake statesalso originatesas snow.It is important that the hydrologist involvedin the the natureand distributionof snowfallandthe mechanisms understand snowmeltprocess. Snowmelt usually beginsin the spring. The runoff derived is normally out of such phase with the periodsof gieatestwaterneed;therefore,variouscontrol schemes An additional this problem. minimize to developed been havJ reseivoirs as storage is that $omeof the greatestfloods result from combinedlargepoint of significanc.e on adesnowmelt.Streamflowforecastingis highly dependent and rainstorms icale the watershed. within flelds snow of characteristics and extent of the knowledge quate fhe water yield from snowfall can be increasedby minimizing the vaporization of snow and melt water. Timing the yield can be managedwithin limits by controlling the melt process' the rate of snowmelt.Early resultscan be be obtainedby speeding by retarding it. The annual or delayed the snowmeltperiod can be extended whereas 14'1' in Fig. is shown States in the United distribution snowfall

&

-v /

S:.-

I O

) \
,4

00 co

I @

()

!1 a !? o

E
q

{)^ > j

5E
p P
.=.Y

o
t

tu5

AND RUNOFF 14,2 SNOW ACCUMULATION

267

factorsis as much a prerequisite of meteorological understanding An adequate The in consideringthe snowmeltpro""rs as it is in dealingwith evapotranspiration. on vapor water of suppliesmoisturefor both snowfall and condensation atmosphere conis a and of energy within a watershed, the snowpack,regulatesthe exchange trolling factor in snowmeltrates. and vegetageologic, topographic, As in the rainfall-runoffprocess, geographic, process. runoff tive factors also are operativein the snow accumulation-snowmelt in estimating used are For rainfall-runoff relations, point rainfall measures areal and time distributions over the basin. A similar approachis taken in snow Mathematical hydrologyalthoughthe point-arealrelationsare usuallymore cornplex. at a given of snowmelt equationscan be usedto determinethe various components also be properties can of depth and other snowpack location. Adequatemeasures amount in estimating , obtainedat specificlocations.The use of thesemeasurements and distribution in area and time of snow over large watershedareasis a much less conditions related to particular areal subdivirigorous procedure.Usually, average sionsover time are usedas the foundationfor basin-widehydrologicestimates'Such proceduresare often in the categoryof index methods(Section 14.6). Snowmelt routines have been incorporated in numeroushydrologic models, some of which also include water quality dimensions.A good accountingof the fundamentalsof the snowmelt process and of contemporary snowmelt modeling may be found in Refs.2-15 listed at the end of this chapter. approaches

AND RUNOFF 14.2 SNOWACCUMULATION

in regionswith heavy snowfall, the runoff Under the usual conditions encountered from the snowpackis the last occurrencein a seriesof eventsbeginningwhen the the ground.The time interval from the start to the end of the process snowfallreaches might vary from aslittle asa day or lessto severalmonthsor more. Newly fallen snow of snowvolumeits waterequivalent hasa densityof about 10 percent(the percentage the settling and compactionincrease would occupy),but as the snow depth enlarges, density.ls The temperaturein a deep layer of accumulatedsnow is often well below after prolongedcold periods.When milder weather setsin, melting occurs fueezing flrst atlhe rno*pu"k surface.This initial meltwater moves only slightly below the surfaceand again freezesthrough contact with colder underlying snow.During the from meltwaterraisesthe snowpack rcfreezingpro""rr, the heat of fusion released t"*p"rutur". Heat is also transferredto the snowpackfrom overlying air and the ground. During persistent warm periods, the temperatureof th9 entire snowpack Iontinually rises'and finally reaches32'F. With continued melting, water begins flowing down through the pack. The initial melt component is retained on snow crystals in capillary films. Once the liquid water-holdingcapacity of the snow is the snow is said tobe ripe. Throughoutthe foregoingprocess'pack density reached, due to the refreezingof meltwaterand buildup of capillary films. After the increases the densityremainsrelatively constantwith continwater-holdingcapacityis reached, the water-holdingcapacity will continueto move ued rnelt. Meltwater that exceeds

268

14 SNOW CHAPTER HYDRoLoGY down through the snowpackuntil the ground is finally reached.At this point runoff can occu.r.Three situationsthat may exist at the ground interfacewhen meltwater reaches it are described by Horton.t6 First, considerthe casewherethe melt rate is lessthan the infiltration capacity of the soil. In addition, downward capillary pull of the soil coupled with gravity exceeds the samepull of the snow lessgravity. The meltwaterdirectly entersthe soil and a slushlayer is not formed. The secondcaseoccurs when a soil's infiltration capacity is greaterthan the melt rate, but the net capillary pull of the snowpackexceeds that of the soil aidedby gravity. Capillary water builds up in the overlying snow until equilibrium is reached at which upward and downwardforces balance.A slushlayer forms and providesa supply of water that infiltrates the soil as rapidly as it entersthe slushlayer. The final situationis one in which the melt rate exceeds the infiltration capacity. A slush layer forms and water infiltrates the soil at the infiltration capacity rate. Excesswater acts in a manner analogous to surfacerunoff but at a much decreased overlandflow rate. As warm weathercontinues, the melt process is maintainedand accelerated until the snowcover is dissipated.

14.3 SNOWMEASUREMENTS AND SURVEYS

Snow measurements are obtained through the use of standardand recording rain gauges,seasonal storageprecipitation gauges,snow boards,and snow stakes.Rain gauges are usually equippedwith shieldsto reducethe effect of wind.3Snow boards are about 16 in. square,laid on the snow so that new snowfall which accumulates periodswill be found abovethem. Care must be taken to assure betweenobservation that adverse wind effectsor other conditionsdo not producean erroneoussampleat the gauginglocation.Snow stakes are calibratedwoodenpostsdriven into the ground for periodic observation to determine of the snowdepthor insertedinto the snowpack its depth. Direct measurements of snow depth at a single station are generally not very useful in making estimatesof the distributon over large areas,since the measured depth may be highly unrepresentative because of drifting or blowing. To circumvent this problem, snow-surveying procedures provide havebeendeveloped. Suchsurveys information on the snowdepth,waterequivalent, density,and quality at variouspoints along a snow course.All thesemeasures are of direct use to a hydrologist. The water equivalentis the depth of water that would weigh the sameamount as that of the sample.In this way snow can be described in terms of inchesof water. Density is the percentageof snow volume that would be occupied by its water equivalent.The quality ofthe snow relatesto the ice contentofthe snowpackand is expressed as a decimal fraction. It is the ratio of the weight of the ice contentto the total weight. Snow quality is usually about0.95 exceptduring periodsof rapid melt, when it may drop to 0.70-0.80 or less.The thermal quality of snow,Q,, is the ratio of heatrequiredto producea particular amountof waterfrom the snow,to the quantity

269 CHABACTERISTICS AREAL SNOW AND 14.4 POINT of heatneededto producethe sameamountof melt from pure ice at 32"F.Valuesof The densityof dry snow is temperatures. Q,may exceed100 percent at subfreezing With variability betweensamples. approximately10 percentbut thereis considerable grPater. percent or of 50 order the on to values increases snow of the density aging, - A snowcourseincludesa seriesof samplinglocations,normally not fewer than about50- 100ft apartin a geometric 10 in number.ttThe variousstationsare spaced in advance.Points are permanentlymarked so that the samelocapattern designed tions will be surveyedeachyear-very important if snow coursememorandaare to be correlatedwith areal snowcoverand depth, expectedrunoff potential, or other significantfactors.Survey dataareobtaineddirectly by forestersand others,by aerial and by automatic recording stations that telemeter photographs and observations, location. information to a central processing In the westernUnited Statesthe Soil ConservationServicecoordinatesmany are also enand private enterprises snow surveys.Various states,federal agencies, in Ref. 14. of snowsurvey dataaresummarized gagedin this type of activity. Sources

ANDAREALSNOWCHARACTERISTICS 14.4 POINT

The estimationof areal snow depth and water equivalentfrom point measurement data is highly important in hydrologicforecasting.

Estimatesof Areal Distributionof Snowfall

or using Thiessenpolygonsdoesnot provide Normally, taking arithmetic averages distribution from point gaugings'This is g snow areal for estimatin results reliable are often pronounced,and gaugingneteffects topographic and orographic because permit the straightforwarduse of normal to enough not dense are frequently works effects are relatively constant orographic regional However, techniques. averaging small whencomparedwith the are that for tracts year to storm year and storm to from This circumstancepermits region.2 in the occurring general storms of areal extent once the basic distribution snow areal the in estimating approaches useful many patternhas beenfound for a region' asOne method used to estimatebasin precipitation from point observations is approximately basin-precipitation precipitation to ratio of station the sumesthat constantfor a storm or storms.This can be statedas'
P o _ Nb Po
No PoNu No

(14.1) (r4.2)

or where
D t b --

Pu:

the basin precipitation

P o : the observedprecipitation at a point or group of stations N t : the annual precipitation for the basin N o : the normal annualprecipitation for the control station or stations

27O

CHAPTER14 SNOW HYDROLOGY

from a map (carefullypreparedif it is The normal annualprecipitationis determined displayingthe meanannualisohyetsfor the region.The precipito be representative) tation is determinedby planimeteringareasbetweenthe isohyets.If the number of depict the basin,Eq. I4.2 canprovide stationsusedand their distribution adequately distributed,weighting coefficients not uniformly a good approximation.For stations used portrayed by a gaugeare sometimes percentage area of the basin basedon the group. in determiningN, for the method.In Another system usedin estimatingareal snowfallis the isopercental percentage as a precipitation is expressed this approach,the storm or annual station on of the normal annualtotal. Isopercentallines are drawn and can be superimposed representing the (NAP) produce new isohyets to a normal annualprecipitation map storm of interest.A NAP map indicatesthe generalnature of the basin'stopographic from this pattern.The advanthe deviations map shows effects,while the isopercental is that relatively consismap directly tageof this methodover preparingan isohyetal as well as into consideration tent storm pattern featuresof the NAP can be taken observedindividual storm variations.

Water Equivalent Estimatesof Basin-Wide

A hydrologistmust be concernednot only with the amount and areal distribution of snowfall, but also with estimatingthe water equivalentof this snowpackover the runoff. Basin water basin, sincein the final analysisit is this factor that determines equivalentmay be given as an index or reported in a quantitative manner such as inchesdepth for the watershed. The customaryprocedurefor determiningthe basin water equivalentis to take data from snowcoursestationsand to provide an index ofbasin conditions. observed to and other approaches weighted averages, Various proceduresemploy averages, of any index this.2The important point to rememberis that the usefulness accomplish the overall basinconditions,not on how favorably is basedon how well it represents it describes a particular point value. Indexesdo not actually provide a quantitative in the factor. evaluationofthe property they cover.Instead,they give relativechanges By introducingadditional data, however,an index can be usedin a prediction equation. For example,if the basin water equivalentcan be estimatedby subtractingthe from the precipitationinput, the index can be correlated runoff and losscomponents with actual basin water equivalentin a quantitativemanner.

Areal Snowcover
Estimatesof the areal distribution of snowfall are very helpful in making hydrologic on the groundat any given A knowledgeof actualarealextentof snowcover forecasts. making seasonalvolumetric in and time is also*applied in hydrograph synthesis generally obtainedby ground are of snowcover forecasts ofthe runoff. Observations approximasurveys, snowcover photography. Between and and air reconnaissance data. hydrometeorological available based on are tions of the extentof the snowcover uniform relatively given normally basin are patterns a within Snowcoverdepletion from data gathered indexes can often be developed from yearto year; thus snowcover stations. at a few representative

14.5 THE SNOWMELTPROCESS

27'I

PROCESS 14.5 THE SNOWMELT

The snowmeltprocessconvertsice content into water within the snowpack.Rates later. Thesediverfactorsto be discussed differ widely due to variationsin causative drainagefrom the snowpack, gencies are not as strikingly apparentwhen considering however,since the pack itself tends to filter out thesenon-uniformities so that the rate. drainageexhibits a more consistent

for Snowmelt EnergySources

radiation, to inducesnowmeltis derivedfrom short-andlong-wave The heatnecessary ofvapor, convection,air and groundconduction,and rainfall. The most condensation and radiation. Rainare convection,vapor condensation, important ofthese sources fall ranks aboutfourth in importancewhile conductionis usually a negligiblesource.

EnergyBudgetConsiderations
as a heat transferprocess,an energybudgetequationcan If snowmeltis considered of the snowmelt.Suchan equationis of the the heatequivalent be written to determine form2 H^: H,r + H," + H" + H" + Hs + He + Hq (14.3)

where H*: the heat equivalentof snowmelt H4 : rrlt long-wave radiation exchange between the snowpack and surroundings H,, : the absorbedsolar radiation H" : the heat transferredfrom the air by convection H": the latent heat of vaporizationderived from condensation Hr : the heat conductionfrom the ground Ho : the rainfall heat content Hn: the internal energychangein the snowpack In this equationH,", Hs, andHo are all positive;I1,1is usually negativein the open; H" and Hnmay take on positive or negativevalues;andH" is normally positive.The actual amountof melt from a snowpackfor a given total heatenergyis a function of thermal quality. The heatenergyrequiredto producea centimeterof the snowpack's waterfrom pure ice at32"F is 80 langleys(g-cal/cm2).Therefore,203.2langleysarc neededto get 1 in. of runoff from a snowpackof 100 percentthermal quality. If the the combinedtotal heat input in langleys,an equationfor snowterm H. represents melt M in inches.is2
MH-

203.2Q,

(r4.4)

where Q, is the thermal quality of the snowpack.For sirbfreezingsnowpacks,Q,, havingsomewatercontent,Q,is lessthan one.A typical 1; for ripe snowpacks exceeds value for these conditions is reported to be 0.97.6Figure 14.2 gives a graphical solution to this equationfor severalvaluesof Q,.

272

CHAPTER14

SNOW HYDROLOGY

w
0 100 200 300 400 500 600
Net heat flux to snow pack, f(langleys)

Figure 14.2 Snowmelt resulting from thermal energy' (After U.S. Army CoJpsof Engineers.2)

EXAMPLE 14.1 with thermal quality of 0.90, determinethe snowmeltin inchesif Given a snowpack the total input is 137 langleys. Solution. UseEq. 14.4 M: H^/203.2Q, : r37/203.2 x 0.90 M : in. rl 0.75 M

Turbulent Exchange is The quantity of heat transferredto a snowpackby convectionand condensation hasbeen Suchan approach equations. from turbulentexchange commonlydetermined mustbe madein and vapor pressure of temperature widely used,sincemeasurements and wind velocity gradithe turbulent zonewherevertical watervapor, temperature, several ents are controlled by the action of eddies.In the following two subsections and convectionmelt are given. Here practical equationsfor estimatingcondensation to acquaintthe readerwith the theory of a combinedtheoreticalequationis presented turbulent exchange. equationcan be written' The basic turbulent exchange

4 o : Adz
where

( 14.s )

property of the air suchaswater Q : thetime rate of flow of a specified area horizontal vapor through a unit dqldT = the vertical gradient of the property

PROCESS 273 14.5 THESNOWMELT

e : the property z : lhe elevation A: an exchange coefficient Property 4 must be unaffectedby the vertical transport.Propertiespertinent to this discussion are the air temperature, water vapor, and wind velocity.Theoretically,the potential temperatureshouldbe used,but air temperatures measured at normal disof tancesabovethe snowpackdo not causeseriouserrors.The potential temperature dry air is that which the air would take if broughtadiabaticallyfrom its actualpressure pressure. to a standard Gradients of the various properties of importance here follow a power law distribution where conditions of atmosphericstability exist.rTThis qualification is Logarithmic profiles are characteristicof the atmosphere'sstate over snowfields.2 more nearly representative conditions.The power of neutral or unstableatmospheric law providesthat the ratio of valuesof a property determinedat two levelsabovethe snow is equivalentto the ratio of the levelsraised to somepower. Thus,
Qz_ Qt

(:31
\Zr,/

/ . \r/n

(r4.6)

where

the value of the property the elevation(with subscripts denotingthe level) power the law exponent

to be the property value at this height, and the If 3, is made equal to I, q, assumed subscript droppedfor the second level,Eq. 14.6becomes
q : qtz\/"

(r4.7)

at the level The magnitu de of q is takenasthe differencein valuesof 4 measured is the For example,if T : 38'F at height z, and temperature z and the snow surface. property of interest,then q: (38 - 32) : 6"p. The gradient of the property dqldz can be obtainedby differentiatingEq. 14.7:

ff: (et),,,_",,"
o : AIL)zo-d/"

(14.8)

equation(14.5),the If this expression is substituted in the basicturbulentexchange followins relation is obtained:

(r4.e)

from Thus,eddyexchange elevation ofthe property at a specified e is determined coefficientis alsorelatedto observations ofthe ptoperty at unity level.The exchange of elevationz. For equilibrium conditions up to the usual levels of measurement gradientsof thesevariablesare suchthat the eddy transfer moistureand temperature, coefficientA mustbe of moistureand heatis constantwith heisht. Then the exchange

274

HYDROLOGY 14 SNOW CHAPTER

related to the prope"t r.:O*r,i inversely - )n,Orr, dqld, h result: the following in Eq. 14.8for dq/dz gives Substitution
A - Ar4"-ttr"

(i4.10)

(14.11)

in since(dq/dz)t : et/nfor I : 1. Now, if the valueof A from Eq. 14.11is inserted Eq.14.9,

n: o,(T)

(r4.r2)

level has been shownto be directly The exchange coefficientat an observation Therefore,it may be proportional to the wind velocity measuredat that elevation.ls written that A': ko1

(14.13)

where o1 : the wind velocity at Level one k : a constantof proportionality for 41 in Eq. I4.I2 gives Substituting

a: \;)a'"'
Using the power law equation(14.7), we find that q, : q;1/"
and D, : 117-1/n

/k\

(r4'r4)
(14.15)
(14.16)

level of o inBq.14.14 and denotingthe observation After making thesesubstitutions thesemay be different),Eq. 14.14becomes as Tu,andthat of q as za(since

n: (!")rr,zu)-'/nq.ou

(r4.r7)

equation. Considerationis now given to turbulent exchange This is a generalized and convectionmelt. for condensation equations specific theoretical developing property to be tlansported The melt. condensation of the consider the case First, the transfer expresses coefficient exchange since the and is water vapor, in this case air mass.This of the content moisture the determine it is necessary to air maqs, of an of the gives weight the which humidity, the specific by using be accomplished can can be used to 14.18 Equation air. of moist in a unit weight contained water vapor humidity: specific calculate q : 0.622e (14.18)

where e : the vapor pressure of the moist air . Po: the total pressure

PROCESS 275 14.5 THESNOWMELT Inserting this expression in Eq. 14.17 for q"yields2

"u, )-'/^ (9J4:), n" : (I)u"z

(r4.re)

," : t s(l)u.,,)-'^(99),"u,

(r4.20)

The proportionality constantk is a complexfunction relatedto the air density and other iactors.Sinceihe densityof air is a function of elevation,k also varieswith of density,and thereforeof elevation, The constantmay be madeindependent height. "introducing a factor to compensatedirectly for the density-elevationrelation' by this, and the equationcan be adjustedby to accomplish serves Atmosphericpressure muttipiying Uy tne ratio pf po, wherepis the pressureat the snowfieldelevationand tevetpr"rrur". Introducingthis ^ratioinBq. 14.20and a new constantk1, po is the se-a gives' which is related to sealevel pressure,

m (99),,u, " : a.s(\){r"r,r','

{r4.2r)

importanceinEq. 14.17is airtemperature. Forconvectionmelt,thepropertyof units, the specific heat of air crmust b-e ttreimat into To convert air temperature H" introduced. Putting these values in Eq. I4.l7,heat transfer by eddy exchange convertsto2

I/n coror), o, : (I) uoz6)-

(14.22)

cgs Sincethe latent heatof fusion is 80 callg, convectivesnowmeltM" in gramsin the is givenby H"l8O, or system / t \/lc\. (r4.23) ,. : (*J ())(,"20)-'h, or"uu Introducing the elevationdensitycorrectionp/po,we obtain

(l)r,",,r',(o1),,r", * : (#)
M"" =

(14.24)

Equations I4.2I and 14.24 can be^ combined into a single convectionmelt M"" equationof the form' condensation

Lk"ru)-,,"(#fir,r"

(r4.2s)

276

CHAPTER14

SNOW HYDROLOGY

theoreticalequationfor snowmeltthat resultsfrom the turbuleni This is a generalized that the exchange tfansfer of water vapor and heat to the snowpack.It is assumed by coefficientsfor heat and water vapor are equal. Their evaluationis accomplished experimentation.r A combined physical equation of the general nature of Eq. 14.25 has been melt by Light.le Widely used,its individual convectionand condensation developed in following sections.The combined form of the Light componentsare discussed equationis

D :
where

pk',

80 rn(alz) ln(b/zo)

+ ( e - 6.IDryf ulc,r P J

(r4.26)

D - the effective snowmelt(cm/sec) air density L _ von K6rm6n's coefficient= 0.38 ^ 0 parameter: 0,25 the roughness and vaporpressure the wind velocity,temperature, at which levels r b : the resPectivelY measured, are U : the wind velocity co : the sPecific heat of air 7 :;the air temperature e : the vapor pressureof the air p : the atmosPheric Pressure

Convection
by convection. to the snowpack Heat for snowmeltis transferredfrom the atmosphere wind velocity. and to temperature process is related The amountof snowmeltby this in inches of snowmelt depth 6-hr the The following equationcan be usedto estimate by convection:2O (r4.27) D: KV(T - 32) where V : the mean wind velocity (mph) Z - the air temperature('F) , a On the basisof the theory of air turbulenceand heat transfer (turbulent exchange), been K of 0.00184 X 10-0'0000156'has coefficient valuefor the exchange theoretical in feet, is usedto reflect the change given by Light.le In this relation h, the elevation is said to in barometric pressuredue to the difference in altitude. The expression of wind and representconditions for an open, level snowfieldwhere measurements abovethe snow.Values are madeat heightsof 50 and 10 ft, respectively, temperature vary from 1.0 at sea level to 0.70 at 10,000ft of of the exprtssion10-00000156' elevation.The actual valuesof K are normally lessthan the theoreticalflgure due to such factors as forest cover. Empirical 6-hr K values have been reported in the literature.20 due to for the houily sno.wmelt Anderson and Crawfordls give an expression convectionas M _

cv(T"_ 32) Q,

04.28)

14,5 THE SNOWMELT PROCESS 277 where M : V : To : Q, : c: the hourly melt (in.) the wind velocity (mi/hr) the surfaceair temperature('F) the snowquality a turbulent exchange coefficientdeterminedempirically

Temperature measurements are at 4 ft, with wind gaugedat 15 ft. The corresponding valueof c is reportedas 0.0002.

Condensation

snow.A total yield of around 8.5 in. of snowmeltincluding the condensate is thus derived. A water vapor supply at the snow surfaceis formed by the turbulent exchange process; consquently, a masstransferequationsimilar to thosepresented for evaporation studiesfits the melt process. An equationfor hourly snowmeltfrom condensation takesthe formrs

- .11) u : W n { r "6
where

(14.2e)

b : an empirical constant eo : the vapor pressureof the air (mb), the numerical value 6.11 : the saturationvapor pressure(mb) over ice at 32'F (e, must exceed 6.1 l)

Also, M, Q,, and v are as previouslydefined.The constantb hasa valueof 0.001 for temperatureand wind measurements at 4 and 15 ft, respectively.ls A similar expression but for 6-hr snowmelt(D) is given as
D:KrV(eo-6.11)

(14.30)

where the theoretical value of K, is said by Light to equal 0.00579 if wind and

- e,) kV(e" ' " Q,

(14.3r)

where E : the hourly evaporationin inches es : the saturationvapor pressure over the snow k : an empirical constant

278

HYDROLOGY 14 SNOW CHAPTER Also, V, eo,andQ,areasdefinedbefore.lsIntheexpressionk : 0.0001,temperature and wind measurements are taken as for Eq. 14.30, and the temperature of the air is assumed equal to that of the snow surface for temperatures below 32oF.

Melt Radiation
radiation receivedby a snowpackcan be a The net amount of short- and long-wave very important source of heat energy for snowmelt. Under clear skies, the most in radiationmelt are insolation,reflectivity or albedoof the snow, significantvariables and air temperature.Humidity effects, while existent, are usually not important. in the amount of radiation from an open When cloud cover exists,striking changes snowfield are in evidence.The general nature of these effects is illustrated in as a function of cloud radiation exchange .2Combinedshort- and long-wave Fig. 14.3 Radiationmelt is shownto be more significantin the height and coveris represented. spring than in the winter. It should also be noted that winter radiation melt tendsto cloudheightasa resultof the more dominant with cloudcoveranddecreasing increase radiation during that period. role playedby long-wave in regulating radiative Forest canopiesalso exhibit important characteristics from thoseexhibitedby the cloudcover, Theseeffectsdiffer somewhat heatexchange. Cloudsand treesboth limit insoradiation is concerned. especiallywhereshort-wave insolation lation, but cloudsarevery reflective,while a largeamountof the intercepted the forestis warmedand part of the incident by the forest.Consequently, is absorbed radiation; an additional energydirectly transferredto snow irl the form of long-wave fraction is transferredindirectly by air also heatedby the forest. Figure 14.4illustratessomeeffectsof forestcanopyon radiation snowmelt.The In conditions for a coniferouscover in the middle latitudes.2 figure typifies average with completeforest cover, and in winter, the maximum radiation melt is associated shouldnot be spring the greatestradiation melt occursin the open. Generalizations effects of forest cover on drawn from thesecurves,which indicaterelative seasonal Another factor affectingradiation melt is radiation melt for the conditionsdescribed. the land slopeand its aspect(orientation).Radiationreceivedby north-facingslopes for example. inclinesin the northern hemisphere, is lessthan that for south-exposure Solar energy provides an important sourceof heat for snowmelt. Above the the thermal equivalentof solarradiation normal to the radiation earth's atmosphere, 3.97 x 10-3 Btu/cm2).The path is 1.97 lingleys/min (1 langleyis approximately actual amount of radiation reachingthe snowpackis modified by many factorssuch and vegetalcover.The importanceofvegetal ofcloudiness,topography, asthe degree has promptedmany forest managecover in influencingsnowmelt,long recognized, snowmelt.t'to'"''o to regulate ment schemes Two basic'lawsare applicableto radiation.Planck'slaw statesthat the temperature of a blackbodyis relatedto the spectraldistribution of energythat it radiates. producesStefan'slaw, Integrationof Planck's law for all wavelengths Ro: cT' where R, : the total radiation constant o : Stefan's [0.813 x 10-10langley/(min-K-')] T : the temperature(K)

(r4.32)

14.5 THE SNOWMELTPROCESS

crurar,.ier,iiTth

M,, = 2.0o[1_ (0.82- 0.0244N] Mt=4.4r[r-(1 -}.\UAM

Amount ofclouds, N (a)

- O.O'U4I\tl M,, = 0.50U- (0.S2 Mt=4.84[1 * (1 _ 0.0242)t{l

J"*qv;
o u.u 4.2 0.4 0.6 0.8 l.t Amount ofclouds, N (b)

Daily radiation melr in the open with cloudy {igure.fa3 skies: (a) during spring, May 20; and (bf during wintei, February 15. (After U.S. Army Corps of Engineeis.r;

280

CHAPTER14 SNOW HYDROLOGY

t t I I

r.)
^ r _t "n"n t : t
lr I

> 10

\
.= rt \

M,= M^

Mrt

-4
\M,

(r 'F))- 3.30 =3.821F + .7s7

0.6 1.0

-o.5
0.0 0.2 0.4 .'
(a)

F cover, canopy Forest

0.5

(l) ,*^=l,r*
I

H o.o
-0.5

'--1

Mf

Mr, + Mr1

1-1.0
0.0

Mrt

? 'qTF r

-n - 3.30 7s7Q.
0.6 0.8

0.4
(b)

F cover, canopy Forest

melt in the forestwith clear - Figure 14.4 Daily radiation skies:(a) duringspring,May 20; and (b) duringwinter, 15.(AfterU.S.Army Corpsof Engineers.2) February Becausesnow radiates as a blackbody, the amount of radiation is related to its temperature(Planck's law), and total energyradiated is accordingto Stefan'slaw. Long-waveradiation by a snowpackis determinedin a complexfashionthrough the interactionsof temperature,forest cover, and cloud conditions'

SNOWMELT PROCESS 281' 14.5 THE Direct solar short-wave radiation receivedat the snow surfaceis not all transferredto sensible heat.Part of the radiationis reflectedand thus lost for melt purposes. Short-wavereflection is known as albedo and ranges from about 40 percent for melting snow late in the season to approximately80 percent for newly fallen snow. This property Valuesas high as 90 percenthavealsobeenrepofted in severalcases.22 of the snowpackto reflect large fractions of the insolation explainswhy the covers persistand air temperatures remain low during clear, sunny,winter periods. That portion of short-wave radiation not reflectedand availablefor snowmelt may becomelong-wave radiation or be conductedwithin the snowpack.Some heat may also be absorbed by the ground with no resultantmelt if the ground is frozen. An expression for hourly short-wave radiation snowmeltis given as2
MH^

203.2Q,

(r4.33)

radiation (langleys) H^ : the net absorbed where ' 203.2 : a conversionfactor for changinglangleysto inchesof water betweenthe snowcover radiation is exchanged When the snowquality is 1, long-wave and its surroundings.Snowmelt from net positive long-wave radiation follows Eq. 14.33.If the net long-waveradiation is negative (back radiation), there is art equivalentheat loss from the snowpack. An approximate methodof estimating12-hr snowmeltDn (periodsmidnight to The noon, noon to midnight) from direct solarradiation has been given by Wilson.2o relation is ofthe form (14.34) Dp = Do(l - 0.75m) where Do : the snowmeltoccurring in a half-day in clear weather (0 for clearweather,1.0 for completeovercast) m : thedegree of cloudiness Suggested valuesfor Do are 0.35 in. (March), 0.42 in. (April), 0.48 in. (May), and 0.53 in. (June)within latitudes 40-48".2o

Rainfall
Heat derivedfrom rainfall is generallysmall, sinceduring thoseperiodswhenrainfall of the rain is probablyquite low. Nevertheless, occurson a snowpack, the temperature at highertemperatures, rainfall may constitutea significantheat source;it affectsthe aging processof the snow and ffequently is very important in this respect. An equationfor hourly snowmeltfrom rainfall isrs

(14.3s)
P : the rainfall (in.) to be that of the rain T- : the web-bulb temperatureassumed This equationis basedon the relation betweenheatrequiredto melt ice (I44 Btu per pound of ice) and the amountof heatgiven up by a pound of water when its temperature is decreased by one degree. where

282

14 SNOW HYDROLOGY CHAPTER

are given by Daily snowmeltby rainfall estimates - 32) Ma: 0.007Pd(T"

(r4.36)

where Md : the daily snowmelt(in.) , P d : the daily rainfall (in.) air taken at the 10-ft To : the meandaily air temperature('F) of saturated

level23

Conduction
are radiation, convection,and condenof heatenergyto the snowpack Major sources sation. Under usual conditions, the reliable determinationof hourly or daily melt plus rainfall ifit occurs.An additional quantitiescanbe foundedon theseheatsources sourceof heat,negligiblein daily melt computationsbut perhapssignificantover an is ground conduction. entire melt season, Ground conductionmelt is the result of upward transferof heatfrom ground to snowpackdue to thermal energythat was storedin the ground during the preceding summerand earlyfall. This heatsourcecanproducemeltwaterduringwinter and eady springperiods when snowmeltat the surfacedoesnot normally occur. Heat transfer by the relation2 by ground conductioncan be expressed dT Hn: K-where K : the thermal conductivity "r'J"n dTlda : the temperaturegradient perpendicularto soil surface (14.37)

small. Wilson The snowmeltfrom ground conductionis generallyexceedingly notes that after about 30 days of continuoussnowcover,heat transferredfrom the The amountof snowmeltfrom ground conducground to the snow is insignificant.20 has been estimatedat approxirnately0.02 in.lday'23 tlon during a snowmelt season Groundconductiondoesact to providemoistureto the soil; thus,whenotherfavorable of runoff can be expected. conditionsfor snowmeltoccur, a more rapid development physicsof snowmelt.The mannerin which heat This sectionhasemphasizedthe -I4.3I and 14.27 Equations wasdiscussed. to initiatethemelt process canbe provided L4.33-I4.37 inclusivecan be usedto estimatethe melt at a given point. The task of in sucha simple computingrunoff from snowmeltin a basin cannot be approached fashion,sincethere are many complexfactorsoperative.The remainderof this chapFigter is devotedto the general subject of runoff from snowmelt investigations. ure 14.5 illustrateshourly variation in the principal heat fluxes to a snowpackfor a cloudy day. EXAMPLE 14.2 existedfor During a completelycloudy April period of l2hr, the following averages 44" N: air latitude of at a lel'el sea ft above a ripe snowpacklocated at 10,000 avenge 65Vo; relative humidity, 10 mph; velocity, temperature50' F; mean wind reading,48oF. Estimate rainfall intensity,0.03 in./hr for I2hr; wet bulb psychrometer radiation, and warm rain condensation, for convection, the snowmeltin in. of water period. for the 12 hr

April 23 100 80 9 6 0
Eo
<2

Apnl24

Aprii 25

1600 2000 2400 0400 0800 1200 1600 2000 2400 0400 0800 1200
--- Short-waveradiation Long-wave radiation ffi
nm

lt ii
l,Incident short-wave radiation (t)

Incident short-wave radiation

Reflected shon-wave raoanon

G *
40

Absorbedshort-wave radiation

Downwad

long-wave radiation (R2)

H . n E z v

l -

-20

=
o

_40

naiatipn tn) J.bwaa]one-wa1e

Reflected shofr wave radiation (1,)

E E o.o4 On?

.,
6

E 0.02
(Each bs reDresents mem value

E o.ot
0
1700-1800hr 2400*0100hr' 0900-1000Itr
no condensation

0.01 0.25 0.20 0.15

E 0.10
o n

no condensation -

no convgctiotr or -

meltllcondensationmeltmelt t t t t l E Radiation melt (Mr) Convrction condensation 6 mett (Mce) --- Radiationmelt total (Hourlyconvection-cotrde$ation meltis added to of subtacbdftom totalhouly radiation meltto arnveattotalhouly
computed melt) I Totalcomputedmelt

Total Snow Melt (M) ud Runoff (O)

Computed melt (net for period0900- 1800 hr = l22ir-022in = I 00 in.)

_o*
0 -0.05 Apil23-----l-Net nighttime loss = 0.22 in

1600 2000 2400 0400 0800 1200 1600 2000 2400 0400 0800 1200 Apfl24
April 25

Figure L4.5 Hourly variation in principal heat fluxesto a snowpackfor a cloudyday.(After U.S.Army Corpsof Enginers.2l

284

CHAPTER14

SNOW HYDROLOGY

Solution a. Convectionmelt, 6 hr

D:KV(r-32) x4 1 0x ( 5 0 - 3 2 ) : 0 . 5 0 i n . D : 2 x 0 . 7x 0 . 0 0 1 8 melt,6 hr b. Condensation

D:KrV(e"-6.11) x 0 . 6 5- 6 . 1 1 ) x 10 x (12.19 D : 2x 0.00578 : 0.21in. . Radiation melt, 12 hr Dn: d. Rainfall melt. hourly M : P(r-'- 32)lt44Q, 14 : 10.03x 12 x (48 - 32)l/(144 x 0.97) : s.64 Thus,total melt is 0.86 in. rr D o ( 1- 0 ' 7 5m ) D r 2: 0 ' 4 2 x ( 1 - 0 . 7 5x 1 ) : 0 ' 1 1

(14.27)

(14.30)

(r4.34)

( 14.3s )

RUNOFF DETERMINATIONS 14.6 SNOWMELT

from snowmelthavebeenfollowed.They to runoff determination Variousapproaches that completelyignore the physical rangefrom relatively simplecorrelation analyses Most methodsusing physicalequations. snowmeltprocessto relatively sophisticated ofrecescorrelations,analyses as basedon degree-day can be considered techniques sion curves, correlation analyses,physical equations,or various indexes.Each is in turn. discussed

Purposesof SnowmeltRunoff Estimates

Snowmeltrunoff estimatesare extremelyimportant for many regionsof the United water yields for a diversity of Statesand other countriesin (1) forecastingseasonal water supply purposes,(2) regulating rivers and storageworks, (3) implementing flood control programs, and (4) selectingdesign floods for particular watersheds. Maximum floods in many areas are often due to a combination of rainfall and of snowmeltrunoff hasthe sameutility snowmeltrunoTf.In effect, the determination as the calculationof runoff from rainfall. In someareasit will, in fact, be the more important of the two.

Condition Snowpack
The mannerin which runoff from eitherrainfall or snowmeltis affectedby conditions Variousviewson prevalentwithin the snowpack is of primary interestto a hydrologist. of a snowpackhave been advanced.These range from the storagecharacteristics

14.6 SNOWMELTRUNOFFDETERMINATIONS

285

conceptthat a snowpackcan retain large amountsof liquid water to the hypothesis that snowpackstoragi is negligible.Thereis no universallyapplicablerelation, and it on a knowledgeof the character important to baseany runoff considerations becomes condition, at the time of study.Winter runoff is relatedto a snowpack's of a snowpack in the spring,onceactivemelt begins,little or no delayin the transportof melt whereas or rainfall through the snowpackoccurs. For drainagebasins in mountainousareas,snowpackstorageeffects may be into relatively uniform areas.Normally, by subdividingthe watershed approximated by usingelevationzones.Snowpackat the lowestlevelsmay this will be accomplished a be conditionedto transmit readily rain or meltwater,whereasin higher elevations be very the snowpackmay liquid water deficit may prevail. At uppermostelevations, dry and cold and thus in a condition for the optimxm storageof water. The storage of the measurements zonesmustbe basedon representative potentialof the watershed water equivalent,moisturecontent,and snowpack snow depth,density,temperature, relatesto the physicalstructureofthe pack' UnforThe snowpackiharacter character. tunately, adequate1;r"usrrr"rof all thesefactors are not always availableor easily obtained. Estimatesof changesbetween sampling periods are usually indexed to variables. readily observedmeteorologic can be The formulation of snowpackstorageand time delay characteristics is relateddirectly to pack.In this case,storage a homogeneous by assuming fashioned the liquid water defici1and cold content of the pack. Time delay is a function of the inflow rate. It is consideredthat the snowpackstoragepotential must be entfuely permits but the assumption In reality this is not the case, beforerunoff begins. satisfied snowpack an analysisto be made. As melt proceeds,the storagepotential of any diminishes. to be the sum of is considered beforerunoff commences of a snowpack Storage to 0'C (cold of the snowpack waterrequilementto raisethe temperature the equivalent If the cold content capacityof the snowpack. contentlV")and the HqJidwater-holding to 0oC,it may be temperature is given in inchesof water neededto bring a snowpack by2 represented

w"T" *r=f fi
where

(14.38)

7] : the mean snowpacktemperaturebelow OoC lyo : the initial watei equivalentof the snowpackin inchesfor an assumed heatof ice of 0'5 specific is thus given by to OoC temperature to raisethe snowpack The time L in hoursneeded , " - r6o1i+*7 WoT,

(14.3e)

whereI is the rainfall intensity(in./hr) andm is the rate of melt (in'/hr)' Storage requiredto meet the liquid water deficit of the snowpackis given by

tr:#(%+w.)
( - : the amount of water stored(in.) where ".1 T _ the percent deficiencyin liquid water of the snowpack JP

(14.40)

286

CHAPTER14

SNOW HYDROLOGY

The time in hours r, neededto fill the storageSyis given by

+ t': fi\wg+ W") 100(i m)

(r4.4r)

It has been specifiedthat the total storagepotential Soto be met prior to the runoff is given as (r4.42) Sr:W'"*Sy to the runoff until sinceit is not available This is also known as "permanent" storage, componenttransitory storage hasfinally melted.An additional storage the snowpack while movingthroughit to becomerunoff. Until S,is that waterstoredin the snowpack as initiation of runoff, the transitory storagein inchescan be expressed

(i+m) ^ D __T__ ",:

where D : the depth of the snowpack (ft) through the snowpack(ftlhr) V : the rate of transmission The delay time of water in passingthrough the snowpackt' is thus
a t

(r4.43)

D
v

(r4.44)

is very small comparedwith Wo,the depth of the for /, in hours.Assumingthat I4z" snowpackis given by

O:%
P'

(r4.4s)

Then of the snowpack. with p" the density wo (r4.46) nv in the total waterS storedin the snowpack, Beforethe runoff commences, inches, is givenby t'-

S:W"+++S,
which can also be written

(r4.47)

t = * ( 1 f t + f i d i. + "m r ,\ r )
before runoff is producedis thus2 The total time in hours that passes t:t"+tr+tt

/ r

(14.48)

(r4,49) (14.50)

z : -, f ! , - r -+ { 1 t = w'1160(i f + r, Too1,-, p,v)

term theonlysignificant runofffromthesnowpack, theactive After establishing with the overallbasinlag and is /,, andthis is usuallysmallcompared in Eq. 14.49

RUNOFFDETERMINATIONS 14.6 SNOWMELT

287

33
End rain

31 -:
'*2
= )

30
Q

6 - ' E
.Z
d o

te

- 2 6

24 23 0 6 12 18 24 30 36 42 48 54 60 66 72 78

(br) Time duringrainfall' in a snowpack balance Figure 14.6 Water can be neglected.With increasedsnowmelt and runoff, additional increments of water prevlouslywithheld by snowblockageto drainageoutlets and other factorsare A deep at present'24 Adequatequantifi;ation of this cannotbe accomplished released. -5oC, could storeabout 4 in. of of temperature a mean having ft, snowpack,say 15 fiquid waterbeforethe onsetof runoff. Figure 14.6illustratesthe nature of the water balancein a snowpackduring a rainstorm. EXAMPLE 14.3 A core sampleof a snowpackproducesthe following information: air temperature, 10 ft; 68"F; relative humidity, 2b percent; snowpackdensity,0'2; snowpackdepth' snowpacktemperature,22"F. a. b. c. d. What is the vapor pressureof the air? on the snowpackoccur, basedon the vapor pressure? Will condensation what is the cold content of one sq ft of surfacearea of the snowpack? Is the snowpackripe? -

Solution

a. From AppendixTableA.2, eo: 233 mb for a relative humidity of 20 percenr,o = 4'66 mb will not occur sincethe air is unsaturated" b. Condensation c. Cold content

160 W": WsT"f

288

CHAPTER14 SNOW HYDROLOGY

: 22oF: -5.6oC Temperature in. W": 0.20x l2Ox 5.6/160: 0.84 is belowfreezing. I I its temperature is not ripe since d. The snowpack

Indexes
Hydrologic indexesare made up of hydrologicor meteorologicvariablesto describe their functioning. The index variable is more easily measuredor handier than the When mean fixed relations are known to exist betweenpoint elementit represents. and watershedvalues,indexescan be used to record both areal and measurements temporal aspectsof basin values. Indexes serve to permit (1) readily obtainable cannotbe which themselves to depict hydrologicvariablesor processes observations and (2) simplificationof computationalmethodsby allowing indieasily measured, valuesin time and to replacewatershed or groupsof observations vidual observations of an index is basedon (1) the ability ofthe index to describe space.The adequacy it represents the physicalprocess , (2) the randomvariability of the obseradequately is typical of actual conditions, point observation (3) the to which the degree vation, and basin means.2 point measurement the beiween (4) variability nature of the and Indexesmay be equationsor simplecoefficients,and variable or constant. thermal budgetstudiesare The types of data requiredto make comprehensive ones.As or part for wut"tJh"dt otherthan experimental in wh^ole normally unavailable a result, a hydrologistmust make the best use of information at hand. The most humidity, commonly availabledata aredaily maximum and minimum temperatures, of thesedata, andfew measurements and wind velocity.Lessprevalentare continuous datacan Hourly cloudiness stationsrecord solarradiation or the durationof sunshine. be obtainedfrom local airport weatherstations. sometimes A completelygeneralindex for reliably describingsnowmelt-runoff relations valid only for includecoefficients Most indexes for all basinshasnot beenestablished. conditionsand aretherehydrologic,and seasonal meteorologic, specifictopographic, Table 14.1 shows some types of fore limitedlnipplicability to other watersheds. in snowmeltinvestigations. indexesthat havebeen used successfully VARIABLES BUDGET THERMAL TODESCRIBE USED INDEXES 14,1 SOME TABLE
component Thermal budget
radiatron Absorbed short-wave Long-waveradiation' Convectiveheat exchange Heat of condensation Index Duration of sunshinedata Diurnal temperaturerange Air temperaturefor heavy forestedareas radiation shouldbe estimated For open areaslong-wave (7" - T)V, where f is air temperature, T6 the snow surface temperatureor basetemperature,and Vthe wind speed of air and snow (e" - e")V,whereeoand e, are vapor pressures surfaceot a basevalue, and V is wind speed

" Figure 14.7 iilustrates an approximatelinear relation betweenmelt and long-waveradiation usedby the U.S, Army Corps of Engineers for index purposes.

14,6 SNOWMELT RUNOFFDETERMINATIONS

289

The snowmeltrunoff equationstatedin terms of thermal budgetindexesis

Y:a+>b$t
where I : a : b, : 4 : the snowmeltrunoff aregression constant the regression coefficients individual indexes

( 14.s1)

Variousindexes usableto represent the termsof Eq. 14.51are selected and a standard regression performedto determine analysis a andb,.It shouldbe notedthat everyterm in the heatbudgetequationis not alwayssignificantfor a particular analysis, and thus the numberof Xr will vary for different basinsand conditions.A final melt equation
1.6
Long-wave radiation melt in forest or with low overcast sky in open Ma= 1Z-L n:{-r'1 ,, toTf,- oT!) {referto Kelvin scale)

-3.355 = 604x lA-1274

E 'c o

n / -v.+

Long-wave radiation melt in open

Ma= zoittlpl rcrf,-orlt = 604x IO-12 Tx- 3.355 = o.o29 (Ta_ 32)
(refer to Fahrenheit scale)

1L O

r l t l M, = long-waveradiationmelt, in./day

-r.o

P = ftee water content of snow (taken as 0.03 in this case) rA = absolute air temperature, K ?s = absolute snow surface temp = 273K

-2.0

o = stefm's constant = 0.813 x 10. lolmgley/min-K-a Io = mean daily air temperatue, F

1 A

10

20

30 40 60 50 Meanair temperature, Zo("F) 270 2'75 280 285 Absolutetemperature, Z1(K) 290 300

260

265

Figure 14.7 Long-wave radiation melt, with linear approximation. (After U.S. Army Corps of Engineers.2)

290

CHAPTER14 SNOW HYDROLOGY

t.6
dataof Equationderivedfrom combined 1954 and1955 (T ^ -'77) G + 0.0245 RO = 0.00238 D=0.90 r=0.95 = 0.36 in. S"- . = 0.1| in'

t.4

'

1 n
z}

ts E U.6
a

"r_"n
.C .l

,/
Mean ru noff 0.59

Z
\C "y-x

ri

5 0.6
0.4

,/,

aa,

'24.
0.2

/ a o

0.2 a .

0.8 0.4 0.6 runoff(in.) Estimated

1.0

1.2

Figure 14.8 Observedversusestimatedrunoff for (X) snowmelt 1954 and (O) 1955. RO = the daily generated runoff (in.) depth over a snow-coveredarea; G : the by snow in the open daily net all-waveradiation absorbed (langleys); 7-u* : the daily maximum temperature for Boise ("F); r = the coefficient of correlation; D : the coefficient of determination;S, = the standarddeviation of observedrunoff (in.); Sr-, : the standarderror of the stimated runoff (in.). (After U.S. Army Corps of

Engineers.2) for the partly forestedBoiseRiver basin above devekiped by the Corps of Engineers2 Twin Springs,Idaho, was

+ 0.0245(T^ * - 77) Q : 0.00238G

(r4.s2)

area where Q = the daily snowmeltrunoff (in.) over the snow-covered in the G = an estimatedvalue of the daily all-waveradiation exchange open (langleys) ZLu, : the daily maximum temperatureat Boise (T) The equationis said to predict the daily snowmeltrunoff valueswithin 0.11 in. of observedvalues about 67 percent of the time. Figure 14.8 illustrates this relation. shouldseek for snowmelt,a hydrologist suitableindexes In attemptingto develop the approachmost closely resemblingthe thermal budget of the area, within the limitations of availabledata.

14.6 SNOWMELT RUNOFF DETERMINATIONS 291

Temperature Indexes
The atmospheric temperature is an extremelyusefulparameterin snowmeltdetermination. It reflectsthe extentof radiation and the vapor pressureof the air; it is also sensitiveto air motion. Frequently,it is the only adequatemeteorologicvariable regulady on hand, so widespreaduse has been made of degree-dayrelations in snowmeltcomputations. I A degreeday is defrnedas a deviation of 1ofrom a given datum temperature consistentlyover a 24-hr period. In snowmeltcomputations, the referencetemperature is usually 32"F. rf the mean daily temperatureis 43oF, for example,this is equivalentto 11 degreedays above 32'F.If the temperaturedoes not drop below freezingduring the24-hr period, there will be24 degree hr for eachdegree departure above32'F. In this examplethere would be 264 degree hr for the day of observation. Variouswaysof estimatingthe meantemperature haveenabledinvestigators to take severalapproaches. one methodis simplyto average the maximumandminimum daily temperatures. Basesother than 32"F are also used.Regardless of the particular attackemployed, a degree hour or degree day is an index to the amountof heatpresent for snowmeltor other purposesand has proved useful in point-snowmeltand runoff from snowmeltdeterminations. The standardpracticein developing snowmeltrelationson the basisof temperature is to correlatedegreedaysor degreehours with the snowmeltor basin runoff. In somecases, other factorsare introducedto defineforest covereffectsand/or other influences. Another approach often usedis to calculatea degree-day factor-the ratio of runoff or snowmeltto accumulated degreedaysthat producedthe runoff or melt.
2.8
1 A

; 1 A o

B
6 r-z

34

38

42

46 50 s4 58 Mean daily temperature, Z('F)

62

66

Figure 14.9 Mean temperature index. The equations are applicable only for the range of temperatures shown in the diagram. (After U.S. Army Corps of Engineers.2)

292

CHAPTER14 SNOW HYDROLOGY

and between factor has been found to vary seasonally Unfortunately,the degree-day values should be used with caution. Pointbasins;therefore, single representative rangefrom 0.015to 0.20 in. per degree basins factorsfor snow-covered degree-day point value of 0.05 can per day when melting occurs.Gartskastatesthat an average be used to representspring snowmelt, provided that caution is used. Linsley and factorsare usuallybetween0.06 and 0.15 othersstatethat basinmean degree-day and at meltingtemperatures. snowcover in./degreedayunderconditionsof continuous for springtimesnowmeltfor clear index equations Figure 14.9illustratestemperature areas.' and forested

BasinSnowmeltEquations Generalized
studiesby the U.S. Army Corps of Engineersat variouslaboratoriesin the Extensive West have produced several general equationsfor snowmelt during (1) rain-free When rain is falling, heattransferby convectionand periods and(2) periodsof rain.2a radiation is of prime importance.Solarradiationis slight, andlong-wave condensation periods rain-free When can readily be determinedfrom theoretical considerations. direct require prevail,both solarand terrestrialradiationbecomesignificantand may rainless during evaluation. Convection and condensationare usually less critical intervals.The equationsare summarizedas follows:2 1. Equations for periodswith rainfall. (coverbelow 10 percent)or partly forested(coverfrom 10 to For open a. 60 percent)watersheds,

- 32) + 0.09 + 0.007P)(7" M : (0.029+ 0.0084fto (over80 percent cover), areas forested b. For heavily M : (0.074 + 0.007P,)(7.- 32) + 0.0s

(14.s3) (14.54)

where M : the daily snowmelt(in./daY) P , : the rainfall intensity (in./day) T o : the temperatureof saturatedair at 10-ft level ("F) wind velocity at 50-ft level (mph) the average t.^ - the basin constant,which includesforest and topographic of the areato exposure average effects,and represents from about 1.0 for clear wind. Valuesof k decrease plains areasto about 0.2 for denseforests ,, Equationsfor rain-free periods. a. For heavyforestedareas, , M:0'074(0.537'"+ O.47fi)

(14.ss)
(14'56)

areas(coverof 60-80 percent), b. For forested

+ 0.7870+ 0.0297" M : k(0.0084u)(0'22r'" areas, c. For partly forested - a) M : k'(I - rx0.00401,)(l r) + 0.78r) + F(0'029 + k(0.0084o)(0.227'"

Q4.57)
)

14.6 SNOWMELT RUNOFF DETERMINATIONS 293 d. For open areas, - a) + (1 -'N)(0.0zIzT: - 0.84) M : k'(0.00s084)(1 + N(0.02gr') + k(0.00S ao)Q.227'" + 0.78Ti) (14.58)

where M, a, k : as previouslydescribed T'" : the difference betweenthe t0-ft air and the snow surface("F) temperatures T'o: the difference between the 10-ft dew-point and ('F) snow-surface temperatures I, : the observedor estimatedinsolation (langleys) a : the observed or estimated meansnowsurface albedo k' : the basin short-waveradiation melt factor (varies , from 0.9 to 1.1),which is relatedto mean exposure of open areascomparedto an unshielded horizontal surface F : the mean basin forest-canopy cover (decimal fraction) Ti = the difference between the cloud-base and (oF) snow-surf,ace ternperatures N : the estimatedcloud cover (decimalfraction) Note that the use of equationsof the type given must be related to the areal extent of the snowcoverif realistic values are to be obtained. Presentmethods of determiningthis are not totally adequate. EXAMPLE 14.4 Use Eq. 14.53to estimatethe snowmeltat an elevationof 3000 ft in a partly forested arcaif the rainfall intensity is 0.3 in./day, the wind velocity is 20 mph, and the temperatureof the saturatedair is 42'F. b. Rework your solution for a denseforest cover and a saturatedair temperatureof 53"F. Solution

- 32) + 0.09 A , M : (0.029+ 0.0084ftr-r + A.007P)(7, - 32) + 0.09 M _ (0.029 + 0.0084 x 0.5 x 20 + 0.007x 0.3)(42
I'tl -

- 32) + 0.05 b . M = (0<074 + 0.007P)(7" M * (0.074 + 0.007 x 0.3x53- 32) + 0.05 M _ 1.65 in./day r:

I.24 in.lday

The Water Budget

The waterbudgetcanbe usedto estimate the snowmeltrunoff from a watershed.2 Such an approach hasparticularmerit for areaswherehydrometeorologic recordsare short. Difflculty with the methodis the usuallack of satisfactory datato quantify the various

294

CHAPTER14

SNOW HYDROI-OGY

properly. A hydrologicbudgetequationfor the earth's surface(Eq. 1.1) components can be written R : P - Z - A S where P : R: L : AS : the grossPreciPitation the runoff the losses the changein storage

(14.se)

For snowmeltcomputationsthis equationis modified somewhat' Grossprecipiiation for a given period P is now definedasthe sum of precipitations in the form of snow P" and rain P,, ot P:P,1-P" This may also be written as P:P,*P":P^+Li where P,: net precipitation L; : intercePtionloss A further refinementyields P: where P,n,P",: L,i, L"i: P-+ L,i+ P"nt L,;

(14.60) (14.61)

(r4'.62)

net rainfall and snowfall, respectively respectively the rain and snowinterception,

Figure 14.10indicatesthe nature of snow interceptionby forestedareas'Additional information on interceptioncan be found in Chapter 3' The total lossI is L : L,i + L,i + L" * Q"where loss L" : the evapotranspiration soil moisture available in the change Q,^: term AS is then given as The storage AS : (l7z - Wr) t Q, $4.63)

04.64)

whereWr,Wr:thefinalandinitialwaterequivalentsofthesnowpack' resPectivelY Qr : the ground and channelstorage Inserting uilo"r for p, L, and AS from Eqs. 14.62-14.64 in Eq. 14.59 gives - L" - Q"^ - (W, - Wt) - Q, R : P,n + L,i + P", + L"i - L"i - L,i and cancelingpositive and negativevaluesof L,i and lr; produces - Q, - L" R = P,n+ P", - (W, - Wt) - Q,-

(14.65) I
(14.66) 1 )

14.6 SNOWMELTRUNOFFDETERMINATIONS

295

4-')

Curv e A
.MC
1\MC o
6 J U

-<

(2) ,SP&
PPR

625 F B ^ ^ o z v
FQ 1<

6,
RF (1) tr RF -Curv B WFM WFM

(4)

PPM tr

o c /rt
:l

,%?r
,/,

r l 0 o
a h 5

./

-10 0 l0 20 30 40 50 60 Canopy density(%) 70 80 90 100

Figure 14.10 Engineers.2)

Snowfall interception loss. (After U.S. Army Corps of

The expression P", - (W, - Wr1represents the snowmeltM; thercforc, R:P,,+M-Q"^-Qr-L" (14.67) If reliable estimates of the terms in Eq. 14.67 canbe secured,the basin discharge R is computable.

Elevation-Band Procedure
Runoff from snowmelton a watershed can be estimatedfrom calculationsof excess watermadeavailtble on a seriesof contributingareas(bands)at variouselevations in the watershed. The practiceis as follows: divide the watershed into severalsubareas or bands;estimatethe quantity of snowmelt,rainfall, and lossesgenerated on each band during a prescribedinterval of time; and usethe weightedsum of thesecontributionsto provide an estimateof the excess wateravailable for runoff. For eachband, it is assumedthat snowmelt, rainfall, and lossesare uniform over the band. The subareas are considered to be either snow-covered or snow-freeand melting or not

296

CHAPTER14 SNOW HYDROLOGY

melting. For eachband, snowmeltis computedusing equationsof the type preSented or historic information, and losses basedon expectations earlier,rainfall is estimated havebeen in Chapterb3 through 5. Ohce theseestimates are estimatedas described made for each band, the following equation servesto provide a weighted value of excess water availablefor runoff from the basin.

)
M _

tte + Mt - Lt)jAt (14.68)

)a,
whereM : snowmelt water availablefor runoff (cmlday), Pr is the rainfall on the 'band, loss,A, is the size of the M, is the snowmeltfrom the band, I, is the subarea subarea,and n is the total number of bands. EXAMPLE 14.5 water Given the data in columns 1-5 of Table I4.2, estimatethe amount of excess method(Eq. 14.68). usingthe elevation-band available for runoff from the watershed 14,5 14.2 DATA FOR EXAMPLE TABLE
(1) Elevation band no.
I

(2)
Subarea slze sq km

(3)
Snowmelt cmid

(4) Rainfall cm/d

(5)

(6)

(7)

Losses cm/d 0.40 0.50 0.70 0.60 0.30 0.10

( 3 )+ ( 4 ) + ( 5 )

(6) x (2)

z
J A

5 6 Totals

230 ))a 289 213 193 167 1316

0.02 0.40 0.60 0.70 0.3s 0.00

0.90 1. l0 1.80 1.90 2.20 "2.40

1,.32 2.00 3.10 3.20 2.85 2.50

303.60 448.00 895.90 681.60 5s0.05 411.50 3296.65

Solution. The solution for the numeratoris the sum bf the productsgiven in is the sum Table 14.2;the solutionfor the denominator column 7 of spreadsheet of the subareas siven in column 2 of the table. : 2.5I cmlday' for runoff : 3269.6511316 The excess-water availabie
II

Recessions Hydrograph
in Chapter 11 and take the generalform curveshavebeen discussed Recession Q: ' where Q : the dischargeat time t Qo : the initial rate of flow k:__afecessionconstant Qoe-o'

(r4.6e)

, : -

RUNOFFDETERMINATIONS 14.6 SNOWMELT

297

o u

Time

hydrograph' of a snowmelt Figure 14.11 Separation permit evaluationof the amountof runoff Studiesof daily streamflowby hydrographs of the derived from snowmelt.The t""hniqutnred is essentiallyone of separation procethe illustrates Figure 14.11(not to scaleandoversimplified) daily hydrographs. fit snowmdlt peaks,respectively, dure. Assumeihat the first, second,and succeeding point A the at a in time, backward days.If the ultimate recessioncurve is extended from recessions between area curveft'om Hydrograph2 will intersectit. The recession Day to attributed melt is the Hydrograph1 and HydiogtipiZ (showncross-hatched) their determine to be studied can 1. In like -unn"r, a seriesofsnowmelt hydrographs By observingsuchhydrographfeaturesas th height to individual melt components. peak X, the height io trough i, andthe form of the recession,volumetric and rate treatmentof this forecastsof snowmeltrunoff can be made.A more comprehensive subjectcan be found in Ref. 25.

SYnthesis Hydrograph
with snowhydrologyareoftwo-types'The ofrunoffhydrographsassociated syntheses of flow distribution is the development kind second The first is a short-termforecist. ' a or for a comPletemelt season forecastingis very helpful in prel of part controls,while the synthesis lating designfloods. To forecasta snowfieldand streamflowneedbe k parameters to havethe reliablepredictionof variousmeteorological ing, it is necessary canbe parameters historic Known conditions. initial in additionto a knowledgeof the satparameters generated or assumed whereas flows usedfor reconstructinghistoric hydrometeorologic common some displays 14.12 Figure isfy designflood syntlieses. data. severalfactors(not of great conceln where In snowmelthydrographsyntheses, First, a drainagebasin with snowconsidered. "utlfully only rainfall exists)mustbe since the areal extent of the system, homogeneous a as cover cannot be accepted

298

CHAPTER14

SNOW HYDROLOGY

6 5 ^ ! 9 > > v

B 5
(h

Short-waveradiationngtesj.^,-- .-

i hsolation obsened by USWB. of Boise Cily' lD 2. Basin albedo of snow surface ^ - = , | L Absorbed shon-wa\e ralialion compukd b) lomula 1ab. ll(

AbsorbedShort-wavefadiation

U
o

9.1 ' H 9 a t o 6

-8
Air Temperature Notes: 1. Veni;al bus show daily range of surface air temperature (max and min) at Idaho City' ID 2. aonnected points indicate daily (0700 hr) 700 mb temperature at Boise CitY, ID

90 80 70 60

t--

s
B

50 o 40

30 a 0.5 120
lrn

u 4 0.4
9 b
q d

gB
E

0.3 0.2

g >
7

2 0 MaY1955

31

20 10 1955 June

Figurel4.l2Hydrometeorologicdataandcomputationofwatergenerated. (AfterU.S.Army Corpsof Engineers.') the contributblanket is highly important. where only snowmeltflows are developed, If rainfall ing areaneel not U"itr" entiredrainage-only that portion with-snowcover' while bare areas from o"-"o., during the snowcoverperiod, contributionscan come may cases in such losses may producecombinedrunoff' The natureof other expanses differ greatly for nonsnowoverlayedand coveredlocations' subThe altitude is an exceedinglypertinent factor in the hydrology of tracts reducgeneral a to due jected to snowfall.Ratesof .no*tnitid"crease with elevation tion in temperaturewith height. orographic effects and the temperature-elevation snowcover relations tend to raise the amount of precipitation with altitude' Greater a result' As rates' melt precipitation and reduced of increased depth occursbecause is apsnowline the as the basin-wide melt and cover-area increase with height

14.6 SNOWMELTRUNOFFDETERMINATIONS

299

proached,then diminish with elevationover the higher placesnormally completely snowcovereduntil late in the season. A snowpack exhibitsanotherimportant trait in relation to rainstorms.In the spring, relatively little runoff occurs from snow-free regionscomparedwith that from a snowfieldfor moderate rainfalls. During very cold weather,the situationduring heavyrains is often reversed, sincea dry snowpackcan retain significantamountsof water. Two basic approachesintroduce elevation effects into procedures for hydrographsynthesis.2 The first dividesthe basin into a seriesof elevationzoneswheie the snowdepth,precipitationlosses, and melt are assumed uniform. A second method considers the watershedas a unit, so adjustmentsare made to accountfor the areal extentof the snowcover,varying melt rates,precipitation, and other factors. To synthesize a snowmelthydrograph,information on the precipitation losses, snowmelt, and time distribution of the runoff are needed.Snowmelt is generally estimatedby index methodsfor forecasting,but in design flood synthesis the heat budget approach, is the most used. Precipitation is determinedfrom gaugingsand historicor generated data.Losses are definid in two wayswheresnowmeltis involved. For rain-on-snowhydrographs all the water is considered a lossif delayedvery long in reaching a stream. This is basically the concept of direct runoff employed in rainstormhydrographanalysis.For hydrographs derivedprincipally from snowmelt, only that part of the waterwhich becomes evapotranspiration, or deeppercolation,or permanentlyretained in the snowpackis consideredto be lost. Assessing the time distribution of runoff from snow-covered areasis commonly done with unit hydrographsor storagerouting techniques. For rain-on-snow events,normal rainfall-type unit hydrographs are applied;for the distribution of strictly snowmeltexcess, special long-tailedunit graphsare employed.Storage routing techniques are widely exeicised to synthesize spring snowmelthydrographs, perhapsdividing them into severalcomponentsand different representative storagetimes. The time distribution of snowmehrunoff differs from that of rainstormsdue mainly to large contrastsin the ratesof runoff generation. For flood flows associated with rainfall only, direct runoff is the prime concern,and time distribution of base flow is only approximated. Big errorsin estimates of baseflow arenot generallyof any practical significance where major rainstorm floods occur. In rainstorm flows, infiltrated water is treated as part of the base flow component and little effort is directedtoward determiningits time distribution when it appearsas runoff. In using the unirhydrograph approachto estimatesnowmelthydrographs, it is customaryto separate the surfaceand subsurface componentsand route them independently. Storagerouting has beenusedextensively for routing floodsthrough reservoirs or river reaches. It is also applicablein preparingrunoff hydrographs. In snowmelt runoff estimates, the rainfall and meltwaterare treatedasinputs to be routedthrough the basin, using storagetimes selectedfrom the hydrologic characteristics of the watershed. Two basichydrologicrouting approaches are relatedto the assumption of ( 1) reservoir-typestorage or (2) storage that is a function of inflow and outflow.These methodswere treatedin depthin Chapter13. Storagerouting techniques that separate runoff into surfaceand groundwater components, assign different empirically derivedstorage timesto each,and thenroute them separately havebeen employed.26 An additional system usesa multiple storage,

300

CHAPTER14

SNOW HYDROLOGY 120

,:

Rn

o 6 U
il

Three 6-hr stages

One 18-hr stage

Time (hr)

storagerouting' Figure 14.13 Example of multiple-stagereservoir-type routstorage reservoir-type iti. ngur" illustratesih" ut" of multiple-storage unit to analogous ttrunn". u in runoff ing for"evaluatingtime distribution of Engineers'2) of Corps tiOrog.uptr. (Af1erU.S. Army

In this method inflow is routed throrrgh two or reservoir-type storagescheduling.2 suchan approach'Any more stagesof storageru"""rtiu!fy. Figure 14.13 illustrates the storagetime and.the desiredtravel time can be obtainedby properly selecting to reflectvarious Retentiontimes betweenstepsmay alsobe varied tf st4ges' of single-stage use that the. clark has suggested drologic characteristics. be^^simplified.27 to rfter translatirrgirp", in time permits Jo=mputations has hydrographs runoff snowmelt synthesizing for method e most practicJd in primarily differs graphs ,,nir hrrrrrnoranh The characterof snowmeltunit 12' chapter in plots' As disclssed time base length fiom that of rainstorm unit single isolated storm events'In from derived are ort"n rainstorm unit hydrograph, snowmeltrunoff,ratesofwaterexcessaresmallandapproximate$continuous'Asa is indicated'2 result, the use of S-hydrographs utility, since it allows (1) adiusthas considerable ri"titoA The S-hydrogrupfr generationrates,(2) adjusting ments to the derivedoni, hydrographfor nonuniform of the areaunder iirn" p"rtoo to; d;ired interval, (3) ready adjustments il;;;t; to get a unit veragingseveralhydrographs

;?ilt1,"1*T,::t1il1""t"1':*"ff methodin adjustingfor nonunirydiograph

form generationrates of water excess' Onceap"r""nrug"S-hyatogtupftisderived'aunithydrographofanydesired are m nig. 14.15.ordinatesof the S-hydrograph periodcanbe obtaineias indlcated

312

HYDROLOGY AND SMALLWATERSHED 15 URBAN CHAPTER havehad wide application;however, of peakflow determination Both categories in applyingthe techniques. two relatively major difficulties are normally encountered First, the rainfall-runoff formulas,suchas the rational formula, aie difficult to apply unless the return periods for rainfall and runoff are assumedto be equal. Also, of coefficientsrequiredby theseformulas are subjectiveand havereceived estimates criticism. The empiric and correlativemethodsare limited in application considerable to they are derivedfrom localizeddata and are not valid when extrapolated because otherregions. methods, The most fundamentalpeak flow formulas and empiric-correlativQ of the and several scene, design the urban ilue to their simplicity,persistin dominating and methods with reader the acquaint most popular forms are briefly describedto 25. in Chapter described are assu-ption*. Urban runoff simulationtechniques

RationalFormula
The rational formula for estimatingpeak runoff rates was introducedin the United Sincethen it has becomethe most widely used by Emil Kuichlingin 1889.18 States Peak method for designingdrainagefacilities for small urban and rural watersheds. flow is found from QO: CIA

(1s.1)

where Qo: the peak runoff rate (cfs) to be dimensionless) C _ the runoff coefficient(assumed rainfall intensity(in./hr), for a stormwith a durationequal I _ theaverage to a critical period of time /" t" : the time of concentration(seeChapter Ii) A : the size of the drainagearea (acres) net rain intensity (in./hr) for a storm with duratiofl: t, cI : the average 1.0 acre-in./hr is because to be dimensionless The runoff coefficientcan be assumed return periods of 5-10-year storms for equivalentto 1.008 ft3lsec.Typical C values are providedin Table 15.1. The rationale for the method lies in the conceptthat application of a steady, uniform rainfall intensity will causerunoff to reachits maximum rate when all parts are contributingto the outflow at the point of design.That condition of the watershed is met after the elapsedtime t", the time of concentration,which usually is taken as At this time, the time for a waveto flow from the most remotepart of the watershed. the runoff rate matchesthe net rain rate. Figure 15.1 graphically illustrates the relation. The IDF curve is the rainfall intensity-duration-frequencyrelation for the areaandthe peakintensityofthe runoff is Q/A: 4, which is proportional to the value of 1 defined at t". The constantof profottionatity is thus the runoff coefficient,C : (QIA)lL Note that QIA is a point yields nothing of the nature of the rest of the value and that the relation, as it stands, hydrograph. The definition chosenfor /" can adverselyaffect a designusing the rational channelvelocity is usedto estimatethe travel time from the formula. If the average most remote part of the watershed(a common assumption),the resulting design

WATERSHEDS 313 FORURBAN 15.2 PEAKFLOWFOHMULAS TABLE15,1 ryPICALC COEFFICIENTS FOR5TO 1O-YEAR FREQUENCY DESIGN
Descriptionof area Business Downtown areas ' Neighborhoodareas Residential Single-family areas Multiunits, detached Multiunits, attached Residential(suburban) Apartment dwelling areas Industrial Light areas Heavy areas Parks,cemeteries Playgrounds Railroad yard areas Unimproved areas Streets Asphaltic Concrete Brick Drives and walks Roofs Lawns; sandy soil: Flat,2Vo Avenge,2-7Vo Steep,TVo Lawns; heavy soil: Flat,2Vo Average,2-7Vo Steep,TVo Flunoffcoefficients

0.70-0.95 0.50-0.70 0.30-0.50 0.40-0.60 0.60-0.75 0.25-0.40 0.50-0.70 0.50-0.80 0.60-0.90 0.10-0.25 0.20-0.35 0.20-0.40 0.10-0.30 0.70-0.95 0.80-0.95 0.70-0.85 0.75-0.85 0.75-0.9s 0.05-0.10 0.10-0.15 0.15-0.20 0.13-0.17 o.r8-0.22 0.25-0.35

Time (min)

Figure L5.1 Rainfall-runoff relation for the rational method.

314

CHAPTER15

URBANAND SMALLWATERSHEDHYDROLOGY

dischargecould be less than that which might actually occur during the life of the project. The reason is that wave travel time through the watershedis faster than average discharge velocity (seeSection 13.1).As a result of using the slowervelocity I{ the peak time (/.) is overestimated, the resultingintensityl from IDF curvesis too small, and the rational flow rate p is underestimated. Rational Method Applications Most applications of the rationalformulain determining peak flow rates utilize the following steps:, 1. Estimatethe time of concentrationof the drainagearea. 2. Estimate the runoff coefficient, Table 15.1. 3. Selecta return period T, and find the intensity of rain that will be equaled or exceeded, once every I years.To produceequilibrium on the average, flows, this design storm must have a locally derived IDF curve such as Fig. 27.I3 or Fig. 15.2usinga rainfall durationequalto thetime of concentration. . 4. Determinethe desired peak flow Q,from Eq. 15.1. 5. Somedesignsituationsproducelargerpeak flowsif designstormintensities for durationslessthan /" are used.Substitutingintensitiesfor durationsless than t" is justified only if the contributingarea term in Eq. 15.1 is also reducedto accommodate the shortenedstorm duration. One of the principal assumptions of the rational method is that the predicted peak dischargehas the same'returnperiod as the rainfall IDF relation used in the
IDF curves for storms in vicinitv of example site

"

0 5 10 15 20

30

40

50

60

70

80

90

100 110 r20

130

Time (min)

Figure 15.2 Intensity-duration-frequency curvesusedin Example 15.1.

15.2 PEAKFLOWFORMULASFORURBANWATERSHEDS 315

prediction.Another assumption, and onethat hasreceivedclosescrutinyby investigais the constancyof the runoff coefficient during the progressof individual tors,re'2o from a list stormsand also from storm to storm. The"coefficientis usually selected surof the drainage capacity infiltration and basedon the degreeof imperviousness : for antecedent is to account if it must vary I,.rf I,the coefficient face. BecauseC moisture,nonuniform rainfall, and the numerousconditions that causeabstractions rainfalls. In practice,a composite,weightedaverand attenuationof flood-producing agerunoff coefficientis computedfor the various surfaceconditions.Times of conof the principal flow centration are determinedfrom the hydraulic characteristics parts, flow and flow in defined overland into two path, which typically is divided obtain /". to are added channels;the times of flow in eachsegment the equation is most is that method Another assumptionwith the rational frequent storms,in the for that exist moisture conditions applicableto antecedent used traditionally of storms interval,representative rangeof the 2- to 10-yrrecurrence frequent less more severe, Because for design of residential storm drain systems. moisture conditions,the rational coefficient is stormsoften have wetter antecedent by multiplying it by a frequencyfactor.The commonly usedmultipliers for increased lessfrequent stormsare:
period (yrs) Return
Multiplier

2-to 25
)U

100

1.0 1.1 t.2 1.25

EXAMPLE 15.I runoff ratesfor the area design and 5O-year to find the 10-year Usethe rational.method 15.2 arc applicable. in Fig. shown shownin Fig. 15.3.The IDF rainfall curves Solution 1. Time of concentration: t,:tt*tz:15+5:20min

At = 3 acres

cr = o'3

tr = Az= Cz = tz =

15min 4acres o'7 5min

Figure 15.3 Hypotheticaldrainagesystem for Example15.1.

316

HYDROLOGY CHAPTER15 URBANAND SMALLWATERSHED

2. Runoff coefflcient: c : [(3 x 0.3) + (4 x 0.7)]lQ + 4) :0.53 for 10-yrevent C : 1.2(0.53): 0.64 for 50-yr event 3. Rainfall intensity-from Fig. 15.2: Irc : 4'2 in'/ht : 5'3 in'/hr 1so 4. Designpeak runoff: Q r c : C I A : 0 ' 5 3x 4 ' 2 x 7 Q s o : C I A : 0 ' 6 4x 5 ' 3 x 7 16 cfs 24 cfs rl

Rational Method Discussion The runoff coefficientin the rational formula is moisturecondition, recurrenceinterval, land on the soil type, antecedent dependent use, slope, amount of urban development,rainfall intensity, surface and channel of and durationof storm.Tablesand graphsgenerallyallow determination roughness, equationscan and regression C from only two or three of thesefactors.Nomographs providerelationsamongmore factors.One suchrelation, applicableonly in the region for which it was derived,is2r 2(0.001CN1 48)0 ts-o't{(P+ I)/zfo j e1-s0 C : j .Z(t1-i)CN3To05[(0.01CN)o

n5.2)
where CN : SCScurvenumber(Chapter4) T : recurrenceinterval ( years)
land slope (7o) average I : rain intensity(in./hr) P : percentimperviousness Yet the a complexhydrologicsystem. The rational formula is a simplemodelto express designby acceptance implying results methodcontinuesto be usedin practicewith results. ers, officials, and the public. The methodis easyto apply and givesconsistent in clear terms demonstrates method the From the standpointof planning,for example, times of because increases surfaces runoff from developed the effectsof development: increase. and runoff coefficients concentrationdecrease is normally askedto estimatethe peak the designer For storm drainagesystems, at leastonce in a given numberof years flow rate that might be equalledor exceeded using the rational (described see Section 10.4). For designs as the frequency to the frequencyof equal is assumed formula, the frequencyof the peak runoff event Studies given duration). in a depth (an rain eventbeing deflnedas some the rain event probabillog-normal cumulative Figure 15.4 shows haveexploredthis assumption.z2 of rainfall and runoff on a 47-acre ity functions (Chapter 26) fitted to observations of 0.44. The imperviousness Surface area in Baltimore, Maryland, with an average and the sequence rainfall observed the to data are partial seriesfitted independently to correspond necessarily not does runoff Thus the largest runoff sequence. observed runoff any between correspondence lack of the largestranked rainfall, and a similar

s:

WATERSHEDS 317 FORURBAN 15,2 PEAKFLOWFORMULAS

Percentage of sample values equal to or greater than indicated value .59998 99 95 90 80 7060504030 20 10 5 2 10.s

Rainfall frequency curve (rr = 7.5 min)

l"

./

!2 k

..t

tt

't'2'
" i/
oo

y"
o

o-'

2
a a

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3

a o

Peak runoff ftequency curve

cd

5 1020
Recurrenceinterval (Years)

Figure 15.4 Distributions of recorded rainfall and runoff' (After Schaake.22)

in and the rainfall that producedit holds for the ranked position of the observations -year of frequency rainfall In Fig. 15.4, the 5 the arraysof the two i"putut" sequences. runoff a indicates theratio of4'0 cfs/acre; io a runofffrequency 6.5in./hr corresponds are eachcloselylogcoefficientof approximately0.6. Although the two sequences that the runoff coefficient increases normal, they tend to converge,which suggests slightly with more intense,iess frequent storms.In the designrange,however,the t"*ttr tend to support the assumptionof the rational method that the recurrence interval of the runoff equalsthe reiurrence interval of the rainfall. It shouldbe noted All IDF in Figs. 15.1and 15.4havesimilarproperties. that the rainfall distributions different many from rainfall intensitiesderived curvesare drawnthrough the average stormsof record; any single IDF curve dses not representthe progressof a single turns to the rational methodto storm.For lack of historicalrunoff data, the designer construct from the rainfall history what amounts to a runoff intensity-durationrelation. frequency by to be caused The most critical (highestpeak) runoff eventis often assumed If the watershed' the a storm havinga duration -qual to the time of concentrationof rainfall IDF curve is steepin the designrange,severaldurationsshouldbe testedfor the given frequency to assurethat no other storm of equal probability producesa higlier peak runoff iate. Most applicationsof the rational method do not includethis with the other thai ihe peak occursat /" is commensurate the assumption testbecause inherent assumptions.

31 8

CHAPTER15

URBANAND SMALLWATERSHEDHYDROLOGY

The rational method is used in the designof urban storm drainagesystems servingareasup to six hundredacresin size.For areaslarger than I mir, liydrograph or other techniques are generally warranted.Considerabl"3udg-"nt is riquir-eAin selecting both the runoff coefficientsand times of concentratloi. a common procedure is to selectcoefficientsand assume that they remain constantthroughout the storm.As the designproceeds from point to point downstream, a compositeweighted C factoris computedfor the drainage areaaboveeachpoint. The time of concentration is composed of an inlet time (the overlandand any channelflow timesto the first inlet) plus the accumulated time of flow in the systemto the point of design. Figure 15.5 is an exampleof a designaid for prJdicting overland flow times. calculation of flow time in stormdrainscan readily be estimatid knowing the type of pipe, slope,size,and discharge.23 Generally,the pipe is assumed to flow full foi this calculation.(see Fig. 15.6.) Nomographs also are available to solvethe Manning equationfor flow in ditchesand gutters.The estimationof inlet time is frequentl| basedsolelyon judgment; reportedvaluesvary from 5 to 30 min. Denselyaevitopei areaswith impervioustractsimmediatelyadjacentto the inlet might be assigned inlet periods of 5 min, but a minimum value of 10-20 min is more uiual.

d o

10n
"-"

4U

Fo F

Figure 15.5 Surfaceflow time curves.(After FederalAviation Agency.23)

15.2 PEAKFLOWFORMUI.AS FORURBAN WATERSHEDS 3.I9

1,500 1,000 800 600 500 400 300 200 2,400 2,000

1,500 1,000 800

0.004 0.005 0.006 0.007

0.4 0.3 o.2

0.7 0.8 0.9

s00
400

600

1.0

300 200

0.1 0.08 0.06 0.05 0.04 0.03 0.02 0.01 0.008 0.006 0.005 0.004 0.003 0.002
V) tl o o

6 cd bo

100 80 60 50 40 30

100 80

60 50 40
)n

po

10 8 6 5
i

s-o.oto r, -0.008 0.006

a

0.10 0.08 0.06 0.05 0.04 0. 0.03 0.1 I , o,o2 0.015 3 g

h + a oo

3
a

10 8 6 5
A

0.001 0.0008 0.0006 0.0005 0.0004 0.0003 0.0002 0.000r 0.00008 0.00006 0.00005 0.00004 0.00003 0.00002

2 c
boo

3 2

: 8
o

1.0 0.8 0.6 1.0 0.5 0.8 0.4 0.6 0.3 0.5 0.4 0.2 0.3 0 . 1 3 0.2

o h o

0.00001 0.000008 0.000006 0.000005 0.000004

Figure 15.6 Flow in pipes (Manning's formula); (After Ref. 24.)

Most designers applyingthis methoddo not usethe time of concentrationin its strictestsense; rather, the largestsum of inlet time plus travel time in the storm drain systemis taken as the time of concentration.Caution is required in app$ing the method' Peak discharge is not the summationof the individual dischapges, b"Jarr* peaks from subareas occur at different times. The runoff from subareas should be recheckedfor eacharea under consideration. The average intensity / is that for the time of concentration of the total area drained. While I decreaiesas the design ploceeds downstream,the size of the contributing area increasesand normally e

320

CHAPTER15

URBANAND SMALLWATERSHEDHYDROLOGY

It shouldbe noted that the designat eachpoint downstream continuously. increases is a new solutionof the rational method.The only direct relation from point to point derivesfrom the meansfor determiningan incrementof time to be addedfor a new time of concentration.The effect is to provide an equal level of protection (i.e., an Example 15.2 is reproequal frequencyof surcharging)at all points in the system. of the rational the application illustrate to references duced from standarddesign method to an urban sYstem.2a EXAMPLE 15.2 of Fig. 15.7a,determinethe outfall discharge. Basedon the storm sewerarrangement Assume that C : 0.3 for residentialareasand C = 0.6 for businesstracts. Use a a minimum 20-min inlet time. 5-yearfrequencyrainfall from Fig. 15.7b andassume Solution. The principal factors in the designare listed in Table 15.2' Additional columns can be provided to list elevationsof manhole inverts, sewer This information is helpful in checkingdesigns inverts, and ground elevations. use in drawingfinal designplans.(see Table 15.3.) lI and for subsequent 'orational" in that thepeak Modified Rational Method Therationalmethodis truly flow rate is simply set equal to the net rain rate after sufflcient time occurs for the to contributerunoff. This resultsfor any storm equallingor exceedentire watershed

design,requiring volume of runoff as well as peak flow rates.

IN TABLE15.3 OF COLUMNHEADINGS TABLE 15.2 DEFINITION Column Comment Line being investigated Inlet or manholebeing investigated Length of the line of the inlet Subarea Accumulatedsubareas Value of the concentrationtime for the area draining into the inlet Travel time in the pipe line WeightedC for the areabeing drained Rainfall intensitybasedon time of concentrationand a 5-yearfrequencyculve Uqitrunoff q: CI Accumulatedrunoff that must be carried by line Slopeof line Size of pipe Pipe capacity Velocity in full pipe Actual velocity in pipe

1 4 5 6 7 8 9 10 t1 T2 t3 14 15 t6 t7

15.2 PEAK FLOW FORMULASFOR URBANWATERSHEDS

o)

I
u\-q

_ 90lt
t
Legend

l
numbers Sewerlinewith manhole Storminlets Gutterflowline

MH1-1 r ,\e / --

contourline - areaoutline Drainage

\
a 5

N
\
J

2:i-yr averagefrequencY
l l r l ! l l

\ {

10-yr averagefrequencY
r

Ij-'ff:::i::::?

K:---t.-

10 20

30

40

50 60 70 80 Duration(min)

90

100 110 120

Figure 15.7 Sample storm drainage problem: (a) typical storm rainfall seier design plan anO (U) intensity-duration-frequency (After 24.) Ref' Iowa. for Davenport, curves

= E

pue lMo'l !y

6 r +.i6
6 A

o r 6ioi
6

n h oioi
@ @

n dt

o ;

puo roddnE
E o C o

1 q o 9 . l n
6 + i O 6 9 6 N O

\ \
6

N 6

! v ?
O @ @

puo JeMo'l di

? - : < : < x = x , x : V ( ] Y Y Y Y - ] : : c l o O O d + F t
O @ @

= o

puo reddnS
(g)]erresut tPl I (g)dorppenul"toqu"N $ qoqu"yl $ (4) sassol (u)^6reuaplol I

E X EE < i i
6 6 h 6 i O i @ o r

E q
* o @ @

? E qE q E q E f r
6 o i < l N : o o o

. n n

. q q

R -

=
c a o

( u0lrop;o qrdeOI (U)peeq,qcoten I (sdD',{lPoPn S

',qcoPn 11n1 $ 'r$cedeg 11n1 !

6 = = s :

F - R 3

\ O @

O \ r

6 r

A \ O

+r;ri
o - 6 $ n $

rid

d+
* : N

ris;
r

n n n

o c J c . ln n d ; 6 h =
N ! f , * - N i N N + O i

qcl S N S s

qq B H
N N

E
J

( uDrelauelc I (%)remes to adols U (sle)llounl lElol 3 ' (arceTsp) #orng $ (rq/'uI) ttBluleuI gounra6ere,rY @ lualclllooc
o
E ^
N

h h h r + d o o

o O o - i o

O O ' + O N 6 o 6 o o o d

k t
ul
T F t

Eq 3 s q E sq q N n + g s 3
i i o a i

- q n ' 1 q . ' l \ 9 n \ 9 q \
0 0

r a c j c ' l
0 0

ua z

6 6 0

0 0

F } E
t!

+ A+ 3 3 $ 3 3 A 3 A : : : : : . i J
n q c ' i ' 1 - ' 1 n
C Sq
O + N d N C ] +

u1 @ uolpes pualeddnoI F teloJ. 0 @ luer.uetcul (D q$uer I

i . ; N N

"

- :
O + N N N N N

09
6 O N N

3 uJ
o

6 6

E o
6

h O h O 9 c . . t. . 1 ' l q N Q N * i

qrl
d

*
N T \o$

6
J

qqo9
N O O

nn

qv-) oqn
o c l X X s s
O h d N a

n \ o

o E
f

: Y : X X + + < J ' * t h r
i i *

e
lf,'

oro
urorl O

o t
o

* t

o l

N l

d
o 3 N i *

o
g E

0)

lrl J
F

Y T i 1 J T l I T
: i : o * * * o o * N : i

eun e

W)

323 WATERSHEDS FOR URBAN FORMULAS FLOW 15,2 PEAK a full hydrographis developqdrather than In the modified rational method., the following reasoning.If the storm using peak rate, flow the simply estimating runoff rate would dse to the rational the concentration, of time duration exceedslhe At that point, runoffrates ceases. rain net until constant stay peak then value, formula Ifthe rainfall-excess basin. the from released is rain excess zero as to would decrease the hydrograph concentration, of time to the (see is equal 11) Chapter releasetime : t",femaining flat peak t al the to rising shape trapezold approximate would havean D I t". until/: therainduration,D,andthenfallingalongastraightlineuntilt: rational modified the incorporate hydrology urban packages for Many software method for hydrographanalysis.the method is approximateand shouldnot be apover 50 acresin size. plied to watersheds

SCSTR-55Method
for estimatingrunoff volprocedures Servicedeveloped The U.S. Soil Conservation They are known collectively as from urban areas.zs ume and peak ratei of discharge TR-55und indiuidually as thegraphical method,chart method, andtabular method. to urbanconditionsby increasin NEH-426 The threemethodsadjustrural procedures ing the curve number CN foi impervious areas and reducing the lag time /1 for and channel improvements.Allowances are also made for various imlperviousness the first two The SCS designed slopes,and times of concentration. shapes, watershed complete synthesizing for methodsto be usedfor estimatingpeak flows,and the third up to (used watersheds small for The tabular method and chart method hydrographs. the evolution explain help to here aredescribed wererevisedin 1986,21but 2000 acres) for use with 24-hr storms'Use with other of the methods.All three were developed storm durationsis not advised. up to 20 mr2 watersheds, for homogeneous The graphicalmethodwasdeveloped curve runoff the by represented in size, on which the land use and soil type may be variable third a is simply number. As shownin Chapter4, the runoff curve number in a graph of rainfall versusrunoff. Tie SCS peak dischargegraph shown in Fig. 15,8 is limited to applications (see where only the peak flow rate ii Oisired for 24 hr, Type-II storm distributions expethunderstorm the 24-hr of Chapter f 6l. A Type-II storm distributionis typical from rienced in all staiei exceptthe Pacific Coaststates.Figure 15.8 was developed Chapter in described model ofine SCSTR-20 eventsimulation numerousapplications time of concentrationin hours is enteredinto 24. To apply Fig. 15.8, the watershed per inch of net rate in cfs/mi2of watershed ttre grapir'to prJdu." the peak discharge 24-hr gross the from rain during tk Z+-nr period. The 24-hr net rain is estimated 4. in chapter amountusingthe scS curve number approachdescribed usingFig. 15.9.Oncethe composite the effectof urbanizationcanbe estimated for the previousarea,a modifiedcurvenumber curvenumber(CN) hasbeenestimated by enteringFig. 15.9 with the value of the percentimperviousareaon is determined to the CN for r"ading vertically to the curve corresponding the modified watershed, the modified determine to the pervious watershed,and then reading horizontally rain depth net the in determining compositerunoff curve numberthat would be used for the urbanizedwatershed' of the tabular Useof the 1975graphicalmethodis restrictedby the assumptions method.This tabular of the case method.The methodii a -ompositeof resultsfor one

324

CHAPTER15

URBANNNO SUNU-WATERSHEDHYDROLOGY

'e 70n Ei 500 qF 400 E 300

o P,O R

E o

200

100 0.1

0.2

0.3 0.4 0.5 0.7 1.0 (hr) Time of concentration

2.5

3.0 4.0 5.0

(cfs/mi2lin.) Figure 15.8 Peakdischarge of runoff versus time of con(AfterU.S.SoilConsercentration /. for 24-hr,Type-IIstorm distribution. vation Service.25) restricts its applicationsto runoff volumes greaterthan about 1.5 in. (if the curve numberis lessthan 60). Time of concentration shouldrangebetween 0.1 and 2.0hr, and the initial abstractionshould not exceedabout 25 percent of the precipitation. The chart method allows determinationof peak flows for 24-hr Type-II storms over watersheds having a fixed length/width relation and no ponding areas.Three chartsareusedfor flat, moderate, or steep slopes of approximately I,4, or 16percent. Tablesof adjustmentsfor intermediateslopesare provided in the technicalrelease. Severalmicrocomputersoftwarepackages for urbanhydrologyhavebeendeveloped.28 Over two-thirdsarebased on SCSprocedures, but cautionshouldbe applied

80
L) e .E

7n

u 6 0

l0

20

30

40 60 70 50 area(7o) Connected impervious

80

90

100

Figure 15.9 Percentage of impervious areas versus composite CNs for given pervious area CNs. (After U.S. Soil Conservation Service.2s)

325 WATERSHEDS URBAN FOR FORMULAS FLOW 15,2 PEAK that the commercialprogramsfully imitate TR-55 or other SCS handin assuming would carry would includeall three methods, An ideal TR-55 package book methods. incorpowould and limitations, and would stateall assumptions SCS endorsement, of percentage for peak coefflcient,percentimperviousness, rate all SiS adjustments slope. and length/widthratio variations, channelimproved,pondingor swampyareas, Its use shoutAako be cautionedfor other than 24-hr stormshaving a Type-II SCS not adheringto these limitations would not be qualified as distribution. Packages TR-55 procedures. is the use A significantproblem in someof the commercial softwarepackages for produce hydrographs of a trianlular-shaped unit hydrographfor convolutionto the to conceptualize stormsof-various durations.The SCS used a triangular shape use of other but has never endorsed peak flow rate of a curvilinear unit hydrograph, hydrg; tabulated in Section I2.5 ot the than either the curvilinear shapediscussed guide2e published a graphsgiven in the TR-55 manual.For further reading,the SCS in the original technical for the useof the 1975TR-55 intendedto clarify procedures release.

SCSTR-55Method Prevailing
for use. is recommended ratherthan the 1975version, The 1986editionof TR-55,27 with the original . It incorporatesseveralyearsof resultsof researchand experiences edition. The revisionsinclude the following:

1. Three additional rain distributions(seeFig. 16'17). ,, Expansionof the chapteron urban runoff curve numbers' 3. A procedurefor calculatingtravel times of sheetflow' 4. Deletion of the chart method.
Modifications to the graphical peak dischargemethod and tabular hydrographmethod. 6. TR-55 computerProgram.

Tablet5.4 andFig.15'10 Ratherthanrelyingtotally on Fig 15.9,the newTR-55uses indicated in the table' instances to provide urban runoff curve numbersfor certain the 24-hr designrain and number For the new graphicalmethod,an urban curve the SCSrunoff from is determined 1, ih"n un initial abstraction depth are estimated, linear interpofrom found is peak flow equation(Chapter4) or from Table 15.5.The on the rainfall depending 15.I4, lation of the curvesin Figs. 15.ll,15.I2, !5.13, or curves,the the outside falls ratio distributiontype (Fig. rcn).If the computedI"f P ponds or of pefcentage a nearestcurvi should be used. If the watershedcontains 15'6' from Table coefficient swampyareas,the peak flow is multiplied by a reduction EXAMPLE 15.3 concentration,CN: 75 A 1280-acreurbanTennesseewatershedhasa6.0-hrtimeof tainis 6'0 in' The25-year,24-ht of the areais ponded. from Table15.4,and5 percent Find the 25-yearpeak discharge.

TABLE 15.4 RUNOFFCURVE NUMBERSFOR URBANAREAS (see Sec. 4.9 foT other values)

Cover descriotion Average percent imperviousareaD

numbers Curve for hydrologic soilgroup' A

B

Cover type and hydrologic condition

c
86 79 74
98

Fully developed urban areas (vegetationestablished) Open space(lawns, parks, golf courses,cemeteries, etc.)" Poor condition (grasscover <50%) 68 Fair condition (grasscover 50-757o) 49 Good condition (grasscover > 757o) 39 Impervious areas Pavedparking lots, roofs, driveways,etc. (excluding right- of-way) 98 Streetsand roads Paved;curbs and storm sewers(excluding right-of-way) 98 Paved;open ditches(including right of-way) 83 Gravel (including righrof-way) 76 Dirt (including right-of-way) 72 Westerndeserturban areas Natural desertlandscaping(pervious areas only)' 63 Artificial desertlandscaping(impervious weed barrier, desertshrub with 1-2-in. sandor gravel mulch and basin borders) 96 Urban districts Commercial and business 89 85 Industrial 81 72 Residentialdistricts by average lot size 77 65 f acre or less(town houses) j acre 61 38 57 30 I acre 25 I acre I acre 20 5l 2 aqes l2 46 Developing urban areas Newly gradedareas(pervious areasonly, no vegetation)" 77 Idle lands (CNs are determinedusing cover types similar to thosein Table 4.7).

79 69 6l 98 98 89 85 82 77 96 92 88 85 75 72 70 68 65

89 84 80

98 92 89 87 85
96 94 91 90 83 81 80 79 77

98 93 9r 89

96 95 93 92 87 86 85 84 82

86

91

'Average runoff condition, and 1" : 9.25. 'The averagepercent impervious area shown was used to developthe composite CNs. Other assumptionsare as follows: impervious areas are directly connected to the drainage system,impervious areas have a CN of 98, and pervious areas are consideredequivalent to open spacein good hydrologic condition. CNs for other combinations of conditionsmay-becomputedusingFig. 15.9 or 15.10 " CNs shown are equivalent to those of pasture. Composite CNs may be computed for other combinations of open spacecovol type. dComposite CNs for natural desertlandscapingshould be comppted using Fig. 15.9 or 15.10 basedon the impervious areapercentage(CN : 98) and the pervious area CN. The pervious area CNs are assumedequivalent to desert shrub in poor hydrologic condition. eComposite CNs to use for the designof temporary measures during grading and construction should be computed using Fig, 15.9 or 15.10basedon the degreeofdevelopment (impervious areapercentage)and the CNs for the newly graded pervious areas. : U.S. Soil Conservation Source Service,2T

0.0

n5

> o

r.o I
I

70

60

50
Total impervious arca (Vo)

Composite CN

Figure 15.10 Graph of 1986 TR-55 composite CN with unconnected imperviousarea,or total imperviousarea,lessthan 30 percent.(After U.S. Soil . ConservationService.2T) TABLEls.s /aVALUES FORRUNOFF NUMBERS CURVE Curve number 40 4l 42 43 44 45 46 47 48 49 50
)l

L fin.)
3.000 2.878 2.762 2.65r 2.545 2.444 2.348 2.255 2.167 2.082 2.000 1.922 r.846 1.774 1.704 1.636 1.571 r.509 1.448 1.390 1.333 1.279 1.226 I.t75 t.t25 1.077 1.030 0.985 0.941 0,899

number Curve 70 71 72
t5

L (in.) 0.857 0.817 0.'778 0.740 0.703 4.667 0.632 0.597 0.564 0.532 0.500 0.469 0.439 0.410 0.381 0.353 0.326 0.299 0.273 0.247 0.222 0.198 0.174 0.151 0.128 0.105 0.083 0.062 0.041

74
IJ

52 53 54 55 56 57 58 59 60 6l 62 63
o+

65 66 67 68 69

76 77 78 79 80 8l 82 83 84 85 86 87 88 89 90 9l 92 93 94 95 96
o1 98

Source:U.S. Soil Conservation Service.

328

HYDROLOGY CHAPTER15 URBANAND SMALLWATERSHED

300

u o P0 R
I

tnn

3 S

100 R o 60

40 01

0.6

0.8 l

Time of concentration, Z; (hr)

(q*) for SCS Type-I rainfall distribution. (After Figure 15.11 Unit peak discharge Service.) U.S. Soil Conservation

9
Po

100
Ro

ci

60

30 0.1

0.2

0.4

0.6 0.8 1

810

Zr (hr) Time of concentration,

(q,) for SCSType-IA rainfall distribution.(After Figure 15.12 Unit peakdischarge Service.) U.S. Soil Conservation

15.2 PEAK FLOW FORMULASFOR URBAN WATERSHEDS 1000 800 600 500 400

329

^ E
o 90

{, s 300 2oo
6 d

= P 1oo
80 60 50
0.1

0.2

0.4

0.6 0.8 1

810

Z" (hr) Time of concentration,

(q,) for SCSType-II rainfall distribution. (After Figure 15.13 Unit peak discharge Service.) U.S. Soil Conservation

700 600 500 400 < 300

-*
d E
E

,nn

= 5 roo
80 60 40 0.1

0.2

0.4

0.6 0.8 1

810

Time of concentration, Z, (tn)

(q,) for SCSType-III rainfall distribution.(After Figure 15.14 Unit peak discharge Service.) U.S. Soil Conservation

330

CHAPTER15

URBANAND SMALLWATERSHEDHYDROLOGY

TABLE 15.6 ADJUSTMENTFACTOR(Fp)FOR POND AND SWAMP AREAS THAT THE ARE SPREADTHROUGHOUT WATERSHED Percentageof pond and swamp areas

0 0.2 1.0 3.0 5.0

Service. Source:U.S. Soil Conservation

1.00 0.97 0.87 0.75 0.72

From Solution. From Fig. 16.17, the Type-II storm appliesto Tennessee. q,: csm/in. 96 I":0.667.Thus I,fP: 0.11.FromFig. 15.13, Table15.5, From Chapter4, tberunoff from 6.0 in. is 3.28 in. Since5 percentof the area usingTable 15.6,giving 4 : 0.72. Thus is ponded,the peak flow is adjusted g : (96 csm/in.)(3.28 in.)(2.0 mr')(0.72): 453 cfs rr only. If a hydrographis needed The graphicalmethodprovidespeak discharges shouldbe used.The event subdivisionis required, the tabular method2T or watershed is very'complexor a higher simulationmodel TR-20 shouldbe usedif the watershed degreeof accuracyis required (seeChapter24). Assumptionsof the graphicalmethod include: The method shouldbe used only if the weighted CN is greaterthan 40. The ?i valueswith the method may range from 0.1 to 10 hr. that is, describable The watershedmust be hydrologicallyhomogeneous, by one CN. Land use, soils, and cover must be distributed uniformly throughoutthe watershed. The watershed may haveonly one main streamor, if more than one, the branchesmust havenearly equal times of concentration' The method cannotperform channelor reservoirrouting. that are not on the The Fofactorcan be appliedonly for pondsor swamps flow path. estimatedby this method will be reducedif Accuracy of peak discharge I"fP va\uesare usedthat are outsidethe range given. When -this method is used to develop estimatesof peak dischargefor usethe sameprogedure presentand developed conditionsof a watershed, for estimating[. Both the graphicalandtabular methodsare derivedfrom TR-20 output.The use within the scopeof the curves of I permits them to be usedfor any size watershed watershedthat is or tables. The tabular method can be used for a heterogeneous for the subwaterHydrographs subwatersheds. dividedinto a numberof homogeneous shedscan be routed and added.

15,3

PEAK FLOW FORMULASFOR SMALL RURAL WATERSHEDS

331

and is not detailedhere. in the technicalrelease The tabularmethodis described employed: are In using the method, the following steps and 1. Subdividedthe watershedinto areasthat are relatively homogeneous reaches' haveconvenientrouting in squaremiles' 2. Determinedrainageareaof eachsubarea procedurefor estimatingI is The in hours. 3. Estimate T"for eachsubarea outlinedinTR-55. 4. Find the travel time for eachrouting reachin hours' 5. Developa weightedCN for eachsubarea. 6. Selectan appropriaterainfall distribution accordingto Fig. 16.17. frequency(Chapter 16). e the 24-hr rainfall for the selected 7. Determin from CN andrainfall (Chapter4)' 8. Calculatetotal runoff in inchescomputed 15.5. from Table 9. Find I,fot eachsubarea selectone of the hydrographs 10. Usingthe ratio of I,f P andT,for eachsubarea, in TR55. tabulated 11. Multiply the hydrographordinates(csm/in.) by the area (mi2) and runoff subarea. (in.) of eachrespective 12. Route and combinethe hydrographs. The SCS recommendsthat TR-20, rather than the tabular method,be used if any of the following conditions apply: Travel time is greaterthan 3 hr. f is greaterthan2hr. differ by a factor of 5 or more. Drainageareasof individual subareas by SCSin a computerprogram.Copies havebeenincorporated The TR-55procedures are availablefrom the U.S. National TechnicalInformation Service.

'

FORSMALLRURALWATERSHEDS 15.3 PEAKFLOWFORMULAS Method SCSTP-149

TR-20 is the unit-hydrograph TR-55 is the SCS procedure for urban watersheds, (see 24), andTP-149wasdevelChapter procedurefor larger agriculturalwatersheds (5-2000 acres)agricultural small from rates peak flow bped to allow esiimation of of a peak discharge the which from 42 charts of It consistsof a series watersheds.3o 24-hr ruinfall can be determined. watershedslope, storm Input to the procedure is the drainage area, average and depth of rainfall. number, curve (I composite distributiontype or II), watershed TP. Shownare type-I in the charts numerous the Figures15.15ind 15.16illustrate : for both. Similar 70 CN with watersheds, sloped and type-Il curvesfor moderately of TP 149 Applications 15.7. Table given in for the combinations chartsare available 15.7, of Table increments 5-unit the than other havingcurve numbers to watersheds or arithmetic by be accomplished percent, can 16 4, or or for slopesother ihan I, values' chart adjacent logarithmic interpolationbetween

332

CHAPTER15

URBANAND SMALLWATERSHEDHYDROLOGY

h\or@e

g ?3eR3=

^^^ e H =HEF==

'/,
1000 800 7N 600 500 400 300 200
o P!

,<\ 7
l
r!

,4

1000 800 700 600 500 400 300 200

.E

rno

'z ..\"\
L'//
l
'//.2

2\

8 0 & 7 0 3 a o 50 40 30

W
L

& t/,, .M
7

100 80 70 60 50 40 30

t0 8 ,7 6 5
h

ffi
/ttl
9 r @ o o o o - N - + o n o o o o r @ o
O

a
v'

20 l$

H
l/,v ,/: R
O

10 8 7
o

) g?fiFAt
O O O O O O O

8
N

Drainage area (acres)

-tP-I49 peak Figure 15.15 ratesof discharge for small watersheds, Type-I storms:24-hr rainfall, moderate slopes, and CN : 70. (After U.S. Soil ConservationService, "A Method of EstimatingVolumeandRateof Runoff in Small Watersheds," U.S. Depaftment of Agriculture,Jan. 1968.)

TABLE 15.7 CHARTSAVAILABLEIN TP-149FOR PEAK FLOW RATESOF SMALL WATERSHEDS

Storm distribution rype

I,il

Slopetype Flat, TVo Moderate,4To Steep,167a

Slope range (%)

Curvenumber, CN 60,65,70,75,80,85,90 60,65,70,75,80,85,90 60,65,10,75,80,85,90

LU
I,il

0-3 3-8 8-30

333 RURALWATERSHEDS FORSMALL FORMULAS FLOW 15,3 PEAK

nre
1000 800 700 600 s00 400 300 200

R se8sRsc

R ==FeF=- F

;90
E IUV

8 0 r.1 70

fi o o
50 40 30

100 80 70 60 50 40 30 20

10 8 7
o 5

10 8 7 6 5

nF-=

3?38RAE
Drainage area(acres)

= = x = = x x X ; + E 6 F d 6

Type-II storms: Figure 15.16 TP-149 peak rates of dischargefor small watersheds, "A 24-hr rainfalT.moderate slopes,CN : 70. (After U.S. Soil ConservationService, U.S.Department Method of EstimatingVolumeand Rateof Runoff in Small Watersheds," of Agriculture'Jan.1968.)

EXAMPLE 15.4 Comparethe peakflow ratesfrom Type-I and Type-II stormsusingFigs. 15.15and and all other conditionsare equal. that only stormtype changes 15.16.Assume with CN : T0lesultsin on a watershed Solution. A 4-in. rain over200 acres : II Qo : 52 cfs for a Type-I storm (Fig. 15.15) and Qp 9l cfs for Type in (Fig. 15.16).Thus the storm distributiontype makesa significantdifference results of peak flow estimationusing SCS techniques. I r

334

CHAPTER15

URBANAND SMALL WATERSHEDHYDROLOGY

FederalHighwayAdministrationscs PeakFlow DesignMethod

The FederalHighway Administration (FHWA) lists in their HydrologicEngineering ( 1995Ed.) a procedurefor estimatingpeak flow ratesfor Circular No. tq, F{ydrology havingtimes of concentrationbe,-ul1-to--"dium sizedwatersheds homogeneour, equationthat has coefficients regression SCS an tween 0.1 and 10 hours.It employs types and ratios of the initial distribution rainfall determinedfrom data on different in metric The peak discharge P. precipitation, total (see 4) and Chapter 1, abstraction units is calculatedfrom
Qp: 4,AQ,

( 1s .3)

A is the drainageareain sq. km., Q is the in m3/sec, where qois the peak discharge from net rain depth in cm, andq, is the unit peak discharge
log qu: Co + Cl log /" f Crlogz t"

(1s.4)

coefficientsare in which /" is the time of concentrationin hours,and the regression obtainedfrom Table 15.8.
FOR FHWA HEC-19SCS TABLE 15.8 COEFFICIENTS PEAK DISCHARGEMETHOD Rainfall

rype

t,/P 0.10 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.10 0.20 0.25 0.30 0.50 0.10 0.30 0.35 0.40 *0.45 0.50 0.10 0.30 0.3s 0.40 0.45 0.50

co 2.30550 2.23537 2.18219 2.10624 2.00303 1.87733 t.76312 1.67889 2.03250 r.91978 1.83842 1.72657 1.63417 2.55323 2.46532 2.41896 2.36409 2.29238 2.20282 2.473t7 2.39628 2.35477 2.30726 2.24876 2.17772

c1 -0.51429 -0.50387 -0.48488 -0.45695 -0.40769 -032274 -0.15644 -0.06930 -0.31583 -0.282t5 -0.25543 -0.t9826 -0.09100 -0.61512 -0.62257 -0.61594 -0.59857 -0.57005 -0.51599 -0.51848 -0.51202 -0.49735 -0.4654r -0.41314 -0.36803

c2 -0.11750 -0.08929 -0.06589 -0.02835 0.01983 0.05754 0.00453 0.0 -0.13748 -0.07020 -0.02597 0.02633 0.0 -0.16403 -0.1t657 -0.08820 -0.05621 -0.0228r -0.01259 -0.17083 -0.13245 -0. l 1985 -0.1 1094 -0.11508 -0.0952s

IA

ilI

Source: Afrer U.S. Federal Highway Administration' Hec-19, Hydrology' FHWAIP-95, 1995.

15.3 PEAK FLOW FORMULASFOR SMALL RURALWATERSHEDS

335

The procedurehas the following limitations: ' use with homogeneous (CNsfrom zone to zone shouldnot differ watersheds CN shouldbe greaterthan 50 /. shouldbe between0. 1 and 10 hours I^/P shouldbe between 0.1 and 0.5 /" shouldbe about samefor any of the main channels, if watershed has more than one main channel . no channelor reservoirrouting is allowed . no storagefacility on main channel ' watershedareain storage ponds and lakes shouldbe lessthan 5 percent . . . '

bYs)

SyntheticUnit-Hydrograph PeakRateFormulas
Peak flow rates from small watersheds can also be determinedusing the synthetic unit-hydrographtechniquesdescribedin Chapter 12. A storm having a duration definedby Eq. 12.22wirl produce,accordingto Snyder'smethodof synthesizing unit hydrographs, a peak discharge for 1.0 in. of net rain given by Eq. 12.17,or

o^ '

64oct'A
t m

( 15.s )

Similarly, the peak flow rate resulting from a storm with duration D given by Eq. 12.22or 12.23is, according to the SCS methodfor constructing synthetic unit hydrographs, equal to

484A ^ U^: -

(15.6)

where /o is the time from the beginningof the effectiverain to the time of the peak runoff rate, which by definitionis the watershed lag time plus half the stormduration. Both of Eqs. 15.5and 15.6 apply to 1.0 in. of netrain occurringin the durationD. Either can be multiplied by P"", for other storm depthswith equal durations.Peak flows for stormswith durationsother than D would need to be determinedby unithydrographmethods.

Discharge-Area and Regression Formulas

A multitude of peak flow formulas relating the discharge rate to drainageareahave beenproposedand applied.Gray3rlists 35 suchformulas,and Maidment32 compares many others.Most of theseempiric equations are derivedusingpairsof measurements of drainagearea and peak flow rates in a regression equationhavingthe form where Q: CA* : peak the discharge associated with a given return period Q A : the drainage area C, ffi : regression constants (15.7)

Popular discharge-area formulas in the form of Eq. 15.7 include the Meyers equation33 ( 1s.8 ) O : 10,000405

336

CHAPTER15

URBANAND SMALLWATERSHEDHYDROLOGY

where A : the drainagearea,which must be 4 mi2 or more Q : the ultimate maximum flood flow (cfs) This examplegives only one flow rate of unknown frequencyand is chosenonly to illustrate the form of flood flow equations.A program of determiningflood magnibasishas been completedby the on a state-by-state tudesfor atange of frequencies in Chapters26 and 27 and discussed techniques USGSusing the multipl- regression are availablefrom the formulas illustrated for Virginii in Problem 27.25. Similar for the equations regression Softwarecontainingall the USGS USGSfor other states. Highway Federal and United Statesis availablefrom the U.S. GeologicalSurvey Administratioh as part of the HYDRAIN softwarepackage'

Method SurveyIndex-Flood Geological U.S.

in Section27.4 is a graphSurveyindex-floodmethoddescribed The U.S. Geological peak dischargerates. The ical regional c&relation of the recurrenceinterval with are outlined in Section stepsinvolvedin the derivationof a regionalflood index curve the mean is to determine to a watershed 27.4.The first stepin applyingthe technique years. period 2.33 of annual flood, defineOas ttre flood magnitudehaving a return equations regression from are found Mean annual floods for ungaugedwatersheds and on flood magnitudes the USGSreport3a similar in form to Eq. 15.7.For example, in Nebraskagives,in cfs, frequencies
Qz'zt = CA0T

(1s.e)

where A : the contributing drainage arcain mi2 obtainedfrom Fig. 15'17 C : aresionalcoefficient Once the mean annualflood magnitudeis obtained,other annual flood magnitudes can easily be determinedfrom the appropriate index-flood curve (see Fig. the USGS data 26.4c).The usqof suchcurvesin urban hydrologyis limited because than 10 mi2. smaller watersheds includes seldom method index-flood network for the in the watersheds for applicable are later, described equations, regression The USGS larger. range and 1-10 mi2

'

CyprusCreekFormula
includingestimates Extreme$ flat areasposeparticular difficulties to the hydrologist, areastends!o these in Flooding rates. runoff and,peak volgme, runoff of infiltration, on the surface stands water low, and are velocities Flow widespread. be shallow and networks of by distinguished often are areas These periods of time. long for relative)y diseventually and store to constructed been have that ihannels straight drainage rain. chargethe excess peak flow to calculatethe'instantaneous a procedure3s developed The Scs be would that of canals capacity the calculating first on based flatland areas prevent would that a duration to storm design for the to limit flat-xeaflooding crop damage,and then to apply a multiplier to this rate to obtain the instant The procedureis illustrateil in Fig. 15. 18. peakfor the designof drainagestructures.

.oe

OOIfr

jtoo

.4 > '-6 *'-1

I' 1 |

'o?

' r l
.t\ t

6 . 1

lo.S

t "

:.9

9; E d
a =

x(.) s ! P>,

ra)

. Y v

- 92,

F\9
i : t6 r
9

I'o

a F = Y

;T{

x
U ',;
E O (J.n

20

x E O o ti= ;r o d c

u d:;< r i t r -

: E.e .$e b"

338

CHAPTER15

URBANAND SMALLWATERSHEDHYDROLOGY

.--Duratio4

of Overbank Flow-->l InstantaneousPeak

o o

Drainage Ditch CaPacitY "Removal" Rate Based on

Iime

Figure 15.18 Illustration of relation befweenCyplys Creek flow. (After Soil ConservationService'35) stJntaneous

"removal" rate and peak in-

to be the maximum allowabletime for duration was} hours,considered The selected to Creekformula, wasdeveloped Cyprus the called inundationof crops.An equation, equation, The rate. removal 24-hr the called rate, flow determinethe canal design curve based on rainfall depth,-contributingdrainage area,and the SCS composite number is
Q: CA5/6

( 1s .10)

channelcapacity fot 24-ht removal (cfs) where Q : reqrrired : drainagecoefflcient C areaGq mi) A: drainage bY coefficient,C,fot Eq. 15.10is found from an equationdeveloPed The drainage and Mi1ls36 Stephens (15.1) C : 16.39+ (14.75Q,",) where designeventfrom Fig' 4'I4' Q"", : the scS direct runoff (in.) for the 24-hr peak flow rate is Once Eq. 15.10is solvedfor the given frequency,the instantaneous from 1 to about areas drainage is limited to from Fig. 15.19.The procedure obtained to the 24-hr rate peak instantaneous that ratios of the 200 squaremiles.It is suggested areas flatland For 1.0. to equal than or canaliemoval rateUetirniLO b values greater peak flows the that that havepart of the areain storm ,"wJts, the SCS recornmends 15.20.TheSCSfurther fromFig. ts.tqu.increasedbytheamountsindicatedinFig. that haveslopesthat are restrictingur" of thit procedureto watersheds recomniends suchas TR-20, TR-55' methods other lessthan 0.002. For stJeperslopewatersheds, equationsare recommended' TP l4g, or regression

- <
I q

i)c<

o11

&

e&EEh+-il1e

t * .
4 q

z ^
;i^ - Y

3lj!.) = , = * d q

;>
=

N N

n $

Q O

o a - N

.e #
o .

(.) F
a

e f q + + t r
t < r o o .

E E b
c6

e
!

>

()
r ,
!

+ '; o c{
t q * c ' t 67)
9 t s
, L

9 H i 5
>, jl ,* -Q
v

- qca !t ,:, l,o

. \ 4

:
N

o
N

0 ) d L r
: r I

3rX

eleX ;;oung e8ere^V JnoH-tZ runrxrxehtr 01 luod snoeupluelsuJJo oqed

340

CHAPTER15

URBANAND SMALLWATERSHEDHYDROLOGY

s40
o !l

o 9 ^ ^ i 4rl

0
Percentage of Area Servedby Storm Sewers

Figure 15.20 Effect of urban storm sewerson peak dischargefor urban areas. (After U.S. GeologicalSurvey.aT)

EXAMPLE 15.5 Usethe Cypro, Creek methodto determinethe peak 50-yr flow rate fron a 1.0 sq mi andhasa50-yr,24-hr drainage areathat hasa CN : 80,is 50 percent stormsewered, rainfall depthof 12.0inches. of rain is 9.45in. The Solution. From Fig. 4.I4, thedirectrunoff for 12 inches drainage coefficient,C, is found from Eq. 15.11,

: 155.t + (14.75X9.45) C : 16.39

FOR SMALLRURALWATERSHEDS 341 15.3 PEAK FLOW FORMULAS

The 24-hr removalrate is found from Eq. 15.10' : 155'8cfs '0)5/6 0 : 155'8(1 rate to removalrate is 2.0, giving a From Fig. 15.19the ratio of instantaneous From Fig. 15.20,it is existed. designflow rate of 311.6cfs if no stormsewers by 35 percentfor shouldbe increased areadischarge found that the unsewered a watershedwith 50 percent storm sewers.The final designflow is 1.35 x 31'1.6: 420.7 cfs. rr

Equations SurveyRegression U.S.Geological for UrbanAreas

The U.S. GeologicalSurvey,in cooperationwith the FederalHighway Administraa nationwidestudy of flood magnitudeand frequencyin urban watertion, conducted includat 56 citiesin 31 states, involved26ggaugedbasins investigation The sheds.37 0.2 to from ranged Basin sizes 15.21. Fig. in are shown locations The Hawaii. ing 100 mi2. parame(see,Chapter 27) of a variety of independent Multiple linear regression to small, applied be could that equations peak flow develop to conducted ters was regression USGS Similar States. United the throughout watersheds urban ungauged in Chaptet27' equationsfor large rural basinsare described involvesthe three most equations regression form of the developed The simplest (mi2),ba-sin^developA area contributing were These identified. variables significant peak RQ,(cfs)for the lth flow corresponding the (dimensionless), and BDF factor ment The watershed. urban as the region same in the basin rural identical an from frequency from be developed can estimates and variations, regional for accounts latter vaiiable reports(seeSection27.4). The threeUSGSflood frequency any of the applicable flowsaregiven for the 2-,5-,IO-,25-,50-, 100-,and500-year parameter equations
as37

ez: l3.2Ao.zt(13 BDFl-o.azpnotz es : 10.6Ao.rz(13 BDF)-o.3eReo18 t0(13- BDnl-otuRQ?dn Qto : 9.5rAo ts(13- BDF)-o'z+P9o'to Qt5 : 8.68Ao ts(13- BDFl-o'zzR03o" Qso: 8.o4Ao ts(13- BOrT-ot'RQ?r!& Qno: 7.70Ao - BDF)-'*RQ1i& : 7.47 A0'16(13 Qsoo

(Ls.r2) (1 s.1 3)
(15.14)

(1s.1s) (1s.16)
(15.17) (1 s.1 8)

from data at 199of the 269 original sites.The other siteswere Thesewere developed presence.of detentionstorageor missingdata. All theseequaof the because deleted above0.90. determination of coefficients tions have used values and observed of estimated the correspondence 15.22 shows Figure deviation standard one within fall values the percent of Forty 15.15. Eq. in devedping line. Graphsfor other recurrenceintervalsare similar to the 1O-year of the regression graphshownin Fig. 15.22.

O N o v N E > li p rii

'

h E

- ^ a s - _ . A

>'
q

,F

i .r t - :

'o 3 r {
g

o
o / o

()

B --l I I I I L,/ I
I

r d \Oa

l n
l V z

I 0".
) .' r U
I j A f
r E l

ol

E'J

,g

q g

--j

, l
: : F

: (,)
7

i I

PROBLEMS 353

15.5,

rainfall duration? is The 4-hr unit hydrographfor a 5600-acrewatershed

Time (hr) 0 (cf9
n 1

400

8 4 6 1000 800 400

10 200

t2 0

: discussthe Rework Example 15.2 basedon a C : 0.2 and C 0'4' Compare and at the outfall' effect of C on the discharge AwatershedhasareaA.Startingwithatriangular.shapedunithydrographwithabase : 484A/to' lengthof 2.67t, and a heightof [0, deriveEq. 15'9 (seealsoEq' t2'25)' Qo derivation' the State and carr units of eachterm usedin

UsingtheSCSdimensionlessunithydrographdescribedinChapter12,determinethe peak for a net storm of 101n.in 2 hr on a 400-acrebasinwith a time to peaidischarge ^of + nt and i lag time of 3 hr. Comparewith Eq' 12'17' with a 100-min time of concentrationreceivesrainfall at a A 10.00-mi2watershed rate of 2.75 in.lhr for a period of 200 min' 1f C : 0'4' (cfs)from the watershed a. Determinethe peak d-ischarge of rainfall. beginning rate lcfs) 150 min after the b. Estimatethe discharge the beginning of after min 40 watershed rate from the c. Estimate the dischar"ge rainfall.
F

E * n

{
*
F

F 6 0

9 4 0
o o

E ) i o - "
s? o

o
-=

100

-Time{min)-after-beianing-of rainfall

354

CHAPTER15

URBAN ANDSMALL WATERSHED HYDROLOGY

1s.13. A storm gutter receivesdrainagefrom both sides.On the left it drains a rectangular
600-acreareaof t" : 60 min. On the right it drains a relatively steep300-acrearea of t" : 10 min. The f index on both sidesis 0.5 in.ihr. Use the intensity-durationfrequency curves in Fig. 15.7 to determinethe peak discharge(cf$ with a25-year recurrenceinterval for (a) the 600-acreareaalone, (b) the 300-acrearea alone,and (c) the combinedareaassuming that the proportion of the 600-acreareacontributing to runoff at any time r after rain beginsis l/60. t5.14. A drainagebasin has a time of concentrationof 8 hr and producesa peak Q of 4032 cfs for a 10-hr storm with a net intensity of 2 in./hr. Determinethe peak flow rate and the time base(duration)of the direct surfacerunoff for a net rain of 4 in./hr lasting (a) 12 hr, (b) 8 hr, and (c) 4 hr. State any assumptions used.

15.15. A 1.0-mi2parking lot has a runoff coefficientof 0.8 and a time of concentrationof
(cf$ by the 40 min. For the following three rainstorms,determinethe'peakdischarge rationalmethod:(a) 4.0 in./hr for 10 min, (b) 1.0in./hr for 40 min, and (c) 0.5 in./hr for 60 min. State any assumption regarding area contributing after various rainfall durations. 15.16. The concentration time varies with dischargebut is relatively constant for large feel confidentin usingthe rational discharges. From this statement, why do engineers formula? 15.17. Determinethe 50-yearflood for a20-mi2 basin at the northwestcorner of Nebraska. that Fig. 26.4 appbes. Use the index-flood method and assume 1s.18. Determinethe entire frequencycurve for the basin in Problem 15.17 and plot it on probability paper.

15.19. Use the index-flood method to determinethe 10- and 50-yearpeaksfor a 6400-acre
drainagebasin near Lincoln, Nebraska.Assumethat Fig. 26.4 applies. 15.20. For the drainagebasin in Problem 15,19 determinethe probability that the 20-year peak will be equaledor exceeded at leastonce (a) next year and (b) in a 4-yr. period. Referto Section26.1.

15.2L. For a 100-mi2drainagebasinnearLincoln, Nebraska,usethe index-flood methodto

determinethe probability that next year's flood will equal or exceed3000 cfs. 15.22. UseFig. 26.4 to determinethe return period (years)of the meanannualflood for that region.How doesthis comparewith the theoreticalvalue for a Gumbel distribution? How doesit comparewith a normal distribution?Refer to Section26.6. for the watershed ts.23. Usethe Cyprus Creekmethodto determine the 25-yr peak discharge is nearly flat. describedin Example 15.3. Assumethat the watershed

15.24. You are asked to determinethe magnitudeof the S0-yearflood for a small, rural
drainage basin (nearyour town) that hasno streamflowrecords.Statethe namesof at leasttwo techniques that would provide estimatesof the desiredvalue. 't5.25. The drainageareas,channellengths,and relevantelevations(underlined)for several subbasins of the Oak Creek Watershed at Lincoln, Nebraska,are shownin Fig. 24.8. The watershed has a SCScurve numberof CN : 75 which may be usedto determine the direct runoff for any storm.Assumethat IDF curvesin Fig. 27.13 applyat Lincoln. Treat the entire watershed the 50-yearflood magnitude asa singlebasinand determine at Point 8 using: a. The rational method. b. The SCSpeak flow graph,Fig. 15.8. Eq. 15.5. c. Snyder'smethod of syntheticunit hydrographs, d. The USGS index-flood method. Figure 26.4 applies.

REFERENCES 355

15.26. RepeatProblem 15.25 with SubareaI excluded.compare the results with Prob-

event of the at Point 8 for the 5O-year lem 15.25 and comment on the effectiveness flood 100-year the store easily BranchedOak Reservoirat Point 9. (This reservoirwill from Area I.) 1s.27. RepeatProblem 15.25 for SubareaA' 15.28. RepeatProblem 15.25 for SubareaI' is deter15.29. Describecompletelyhow the magnitudeof the 30-yearflood for a watershed mined by the USGS index-flood method. 15.30. A rural watershedwith a composite cN of 70 is being urbanized. Eventually' 36 percentof the areawill be impervious.Determinethe increasein runoff that can for a 6.2-in.rain. be expected unit hydrographin Ch. l2,_determine peak flow for the SCS dimensionless 15.31 Using the for a net storm of 10 in. in Zhr on a 400-actebasinwith a time to the piak discharge peak of 4 hr and a lag time of 3 hr. shown L5.32. A timber railroad bridgein Nebraskaat Milepost 27I.32 ontherailroad system in the sketchis to be replacedwith a new concretestructure.The 50- and 100-year are neededto establishthe low chord and embankmentelevations, flood magnitudes respectively.Determine the designflow rates using the scs TP-149 method. The bridge drainsthe zone marked,about45 acres.The moderatelyslopedbasinlies in a rainfall rype-n stormregion,the curve numberis 70, and the 24-hr 50- and 100-year g.4" respectively' depthsare 8.6" and

Bidge27l.32

method.The problem 15.32usingthe FHWA HEC-19 peakflow SCSdesign 15.33. Repeat relationships the from be determined 1o can of Values hrs. is 0.2 time of concentration in both metric and English units. , in Fis. 414. Provide the answers

REFERENCES
"A Critique of Current Methods in Hydrologic Systems 1 . J. Amorocho and W. E. Hart, Investigations ," Trans.Am. Geophys.Union 45(2),301-321(Jwe 1964)' ..NonlinearInstantaneous jingh, p. Unit-HydrographTheory," ASCE J. Hyd. Div. 2. f. -347(Mar' 1964). 90(HY2), Par I, 313

356

HYDROLOGY 15 URBAN ANDSMALLWATERSHED CHAPTER "ContinuousHydrographSynthesis with an and J. C. Monro, 3. W T. Sittner, C. E. Schauss, Res.5(5), 1007- 1022(1969). API:Iype HydrologicModel," WaterResources 4. J.E.Nash,"TheFormoftheInstantaneousUnitHydrograph,"Int.Assoc.Sci.Hyd.3@5), "Mathematical Models of CatchmentBehavior," Proc. 5. D. R. Dawdy and T. O'Donnel, ASCEJ. Hyd. Div.91(HY4), 124-127(Iuly 1965). J. Geophy. 6. S. L. S. Jacoby,"A MathematicalModel for Nonlinear Hydrologic Systems," Res. 7l(20), 48t | - 4824(0ct. 1966). Model," Proc. ASCE J. Hyd. 7. R. Prasad, "A Nonlinear Hydrologic System Response Div. 93(HY4)(1967). "Hydrology of Urban Runoff," J. ASCE 85, 418. A. L. Tholin and C. T. Keifer, 1959). 106(Mar. "Digital Simulationin Hydrology:StanfordWater9. N. H. Crawfordand R. K. Linsley,Jr., shedModel IV," Department of Civil Engineering,Stanford University, Stanford, CA, Tech.Rep.No. 39, July 1966. 10. JohnC. Schaake, Jr., "synthesisof the Inlet Hydrograph,"Tech.Rep. 3, Storm Drainage Project, JohnsHopkins University,Baltimore, MD, June 1965. Research "WaterPollution Aspectsof UrbanRunoff," Federal 11. AmericanPublic WorksAssociation, Water Pollution Control Administration, 1969. "Urban Water Resources Re12. Arnerican Society of Civil Engineers,First Year Report, search,"Sept. 1968. 13. W. Viessman,Jr., "Modeling of Water Quality Inputs from Urbanized Areas," Urban Water ResourcesResearch, Study by ASCE Urban Hydrology Research Council, pp. A79-A103. Sept.1968, "Characterization, 14. S. R. Weible,R. B. Weidner,A. G. Christianson,and R. J. Anderson, of the Third InternaTreatment,and Disposal of Urban Storm Water," in Proceedings (S. H. Jenkins, International AssociationonWater Pollution Researcft tional Conference, Press, 1969. Elmsford, NY Pergamon ed.). "Pesticides and Other 15. S. R. Weible,R. B. Weidner,J. M. Cohan,and A. G. Christianson, Contaminants in Rainfall and Runoff," '/. Am. Water Works Assoc. 58(8), 1675(Aug. 1966). Departmentof Civil Engineering,University of Cincinnati, 16. Division of WaterResources, SeWater Pollution Control Research Cincinnati. OH. "Urban Runoff Characteristics," ries. EPA. 1970. Engineers, 17. Metcalf and Eddy, Inc., University of Florida, Gainewille, Water Resources "Storm Vol. 1, Agency, Protection Model," Environmental Management Inc., Water 18. E. Kuichling, "The Relation Betweenthe Rainfall and the Dischargeof Sewersin Populous Districts,"Tians.ASCE,20(1889). 19. W. W. Horner, "Modern Procedurein District SewerDesign," Eng. News 64,326(1910). "Relation BetweenRainfall and Runoff from Small Urban 20. W. W Horner and F. L. Flynt, Areas," Trans.ASCE 20(140),( 1936). 21. R. L. Rossnriller,"The Runoff Coefficient in the Rational Formula," EngineeringResearchInstitute, Iowa State University,Feb. 1981. "Experimental Examination of the 22. J. C. Schaake,Jr., J. C. Geye1,and J. W. Knapp, RationalMethod," Proc.ASCEJ. Hyd. Div.93(HY6) (Nov. 1967). "Airport Drainage," Advisory 23. FederalAviation Agency, Departmentof Transportation, PrintingOfflce, 1970. D.C.: U.S. Government Washington, Circular, AIC 150-5320-58.

r14-L2r(r9s7).

r971.

Chapter16

Design Hydrologic

Prologue
of this chaPteris to: The PurPose for designing usedin the United States . Introducethe hydrologist to procedures of flood flows' structuresfor safe and effectivepassage , Give sufficient information for the designerto selectthe applicablecriteria for hYdrauhcstructures' designing and provide methodsfor , provide a discussionof designstorm hyetographs design. selectingthe duration, depth, and distribution of plecipitation for precipitation without using . DemonJtrate how designiloods can be developed data' ' Discussparticular designmethods including airport drainage'urban storm sewerdeiign, and flood control reservoirdesign' . Describethe U.S. Federal EmergencyManagementAgency (FEMA) flood systemand piesent the hydrologicfundamentalsof flood plain management Plain analYsis. 26 and27 piotto studying to reviewthe materialin Chapters Readersare encouraged in this presented designprocedures rater discharge PredictingPeak ma and minor for use in designing I hydrology' aspects of engineering dt levees, small crodsroadculverts, akPort drainage structuresto the lumped together'with major structr design information. GenerallY,a h dischargefor a designfrequencY,a dischargehYdrograPhfor a design rates,low-flow frequencYanalysis, are often conductedas part of a designproiect'

360

DESIGN 16 HYDROLOGIC CHAPTER Mostdesignsinvolvinghydrologicanalysesuseadesignfloodthatsimulates are futirre eventorlmitates ime historicalevent.If streamflowrecords somesevere records storm from available are synthesized unavailable,designflood hydrographs are 15. Only in rare cases and 12, Chapters2, of procedures usingthe rainfall--runoff watersheds' in small particularly for complex designs, streamflowrecords adequate in Chapter 15 are discussed methods empiric-coirelative the and analyses Regional in chappresented Methods sites' peak flow ratesat ungauged usefulfor determining for necessary hydrographs entire ter 12 andin this chapterare used for developing designs. manYengineering in minor and major structuresare described uyirologic;ethJds for designing for rnethods levels, this chapter.included are discussions;f data needs,frequency for floodplainsand dams' designstorms,and hazard assessments synthesizing

PROCEDURES DESIGN 16.1 HYDROLOGIC

peak flow for estimatingdesignflood flows (interestcan be in either the Procedures projected or historical rate or the entire hydrogiaph)includemethodsthat examine that methods and (flow-basedmethods), estimate flood flowsto arrive at i sultaute rates flow flood to storms the stormsthat producefloods,andthen convertthe evaluate a on selecting canbe based the analysis In eachcase, (precipitation-basedmethods). meth' (callfrequency-based flood i"rign rr"qo"ncy and determiningthe ass,ociated narrowing the final and_ frequencies flood of ringe a for designs ods), developing methods)'or (called risk-based choice on the basisof long-term c6sts and benefits designingonthebasisofanestimateoftheprobablemaximumstormormanmum RooAtnit could occur at the site (calledcritical-event methods). basedon frequencyMinor Structure Design Minor structuredesignis largely to approach hydrologic the in steps Several risk]basedmethods. basedor sometimes techadopted and handbooks design most to design are comrnon minor structtJre are: niques.The generalsteps(eachis illustrated subsequently) to the time 1. Determinethe duration of the critical storm,usually equated concentrationof the watershed. t Choosethe designfrequencY. frequencyand duration' on the selected 3. Obtain the storri OeptltUasea in 4. Qomputethe net direct runoff (severalmethodswere presented ter 4). 5. Selectthe time distribution of the rainfall excess' 6.Synthesizetheunithydrographforthewatershed(seeChapterl2). T.Applythederivedrainfall_excesspatterntothesyntheticunithydrograph get the runoff hYdrograPh. equalto t flood (usuallyassumed the frequen-cyiftn" calculated Establish 8. designstorm frequencY).

DESIGNPROCEDURES 16.1 HYDROLOGIC

361

are considof maior structures Major Structure Design Hydrologicdesignaspects drainage or urban culvert, crossroad erably more complexthan thoseof a small dam, it is put to but is required still system.A designstorm hydrographfor a'large dam of adequacy the to determine is routed greater use. The designstorm hydrograph economic The storage. reservoir with spillwaysand outlets operatedin conjunction dictatesthe final designand ofthe spillwaysizefrom the variouspossibilities selection life and property, provided for downstream protection is a function of the degreeof operational reservoir and standards, policy project economy,agency andconstruction premethods critical event on largely based Major structuredesignis requirements. in Section16.5. sented in presented Water Resource System Design Most information and techniques large strucand of small protection aspect this chapterare directedtoward the flood for more than just flood protecto say,a major structureis designed tures.Needless for tion; it is multipurposeand may provide storage irrigation, power, water supply, to theseuses The proper allocationof storage navigation,and low-flow augmentation. frequency of the in terms history of the entire streamflow requiresan understanding yearly as well flows, and monthly, seasonal, of occurrenceof low flows and average provide a Five to presented in Part as the historical and designfloods. Material is complex for a histories hydrologistwith the tools to developcomplete streamflow multipurposesysteminvolving various combinationsof minor and major structures, practlces. projects, and management water development

Methods Flow-Based
or whereflows from For designlocationswhererecordsof streamflows are available, magnitudecan flood a design location, the design to can be transposed anotherbasin methods: following the of flows by any from the stream directly be estimated ' 1. Frequencyanalysisof flood flows at the designlocation or from a similar basin in the region. from regresnormally developed 2. Useof regionalflood frequencyequations, gauged data. flood sion analysis(seeChapter 26) of 3. Examination of the stream and floodplain for signs of highest historical of the crossfloods and estimationof the flow rates using measurements sectionand slopeof the stream.

Methods Precipitation-Based
for streamflow.estimaor inadequate recordsare unavailable Where stream-gauging precipitation that would prothe by evaluating be estimated floods can tion, design ofthe rainfallany runoffby into the flood, and then converting ducethe frecipitation include: methods Typical 2l-27. 10-15 or in Chapters described runoff methods 1. Design using the greateststorm of record at the site, by converting the precipitation to runoff. in historicalstormfrom anothersimilar watershed of a severe 2. Transposition the region.

362

CHAPTER16

HYDROLOGIC DESIGN

3. Frequency analysis of precipitation and conversion of design storm to runoff. 4. Useof a theoreticalprobablemaximumprecipitation (PMP), or fraction of PMP, basedon meteorological analyses. methodsare Because the flood flow rate is desiredin all cases, the flow-based preferred over conversionof precipitation to runoff. Due to the relatively longer period of time and greaternumberof locationsat which precipitation amountshave been recorded, precipitation-basedmethods are used in the majority of designs, especiallywith small and very large basins.Flow-basedmethodsare typically used in the midrangeof basin sizes.

Frequency-Based Methods
most often proceed Regardless of whetherflow or precipitation dataareused,designs by selectinga minimum acceptable recurrenceinterval and using proceduresfrom Chapter27 to determine the corresponding worst condition storm or flood that could be equalledor exceeded recurrenceinterval. Criteria for selecting during the selected designrecurrenceintervals are summarized in Section 16.3. Resultsfrom frequency analysisof flood flow data normally provide reliable estimatesof 2-, 5-, 10-, and 25-yearflows.Extrapolationbeyondthe rangeof the period of flow recordsis allowed, but is lessreliable.

Risk-Based Methods
Recenttrends in designof minor (and major) structuresare toward the use of ecoThe risk methodselects the nomic risk analyses rather than frequgncy-based designs. costs.Tfreseare madeup of the structuresizeas that which minimizestotal expected with the particular structure. structurecostsplus the potential flood losses associated The procedureis illustrated in Fig. 16.1.The total expectedcost curve is the sum of

o b0 q

Optimal structure size, S* (least total expected cost) S^in Structure size, S

forstructure Figure 16.1 Principlesof economicriskanalysis size selection. (U.S. Federal Highway Administration, Hydraulic EngineeringCircular No. 17).

16.2 DATA FOR HYDROLOGIC DESIGN 363 the other two curves.Risk costs(flood damages, structuredamages, road and bridge losses, traffic interruptions)and structurecostsare estimated for eachof severalsizes. The optimal sizeis that with the smallestsum.Structuresselected by risk analysisare normally constrained to sizesequal to or larger than thoseresultingfrom traditional frequency-based methods.

CriticalEventMethods
Becauseof the high risk to lives or property below major structures,their design generallyincludesprovisionsfor a flood causedby a combinationof the most severe meteorologicand hydrologic conditions that are possible.Instead of designingfor somefrequencyor leastexpected total cost,flood handlingfacilities for the structures are sizedto safelystoreor passthe most critical storm or flood possible. Methodsfor designing by critical eventtechniques include: Estimating the probable maximum precipitation (PMP) and determining the associated flood flow rates and volumesby transformingthe precipitation to runoff. ) Determiningthe probablemaximum flood (PMF) by determiningthe PMP and convertingit to a flood by applicationof a rainfall-runoff model,including snowmeltrunoff if pertinent. 3. Examining the flood plain and stream to identify palaeo-floodevidences such as high-water marks, boulder marks on trees or banks, debris lines, historical accountsby local residents, or geologicor geomorphologic evidences. 4. In somecases, the critical eventmethodinvolvesestimatingthe magnitude of the 500-yr eventby various frequencyor approximatemethods.Often, suchas in mappingfloodplains,the 500-yr flood is estimatedas a multiple of the 100-yr event, ranging from 1.5 to 2.5. Due to lack of longer-term records, frequency-based estimatesare seldom attemptedfor recurrence intervals exceeding 500 years.

16.2 DATAFORHYDROLOGIC DESIGN

The designof any structurerequires a certain amount of data, even if only a field estimate of the drainage areaand a description of terraintype and cover.The following material identiflessomegeneraldata types and sources.

Physiographic Data
The hydrologic study for any structurerequires a reliable topographicmap. United StatesGeologicalSurveytopographic mapsusually are available. The mappingof the United Statesis almost completewith 15-minute quadrangles, and many of these areasaremappedby 7.5-minutequadrangles. County mapsand aerialphotoscan also be usedto advantage in making preliminary studiesof the watershed.

364

DESIGN CHAPTER16 HYDROLOGIC

drainage Based on an area map, a careful investigation of the watershed's maps USGS from obtained be can information behaviormust be made. Additional erosive and inflltration the and types Soil that depict predominantrock formations. districtsor univerfrom U.S. Soil Conservation of soilscanbe secured characteristics divisions. sity extension of an The drainageareascontributing to large dams require stricter analysis of a possibility The structures. minor designing in area,shydrotogyitranis necessary large for uniformly intenserainfall over the entire basin is an unrealisticassumption thus should the rainfall of variations spatial and The influenceof temporal watersheds. "worst possible" rainfall values are estimated the For major dams, be considered. which is then usedin reservoir hydrograph, generallyconvertedto a designdischarge and storage, size, surcharge spillway and reservoii routing calculationsto propoition downsustained or power requirements maintain any additional outlets neededto irrigation, or watersupply.The basicconcernin hydrologic streamflow for navigation, interestsusing a realistic estimatefor designof a large Aamis to protect downstream the designstorm hYdrograPh. purpose the Topographic'rnuf o"tuit necessarilyshifts with the type and .of of understanding the pi"ta increases always reconnaisiance structurebeing design"a. be. might structule the insignificant matter how an area,shydrology*no

Data Hydrologic
datafor the regionunder one difficulty in hydrologicdesignis that of gettingadequate by issued published-reports from pr-eviously data canbe-acquired study.ConsiOerabie agencies federal of list is a and/or universities.The following agencies governmental data: hYdrologic that Publish ' Service egricultural Research Service Soil Conservation Forest Service U.S. ArmY CorPsof Engineers National Oceanicand AtmosphericAdministration Bureauof Reclamation DePartmentof TransPortation Division U.S. GeologicalSurvey' Topographic Division WaterResources U.S. Geological.Survey, interof stategovernments, Additional dlta often canbe procuredfrom departments and regional and local agencies' statecommissions,

MeteorologicData
and Atmospheric The National WeatherService, couchedin the National Oceanic in a variety published data meteorologic of source Administration, is the primary 16.2 Figure (HMR) series. Report of forms, including their Hyirometeorologic showstheapplicablereportsforvariousgeographicandtopographicregionsofthe

CRITERIA DESIGN-FREQUENCY 16.3 HYDROLOGIC

365

Figure16.2HydrometeorologicalreportseriescoverageofconterminousUnitedStates. of Reclamation') (U.S.Bureau collect and analyze state,and local agencies United States.lNumerousother federal or regulatelarge inspect, who design, precipitation information-"tp""iuriv-tr'rore

"*"8l;.H[1;:1;ce

in the region,maximum amount ttre meteoroioii" .huru"t"ristics of Jtot-t of total storm ;;r;r of precipitation,frequencies itable moisturein the atmospheJo over the storms for snowmelt of and influence. durationsbf U;t, depthsf*;;;i"", topography chains, mountain of major region.tn ,o1n" areassuch", f;;;iii'*gion, precipitation' on has a very distinct impact

require:.5::S:1g:,:f stormhyetographs design forestimating or precrp-

CRITERIA DESIGN-FREQUENCY 16.3 HYDROLOGIC

S e l e c t i o n o f f r e q u e i s m o s t o f t e n b a s e d o n p o t e n t i a l d a m a g e t o p r o p e r r : l o s s e s s u c h a s i n t e r r u p t i o n o f A stanc commefce' the worst conditio involved,a greal a projects involve somerisks to property an A11 thror.t human tife li absent'the designcan proceed tn structure cost least the of quencylevel and design alternativetoleastStructurecost,economicriskana] rather than the final designfrequencyis optimized trequencret several for storms would accommodate includenot only the actualconstfuction costs itr"r" u*"J. is cost leasttotal expected "o** du9 to interruption of servicesand costsbut also the flood dama!;;irk una can be used' worth economicanalyses commerce.Either annual or p"resent

366

CHAPTER16

HYDROLOGIC DESIGN

MinorStructures
shown in Table 16.1 are typical of levels generallyencounThe designfrequencies tered in minor structuredesign.An exampleof variationsthat do occur is the design could effectivelyhalt backwater of excessive frequencyof a culvert,which undercases trafflc. the use of a2l-year frequencyfor Servicerecommends The Soil Conservation minor urban drainagedesignif there is no potential loss of life or risk of extensive damagesuch as first-floor elevationsof homes.A 100-yearfrequencyis commonly property damagemay occur.t recommended when extensive
DESIGNFREQUENCIES TABLE 16.1 MINORSTRUCTURE

Typeof minor structure Highway crossroad drainage" ADT' 0-400 400-1700 ADT 1700-5000 ADT ADT 5000Airfields Railroads Stormdrainage
Levees Drainage ditches

period, Return 4 10yr 10-25yr 25 yr 50 yr 5yr 25-50 yr 2-10 yr 2-50 yr 5-50 yr

= 1/7, Frequency 0.10 0.10-0.04 0.04 o.o2 0.20 0.04-0.02 0.50-0.10 0.50-0.02 0.20-0.02

'ADT : averagedaily traffic. (After Ref. 3).

Large Dams
of the original structureandduring duringthe design Damsrequirehydrologicanalysis when arepossible periodic safetyevaluations. Significanteconomicand humanlosses from storage. released quantities rapidly of water are large the total volume Initial heightsof retardedwater behind the dam, disregarding distance for a considerable flood waves produce destructive water, can of stored Floods recomDesign on Spillway Task Force the Based on two criteria, downstream. of construc16.2. The type Table li,sted in as of large dams the classification merided tion has not been included in this grouping, althoughit affects the extentof failure resulting from overtopping. of haveadopteddefinitionsfor hydraulic elements Many of the federalagencies Service: Soil Conservation list is by the The following used dams. A spillwuy is an open or closedchannel,or both, used to convey excess water from a reservoir.It may contain gates,either manually or automatwater' of excess ically controlled, to regulatethe discharge to conveythe water Theprincipal spillwayis the ungatedspillwaydesigned for the structure. rates established from the retarding pool at release to conveywater spillway of a dam is the spillwaydesigned The emergency of that impoundedfor flood control or other beneficialpurposes. in excess

DESIGN-FREQUENCYCRITERIA 16.3 HYDROLOGIC

367

DAMS FORI-ARGE CRITERIA TABLE16,2 DESIGN danger lmpoundment Potential Category (1)
Major; failure cannot be tolerated Storage (acre-ft)o (2)

Failure damage Potential' Loss of life (4) Considerable

Height (ft) (3)

Damage (5)
Excessiveor as matter of PolicY

Spillwaydesignflood (6) Probablemaximum; most severeflood considered possible reasonably on the basin Standardproject; based on most severe storm or meteorological conditions considered reasonablycharacteristic of the sPecific reglon basis; Frequency 50-1O0-year recurrence interval

>50,000

lntermediate

1000-50.000 40-100

Possiblebut small

Within financial capability of owner

<1000

<50

None

Of samemagnitude as cost of the dam

and future potential oBased on consideration of height of dam above tailwater, stoarags volume, and length of damage reach, present floodplain' of population, and economic development tstorage at design spillway pool level. Sozrce: After SnYder.3

The retarding pool is the reservoir spaceallotted to the temporary rmpoundmentoi floodwater.Its upperlimit is the elevationof the crestof the emergencyspillway. Retardingstorageis the volume in the retarding pool' pool is the reservoir spaceallotted to the accumulationof The sediment incoming sedimentduring the life of the structure' storageis the volume allocatedto total sedimentaccumulation' Sediment sedimentpool elevationis the elevationof the surfaceof the anticipated sedimentaccumulationat the dam. Anearthspitlwuyisanunvegetatedopenchannelspillwayinearthmaterials. Avegetatedspillwayisavegetatedopenchannelspillwayconstructedof earth materials. spillway constructedon the downstream A ramp spillway is a vegetated dam. faceof an earth where The control section in an open channel spillway is that section through critical depth' flow passes accelerated spillwayis the channelupstreamfrom The inlet channelof an emergency the control section.

368

CHAPTER16

DESIGN HYDROLOGIC Top of dam

/ :Minimum crestJ spillway Emergency x r freeboard Surcharge poollevel Normal \ v I storage

Flood control retardingstorage

Emergency spillway

Reservoir Minimum Pool

Qo

zones' andstorage poollevels reservoir Figure 16.3 Multipurpose The exit channelof an emergencyspillwayis that portion of the channel from the control sectionwhich conductsthe flow safelyto a downstream without jeopardizing the integrity of the point where it may be released structure. spiltway lrydrographis that hydrographusedto establish The emergency of the emergencyspillway' the minimum designdimensions Thefreeboard hydrographis the hydrographused to establishthe minimum elevationof the toP of the dam. Severalof thesefeaturesare illustrated in Fig' 16'3'

SmallDams
usingtwo or more levelsof frequencyto provide Small damscustomarilyare designed 16.3 an emergencyspillway and ensure an adequateallowable freeboard. Figure (MF)showsa iypicit small dam with normal freeboard(NF) and minimal freeboard The freeboardvaluesfor earth dams with riprap protection on the upstreamslope' by storm winds with 100-mph shownin Tablq 16.3, atebasedon waverunup caused wind velocities.Minimal freeboardpertainsto wind velocitiesof 50 mph. The fetch If is definedas the perpendiculardistancefrom the structureto the windward shore. values smoothconcrerc;atirerthan riprap is usedon the upstreamface' the freeboard shown shouldbe increased50 percent'"

CFIITERIA DESIGN-FREQUENCY 16.3 HYDROLOGIC TABLE 16.3 USBR RECOMMENDED NORMALAND MINIMUM VALUES,FT FREEBOARD Fetch (mi)

369

<1
I

4 6 8 10

3
+

2.5 5 10
Soarce:After Ref. 4.

5
o

of The U.S. Soil conservation Service designcriteria for principal spillways be No. 60 should small dams are given in Table 16.4. The SCS TechnicalRelease are frequency_requirements Design table.s this of consultedfor full interpretation use of the structures.The SCS classifles selectedto fit the planned or foreseeable grouPs:u into three structures Class a. Structureslocated in rural or agricultural areas where failure mightdamagefarmbuildings,agriculturalland,ortownshiporcountry roads. ctass b. Structureslocated in predominantlyrural or agricultural areas where failure might damageiiolated homes, main highways.or minor railroads, o, "urrr]"interruption of use or serviceof relatively important public utilities. Classc. Structureslocatedwhere failure might causeloss of life, serious public damageof homes,industrial and commercialbuildings,important utilities, main highways,or railroads, ft in heightbut generally The physicalsizeof a small dam can rangeto over 100^ of storageat the emeracre-ft is restrictedto structuresretarding lessthan 25,000 are gency spillway crest. Small dams generally receive.special attention if they Many life' of loss the cause could ion.i.o"i"d in populatedareaswhere dam failure by dam or leveefailure. When this possibilityexists,the havebeencaused flood deaths precipidesignstorm for small damsis lstablishedby useof the probablemaximum of the maximization reasonable the tatio'n, PMP. The PMP is generally defined as definitions Other storm. maximum factorsthaioperate to producea meteorological including: havebeen proPosed,T canbe 1. The p1itp is the rnaximumamountand duration of precipitationthat expectedto occur on a drainagebasin' combinafrom the most severe 2. ThePMP is the flood that may be expected reasonably are that conditions and hydrologic tion of critical meteorologic possiblein the region. T'he pMp Las a low, but unknown, probability of depthat the designlocation the maximumobserved It is n-either occurrence. to exceedance' immune or region nor a value that is completely

370

CHAPTER16

HYDROLOGIC DESIGN

SPILLWAYS TABLE 16,4 SCS DESIGNCRITERIAFOR PRINCIPAL OF SMALL DAMS Precioitation data for maximum frequency2of use of spillwaytype: emergency

Class of dam
(a)

Purpose of dam
5lngle-

V"HI
Less than 30,000 Greaterthan 30,000

Existing or planned upstream dams

None

Earth

Vegetated

irrigation only

0.sDrJ 0.75DL
!D 50

O.5DL

None

0.75DL
D6 r25

SingIe or multiples

Lessthan 30,000 than Greater 30,000 All

All

None None Anyt None or any None or any

0.5(Pso + Ploo)
D r 100

0.5(Prs+ Pso)
rD 50

(b)

Single or multiple Single or multiple

100

(c)

All

Proo

I Product of reservoir storagevolume V, (acre-feet) times effective height of dam 11, (feet). 2Precipitation depths for indicated return periods (years). 3Applies to irrigation dams on ephemeralstreamsin areaswhere mean annual rainfall is less than 25 in. aDL = designlife (years). 5Class (a) dams involving industrial or municipal water are to use minimum criteria equivalent to that of Class (b). 6In the case of a ramp spillway, the minimum criteria should be increasedfrom Prr to Pt66. ?Applies when the failure of the upstream dam may endangerthe lower dam. Soarce.'Soil Conservation Servtce.

Estimatesof PMP are basedon an investigationby the U.S. WeatherBureau conducted to establish the maximumpossibleamountof precipitablewaterthat could of precipFigure 16.4providesestimates be achieved throughoutthe United States.s'e itable water over watersheds betweensealevel and 8,000 ft. Figure 16.5 extendsthe estimatesabove 8,000 ft.1 Point valuesof PMP for the samelocale may vary with for duration of storm causingthe precipitation.Figure 16.6 providesPMP estimates 6-hr storms.These and similar publishedcharts for other durations are helpful in selectingthe PMP for any region in the United States. the deiign frequenlies-for principal spillwaysfor small SCS Class aob, or c damsare providedin Table 16.5.Theseare basedon 6-hr rainfall depthsfor ( 1) the for (Fig. 16.7) and(2) the PMP (Fig. 16.6).Designstormdepths 1OO-year frequency all watershedshaving a time of concentration less than 6-hr are establishedin are with greatertime of concentration,adjustments Table 16.5.For thosewatersheds madeto the 6-hr storm depth to accountfor the gteateramountsof direct runoff in a in Section 16.4. longerperiod of time. Theseadjustmentsare discussed

16.3 HYDROLOGICDESIGN-FREQUENCYCRITERIA 371

DEPTHSOF PRECIPITABLE WATER IN A COLUMN OF MILLIBARS AIR OF GIVEN HEIGHT ABOVE 1OOO
Assuming saturation with a pseudo-adiabatic lapse mte for the indicated surface temperatures Adapted ftom the U S Wther Bueau Hyalrometeorological Repon No 23 . 142230343842 46 505254 565860 TEMPEMrure 62 64 66 68 '10 '72 74 '76 78

80 .F

800
a

.: =
p
h

z
a
H d

Extended to 200 mbar on Figure 16.5

1.5 1.0 ABLE WATER IN INCHES

Figure 16.4 Diagram for precipitable water determination from 1,000 to 700 millibars.(U.S. Bureauof Reclamation.)

372

16 HYDRoLoGIc CHAPTER DESIGN

DEPTHS OF PRECIPITABLE WATER IN A COLUMN OF AIR OF GIVEN HEIGHT ABOVE 1OOO MILLIBARS lapse Assuming saturation witha pseudo-adiabatic
rate for the indicated surface temDeratwes Adapted from the U-S. Weather Bureau Hydrometeorolgical Report No 23 EMPEMTre

l8 26

66 70

74

200

2.0 2.5 3.0 3.5 1.5 WATERIN INCHES PRECIPITABLE

Figure 16.5 Diagram fbr precipitable water determination from 800 to 200 (U.S. Bureauof Reclamation.) millibars.From 1281. 103-D-1908.

16.4 DESIGNSTORMS

373

MaiorStructures

STORMS 16.4 DESIGN

the next stepin a structuredesignis Once the designfrequencyhasbeenestablished, storm duration,the durationof rainfall the puru-eters: six ,io.of the determinaiion the storm intensity and time adjustment, depth ireal point any depth, the excess, pattern' distribution the areal and distribution,

Duration
The length of storm usedby the SCS in designingemergencyand freeboardhydrographsior small damsis of 6-hr durationor /c, whicheveris greater' Often, the minor cannotbe justified economicallyon the basisof this length it*"tu.. being designed of storm. Foi many minor structures, particularly urban drainage structures, a designflood hydrogiaph is basedon a storm duration equal to the time of concenfnis procedureusesthe rational method of Chapter 15 or trati-onof the wateisneA. the synthetic unit hydrographs of Chapter 12 along with a critical storm pattern pattern into the most critical sequence. produced by arranging the rainfall excess fn" SCS uies 24-hr durations for all urban watershedstudies' but for small watersheds, Durationsof approximately6hr or lessare satisfactory 10 days' to up periods of for depths the lengthsof stormiln large areasrequire storm valuesare availablefor durationsof from 2 to I0 daysfor locations Freque-ncy-based Similar data are also availablefor other selectedareas within the United States.to outsidethe United States.Generally,however,designcriteria for large damsrequire of storm depthsthat do not havefrequencylevelsassigned. estimates

Durationof RainfallExcebs
and the duraInitial rainfall during most stormsinfiltrates or is otherwiseabstracted, to the an amount.equal by duration rain actual the than less rainTf,is tion of excess for a estimated be 7s can duration rain Excess occur. abslractions initial time that from P precipitation and CN number curve the of function a as 6-hr storm

374

DESIGN CHAPTER16 HYDROLOGIC 125" 120"

i 7t?
"d, \
tv
,-o-

Bureau, Figure 16.6 The 10-mi2or less PMP for 6-hr duration (in.). (U.S. Weather NOAA.)

by the Sclrlt y-h:I" P is the storm Fig. 16.8. This family of curves was developed zero una CNis a losi parameterdefined1lhaptgr {. A CNof 100 represents O""pttr lossessothatZg:6hrforCN:l00.Table16.6isusedtofindthedurationof is the excess rain for any storm duration greater than 6 hr. The rainfall ratio precipitation total the by divided (T'iUle 16.7) absrractionP* losi before runoff to amount P. The time ratio from Table 16.6 is multiplied by the rainfall duration obtain Ze.

0 s0 r00
95"

2?0 3?0 490

90.

\ \

5P
85' 6u

Figure 16.6 Continued

Depth

available.

386

CHAPTER16 HYDROLOGIC DESIGN 1.0 0.9 0.8 X \ o 0.7 0.6

.=a

0.5

0.3
v.z

0.1

3 4 Time(hr)

Figure16.18 A6-hrdesignstormdistribution for SCSdam design.(After Ref. 12.)

country. For more preciseinformation on boundaries in a statehavingmore than one storrntype, contact the respectiveSCS State Conservation Engineer. The greatestpeak flows from small basinsare usually causedby intense,brief rains. Thesecan occur as distinct eventsor as portions of a longer storm. The 24-hr storm duration is longerthan neededto determinepeaksfrom small watersheds but is appropriatefor determiningrunoff volumes.In light of this, the SCS usesthem to studypeak flows, volumesof runoff, and direct iunoff hydrographs from watersheds normally studiedby the agency. Time distributionsfor PMP and other stormsusedin major structuredesigncan be constructed from Fig. 16.20.This family of curvesis usedby the U.S. Burlau of Reclamation6 in threegeographical zonesshownin Fig. 16.6.The corps of Engineers usesa distribution curve similar to Fig. 16.18for 6-hr SpS analyses. Triangular Distribution The simplest design storm distribution is a triangular shape. Because the depth,P, andduration,D, of rain are alreadyestablished, the peak intensity,l-u,, is 2PfD, foundby solvingfor the height of the triangularhyetograih as shownin Fig. r6.2L.The only remaining decision is the time to the peak,40. The ratio to/D has been investigated for a large number of storms at locationsin California, Illinois, Massachusetts, New Jersey,and North carolina. values range from about 0.3 to 0.5.17 Once the triangle is constructed, the intensitiesat regulaiintervals may be graphicallyor analytically determined for input to the rainfall-runoff rnodelbeing usedfor design. Blocked IDF Distributions A frequentlyusedprocedurefor developing a design storm distribution for short duration storms (up to about 2 hr) is to successively construct blocks of a design storm.histogramby using the appropriate intensity-

16.4 DESIGNSTORMS

Rainfall distribution

f--l ryp"r ffi rvwIe

i--l rypeu

llffil rvnerrl
rFooo

V v

(b)

for zonesI' Figure 16.19 SCS 24-hr rainfall distributions:(a) 24-hr rainfall distributions (After Ref' 16') distributions' rainfall for scs boundaries IA, II, and III and (b) approximate

388

CHAPTER16

HYDROLOGIC DESIGN 1.00 0.90

r ORO

;
E tr ;

o.7o
o6n nsn 0.40

E o.3o
o.20 0.10

Time (hr)

Figure 16.20 Distribution of 6-hr PMP for any area west of the 105' meridian. (After Ref. 6.)

Time, hr Figure 16.21 Triangular hyetograph. design duration-frequency curveto find therain intensities for A/, 2 At,3 L,t,etc.,increments of time and then to organizethese"blocks" of rain intensitiesin somepattern,usually symmetrical,.around the center of the storm, making sure that the area under the hyetographis equal to the designstorm depth, P, spreadover the designstorm duration,D. To apply the procedure, successive depths of equal-probability storms with durations of A,t,2Lt,3A,t,4Lt,etc.,aredetermined from the IDF curveandtabulated. Next, any of a variety of procedures,such as the alternating block method, the Chicagomethod,or the balancedmethod,are availablefor distributingtheseblocks and assuring that the total rain depth equals P. Most assumethat the highest

16.4 DESIGNSTORMS

389

highestoccursnext, and so on, intensityoccursin the middle of the storm,the second The balanced method, for block. center the from directions working out in both the IDF curvecould occur, from ia, intensity with storm a Ar-hr that exampli, assumes storm. This intensity is design D-hr the of middle the probability, during with equal the rain depth for Next, hyetograph. storm design of the block plotted as the middle is assumedto be a Its distribution curve. IDF from the obtained duration 2Lt is Ar-hr storm; the of the intensity the matching half first the with two-bar histogram rain depth for the 2 the of rest the spreading by is calculated secondhalf intensity for rain process is repeated The interval. Ar second the over unifoimly Al-hr duration a goal is to develop The D. to up . .. 4Lt, 3A/, of durations with , depthsfor storms of the middle the at centered any duration, of a storm such that storm hyetograph will havea total rain depthmatchingthe rain depthfrom the blockedIDF hyetograph, duration. IDF curve for the siven

Areal Distribution
precipitation depths can and do vary from point to point during a storm. Areal exceptin major structure variation in designstorm depth is normally disregarded in major structureanalysisis to selecta design(usually The usual approach designs. isohyetalpattem for the PMP or SPS depth and elliptical) or historic (transposed) precipitation depthsio the isohyetsin a fashionthat givesthe desiredaverage assign depih over the basin. The averagedepth is determinedby the isohyetal method illustrated in ChaPter2. Four majorlypes of storm patterns are shown in Fig. 16.22 fot areasup to of midwesternstormpatterns.ll 400 mir. Thesewere identifiedby iluff in his analysis The letters H andL representareaswith high and low precipitation depths,respecas generally tively. The typical isohyetalpatternfor SPSstormshasbeenestablished Valley Tennessee elliptical in itrapeas shownin nt. 16.23.This patternis usedby the

?b

Figure 16.22 Major types of storm patterns: (a) closedelliptical, (b) open elliptical, (c) multicellular; and (d) banded.(After Huff.tl)

390

CHAPTER 16 HYDROLOGIC DESIGN

t t Scale: miles

Figure 16.23 Generalized pattern storm.

lsohyet

Area enclosed(mi2) l1 45 114 279 546 903 1349 2508 4458

B C D E F G H
(After Ref.18.)

Authority (TVA)18for areasup to 3000 mi2.Variationsin the rainfall depth found in a standardproject storm will divergefrom a maximum at the storm centerto a value considerablylessthan the average depth at the edgesof the watershed boundaries. This variation can be determinedand incorporatedin the designstorm. A slightly modified isohyetalpattern for SPS storms is used by the Corps of Engineersre percentages asshownin Fig. 16.24.The shownfor isohyets A, B, . . . , G are multiplied by the 96-hr SPS depth to give an elliptical pattern with the desired average depth.Similarmapsfor 24.,48-, or 72-hr stormscanbe obtained simplyby modifyingthe 96-hrpercentages of Fig. 16.24.Thisis accomplished usingthe deptharea-duration curves in Fig. I6.24.For example, if a24-hr stormis used, first notethat theA isohyetof Fig. 16.24encloses an areaof 16nrr?. From Fig. 16,25thecorresponding SPSpercentage for a24-hr stormis 116 percentrather than the 140percentvalue used with a 96-hr storm. Thereforethe pattern percentages vary with the selected designstorm duration. An additional aid for constructingdesignstorm distributionsover smallermid(up to 400 mi') is presented westernl8 watersheds in Table 16.10.The ratio of maximum point rainfall to mean rainfall over the basin is provided and can be used to estimatethe maximum depth occurring at a storm centerif the mean areal depth is

16.5 CRITICALEVENT METHODS

391

: miles

Figure 16.24 GeneralizedSPS isohyetal pattern for a 96-hr storm' The pattern may be orientedin any direction and may correspondto the depthby a 96-hr storm. arearelation represented

lsohyet
A

Area (mi2)

B C D E F G
(After Ref. 19.)

l6 100 320 800 1800 3700 7100

known. Ratios for 50-, 100-, and 200-nr2 areasare equal to those in Table 16.10 multiplied by 0.91, 0.94, and 0.97, respectivd. For uniform rainfall the 95 percent With extreme Variability the 5 percent ratio ratios of the table are recornmended. conditions' The 50 percentratios approximateaverage applies.

ULrnoos EVENT 16.5CRTilCAL

392

16 CHAPTER

DESIGN HYDROLOGIC 20,000 10,000

\
+ 72-hperiod

5,000 4,000 3,000 2,000 1,000

o

\\\

s=
,*Nr"

500 400 300 200 100 50 40 30 20

10 40

\\ \ \

o I

\ \ \

60

120 100 80 value of SPS Isohyet Percentage

140

Figure 16.25 SPS depth-area-duration curves by 24-hr storm increments. (After Ref. 19.)

future event, and then design accordingly. These methods include the use of the probable rnaximum precipitation PMP, probable maximum flood PMF, record floods, and high storm depths, record high floods, multiples of frequency-based paleohydrology.

MaximumPrecipitation Probable
Probablemaximum precipitation depthsfor drainagebasinsin the United Statesare identified in Fig. 16.2. providedin the respectiveNational WeatherServiceHMRs2o reasonconsidered severe storm most The probablemaximum stormis deflnedasthe obtained is customarily flood probable maximum to occur.The resulting ably possible of the PMP preparedby the National and rainfall estimates by usingunit hydrographs rily'eather (see Figs. 16.6 and 16.26). Service2l

EVENT METHODS 393 16.5 CRTTTCAL ON 400 miz POINT TO MEANRAINFALL OF MAXIMUM TABLE16.10 RATIO
Mean rainfall(in.) Rainfallperiod (hr)

1.0

1.5

2.O

2.5

4.0

5.0

5% Probabilitylevel ratios (Storms with extreme variation in intensity)

0.5 1 2 3 6 l2 18
)4

5.20 5.50 5.80 6.05

48

0.5
I
L

3 6 12 l8
JA

2.66 3.03 3.46 3.77

48

0.5
I

2
J

2.38 2.75 3.15 3.46

6 t2 l8
)4 48 Sorrce.'After Huff.lr

r.4r 2.r8 1.70 1.48 1.80 2.29 1.55 1.90 2.44 1.99 1.61 2.53 r.72 2.r2 J.t I 2.69 1.83 2.25 2.86 4.01 1.90 2.33 2.96 4.14 1.96 2.40 3.05 4.27 2.08 2.55 3.25 levelratios 50% Probability (Storms timedislributions) with average 1.22 1.32 1.5'1 2.02 r.27 1.65 L39 2.r5 1.32 1.46 r.75 2.29 1.38 1.85 2.42 1.52 r.43 r.63 1.98 2.59 1.50 r.75 2.12 2.78 1.57 1.81 2.20 2.89 1.60 2.28 1.87 3.00 2.44 3.r7 .1.68 ,1.99 levelratios 95%Probability (Storms intensities) with uniform r.16 r.28 1.18 1.53 1.20 1.23 1,38 1.72 1.24 1.28 1.47 1.90 1.27 1.53 1.33 2.02 1.31 r.43 2.24 t.67 1.38 1.78 2.50 1.50 r.4l 1.53 1.89 2.67 r.43 1.58 1.92 2.77 r.47 r.64 2.04 3.07 3.00 3.21 3.38 3.54

1.30 1.35 1.41 1.46 1.52 1.60 1.65 r.69 1.77

t.26 1.30 r.33 r.36 1.43 1.50 1.54 1.57 1.63

1.22 1.25 r.28 1.30 1.35 1.40 1.43 1.45 1.50

1.16 1.20 1.24 t.28 1.33 r.39 1.43 1.47 1.53

|.14 1.18 Lzr 1.23 r.28 1.32 1.35 1.38 r.46

l.l2 1.16 1.19 t.22 1.26 1.30 1.32 1.33 1.38

1.13 l.l7 1.20 1.22 1.27 1.3 I 1.33 1.35 1.40

1.11 1.15 1.18 r.20 1.24 1.28 1.30 t.32 1.36

1.10 1.r4 1.16 1.18 1.21 1.25 1.27 1.29 1.33

that the suggests by Hershfield2l PMP advocated methodto estimate A proposed 24-hr PMP at a point be computedby the equation
PMP24:P*KS,

(16.1)

where PMPZ : . F : K: , & :

the 24-hr probablemaximum precipitation the meanof the 24-hr annualmaximumsover the period of record a constant equalto 15 the standarddeviationof the 24-hr annual maximums

Adjustmentsto the value of F and S, for the record length are noted by Hershfield. However, for appraisalpurposesthese adjustmentsprobably will not significantly alter resultsmore than 5-10 percent.

394

CHAPTER16

HYDROLOGIC DESIGN

zo.t

,/

\\ - - - - -"..20.1(3)

ptvtp(in.). (1) Alexandria, Figure16.26 Twenty-four-hour 2000-mi2 LA, (3)Elba, (2)Eautaw, June 13-17,1886. AL, April 15-18,1900. AL, March (4) Yankeetown, (5) Altapass, l1- 16,1929. NC, FL, September 3=7, 1950. (AfterRef. (6)Jefferson, July13-17, 1916. 10-13,1878. 18.) OH,September The U.S.Bureauof Reclamation underwentconsiderable evaluationof its design criteria for new damsand for safetyevaluationof existingdams,following the Teton, Idaho, dam failure in 1976.The policy adoptedfor modification of existingdamsis first to determinewhether they will accommodate the peak dischargeof the PMF without overtopping.In addition, the dam and appurtenant featuresmust accommodateat leastthe first 80 percentof the PMF volumewithout failure. For embankment dams,failure is assumed to occur if overtoppinglevelsare reached.

Recordedbdremes-Creager, and Crippenand Bue EnvelopeCurves

Where frequency-based methodsof PMP/PMF studiesare unwarranted,designfor critical eventscanbe basedon the greatest recordedrain or flood flow for the location. Similarly, tablesor curvesof flood data can be developed to give the maximum floods of record in the region under study; seeCreagerflood envelope curvesinFig. 16.27.

16.5 CRITICALEVENT METHODS 10.000 5,000 3,000 2,000

1,000

395

$ zoo
q? 100

500 300

s 0
ii JU

20 10 5 3 2
I

= {6gtrQs9ae

o 048 -11

N-h

o R93' 8 8 8 8 8 3 8 3 I i;;". . 6 -A N 6
Drainage area (mi2)

o o *

o o o o N o

o o h

o o o

Figure 16.27 Creagerenvelopecurves: O peak inflow for Harza Projects; 'recorded (1) Congo at Inga, Congo.(2) Tigris at Samarra,Iraq. (3) unusualflood discharges. (4) Tigris at Eski Mosul, Iraq. (5) Jhelumat Mangla, Pakistan. Caroni at Guri, Venezuela. (6) Diyala at DerbendiKhan, Iraq. (7) GreaterZab at Bekhme,Iraq. (8) Surinameat (9) Lesser Brokopondo, Suriname. Zab at DokenDam,Iraq. ( 10)PearlRiver,U.S"A.(11) Cowlitz at Mayfield, U.S.A. (12) Cowlitz at Mossyrock,U.S.A. (13) Karadj, Iran. (14) Agno at Ambuklao,Philippines. (15) Angat, Philippines. (16). Tachien, Formosa. (Nole.'Curves taken from Hydroelectric Handbook, by Creagerand Justin. New York: Wiley, 1950.)

In cases where estimates of PMP have not been made. volumes of rainfall to be expected can also be approximated from Creager rainfall envelope curves of the world record rainfalls as depicted in Fig. 16.28. Maximum flood flow data for 883 sites up to 25,900 sq km formed the basis for the Crippen and Bue envelope equation given by
4o : A)31 6rlosA+caGog A)2+ca(log lgfc r+

(16.2)

where qo is the maximum flow (m3/sec), A is the drainage area (sq km) and the coefficients are from Table 16-11 usingFigure 16.29.

StandardProjectStorm
The standardproject stormis anotherrainfall depththat is usedin the designof large dams. This value is usually obtainedfrom a survey of severestormsin the general vicinity of the drainagebasin. The storm selectedas the SPS may be oriented to produce the maximum amount of runoff for the SPF. Alternatively, severestorms experiencedin meteorologically"similar" areascan be transposedover the study
atea.

396

CHAPTER16

DESIGN HYDROLOGIC 2000 1000 600 400 200

/:fr>y%^"1",
ndia

? 1oo
; 6 0 s 4 0
& 2 0 10 6 4 2
Fr-

g
Ihra]l, TX imethport, I A -

runkiKO , Formo t a aquio, Pldl .ppineIr landr

ffi"*"[,,"1,

anlalca

I Unionville \4d. 2 4 6810 20 4060

tri--ftTl- tIT ft--J-f-1lHoit,MOreaDe Arses, Bglqeqla ?oint,Jamaica I | | I Plumb.

L{

-n"irpfii,W

ssen, Bal

3
j

6 Hours

t2

24

10 2030

9t2
Months

24

+.>+/>

DaYs
Dulatron

Figure16.28Creagercurvesofworld'sgreatestrainfalls.(AfterRef.18')

AND BUE PEAK DISCHARGE FOR CRIPPEN TABLE 16,11 COEFFICIENTS CURVES ENVELOPE Fig.16.29 Region
I

Coefficients

limit(sqkm) Upper 26000 7800 26000 26000 26000 26000 26000 26000 26000 2600 26000 18100 26000 26000 50 2600 26000 2600

c1
3.203865 3.4'10923 3.330746 3.258400 3.126412 3.500489 3.326333 3.236183 3.503734 3.314692 3.231389 3.596209 3.461373 3.07349'l 3,451746 3.s65536 3.389030 3.743026

c2
.8049163 .74'72908 .8443r24 .8906783 .796472r .9123848 .8503960 .9193289 .8054884 1.0386350 .8867450 .8806263 .8519276 .64'727rO .9718339 .9699340 .9445212 .7918884

c3

c4

2 3
4 5

6 '7 8 9 10 11 12
IJ

I4 15 16
L I

Nationwide

-.002975'7 -.0394382 - . 0 5 5 1 7 8 0 -.0000965 -.0642062. -.0021362 -.0870959 .0022803 -.0899000 .0022'744 -.1013380 .00496r4 -.0998'74'7 .0042129 -.0947436 .0029486 -.0890172 ,0018961 -.059'7463 -.0042542 -.102053s .0045531 -.0747598 .0000138 -.1094456 .0058948 -.0038285 -.0252243 -.00s7110 -.0617496 -.0034'776 -.0649503 - . 0 6 7 8 1 3 1 -.002'7647 .0244991 -.0192899

Source: AfIer Crippen,J. R., and C. D. Bue, PaPer1887, 1977. WaterSupPlY

..Maximum Flood flows in The ConterminousUnited States,''U.S.G.S

o !

3 F 6'* *
o F F : o o

r ll

k c
6 -

S B (h E
u)
f O

+
r N

+ Q E ( o

e a e
6 Fcr
F * ,'! o

h d 6 h

6 o d d n

+ $ i r N n

i h n i o N d d r ; O N N n h n

oa

3 .* @'a
9 -

q q q q q
@ n h a r : . f + d r r r r @ r
i O O i O

FA g 3 " ' o
A * = v O a ' o ^

: < = x x x
Y Y Y - -

!s E

Jjas5
:

6.?f)

o + s o @ 6 h n o

b.E 9q g 8d' ! o F
o 9

@ + d r * @ d O N N O

$ N O

>

f J

g =

E =

a ^ z +

o i n N O O N 6

* E -

r r;+\c;-; d o d i r
E
d N i h O o o d s

= ^^ E o e*v 9
d

r d o i v i . j

; s H o+ + . . , j . . l 6
. L ( E v
: d N i

F< a'

6 N r o n r 6 h o

= E ^
E E e

qii:E *E5o
N d c n N d

6
t r

Eec'
-

o 6 n h n n n d l q d l

^ ^ ^ o d

F 6 -

E= P * ' E 9 - E =

- . o

c^

d O @ d * : f + $ h

\ \ 9 n
o r

E PF a tr E g6 I 3 = E . Eg J
E o =

\ q c t \ q
N h o o a

4 . 2

? X

@ 9 0 - o * < i d + s o { 6

6 Pu "
. E o g S t
J @ E

* h h o n n r

s i

* i
i

d * o o
* * i 6

O * 6

9
t r F F {

o ^ c +

N : O O A E

5 5 - E d

408

DESIGN CHAPTER16 HYDROLOGIC

V:

i:::\

including storm Figure 16.36 Children's artist rendering of urban undergroundsysteT, with perReprinted Macaulay. David by 1976 A driins. From UNDERGROUND,Copyrighl missionof HoughtonMifflin Co' Al1 rights reserved'

and and the storm sewer systems, detention in gutters, house drains, catchbasins, locations' landscaped interceptionin extensively in urban storm drain designare: for accounted normally Two items many 1, Infil,tration. The ability of the soil to infiltrate water dependson of the soil as noted in chapter 3. The rangeof valuesgiven characteristics in the following table is typical of variousbare soils after t hr of continuous rainfall' RATEs rvircnr- |NFILTRATIoN
Soilgroup
High (sandy,oPen-structured) Intermediate(loam) structured) Low (clay,close-

(in./hr) Infiltration 0.50-1.00 0.10-0.50 0.01-0.10

thesevalues3 to 7'5 times' The influenceof grasscover increases

ANALYSIS 16.8 FLOODPLAIN

409

such to be 0.10 in. for pervioussurfaces 2. Retention This is usually assumed surfaces' as lawns and normal urban pervious Developmentof hydrologicparametersfor designof storm sewerpipes, street gurrers, or detentionbasinsis by the rational method(or modified rational rnethodor are adequate, hydrographs iee Chapter25) when peak flow ratesand approximate greater when methods synthesis unit hydrographand kinematic wave hydrograph The latter usually involveuseof public domainor vendor-developed detail is needed. stormwaterdesignsoftware.The hydrologicaspectsof computerizedhydrologicdein this text' signtools are deiailedin Chapter25. In addition to the material presented ILLUDAS' method, rational modified descriptionsof usesof the rational method, TR-sj, SWMM, DR3M, and other tools in designingurban storrrtdrainagefacilities Additionally, in numerousurban drainagedesigntextsand handbooks. are addressed offices have engineer's county and of transportationor city many statedepartments Society American the well, As tocattyapplicabledrainagedesignmanuals. developed i'model" adaptalocal for manual design drainage a of Civil Engineeishas developed of urhan discussion the Finally, tion, availableby contacting ASCE in New York. "shopper'sguide" to urban drainageanalysis modelsin Chaptlr 25 includesa useful and designsoftware.

16.8 FLOODPLAIN ANALYSIS over the years,the federalgovernDue to heavymonetary and other floodplainlosses and meththe nation's waterways of floodplains of ment hasbeenconductingstudies incauses that overdevelopment pieventing prop"tty and ods of protecting life und in ingredient key is a Hydrology flooding. widespread water levelsand more creased open dams, of effects studying rates, flow potential these studies for identifying volumes and determining on hydrographs water control structures and.other channels, and waterways the by conveyed and stored safely be of floodwatersthat will needto floodplains.

U.S.NationalFlood lnsuranceProgram(NFIP)
In 1968, the U.S. Department of Housing and Urban Development(HUD), later to called the FederalEmlrgency ManagementAgency (FEMA), initiated the NFIP of mapping with floodplains of identify flood hazard areasand to provide occupants tolorv-costflood insurance'The NFIP requireslocal the flood-proneareasand access programsthat preventdevelgou"rn-"nt5, to adoptand implementflood management of national standards' opmentsin excess Since the inception of the National Flood InsuranceProgram, flood hazatd The areas have been mapped in over 18,000 communities in the United States. maintenance a to converted since has and programcost over $t.O Uittion to complete occur'-Each of these effort of updating and expandingthe maps as developments peak flow rates for of evaluation detailed studieshas required eithei approximateor baseflood, has the called discharge, a range of recurrenceintervals. The 100-year the baseflood by occupied floodplain portion ofthe The in all cases. beendetermined

4'10

oHAPTER 16 HYDRoLoGIc DESIGN

has been mapped,allowing communitiesto determinewhether a property is in the 100-yr floodplain,and in many cases, what water surfaceelevationwould be experiencedat the property during the baseflood. Figure 16.37illustrates the typical NFIP mappingand floodplain management procedure.Surveyedvalley and channelcross-sections are used in determiningthe 100-yr flow depth, allowing the hydrologistto delineatethe lateral extentof flooding duringthe 100-yrflood. Then afloodwaywidth is generallydetermined asthat portion of the floodplain that is reservedin order to dischargethe 100-yearflood without cumulativelyincreasingthe water surfacemore than 1.0 ft. This procedureis illustrated in Fig, 16.38.The floodwayis most often centeredover the main stream channel,but can be offset or even split into severalzones. Development within the floodwayis allowed only if compensated by relocating the floodway or mitigating the water surfaceincreasedue to the development. The portion is that of the floodplain outside the floodway in which developflood fringe ment is allowed,up to a point of full encroachment by buildings,roadbeds, berms,and so forth. As much as sevento ten percentof the total land area of the United States lies within the 100-year floodplain.The largestareasof floodplainare in the southern parts of the country, and the most populatedfloodplainsare alongthe north Atlantic coast,the GreatLakesregion,and in California. The floodplain mapping effort produced a large amount of data and analyses useful to designhydrologists. The productsof the program include: 1. The 10-, 50-, 100-, and 500-year frequencydischarge for streams. 2. The 10-, 50-, 100-,and 500-year flood elevations for riverine,coastal, and lacustrine floodplains.

Floodway Flood Fringe Flood Fringe

',100_year" Floodplain

Channel

Figure 16.37 Definition sketchof floodplain delineations.

ANALYSIS 16,8 FLOODPLAIN

100Year FloodPlain Floodway Encroachment Flood Fringe

411

width' the floodway for determining Figure 16.38 Procedure 3.The100-and500-yearmappedfloodplaindelineationsatscalesranging to 1:24,000' from 1:4800 4. The 100-yearfloodwaydata and mapping' wave haz5. Coastalhigh hazard irea mapping 1*"u* subjectto significant ards). 6. FloodwaYflow velocities' risk zones' 7. Insurance This information is provided in the form of three products: |.FloodlnsurancestudyReportsprovidegeneralprogramandcommunity floodway data,tabulated. flood dischatge information,tauutatea background tabuinformation' surcharge datalncluding velocity, floodwaywidth, and and 100-' 50-' 10-' the of dati, and profi'les lated flood insurance'zone flooding' riverine for 500-yearflood elevationversusstreamdistances of the 1002. Flood InsuranceRateMaps (FIRM maps)provide delineations and500-yearfloodplains,basefloodelevations'coastalhighhazardareas' andinsuranceriskzonesonaplanimetricbaseatascalebetweenl:4800 and 1:24,000. 3.FloodBoundaryHazardMapsprovidedelineationsofthel00-and500floodplainand channelcrosssections locationsof surveyed yearfloodplains, floodwayon a of the 100-year and delineations usedin hydraulic analyses, 1:24'000' and 1:4800 baseat a scalebetween planimetricor topographic

Studies for FloodPlain HydrologY

NFIP studiesare basedon Flood flow frequency estimatesfor gaugedlocationsin Annual peak Type III (seechapter 27; analysisof streamflowrecords. log-pearson by FEMA' recommended flows and historical data arcfitted accordingto procedures

412

CHAPTER16

HYDROLOGIG DESIGN

through For ungauged locations,flood flow frequency'estimates are developed regionalfrequencyanalysis or throughrainfall-runoff modeling.Equationspublished to various by the U.S. Geological of variousfrequencies Surveyrelatepeak dfscharges shape,andland use.These drainage basincharacteristics suchassize,slope,elevation, equationsare developedusing multiple regressiontechniques(see Chapter 26) at gaugedsitesthroughoutthe region. (Chapter24) use syntheticrainfall hyetoRainfall-runoff modelingtechniques graphs.Storm-eventmodels, such as the Corps HEC-I and SCS TR-20 packages, employdesignstormsof particular frequencies and then mathematicallysimulatethe physical runoff process.The resulting peak dischargeis assumed to have the same frequencyas the rainfall.

U.S.Flood Hazards
Despiteconsiderable effort and expenditurein identificationof floodplainsand flood damage hazardareas, continueto resultin severe dam failuresand other catastrophies property, hurricanes, intense rainstorms, life, Floods from to and the environment. and rapid snowmelt or structure failure have all contributed to the loss of life. A tabulationof eventscausingmore than 100 deathsin the United Statesis providedin Table 16.15. As indicated, the majority are hurricane related, principally concentrated in the east-coast and Gulf of Mexico regionsas sfown in Fig. 16.39. Monetary losses from floodsare also large.Table 16.16showsa numberof past each,given in 1966dollars. U.S. floodsproducingover $50 million in flood damages produced in billions of dollars, disthese floods flood dam4ges Collectively, have I tributed through the yearsas shown in Fig. 16.40. The Federal InsuranceAdministration evaluatedthe floodplain areas in the and economicinformation, communitiesmappedby FEMA. By using demographic projectionsof future property at risk of flooding could be made.Results.suggest that property in have occurred of investments in flood damageable floodplains. billions value that Table 16.17lists the breakdown,by state,of estimated1990 development will be in harm's way.

Dam Break Hazards

' .-"-fabir- 16.18lists the outflow rates,peak depth, and storageat the time of failure for 18 significantdam failures in the United States.The death rate for dam failures is thosepeoplewho would relatedto the polpulation at risk (PAR). This term describes need to take someaction to avoid the rising water. Figures 16.41 and 16.42showthe losses as functions of PAR for low (lessthan 1.5 hr) and high (greater than 1.5 hr) advancewarning times, respectively.The high-warning-time losses are significantlyless.This strongly supportsthe incorporation of early warningand flood delayfeaturesin the designof any structure.Data used in plotting Figs. 16.41 and 16.42 are given in Table 16.19. resiTable 16.20providesa typical time line requiredfor alerting downstream dentsof a severestorm and potential dam failure. The valuesgiven are hypothetical, and apply to an assumed15-mi reach betweenthe storm center and the populated atea.

SUMMARY 413
1OO OR MOREDEATHSIN THE TABLE 16.15 FLOODSCAUSING UNITEDSTATES

Year
1831 1856 1874 1875 1886 1889 1893 1899 1900 1903 1903 1906 1909 1913 1913 1915 l9l9 l92l 1926 1927 1927 1928 1928 1928 1932 1935 1935 1936 1937 1938 1955 195'l 1960 1972 ' 1972 t1976

Streamor place BaratariaIsle, LA LA Isle Derniere, ConnecticutRiver tributarY TX Indianola, TX Sabine, PA Johnstown, Vic. GrandIsle,'LA PuertoRico TX Galveston, Central States HePPner,OR Gulf coast Gulf coast-New Orleans Miami, Muskingham,and Ohio Rivers Brazos River, TX Louisianaand TexasGulf coast Louisianaand TexasGulf coast Upper ArkansasRiver Miami and Clewiston,FL River Lower MississiPPi Vermont Puerto Rico FL Lake Okeechobee, San Francisco,CA PuertoRico Florida KeYs River, KS, NE RePublican United States Northeastern Ohio River New Englandcoast United States Northeastern Westcoast,LA PuertoRico Buffalo Creek, WV RaPid.Creek,SD Big Thompson,Co

Lives lost

Cause
Hurricane tidal flood Hurricane tidal flosd Dam failure Hurricane tidal flood Hurricane tidal flood Dam failure Hurricane tidal flood Hurricane tide and waves Hurricane tidal flood Rainfall-river floods Rainfall-river floods Hurricane tidal flood Hurricane tidal flood Rainfall-river floods Rainfall-river floods Hurricane tidal flood Hurricane tidal floo<l Rainfall-river flood Hurricane tidal and river flood Rainfall-river floqd Rainfall-river flood Hurricane tide and waves Hurricane tidal flood Dam failure Hurricane tide and waves Hurricane tidal flood Rainfall-river flood Rainfall, snowmelt-river fl oods Rainfall-river flood Hurricane tidal and river flood Hurricane rainfall-river floods Hurricane tide and river floods Hurricane rainfall-river floods Dam disaster Rainfall Rainfall

150 320 t43 176 150 2t00 2000 3000 6000+ 100+ 247 151 700 46'7 177 550 284 t20 350 100+ t20 300 2400 350 ,))\ 400 110

ro7
137 200 115 556 r07 t25 245 r39

414

DESIGN 16 HYDROLOGIC CHAPTER

Figure 16.39 Number by state of major hurricanesin the United States, NaI 899-1989.(Soarce.' tional Hurricane Center, National WeatherService, NOAA.)

Figure 16.40 Average in annual flood damages the U.S., 1916-85. (Source:National WeatherService,NOAA.)

Damages Damages (1985 $) Damages/200 million population (1985 $)

rg2o 25

30

35 40

45

50

55

60

65

70

75

80 8s

of five'Year Last Year Period

$50 MILLION TABLE 16.16 FLOODSRESULTINGIN DAMAGESEXCEEDING IN THE UNITEDSTATES Damage($ millions)

Year

Streamor place

dollars 1966 Dollars Contemporary

N.A."
.A

Cause
Rainfall-river flood Dam failure Hurricane tidal floods Rainfall and dam failure Rainfall-river flood Rainfall-river flood Hurricane rainfall-river floods Rainfall-river flood Hurricanetidal and river floods Rainfall-river floods Rainfall-river flood Rainfall-river flood Hurricane tide and waves Rainfall-river flood Rainfall-river flood Rainfall- snowmelt fl ood Rainfall-river flood Hurricanetidal and river floods Rainfall-river floods Rainfall-river floods Rainfall-river floods Huriicane tidal and river floodb Rainfall-river floods Rainfall-river floods Hurricane tidal and,rivet floods Rainfall-river floods Hurricanetidal and river floods Rainfall-river floods Rainfall-river floods Rainfall-river floods Rainfall-river floods Snowmeltfloods Rainfall-river floods Hurricane tidal floods Huiricane tidal and river floods Rainfall-river floods Rainfall-river floods Rainfall-river floods Rainfall-river floods Hurricanetidal and river floods Hurricane tidal floods Hurricanetidal and river floods Rainfall-river floods Rainfall-river floods Rainfall-river floods Hurricanetidal and river floods Rainfall-snowmelt river flood Rainfall-river floods Rainfall-river floods Hurricane tidal flood

River t844 Upper MississiPPi PA 1889 Johnstown, 1900 Galveston,TX and DelawareRivers 1903 Passaic 1903 Missouri River basin 1 9 1 3 Ohio River basin t913 Brazos and ColoradoRivers, TX 1921 ArkansasRiver 1926 Miami and Clewiston,FL t926 Illinois River 1927 New England 1927 Lower MississiPPi 1928 Puerto Rico and DelawareRivers 1935 Susquehanna r936 NortheasternUnited States t936 Ohio River basin 1937 Ohio River basin 1 9 3 8 New England sffeams 1938 California streams t942 Mid-Atlantic. coastalstreams 1943 Central states 1944 South Florida 1944 Missouii River basin 1945 Hudson River basin 1945 South Florida 1945 Ohio River basin 1947 South Florida t947 Missouri River basin 1948 Columbia River basin 1950 San JoaquinRiver, CA KansasRiver basin I 195 1952 Missouri River basin 1952 Upper MississiPPiRiver 1954 New England streams 1955 NortheasternUnited States 1955 California and Oregonstreams 1957 Ohio River basin 1957 Texas rivers 1959 Ohio River basin 1960 South Florida 1 9 6 1 Texascoast 1964 Florida 1964 Ohio River basin 1964 California streams 1964 Columbia River -North Pacific t965 South Florida 1965 Upper MississiPPiRiver 1965 Platte River, CO, NE 1965 ArkansasRiver, CO, KS 1965 New Orleans and vicinitY
,N.A. : not available. Council, 1968. Source:IJ.S.WaterResources

25 25 50 150 t28 t3
'7rl

1,161 84 100 273 N.A, 516 349 64 130

]I

N.A, 50 284 50 36
zzl

178 N,A. 90 185 374

3tl

150 418 125 100 28 172 63 52 24 54 34 60 178 to2 883 180 198 180 684 271 65 144 lt4 78 300 32s 106 t'73 289 r39 158 19l 61 322

996 3'76 294 103 N.A. rt7 N.A. 75 98 6l 88 N.A. 226 57 N.A. N.A. N.A. 216 879 405 72 188 r20 86 336
3+Z

r12 183
3lL 144 r62 N.A, 65 338

FLOODING, AT RISKFROM VALUE PROPERry TABLE 16.17 ESTIMATED ON 1990COSTS BASED RANKED IN DECREASING ORDER. Rank 1 2 3 5 6 7 8 9 l0 tl 12 13 L4 15 16 l7 18 1,9 20 2l 22
z3
/)A

State
California Florida Texas Louisiana Arizona New Iersey New York Illinois Massachusetts Pennsylvania Virginia Maryland Washington Ohio Michigan North Carolina Wisconsin Georgia Connecticut Missouri Indiana Minnesota Nebraska Oklahoma Alabama South Carolina Tennessee Colorado Oregon Mississippi New Mexico Kansas Iowa RhodeIsland Kentucky North Dakota Urah Nevada Arkansas Delaware Maine West Virginia New Hampshire South Dakota Idaho Hawaii Vermont Wyoming Montana Alaska

value,X $1000 Property 163,323,1,92 131,548,814 72,376,950 45,402,322 45,094,183 38,945,265 32,005,900 26,880,755 23,8t3,115 18,888,390 17,441,420 16,330,448 t6,245,009 t5,273,r47 13,449,078 12,993,067 12,r8r,725 11,832,494 tt,1r7,290 11,654,861 r0,786,741 10,655,t64 t 10,360,574 9,501,778 9,274,903 9,220,305 8,037,425 7,137,757 6,861,790 6,134,073 5,519,278 5,279,t94 5,26r,678 4,312,117 4,r70,637 3,924,872 3,812,936 3,437,813 3,005,rs0 2,954,467 2,416,322 2,098,262 t,991,453 1,430,610 1,39t,498 1,323,90s 1,091,099 1,081,460 881,661 647,81,8

25 26 2'7 28 29 30 3l
JZ

33
3+

35 36
3 t

38 39 40 4I
4)
43

44
4f

46 47 48 49 50

"status of Floodplain Hazard Evaluation Under the National Flood Insurance Program," Source: B. R. Mrazik, Emergency ManagementAgency, Washington,DC, 1986. _ fed9r91

17.3 SUBSURFACE DISTRIBUTION OFWATER 429 locations it has become more important than overdrafts of groundwater supplies. Today,the hydrologistmust be concernedwith both the qualitli and quantity aspecrs of groundwater.Furthermore, there is emerging an increasing specialization in groundwaterquality modeling.This latter type of modelingis ge;erally beyondthe scopeof this text but information on this topic may be founOin'Refs. j-6.

17.2 GROUNDWATER FLOW-GENERAL PROPEFTIES

Understanding the movementof groundwaterrequiresa knowledgeof the time and spacedependencies of the flow, the nature of thJporous medium and fluid, and the ,boundaries of the flow system. Groundwater flowsareusuallythree-dimensional. Unfortunately, the solutionof suchproblemsby analytic methodsis complexunlessthe systemis symmetric.7,8 In othercases, space dependency in oneofthe ioordinate direciions.uy 6" so slightthat assumption of two-dimensionalflow is satisfactory. Many problemsof practical importancefall into this class. Sometimes one-dimensional flow can be assumed, thus further simplifying the solution. Fluid propertiessuchas velocity, pressure, temperature,density,and vis6osity often vary_intime and space.When timi dependency occurs,the issueis termed an unsteady flow problem and solutionsare usually difficult. On the other hand, situations wherespacedependency aloneexistsare iteady flow problems.Only nomogeneous (single-phase) fluids are consideredhere. For a discussion of muliiple phase flow, Refs.5 and g are recommended. Bou-ndaries to groundwater flow systems may be fixed geologicstructures or free water surfaces that are dependent for their position on the stateol the flow. A hydrologistmust be ableto definetheseboundaries mathematicallyif the groundwater flow problemsare to be solved. Porousmedia through which groundwaters flow may be classifiedas isotropic, anisotropic,heterogeneous, homogeneous, or severalpossiblecombinations ofthese. An isotropic medium has uniform properties in all directions from a given point. Anisotropic mediahaveone or more propertiesthat dependon a given diiection. For example, permeabilityof the medium might be greateialong a horizontal plane than alonga vertical.one.Heterogeneoas mediahavenonuniform-properties of umrotropy or isotropy, while homogeneous media are uniform in their ihaiacteristics.

17.3 SUBSURFACE DISTRIBUTION OF WATER

4;

GEoLocy sotLs,AND 17 GR.,NDWATER, cHAprER and extends l. Soilwater zone.A soil water zonebeginsat the ground surface downward through the major root band. Its total depth is variable and

'

in this region: hygroscopicwater, which is adsorbed may be encountered from the air; capillary water, held by surfacetension; and gravitational soil water draining through the soil' water, which is excess from the bottom of the soil-watetzone 2. Intermediate zone.This belt extends to to the top of the capillary fringe and may ghangefrom nonexistence link The zoneis essentiallya connecting severalhundredfeet in thickness. between a near-ground surface region and the near-water-tableregion through which infiltrating fluids must pass. from the watertable (Fig.I7 '2) to 3. Capiltary zone. Acapillary zoneextends in the soil. by the capillary rise that can be generated a height determined The capillary band thicknessis a function of soil textureand may fluctuate not only from region to region but also within a local area. 4. Saturatedzone. In the saturatedzone, groundwaterfills the pore spaces completelyand porosity is thereforea direct measureof storagevolume. Part of this water (speciflcretention) cannot be removedby pumpirygor tensionforces.Specificretention ofmolecular and surface because drainage is the ratio of volume of water retained againstgravity drainageto gloss volume of the soil.

:ffi::J; ,:'##l":ff l;"lhr?ff fi',,nH#,:?'l*?,T#*.Ji'.i,'"1:f,

Waterthat can be drainedfrom a soil by gravity is known as the specificyield. asthe ratio of the volumeof waterthat can be drainedby gravity to the It is expressed grossvolumeof the soil. Valuesof speciflcyield dependon the soil particle size,shape and distribution of pores, and degreeof compactionof the soil. Averagevaluesfor alluvial aquifersrange from 10 to 20 percent. Meinzer and others have developed procedures for determiningthe specificyield.12

CONSIDERATIONS 17.4 GEOLOGIC

volumesand flow ratesrequiresa thoroughknowlof groundwater The determination basin.In bedrock areas,hydrologiccharacterisedgeof the geologyof a groundwater tics of the rocks,that is, their location,size,orientation,and ability to storeor transmit to basinsoften containhundreds rock areas, water,mustbe known. In unconsolidated fill depositsthat originated to unconsolidated of feet of semiconsolidated thousands quantitiesof areas.Suchfllls often contain extensive from the erosionof headwater of thesebasin fills must be evaluated. storedwater. The characteristics A knowledge of the distribution and nature of geohydrologicunits such as or to proper planningfor development aquifurs,aquifugis, andaquicludesis essential must be In addition,bedrockbasinboundaries supplies. of groundwater management locatedand an evaluationmadeof their leakagecharacteristics. An aquifer is a water-bearingstratumor formation that is capableof transmitAquifers may be considered ting waterin quantitiessufficientto permit development.

CONSIDERATIONS 17,4 GEOLOGIC

P^^L ...u,ritlge

431

arc?

Discharge area

E] F a

(/)
TT]

F z p

(b)

Figure 17.2 Deflnition sketches of groundwater systems and mechanisms for rechargeand withdrawal: (a) aquifernotationt0and (b) componentsofthe hydrologic II cycleaffectinggroundwater. as falling into two categories, confined and unconfined, depending on whether a water table or free surface exists under atmospheric pressure. Storage volume within an aquifer is changed whenever water is recharged to, or discharged from, an aquifer. In the case of an unconfined aquifer this may easily be determined as AS:SIAV

(17.1)

432

CHAPTER17

SOILS,AND GEOLOGY GROUNDWATER,

where AS : the changein storagevolume specificyield of the aquifer S, : the average the original watertable andthe AV : the volumeof the aquiferlying between water table at somelater specifictime For saturated, confined aquifers, pressure changes produce only slight modificationsin the storagevolume. In this case,the weight of the overburdenis supportedpartly by hydiostatic pressureand somewhatby solid material in the uqoif"t. When hydrostaticpressurein a conflned aquifer is reducedby pumping or with the result its compression, causing theioad on the aquiferincreases, othermeans, a small causes pressure also hydrostatic the that somewateris forced out. Decreasing confined For of water. release additional an expansion,which in turn produces in terms of a storagecofficient S", definedas the aquifers,water yield is expiessed per unit surfaceareaof aquiferper unit volumeof wateian aquifei takesin or releases of 17.2 illtstrates the classifications Figure changein head normal to the surface. aquifers. ratesofyield, thereare strataexhibitingsatisfactory In additionto water-bearing large quantitiesof may contain that strata and impermeable also non-water-bearing permit effectivedevelopto enough high not rates are water but whosetransmission an aquicludeis of water; devoid and ment. An aquifugeis a formation impermeable an imPerviousstratum. In the following three chapters,the mechanicsof groundwaterflow and the presentedall elementsof groundwatermodelingwill be introduced.The techniques a numerical Before modeled. be to physical system of the dependon a=knowledge framework This be devised. must framework a conciptual model can be developed, of aquifers, types geology; the and must take into accountthe region's topography characteristics; and variations, lithological lateral extent,boundaries, their thickness, areas,their rates of dischargeand the nature and extent of rechargeand dischar_ge table'' recharge;and the elevationof the water

Topography
to know something it is essential operates, how a groundwatersystem To understand showingall surface compiled map shouldbe A topographic aboutthe region's surface. ponds,as well and/or water bodies,including streami, iakis, and artificial channels observation pumping wells, as land surfacecontours.Furthermore,an inventory of types of identifying as such wells, and explorationwells shouldbe madefor purposes table water determining locationsandrates;and soilsand rocks,pinpointingdischarge elevations.

GeologY Subsurface
governsthe occurrenceand movement The geologicstructureof a groundwater'basin of the grirndwater withinlt. Specifically,the number and types of water-bearing formations, their vertical dimensions,interconnections,hydraulic properties, and Once the before the systemcan be analyzed.u outcrop patternsmust be understood lower and conditions have been identified, contour maps of the upper subsuriace

SUMMARY 433 boundariesof aquifers,watertable contourmaps,and mapsof aquifercharacteristics can be prepared.Well-drillers logs, experimentaltest wells, and other geophysical explorationmethodscan be usedto obtakr the neededgeologic data.s-e'13'14

17.5 FLUCTUATIONS IN GROUNDWATER LEVEL

Any circumstance that alters the pressureimposedon undergroundwater will also factors,changes in stream and causea variation in the groundwaterlevel. Seasonal pressure river stages, changes, winds, tides,external evapotranspiration, atmospheric all may produce loads,various forms of withdrawal and recharge,and earthquakes fluctuationsin the water table level or piezometricsurface, depending on whetherthe aquifer is free or confined.eIt is important that an engineer concernedwith the development and utilization of groundwatersuppliesbe awareof thesefactors.The engineershould also be able to evaluatetheir importance relative to operation of a speciflcgroundwaterbasin.

17.6 GROUNDWATER-SURFACE WATERRELATIONS

Notwithstandingthat water resourcedevelopment has often been basedon tL pr"it mustbe emphasized that these dominantuseof either surface wateror groundwater, Changesin one two componentsof the total water resource are interdependent. and componentcan havefar-reachingeffectson the other. Coordinateddevelopment management of the combinedresourceare critical. Linkage betweensurfacewaters effects in all regionalstudiesso that adverse and groundwaters shouldbe investigated understood. can be noted if they exist and opportunitiesfor joint management In Part Three it was shown how surface stream flows are sustainedby the groundwater are replenished resource,and it was also pointed out that groundwaters by infiltration derived from precipitation on the earth's surface. Undergroundreservoirsare often extensiveand can serveto store water for a recharge, multitude of uses.If withdrawalsfrom thesereservoirsconsistentlyexceed results.By properly coordinating mining occursandultimate depletionof the resource optimum regional water resource the use of surfacewater and groundwatersupplies, development seemsmost likely to be assured.Several studiesdirected toward this usehavebeeninitiated.r5'16 coordinated

r Summary
The importance of groundwaterto the health and well-being of humans is well documented. Groundwateris a major sourceof freshwaterfor public consumption, industrial uses, and the irrigation of crops. For example, more than half of the freshwaterused in Florida for all purposescomes from groundwatersources,and about 90 percent of that state'spopulation dependson groundwaterfor its potable is clear.Quantity and quality dimenthis resource watersupply.The needto husband sionsare both important.

4g4

CHNPTENTZ

SOILS,AND GEOLOGY GROUNDWATER,

Groundwaterprotectionandmanagementpracticesmustbebasedonanunder. groundwateris^distributed standing of groundwatersources,the rianner in which of the region' topogiaphic,and soil characteristics geologic, below the earth's surface, andtheinterconnectionsbetweengroundwaterandsurfacewatersources.

REFERENCES
"Ground Water,"Mech'Eng' (Jan' 1960)' 1 . J. G. Ferris, ..GrouniwaterProtection," Final Report of the National The Conservationrounoaiion, L. D'C'' 1987' GroundwaterPolicy Forum, Washington, p"rru.u, diUiu- G' Gray' and Gegpe-F Pirltder'.Groundwater e. Ra'ymond F. Wood, E. CoitaminationfromHazardousWastes'EnglewoodCliffs'NJ:Prentice-Hall'1984' in the United States. R'pu,ri"t, unJ f. euurt" s, Grouidwater Contarnination ;. ffi;;; Press' 1983' ehitadelphia:University of Pennsylvania R.A.FreezeandJ.A.Cir"..y,Groundwater'EnglwoodCliffs'NJ:Prentice-Hall'19'19' of llydrology' New York: McGraw-Hill' 1993' D. R. Maidment (ed.), ii"aLook ,,Niurnerical Basins,"InternaModeling of Groundwater J. Boonstraand N. A. O" nlJa"r, 1981'' Netherlands' The tional Institute for Land Reclamationand Improvement' 1965' Wiley' New York: R. J. M. DeWiest,Geolrydrology' Wiley' 1960" D. K. Todd, GroundwaterHydrology' New York: ..croonowu# n"gi#, of the United States," GeologicalSurvey water R. c. Heath, U'S' GovernmentPrinting Office' 1984' Supply PaperNo. 22a2' iishiniton, D'C-: "The Report to Congress:^Waste DisposalPracproteciion ig"""y, 1 1 . U.S. Environmentat summary'U'S' EPA' PB 265-364' tices and Their Eft'ectson Groundwater,"Executive 1977. States,"U'S' Geological tz. o. E. Meinzer, "The occurrence of Groundwaterin the united 1923' Survey,Water-SupplyPaperNo' 489' "A GeneralizedGraphical Method Evaluating -for 1 3 . H. H. Cooper, Jr. and i. n. lacoU, Union Geophys' Am' Trans' Formation constants una so*-u.izing well-Field History," 27, 526-534(1946). "Outline of Methodsfor EstimatingCrounlw11e1fupplies"' U'S' Geolog1 4 . O. E. Meinzer, D C:' 19?2ical Survey,Water-SupplyPaper638-C, Washington' "Conjurrliu" Op"tution of Dams and Aquifers"' Proc' ASCEJ' Hyd' Div' 1 5 . Nathan Buras, S9(HY6) (Nov. 1963). 16. F'B.Clendenen,..AComprehensivePlanfortheConjunctiveU{ilizationofaSurface water SupplyDevelopment:Solano Reservoirwith undergroun'Js,orug" for Basin-wide 1959. project california,,, po.to. oi Eni thesis,university of california, Berkeley,

1B Chapter

of Flow Mechanics

r. Prologue
The purposeof this chaPteris to: i I . . I Presentthe principlesof groundwaterflow. Describesoil propertiesthat affect groundwaterstorageand movement. Describethe relevanthydrodynamicequations. Relatethe mechanicsof groundwaterflow to modelingregional groundwater systemsand calculating flows to wells and other groundwater collection devices.

18.1 HYDROSTATICS
of a saturatedmedium is under pressure(calledpore Water locatedin pore spaces in the mediumat a point by insertinga piezometer be determined cin pressure), which seenthat pore pressure can be it (Fig. considered, is 18.1) A If Location of interest. given by is P:h"l where p : the pore pressure(gaugepressure) from the point to the water table h o : the headmeasured water of weight the specific v: positive or negative,dependingon whetherthe pressure Pore pressureis considered the point under consideration. above(poJitive)or below (negative) headis measured the total head or piezometric head abovethe If an arbitrary datum is established, datum is (18.2) Po:Z'th wherePois known as the piezometricpotential. In Fig' 18.1 this is equal to ho 1- 7o zone.The zone andza - h6forPointB in the unsaturated for poiniA in the saturated

(18.1)

43r.

436

CHAPTER18

OF FLOW MECHANICS

preshydrostatic showing Figure 18.1 Definitionsketch medium. in a porous sures or vacuum(negative,pore tension of A while -hu denotes Iermh"is the pore pressure pressure) at B.

FLOW 18.2 GROUNDWATER

Analogiescan be drawn betweenflow in pipesunder pressureand in fully saturated confinid aquifers.The flow of groundwaterwith a free surfaceis also similar to that flow system in an open channel.A major differenceis the geometryof a groundwater The systems' or channel pipe flow channel as compared with common hydraulic comsection cross a discharging problem "un eutily be recognizedby envisioning posedof a numbei of small openings,eachwith its own geometry,orientation, and iir" ,o that the flow velocity issuingfrom eachpore varies in both magnitudeand direction. Difflculties in analyzingsuch systemsare apparent. Computations are of fluid and medium propertiesover a given usually basedon macroscopicaverages area. cross-sectional flow problemsare density, in groundwater Unknown quantitiesto be determined to exist'r la In are assumed conditions pressure, and velocity if constanttemperature so the number of working variablesis general,water is consideredincompressible, later relative to the storagecoefficientfor ieduced.An exceptionto this is discussed here will be placed on the flow of water in a a confined aquifei. Primary emphasis porous medium. saturated

18.3 DARCY'SLAW
Darcy's law for fluid flow through a horizontal permeablebed is statedas' O: - K A dh dx (18.3)

LAW 18.3 DARCY'S areaincluding the spaceoccupiedby the where A : the total cross-sectional porous material K : the hydraulic conductivity of the material the control areaA Q: the flow across In Eq. 18.3 h : z * where h - the piezometrichead the elevationabovea datum p : the hydrostaticpressure C _ an arbitrary constant in Eq. 18.3, If the specificdischarge S = QIA is substituted q: -K*(,.t)

437

P-+c v

(18.4)

( 18 .s )

, Note that 4 also equalsthe porosity n multiplied by the pore velocity Vo.Darcy's law is widely used in groundwaterflow problems.Severalapplicationsare illustrated in later sections.

No is definedherein as
No: PQd lL

(18.6)

where q d P p

: : : :

the specificdischarge the mean grain diameter fluid densitY dynamic viscositY

For many conditions of practical importance (zones lying adjacent to collecting devicesare an exception),Darcy's law has been found to apply' to Ohm's law Of specialinterestis the fact that the Darcy equationis analogous

' : (*)u
where i:thecurrent R : the resistance E : the voltage

(18.7)

asareK andI f R, andE anddhldx.The similarity current and velocity are analogous, of the two equationsis the basis for electric analog models of groundwaterflow svstems.2'3

438

CHAPTER18

OF FLOW MECHANICS

EXAMPLE 18.1

: . Water temperaturein an aquifer is 60'F and the rate of water movement t .2 ft I day particle diameterin the porous medium is 0.08 in. Find the Reynolds The average number and indicate whetherDarcy's law is applicable. . Solution. Equation 18.6 gives the Reynoldsnumber as

X*: #
This may also be written as
qd N--'
t)

Converting From TableA.2 in Appendix A, o is found to be 12l X 10-s ft2lsec. : : X 10-5' The 1.39 gives I.2/86,400 the velocity4 into units of ftlsec 4 : : in the values these Substituting 0.0067. 0.08/12 meangrain diameterin ft equation,we obtain
Nn:

r.39x10-sx0.0067
1 . 2 1X 1 0 - s 0.0077

SinceN,, < 1.0,Darcy's law doesapply' rl

18.4 PERMEABILITY
into The hydraulic conductivity K is an important parametrthat is often separated product The fluid. the to other the medium, one related to the two components,
k:Cdz

(18.8)

can be written

X:4
lL

(18.e)

of intrinsic permeability areL2. Sincevaluesof k given asft2 or cm2 Dimensions known as the darcy has beenwidely adopted. are extremelysmall, a unit of measure 1 darcy : 0.987 x 10-s cm2 or 1.062x l0-1r f* hydraulic conductivity are reportedin the literature. Severalwaysof expressing the standardcoefficientof permeabilityK" as defined has Survey The U.S. Geological passingthrough 1 ft2 of medium under a per water gallons of day - the nunber of

446

CHAPTER18

OF FLOW MECHANICS

" If Eqs. 18.46 are substituted into Eq. 18.50,then

#o;+frat:o

(18.s2)

must be a constant.A seriesof is equal to zero,and ry' The total differential dry' of constantscan be drawn and will be tangentat curves {(r, y) equalto a succession all points to the velocity vectors.Thesecurvestrace the flow path of a fluid particle and are known as streamlinesor flowlines. An important property of the stream function is demonstratedwith the aid of Fig. 18.2. Consider the flow crossinga across definedas f1 and tltr.lf the discharge vertical sectionAB betweengfueamlines as Q, it is apparentthat the sectionis designated

n: r,'
f*t

udy

( 1 8.s 3)

Q : and

I d,t' J,t,^

(18.s4)

Q:Qt-Q,

(18.55)

Equation 18.55illustratesthe important property that flow betweentwo streamlines is ionstant. Streamline spaclngreveals the relative magnitudesof flow velocities and vice versa. with narrower spacings, betweenthem. Higher valuesare associated as @rand Q,, calledeqaipotentiallines, ate The curvesin Fig. 18.2 designated determinedby velocity potentials Q@, y): constant. These curves intersect the flowlines at right angles,illustrated in the following way. The total differential d@is given by

a 0 ,+ a + , d0: frdx fidt

substituting for terms aslax and 0Q/sy their equivalentsu and o gives us udx-lody:O
and

(18.s6)

(18.s7)

dy dx

u
t)

(18.58)

andequiposf flowlines Therystem lineqarenormalto flowlines. Thusequipotential net. tentiallinesformsa flow

CONDITIONS 18.8 BOUNDARY

447

'

is that equipotenOne significantpoint of differencebetween{ and ry'functions flow the For two-dimensional is irrotational. tial lines exist only when the flow (. is of vorticity component the when z condition of irrotationality is said to exist
ZA[O; Ot

:o t,: (x-,4)
'
d-Q

(18.se)

substitutingforu and o in Eq. 18.59in termsof Proof of this is givenby Eskinazi.6 @,we obtain

_ a26:0
EY0x

(18.60)

6xdy

This indicatesthat when the velocity potential exists,the criterion for irrotationality is satisfied. or equipotentiallines in a flow domain are determined, Once either streamlines of the relations in Eq. 18.48.Thus r the other is automaticallyknown because

and

r: [(X*- H,o.) - H*) r: [(Xo.

(18.61a) (18.6lb)

It is enough then to determineonly one of the functions, since the other can be obtainedusing relationsEqs. 18.61aand 18.61b.The complexpotentialgiven by

(r8.62)
Of where l, the squareroot of -1, is widely used in analytic flow net analyses'a's specialimportanceis the fact that (18.63) y2w:V26+iYzt1r:g the conditions of continuity and irrotationality simultaneously. satisfies Equations presentedin this section have been limited to the case of twodimensional flow. Extension to three dimensionswould be obtained in a similar fashion.

CONDITIONS 18.8 BOUNDARY

surfaces). Impervious boundariesmay be artificial objects zuch as.concretedams, rock boundary In Fig' 18.3the impervious strata,oi soil stratathat arehighiy impervious.

448

CHAPTER18

OF FLOW MECHANICS

AB represents sucha limit. Sinceflow cannotcrossan imperviousboundary,velocity In other components normal to it vanishand the imperviousboundaryis a streamline. : constant. words,at the boundary,V Next look at the upstreamface of the earth damBC. At any point of elevation hydrostatic,or y along BC the pressurecan be assumed p:y(h-y) The definition of a velocity potential statesthat (18.64)

*. /) o ' :-*\ (v z *' r

in Eq. 18.65yields for pressure Substituting

(18.65)

o:-*l
and

*r]*t

(18.66) (18.67)

6:-Kh+C medium, levelh andan isotropic reservoir Thusfor a constant

d : constant

and surfaceBC, often termed a reservoir boundary, is an equipotentialline. CD in Fig. 18.3 is seento be a boundary The free surfaceor line of seepage Since flow doesnot occur acrossthis zones. unsaturated betweenthe saturatedand along this free surfacemust be Pressure also a streamline. boundary,it is obviously constant,and thereforealong CD 6 + Ky: constant (18.68)

This is a linear relation in $, and therefore equal vertical falls along CD must be equipotential drops. One important groundwater flow associatedwith successive problem is to determinethe location of the line of seepage. the location at which water DE of Fig. 18.3 represents The surfaceof seepage trickles toward point E. The and the dam face of downstream seepsthrough the is neither a flowline nor an The surfaceof seepage pressurealongDE is atmospheric. equipotentialline.

ImPervious laYer

Figure 18.3 Some common boundary conditions.

NETS 18.9 FLOW

449

18.9 FLOWNETS
of families of streamlinesand equipotential Flow nets, or graphicalrepresentations lines, are widely used in groundwaterstudies to determine quantities, rates, and flow at directions of flow. The use of flow nets is limited to steadyincompressible media or for regions that can be constant viscosity and density for homogeneous Darcy's law must be applicableto segments. into homogeneous compartmentalized the flow conditions. The mannerin which a flow net canbe usedin problem solvingis bestexplained with the aid of Fig. 18.4. This diagram showsa portion of a flow net constructedso by a pair of streamlinesand equipotentiallines is approxithat eachunit b.ounded mately square.The reasonfor this will be clear later. A flow net can be determinedexactly if functions Q and $ are known beforeflow nets are hand.This is often not the case,and as a result, graphicallyconstructed widely used. The preparation of a flow net requires application of the concept of square elementsand adherenceto boundary conditions. Graphical flow nets ard practice an acceptable usually difficult for a beginnerto create,but with reasonable methodsfor graphicalflow net constructionare net canbe drawn.Variousmechanical here.5'e presented in the literature and are not discussed usingthe geometryof it can be analyzed After a flow net hasbeenconstructed, the net and by applylngDarcy's law. that the hydraulic that h : p/y + z, we find thatFig' 18'4 shows Remembering gradient G2betweentwo equipotentiallines is given by

o r: x
Lq: '*(*)

(18.6e)

Then by applyingDarcy's law, in the mannerof Todd,2the flow incrementbetween is adjacentstreamlines (18.70)

area for a net of unit width normal to the the cross-sectional whereLm represents plane of the diagram. If the flow net is constructedin an orthogonalmanner and
h- Ll
Equalpotential lines (0 = constant) Lm

r d " Qz

Figure 18.4 Segmentof an orthogonalflow net.

450

CHAPTER18

OF FLOW MECHANICS

composedof approximatelysquareelements, Lm: L,s and Lq = K Lh (18'71) Now if there are n equipotentialdrops betweenthe equipotentiallines, it is evident that h Lh: IL

If the flow is divided into m sections wherc his the total headloss over the n spaces. of the medium is per width unit by the flowlines,then the discharge

o:iu:Y

(r8.72)

can be computed when the medium's hydraulic conductivityis known, the discharge using Eq. 18.72 and a knowledgeof flow net geometry' the entranceand exil Where the flow net has a iree surfaceor line of seepage, of these discussion conditions given in Fig. 18.3 are useful. A more comprehensive conditionsis given in Ref. 10. Some trouble arisesin flow net construction at locations where the velocity Suchpoints are known assingularpoints andaccording infinite or vanishes. becomes In the first classification to De'Wiestmay be placed in three separatecategories.a lines do not intersectat right angles'Sucha situationoften flowlinesand equipotential with a flowline; PointA in Fig. 18.5is-anelamnl-e. occurswhen a boundarycoincides has a discontinuityalongthe boundary that abruptly The secondclassification the slope of the streamline.In Fig. 18.6 PointsA, B, and c representsuch changes At PointsA and c the velocity is infinite, while at Point B it is zero' discontinuities. direction insidethe flow in a counterclockwise If the angleof discontinuitymeasured the velocity is zero; if larger than 180o'it is inflnite' The angle than 180o, field is le-ss atA is 270",for examPle. net' The third categoiyincludesthe casewherea sourceor sink existsin the flow net flow of the Under these circumstancesthe velocity is infinite, since squares

Line of seepage Tangent Surface of seepage

tt

a<90o,0<90"
(a)

ft)

(After Figure 18.5 Some entranceand exit conditions for the line of seepage' Casagrande.lo)

18..!O VARIABLEHYDRAULICCONDUCTMry

451

discontinuities' Figure 18.6 Flowlineslope Wells and rechargewells approach zero sizeas the source or sink is approached. ' later. discussed are and sense practical in a representsinks and sources

CONDUCTIVITY HYDRAULIC 18.10VARIABLE

It is commonfor flow within a porousmedium of one hydraulic conductivityto enter another region with a different hydraulic conductivity. When such a boundary is crossed.flowlines are refracted.The changein direction that occurs can be determined as a function of the two permeabilitiesinvolved in the manner of Todd and this. Figure 18.7illustrates DeWiest.2'a K, and K, which are separatedby the permeabilities of consider two soils of the flowlines before and after The directions 18.7. in Fig. boundary lR shown 0r' and 0, angles by defined crossingthe boundary are

Figure 18.7 Flowline refraction.

452

CHAPTER18

MECHANICS OF FLOW

"

For continuity to be preserved,the velocity componentsin media K, and Kt, areaat the which are normal to the boundary,mustbe equal,sincethe cross-sectional boundaryis AB for a unit depth.UsingDarcy's law and noting the equipotentialdrops h" and hr, we flnd

o#cos 92 K,L*cos o' : u, From the geometry of the flgure it is apparentthat AC : AB sin 0t BD : AB sin 0z

(18.73)

The headlossbetween A andB is shownon the figure to be equalto both Lh" andA,hu, and sincethere can be only a singlevalue, | Lh": Ah6 in Eq. 18.73produces Introducingtheseexpressions
K t _ K , tan 0, tan 02

(18.74)

For refractedflow in a saturatedporous medium, the ratio of the tangentsof anglesformed by the intersectionof flowlineswith normals to the boundary is given by the ratio of hydraulic conductivities.As a result of refraction, the flow net on the DB if the equipotentialline spacing K2 sideof the boundarywill no longerbe squares is maintained.To adjust the net on the K2 side,the relation
Lhu Kl

Lt%- It

( 18 .7s )

can be usedwhere Lhb + Lh". The Equipotentiallines are also refractedin crossingpermeability boundaries. relation for this is
K, K2 _ tan at tan a,

(18.76)

where a is the anglebetweenthe equipotentialline and a normal to the boundary of permeability.a

18.11ANISOTROPY
on the direction of flow within a In many cases hydraulic conductivity is dependent deposits often givenlayerof soil. This condition is saidto be anisotropic.Sedimentary fit this aspect,with flow occurring more readily along the plane of depositionthan acrossit. Where the permeability within a plane is uniform but very small acrossit as comparedto that along the plane, a flow net can still be usedafter proper adjustNonhomogeneous ments are made. A discussionof this is given elsewhere.a's'11'12

THEORY 453 18.12 DUPUIT'S 'sometimesbe analyzed by Yslng aquifers require special considerationbut may parameters'A detailid study is outsidethe scopeof this or average representative
book.3-5'r2

THEORY 18.12DUPUIT'S
free surfacecan be analyzed Groundwaterflow problemsin which one boundaryis a theory is foundedon two This flow. on the basis of Dupuit's theory of unconfined is only slightly seepage of line ptptl the if in tgO:.t' First, madeby assumptions equlp:tencorrespondingly, and, horizontal r""y b" considered inclined, streamlines tl" seepage of line the of slopes -utfd tial lines will be essentiallyvertical. Second, satisfactorily be to known at" hydraulic gradient *r "fit. Whel fleld conditioos Dupuit's theory by theseusru-ption*, the resultsobtainedaccordingto represented "o'-pur"veryfavorablywiththosearrivedatbymorerigoroustechniques' into a mathemdtFigure 18.8 is u,"fol in translatingthe foregoingassumptions base'ateadx dy a has which figut" th" Consideran elementgiien in ical statement. the x direction and considand a vertical height h,Writing the cJntinuity equaiionin flow to be the case, ering steadY (18.77) infloqo : velocitSo X area,o The velocity at x : 0 is given by Darcy's law as
,ro: _K*

(18.78)

acrossthe elementat x : 0 is Thus the discharge - - a h. -K dY Qo: *h expansionas The outflow at x = dx is obtainedby a Taylor's series -K*.h _ _ *(-**n .. dv " + o,_ Qor: dr 6x\ Ax,. ' ' aY\+ _ /

(18.7e)

(18.80)

'

Figure 1,8.8 Definition sketch for development of DuPuit's equation'

454

CHAPTER18

MECHANICS OF FLOW

Subtractingthe outflow from the inflow if K is considered constant,we obtain

I* - o,: K dxa, " !(n*)

dx\

6x/

tt*.

or

r,_ o.: rydr,A+fg) 2 Ax\6x/

'x

(r8.

wheredx anddy are considered in the y fixed lengths.A similar consideration tion yields

r,-o,:ryy*W)
i

, (18

; )

Assuminsthat thereis no movementin the vertical direction.theseare the components of the inflow and outflow.Furthermore,still dealingwith steadyflow, I changein storagemust be zero. As a result,

Kd:dy Kd:dy : o * +fg) *(#) 2 lx\Ex / 2 ay\6y /

and since (K dx dy)12 is constant,this reducesto A2h2 -:0 - r . Azhz 0y' 6x' :0 Y2hz

o8

or

(18.

Consequently,. Laplace's equation for accordingto Dupuit' s assumptions, functionh2 must be satisfied.la In the particular casewhererecharge is occurring as a resultofinfiltrated reachingthe water table, a simple adjustmentmay be made to Eq. 18.85. If recharge intensity(dimensionallyLT-t) is specified asR, thenthe total recharge to elementof Fig. 18.8 is R dx dy and the continuity equationfor steadyflow beco

*4!*(#.#).Rdxdy:s
or more simply,

(18

v2h2+?o:o
Now, applyingDupuit's theory to the flow problem illustratedin Fig. 18.9, assuming one-dimensional flow in the x direction only, we obtain the discharge unit width of the aquifer given by Darcy's law:

O: -Kh#
In this instanceh is the height of the line of seepage at any position x along imperviousboundary.For the one-dimensional exampleconsidered here, Eq. 18

18.12 DUPUIT'STHEORY

455

fto=50ft

free surface

(b)

Figure 18.9 Steady flow in a porous medium between two witer bodies: (a) free surface with infiltration and (b) free surfacewithout infi ltration. becomes d 2h2 --;-;: clx'

(18.e0)

Upon integration,
h2:ax*b

( 18.e 1)

wherea andb are constants. Then for boundarv conditions at x : O,h : hs,

b=ht
Differentiationof Eq. 18.91yields ^. dh
Ztl';-

(18.e2)

ax

= A

( 18.e 3)

-QIK. Making this substitution,we obtain Also from Darcy's equation,h dhldx :

o:-T-

-2Q

(18.e4)

and inserting the valuesof the constantsin Eq' 18.91,we obtain

h2=-2f*+nt

(18,e5)

456

18 .CHAPTER

MECHANICSOF FLOW

It is a parabola(often calledDupuit' s parabola)' This is the equationof a free surface. andonotingthat at * : L, If the existenceof a surfaceof seepage.afBi. ignor.d, h : hr, we f,nd that Eq. 18'95becomes 2QL hL: _ x n ra

(18.e6) (18.e7)

or

o ol-n?) - =L 2L'"

which is known as the Dupuit equation' EXAMPLE 18.3 RefertoFig.ls.ga.GiventhedimensionsshownandarechargeintensityRof : 1000 ft using Dupuit's equation'Assumerthat 0.01 ftlday, find the Oi."ft"tg" ut x K: 8. Solution. Note that

Q=n
dx or
Q=Rxi.C

Atx:0,
. Q: Qo

therefore,
Q: Rx -l Qo

Also,
Q dh -Kh=

- oh
IntegratingYields

nh

i:

Rx ' r Qo

- Kh,l,, : o*,il' * o^*1,

2lo" 2lo -"lo

and inserting the limits, we obtain

o L

-K(h'-Lh7) :ry
r

* QoL
RL 2

1)^:----:----

K(h?- h'r) z L

PROBLEMS

457

Then since Q : Rx + Qo,

- h?\ K(h'^ . ,) 2L R 0 . 0 1 7 . 5 : 0.075gpd/ft2 _ soo)+ 8(50, 40r) o : 0.075(1000 2000 0.075x 500 , 8 x 9 0 0 2000 3 7 . 5+ 3 . 6 4Ll gpdlftz T I

R(.

r summary
Understanding the movementof groundwaterrequiresa knowledgeof the time and space dependencyof the flow, nature of the porous medium and fluid, and the boundariesof the flow system. In particular, groundwaterdevelopment and management dependon understanding propertiesofthe associated the storage soils and rocks ' and the ability of thesesubsurface materialsto transmit water. Fundamentalto the mechanics of groundwater flow is Darcy's law (Eq. 18.3).Usingthis equationalong with a knowledgeof the hydraulic conductivity K, estimates of flow can be had. The hydrodynamicequationspresentedin this chapter serve as models for a variety of groundwatgrflow calculations.Applications are given in Chapters19 and 20.

PROBLEMS'
18.1. What is the Reynoldsnumber for flow in a soil when the water temperatureis 55oF, the velocity is 0.5 ftlday, and the mean grain diameteris 0.08 in.? 18.2. The water temperaturein an aquifer is 60'F, the velocity is 1.0 ftlday. The average particle diameterof the soil is 0.06 in. Find the Reynoldsnumberand indicatewhether Darcy's law applies. 18.3. ReworkProblem 18.2assuming the temperature is 65"F and the velocity is 0.8 ftlday. 18.4. A laboratory test of a soil gives a standardcoeff,cientofpermeability of 3.8 x 102 gpdlft2.If the prevailingfield temperature is 60"F,find the field coefficientof permeability. " 18.5. ReworkProblem 18.4 assuming K" is 3.8 x r02 gpd/ft2and the temperature is 65'F. 18.6. Given the well and flow net datain the following flgure, find the discharge usinga flow net solution.The well is fully penetrating; K : 2.87 X 10-4 ftlsec, a: 180 ft, b = 43 ft, and c : 50 ft. 18.7. ReworkProblem 18.6assuming K : 8.2 x 10-5 mlsec,a: 85 m, b : 2l m, and c:26m.

458

CHAPTER18

MECHANICS OF FLOW
Stagnation flowline

Axis of

5 4 3 2 I

symmetry -1 -2 -3 /

Kg : 8 ' 4 X 1 0 - 5m l s e c , a : 1 0 0 m ,b : 2 2 m , a n d 18.8. ReworkProble1 m8 . 6 a s s u m i n c:35m. 18.9. A stratum of clean sand and gravel 15 ft deep has a coefficient of permeability of to K : 3.25 X 10-3 ftlsec, and is suppliedwith waterfrom achannelthat penetrates the bottom of the stratum.If the water surfacein an infiltration gallery is 2 ft above the bottom of the stratum.and its distanceto the channelis 50 ft, what is the flow into a foot of gallery?UseEq. 18.97.

REFERENCES
1. 2. 3. 4. 5. 6. 7. Henri Darcy, Lesfontaines publiquesde la ville de Dijon. Paris: V. Dalmont, 1856. D. K. Todd, GroundwaterHydrology. New Yorkl V/iley, 1960. Evaluation.New York: McGraw-Hill, 1970. William C. Walton, GroundwaterResource Wiley, 1965. New York: R. J. M. DeWiest,Geohydrology. New York: McGraw-Hill, 1962. M. E. Harr, Groundwaterand Seepage. SalamonEskinazi,Principles of Fluid Mechanics.Boston:Allyn and Bacon,7962"Flow of Groundwater,"in Engineering Hydraulics (Hunter Rouse,ed.) New C.E. Jacob, York Wiley, 1950.

Chapter '1 I

Wellsand Collection Devices

r Prologue Thepurpose of this chapter is to:

. Presentmethodsfor calculatingconfined and unconfinedsteadyradial flow toward a well. . Describeproceduresfor dealingwith unsteadygroundwaterflow conditions. . Describea method for estimatingflow to an infiltration gallery. Groundwater is collectedprimarily by wells, althoughinfiltration galleriesare sometimes usedwherethe circumstances are appropriate.tOutflows from natural springs are also amenable to collection,but once thesewatersexit the ground, they become surfaceflows and are handled as such. Wells are holes or shafts, usually vertical, excavatedin the earth for the purpose of bringing groundwater to the surface. Infiltration galleriesarehorizontalconduitsfor interceptingand collectinggroundwater by gravity flow. Problemsof groundwaterflow to wells and infiltration galleries can be solved by applyingDarcy's law. ,

19.1 FLOW TO WELLS A well system can be considered as composed of three elements-the well structure, pump, and discharge piping.2The well itself containsan open sectionthrough which waterentersand a casingto transportthe flow to the ground surface. The open section is usually a perforatedcasingor slottedmetal screenpermitting waterto enter and at the sametime preventingcollapseof the hole. Occasionally,gravel is placed at the bottom of the well casingaround the screen. When a well is pumped,wateris removedfrom the aquiferimmediatelyadjacent to the screen. Flow then becomes established at locationssomedistancefrom the well in order to replenishthis withdrawal. Because offered by the soil. of flow resistance a head loss results and the piezometric surfaceadjacent to the well is depressed. producinga coneof depression (Fig. 19. 1), which spreads until equilibriumis reached and steady-state conditions are established.

RADIALFLOW TOWARDA WELL 19.2 STEADYUNCONFINED

461

Impervious

Figure 19.1 Well in an unconfinedaquifer.

by the storage of an aquifer (which are described The hydraulic characteristics or field tests' by laboratory determined be can permeability) coefficient and aquifer the use of tracers, application thp are methods field used "o111*only The three 111ort per!o.r. of aquifer A discussion tests.3 performance aquifer uni of field permeameterr, wells.2'a's for equations of flow development the with given here along mancetestsis testsmay be either equilibrium or nonequilibriumtests.In Aquifer performance mustbe stabilizedfor a flow equationto be an equilibrium testthe coneof depression includesa condition that steady-state derivation tesithe derived.For a nonequilibrium tests publishedthe first performance Thiem Adolph reached. conditionshavenot been 1906.6 in basedon equilibriumconditions

A WELL RADIALFLOWTOWARD UNCONFINED 19.2 STEADY

Q: hrxYKrfi
where 2rrxy : . Kr: dyfdx : Q:

(le.1)

the areathrough any cylindrical shell, in ft2 with the well as its axis the hydraulic conductivity (ftlsec) the water table gradient at any distancex the well discharge(ft'lsec)
f,2 r..
n

Integratingover the limits specified,we find that

fo, : 2rrK1l y ay gY I l r
Jhl

(re.2)

Jt

462

DEVICES CHAPTER19 WELLS AND COLLECTION

Q n f1 2_
and

- h?) 2rrKr(h2z 2 : nKr(hi hl)

^

( 1e.3)
g

ln(rr/r1)

(te.4)

converting K, to the field units of gpdlftz, Q to gpm, and ln to 1og,we can rewrite Eq. 19.4as
,ar: " ------;-;----;;-

1055Qlog(r2lrr) n;- ni

(1e.s)

one half of the original aquiferthickness If the drawdownin the well doesnot exceed by usingEq.19.4 or tr9.5,even of Q or Krcanbe obtained estimates ho,reasonable at the well peripherywhere1L: r*, the radius of the well if the heighth, is measured boring. EXAMPLE 19.1 an unconfinedaquifer of 100-ft depth.Two observa: An 18-in. well fully penetrates tion wells located100 and 235 ft from the pumpedwell areknown to havedrawdowns of 22.2 and 21.ft, respectively.If the flow is steady and K1 : 1320 gpdlft2, what would be the discharge? Solution. Equation 19.4 is applicable,and for the given units this is 1055Iog(r2/r1) :0.37rlt = log(r2/r') rwQ35l1'00) hz: 100-21=79ft
ht=

Q :

K(h?- h?)

100- 22.2: 77.8ft - 77.82) t32O(792 Q : 1055x 0.37107 gPm lr 634.44

A WELL RADIALFLOWTOWARD CONFINED 19.3 STEADY

The basic equilibrium equationfor a confined aquifer can be obtainedin a similar apply. Mathematimanner, using the notation of Fig. I9.2. The sameassumptions from cally, the flow in ft3lsecis found

o : 2nxmXr! "dx
Integrating,we obtain h"-h, O:2rrK'm " :# ln\r2/h)

(1e.6)

(re.7)

.I9.4 WELL IN A UNIFORM FLOWFIELD

--12+l

463

aquifer. Figure 19.2 Radialflow to a well in a confined The coefficient of permeability may be determinedby rearrangingEq. 19.7 to the form - _ 528Q log(rz/r'\ nr *&; h) where Q : gpm K f : the permeability (gpd/ft') r , h : ft

(1e.8)

EXAMPLE 19.2 Determinethe permeabilityof an artesianaquiferbeingpumpedby a fully penetrating well. The aquifer is 90 ft thick and composedof medium sand. The steady-state pumpingrate is 850 gpm. The drawdownof an observationwell 50 ft awayis 10 ft; in a secondobservationwell 500 ft awayit is 1 ft. Solution
Kf:

52SQroe?Jr')
m(h2 - h')

528 x 850 x log 10 90x(10-1) : 554gpdlfr2 rl FIELD 19.4 WELL IN A UNIFORM"FLOW

well in a uniform flow field wherethe original piezometricsurface For a steady-state preis not horizontal, a somewhatdifferent situation from that previously assumed vails. Consider the artesian aquifer shown in Fig. 19.3. The heretofore assumed circular area of influencebecomesdistortedin this case.A solution is possibleby or, if the slopeof the piezometric applyingpotentialtheory,by usinggraphicalmeans, surfaceis very slight, Eq. 19.7 may be employedwithout seriouserror.

464

DEVICES CHAPTER19 WELLS AND COLLECTION

Original piezometric surface

Figure 19.3 Wellin a uniformflow fieldandflow net definition' . Figure 19.3providesa graphicalsolutionto a uniform flow field problem.First, an ortholgonalflow net consistingof flowlines and equipotentiallines must be constructed.This shouldbe done so that the completedflow net will be composedof a in shape.Once the net is complete, number of elementsthat approachlittle squares Uy consideringthe net geomtry and using Darcy's law in the if can be analyzed manner of Todd.3 EXAMPLE 19.3 to the well of Fig. 19.3by using an applicableflow net. Consider Find the discharge as shown' the aquiferto be 35 ft thick, 4 : i.65 x 1d-4 fps, and other dimensions Solution. UsingEq. 18'72,we find that

n : * *n'
whereh:35*25:60ft m : 2 X 5 : 1 0 , n:I4 3.65X10-4x60x10

t4 : 0.0156cfsper unit thlckness of the aquifer

The total dischargeQ is thus r r Q : 0.0156 X 35 : 0.55 cfs ot 245 gPm

19.5 WELLFIELDS 465

19.5 WELL FIELDS When more than one unit in a well fleld is pumped,there is a compositeeffect on the is illustratedby Fig. 19.4in which the conesof This consequence free water surface. are seento overlap.The drawdownat a given location is equalto the sum depression of the individual drawdowns. If, within a particular well field, pumpingratesof the pumpedwells are known, the compositedrawdownat a point can be determined.In like manner,if the drawdown at one point is known, the well flows can be calculated' 1,2, ' as m, andsubscripts 'n If the drawdownat a given point is designated the drawdown to mt fefers ., well(e.g particular are usedto relate this drawdownto a for I7,), for the total drawdownmr at somelocation'
n

lTlr:

sr
^Z

lTli

(1e.e)

The numberof wells, their rate of pumping,and well-fieldgeometryand charac' teristicsdeterminethe total drawdownat a specifiedlocation. Eq. 19.4,we obtain Again considering

h3- h'

: #*t"1

(1e.10)

lt can be seenthat the drawdownfor a well pumped atrate Q canbe computedif ho, ro, and r are known. It follows then from Eq. 19.9 that for n pumped wells in an unconfinedaquifer

h 3h - ':2*."?

( 1 e .1 1)

Figure 19.4 Combinedeffect of pumping severalwells at equal rates'

466

CHAPTER19 WELLS AND COLLECTION DEVICES

where ho : the original height of the water table h : the combined effect height of the water table after pumping n wells r Q, : the flow rate of the ith well foi : distance of the lth well to a location at which the drawdown is considered negligible r, : the distancefrom well i to the point at which the drawdownis being investigated Todd indicatesthat valuesof rousedin practiceoften rangefrom 500 to 1000ft.3 The impact of this assumption is softenedbecause Q inEq. 19.10is not very sensitiveto are relativd small. 16.Equation19.11shouldbe usedonly wheredrawdowns for combineddrawdownbecomes For flow in a confinedaquifer the expression
n

h o - h : 2 =Q:, ,n'o'
z1rKm

f1

(r9.r2)

Equationsfor well flow covering a variety of particular well-field patternsare reported in the literature.3'7 Those given here are applicablefor steadyflow in a homogeneous isotropic medium.

19.6THEMETHOD OF IMAGES
Some groundwaterflow problems subjectedto boundary conditions negating the fitting these into infinite systems direct useof radial flow equations canbe transformed equations by applyingthe methodof images.2'8'e When a streamis locatednear a pumped well and the streamand aquifer are interconnected, the drawdowncurve of a pumpedwell may be affectedas shownin Fig. 19.5. Another boundary condition often affecting the drawdownof a well is an of imperviousformation that limits the extentof the aquifer.The cone of depression After that, the shape a pumpedwell is not affecteduntil the boundary is intersected. of the drawdown curve will be changedby the boundary. Boundary effects can frequently be evaluatedby meansof "image wells." The boundary condition is replacedby either a rechargingor a discharging well that is pumped or rechargedat a rate equivalentto that of the pumpedwell. That is, in an infinite aquifer,drawdowns of the real and imagewells would be identical.The imagewell is locatedat a distance from the boundary equal to that of the real well but on the oppositeside(Fig. 19.5). are supplanted boundaries Streamsare replaced by recharge wells while impermeable for the caseof a well and imperviousboundary by pumpedimagewells.Computations directly follow the procedures outlined under the sectionon well fields.For the well to havea negative discharge. and streamsystem, the recharge imagewell is considered The headsare then addedaccordingto this sign convention. The procedurefor combiningdrawdowncurvesof real and imagewells to obtain an actual drawdown curve is illustratetl graphically for the example shown in Fis. 19.5.More detailedinformationon othercases can be found elsewhere.e'10

19.7 UNSTEADY FLOW

467

Cone of depressionof real well without sffeam

connectedto Figure L9.5 Drawdown in a pumping well whose aquifer is a stream'

FLOW 19.7 UNSTEADY

comesdirectly from , alargeportion of the discharge when a new well is first pumped u.nder thesecircumas the cone of depression volume released the storage _d91elons, permeabilityandthereforethe yield of overestimate the equilibrium equations stances conditionsare not encountered-as is usually the situathe well. When steady-state can,be approaches tion in practice-a nonequilibrium gOgltlgl must be used' Two asthat such ratherrigorousmethodof C. V. Theisor a simplifiedprocedure taken,t^he by Dcob.l1'12 proposed that takesinto consideration In tgjS treis publisheda nonequilibriumapproach of the aquifer.ll His methodusesan analogybetween characteristics time and storage by the Biot-Fouiier law and groundwaterflow to a well' The heattransferdescribed boundary conditions' method providesa solution to Eq. 18.41 for given initial and constantthickness' Application of the methodis appiopriatefor confinedaquifersof flow, vertical componentsof flow must be For'use under conditions of unconhneO

468

CHAPTER19 WELLS AND COLLECTION DEVICES

negligible,and changes in aquifer storage through water expansion and aquifer compressionmust also be negligiblerelative to the gravity drainageof poresas the water table drops as a result of pumping.l3 Theis statesthat the drawdown(s) in an observationwell locatedat a distance r from the pumpedwell is given by

O f * " - o" ', s : -4rrT J, ;

(1e.13)

in which Q : (constant) pumping rate (L3T-r units), Z : aquifer transmissivity (LtT 'units), and a is a dimensionless variabledefinedby

, ,s" u : r--4tT

(re.r4)

where r is the radial distancefrom the pumpingwell to an observation well, S. is the aquiferstorativity(dimensionless), and r is time. The integralin Eq. 19.13is usually known asthewellfunctionof u andiscommonlywritten asW(u).It maybe evaluated from the infinite series

- ln u * u - =+ - + -]-w(u): _'0.577216 2x21 3x3l

Usingthis notation,Eq. 19.13can be written as ( : -Qw(u) " 4irT

(1e.1s)

(1e.i6)

The basic assumptiops employedin the Theis equation are essentiallythe sameas thosein Eq.I9.7 exceptfor the nonsteady-state condition. Some valuesof the well function are given in Table 19.1. In American practice,Eqs. 19.13 and 19.I4 commonly appearin the following form,

t:

rI4.6Qf* "-' , , J,;o'

I . 8 71 2 S"

(re.r7)
(1e.18)

,:

where 7 is given in units of gpd/ft, Q hasunits of gpm, and / is the time in dayssince the start of pumping. Equations 19.13and 19.14canbe solved a log-logplot of a versus by comparing I4z(z) known as i type curve, with a log-1ogplot of the observeddatar2ft versuss. In plotting type curves,W(u) ands are ordinates, The two curves z and rt ft areabscissas. are superimposed and movedaboutuntil segments coincide.In this operationthe axes must remain parallel. A coincidentpoint is then selected on the matchedcurvesand both plots marked.The type curve then yieldsvaluesof u and W(u) for the desired point. Corresponding from a plot ofthe observed valuesofs and r'ft aredetermined

FLOW 1g7 UNSTEADY OF U VALUES OF W(U}FORVARIOUS TABLE19.1 VALUES 'L0 5.0 4.0 3,0 2.0 XI x 10-r x 10-2 x 10-3 x 10-4 x 10-5 x 10-6 x 10-7 x 10-8 x 10-e x 10-10 x 10-11 x 10-t2 x 10-13 x 10-14 x 10-15 0.219 t.82 4.04 6,33 8.63 10.94 t3.24 15.54 17.84 20.15
aa A<

469'

6.0 0.00036 0.45

z,Jv

8.0 -' 0,00012 0.37 2.1,5 4.39 6,69 8.99 rt.29 13.60 15.90 18,20 20.50 22.8t 0,000038 0.31 2.03 4,26 6.55 8.86 I 1.16 t3.46 t5.'16 18.07 20.37 22.67 24.9'1 27.28 29.58 31.88

9.0 0.000012 0.26 1.92 4.14 6.44 8.74 1l.04 13.34 15.65 17.95 20.25 22,5s 24.86 27.16 29.46 3t.76

24.75 27.0s 29.36 31.66 33,96

0.049 1.22 3.35 5.64 7.94 t0.24 12.55 14.85 t7,15 19,45 21.76 24.06 26.36 28,66 30.97

0 . 0 1 3 0.0038 0.70 0.91 2.68 2.96 4.95 5.23 7.25 1.53 9.55 9.84 12.14 11 . 8 5 14.15 14.44 16.46 16.74 19.05 t8.76 2 r . 3 5 2r.06 23.65 2 3 . 3 6 25.96 25.67 27.97 28.26 30.56 30.27 32.86 32.58

0.0011 0.56 2.47 4.' 13 7,02 9.33 11 . 6 3 1.3.93 16.23 18.54 20.84 23.14 25.44 27.75 30.05 32.35

4.54 6.84 9.r4 r1.45 13.75 16.05 18.35 20.66 22,96 25.26 27.56 29.8'7 32.1,7

25.rr
2'1.41 29.71 32,02

..Methodsfor Determinlng Permeability of water Bearing M.aterialswith Special Referenceto Dischargt.fig Source: AfterL,K. Wenzel, Washington'DC' 1942' Well Methods," U. S. GeologiJSt.""V, W"t*-Supp-ly Paper 887'

values for data. Inserting these values in Eqs. 19.13 and 1,9.1'4and,rearranging, found' coefficientS" can be transmissibilityI and storage "can be shortenedand simplified' when r is small and t often this procedure Thus terms in the seriesof large, Jacobfound that valuesof u arc generallysmall'12 for Zbecomes expression the and fql te.tS bpyondthe secondonebecomenegligible 264Q(loet2 - log tt) (1e.1e) ' h"- h which can be further reducedto

264Q T _ Lh
where Ah Q ho,h T : : : : drawdownper log cycle of time l(ho (gPm) well discharge as definedinFig' I9.2 (gPd/ft) the transmissibilitY - n)lloe t' - log t')l

(re.20)

paper' The Field data on drawdown(ho h) versust arc dtaftedon semilogarthmic a straightline drawdownis plotted on an arithmetic soale,Fig. 19.6. Thisplotforms and *to*" slopepermits computingformation constantsusing Eq' 1'9'20 ^ : - 0.3Tto \ - r '

(1,9.2t)

Equation 19.21is obtainedthrough to zerodrawdown. with rothe time corresponding of Eq. 19'13. manipulation

470

DEVICES CHAPTER19 WELLS AND COLLECTION

, = t4

r\too'

= 4e,8oo sd/ft

tl

I I
I

Time since PumPing began (min)

Figure 1.9.6 Pumpingtest data, Jacobmethod. EXAMPLE 19.4

for an aquiferusinga graphical Usingthe following data, find the formation constants solution to the Theis equation.Dischargeequals540 gpm.
from Distance pumped well, r (ft) 50 100 150 200 300 400 500 600 700 800
Average drawdown, s (ft)

12 /t

1,250 5,000 il,250 20,000 45,000 80,000 125,000 180,000 245,000 320.000

3.04 2.16 1,.63 r.28 0.80 0.51 0.33 0.22 0.15 0 .l 0

FLOW 19.7 UNSTEADY

o.o ? 0.s E o.+

B

1.0 0.9 0.8 0.7

't I | , I W(u)vs. u

E 0.3

5 6 7 891x105

I rrzldav)
equation' to Theis's solution ' Figure L9.7 Graphical Solution.Plotsversusr2ftand}V(a)versusaasshowninFig.lg.T.Determine Eqs' 19'7 and 19'8:the match point as noted and computeS" and Z using T : lr4t6Q w(u) s X 540114.6

: --18---c u"

^ 1.9 :

91,860gpdift

uT
1.87r'1t

_ 0.09x 91,860 : 0,22 tr 1.87X 20,000

EXAMPLE 19.5

7"andstorage of transmissibility in Fig. 19.6,findthe_coefficient usingthedatagiv.en : ft' 300 Q: 1000gprn andr S, for an aquiflr, given coeffrcient Solution.FindthevalueofAhfromthegraph,5.3ft.ThenbyEq.|9.20 2640 264 - - Tx31000

T: -ff:

: 49,800gpd/ft we find that Using Eq. 19.21.,

s":

O.3Tto r'

472

CHAPTER19 WELLS AND COLLECTIONDEVICES

Note from Fig. 19.6 that ts : 2.6 min. Converting to days, we find q6il)ti\--l becomes
, 0 -

and

s":

' x lb-'days 1.81 0 . 3 x 4 9 , 8 0 0 X 1 .X 81 0 - 3

(3oo)'

: 0.0003 I T
EXAMPLE 19.6 point 200 ft awayfrom a pumping well. Given Find the drawdownat an observation that T : 3.0 x 10a gpd/ft, the pumping time is 12 days,S" : 3 X 10-4, and

gpm, 0 = 3oo

Solution. From Eq. L9.L8,u can be computed,

x l2l : 6.23x I}-s x 104 u : u.87 x (200)'x 3 x L0-41113.0

we estimateW(u) to be 9.1. Then, Referringto Table 19.1 and interpolating, using Eq. 19.17 , the drawdownis found to be ft rr s : [114.6x 9.1 x 300]/[3.0x 104]: 10.41 EXAMPLE 19.7 Giventhat I : 0.0028m2ls, A well is beingpumpedat a constantrate of 0.0038 m3/s. : : r 90 meters, and the storagecoefficient 0.00098, find the drawdown in the observationwell for a time period of (a) 1,000 sec.and (b) 20 hours. Solution a. Using Eq. 19.14,u canbe computedas follows, x 1000 x 0.0028] u :190 x 90 x 0.000981/[4 u : 0.71 Then from Table 19.1,W(u) is found to be 0.36. Applying Eq. 19'16,the drawdowncan be determined,

x n x 0.0028] x 0.367114 r : [0.0038 s : 0.039m

b. Follow the procedureusedin (a) ' u : 1 9 0 x 9 0 x 0 . 0 0 0 9 8 1 / [x 47 2 , 0 0 0 x 0 . 0 0 2 8 ] : a 0.0098 Then from Table 19.1, W(u) is found to be 4.06 Applying Eq. 19'16,the drawdowncan be determined,

x n x 0.00281 x 4.06]114 s = [0.0038 : 0 . 4 4 m r r s

GALLERY 19.10 FLOW TO AN INFILTRATION

473

AQUIFERS 1e.8LEAKY

havedealtwith free aquifersor thosecopfinedbe#een imperThe foregoinganalyses t"ulity, many casesexist wherein the confining strata are not comvious str-ata.1n PletelYimPerviousand water is at aquifer. The flow regime is altere< about 1930,leakYaquifershaveber De Glee, Jacob,Hantush, DeWies ers.t'to-'uA thorough treatmentc interestedreadersshouldconsulttheindicatedreferences.

a_)

WELLS PENETRATING 19.9 PARTIALLY

of the well' The ln many actual situations there is only partial penetration previouslyfor full developed question then arisesasto the applicability of procedures Penetration. ln 1957Hantush of this problemhavebeenconducted'7'2't'28 studies Numerous aquiferbecomes reportedthat steadyflow to a well just penetratingan infinite leaky from the well of about1.5timesthe aquiferthickness'28 very nearlyradial at a distance the approachto,radial flow becomesincreasingly As depth of penetrationincreases, apparent.Therefore,computationofdrawdownsforpartiallypenetrating'wellsare provided that the made using equationsfoi total penetration with relative safety, aquifer thickness'At distancefr"o-ifr" pumped we11ls greaterthan 1.5 times the other relations points closerto the weli, it is frequJntly possibleto use a flow net or for this region. develoPed

GALLERY 19.10FLOWTO AN INFILTRATION

conduit constructed An infiltration gallery may be defined as a partially pervious part of this flow will be acrossthe path of t6" to"uigroundwater flow iuch that all or often built in a valley areaparallel to a streamso that Thesegalleries-are intercepted. location under gravity-flow they can convey the collected flow to some designated .ondition*.Figurelg.Sshowsatypicalcrosssectionthroughagallerywithone 'perviousface. pervious wall corrrputation of dischargeto an inflltration gallery with one assumptions Several inihe manneroutlinedby Dupuit-2e (Fig. f g.Sjis accomplished of tangent the angleof mustbe madelo effectthe solution.They are that the sineand that the velocity vectorsare evefyinclination ofthe water table are interchangeable; is incompressibleand where horizontal and uniformly distribuied; that the soil effects are negligible' isotropic; and that the gallery is of sufficient length that end do limit the utility of While permitting u ,oluion oi the problem, theseassumptions the results. Basedontheseassumptions,andfollowingtheproceduregiveninSecper unit width' using the the discharge tion 18.12,Eq. 18.97canbe usedto calculate

474

CHAPTER 19 WELLS AND COLLECTIONDEVICES

Ground surface

Intersection of assumed and actual water tables

X+

Figure 19.8 Cross-section throughan infiltrationgallery. nomenclature of Fig. 19.8,Eq. 18.97becomes

s :;(hi

K . - "-

he)

This equationindicatesthat the computedwatertableis parabolic.This is often called Dupuit's parabola.Figure 19.8 showsthat the computedwater table differs from the actual water table in an increasingmanner as the gallery face is approached. It is thereforeapparentthat the computedparaboladoesnot accuratelydescribethe real water table. The differences,however,are small except near the point of outflow, providing the initial assumptions The calculateddischarge approximates are satisfled. the true discharge more closelyas the ratio of Z/h, increases.
EXAMPLE 19.8

A stratum of clean sand and gravel20 ft deep has a coefficient of permeability of K : 3.25 X 10-3ftlsec,andis supplied that penetrates to with waterfrom a channel the bottom of the stratum.If the water surfacein an infiltration gallery is 3 ft above the bottom of the stratum,and its distanceto the channelis 50 ft, what is the flow into a foot of gallery?UseEq. 18.97. Solution q : 0.5(3.25 x 10-3)(20 x 20 - 3 x 3)/50 : 0.072 cfs, the flow into one foot of gallery I r \

19.11SALTWATER INTRUSION
The contaminationof fresh groundwater by the intrusion of salt water often presents a seriousquality problem. Islands and coastalregions are particularly vulnerable. Aquifers locatedinland sometimes contaiq highly salinewaters as well. Freshwater

BASINDEVELOPMENT 475 15.12 GROUNDWATER

BASINDEVELOPMENT 19.12GROUNDWATER
permitting the maxiefficiently while simultaneously resources To use groundwater betweenwilhmum dJvelopmentof the resource,equilibrium must be established Economic, legal, political, social, and water quality drawals and replenishments. require full consideration. aspects only whenlong-termwithdrawls will be assured of groundwater Lasting supplies period. The potentialof a groundduringthe corresponding arebalanceJby recharge by employingthe water budgetequation, water basin can be assessed

)r-)o:As
where the inflow ) l includesall forms of recharge,the total outflow ) O includes duringthe accounting in storage the change everykind ofdischarge,and AS represents are those listed in discharge and recharge of forms perioO. The most significant Table t9.2. A groundwaterhydrologistmust be able to estimatethe quantity of water that time basinin a specified canbe economicallyand safeiyproducedfrom a groundwater of imposing the consequences to evaluate competent period. He or sheshouldalsobe variousrates of withdrawal on an undergroundsupply. Developmentof groundwaterbasins should be based on careful study, since groundwaterresourcesare finite and exhaustible.If the various types of recharge balance the withdrawals from a basin over a period of time, no difficulty will be to a drafts, however,can depleteundergroundwater supplies Excessive encountered. ultimately will water of mining The feasible. is not point whereeconomicdevelopment depletethe entire supply. DISCHARGE AND OFRECHARGE TONT"TS 19,2 SdUE TABLE
Recharge from streams, Seepage Ponds,lakes inflows Subsurface Infiltrated precipitation Water rechargedartificallY Discharge to lakes, streams,spnngs Seepage outflows Subsurface Evapotranspiration Pumping or other artificial meansof collection

i-_

476

DEVICES CHAPTER19 WELLS AND COLLECTION

r Summary
primarily through the constructionof is accomplished The collection of groundwater estimation"of their performance. numerical the influence wells, and many Tactors utilization of relatively simple the through to solution Some situation, ur" amenable applicationof the hysophisticated upon depend Others mathematicalexpressions. of aquifermaterials nonuniformity of Conditions various under drodynamicequations to be misledby the not cautioned is reader The conditions. and a variety oiboundary of theserl\ many that to observe presented and thqsolutions simplicity oi ,o-" of situations.(. groundwater-flow all to not applicable are to speciaiconditions and ,) The rate of movementof water through the ground is of a different magnr'tudd than that through natural or artificial channelsor conduits.Typical flow ratesrange the impact of from 5 ft/day ti afew feet per year. Theselow ratesof flow exacerbate natural since cleanup complicate and sources groundwater contaminani spills on years occur' to many flushing from the site may take in this chapterfor estimatingflows to collection devices The methodsdescribed principles of fluid flow embodiedin Darcy's law. Applicafions arebasedmainly on the in the but undermostconditionsencountered range, laminar arelimited to flowsin the mechanics the describing equations of ure the of Examples field, Darcy's law applies. of flow to wells andinfiltration gilleries were given in this chapter.Both steady-state as well' and unsteadvflow conditions were addressed

PROBLEMS
coefflcient of 19.1. A 12-in. well fully penetfatesa confined aquifer 100 ft thick. The permeabilityis 60d gpd/ft'. Two test wells located40 and 120 ft awayshow a difference in drawdown U-"i*een them of 9 ft. Find the rate of flow delivered by the well. of 1g.2. A l2-in. well fully penetratesa confined aqpifer 100 ft thick. The coefficient permeability is 600 gpd/ft2. Two test wells located 45 and 120 ft away show a the difference in drawdoin between them of 8 ft. Find the rate of flow delivered by well. well' The 19.3. Determine the permeability of an artesianaquifer for a fully penetrating pumping steady-state The ft thick. aquifer is composedof medium sand and is 100 ft, and the 14 is away 75 ft well observation in an rate is 1200 gpm. The drawdown per day gallons Kin Find is l.2ft. away ft 500 well observation drawdownin a second foot. per square ftj/day/ft' Lg.4. Considera confined aquifer with a coefficientof transmissibilityT of 680 Att:5min,thedrawdowns:5.6ft;at50min,s:23'Ift;andat100min's: well is 75 ft awayfrom the pumpingwell. Find the discharge 28.2 ftrThe observation of the well. 19.5. Giventhefollowingdata:0:59,000ft3lday'T:630ft3/day't,,3}days'r:1ft' the drawand s" : 6.4 x L6,4. Considerthis to be a nonequilibriumproblem.Find down s. Note that for a:8.0x10-e u:8.2 X 10-e

: 13'sa W(a) W(u)= 19.94

'

PROBLEMS 477

19.6. ' Determinethe permeabilityof an artesianaquiferbeingpumpedby a fully penetrating pumping of medium sandis 130ft thick. The steady-state well, The aquifercomposed ft, and in a is 12 away well 6{ft in an observbtion gpm. The drawdown is 1300 rate secondwell 500 ft awayis 1.2 ft. Find Ky in gpdlft2. 1g.7. Considera confined aquifer with a coefficientof transmissibilityT : 700 ft3lday-ft. : 5.1ft; at50min,s = 20.0ft; at 100min,s = 26'2ft. Atr : 5 minthe drawdown of the well. The observationwell is 60 ft from the pumping well. Find the discharge pumping gpm and is confined of 300 pumped at atate 19.8. Assumethat an aquiferbeing testdata are given as follows.Find the cbefficientof transmissibilityTand the sttragq coefficientS. Assumer = 55 ft. | )
Time sincepumping started (min) Drawdowns (ft)

1,3 4.6

2.5 8.1

4.2 9.3

8.0 12.0

11.0 15.1

100.0 29.0

19.9. We are given the following dat4: Q = 60,000 ft3ldaY t:30days r:lft s"=6.4 x 1o-4

fcl(dar(fo r : 650
a : 8.0 x 10-e u : 8.2 X 10-e u : 8.6 x 10-e

Assumethis to be a nonequilibriumproblem. Find the drawdowns. Note for lV(u) : 18.06 W(u) = 19.94 W(u) : 17.99

19.10. An 18-in. well fully penetratesan unconfinedaquifer 100 ft deep.Ttvo observation wells located90 and 235 ft from the pumped well are known to havedrawdownsof 225 and20.6ft, respectivd. If the flow is steadyandKy: 1300gpd/ft2, what would be the discharge? lg.1l. A confinedaquifer 80 ft deepis being pumpedunderequilibrium conditionsat a rate ' the aquifer. Watpr levelsin observationwells of 700 gpm. The well fully penetrates Find the field 150 and 230 ft from the pumped well are 95 and 97 ft, respectively. coefflcient of permeability. lg.112. A well is pumpedat the rate of 500 gpm undernonequilibriumconditions.For the data listed, find the formation constantsS and Z. Use the Theis method.

r"/t
1,250. 5,000 tt,250 20,000 45,000 80,000 125,000 180,000 245,000 320,000

Averagedrawdown, h (ft)
3.2+

2.t8 .1.93 r.28 0.80 0.56 0.38 0.22 0.15 0.10

478

DEVICES ANDCOLLECTION 19 WELLS CHAPTER "We are given a well pumping at arate of 590 gpm. An observationwell is locatedat 19.13. r : 180 ft. Find S and Z using the Jacobmethod for the following test data.

Drawdown (ft)

Time (min)

Drawdown (ft)

Time (min)

0.43 0.94 l 08 1.20 r.34 |.46 1.56 1.63 1.68 1.71 l 85 1.93

26 78 99 t3r t'73 2t8 266 303 331 364 481 5'13

2.00 2.06 2.12 2.15 2.20 2.23 2.28 2.30 2.32 2.36 2.38

66r 732 843 926 IO34 r134 1272 1351 I4t9 r520 161 1

tg:t4.

the full depthof an unconfinedaquifer.The original A24-1n.diameterwell penetrates belowthe were located50 and 150ft, respectively, aquifuge table and a bedrock water land surface.After pumping al arate of 1700 gpm continuously for 1920 days. and the original water levels in equilibrium drawdownconditions were established, of the pumpedwell were the center ft from 100 observationwells located 1000 and respectively. 20 ft, 10 and lowered a. Determinethe field permeability (gpdlftz) of the aquifer. b. For the samewell, zero drawdownoccurredoutsidea circle with a 10,000-ftradius drawfrom the centerof the pumpedwell. Insidethe circle, the average measured down in the water table was observedto be 10 ft. Determine the coefficient of storageof the aquifer. 19.15. A well fully penetratesthe 100-ft depth of a saturatedunconflned aquifer. The using drawdownat the well casingis 40 ft whenequilibriumconditionsare established of 50 gpm. What is the drawdownwhenequilibriumis established a constantdischarge of 66 gpm? using a constantdischarge by 8 cfs from period, the flow in Wahoo Creek decreases long rainless a 19.16. After an unconfinedaquifer, 8 mi to Ashland.The streampenetrates Memphisdownstream wherethe water table contoursnear the creek parallel the westbank and slopeto the streamby 0.00020, while on the east side the contours slope awayfrom the stream toward-theLincoln wellfield at 0.00095. Computethe transmissivityof the aquifer knowing Q : TIt, where1 is the slopeand I is the length. 19.17. The time-drawdown data for an observationwell located 300 ft from a pumped well (500 gpm) are given in the following table.Find the coefficientof storage artesian (ft3 of water/ft3of aquifer)and the transmissivity(gpd/ft) of the aquiferby the Theis method. Use 3 x 3 cycle log PaPer.

REFERENCES

479

Time (hr) 1.8 2.\

. A

Drawdown (ft)

Time

(h0
9.8 12.2 t4.7 16.3 18.4 21.0 24.4

Drawdown (ft)

3.0 3.7 4.9 7.5

0.27 0.30 0.3'l 0.42 0.50 0.61 0.84

r.09 t.25 1.40 1.50 l 60 t.70 1.80

level of the water table for an unconfined 19.18. Over a 100-mi2surfacearea, the average

of the removalof 128,000 area-ftof waterfrom the aquiferhas dropped I 0 ft because for the aquifer.The specificyield is 0.2 and coefficient the storage aquifer.Determine the porosityis 0.22. 19.19. Over a 100-mi2 surface area, the averagelevel of the piezometric surface for a confined aquifer in the Denver area has declined 400 ft as a result of long-te1m pumping.Determinethe amountof the water (acre-ft) pumpedfrom the aquifer' The porosity is 0.3 and the coefficient of storageis 0.0002. L9.20. Find the drawdownat an observationpoint 250 ft awayfrom a pumping well, given -: thatT = 3.1 x 104 gpdlft,the pumpingtime is 10 days,S" 3 x 10-4, andQ 280 gpm. 19.21. Find the permeabilityof an artesianaquiferbeingpumpedby a fully penetratingwell. pumpof medium sand.The steady-state The aquiier is 130 ft thick and is composed ft' and is 12 ft away ing raie is 1300 gpm. The drawdownin an observationwell 65 gpdlft2. Kyin Find well 500 ft awayit is 1.2 ft. in a second an unconflnedaquifer 100 ft deep.Two observation fully penetrates in. well An 18 19.22, wells located90 and 235 ft from the pumpedwell are known to havedrawdownsof 22.5 ft and20.6 ft respectively.If the flow is steady arld Ky: 1300 gpd/ft2, what would be the discharge? : 0.0028m2ls, 19.23. A well is being pumpedat a constantrate of 0.004 m3/s.Given thatT : in the obserdrawdown find the : 0.001, coefficient 100meters,and the storage r (b) hours' 24 (a) hr, and t period of time vation well for a = 0.0028mzls, 19.24. A well is being pumpedat a constantrate of 0.003 m3/s.Given thatT is 12 hours,find pumplng began = since time 0.001, and the the storage"oiin"i"nt (b) 500 m' (a) m, and 150 of distance radial for a well observation the drawdownin an

REFERENCES
1 . J. W Clark, w. viessman, Jr., and M. J. Hammer,water supply and Pollution conffol, 2nd ed. New York: Thomas Y. Crowell' 1965. "Field TestsDeterminePotential Quantity, Quality of Ground Water 2. John F. Hoffman, Supply," Heating, Piping, and Air Conditioning(Aug' 1961)' 3 . D. K. Todd, GroundwaterHydrology, New York: Wiley, 1960'

Chapter20

Regional Modeling SYstems Groundwater

r Prologue
of this chaPteris to: The PurPose ' Describethe featureS of large-scalegroundwatersystems. to modelingregional . Introduce the principlesof finite differenceapproaches groundwatersystems. . Iilustrate the application of groundwatermodeling techniquesto the Upper Big Blue basin in Nebraska. flow of The analytical methodsdescribedso far havebeen applicablemainly to the groundregional warcr rc individual wells. In this chapter,the conceptsof analyzing are requisitesfor the wise development, are introduced.Suchanalyses water systems management,andoperationofexpansivegroundwaterresources' mustbe dealt with jointly in most quality aspects Given that waier quantity an-d modelsmrrstoften prol"tt"t, t"gional groundwater decision-making waterresources of groundaspects The fluid flow to includeboth of tfuse dimensions.l-2s be designed and complex are in this chapter.Solutetransportmodels. water modelsare presented end of the given at is this book, but a biief introduction to them beyondthe scope-of of role and the chapter.It is important for the readerto understandthe importance thesewater quality-orientedmodels'

MODELS GROUNDWATER 20.1 REGIONAL

(mathematical) Groundwatersystemsmodels may be of the analog or the digital type most variety. The focus of this chapter is on the digital type of lodel, the equations of a set by commonly employedtoday. Such models are characterized be models.may These ,"pr"r"nting the ihysical i.o""*"* occurring in an aquifer. discussed are in nature, but only deterministic-models deierministt or pio-babilistic relations stemmingfrom known featuresof the cause-effect the describe They here. physicalsystemunder study'

482

CHAPTER20

SYSTEMS MODELINGREGIONALGROUNDWATER

Simplify equation so that solutions may be obtained by analytical methods

l _

Approximate equations numerically resulting in a matrix equation that may be solved using a computer

t_

model' a mathematical Figure 20.1 Logic diagramfor develqping OH') Worthington' Association, Well Water (Courtesy National of the

a deterministicmathethe procedurefor developing Figure 20.1 characterizes matical model. A conceptual model is formulated based on a knowledge of the of the mechanicsof groundwater of the region and an understanding characteristics flow. The next stepis to translatethe conceptualmodel into a mathematicalmodel, accompanied by a partial differential equationor setof equations usuallyrepresented boundaryand initial conditions.Conditionsof continuity and conserUy approplia@ uution of momentum, usually describedby Darcy's law, are incorporated in the and includeartesianor watertable condition designation model,Other modelfeatures transheat quality and/or If water dimensionality(one-,two-, or three-dimensional). describing areto be incorporatedin the model,additionalequations fer considerations are energy of involvedand conservation of massfor the chemicalspecies conservation and Fourier's required.Typically usedrelationsare Fick's law for chemicaldiffusion law for heat transport. Once the mithematical model has been formulated, it can be applied to the situation at hand. This requiresconvertingthe governingequationsinto forms that through the use of numericalmethods facilitate solution.Ordinarily this is achieved partial differenthe applicable to represent or dnite elements suchasfinite differences is divided region the for example, In using a finite differenceapproach, tial equations. variables discrete as into grid elementsund th" continuousvariablesare represented by replaced is at the nodal points. In this manner,the governingdifferential equation Models way. iterative that canbe solvedin an a finite nu-b"r of algebraicexpressions of this type find wide applicationin the estimationof site-specificaquifer behavior. They havepro\iento be effectiveunderirregularboundaryconditions,wherethereare and where highly variablepumping or rechargerates are expected'l heterogeneities, Severil types of groundwater models and their applications are summarizedin Figure 20.2. region' A numberof stepsmustbe followedin modelinga targetedgroundwater be may They Figure 20.3 is illustritive. The f,rst step is to define the boundaries. politically, a of choice ptiysical,suchas an imperviouslayer, or arbitrary, suchas the

Applications Watersupply
Regionai aquifer analysis Near-well performance Groundwatelsurface water interactions Dewatering operauons Seawaterintrusion Land fills 'Geothermal Thermal storage Land subsidence

Waste injection

Heat PumP Thermal poliution

Radioactive waste storage Holding pontls Groundwater pollution

Figure20.2Typesofgroundwatermodelsandtypicalapplications.(Courtesyof OH') Worthington, WellAssociation, theNationalWater by Next, the region is divided into discreteelements or otherwise,definedsubregion. 20'4)' (see Figute grid a rectangularor polygonal superimposing ' (s. and z) and the controlling aquiferparameters once th-egrid is determined, is includedin the transport solute If grid element. the initial conditionsare setfor each must also properties dispersion suchai hydrodynamic model, additional parameters operated be can model the met, been hive Atter atl of thesespecifications be specifigd. of comparisons matching). (history and its output comparedwith recordedhistory permit predictions model counterpart with recordedvaluesof headand other features and computeddata are considered to be madeuntil observed parameteradjustments by the modelerto be in closeagreement. to analyzea Upon completion of thelodel's calibration, it can be applied of the prediction model's The options. and/or development variety of management decision-making to aid valuable a be can outcomesof these alternative strategies include: the Examplesof the types oI problems that can be addressed processes. aquifer of an on impact the use; of utilty of an aquifer to suppoit various levels storageof underground on effects the varying naturaiand artificial rechargerates; contamsubsurface of movement of rate andpumpingrate; the spacing, weli lo-cation, intrusion. inants;and saltwater caution while numerical groundwatermodelshavemuch to recommendthem, Prickett appropriately. interpreted and to enJurethat they are used must be exercised three waysin notesthat overkill, inappropriatepiediction, and misinterpretationare both the modeler pitfalls, these avoid To -#"t, can Uemisused.s which groundwater

484

CHAPTER20

MODELINGREGIONALGROUNDWATER SYSTEMS

illiiT3:':,",*itlffi,ffi:ff tlSl

(courtesv Nation ar water or the use'

upon which the model was and user must understandthe underlying assumptions founded,its limitations, andits sources of errors.Usedwisely,modelscanbe powerful decision-makingaids. Used inappropriately,they can lead to erroneousand sometimes damagingproposals.

20.2 FINITE-DIFFERENCE.METHODS
Digital simulation requires an adequatemathematical description of the physical processes to be modeled.For groundwaterflow this descriptionconsistsof a partial differential equationand accompanying boundaryand initial conditions.The governing equationis integratedto produce a solution that gives the water levels or heads associatedwith the aquifer being studied at selectedpoints in space and time. so that the The model can simulateyears of physical activity in a span of seconds,

METHODS 485 2A.2 FINITE-DIFFERENCE

(a)

A.r Ly ..,--V ^t ,
/_-/ | 1"

', / 1 . 1 , / 4Finitedifference
/
/

srld DlocK

(b) (b) Finite-difboundaries. Figure 20.4 (a) Map view of aquifershowingwell field and is the spacing At' t i"r""n"" grid{oi aquifir study,wGre Ax is the spacTqii lhu *ection' nodes;open block-center dots: Solid thicknesg. in tt" y-at"rtion, and f is itre aquifer worthingAssociation, well Water National the of circles:source-sink nooo tcoo*sy ton. OH.)

486

CHAPTER20

SYSTEMS MODELINGREGIONALGROUNDWATER

of proposedactionscan be evaluatedbefore decisionsinvolving conconsequences struction or social changeare implemented.The expectationis that the model runs * will lead to wiser and more cost-effective'decisions. of an aquiferinto a grid The finite-differencemethodis basedon the subdivision with zonesof the aquifer.The equationthat must and the analysisof flows associated and Darcy's law for groundwater be solvedis derivedfrom continuity considerationq motion. This yieldsthe followirigpartial differential equation(a versionof Eq. 18.41), aquifer.Note that the equationpresented flow through an areally extensive describing here describes the two-dimensionalcase:

s#P.N#ot:s!+w
where h : x : y : S: Z : V[ : total hydraulic head (L), x direction in a cartesiancoordinatesystem(L), y direction in a cartesiancoordinatesystem(L), specificyield of the aquifer (dimensionless), transmissivity of the aquifer(I]lT) sourceand sink term (L/T)

(20.r)

In the aboveequation,vertical flow velocitiesare consideredto be negligible areimplict in the derivation:the in the aquifer.The following assumptions everywhere hydraulicconducand space; in time is constant fluid density flow is two-dimensional; specificyield of law; and the Darcy's flow obeys within the aquifer; tivity is uniform nonlinear for unconfined 20.1 is time. Equation is constant in space and the aquifer dependent variable. thus the head and is a function of transmissivity aquifersbecause saturated head, transmissivity, of initial values to integrate Eq.20.I, In order and sinksmust thickness of the aquifer,andthe amountsof waterproducedby sources yield and The specific integration. point region of the for every in the be identified analytic Unfortunately, be defined. geometric must also boundaries location of It is trivial cases. for the most to obtain except Eq. 2O.l are impossible solutionsto desired obtain the to techniques integration necessary to resort to numerical thus
answers.t'to-to

to groundwaterflow problems reApplication of finite-difference techniques or elements quiresthat the region of concernbe divided into many small subregions valuesof all the variablesin (Fig. 20.5). For each of theseelements,characteristic which to the centersof the elements, Thesevaluesare assigned Eq. 20.I are specified. a finite-difference through are related in nodes nodes. The heads adjacent are called equation,which is derivedfrom Eq. 20.1. Thesedifferenceequationscan be derived The or by massbalanceconsiderations.8 by an appropriateTaylor's seriesexpansion yield heads at to the simultaneously then be solved equations can resulting algebraic for each time step considered. eachnode that the simulation methodspresentedin this chapter It should be understood of regional rather than localized groundwaterprobpointed the analysis toward are the prediction of the drawdownat a particular well. In suchcases, as the lems such we,are Here most appropriate. usually the in 19 Chapter are methodsdiscussed that might occur over a large area with waterlevel or headchanges mainly concerned practices. prescribed water-use due to

2O.2 FINITE-DIFFERENCEMETHODS 487

t, I

o m,1
J

{
eAx'---

Node

Subregion

I Ay

l,,i

I
\ \ l, nd Boundary ofregion of integration a m,n

for a finiteFigure 20.5 Subdivisionof a region of integrationinto computationalelements differenceProblemformulation.

BoundaryConditions
In order to integrateEq.20.l,the governingboundary conditionsmust be-specified' in Chaphere.Otherswere presented Two typesof bJundary condition aie discussed ters 18 and 19. chosen Where the region of integrationis limited by a political or arbitrarily condition'10 boundary constant-gradient boundary,it is oftei the policy to employ a change In this "ur", unar.umption is madethat the gradientof the watertable will not with streams Where fall. or rise may llvel alongthe boundaryeventhoughthe water condiboundary stream encountered, are system to the groundwater interconnections mathematicallyas boundariesare expressed tions are employed.Constant-gradient
a h l \

(20.2)

*:81x'!) at the location x, y throughoutthe period of specified y): a constant g\x, where simulation (dimensionless) /z : hydraulic head (L) s : direction perpendicularto the boundary (L) as Streamboundariesare expressed h : f(x, Y, t) wheref(x,y,/)=anunknownfunctionoftimeatthelocationx'y (dimensionless) = hydraulic head (L) h

(20.3)

488

CHAPTER20

SYSTEMS MODELINGREGIONALGROUNDWATER

by boundariesdescribed The volumetric rate of flow acrossthe constant-head Darcy equ-ation:ro the Bq.20.2 can be modeledat eachtime stepusing

o: rfrtr

(20.4)

where h:head(L) Al = dummy variable denoting the length of the side of the subregion perpendicularto s (L) s : dummy variable denoting the direction of flow perpendicularto the boundary (L) (LilT) p : volumetricdischarge Z : transmissivityat the boundary (IllT) Use of this equationat a boundary is illustrated by the notation of Fig. 20.6. Considerthe flow from left to right in the x direction acrossthe left-handsideof the elementalregiondepicted.The node i - I, j lies outsidethe regionof integrationand may An assumption that no information aboutit is available. thus it may be assumed the boundary across be madeto circumventthis problem.It is that the transmissivity is uniform and equal to Tt,t. In finite-differenceform the head changeterm in Eq. 18.26 can be statedas
ah _hi.j-hi-l,j

6x

A,x
hi*r,j -

(20.s)

But the headh,-r,, doesnot exist, and anotherapproximationis required,

h,,i h,-r,i : hi,j

(20.6)

in Eq. 20.4 to yield are then substituted Thesetwo expressions

f,,,u#'LY Q;-r/",i:

(20.7)

Boundary where 6 h= r . dx

_ 1 'i a ^ ' i'

Figure 20t6 Subregions adjacent to a constantgradientboundary.

2O.2 FINITE.DIFFERENCEMETHODS 489

At the beginningof each time step, a new volumetric flux is calculatedalong each by usingthe headsand transmissivboundary.This is accomplished constant-gradient previous interval. . time ities computedin the in groundboundaries as constant-head treated are sometimes Surfacestreams body in the surface level water the where problems. is adequate The assumption water process. In modeling period of the the time during remain unchanged is expected to by affected are significantly heads, hence flows, and surface however, many instances, They may then groundwatersystem. to the interconnected withdrawalsor recharges the To accommodate groundwater. system. for the supply of water be a limited source expression This be applied.to may leakage term linkage, a water-groundwater surface may take the form

= -fi{n,.,.r leakage,,r,1
where

h,i,o)

(20.8)

b, , : thicknessof the streambed(L) hi,i.t : head in the aquifer at node i, j, at time k; k : 0 indicatesinitial conditions(L) t k;,; : hydraulic conductivity at node i,.i (LlT).

to coverthe entire arearepreWhen Eq. 20.8 is used,the streamis considered from the streamto the leakage the time step After each node. sentedby the related If the streamflowat a accordingly. are depleted and streamflows aquiferis calculated particular node becomeszero, the model can be madeto note that the streamis dry and break the hydraulic connection atthat point.l0

Time Stepsand ElementDimensions

of any finite-difference schemedependson the incremental valuesasThe success and the time steps.In general,the smallerthe dimensignedthe elementdimensions sionsof elementsand time increments,the closerthe finite-differenceapproximation to the differential equation.However,as thesepartitions are madesmaller,a price in may computationalcostsand data needsmustbe paid. Furthermore,oversubdivision degree evenbring about computationalintractability. Thus the object is to selectthe of the systemwhile keeping represehtation of definition that results in an adequate data and computationalcosts at a minimum. There are proceduresfor making such but, exceptfor a brief discussionin the following section, they are not selections, presented here.lo-la

Flow Model One.Dimensional

To illustrate the finite-difference approachto groundwaterproblem solving, a onemodelsare Although mostpractical-scale is discussed. conceptualization dimensional of the an extension is only their development in character, two- or three-dimensional reader the models complex more the of some of For details case. one-dimensional The book by McWhorter and should consult the appropriate references.6-8'i0-1s problems.8 The treatment of example excellent includes to read and Sunadais easy reference. of that the approach here follows taken flow one-dimensional

490

CHAPTER20

MODELINGREGIONALGROUNDWATER SYSTEMS Upstream headHn

Downstream head I{, H6*Huatt=0 Flow
hi*t

-----w

IlD-constantatt>0

Confined aquifer

l+Ay+l (After groundwater flowcase. Figure20.7 Gridnotation for a one-dimensional McWhorter andSunada.8) suchasthat Let us considera one-dimensional flow in a confinedaquifersystem illustratedby Fig. 20.7.lt is assumed that the flow is unsteadyand that the flowlines On this basis,a unit width of the aquifercan be are parallel and not time dependent. As madeaboutit can easilybe translatedto the total system. studiedand observations shownin the figure, the unit width of the aquiferis A.r. The flow regionis overlaidby a grid, and for eachgrid element,valuesof hydraulic conductivity Kr, elementlength y,, aquiferthicknessb,, storage coefficientSi, and the initial valuesof headh, must be specified.The massbalancefor grid elementI requiresthat the inflow (Qr-t-,) from I * I element I - 1 to element I minusthe outflow{Qt-,*r)fromelementlto element which occursin elementi, LV,/LI. must be balancedby the rate of changein storage To simplify the problem, let us further considerthat the aquifer is of uniform b, and Ay and isotropic. Thus the valuesof K, ,S, thicknessand that it is homogeneous are constant,and we shall considerthat from studiesof the aquifer properties,they are also known. Thereforeit mav be statedthat
K t : K z : ! . . : K ^ : K

Sr:Sz:..':S.:S bt:bz:',,:b*:b N r : L Y ,= ' ' ' : L Y - = N

(20.e)

Usingthis notation ln represents the total numberof grid elements. wherethe subscript and Fig. 20.7, we can seethat the flow from elementi - 1 to I is
. Oi-r-i: -UOO'| ;i '

(20.10)

where i : the elementnumber time n : the selected It is assumed in this repreas Darcy's equation. Equation20.10is recognized sentationthat the head generatingthe flow at time n is the differencebetweenthe averageheadsat the two adjacent elementsdivided by the distancebetweentheir as Ay diminishesto zero. exactness centers(nodes).This approximationapproaches

2O,2 FINITE.DIFFERENCEMETHODS 491

area of flow and is in Eq. 20.10 is the cross-sectional The ireaA appearing obtainedasthe productof Al and b. Sincewe are dealingwith a unit width of aquifer, A.r : 1 and sinceb is a constantby definition here, Eq. 20.10 mtry be written
o,- -': hl - " h,! i-t -7"i '

Ay

(20.11)

for the flow from element ito i * 1 may be where 7 : Kb. A similar expression obtained:

(20.r2)
the inflow and outflow from elementi. ConsidEquations20.11 and 20.12 represent ering that continuity conditionsmust be met, this changein flow acrossthe element which occursduring the time step.This is *urt b" balancedby the changein storage siven as

in the continuity equation(inflow - outflow : Now insertingthesethreeexpressions we get changein storage),

fr:'.(ry)

' (zo.tz)

t''i) (-rhi -,ni ) - f -rhi-,-- hi) : , 6,(h'lo'^. '\ ^t AY / / Av / \ \

becomes the equation By rearrangement,

1zo.r+i

(20.1s)
which is known as the explicit or forward difference form of the finite-difference as the current value of time. If, on the other hand, n is equationif n is designated definedas / + Ar, thin the equationis the implicit or backwarddifferenceequation. The explicit solutionto Eq. 20.15 Eachof theseforms hasits own solutiontechniques.8 here. will be discussed By letting n : t inF;q.2O.l and rearranging,one obtains

+ tur+^, ffirr;, +hi-) nllr ##]

(20.16)

derivativesare centeredat the beginningof the time stepand the In this casethe spabe singleunknownls h!*^'. Equation 20.16 canbe solvedexplicitly at eachelementfor only on a knowledgeof the the headat the next period of time. The solutiondepends

492

CHAPTER20

SYSTEMS GROUNDWATER MODELINGREGIONAI-

here, case discussed an unstable condition. Irt the one-dimensionalhomogeneous if stability is assured

<1 s/,vY t
TLt

(20.r7)

The equationshowsthat the choiceof time and spaceincrementsis not independent. an accurateapproximation,however;it Satisfactionof Eq. 20.17 doesnot guarantee solution.8 for a stable provides only EXAMPLE 2O.T Refer to the one-dimensionalflow problem of Fig: 20.8. Let us assumethat the elementlength is 4 m and that the thicknessof the confinedaquiferis 2 m. It is further thit the headat the left and right sidesof the region is 8 m at / : 0 and that assumed : the head on the right side takes on the value 2 mfot all t greatet than zero. K the Using : 0.5 m/day and S 0.02. As shown in the figure, there are five elements. notation of eq. 20.16,the initial condition is hf; : 8.0 m. Use the explicit methodto determinefuture heads. Solution 1. First a determinationmust be made of the time step to use. This may be usingEq. 20.17. accomplished Lt <1 s(Ay)' :;W:

0 . 1d 6a v s

was obtainedusingthe relation The valueof Zused in the aboveexpression T:Kb: Z:0.5X2.0:1.0 a time stepof 0.1 days,althoughany To be on the safeside,we shall choose stability. value lessthan 0.16 would have assured

8m

A)=4m Figtrre 20.8 Sketch for Example 20.1.

493 20.3 FINITE-ELEMENTMETHODS ^' 2."For the first time step,t : 0.1, we can calculateh'a* and corresponding headsfor the other elementsusing Eq. 20.16.Thus

: h,;",

* 0f,rful ," -s(Ay),1 + hg) ,,411 f# @?

and substitutingnumerical values,we get

^f. - 2(1.0)(0.1)l ,n, : 1.0(ol) + 8.olt /,1' (ffi(2.0 + 8.0) drn+Fl

: 3 . 1+ 3 . 0: 6 . 1 3 m Sinceft? andho2: 8.0 m and sinceh, : 8,0 by definition,it can easilybe from shownusingEq. 20.16 thatthe valuesof hlr andhl't ate not changed step. their original level of 8.0 during the first time 3. Now considerthe secondtime step,t + Lt = 0.2 days.For element4,

: h2,
For element3

: 0.31Q.0+ 8.0) + 6.13(0.37 ) : 5.4 m

#(h2,

+ nZ'lt + hg\

#]

hg': ffi] ffi(h:'+h9\+l,l'[r

:7 : 0 . 3 1 ( 6 . 1+ ) '4m 3 8 . 0 )+ 8 . 0 ( 0 . 3 7 Element 2 doesnot have a head changeuntil the third time step. until the headshavebeencalculated is repeated 4. The processdemonstrated for the total time period of interest.For this example,they will ultimately reachequilibriumconditions. rl This exampleproblem illustratesthe mechanicsof the finite-differenceprocedure. Problemsof practical scalewould require the use of a computer,but the approach would still be the same.

METHODS 20.3 FINITE.ELEMENT

flow problemsare for solvinggroundwater The mostwidely usednumericaltechniques The finite-elementmethodis similar the finite-differenceand finite-elementmethods. in lead to a set of N equations to the finite-differencemethodin that both approaches Nodesin the finite-elementmethodare Nunknowns that canbe solvedby relaxation.6 usually the corner points of an irregular triangular or quadrilateral mesh for twobricks or tetraheapplications, while for three-dimensional dimensionalapplications, are arbitrary. drons are commonly used.The sizeand shapeof the elementsselected to fit the applicationat hand.They differ from the regularrectangular They are chosen grid elementsusedin finite-differencemodeling.Elementsthat are closestto points of flow concentratiensuchas wells are usually smaller than those further removed

494

CHAPTER20

MODELINGREGIONALGROUNDWATER SYSTEMS

from suchinfluences. Aquifer parameters suchashydraulic conductivity may be kept constantfor a given elementbut may vary from one to another. To minimize the variational function, its partial derivativewith respectto head,,is evaluatedfor each node and equatedto zero.The procedureresultsin a set of algebraicequations that can be solved by iteration, matrix solution, or a combination of thesemethods.ra Finite-elementmodelersmust understand partial differential equationsand the calculus of variations.6 The finite-elementapproachoffers someadvantages over the finite-difference technique.Often, a smaller nodal grid is required, unA tnir offers economies in computer effort. The finite-elementapproachcan also accommodate one condition that the finite-difference approach is unable to handle.6When using the finitedifferencemethod,the principal directionsof anisotropyin an anisotroplcformation are parallel to the coordinatedirections.In caseswhele two anisotropicformations having different principal directions occur in a flow field, the finite-difference approach cannot produce a solution, whereasthe finite-element approach can. TLe finite-elementtechniquecan be used to simulate transient aquifei-performance. A detaileddiscussion of the finite-elementtechniqueis beyondthe scopeof this boor<, but there are many good references for the interestedreader.6'rsts.zi ,

20.4 MODEL APPLICATIONS

To illustrate how simulation modelscan be usedto provide insightsinto water managementschemes, a model analysisof the Upper Big Blue basin aquiferin Nebraska is presented' The studywasconducted by the Conservation and SurveyDivision of the University of Nebraskaunder the direction of Huntoon.l0 The useof groundwater for irrigation in the Upper Big Blue basinwas observed to be rapidly increasingand by 1972 about3.: wetlslmit iere in operation.At that time farmerswere becomingconcernedaboutthe progressive declineof water levels and were seekingguidanceaboutthe efficiencyof implementingsomeform of basinwide water management proram. The Universiry of Nebraski designeda model to evaluate the situationand to explorevariousproposalsfor recharging:the aquiferand for estimatingthe long-term consequences of siveral scenariosof water use in the basin. The study area is shown in Fig. 20.9. Generallythe water table is free in the regionof interest'Figure 20.10 showsthe configurationof the watertable asobserved in 1953.For modelingpurposes, the water-levelcontoursshown were considered to be representative of predevelopment conditions.This assumptionwas basedon the fact that groundw-ater withdrawals before this time were not extensive. It was also surmisedthat the Lonburs represented a water table in which an equilibrium existed between natural recharge and discharge in the region.Transmissivities wereestimated from drill-hole sample logs recorded in the area. These values are needed for modelingand are also important indicesof the potential yield of wells that might be constructed. As might be suspected, the information of most concernto the local landowners and water plannerswas the rate of decline of the water table. In particular, it was

.-/l

t-

z
()
I

6)

bo

o\
N

b0

,-

H H

3
d
C) C) a

o
C)

o
a

b
Q

?
a

ca ra

U
o
6 H

e F (.)

-.----d

e o rrl

a o d

N
ra

p R

.$E

20.4 , MODELAPPLICATIONS

497

where and resourcewould Le depleted, desiredto knbw how rapidly the groundwater use, and what on water when waterlevel declineswould posean economicconstraint of decline. on thdrate would have and/or management impactsfuture developments represento explorethesefeatureswas a two-dimensional The model developed tation of flow through an areally extensiveaquifer.loEquation 20.1, alongwith the appropriate boundary conditions, constituted the model. The region shown in was divided into a finite-differencegrid and, after substitutionof the nodal n1g.Z0.S to the year ualu"t of Z and S, the model was operatedto predict water-levelchanges Calibradevelopment. of 2020 for variouspolicies of rechargeand for severallevels over was operated model The usinghistoricdata. tion of the model was accomplished net the average and wells the period 1953-1972 using the known distribution of water-level and estimated pu-pug" per well to establisha match betweenobserved the simulation of future trends proceeded. "ttung"i. Once this was accomplished, process. in the matching achieved Figures20.II and20.12showthe correspondence study in the levels water that it wasdetermined On the basisof the model studies, It l912level. to the was limited areawould continueto declineevenif development groundwasevere wasfurther predictedthat someparts of the areawould experience ter shortageiby the year 2000. It was found, however,that by employingartificial the To assess rechargefrethods, permanentgroundwatersuppliescould be assured. of these Both modeled. were two water delivery systems effectsof artificial recharge, delivered water from Platte River Valley sourcesto rechargewells located in the were three rechargeschemes project area. Using thesetwo water delivery systems, of cancellation the was ii-ulut"d. The grosseffect of introducingthe rechargewells the 2O-I3 shows Figure the effects of the proportionatenumber of pumping wells. plan computedwater-levelchangesat one location under a graduateddevelopment and.then recharge no with (projected on the basisof the 1972rate of development) The continual for eachof the three rechargeschemes. wlth graauateddevelopment the nature of (curve shows clear$ 1) downwardtrend in waier level with no recharge artificial the three depicting the problemin the UpperBig Bluebasin.The othercUrves is taken. approach an such ,""hurg" options show that stability can be achievedif it is excessive, be might While the costs of implementing artificial recharge of problem, short table water apparentthat any long-terrn solution to the declining water. sourceof reducinguse, would require a supplemental Operationof the modelprovidedusefulinsightsinto the natureof the watertable that irrigators shouldbe making someimportant water manproblem and suggested agementdecisionsabout their future mode of operation..^ ^In structuring models is complex.lo-25 The modeling of groundwatersystems be made.Thesehave usually must simplifying assumptions suchasthat just discussed, (for unconfined yield to do with aquifer pbrameterssuch as transmissivity,specific the boundary Furthermore, aquifers),and storagecoefficient (for confined systems)' physical system.and in the conditions are normally approximationsof what occurs are somestrata subsurface aboutthe uniformity of materialsin various assumptions to yield expected be cannot models times crude.This doesnot meanthat groundwater how about cautious be must useful results.It doesimply that the usersof the models that as such model aquifer they interpretthe output.For example,an areally extensive give to expected be problem can by Huntoon for analyzingthe Blue River developed

o E

Z-u---sI

E.l(

F \ $

o
a
H H

()
o

e
a

:d65'
\ -o-

\=e 1t "-

r-O

t\-z
,
lo

U U
o

9pc
H ' A

o o i * O d O

E =
H E

d 9
: : 4 J

>.o
N X o ! l L ; hn:\ a;-Y

F I X ti ;i

2O.4 MODELAPPLICATIONS ^ 8 2 o

499

.8 s4
d 8 6 e 6 6
A 1/

rl--

Computed trend

Eno
P, 92 e 9 4 6

\/
V
Measured water levels

tse6
h q R

v
Y
1958 t959 1960 t961 t962 1964 1965 Year

F 100 1,02
1956

t967 1968 1969 1970 r97I 1972 1973

trends'(AfterHuntoon'10)' water-level andcomputed Figure20.1.2 Measured reliable information about water-leveltrends for various conflgurationsof developan accuratepredictivetool for ment. It shouldnot, on the other hand, be considered at somespecificpoint in the regionof concern.This monitoringthe water-levelchange type of information could be derived only from a more detailed modeling of the the point. The informationprovidedby the Blue River modelwas lolafif surrounding what the future might hold for severaldevelopment to showlocal landowners targeted options.The actual water levelspredictedby the levelsand for severalmanagement model were not of central concern;what was of interestwas the determinationthat water provided,or unless was restrictedand supplemental unlessfuture development current usescould be significantlyreduced,the outlook in the next 50 yearswas not good for irrigated farming. The model thus providedthe basisfor making somequantitativeobservations from aboutthe future. It also providedinsightsinto the relief that might be expected management Beyondthat, it could be usedto model other possible artificial recharge.

2010

2020

20.13 Computed water-level Figure under a plan of graduateddevelopchanges ment for conditions of (1) no techatge,(2) rechargeunder Scheme1, (3) rechargeunder Scheme 2, and (4) recharge under 3. (After Huntoon.lo) Scheme

500

CHAPTER20

MODELINGREGIONALGROUNDWATER SYSTEMS

options.A,model suchas this, carefully usedand properly interpreted,can thus add a powerful dimensionto decision-making processes.

20.5 GROUNDWATER QUALIW MODELS

quality hasbecomea major sourceof concernin recentyears.This has Groundwater comeaboutfrom the realizationthat many groundwater sources that wereat onetime consideredalmost pristine have now been degradedin quality by seepages from dumps, leakagefrom industrial waste holding ponds, and by other waste disposal and/or industrial and agricultural practices.To deal with suchproblems,'therehas been an expandingmovementto developquantitative techniques to understandthe mechanicsof groundwaterquality. Thesemodels,althoughnot as advanced as their surfacewater counterparts,are now beginningto play an important role in water quality management. The subjectof groundwater quality modelingis complexand underrapid development.Accordingly,a thoroughtreatmentof the subjectis beyondthe scopeof this book. The importance of this topic cannot be overemphasized, however, and the readeris encouraged to consultthe references at the end of the chapter,specifically Refs.6 and 26-30. In 1974, Gelhar and Wilson developed a lumped parametermodel for dealing with water quality in a stream-aquifersystem. The nomenclature and conceptualization of their model are shown in Fig. 20.14.2e The rationale for using a lumped parameterapproachwasthat when dealingwith changes quality over in groundwater long periods of time, temporal rather than spatial variationsare most important. Changes in water table in the Gelhar-Wilson (GW) model are represented by the following equation:
dh 'nd- t=

-q+e*q,-ep

(20.18)

where

h - average thicknessof the saturatedzone p : average effectiveporosity e : natural rechargerate natural outflow from the aquifer 4 , : artificial recharge/unitarea Q p : pumping ratelunit area T _ time

This is just anotherform of the continuity equationrelating inflow, outflow, and the (lefrhand term in Eq. 20.18).The changein concentration changein storage of a constituentis given by
,dc Pndt(e -l q, -t aph)c : ec. * q,c,

(2o.re)

QUALITYMODELS 20.5 GROUNDWATER

. Ct

(b)

model'(After of the Gelhar-Wilson Figure 20.L4 Schematic andChesters.2a) Novotny where c: c; : c, : c: concentration coflcefltrationof the natural recharge concentrationof the artificial recharge of the contaminant a first-orderrate constantfor degradation

may be that dispersionis negligible.This assumption The GW model assumes regional-average iS to estimate model of the objective made on the basis that the The model also provides for the determinationof hydraulic and concentratiolrs.2e of the lag that occursin the Theseare measures times for the system. soluteresponse Gelhar and Wilson the system. to inputs constituent moue-"ni of both water and likened to that of a can be input specific to a an aquifer of that the response assume determinationof model's the that showed studies Their well-mixed linear reservoir. of the average representative is an aquifer leaving the concentrationof constituents that the model appears basis,,it On this aquifer. in the concentrationof the constituent

502

CHAPTER20

MODELINGREGIONALGROUNDWATER SYSTEMS

to a surface stream, is well suitedto estimatingthe quality of groundwater discharging providing-ihe aquifer is narrow relative to the length along which discharge occurs.

r summary
in a regionalaquifer system is constantlyin motion. The amountstored Groundwater evapotranspiration, flow to at any time is affectedby artificial and natural-recharge, springs and surface water courses,and by collection devices such as wells and infi ltration galleries. Natural hydrologic statesmay be significantly affected by human activities. Aquifer depletionshaving regional and national economicimplications are not uncorlmon. Depletionof the Ogallala aquiferin the central United Statesby long-term and extensive water withdrawalsfor irrigation is a good example.On the other hand, by humanintervention. inadvertently, waterlevelshavebeenmadeto rise, sometimes for groundLeaky irrigation canalsin central Nebraskawere at one time responsible waterlevel risesin somefarming locationsof a magnitudesufficientto jeopardizeuse occur, they are of the land. Once major problemsof depletionor over-replenishment policy for groundwater has management not easilydealt with. In general,a safe-yield merit and shouldbe considered.6'30. Regional groundwater flow problems are usually modeled by an equation combiningDarcy's law and the equationof continuity.The resultingpartial differenthe hydraulicrelationswithin the aquifer. tial equationo or setof equations, describes To effect a solution to the governingequation(s),the aquifer's hydraulic features, geometry,and initial and boundary conditions must be determined.Unfortunately, many groundwaterproblems exist for which exact analytic solutionscannot be obtained.In suchcases, it is necessary to rely on numericalmethodsfor modeling.Under suchcircumstances, an approximatesolutionis obtainedby replacingthe basicdifferential equationswith another set of equationsthat can be solved iteratively on a computer.Both finite differenceand finite elementmethodsare applicable. The finite differenceapproachdescribed in this chapterreplacesthe governing Thesecanbe solvedon partial differential equations with a setof algebraicequations. at a finite numberof locations the computerto producea set of water table elevations in the aquifer. Once the groundwatermodel has been calibrated,it can be usedto predict the prostrategies and/or management outcomes(impacts) of alternativedevelopment posed for an aquifer. Such analyses are valuable adjuncts to decision-makingproModels can, for example,simulate the effects of opening new well fields, cesses. analyzechangedbperating practices for existing well fields, explore schemesfor plans. artificial recharge,and predict the impactsof proposedirrigation development Groundwatermodelscan be applied to unconfinedaquifers,semiconfinedaquifers, large variaconfined aquifers,or any combinationthereof. They can accommodate coefficient,and tions in aquiferparameters suchashydraulicconductivityand storage they can be usedto analyzeunsteadyas well as steadyflow problems.

REFERENCES 503

PROBLEMS
20.t. Refer to Fig. 20.8. Assumethat the elementlength is 5 m and the thicknessof the
confined aquifer is 2.5 m. The head at the left and right sidesis 8.1 m at r : 0, and the head on the right is 2.5 m for all r > 0. K : 0.5 m/day and S : 0.02. Use the explicit method to determineheadsat future times. 20.2. Referto Fig. 20.8.Assumethe elementlengthis 10 ft andthe thicknessof the confined aquiferis 8 ft. The headat the left and right is 21 ft at t : 0, and it drops on the right sideto 8 ft for all t > 0. K: 1.5 ftlday-ands : 0.02. use the explicitmethodto calculatefuture heads. from studyingthis 20.3. Referto Fig. 20.12.Asidefrom the trend, what elsecanyou deduce figure? 20.4. Discusshow you would go about designinga grid for a regional groundwaterstudy. What types of boun{ary conditions might you specify?Why?

REFERENCES
Modeling.Worthington,OH: National Water 1. J. W. Mercer and C. R. Faust,Ground-Water 1981. Well Association, 2. C.A.AppelandJ.D.Bredehoeft,"statusofGroundwaterModelingintheU.S.Geological ey Circular 737(1976). Survey," U.S, Geol. Surv "Utilization of Numerical Groundwa3. Y. Bachmat,B. Andres,D. Holta, and S. Sebastian, ter Models for Water ResourceManagement,"U.S. Environmental Protection Agency ReportEPA-600/8-78-012. "Contribution of Ground-waterModeling to Planning," J. Hydrol' 43(Oct. 4. J.E. Moore,

r979).
5. T. A. Prickett,

"Ground-water Computer Models-State of the Att,'l Ground Water

t2r-r28(r979). r7(2),
6. R. A. Freezeand J. A. Cherry, Groundwater.EnglewoodCliffs, NJ: Prentice-Hall,1979. 7. G. D. Bennett, Introduction to Ground Water Hydraulics, book 3, Applications of HyD.C.: U.S. GeologicalSurvey,U.S. GovernmentPrinting Offlce, draulics. Washington, 1976. 8. D. B. McWhorter and D. K. Sunada, Ground Water Hydrology and Hydraulics. Fott 1977' Publications, Collins,CO: WaterResources Hydrology,2d ed.New York: Wiley' 1980' 9. D. K. Todd,Groundwater "PredictedWater-LevelDeclinesfor Alternative GroundwaterDevelop10. P. W. Huntoon, Rep.No. 6, Conservation Big Blue River Basin,Nebraska,"Resource in the Upper ments and Survey Div., University of Nebraska,Lincoln, 1974. "The Numerical Solution of Parabolic and and H. H. Rachford, Jr., 11. D. W Peacemen Afpl. Math. J.3,28-4I(I955)' Indust. Soc. Elliptic DiffetentialEquations," "Application of th Digital Computer for Aquifer D. Bredehoeft, and J. F. Pinder 12. G. - 1093( 1968). 4(4), 1069 Res. WaterResources Evaluation," Hydrol13. I. Remson,G. M. Hornberger,and F. J. Molz, NumericalMethodsin Subsurface ogy. New York: Witey, 1971. Techniques Digital Computer for Groundwa14. T. A. Prickett and C. G. Lonnquist,Selected Evaluation,Illinois StateWater Survey Bull. No. 55, 1971. ter Resource

PARTFIVE

MODELING HYDROLOGIC

Chapter21

to Hydrologic lntroduction Modeling

r Prologue
The purposeof this chapteris to: . Introduce the types and classes of hydrologicmodels. . Illustrate the limitations, alternatives,steps,general components,and data needsof hydrologicsimulation models. . Presenta philosophicalprotocol for performing successful modelingstudies. . Give an overview of groundwatermodel types. . Distinguishthe need for separate, specificproceduresdetailedi4 subsequent 25. 23, 24, and 22, Chapters ratesandvolumesof flow at any point of interestalonga stream Information regarding in the analysisand designof many types of water projects. Although is necessary many streamshavebeen gaugedto provide continuousrecords of streamflow,planinformastreamflow facedwith little or no available are sometimes nersand engineers flow generate artificial to tools as simulatlon and must rely on synthesis tion and of effects the sizes, structure regarding decisions in rationalizing for use sequences of quality, effects and the water supplies, water control measures, land use, flood changes. climatic watershed or natural or induced making in both the designand operationof large-scale The problemsof decision have and water supplysystems canals,aqueducts, reservoirs, of flood control systems to and synthesis simulation such as approaches mathematical resultedin a need for of description mathematical as the defined project. is Simulation the total investigate the responseof a'hydrologic water resourcesystemto a seriesof eventsduring a time period.For example,simulationcanmeancalculatingdaily, monthly, or selected resulthydrograph streamflowbasedon rainfall; or computingthe discharge seasonal in a missing values in the fllling or simply storm; or hypothetical ing from a known streamflowrecord. Simulation is commonly used in generating streamflow hydrographsfrom rainfall and drainagebasin data. The philosophiesand overall concepts used in simulation are introduced in this chapter. Chapter 22 summarizesconcepts of

508

21 rNTRoDUcroN To HyDRoLocrc cHAprER MoDELING streamdowsynthesis by stochasticmethods.Chapters23-25provide detailsregarding determinislic continuousmodels, single-eventmodels,urban runoff and storm sewerdesignmodels,and water quality models. Stochastic techniques usedto extendrecords,either rainfall or streamflow,are classifiedas synthesis methods.This procedurerelies on the statisticalpropertiesof an existingrecord or regionalestimates An overviewof synthesis of theseparameters. techniques is presented in Chapter22.

21.1 HYDROLOGIC SIMULATION

In this chapter, simulation of all or parts of a surface,groundwater,or combined system implies the use of computersto imitate historical eventsor predict the future response of the physicalsystem to a specificplan or action. Physical,analog,hybrid, and or other modelsfor simulatingthe behaviorof hydraulic and hydrologicsystems imitating systemcomponentshave had, and will continue to have, application in prototype behSviorbut are not discussed here. A few of the numerous event,continuous, andurbanruneff computermodelsfor simulatingthe hydrologiccycle are comparedin Table 21.1. As shown in the tabfe, All have mostof the modelsweredeveloped or federalagencies. for, or by, universities percent from 1 10 moderate-to-extensive and all have to of input data requirements,
TABLE 21.1 DIGITALSIMULATION PROCESSES MODELSOF HYDROLOGIC

Code name

Model name

Percentage of inputsby Date of original Agency or organization judgmenf development

Private ARS Stanford University EPA Corps USGS USDA 24
I 5
I

23 Continuous simulation models-Chapter streamflow

API USDAHL SWM-IV HSPF NWSRFS SSARR PRMS SWRRB HEC-1 TR-20 USGS HYMO SWMM UCUR STORM MITCAT SWMM ILLUDAS DR3M PSURM Antecedent Frecipitation Index Model 1970,1973,1974 Revised Watershed Hydrology Stanford WatershedModel IV Hydrocomp Simulation Program-FORTRAN National Weather Service Runoff Forecast System Streamflow Synthesisand Reservoir Regulation Precipitation-Runoff Modeling System Simulator for Water Resourcesin Rural Basins

I 10 10 10 3 5 10

1969 t970 1959 196l t972 1958 1982

r990 1973 I 965 t972 r972 t971. t972 r974 t970 I9'71 1972 1978 1979

Rainfall-runoff event-simulation models-Chapter HEC-I Flood HydrographPackage Corps Computer Program for Project Hydrology scs USGS Rainfall-Runoff Model USGS Hydrologic Model Computer Language ARS Storm Water ManasementModel EPA

10 1 5 z 3 5 5 I 5 5

runoff models-Chapter 25 Urban simulation

University of Cincinnati Urban Runoff Model Quantity and Quality ofUrban Runoff MIT Catchment Model Storm Water ManagementModel Illinois Urban Drainage Area Simulator Distributed Routing Rainfall-Runoff Model PennsylvaniaState Urban RunoffModel University of Cinci4natr Corps MIT EPA Illinois State Survey USGS PennsylvaniaState University

"Judgment percentagesare from U.S. Army WaterwaysExperiment Station.r

SIMULATION509 21.1 HYDROLOGIC trials Theseare normally validatedby repeated inputs that arejudginent pafameters. but models simulation event primarily are with the models.The urban runoff models deferred are models urban of descriptions the havebeenisolatedin Table2 1. 1 because 25. to ChaPter streamflowsimulationmodelsshown Severalof the major eventand continuous StanfordandHEC-1 models and24.The 23 in Table2!.1aredescribedin chapters in Table 2I't are briefly listed models most For further referince, are emphasized. "Models and Methpublication in the models, alongwith about 100 other described, Fleming's text plesentscomplete ods Applicable to Corps of EngineersStudi-es."1 and other models'2 USDAHL' descriptionsof the SSARR, S'[iM, HSP,

Models of Simulation Classification

the scienceof ct In recent decades has Passed water resourcesystems procedure.The variedni engineering has causeda proliferation of catego are Presented. classiflcations physical vs. Mathematical Models Physicalmodelsinclude analogtechnologies models'In contrast' mathematical and principlesof similitude appliedto smali-scale the system'A laboratory flume represent to models,"iy on mathematicafJtut"-.nt, unit hydrographtheory of the while rhay be a 1:10 physical model of a stream, to variouseffective watershed a of response modelof the Chapter12 is a mathematical rain hYetograPhs. by consideris achieved classification Continuous vs. Discrete Models A second processes the because continuous as models ing physical,analog,and some digital occur and are modeledcontinuous of slicing necessityand advantages qualifY as discrete models' A we indication method for routing a flo rz reservoirdischarge instantaneous time. over time and timethat involve changes Dynamic vs. static Models Processes modelsthat contrast' In models' dynamic by varying interactionscan be simrrlated hydrologic Few static' called irequently ui" examine time-independentprocer*"* simulation modelsfall into the latter category' havehad the greatest Descriptive vs. conceptual Models Descriptivemodels they are because appticaiionand are of particular interestto practicinghydrologists through empiricisrnand the useof basic phenomena to accountfor observed designed concepconservationassumptions' momentum fundamentalssuchas continuity or rather interpletphenomena to theory on heavily tual models,on the other hand, rely on based models include latter the of Examples than torepresentthe physicalpto""tt.

'

510

CHAPTER21

INTRODUCTION TO HYDROLOGIC MODELING

probabiliiy theory.Recenttrendsin the useof artificial intelligence and expertsystems in water system modelingwould classifyas conceptual methods. Lumped vs. Distributed Parameter Models Modelsthat ignore spatialvariations in parameters throughoutan entire systemare classifiedas lumpedparameter models.An exampleis the use of a unit hydrographfor predicting time di;tributions of surface runoff for different stormsover a homogeneous drainagearea.The "lumped parameter" is the X-hour unit hydrographusedfor convolutionwith rain to givelhe storm hydrograph.The time from end of rain to end of runoff is also a lumped parameteras it is held constantfor all storms.Distributedparametermodelsaccount for behavior variations from point to point throughout the *yst"*. Most modern groundwater simulationmodelsare distributedin that they allow variationsin storage andtransmissivity parameters overa grid or lattice system superimposed overthe plan of an aquifer.More recently,surface water systems are being analyzedthroughuseof distributedparameterGeographical Information System(GIS) technologies. Black-Box vs. Structure-lmitating Models Both of thesemodelsacceptinput and transform it into output. In the former case,the transformationis accomplished by techniques that havelittle or no physicalbasis.The alchemist'spurported ability to transform lead into gold or plants into medicinewas accomplished in a black-box fashion- In hydrology, black-box models may sometimestransform "plants" into "medicine" even though the reasonsfor successare not clearly understood.For example, a modelthat accepts a sequence ofnumbers,reduces eachby 20 percent,and outputsthe resultsmight be entirely adequate for predictingthe attenuationof a flood waveas it travelsthrough a reachof a given stream.In contrast,a structure-imitating modelwould be designed to useaccepted principlesof fluid mechanics and hydraulics to facilitate the transformation. Stochastic vs. Deterministic Models Many stochastic processes are approximated by deterministic approachesif they exclude all considerationof random parametersor inputs. For example,the simulation of a reservoir systemoperating policy for water supply would properly include considerationsof unceitainties in natural inflows, yet many water supply systems are designedon a deterministic basisby masscurve analyses, which assume that sequences of historical inflows are repetitive. Deterministic methods of modeling hydrologic behavior of a watershedhave becomepopular. Deterministic simulation describes the behaviorof the hydrologic cycle in terms of ma[hematicalrelations outlining the interactionsof variousphases of the hydrologicclcle. Frequently, the modelsare structuredto simulatea streamflow value,hourly or daily, from given rainfall amountswithin the watershed boundaries. The model is "verified" or "calibrated" by comparingresults of the simulation with existingrecords.Oncethe modelis adjusted to fit the known period of data,additional periods of streamflowcan be generated. Event'Based vs. Contlnuous Models Hydrologicsystems can be investigated in greaterdetail if the time frame of simulationis shortened. Many short-termhydrologic modelscould be classifiedas event-simulation modelsas contrastedwith seauential

'

' 21.1 HYDROLOGIC SIMULATION

511

or coniinuousmodels.An exampleof the former is the Corps of Engineers-singlemodel eventmodel, HEC-1,3 and an exampleof the latter is the Stanfordwatershed three' simulate to operated normally is which by Crawford and Linsley,a developed use might model simulation event A typical four, five, oi *or" yearsof streamflow. min' I perhaps even a time incrementof t hr or have arisen water Budget vs. Predictive Models sevelal model classiflcations comparison important One of the model types. that distingtish betweenthe purposes precipifuture conditionsusing synthesized predict proposes to model the is whether tation and watershedconditions or model is definedas a model or set of in of inflows, outflows,and changes studier model simulation that advised that affirm the b use the Parameters meteorologicdat example, shed.For might be known for a given agricultural watershed'A water amounts application (ET) formula the correctevapotranspiration budgetmodelwould be uied to determine equation continuity in the parametersby testing a range of valuesuntil a balance month-byor day-by-day occursfor all time increments.This is often performedon a budget model' month basis. once the ET parametersare derived from the water or farming conditions' predictive simulationsof diffirent crop patterns,meteorologic the model in relationships that the practicescould be performedwith the satisfaction corroboratehistorical water budget (precipitation outputsare measured dt simultaneous the require studies infil ET, as such ondarY Processes spatial distribution of water applications'

of Simulation Limitations
systems'some Becausesimulation entails a mathematicalabstractionof real-world to which the extent The can^occur. behavior system degreeof *i,,"p,",entation of "depends a developed of test The factors' many on model and systemoutputs vary is consisbehavior the that demonstrating by of u"iifi"uiion simulationmodelconsists system' physical the tent with the known behaviorof resources Even verified simulation models have limitations in usesfor water of assessments performance allow will models planhing and analysis.Simulation options,particularly optibut cannotbe usedefficiently to generate specificschemes plan is formulated^bysome near-optimal a once mal plans, for stated objectives. effectivefor testing normally are runs simulation of a limited number othertechnique, using ranvariables decision of combinations and improving the plan by modifying plans are optimal generating for Techniques dom or systeriaticsamplingtechniques. in Section21'3' described proceAnother limitation of simulationmodelsinvolveschangingthe operating Programming modeled' being system the of duresfor potentialor existingcomponents for example,requires u "o*pot", to handle reservoir storageand releaseprocesses' is rereprogramming considerable and rules, large portions to define the operatin! qoir"a if other operatingproceduresare to be investigated.

t-

512

oHAPTER 21 INTRoDUoTIoN To HYDRoLoGIc MoDELING A'fourth limitation of simulationmodelsis the potential overreliance on sophisticated output when hydrologicand economicinputs are inadequate. The techniques of operational hydrology can be used to obviate data inadequacies, but these also require input. Controversyover the use of syntheticdata centerson the questionof whetheroperationalhydrologyprovidesbetter information than that containedin the input.

Utilityof Simulation
Computersimulation of hydrologicprocesses has severalimportant advantages that shouldbe recognizedwheneverconsidering the merits of a simulation approachto a problem that has other possiblesolutions.One alternativeto digital simulationis to build and operateeither the prototype systemor a physically scaledversion.Simulation by physical modeling has been applied successfullyto the analysisof many components of systems suchas the designof hydraulic structuresor the investigation of streambank stability.However,for the analysis of complexwaterresourcesystems comprisedof many interactingcomponents, computersimulation often provesto be the only feasibletool. Another alternative to digital simulation is a hand solution of the governing equations.Simulation models,once formulated, can accomplishidentical results in lesstime. Also, solutionsthat would be impossibleto achieve by hand are frequently achieved by simulation.In addition, the systemcan be nondestructively tested;prdposedmodificationsof the designs of systemelementscan be testedfor feasibility or comparedwith alternatives; andmanyproposals canbe studiedin a shorttime period. An often overlookedadvantageof simulation includes the insight gained by gathering,organizing,and processing the data, and by mentally and mathematically formulating the model algorithmsthat reproducebehaviorpatternsin the prototype.

Stepsin DigitalSimulation
A simulationmodel is a set of equationsand algorithms(e.g.,operatingpolicies for reservoirs)that describethe real systemand imitate the behaviorof the system.A fundamentalfirst stepin organizinga simulationmodelinvolvesa detailedanalysis of all existingand proposedcomponentsof the systemand the collection of pertinent data. This stepis called the systemidentification or inventory phase.Includeditems of interestare site locations,reservoircharacteristics. rainfall and streamflowhistories, water and power demands, and so forth. Typical inventory items requiredfor a simulatiol study and data needsthat are specificto someof the modelsare detailed in subsequent paragraphs. The second-phase is model conceptualiTation, which often providesfeedbackto the first phasebf defining actual data requirementsfor the planner and identifying system components that areimportant to the behaviorof the system. This stepinvolves (1) selectinga techniqueor techniques that are to be used to representthe system (2) formulating the comprehensive elements, mathematics of the techniques, and (3) translatingthe proposedformulation into a working computerprogramthat interconnects all the subsystems and algorithms. Following the systemidentification and conceptualizationphasesare several stepsof the implementationphase. Theseinclude(1) validatingthe model,preferably

SIMULATION 513 21.1 HYDROLOGIC for the by demonstrating that the model reproduces any available observed behavior the improve to necessary as algorithms (2) the modifying actual or a similir system; simulathe out carrying by work to model (3) putting the accuracy of the model; and tion exPeriments.

Model Protocoi
from recommendations adapted modelstudies, Five axiomsfor performingsuccessful by Friedrich,5are: 1. Evaluatethe data beforebeginning. t Document assumPtions. of computerruns' 3. Plan and control the sequence of output. 4. Insist on reasonableness 5 . Document,document,document. An annotated,bibliographic Examining and evaluatingthe basic data are essential. program \ record of the data ,orrr."* shouldbe maintained.It is alwaysgood adviceto numerical of the modelsthat output (echo)datavaluesasthey arereadin. Verification from the echo. valuesand proper entry of the data can be established Statisticssuchasih" 1n"un,mode,median,range,standarddeviation,skewness, for kurtosis, and rank order are often helpful in locating entry errors' Checking Are the rainfall? of ahead can identify errors. Didthe runoff occur inconsistencies characters Do alphabetical waterlevelsgraduallyvaried,or are there discontinuities? or missing as be interpreted appearin th! data?Will blank valuesin the data sets hyFor zeto)? (division by zeros?Will zeros in the data sets result in overflows fall betweenthe limits recommended drographrouting, doesthe time interval selected of the numerical method usedto solvethe differential for staUltityand convergence equations? ^ of a simulation' Assumptions Assumptionsare also important to the success and additional assumpthe mod91, were madeby the p.og.u--"iwhen developing deviation the standard tions aremadeby userJ.Fo, "*u-ple, a programthat calculates must be size sample the from an unbiased estimating equation assumesthat = considered often is 30 N sufficiently lafge to validate ihe estimate.A value of used' as minimal. For a TP-149 (Chapter 15) application, is a 24-hr storm being and reading prior to used be by the method?No computerprogram should assumed of the aware becoming and madeUy ttre programmer undershndingthe assumptions programmed' implicit in the hydrologicprocessthat was assumptions 'ih" lor cpst of simulationcan reiult in unnecessary runs and may enticeusers information originally the bstantively to working the plan can and purposeofruns)

r alproximate time and in a single useasa guideduringa simulationproject. combining severalinvestigations available models of the Some run is anotherway to conductarrefficient simulation. allowthis.Forexample,TR-20(seeChapter24)allowsthegenerationofflood

asmarr isonrY dtli|l,""i:#THfftime monetary limits to

514

21 CHAPTER

MODELING TO HYDROLOGIC INTRODUCTION

'

to generatethe2-,5-, from severalstormsat once.It is often desirable hydrographs location. a single watershed at flood discharge 10-, 25-, 50-,100-, and 500-year can hydrologist than the output generate far more is able to The computer quantity. output to specify the user allowing options incorporate Most models analyze. shouldbe madeof which specific In additionto controlling outpul, a predetermination data can be developedto key output of performed. A tabulation analyseswill be (and after eachrun. Because corrections) coursg make trends compile and evaluate opportunitiesexist in many percent acbounting, is 80 hydrology about deterministic to an aquifer is recharge If the total balances. water budgel simulation for assessing are rising, a water tables but simulated and withdrawls, less than the total outflow conclusions important Writing made. parameters should be checkof input and model guide revisions and the study helps document runs on the printed output of simulation in future runs. Documentation of simulation studies is generally deficient in practice. The record should communicatethe findings in a way that provides a later reviewer for each made,andreasons generalunderstanding of the work plan followed,decisions input of the provide samples made, state assumptions should run. The documentation are results the how sensitive state preparation requirements, and output,explaininput parameout-of-range reasons document and and assumptions, to parameterchanges ters were accepted. Documentationis an ongoingand continual task. It is especiallycrucial if the model will be employedin regulatory proceduresor litigation. A comprehensive process would6: documentation Include an outline descriptionofthe problem being studied. and methodsused. Identify the equations,techniques, Demonstratethe model's validity to this problem. Discuss the code. usedin the code and in preparingthe input. Include all assumptions List publishedor known limitations or rangesof the applicability of the model. the uncertaintiesin the model; describesensitivitytests. 7. Characterize and data setsused. 8. Describeparameters 9. Statethe regulatory or legal criteria incorporatedin the model. 10. Describe the verification, whether with test data or analytical solutions. or 11. Include a narrative descriptionof the results,indicating any unexpected unusualoutcomes. 12.Presentany other details deemedrelevant. 13.Discusithe modelused. 14.Documentchansesmade in the model code. 1. 2. 3. 4. 5. 6.

Models Simulation of Hydrologic Components

for the purposeof simulating Numerousmathematicalmodelshavebeen developed general conceptualmodel including A phenomena systems. and various hydrologic othersare described 21.1; several Fig. in is shown components mostof the important or groundreservoir input to be leJt could the Irnported water in subsequeht$. lorygl

SIMULATION 21,1 HYDROLOGIC

515

System outflow

Snow accumulation and melt Depression storage Overiand flow direct runoff

System outflow

SYsteminflow

System rntlow

Figure2l..l.Componentsofasurfaceandsubsurfacewaterresourcesystem. allocationson water storageor channelflow, or it might be guideddirectly to water The routing of unnecessary. or distribuiion were deemed the far rlgtriif either storage by simple lumped parameter channelflo* o, overlandflow could be accompdbhed of for discretesegments flow eq-uations or solutionsof the unsteady-state techniques, algorithms and the channelcould be used.In other words,the selectionof techniques as output desired refinement of d9ere9 the on depends to fepresenteachcomponent justifled is and also on knowledgl of the system.A distributedparameterapproach in described of modelsare Components information is adequate. only when available 22-25. ChaPters

Simulation DataNeedsfor Hydrologic

requiresa data inventoryaspart of the The simulation6f all or part of a water system (90 percentor mofe) are initial planningproc"ts. Most modelinput datarequirements from engineering obtained or or canbe empiricallydetermined map or tield ariaitable, encompasses that topics inventory handbooksand equations.A generallist of data - economicmodelingneedsfollows' most hydrologic Characteristics A. Basin and Subbasin times of concentration' l, Lagtimes, travel times in reaches, 2.Contributingareas,depressions,meanoverlandflowdistancesandslopes' \--

516

CHAPTER 21

INTRODUCTION TO HYDROLOGIC MODELING

3. Designstorm abstractions: evapotranspiration, infiltration, depression, and interception losses.composite curve numbers,infiltr_ationcapacitiesand parameters, @indexes. 4. Land-usepractices,soil types, surfaceand subsurface divides. 5. Water-usesites for recreation,irrigation, flood damagereduction, diversions,flow augmentation, and pumping. 6. Numbering systemfor junctions, subareas, gaugingand precipitation statrons. 7. Imprevious areas, forested areas, areas between isochrones, irrigable acreages. B. ChannelCharacteristics 1". Channelbed and valley floor profiles and slopes. 2. Manning or Ch6zycoefficients for variousreaches, or hydrhulicor field data from which thesecoefficientscould be estimated. 3. Channeland valley cross-sectional data for eachriver reach. 4. Seepage information; channellossesand baseflows. 5. channel and overbank storagecharacteristics, existingor proposedchannelization and leveedata. 6. Sedimentloads,bank stability, and vegetative growth. c. MeteorologicData 1. Hourly and daily precipitation for gauges in or near the watershed. 2. Temperature, relative humidity, and solar radiation data. 3. Data on wind speedand direction. 4. Evaporationpan data. D. WaterUse Data 1. Flows returned to streamsfrom treatmentplants or industries. 2. Diversionsfrom streamsand reservoirs. 3. Transbasin diversionsfrom and to the basin. 4. Stream and ditch geometricpropertiesand seepage characteristics. 5. Irrigated acreages and irrigation practices,including water useefficiencies. 6. Crop types and water consumptionrequirements. 7. Pastconservationpracticessuchas terracing,insfallation of irrigation return pits, and conservationtillage. 8. Stock wateringpractices. 9. Presenceand types of phreatophytes in stream valleys and along ditch banks. E. StreamflowData 1. Hourly, daily, monthly, annual streamflow data at all gauging stations, includin$ statisticalanalyses. 2. Flood frequencydata ani curvesat gaugingstations,or regionalcurvesfor ungauged sites,preferablyon an annual and seasonal basis. Flow duration data and curves at gauging stations (also any synthesized _3. data for ungauged areas). 4. Rating curves; stage-discharge, velocity-discharge, depth-discharge curvesfor certain reaches.

SIMULATION517 21.1 HYDROLOGIC $, Flooded area curves. 6. Stageversusarea flooded' curves' frequencY 7. Stageversus basis. 8. Stageversusflooi damagecurves,preferablyon a seasonal curves' 9. Hydraulic radius versusdischarge sitesas fractions of gaugedvalues. 10. Sireamflowsat ungauged ll. Returnflows as fractionsof water-useallocationsdivertedfor consumptive use. distributionsof allocationsto users' 12. Seasonal 13. Minimal streamflowto be maintainedat eachsite' 14.Masscurvesandstorage_yieldanalysesatgaugedsites' F. Design Floods and Flood Routing temporaland spatialdistributionand l. Designstormand flood determination; mtenslty. 2. Maximum regional stormsand floods' to be usedand necesverificationof flood routing techniques 3. Selection.and sary routrng parameters. during designfloods' 4. Baseflow estimates 5. Availablerecords of historic floods' G. ReservoirInformation L. List of potential sitesand location data' 2. Elevation-storagecurves. 3. Elevation-area curves. 4. Normal, minimal, other pool levels' loss data or estimates' 5. Evaporationand seepage 6. Sediment,dead storagerequirements' 7. Reservoireconomiclife. 8. Flood control operatingpolicies and rule curves' weir and outlet equations,controls. 9. Outflow characteristics, recreationbenefit functions' 10. Reservoir-based L1. Costsversusreservoirstoragecapacities' and benefits. of eachreservoirand beneficiaries 12. Purposes

Assessments Nonmodeling
the availabledata and information, the needfor simulationcan be After researching If a decisionis made to proceed,the appropriatesimulation model can be assessed. planned,and data prepared' a sequence selected, Transformationof raw data into usabieform doesnot alwaysrequire a simulacan assessments for waterresources tion model.Much of the usualinformationneeded microcomputer in be preparedby hand or by using analytical proceduresavailable format.Typicalnonmodelinganalysesincludethefollowing: 1. Identify water-user groups and all basin sites for hydropower producirrigation, flood damagereduction from tion, reservoir-based-recreation,

518

CHAPTER 21

INTRODUCTION TO HYDRoLOGIC MODELING

reservoircapacity,industrial and municipal water supply,diversions,and flow augmentation. t Compile annualand seasonal and flood recordsat eachgauged streamflows site for the period of record at eachsite. 3. Determine the fraction of the allocation to each consumptiveuse that is assumed to return to the streamat eachuser site in the basin. of annual streamflowand flood valuesat each 4. Perform frequencyanalyses gaugesite in the basin. losses. and seepage Determinefor eachreservoirsite the eVaporation yield probabilities secondary analyin the firm and mean to be used 6. Select
SES.

7. Developflood peak probability distributionsat eachpotential flood damage center or reachin the basin. 8. Determinethe fraction of water to be allocatedduring eachperiod to each water-usesite in the basin. 9. Determine existing and proposed hydropower plant capacitiesand load factors. to be maintainedfor flow aug10. Identify any minimal allowablestreamflows in the basin. reach mentation at eachflow augmentation on any of the annual or seaconstraints 11. Specify any maximal or minimal yields. target capacities, or sonal water allocations,storage dead storage,active storor minimal 12. Specify any constraintson maximal capacities at any or all of the total storage age, flood control storage,or reservoirsitesin the basin. 13. Determine annual capital, operation, maintenance, and replacement (OMR) costsat eachreservoirsite as functions of a rangeof total reservoir 'capacities or scalesof development. 14. Determinebenefitsas functions of energyproduced. L5. Determineannual capital and OMR costsat eachhydropowerproduction site as functions of variousplant capacities. 16. Determine benefit-loss functions for a variety of allocationsto domestic, commercial,industrial, and diversionuses. 17. Determine short-run lossesas functions of deviations (both deficit and surplus)in plannedor target allocationsto user sites. 18. Developbenefit functions at eachirrigation site in the basin.This analysis requiresinformation on the area of land that can be irrigated per unit of water allocated,the quantitiesof eachcrop that can be producedper unit areaof land, the total fixed and variablecostsof producingeachcrop, and the unit"pricesthat will clear the market of any quantity of eachcrop. 19. Developflood-damage-reduction beneflt functions at eachpotential flood records of historical and/or simulated site. This analysis requires damage floods, channel storagecapacities,and flood control reservoir operating policies. 20. Developreservoir-based recreationbenefitfunctions at eachrecreationsite in the basin.

SIMULATION 21.2 GROUNDWATER

519

21.2 GROUNDWATER SIMULATION Digital simulation models are used in a different manner to study the storageand movementof water in a porous medium. Distributedrather than lumped parameter models are used to imitate observedevents and to evaluatefuture trends in the the The equationsdescribing of groundwatersystems. and management development regional of modeling and 18 Chapter in derived were poroui medium flow of waterin a in Chapter 20. This section deals primarily with techniques was discirssed systems used in solving the hydrodynamicequationsof motion and continuity, followed by of calibrating and (2) techniques of (i) typical input requirements, brief discussions to parameter models groundwater of (3) sensitivity the verifying the models, and modelis also groundwater a of application and of the calibration changeslAnexample provided.

ModelTypes
Groundwaterstudiesinvolve the adaptationof a particular code to the problem at hand. Severalpopular public domain computer codesfor solving various types of becomemodelswhen flow problemsare listedin Table2L2.The codes groundwater geometryand system the inputting by code to the beingsiudiedis described the system boundary and initial parameters, field flow (aquifer and known internai operandi parts of or to all in time applied stresses flow conditions, and water use and groundwater general categories: flow four in the system).Codes have emerged analysis data test aquifer and codes, tracking particle codes,solutetransport codes, programs.to Groundwaterflow codesprovide the user with the distribution of headsin an aquifer that would result from a simulated set of distributed recharge-discharge any two points From Darcy's law, the flow passing at cells or line segments. stresses model both to used are codes The differential. can be calculated from the head flow, or regional model to be structured can Each confined and unconfinedaquifers. conditions transient and Steady-state wellfield. or well flow in proximity of a singG and Boundaiiescanbe barriers,full or partially penetratingstreams canbe evaluated. By application perimeters. gradient constant or head lakes,leaky zones,or constant after solving can be determined velocitiesof groundwater the seepage of Darcy's 1aw, for the head differentials. and velocitiesareknown, the advection,dispersion, seepage When groundwater on build models transport Solute modeled. be can in concentrationof iolutes changes chemical and/or dispersion, advection, of addition groundwaterflow modelsby the due or dilution concentrationchanges If the chemical,dispersion, reactionequations. by transport model codes particle tracking m groundwaterflow are not important, path the track to models transport solute than method advectionand providean easier that moveunderthe influenceof headdifferentials.Aquifer and traveltimes of solutes test data programsprovide userswith computersolutionsto many of the hand calculations (Ctrapter li) neededto graph and interpret aquifer test data for determining aquifer and well Parameters'
I

\.

520

CHAPTER 21

INTRODUCTION TO HYDROLOGIC MODELING

TABLE 21.2 GROUNDWATER MODELING CODES

Acronym for code PLASM MODFLOW AQUIFEM-1 GWFLOW GWSIM-II GWFL3D MODRET SUTRA RANDOMWALK MT3D AT123D MOC HST3D FLOWPATH PATH3D MODPATH WHPA TECTYPE PUMPTEST THCVFIT TGUESS

Description
Groundwater flow models Two-dimensionalfi nite difference Three-dimensionalflnite difference T!wo-and three-dimensionalfinite element Package of1 analytical solutions Storageand movementmodel' Three-dimensionalfinite difference from retention ponds Seepage Solutetransooftmodels Dissolvedsubstance transport model Two-dimensionaltransientmodel Three-dimensionalsolutetransport Analytical solution package Two-dimensionalsolutetransport 3-D heat and solutetransport model Particletrackingmodels Two-dimensionalsteadystate Three-dimensional transient solutions Three-dimensional transient solutions Analyticai solution package Aquifertest analyses Pump and slug test by curve matching Pumping and slug test Pumping and slug test c capacitydetermination Specifi

Source

ru.sws
USGS MIT IGWMC TDWR TDWR USGS USGS ill. SWS EPA DOE USGS USGS SSG Wisc GS USGS EPA

1971 1988 r979 1975 198 1 r991 1992 1980 1981 1990 1981 t978 r992 1990 1989 t991 1990 1988 1980 1989 1990

ssG
IGWMC IGWMC IGWMC

Note: IGWMC : International Groundwater Modeling Center; Ili. SWS : Illinois State Water Survey; SSG : Scientific Software Group; EPA : Environmental Protection Agency; USGS : U.S. Geological Survey; Wisc. GS = Wisconsin Geological Survey; MIT : Massachusetts Institute of Technology; TDWR : Texas Department of Water Resources;DOE : Department of Energy.

SolutionTechniques
With few exceptions,the hydrodynamic equations for groundwaterflow have no analyticalsolutions, and groundwater modelingreliesonfinite-dffirence and,finiteelementmethodsto provide approximatesolutionsto a wide variety of groundwater problems.The choice of method is normally driven by the systemto be modeled. Other numericalmethodsincludeboundary integral methods,integratedfinitedffirence methods,and analytic elementmethods. Thesesolutions,as with streamflowsimulation models,are facilitated by first subdividingthe region to be modeledinto subareas. Groundwatersystemsubdivision depends more on geometriccriteria and lesson topographiccriteria in the sense that the region is overlaidby a regular or semiregular pattem of node points at which (or betweenwhich) specificmeasures of aquifer and water systemparameters are input and other parametersare calculated.Approximate solutionsof simultaneous linear and nonlinearequationsare found by making initial estimates of the solution values, testingthe estimates in the equationsof motion and continuity, adjustingthe values, andfinally accepting minor violationsin the basicprinciplesor making further adjustments of-the parameters in an orderly and convergingfashion.

SIMULATION 21.2 GROUNDWATEFT

521

r '

or unsteadyof the steady-state Theorderly solutionof finite differenceanalogs in a confined groundwater of flow for motion of equation partial difierential stare relaxAn eady methods' by rela_xation is obtained aquifer unconfined aquifer or an problems, For two-dimensional by Jacob.T is discussed equation of the ation solution and by Peaceman (ADI) methoddeveloped the iterativealternating-direction-implicit adoPted. is often Rachfords prickett and Lonnquisteused the ADI techniqueto calculate fluctuations in at all nodesin an aquifermodelby proceedingthrough time in watertable elevations a known initial state.Their modelis computationallyefficient from small increments is particularly attractivefor usewith problemsinvolvingtime and applied and readily are the permeability nodes.The primary aquiferparameters nu-"tous variablesind plan, result in a aquifer the over constant if assumed which, and storagecoefficient, methods'the relaxation with familiar those For condition. isotropig and homogenJous application had have (SOR) methods over-relaxation the successive Gauss-seideland equations. in solvingdifference as spatialand temporal' Spatial modelsmay be classified system Input to groundwater thicknessdata over the saturated maps, table water projected initial oi input inciudes variationsin storage regional maps, transmissivity maps, contour ."gion, land surface types of aquifer and canals, and locations of wells and typei coefficients,locations or net pumpage actual system, coordinate a node vertical, and lateral both boundaries waters,logs applied and other precipitation for rates recharge percolation and rates, patterns. cropping and types soil and geologic stratigraphy, of drilled wells, Time-dependenidatarequirementsfor aquifer models involve principally the using a rangeof time incrementsfor suchvariablesas formulation of ti-" schedules, groundhydrographs, canaland streamflow hyetographs, precipitation pumpingrates, of added timing as the such variables development and rates, waterevapotranspiration only apply can schedule temporal each Because system components. wells or other are also requirements time-dependent the positions, node of particular iubset to a spatial. aquifer modelsrequire reliable estiIn addition to the listed input parameters, phase that actually percolateto the land the in of waters percentages matesof the knowledgeof the physical on based be can estimates These modeled. aquifer being but are most often medium porous through flow in unsaturated involved processes phase of the calibration the during modified judgment that are parameters obtainedas piezometer or in changes the and movement lateral the stated, simulation. Simply rates or (withdrawal stresses node-by-node if the modeled easily are watertable leneli movecomplex the governed by are parameters latter The known. aie rates) recharge ment of water in the unsaturatedsoil zone and by the random precipitation and lies systems usepatternsof the region.The art of modelinggroundwater consumptive parameters' these to evaluate in the ability

DataRequirements

Calibration
involvedin parameter modelcalibrationremovessomeof the guesswork Groundwater knowledgeof available on based parameters, of combinations Several determination. records are which period for a during model in the tested are physical syqtem, the

I L,

522

21 INTRODUCTIoN CHAPTER To HYDRoLoGIc MoDELING availabi-e. Simulatedresultsare then comparedwith historical events.After structuring the model,calibrationis achieved by operatingthe model during the studyperiod by imposinghistoricalprecipitationamounts, evaporation and evapcanaldiversions, otranspirationrates, streamflowsand stream levels, pumping rates during known periods, and other stresses on the aquifer. Calibration is achievedafter the flow, storage, and other parameters lirnits to produce havebeenadjustedwithin reasonable the best imitation of recordedevents.

CaseExample
modelingin A typical finite-differencestudyinvolvingsurface waterand groundwater central Nebraskawas performed by Marlette and Lewis.tt The region involved is shownin Fig. 2I.2.In additionto the surface irrigation system represented by the severalcanalsand laterals,over 1200wells withdraw waterfrom the aquiferbetween the PlatteRiver and the Gothenburg and DawsonCounty canals.The aquiferrecharge and withdrawal amountsas percentages of precipitation, snowfall, pumped water, deliveredcanal water, evaporation,and evapotranspiration were estimatedusing a mix of judgment andphysicalprocessevaluations. The resultingsetthat producedthe wells showninFig.2l.2 bestcomparison with recorded events at the six observation is summarized in Table 21.3. Samplesof the comparison betweenrecordedand simulatedwaterlevelsin the DawsonCounty study during aZ-yearcalibrationperiod are shownin Figs. 21.3 and2l.4. The Prickettand Lonnquistmodelwasappliedin the DawsonCounty study.The by calibrationtrials as storage coefflcientfor this unconfinedaquiferwas established 0.25 and the adopted permeability was 61 mlday. Other trials were made using variouscombinations of S andK, with S rangingfrom 0.10to 0.30andwith Kranging between4I and I02 mlday. As with most unconfined aquifer models,water table elevations were most sensitive to fluctuationsin the storasecoefflcient.Fisure 21.5 is

tt'iorX
\F 'q;

Figure 21.2. Grid coordinatesfor Dawson County, Nebraska,aquifer model. o : observationwell.

SIMULATION 21.2 GROUNDWATER ALLOCATIONS FORWATER CRITERIA RECHARGE TABLE 21.3 ADOPTED .. AQUIFER THEDAWSON COUNTY OVER

523

component System
Rainfall Snowfall Pumpedwater Delivered canal water

and appliedamounts Allocation Recordeddepth if daily amount exceeded 0.25 cm at all nodes 25Voofrecorded depthsat all nodes Averagerate of 50 l/sec at-all well nodes during irrigation seasons Recordeddaily rates,applied to land surfaceone node laterally uphill and two nodesdownhill from canal Observeddaily lake evaporationdepth at all marsh and water surfacenodes I25Voof daily lake evaporation,applied at all alfalfa nodes

recharge/withdrawal Aquifer ot as a percentage amount applied 30 30 50 30 100 15

Evaporation Evapotranspiration

o o
6

B
d

0.999

0.998
A M J J A S O N D J F M A M J J A S O N

1970

1971

at observation waterlevels andrecorded Figure2L.3. Simulated w e l l Di n F i g . 2 1 . 2 .:I8 2 ; i : 3 7 . a typical summary of the calibration results at a single observationwell located at F in Fig. 21.2. Position After vefification, the Dawson County model was applied to investigatethe Included among the schemes schemes. short-terminfluenceof severalmanagement involving the completeremoval or shutdownof the surfacewater were investigations canals,and other testsin which isolatedcanal contributionsto rechargewere determined by operatingthe model with singlecanals and comparingresults with water table fluctuationsfor identicalconditionswith all canalsremoved.Many other applicationsof the model are possible.This particular study revealedthat rechargefrom

524

CHAPTER21

MODELING TO HYDROLOGIC INTRODUCTION

o a
d

0.999

0.998 A M J J A S O N D J F M A M J J A S O N 1970 l97l

Figure 21.4. Simulated and recorded water levels at observation well F in Fig. 21.2. I : 97; i : 42.

lu3

P (J

o
q) 'F1

(.1

H 0)

i3 -

702

t9'70

r97|

permeability andvarying with constant changes Figure 21.5. Water-level Well4 Fig. 21.2;K : 6Im/day;I : 97;J : 42. storage coefficients. contributesto the waterbalancqof the aquiferbut is not the the existingcanal system from precipitationandfrom the dominantfactor\n the shortrun. The naturalrecharge Platte River accountfor the long-term water table stability in the region.

PROTOCOL SIMULATION 21.3 HYDROLOGIC

processis not new The useof hydrologicsimulationas a tool in the decision-making form. A model is a and more encompassing but is of a different, more sophisticated that permitsthe evaluationand-rmanipof an actualor proposedsystem rgplesentation

PROTOCOL SIMULATION 21.g HYDROLOGIC

525

useof ulation of-manyyearsof prototypebehavior.This is the featurethat makesthe largest, the of even thesetooii so attractiveand hoids suchpotential for the analysis so well It is alsothe prin-ipal featurethat makesthis approach mostcomplexsystems. suitedto water resourcessystemplanning and analysis' models, Apart from the useof cot1entionalhandmethodsand someelementary as however, is changing, judgment. This planninghastraditionally beena practiceof alternaof numbers large of that permit the analysis quantititive tools are developed is not ruled out but is elementof the process, tives and plans.Judgment,a-nessential in the planning to those through new insights that were not available strengthened professiona few years ago. "What if ?" Plannersare conti;ually required to anticipatethe future and ask suchas simulation and "What's best?" questions. Quantitativeplanning techniques, for lesscost than alternatives can provide detailedinformation aboutmore planning principally at occurred has of thesetools uny oth", approachavailable.Development universitiesand federal agencies.

Models and Optimization CombinedUse of Simutation

at this An important secondtype of quantitativeplanningtool shouldbe mentioned a select to information to utilize llmiPd system modelsare designed point.Screening Hence objectives. of set or objective for a specified bestplan u*oni many alternatives modJs, or optimizationmodelsasthey are often called,are orientedtoward screening plan formulation rn contrast to th Simulation models are suited to de reliable information on which to bi "If the question, out modelsaddress tion models,on the otherhand,ask, look like after we arr will the system special merits of each, these two Completeder planningtechnologies. ing, oPtimization,and simulation r Final designvaluesshouldbe and operatinga detailedsimulation model over time using a to the systemel-ements while at and/or streamflows, precipitation-amounts of known or synthesized sequence streamand reservoir control, benefiisovertime for flood the sametime accumulating siderecreation,wateryields,strean and anY other factorsnot consider model. Severalsimulationruns wi resultin a plan that bestmeetsthe c by coilventionalmethods' generated inforResultsof optimization modelswill provide readily obtainedand useful promismost the test to in order mation for initiating more refinedsimulationanalyses and arrive at final plans for the design,construction,and operationof ing measures Even thor a water resourcesystem. regardingboth the de for decisions postoptimizationsimulation is rec often requiredin prel assumptions

526

CHAPTER21

MODELING TO HYDROLOGIC INTRODUCTION

alternativesowing to time and cost limitaof development for preliminary screening currenttime and sizelimitations evolves, of computers tions.Unlessa new generation by simulating all iilternativesunlesssubstantialsacrificesin do not allow screening followedby detailedsimularealismare made.For the present,preliminary screening to be the most effectivemeansfor arriving at optimal waterdevelopment tion appears and management Plans.

MODELS SIMULATION OF ENGINEERS 21.4 CORPS

a specialtybranch locatedat In 1964,the U.S. Army Corps of Engineersdeveloped The facility provides (HEC) California. Davis, in Center the HydrologicEngineering of the Corps fleld needs practical to results research a centerforipplying academic to governassistance technical and provides training offices.In addition, the center operations. reservoir and hydraulics, hydrology, in advanced menr agencies Over the years, a large number of analytical tools were developedat HEC. river/reserof hydrology, the computerprogramsin categories Table 2I.4 summarizes water river/reservoir hydrology, stochastic operations, voir hydraulics, reservoir
TABLE 21.4 HEC WATER RESOURCECOMPUTERPROGRAMS Name Date of latest version HydrologyModels 1980 September

Purpose
Simulatesthe precipitation runoff river process in any comPlex

HEC-I, Flood HydrograPhPackage

Basin Rainfall and SnowmeltComputation

July 1966
many subbasilsof a river basin using gaugedata and weightings (includedin HEC-1).

Unit Graph and HydrographComputation

Iuly 1966

Unit Graph Loss Rate OPtimization

August 1966

HydrographCombining and Routing

August 1966

Computessubbasin interception/infiltration, unit baseflow, and hydrographs runoff hydrograph(included in HEC-1). Estimatesbest-fitvaluesfor unit graph and lossrate Parameters from given precipitation and runoff (included in subbasin HEC-1). at Combinesrunoff from subbasins and routes confluences througha river hydrographs network using hydrologicrouting methods(includedin HEC-1).

MODELS SIMULATION 21.4 CORPSOF ENGINEERS TABLE21.4 (Continued)*

527

Name
Streamflow Routing OptiffIization

Dateof latestversion
July 1966

Purpose
Estimatesbest-fit valuesfor hydrologic streamflow routing parameterswith given upstream' downstream,and local inflow (includedin HEC-l). hydrographs gravitY and Computes seepage, pressue flow, pumping and for Pond overtopping discharges areasbehind leveesor other flow obstructions.Main river elevation and ponding area elevation-area-capacitydata are usedin computingdischarges. Simulatesthe precipitation runoff processfor a single, usuallY urban, basin for manY Yearsof hourly precipitation data. Simulates qualitY of urban runoff and dry weather sewageflow. EvaluatesquantitY and qualilY of overflow for combinations of sewagetreatmenl plant storage and treatment rate.

Interior Drainage Flood Routing

November1978

Storage,Treatment, Overflow, Runoff Model (..STORM")

JuJy1976

hYdraulics River/reservoir
HEC-2, Water SurfaceProfiles August 1979

super-critical. AnalYzes for a allowableencroachment given rise in water surface. Gradually Varied UnsteadyFlow Profiles Iarnary L976 Simulatesone-dimensional, unsteady,free surfaceflows in a branching river network. Natural and artificial cross sectionsmaY be used. Uses an exPlicit centered difference computational scheme. Computestables of hYdraulic elementsfor use bY the GraduallY Varied UnsteadYFlow Profiles or other programs.InterPolates values for area, top width, n value, and hydraulic radius at evenly spacedlocations along a reach.

Geometric Elements from Cross Section ("GEDA') Coordinates

June 1976

534

CHAPTER21

TO HYDROLOGIC MODELING INTRODUCTION

PROBLEMS
in Chapters'22and 23. List the most are treatedseparately 21.1. Simulation and synthesis of eachmethod and give an exampleof each. distinguishingcharacteristics 21.2. Listatleastthreereasonsmanyofthedevelopedmodelsoftherainfall-runoffprocess might not be usedby hydrologists. 21.3. You are askedto determinea designinflow hydrographto a reservoirat a site where I,!st generalstepsyou would take as a hydrolno recordsof streamfloware'available. ' ogist in developingthe entire designinflow hydrograph.

REFERENCES
ExperimentStation,"Models and MethodsApplicableto Corps of 1. U.S. Army Waterways PaperH-74-8, National TechnicalInformation Engineers Urban Studies,"Miscellaneous Service,Aug.1974. in Hydrology. New York: American 2. GeorgeFleming, ComputerSimulation Techniques Elsevier,1975. "HEC-I Flood HydrographPackage,"Users and ProU.S. Army Corps of Engineers, grammers Manuals,HEC Program723-X6-L20I0, Jan.1973. "Digital Simulationin Hydrology:StanfordWater+ . N. H. Crawford and R. K. Linsley,Jr., shedModel IV," Department of Civil Engineering,Stanford University, Stanford, CA, Tech.Rep.No. 39, July 1966. 5 . A. J. Friedrich, "Managementof ComputerUse in SolvingEngineeringProblems,"U.S. Army Corps of Engineers,Hydrologic EngineeringCenter,Davis,CA, 1979. Council, Ground WaterModels-Scientific and RegulatoryApplica6 . National Research Mathetions. Water Scienceand TechnologyBoard, Commissionon PhysicalSciences, D.C., 1990. Washington, National AcademyPress, matics,and Resources, ed.)' New Hydraulics (HunterRouse, 7 . C. E. Jacob,"Flow of Groundwater,"in Engineering York: Wiley, 1950. "The Numerical Solution of Parabolicand Elliptic and N. H. Rachford, 8 . D. W. Peaceman Indust. Appl. Math.3' (1955). J. Soc. DifferentialEquations," "selectedDigital ComputerTechniques for Ground9 . T. A. Prickett and C. G. Lonnquist, Ilinois StateWaterSurveyBull. No. 55,197I. Evaluation," waterResource 1 0 . D. R. Maidment, (ed.), Handbookof Hydrology.New York: McGraw-Hill, 1993. "Digital Simulationof Conjunctive-Useof Groundwater 1 1 . R. R. Marlette and G. L. Lewis, in DawsonCounty, Nebraska,"Civil EngineeringReport, University of Nebraska,Lincoln, 1973. 12. W. K. Johnson,"Use of SystemsAnalysis in Water ResourcePlanning," Proc- ASCE J. Hyd. Div. (1974). Planning and Design CaseStudiesin Modeling 1 3 .R. deNeufvi[e and D. H. Marks, Systems Cliffs, NJ: Prentice-Hall,I914. . Englewood Optimization and EvaluatioiT 14. D. P. Loucks, "stochasticMethodsfor Analyzing River Basin Systems."Cornell Univer', Center,Ithaca, NY, Aug. 1969. and Marine Sciences sity Water Resources MA: HarvardUniversity System*Cambridge, 1 5 . A. Maass,(ed.),Designof WaterResources Press, 1962. t6. A. F. Pabst, "Next Generation HEC Catchment Modeling," Proceedings,ASCE HyHydrology,SanFrancisco,CA, July 25-30' on Engineering draulicsDivision Symposium 1993.

Chapter22

Time":?J5: Hydrologic

r Prologue
The purposeof this chaPteris to: . Show how time seriesanalysisis used for generatingsynthetic hydrologic records. of hydrologic aspects . Give definitionsof termsusedto describe the stochastic series. including masscurve analy. Introduce fundamentalsof streamflowsynthesis and sesequences, serial-dependent sis, random generationof sequences, quences havingprescribedfrequencydistributions' Time-seriesanalysisof hydrologicvariableshas becomea practical methodolof precipitation or steamflowvaluesthat can ogy for generatingsyntheticsequences ui-usedlor u tung" of applicationsfrom filling in missingdata in a gaugedrecord to, extendingmonthiy streamflowrecordsl, and from analyzinglong-term reliability of floodsor snowmeltrunoff quantito forecasting or reservoirst'o yields of-watersheds2 Thesesynthetic hydrologytechniques iies from syntheticprecipitation sequences.5 widein Chapter21. Both have,experienced augmentttre simulationtools described generation of a Synthesisinvolves the spiead use by hydrologistsand engineers.6 or annual). ,"qu"n"" of valuesfor somehydrologicvariable(daily, monthly, seasonal, for use in are most often applied to produce streamflowsequences The techniques segenerate rainfall reservoir designor operation studiesbut can also be used to be input to simulation models' quences that can subsequently of all possiblefuture to be representative If historioalflows iould be considered during its lifetime, there would be little variationsthat someproject will experience for predicting The hiitorical record is seldomadequate needfor synthetichyOroiogy. to recur, pattern is unlikely future eventswith certainty,however.The exacthistorical may they as severe of dry years (or wet years) may not have been as sequences the of knowledge beiome, and the singlehistorical record gives the planner limited magnitudeof risks involved.

536

CHAPTER22

TIME SERIESANALYSIS HYDROLOGIC

particularly if Syhthesisenableshydrologiststo deal with data inadequaci.es, of hydrologic records historical Short record lengths are not sufficiently extensive. synhydrologic sequenceslrsing longer to variables*"fu u, streamfloware extended hydrology'1 as operational known science of the broad thesisand other techniques either preservethe statisticalcharacterof the historThesenew, syntheticsequences ical records or follow a prescribedprobability distribution, or both. When coupled provide hydrologistswith imthe techniques with computer simulation techniques, proved designand analysiscapabilities. The rnethodsdescribedin this chapterare basedon probability and statistics. The material presentedin Chapter 26 should be reviewed prior to studying this chapter.

HYDROLOGY 22.1 SYNTHETIC

techniquesare classifiedas (I) historical repetition methods, Hydrologic synthesis that historical recordswill repeatthemwhich assume suchas masscurve analyses, sevlesin as many end-to-endrepetitionsas required to bracketthe planningperiod; which assume such as Monte Carlo techniques, (2) random generationtechniques, any of which events, independent random, that the historicalrecordsare a numberof methods, (3) persistence and could occur within a definedprobability distribution; are in sequence flows that assume such as Markov generationtechniques,which of the subset by some influenced is dependentand thit the next flow in.sequence are normally previousflows. Historical repetition or random generationtechniques intervals time shorter for flows flows. Successive applied only to annual or seasonal method. generation Markov the by analysis are usually correlated,necessitating of hydrology,a number of computerprogramsfor timeAs with most subfields One ofthe first, and havebeendeveloped. andhydrologicdatasynthesis seriesanalysis model HEC-4 of Engineers Corps Army wasthe U.S. one of the most widely app1i"d, synthesizing to though, limited, is Its use in I97IJ (seeSection21.a) pubfished codess'e Other reach. river in a streamflows monthly of seriaitydependent sequences of descriptions and models Additional however. are avarlableto thi hy-drologist, are streamflow and precipitation of theory and applicationsof time-seriesanalysis detailedin a number of availabletexts and publications'10-l3

MassCurveAnalysis
by Ripplla to inveswas devised techniques One of the earliestand simplestsynthesis that the future His analysisassumes tigate reservoirptoragecapacityrequirements. in its entirety inflowsto a reseivoirwill be a duplicateof the historicalrecord repeated reservoir. the life of useful the span to is necessary as fo end end times many as critical during release for waters surplus to hold selected is then storage Sufficient Reservoirsize selectionis easily accomperiods when inflows fall short of demands. pmfr"a from an analysisof peaks and troughs in the mass curve of accumulated Future flows can be similar, but are unlikely to be syntheticinflow versustime.15-17 produce identicalto pastflows.Randomgenerationand Markov modelingtechniques of, historical flows. t-hatare difJerentfrom, although still representative seqUences

HYDROLOGY 537 22.1 SYNTHETIC

EXAMPLE 22.1 Streamflowspast a proposedreservoir site during a 5-year period of record were, acre-ft.use Rippl's 8000,and 12,000 10,000,6000, 14,000, in eachyear respectively, provide a yield of to needed reservoir of size the to determine method ,nur. Crrru" 9000 acre-ft in eachof the next 10 years. is of syntheticflows,usingRippl's assumptions, solution. A lo-year sequence repeated record historical the to equal shown in Table Z2.l.Inflows are set twice. rl EXAMPLE22'1 FOR 22.1 STREAMFLOWS TABLE
Flows (thousandsof acre-ft) Year Inflow Cumulative inflow

r 1 14

2 4 1 24

3 4 0 6 8 38 30

5 6 1 2 1 64 50

7 8 9 1 0 4 1 0 6 8 1 2 100 88 80 74

draft of 9000 acre-ft per year for 10 years.

90,000 80,000

^
I

70,000 60,000 50,000 40,000

4000 acre ft storagerequired Storagerequfued for 9000-acre-ft/Yr draft is maximum ,' , Cumulative draft, slopeof9000 acre-fl/Yr

30,000 . 20,000 10.000

1 2 3 4 5 Year 6

Eigarc 22.1 Mass curve for ExamPle 22.1: inflow; --- cumulativedtaft.

cumulative

538

CHAPTER 22 HYDRoLoGIc TIME SERIES ANALYSIS

RandomGeneratiOn
one method of generatingsequences of future flows is a simplerandom rearrangement of past records.If the streamis ungauged and recordsare not available, a probability distributioncan be selected and a sequence of future flows that follow the distribution and haveprescribedstatisticalmomentsis generated. Wheneverhistorical flows are available,a reasonable sequence of future flows can be synthesized by first consultinga table of randomnumbers,selecting a number, matchingthis with the rank-in-file numberof a pastflow, and listing thecorresponding flow as the first value in the new sequence. The next random numberwould be used in a similar fasion to generate the next flow, and so on. Randomnumbershavingno corresponding flows are neglected and the next randomnumberis selected. Table-B.3 in Appendix B is a table of uniformly distributedrandom numbers(eachsuccessive numberhas an equalprobability of taking on any of the possible values). To illustrate the use of Table B.3 in the random generation process, the first three yearsof a syntheticflow sequence could be generated by selecting the 53rd, 74th, and23rdfrom the list of past flows. Alternatively, the flows in 1953, 1974, and 1923 could also be selected as the new randomsequence. Most computershave random number generationcapabilitiesin their system libraries. Rather than storing large tables of numberssuchas Table B.3, successive random integersare usually generated by the computer. EXA]I/IPLE22.2 Annualflowsin Crooked Creekwere 19,000, 14,000, 21,000, 8000,11,000, 23,000, 10,000,and9000acre-ft,respectively,fo y,e rs l ,r2 3a ,4 ,5,6,7,and8.Generatea 5-year sequence of annual flows, O,, by matching five random numberswith year numbers. Solution. Randomintegers between0 and 9 are generated from the computer. The Q, valuesin Table22.2 areselected from the eight given flowsby matching the respective year numberwith the random number.The digit t has no correspondingflow in the 8-year sequence, so the next random number, 2, places the 14,000-cfs flow in Year 2 in the first position of the synthetic 5-year sequence.rI
TABLE 22.2 DEVELOPMENT OF s.YEAR SYNTHETIC SEQUENCE Randomdigit
I o

Q(acre-ft)

2 3
I

5 6

2 5 8 I
I

Skip 14,000 11,000 9,000 19,000 8.000

544

CHAPTER22

TIME SERIESANALYSIS HYDROLOGIC

'

coefficient of the standarddeviation for the daily flow loga3.- a regression rithms within each month of record to the logarithm of the monthly total flow., the simulationof daily flows could be structured Given the calculatedstatistics, in the following manner: from Eq. 22'1,0. standardized variates 1. Generate 2. Use the logarithm of the monthly mean flow as an initial estimateof the mean of the logarithmic daily flows for the month' 3. Calculatethe standarddeviationof flow logarithmsby the previouslydetermined regression equation. 4. Apply the inverseof Eq. 22.1I, k--

?lz('
-:) .

'l-;
13 1

(22.r2)

variatesto flows,multiplying by the approprito transformthe standardized ate standarddeviationand addingthe mean. and the given 5. Add the differencebetweenthe total monthly flow generated the simulation. repeat given flow, and monthly monthly flow to the 6. Multiply daily results of the secondsimulation by the ratio of the given monthly total. monthly total to the generated Each simulation techniquecommonly requiresmodificiationswhen applied to individual problems.Methodsoutlined thus far can be utilized as guidesin establishflows.Simulationof flows for a given station ing a procedureto follow in synthesizing means,and standarddeviations with serial correlation,skewness, hasbeenpresented the preserfor an entire system, rnaintained.When generatingstreamflowsequences betweenstationsbecomesa significantfactor. vation of cross-correlation are employedto determinethe capacrunoff sequences Syntheticallygenerated flow magniities of reservoirsto satisfy specifieddemands.Individually generated of flows. Hydrolosequences tudesare uncertain, as are the syntheticallygenerated severalequally likd sequences probabilitiesof flowsby generating gistscanestimate of certain values.Herein lies one of the most recurrences of flowsand then evaluating useful applicationsof Markov generatingtechniques.

i Summary
as comprisingonly the deterministicmodels Hydrologicmodelingis often presented in Chapters2I,23,24, and 25. The fully described of the rainfall-runoff process equippedhydrologistincorporatesthe synthetichydrology models describedin this A growing numberof chapterin the analysisand designof water resotrces systems. hydrology and timeof synthetic basis projects are constructedor operatedon the year. sedesanalysiseach

PROBLEMS

545

PROBLEMS
record in Example22'2 and detet22.1. plot cumulativeinflows versustime for the S-year yield of neededto provide a mine by mass curve analysisthe size of the r_eservoir maximumyield possible? is,the What 12,000acre-ft in eachoithe next 24years. of synthetic sequence 'r, t use the annual rainfall trom Table 26.2to generatea lO-year assumption' annual rain depthsfor Richmond using Rippl's masscurve curve methods'Use 22.3. RepeatProblem 22.2 wingrandom generationrather than mass with the last two digits of these match and taUl'n.: numberJfrom i*o-Olglt.unaom in Problem26.32. the yearnumbers

22.4. RepeatProblem22.2wingrandomgenerationtogenefateal0-yearsyntheticsestandard
tr<

and qu"n"" of annual rain defths that has.a normal CDF with a mean 1 26'32' Problem from data rain annual of ihe to that equal deviation follow a log-Pearson Repeat Problem 22.4 assumingthat the annual rain depths and skewofthe deviation, standard mean, the use statisiics Type III distribution.For logarithmsof annualrain at Richmond' (c) Pearson using (a) normal distribution, (Uitog-nbrmal distribution' and distribution.

22.6. SelectagaugedStreaminyourgeographiclocationandprepareaquarterlymodel Type III

)a1

CanyouconvertthesimulationprobleminExamp|e22.4toalog_normaldistribution given in the example? simulation?What difficulties aie encounteredwith the data duration of rainfall data, flt a distributionto the time betweenstorms,and a computerpfogram Prepare month. selected the covering data 20 yearsof recorded the times betweenstormsand the durationsof storms' to randomly generate

22.8. Selectamonthofthunderstormactivityinyourregion.FrompublishedNoAAhourly storms'for

22.9. Flowsduring6yearsofrecordwereusedinsynthesizingthemasscurveshownonthe
following page. a.UseRippl'sassumptionandthegraphtodeterminethemissingmagnitudeofthe flow for the 12th Year. yield of 2000 acreb Determinethe reservoircapacityrequired to allow an annual acte-ftlYr' for 500 ftlyr. RePeat does this value relate c. Determine the maximum yieid possibleat the site. How statisticallYto the flows? how you would developa table of randomprecip22.10. Describewith words and equations of 4 in' and a standard itation depthsthat follow a normal distributionandhavea mean deviationof3in.Useyourmethodtocalculatethefirstthreedepths. use random of uniformly distributedrandom numbersis given below' 22.LI. A sequence follow a will that depths rain annual of generationto generatei 5_y"u, sequence deviation standard a in'' 25'8 of mean a have F"u..on fype iU distribution and will of4'0in.,andaskewcoefficientof_2.2o.Randomnumberstobeusedare20,0I' 9 0 . 0 3 .a n d 8 0 . Type III to a Pearson 22.L2. Total July runoff from a basinis randomly distributedaccording the acre-ft' 1000 is deviation standard the acre-ft', 10,000 is distribution. The mean : with Start is 0.50. Q1 coefficient skewis -0.6, and the lag-oneserialcorrelation selected randomly of sequence if a flows 10,000and find nu" -oiJuu.kov-generated return periodsgives2, 100, 10,2, and50 years'

546

CHAPTER22

HYDROLOGIC TIME SERIESANALYSIS 30,000

25,000

20,000
9l B

15,000

10.000

101112

Year Figtre 22.9 Mass curve.

REFERENCES
1 . U.S. Army Corps of Engineers,*HEC-4 Monthly Streamflow Sirnulation," Hydrologic EngineeringCenter, 197l. 2. R. M. Hirsch, "synthetic Hydrology and Water Supply Reliability," Water Resources Research, v. 15, no. 6, L979. "The Valueof Stochastic StreamflowModelsin Overyear J. R. M. Vogel,andJ. R. Stedinger, v. 25, no. 9, 1988. Research, ReservoirDesign Applications," WaterResources A Hydrologyin ReservoirOperation," J. Irrigation D. K. Frevert,et al., "Use of Stochastic ASCE,v. 115,no. 3, 1989. and DrainageEngineering, "An Evaluationof the Practicality and Complexityof SomeRainfall 5 . J. W. Delleur, et al., v. 12, no. 5, 1976. Research, and Runoff Time SeriesModels," WaterResources 6. J. D. Salas, et al., "Applied Modeling of Hydrologic Time Series," Water Resources Littleton, CO, 1980. Publications, "Operational HydrologyUsing Residuals," J. Hydraulics 1 . G. K. Young,and W. C. Pisano, Division,ASCE,v.94, no. HY4, 1968. "Applied StochasticTechniques, PersonalComputer Ver8 . U.S. Bureau of Reclamation, Division,Denver,CO, 1990. sion 5.2. User'sManual," Earth Sciences "SPIGOT, A Synthetic StreamflowGenerationSoft9 . J. C. Grygier, and J. R. Stedinger, ware Package," School of Civil and Environmental Engineering, Cornell University Ithaca.NY. 1990.

Chapter23

Models Simulation Continuous

r Prologue
The purposeof this chaPteris to: , Introduce and describecomputercodesavailablefor performing contrnuous simulation of surfacerunoff and streamflow' model. Presentin detail how one of the programs-the Stanfordwatershed cycle' hydrologic the of "otnponents simulatesthe miscellan"ou, .Showthemajorsimilaritiesanddifferencesoftheleadingmodels. from are developed . Providea detailedcasestudyof how the modelparameters availableinformation. . Illustrate, by using the casestudy,the stepsinvolvedin calibratinga continuous model and verifying the results' . show how well the moleh are able to replicate gaugedstreamflows. in this chapterprovide hydrologistswith tools for estiSimulation modelsdescribed in tittt" for precipitation'direct runoff ' accounting by continuously mating streamflow percolation' baseflow' and streamdeep interflow, infiltration, evaportranspiration, continuoussimulationmodelstrack stormi, flow. During rain-free intervalsbetween deeppercolation,and baseflow, evaporation, to of water.andits depletion the storage until the next rain or snow eventoccurs' ThemodelsarebasedonthephysicalprocessesdescribedinChaptersl_14.As

yasdeterministiciools't1;:H':1;ffi ff;ff t: such,theycrassif :T:TLi""ll':'ff

rain and snow into runoff and streamflow' can be usedto or other similar procedures' r are then input to continuoussimulation in Table2l'l ate Severalof the continuoussimulationmodelsidentifiedearlier is presented (swM-IV)' IV version model, here.The Stanford watershed described indetailastypicaloftheothermodels.Manyoftheothersare,infact,basedon

MODELS 549 SIMUIATION STREAMFLOW 2g.1 CONTINUOUS in the SWM{V, and several simulate various componentsof the hydrologic cycle of studies case independent and comparestwo samemanner. Section 23.2 ptesents the how and were {etermined Stanford model studies,showinghow the parameters modelswere calibratedand applied to the problemsbeing assessed'

MODELS SIMUI-ATION STREAMFLOW 23.1 CONTINUOUS API Model

simulation of a This model was one of the earlieststructuredto give a deterr.ninistic of 68 and watersheds on tested originally was It continuousstreamflowhydrograph. g:i miz and must be calibrated^ to each watershedto obtain a reliable method of r givenin Fig. 23' 1' simulatint the streamflow.lA flow diagramshowingthe structureis of to pertaining interrelations ttie Four basic components describe .this .model index, precipitation (antecedent API an hydrograph, streamflowin a river: a unit groundwater introduced in chapter 2 and iilustrated in Sec. 10'3), a relation for asa function hydrograph flow groundwater the and a relation for computing recession, flow and groundwater both generates model This of the diiect runoff hydrograph. to enjoy continues model API The values. precipitation from direct runoff discharge modeling' simulation in popularity and use widespread

Direct runoff hydrograph method) lunit-hydrograPh

Groundwater outflow hydrograph

Figure 23.1 Schematic diagram of API-type hYdrologic model. (After Sittner et al'r)

550

CHAPTER23

SIMULATION MODELS CONTINUOUS

StanfordWatershedModel lV (SWM-|V)
Crawford and Linsley designedthis digital computerprogram to simulateportions (the land phase)of the hydrologic cycle for an entire waftished.2The model has from the muchdevelopment sinceits conceptionand is currently available undergone U.S. Environmental Protection Agency under the name HSPF, which is a public of the original program. The subsequently) domain FORTRAN version (discussed of hydrograph a continuous as a tool to synthesize SWM-IV hasbeenwidely accepted parameter approachis A lumped hourly or daily streamflowsat a watershed,outlet. used and data requirementsare much less than for alternative distributedmodels. data,and a variety of watershed Hour$ and daily precipitationdata,daily evaporation parameters are input. of SWM-IV are shownin The relationsand linkage of the variouscomponents Fig.23.2. Hydrologicfundamentalsare usedat eachpoint to transformthe input data data are into a hydrographof streamflowat the basinoutlet. Rainfall and evaporation first enteredinto the program.Incoming rainfall is distributed,as showninFig.23.2, amonginterception,imperviousareassuchas lakes and streams,and water destined to be infiltrated or to appearin the upper zone as surfacerunoff or interflow, both of evenwhich contributeto the channelinflow. The infiltration and upper zone storage storage. and to activeand inactive groundwater tually percolateto lower zone storage parametersgovern the rate of water movementbetweenthe storage User-assigned zonesshownin Fig. 23.2. condiThree zonesof moistureregulatesoil moistureprofiles and groundwater for is accounted in smaller watersheds encountered tions. The rapid runoff response in the upper zone,while both upper and lower zonbscontrol suchfactorsas overland for longerThe lower zone is responsible flow, infiltration, and groundwaterstorage. term inflltration and groundwaterstoragethat is later releasedas base flow to the flow, and is a combinationof overlandflow, groundwater stream.The total streamflow interflow. of computationroutines Model Structure The SWM-IV is madeup of a sequence in the hydrologiccycle (interception,infiltration, routing, and so on). for eachprocess of each componentare provided in the following paragraphs. Separatediscussions Actual calculationsproceedfrom processto processas ilfustrated by the arrows in or wasinput as Fig.23.2. All the moisturethat was originally storedin the watershed precipitation during any time period is balancedin the continuity equation P : E + R + A S where

(23.r)

P : precipitation E :.evapotranspiration R : runoff AS : the total change in storagein the upper, lower, and groundwater storagezones The changein storagefor each zone is calculated as the difference between the volumesof inflow and outflow. Furthermore,all hydrologicactivity in a time interval to the next time interval. The is simulatedand balancedbeforethe programproceeds simulationterminateswhen no additional data are input.

MODELS SIMULATION STREAMFLOW 2g.1 CONTINUOUS

551

maximumsare providedin Table23.1. MAXIMUM 23.1 ryPICAL TABLE

RATES INTERCEPTION Watershed cover Glassland Moderateforest cover Heavy forest cover
Source: After Crawford and Linsley.2

rate (in./hr) Interception

0.10 0.15 0.20

(ET) is assumed to occur at Evapotranspiration In SWM-IV evapotranspiration "upper" storage The upper zone' the and the potential rate from interceptionstorage soil zone lower The soils. surface permeable andhighly the depressions zonesimulates zone' storage , simulatesthe linkage to the groundwater from the lower zone is set equal to the ET opportunity, Evapotranspiration defined in fig. ZZ.Z.W opportunity is defined as the maximum amount of water availablefor ET at a partiiular location durrng a prescribedtime interval. In the modelinglogic, ET occursfrom severallocations(seeFig. 23'2) includingthe interand streamand lake surfaces, lower zonestorage, upperzonestofage, ceptionstorage, is storage zone arrdlpper frominterception ttotug". Evapotranspiration groundwater rate evaporation lake the to be is assumed Jet equal to the potential rate, Eo, which calcuiatedasthe productof a pan coefficienttimes the input valuesof the evaporation to occur at a rate equal wateris assumed of any intercepted pan data.The evaporation has storage interception the when ceases and rate io the potential evapotranspiration been depleted. alsooccursat the potentialrate. The from streamand lake surfaces Evaporation of streamsand lakes(ETL) defined area surface by the total total volgmeis governed to the total watershed watershed the in area lake and as the ratio of the total str-eam at the potential rate occurs also storage groundwater from area.Evapotranspiration to a factor K24EL equal area a surface using and is caiculatedin a similar fashion the fraction represents K24EL parameter the area.Thus multiplied by the watershed groundwater the from evapotranspiration of tfre total watershedarea over which at avalue equalto the fraction setthis parameter will occur.Most investigators storage is normally small but can value Its phreatophytes. area coveredby of the watershed acresof subirrigated many has that area be large, for eiample, in an agri-ultural alfalfa. the modelwill attemptto satisfythe potential is depleted, If interceptionstorage at the potential rate. Once the upper for ET by drawingfrom the upper zone storage zonebut not at the potential rate; lower zone sto;ageis dJpleted,ET ociurs from the Eo. When interception and the than less the ET ratb from the lower zone is always entersas EoinFig-23.3, excess any potential, do not satisfythe uppef zone storage

MODEI.S 23 CONTINUOUS SIMUIATION CHAPTER

/-------'-------\
/r \.
\

Actual evaootransoiration -'-r--\--------1-------J I

t., /
/

l* - - - - - - - - - - - - - -----\

model IV flowchart. (After Crawford and Linsley.2) Figure 23.2 Stanford watershed

MODELS SIMULATION STREAMFLOW 29.1 CONTINUOUS

553

F".""I
(: "iPrf) @
Channel inflow

@
Channel inflow Channel inflow

zi
\

-k^-*ff^." \ - - - - ; - - J

Simulated streamflow

tt ,'
t

Eigure 23.2 Continued

"7
554
CHAPTER23 CONTINUOUS SIMULATION MODELS

''E. h a'.: va' x x

n l -

25

50

75

100

Percentase oraretxTi* SfttaffiTffi,:'JTl?J."pportunitv

Figure 23.3 Evapotranspiration relation used in the Stanfordwatershed model. (After Crawford and Linsley'1 and the rate of evapotranspiration from the lower zone is determined from the shaded afea, or

E:Eo_% Q3.2) 2r The variable r is the evapotranspiration opportunity, deflnedas the maximum water amountavailable for ET at a particular location during a prescribedtime period. This factorvariesfrom point to point over any watershed from zero to a maximumvalue of -.^ LZS (23.3) r : ,''J LZSN where LZS : the current soil moisture storagein the lower zone (in.) LZSN : a nominal storage level, normally set equal to the medianvalue of the lower zone storage(in.) K3 : an input parameterthat is a function of watershed cover as shown inTable 23.2 The ratio LZSILZSN is known as the lower zone soil moisture ratio and is usedto comparethe actuallower zone storage with the nominal value at any time. Valuesof ET opportunity are assumed to vary over a watershed from zero to r along.the straight line shownin Fig. 23.3. This assumed linear cumulativedistributionof the parameter over an area is also usedin evaluatingareal disftibutionsof infiltration rates. Infiltration Like the erapotranspirationopportunity, the infiltration capacity of a watershedis highly variable from point to point and is assumedto be distributed accordingto a linear cumulativedistributionfunction shownas a line from the origin to Point b inFig.23.4. TABLE 23.2 ryHCALLOWER ZONE
EVAPOTMNSPIRATION PARAMETERS

Watershed cover
Open land Grassland Light forest Heavy forest

0.20 0.23 0.28 0.30

MODELS SIMULATION STREAMFLOW 23.1 CONTINUOUS

555

Percentage of area with an infiltration capacity equal to or less than the indicated value

Figure 23.4 Assumedlinear areal variation of inflltration capacity over a watershed. (After Crawford and Linsley.2)

Infiltration into the lower and groundwaterstoragezones is determinedas a function of the moisture supply 7 available for infiltration. Steps to determine infiltration for a given moisture supply7 are: "infiltration" in 1. The net infiltration is determinedfrom the area labeled to infiltrate into the lower and groundwater Fig. 23.4.This wateris assumed by the trapezoidis given by the equations zones.The areaenclosed storage the maximum in the first row of Table23.3.If themoisturesupply7 exceeds infiltration capacityb, the maximum allowed net infiltration is b/2, which is the median infiltration capacity. 2. Some of the moisture supply contributesto an increasein the interflow detentionduring any time increment and is calculatedas the region indicatedby an arrow in Fig. 23.4.Eqtationsfor this areausingvariousranges in x are providedin the secondrow of Table 23.3. The volume of water in asinterflow at any instantis calledthe interflow a stateofbeing transported detentionor detainedinterflow. to increasing 3. Any remainingmoisturesupplied,AD in Fig. 23.4, contributes for this Equations the surface detention during the time increment. of 7. values for various area are included in Table 23.3 triangular-shaped
TABLE 23,3 EQUATIONSFOR THE SHADEDAREAS IN FIG. 23.3

Component
Net infiltration Increasein interflow detention Increasein surfacedetention of increaseddetention Percentage to interflow assigned
Searce; After Crawford and Linsley.2

x<b

b<i<cb

x>cb
b 2

x-

zb

t
b 2 i2 2cb

i2/ r\ - l l - - f c/ 2r\ 2rb

u-t, - ts
2' - c b

2rb
/ r \

x-t
/ , t \

r oo(r:)

r oo(r "*

_ r tr )

too2- _ ilb j

c - l

556

MODELS CHAPTER23 CONTINUOUSSIMULATION

ThEquantity of net infiltration is contrblledlargelyby the maximuminfiltration because the capacityb, while the pbrameterc significaltly affectshydrographshapes The increment. parametercoritrols the amount of water'detainedduring thelime valuesof b andc foi ahy time intetval dependon the soil moistureratio, LZS/LZSN, and on the input parametersCB and CC; CB is an index that controls the rate of on the soil permeabilityand the volume of moisturethat can infilffation and depends be storedin the soil. Valuesin the rangefrom 0.3 to 1.2 arecommon.The parameter CC is an input value that fixes the level of interflow relative to the overland flow. Values of CC rangefrom 1.0 to 5.0. If the soil moistureratio is lessthan 1.0, the variable b is found from

u "

cB
2ALzs/LzsN)

(23.4)

and whenLZS/LZSN is greaterthan 1.0,the equationfor b is

. 0 :

,n-rr"tvttt*,

*n

rr?5)

by Crawfordand Linsley from num0rousttials using Theseequations were developed a value When the soil moistureratio reaches SWM-IV in many different watersheds. parameter The c is of 2.0, the variable b reachesits minimum value of *r of CB. determinedfrctm
c : .(CC)ZQzslLzsN)

(23.6)

Variations in parametetsb and c with changesinLZS/LZSN aie shown in Figs. of CB : 1.0and CC : 1.0wereusedin developing 23.5 and23.6.Midrangevalues thesecurves. Figure 23.7 is a graph of distributionof water amonginfiltration, intefflow,'and overland flow for various valuesof the moisture supply 7. Different valuesof b and c would producea different set of curves. Water stoied as overland flow surface detention will either contribute to in Fig. 23.2.The portion that as'depicted streamflowof enterthe upper zone storage

0.0

0.2

0.4 0.6 0.8

1.0 1.2

r.4

1.6 1.8 2.O

ratt" (-!ZL'J zone Lower soilmoisture b for variousvaluesofthe soil Figure 23.5 Variationin patarneter moisture ratio. (After Crawfordand Linsley.2)

0.2

0.4 0.6

0.8

1.0

1.2

1.4

1.6

zone soll molstufe rauo Lower \-I_ZSN'

/ Lzs)

Figure 23.6 Variation in parameterc for various values of the soil moisture ratio. (After Crawford and Linsley.2)

o o I
E

lncrease in overland flow surface detention

b=L.0 c=1.5

0.2 0.4 0.6 0.8 1.0 1.2 1,.4 1.6 supply 7 Moisture

to moisture supFigure 23.7 Typical SWM-IV response ply variations.(After Crawford and Linsley.2)

100
; Y

8F
o 9 ^

80 60
, Inflection

bR

0 0

0.5

1.0

1.5

2.0

2.5

3.0

rati" (ffi) zone soilmoisture Upper Figure 23.8 Delayedinfiltration as a function ofupper zone soilmoisture ratio. (After Crawford and Linsley.z)

558

CHAPTER23

CONTINUOUS SIMULATION MODELS

entersthe upper zone storageis called delayedinfiltration and is a function of the upper zone soil moistureratio,\JZS/UZSN, as shownin Fig. 23.8. The inflection point occursat a soil moistureratio of 2.0. If the ratio is lessthhn 2.A, thepercentage retainedby the upper zone is given by

- (ffi)(--ri-)"'' e: roolr' ]
where UZI| is determinedfrom

(23.7)

rJZrr:rrl#k-r.o ]r . o +
The curve is definedto the right of the inflection point by

(23.8)

P': lool(tt . uzLz/ r

whereUZI2 is determinedfrom

f/

t.0_)r,',-l

(23.e)

rrzr2:rrlffi-z.+ o] r.o

(23.r0)

zone,as shownin Fig. 23.2; receives a Upper Zone Storage The upper storage large portion of the rain during the flrst few hours of the storm, while the lower and groundwaterstorage zonesmay or may not receiveany moisture.The portion of the upper zone storage that is not evaporated or transpiredis proportionedto the surface runoff, interflow, and percolation.Percolation(upperzone depletion)from the upper LZS/LZSN. zoneto thelowerzonein Fig.23.2 ocurs only whenUZS/UZSN exceeds When this occurs,the percolationrate in in./hr is determinedfrom

: o.oo3(cBxuzs$(ffi PERC ffi)'

(23.rr)

where CB is an index that controls the rate of infiltration. It rangesfrom 0.3 to 1.2 depending on the soil permeability and on the volume of moisturethat can be stored in the soil. The variablesUZS and UZSN are definedas the actual and nominal soil moisturestorageamountsin the upper zone.The nominal value of UZSN is approximately a function of watershed to be topographyand cover and is alwaysconsidered much smaller than the nominal LZSN value.The initial estimatesof UZSN relative to LZSN are found ftomTable 23.4.
TABLE 23.4 VALOESOF UZSN AS A FUNCTIONOF LZSN FOR INITIAL ESTIMATESIN SIMULATION WITH SWM-IV Watershed Steepslopes,limited vegetation,low depression storage Moderateslopes,moderatevegetaion,moderatedepression storage Heavy vegetalor forest cover, soils subjectto cracking,high depression storage,very mild slopes
Source: After Crawford and Linesley.2

O.O6LZSN O.OSLZSN O.14LZSN

MODELS SIMULATION 29.1 CONTINUOUS STREAMFLOW

559

The parametersLZSN and CB must also be estimatedat the beginningof a simulation study.The combinationthat will most satisfactorilyreproduceboth longto hydrologic inputs can be determinedby the and short-termhistorical responses following procedure:2 1. Assumean initial value for LZSN equal to one quarterof the mean annual rainfall plus 4 in. (usedin arid and semiaridregions),or one eighth of the annual mean rainfall plus 4 in. (usedin coastal,humid, or subhumidclimarcsJ. ) Determinethe initial value of UZSN fromTable 23.4. 3. Assumea valuefor CB in the normal rangefrom 0.3 to I.2. 4. Simulatea period of record usingthe streamflow,rainfall, and evaporation data and systematicallyadjust LZSN, UZSN, CB, and other parameters is satisfacand recordedstreamflows betweensynthesized until agreement CB is LZSN is adjusted; do not balance, tory. If the annualwaterbudgets and recorded adjustedon the basis of comparisonsbetween synthesized flow rates for individual storms. zonein storage Lower Zone Storage and Groundwater The lower groundwater The percolaFig.23.2 receives waterfrom the net infiltration and from percolation. of net infiltrationthat reaches from F;q.23.I1.Thepercentage tion rateis determined groundwaterstorage depends on the soil moisture ratio LZS ILZSN as shownin Fig. P, is found from 23.9.If this ratio is lessthan 1.0,the percentage

P,:1oo[#(-#t,"',
is andif LZS/LZSN is greaterthan 1.0,the percentage

(23.r2)

equations, tn",*:r* rn both :, _ i'E;il,

+:roofr'-(--rg l"]
] .,.,

(23.r3)

(23.r4)

LZS LZSN equalsthe lower zonestorage Notefrom Fig.23.9 thatthe nominal storage when half or 50 percent of all the incoming moisture enters groundwaterstorage. The outflow from the groundwaterstorage,GWE at any time is basedon the versustime. This commonly usedlinear semilogarithmicplot of baseflow discharge in Fig. 11.8.In modifledform technique wasdelcribedin Section11.4andillustrated the baseflow equationis GwF : (LKK4)[1.0 + KV(Gws)](sGw) where LKK4 is definedby LKK4:1.0-(KK24)t/e6

(23.rs)
(23.16)

in which KK24 is the minimum of all the observeddaily recessionconstants(see rateto the discharge 11.4), whereeachconstantis the ratio of the groundwater Secti,on

560

CHAPTER23

MODELS SIMULATION CONTINUOUS

Inflection 9oo
oF \ i ltt

0.5

1.0

1.5

2.5

*u" (;R) moisture Soil of infiltratedwaterthat reaches Figure 23.9 Percentage andLinsley'2) (After Crawford storage. groundwater (K n groundwaterdischargerute 24 hr earlier. Thus the recessionconstantKK24 values has The variableGWS in Eq.23.15 usingt : I day. Eq. 11.1)is determined Its value on any given storage. groundwater to inflows long-term on the that depend day'svalue,adjusted previous as97 percenLofthe day (e.i., the lth day)is calculated : or GWS; 0'97 (GWS'-1 * inflow to groundstorage, foi any inflow to groundwater water storageduring daYl)' In Eq. 23.J{, SGW is a groundwater storage parametel that reflects the term fluctuationJin the volume of water storedand rangesfrom 0.10 to 3.90 in' The groundwater KV in Eq. 23.15 allows for changesthat are known to exist in the to Eq. 11.1and when KV is zero, E,q.23.15reduces ratesas time passes. recession follows the linear semilogrelation- If the usual dry season recessilon the groundwater arebeing slofaggs rate KKz4is too largefor wet periods(whengroundwater recession so KV is hand-adjusted that the ihe parameter from the streams) by seepage recharged during term i0 +-KVaGWS) will reduce the effective rate to some desiredvalue receseffective periods.Table 23.5illustratesthis computationby showing recharge to 1'0' equal set sion rates for variouscombinationsof KK24 and GWS when KV is or to deep lost that is either The fraction of activeor deepgroundwaterstorage drainage the or is diverted as flow across inactive groundwaterstorage Gi;.b.D This fraction is the total inflow to K24L. pu.u*tt"r ur input is basin boundary FOR RATES RECESSION 23.5 EFFECTIVE TABLE
OF KK24 VARIOUSCOMBINATIONS AND GWS WHEN l(/ : 1'0 GWS

0,5 0.99 0.98 0.97 0.96 0.99 0.98 0.97 0.96 0.985 0.970 0.955 0.940

1.0
0.98 0.96 0.94 0.92
n07

0.94 0.91 0.88

Source: After Crawford and Linsley.2

MODELS 561 SIMULATION S1REAMFLOW 23.1 CONTINUOUS groundwaterand representsall the active groundwaterstoragethat does not contribute to streamflow. Overland Flow The overlandflo* pro".r. hasbeenstudiedLy -uny investigators.

hydraulictechniques. of overland flow in the Averagevalues of lengths, slopes,and roughnesses calculatethe Manning and continuity equationsare usedin SWM-IV to continuously rate q is then relatedto D,. D".The overlandflow discharge detentionstorage surface As the rain supplyrate continuesin time, the amountof water detainedon the The amount of surface surfaceincreasesuntil an equilibrium depth is established. detentionat equilibrium estimatedby SWM-IV is D":
6 0.000818i0 no'6L1'6 s0'3

(23.r7)

where D" : the surfacedetention at equilibrium (ft3lft of overland flow width) j : the rain rate (in./hr) S : the slope(ftlft) L : the length of overlandffow (ft) coefficient ru : Manning's roughness The overland flow dischargerate is next determinedas a function of detention storage from

0* o,o(r2)']"' , : # y,,(?)"'[,
where e : the overland flow dischargerate (cfs per ft of width) D : the averagedetention storagedurrng the time interval

(23.18)

but the that occurs after rain ceases, The equationalso appliesduring the recession coefficients afe roughness flow overland Typical to be 1.0. ratio blD" is assumed providedin Table23.6. The time at which detention storagereachesan equilibrium is determirtedfrom
t

.e

: -

0.94L3/5n3/s
i2/s S3/ro

(23.te)

where /, is the time to equilibrium (min). Crawford and Linsley show that these overlandflow hydrographs.2 equationsvery accuratelyieproducemeasured For eachtime interval Lt, an end-of-intervalsurfacedetentionD, is calculated from the initial value D, plus any water addedAD (Fig. n.q to surfacedetention from during the time interval, lessany ovedandflow dischargeQthatescapes storage
.4.

562

23 CHAPTER

MODELS SIMULATION CONTINUOUS ryPICAL MANNINGEQUATIONOVERLANDFLOW NN PARAMETERS, ROUGHNESS n for overlandflow Manning's

TABLE;3.6

Watershed cover Smooth asphalt Asphalt or concretePaving Packedclay Light turf Denseturf DenseshrubberYand forest litter
Source: After Crawfordand Linsley.2

0.012 0.014 0.03 0.20 0.35 0.40

of continudetentionstorageduring the time interval. This is simply an expression itv. or Dz:Dt+LD-4Lt

(23.20)

(D: + D)/2' Equao! D The discharge . @is found from Eq' 23.18usinga value flow using easily of overland tions 23.17- 2i.20 allow the completedetermination overlandflow' roughness and length, slope, found basin-widevaluesof the average is treatedin the same Interflow The watertemporarilydetainedasinterflow storage detention was interflow to inflow fashion as overland flow detention storage.The similar constant recession a daily using is simulated in Fig. 23.4.Theoutflow defined is the IRC constant recession The interflow discharge. to that definedfor groundwater hr 24 discharge interflow to the at any time ratio of the interflow discharge average is storage detention from outflow earlier.For each15-min time intervalmodeled,the

rNTF : LIFC4(SRGX)
where

(23.2r)
()1n\

LIFC4:1.0-(tRc;'inu

Its The variableSRGXis the water storedin the interflow detentionat any time. time each to applied value continuouslychangeswhen the continuity equation is accordingto continuity, on the interval. The end-of-inteival value of SRGX depends, from the interflow valueat the beginningofthe interval and any inflow to or discharge detentionduring the interval. modelutilizesa hychannel Translation and Routing The Stanfordwatershed to translatethe channelinflow to the watershed routing technique drologicwatershed in Section 12.6is adoptedalmostas outlei. Clark's IUH time-ar"a to"ihod described l2.Inplaceofthenetrainhyetograph,theStanfordmodelviews presentedinchapter "inflow" hyetograph. This inflow is the sum of all channelinflow componentsas an is next routed it where then translatedin time through the channelto the basinoutlet, by storage caused to accountfor the attenuation system through an equivalentstorage sensethat (linear in the Roulingthrough the linear reservoir in the-channelsystem. is accom12.35) Eq. to be directly proportional to the outflow' storageis assumed plishedfrom (23.23) I or:7-KS1(1 -01) )

MODELS 563 SIMUI-ATION STREAMFLOW 29.1 CONTINUOUS where Oz: the outflow rate aI the end of the time interval

o t : the outflow rate at the beginning I _ the average inflow rate during the time interval
Also,

KSI

- Ltlz K* :v 6*

(23.24)

from watershed parameters Examplesof the determinationof K and othernecessary 23.2. in Section data areincluded Applications of the SWM-IV Applicationsof the model typically beginwith data for a three- to six-yearcalibration period for which rainfall and runoff data arc of severalparameters adjustments Thesedata are usedto allow successive available. agree.If sufficient streamflow the of hydrographs recorded until the simulatedand for use as a control to reserved be may record period of data areavailable,a second the first half of with a calibration from paiameters derived checkthe accuracyof the the data. in 1959and hasundermodelwasoriginally developed The Stanfordwatershed Crawfordand Linsthe translated James3 time. that since goneseveralmodifications of the FORTRAN modifications Several to FORTRAN. ley model from ALGOL theseare the among Included investigations. of a variety versionhave evolvedfrom self-calibrating Kentucky (KWM),4'5 the model Kentucky watershed _version (OpSETi,4the Ohio State University version, the Texas version,6the Hydrocomp nonproprietary SimulationProgram (HSP) written in PL/1, and EPA-produced, Servicerunoff Weather National the and HSPF, HSP called FORTRAN u"rrion of below. included are these of several of descriptions forecastingmodel. Brief

ARS RevisedModelof WatershedHydrology(USDAHL)

and farming Growing interest in the effects of soil types, vegetation,pavements, USDAHL growth of the practicei on infiltration and overlandflow hasresultedin the by developed was model continuoussimulation model. The 1974 versionTof this Laboratory. ServiceHydrograph at the Agricultural Research investigators Input data to the model are relatively extensive.Continuousrecords of the pan-evaporation the weekly average temperatures, precipitation,the weekly average practicesare and cultural land use, amounts,and detaileddata onioils, vegetation, required' is initially dividedinto asmany asfour distinctland-useor The studywatershed and a main calling routine computefor each soil-type ,on".. Fourteen subroutines zone the snowmelt, inflltration, overland flow, channel flow, evapotranspiration, recharge,and return flow' groundwaterevaporationand movement,groundwater to coefficients by app$ing assigned potentialsare estirnated Evapotranspiration a using is computed zone pun-"uuplrution data. Infiltration for eachsoil or land-use function a as is simulated soils Waterstoredin cracksin dry modifiedHoltan equation. of soil moisture.Manning's equationand the continuity equationare used to route solution of the continuity overlandflow. The streamflowis routedby a simultaneous

564

CHAPTER 23

MODELS CONTINUOUS SIMULATION

movementsare calculatedby Darcy's function. Groundwater equdiion and a storage zoneis calculatedas a function of the temperequation.The daily snowmelton eachdensityfor the zone, vegetative ature at which snowmeltstarts,the weightedaverage the weekly average air temperature,and the potential snowmeltper day in the zone snowpack. Precipitation falling during a snowmelt day also contributes to the snowmeltequation. Service Among otheruses, the modelhasbeenappliedby the Soil Conservation Figure 23.10 showsthe results of in preparing environmentalimpact statements. and widely applying the 1974 version to annual runoff'from four widely separated In addition to the runoff, the model comdiversifiedARS experimentalwatersheds. putes the evapotranspirationamounts, soil moisture changes,return flows, and groundwaterrechargedepthsfor eachof the zones. Although other modificationsare possible,the USDAHL model is specifically generallyunder 20 squaremiles. for relatively small rural watersheds, designed

$-i

E U

Cumulative computed runoff, O (in.)

Figure 23.10 Chart showing the accuracy of USDAHL-74 model for estimatingthe cumulativecomputedrunoff as comparedwith the cumulao W-97, Coshocton, OH: A W- 11, tive measured runoff at four watersheds. Hastings,NE; I W-3, Ft. Lauderadale,FL; x W-G, Riesel, TX (After Holtan andLopez.l)

MODELS SIMULATION STREAMFLOW 2g.1 CONTINUOUS

565

NationalWeatherServiceRiver Forecast System(NWSRFS)

by the Hydrologic modelwasdeveloped yet anotherversionof the Stanfordwatershed The Hydrology.8 of Office Service Weather Laboratory staff at the National Research by the stages and flows river forcecasting in use for NWSRFS model wis developed river several to successfully applied been has National WeatherService.The model River basinsrangingin sizefrom 70 mi2in North Carolinato 1000mi2in Oklahoma' for SWM in incorporated detail the require not does forecastin! in"largeriver basins changes major two includes model NWS the For this reason' smaller watersheds. fewer process involvingthe useof a longertime increment,simplifiedprogramming, parameters watershed optimal determining for comput;ions, and a rapid procedure accurately. flows historical that allow the model to reproduce inputs and A 6-hr time incrementis usedby the model, allowing fewer rainfall flow that overland as such processes of ,nor" i.po.tant, fewer detailedcalcuiations with than rapidly mo_re completed thus are occur in'shorter time periods.Iterations optimization Service Weather National the the SWM. As with the OPSETmodel, available procedurefor determiningparametervaluesgivesthe model a strengthnot with the SWM-IV. Other modificationsincludeh uPPerand lower zone retentionand the uPPersoil zone to groundwate al groundwaterevapotranspiration jointlY comPuted in the NWS versi runoff, interflow' and groundnatedand is replacedby three types ofrunoff: surface response' slow and water flow-representing fast, medium, and instantaInput data for modll calibration consistof meandaily discharges a continuous as is input Rainfall events. runoff for a few selected neoushydrographs Betechniques. averaging areal from record of 6-hr basin-widemeansdetermined the process simulation' of detail in the and in routing increment causeof the changes outputfromtheNwsiersionissimilarinmakeuptotheSWM-IVoutput.

Regulation and Reservoir Synthesis COEStreamflow Model (SSARR)

for large Another widely used continuous streamflow simulation model designed developed was model SSARR The basinswas devllopedby the corps of Engineers.e operation primarily for streamflowand flood fo.ecastingand for reservoirdesignand Weather National at the hydrologists of NWSRFS, i,rior to the development studies. rain and both to applied been has model Service used.the SSARR model. The snowmeltevents. basin into Applications of the model begin with a subdivisionof the drainage subdivides, with consistent character and hydrologic units of i size homogeneous rJservoirsites,diversionpoints,soil types,and otherdistinguishchannelconfluences, are computedfbr ail significantpoints throughoutthe ing features.The streamflows river sYstem.

566

MODELS 23 CONTINUOUS SIMUI-ATION CHAPTER duintutt data can be input at any numberof stationsin the basin.The part that or interflow, and surfacerunoff. will run off is divided into the baseflow, subsurface The division is based on indices and on the intensity of the direct runoff. Each and all are then comcomponentis simply delayedaccordingto different processes, runoff is then outflow hydrograph.This subarea bined to producethe final subbasin routed through stream channelsand reservoirsto be combined with other subarea hydrographs, all of which becomepart of the output. by the sametechRoutingsthrough channelsand reservoirsare accomplished allowances and occasional of shortstreamreaches, nique.This requiresan assumption in the channelrouting process.Streamflowsare for backwatereffectsare necessary on the basisof rainfall and snowmeltrunoff. Snowmeltcanbe determined synthesized on the basis of the precipitation depth, elevation,air and dew point temperatures, albedo,radiation, and wind speed.Snowmeltoptions include the temperatureindex method or the energybudgetmethod. indicesfor subdiInput includesthe precipitation depths,the watershed-runoff and outflows, processes, elevations initial reservoir flow among the three viding from reservoirs, discharge allowable storage and areas, bounds on usable drainage to control instructions other special periods, intervals, and routing total computation plots, prints, and other input-outputalternatives. This model was one of the earliestcontinuousstreamflowsimulation models and has its primary strengthin its verified using a lumped parameterrepresentation large drainagebasinsincluding the in several indicated by tests conducted accuracy River basin. Mekong River basin and the Columbia

Program(HSP) Hydrocomp Simulation

at Hydrocomp, A commercial version of the Stanford water model was developed incorAmong severaladvantages Inc., namedthe HydrocompSimulationProgram.to chanand kinematic-wave poratedin HSP are hydraulic reservoirrouting techniques Other major changesinclude the addition of water quality nel routing techniques. simulation capabilities.Due to theseadditions,the model is often referredto as the Hydrocomp water quality model. The HSP model has been usedroutinely for severaltypes of hydrologicstudy including floodplain mapping,water quality studies,storm water and urban flooding studies,urban drainagefacility design,and water quality aspectsof urban runoff. The model consistsof three computerroutines: to handleinput datawith disk storage l, Library allowsthe useof direct access routines. efficient data management in 2. Lands handlesthe usual SWM lands phasealong with addedprocesses calculating soil moisture budgets,groundwaterrecharge and discharge, inflow to streamchannels,and eutrophication. 3. Channel is responsiblefor assemblingand routing all channel inflow through channelnetworks,lakes, and reservoirs. water balanceby trackingprecipitaThe HSP model incorporatesa continuous of the hydrologicand water resourcesystem.The tion through all possibleavenues

29j

MODELS SIMULATION STREAMFLOW CONTINUOUS

567

inflow, interflow, and surfacerunoff are individually simulated,lagged, groundwater ind combinedat appropriatetimes as the channelinflow. The routing of computed model.Water inflow through the ihannel networkutilizes a modified kinematic-wave parameters water other and rates discharge variable related to are quality constituents is accomplished. runoff quality of the quantity and so that the coupling of radiation,wind, and Inputs foisimulating waterquality includethe temperature, form the basis for which study under factors of the humidity and observedvalues however,calibrafor calibration; preferable are years data of calibration.At lest two input includes required Other year data. of one than with less tion hasbeenachieved and potential increments greater time or or 30-min 15-, in 5-, hourly precipitation temperature, the is desired, quality simulation or water if snow evapoffanspiration; needed' are factors radiation, wind, and humidity Outputsfrom HSP can be obtainedfor any desiredpoint within the watershed. Included in the output options are valuesof quality dataat outfalls or other points, rates,streamand lake ,"r"ruoi, levels,hourly and meandaily discharge river stages, counts,phosphorus' algae solids, dissolved total and dissolvedoxygen remperarures, BOD, colipH, carbonaceous phosphate, nitrogen, nitr;te, nitrite, ammonia,iotal budgets, water annual and monthly, daily, usual the metals,and forms, conservative equivalents' moisture snow depths,and end-of-period Typical simulation periods in HSP applicationsrange from 20 to 50 years. of future flows. Rather, the Hour-by-hour data are not viewed as an exact sequence of rangesin the factorsof occurrences of probability data arcusedfor analysisof the with a purposesimilar functioning is model the interest.When usedin this -unir"r, in Chapter22. described techniques hydrology to that of someof the operational

Program-Fortran(HSPR Simulation EPA Hydrocomp

of the HSP versionof the Stanfordmodel,the U.S. EnvironFollowing development mental Piotection Agency contractedin 1980 to have public-domain version made for continuousstreamflowsimulationand water quality modeling.The origavailable The HSPF code inal program,written in ALGOL, wasconvertedto FORTRAN 77'11 quality modeling alwater of portions is availa|le for PC applications.Substantial processes, cycle hydrologic The HSPF. gorithms were addedio HSP in developing is the exception One HSP' and however, are essentiallythe same as in SWM-IV available. previously addition of severalrouting proceduresnot for all the hydroparameters that drive ihe routinesin HSPFmustbe estimated of the numerous because difficult making verification of the model logic processes, and caliparameters these estimating combinationsof paramiter values.Methodsof (Table defines 23.8 23.2 in Section bratingthe modeiareillustratedin the casestudies usedin the Stanfordmodel)' over 35 parameters

ModelingSystem(PRMS) USGSPrecipitation'Runoff
their urban storm-eventmodel, DR3M (seeChapter 25), the U.S' After developing severalother computercodesto model continuoushyGeologicalSurv"y developed for a variety pRUS performssimulationof daily streamflows drologicpro""rr"r. fne USGSin the from is available It of precipitation,climate,and land usecombinations.
I

L.

568

CHAPTER23

MODELS SIMULATION CONTINUOUS

During storms,the model givesoutput on PC, riinicomputeroor mainframeformat.12 any prescribedtime interval. Betweenstorm periods,the model tracks soil moisture zones on a daily basis until the next storm interval. and other storage/depletion Streamflowis output as mean daily flow rates. of is utilized in PRMS.The smallestsubdivision approach A lumped-parameter to behaveas a the study watershedis a hydrologic responseunit that is assumed. HRUs in most areasof hydrologicelement.The USGS has delineated homogeneous and reservoirs, R3M model,the streams,storm sewers, the United States.Like the D. detentionponds in the watershedare modeledas nodes and interconnectinglinks' kinematic-wave Hydrographrouting in channelsand reservoirsis accomplished.by methods,respectively. and storage-indication

in Rural for WaterResources ARS Simulator Basins(SWRRB)

Through a cooperativeprogram with TexasA&M University, the Agricultural Research Service, U.S. Department of Agriculture, developeda continuous daily streamflow simulation ptog.a- for use in modeling ungaugedagricultural areas.t3 This FORTRAN 77 water budgermodel is availablefrom the ARS in PC format. Its focusedon a model that would allow the user to predict impacts of development practicessuchas crop rotation, fall plowing, urbanmanagement variouswatershed ization, conservation tillage, terracing, fallowing, and floodwater detention on monthly and annual water and sedimentyields from rural basins. and for the total basin, are Sedimentyields for each of the subwatersheds, hydrolThe methoduseswatershed usingthe universalsoil lossequation.la calculated ogy outputsfrom the rainfall-runoff portion to estimatesedimentyield, using inputs of runoff volume,peak flows,soil type anderodibility, crop types,erosionmanagment factors,and watershedslopeand length. style.Oncerainfall or snoware modeledin a lumped-parameter The subbasins are linked as shownin Fig. 23.11 algorithms fall recordsare input, physical-process sedimentprolosses, for solar radiation, snowmelt,surfacerunoff, ET, conveyance percolation,and soil from watersurfaces, evqporation ductionfrom individual storms, basedon crop types,temperCrop productionis alsocalculated moistureaccounting. ature, consumptiveuse of water, and irrigation practices. by the SCS curve are estimated for eachsubwatershed Hydrologicabstractions number(CN) method(Chapter4, Section4.9). Soil waterbudgetingis performedby adding the net moistureinput to the soil profile from precipitation after subtracting from soils),percolauseby the cropsand evaporation direct runoff, ET (consumptive runoff is Direct surface tion, and retu{n flow (groundwaterflow back to the streams). set equal to the net rain from the CN method. For input to the sedimentyield component, a peak flow rate for individual storms is estimatedusing a modified rational method (Chapter 15). Snowmelt is calculatedby the degree-daymethod ' in Chapter14. described

ExpertSystems
of expert system(ES) analysisare leadingto developmenl Recenttrends in systems planning and designof water in use that rely on artificial intelligencefor techniques
-, .oo^,,r^^o -r^io;id--7[mrfian-.m h-c-ciw?i'- ii-FiYmqfinn nhfqined frnm extencive

MODELS SIMULATION STREAMFLOW 29.1 CONTINUOUS

C*.')
' /READTNPUTDATA/
INITIALIZE PARAMETERS

READ OR GENERATE PRECIPITATION AND MAX/MIN TEMPERATURE GENERATE SOLAR RADIATION

CoMPI.ITE SOIL TEMPERATTTRE COMPUTE SNOWFALL AND SNOWMELT

ACE

COMPUTEPEAK RATE TRANSMISSIONLOSSES. OUTI-ET SBDT,M}{T YIEIO ENO ROUTE SEDIMENT TO BASIN

COMPUTEANDPRINT FINALBASIN STATISTICS

Figure 23.11 Flowchart of the SWRRB hydrologicpr6""*, algorithms.(After Arnold't3)

570

CHAPTER23

MODELS SIMULATION CONTINUOUS

interviJws of one or more experts in some field. The computer can then make "decisions"in muchthe sameway asthe,experts, applyingtheir judgment and experithe expert systemmodel. through to others ence and making theseavailable simulation,incorperform continuous that those Streamflowmodels,especially judgment. and usersof Developers parameters require considerable that porateinput in assigning of experience decades have accumulated ahe watershed models judgment by an extracted can be and parameters. Their experience coefflcientsand mode'. system expert build an questions to of hundreds interview processinvolving about uncertainties but alsoaddresses direct answers The modelnot only incorporates promise.ls considerable show technique modeling with this each.Early applicdLtions have the potential to be In addition to streamflowsimulation, expert systems of dams,resersystems basin river of complex useful in the designand management for such Operations structures. control flood voirs,powerplants,diversioncanals,and professionals' of dozens by decisions systemsinvolve independentand collective contact with numerousother conTheseexpertsare normally in radio or telephone from theseteams,the developed be could If ES data trollers and decision-makers. implementingexpert of prime incentive A exists. potential for improvedmanagement profesexperienced of insights capturing involves systems iystemsin waterresources positions. sionalsbefore they retire or move into other

STUDIES MODEL SIMULATION 23.2 CONTINUOUS

applicationsof the Kentuckyversion in detail two independent This sectiondescribes in Kentucky and Nebraska.Results basins small to model watershed of the Stanford (CC) in Kentuckyand by watershed Creek Cave the in modeling Clarkel6 obtainedby (BBC) in Nebraska watershed Creek Big Bordeaux for the KWM using the authorslT good relative$ having watersheds homogeneous small, Both are are compared. simultaneously described are studies case The two runoff. precipitation and recordsof input data requiredto develop to showhow different analystsdealt with the decisions parameters. and

Selectionof WatershedSize and Study Period

size and time period to be subarea Severalguidelinesexist for selectinga watershed watersheds homogeneous small, relatively use of modeledin a simulation study.The difficulty any minimize to recommended is or subdivisionsof larger watersheds also This areas. larger over precipitation in causedby ignoring spatial variations soil types, as soil such characteristics watershed minimizes the'effects of lumping entire the parameters representing into single uses andland profiles,imperviousareas, drainage an upper limit of 25 fii12for study-watershed catchment.Ross suggests areas.18 sizearisesfrom the fact that few strearns One difficulty in restrictingwatershed gauged, and refiable streamflow data are continuously for small drainageareasare it was one because was selected BBC watershed the difficult to obtain. For example, waterCreek The Cave records. sufficient with in Nebraska of few small watersheds criteria'1e the following of the basis on shedwas selegted

MODELSTUDIES SIMULATION 2g.2 CONTINUOUS

571

minimum of 10 yearsof continuousrunoff recordsin order to establish the existingrainfall-runoff relation. 2. A drainas" ur"u of less than 5 mi2 so as to be reptesentativeof small drainage6asinsfor which better runoff coefficientsare needed' 3. A locaiion in closeproximity to a rain guagefor which hour$ precipitation dataarc available. under study. 4. The availability of soil surveysfor the watershed 1.

'A

DataSources
and Hourly precipitation data were availableat sites L20 fiil from the CC watershed were data runoff 8.0 mi fromthe BBC gaugingstation.Soil surveyrecordsand Daily pan evaporationdata were availableapproxiavailablefor both watersheds. mately 30 mi south of the BBC watershedand 25 mi south of the CC watershed. d and9 .22 mi2for the BBC watershed Drainageareasof 2.53 mi2for the CC watershe for the maps.Input parameters were found from U.S. GeologicalSurveyquadrangle Numerical 23.7. Stanfordand Kentuckyversionsare comparedand definedin Table names,aretabulatedfor both watersheds usingthe Stanfordversionparameter values, in detail in the following sections. in Table 23.8. Each parameteris described

Time-Area HistogramData
The time-area histograms for the BBC and CC watershedsare developedin for the Traveltimes and times of concentration Figs.23.12 and23.Ii, respectively. larger (for watersheds equation Kirpich the from found model are Stanfordwatershed s1 than 15 acres),ie

a : ooo78(#)"

(23.2s)

where T" : the time of concentration(min) L : the horizontal projection of the channellength from the most distant point to the basin outlet (ft) : S the slopebetweenthe two points the time-area histogramfor Big BordeauxCreek, 30 points and In developing times (min) were plottedas shownin Fig. 23.12.The dashed correspondingi.auet by linear (lines of equal travel time to the basin outlet) were constructed isochrones were paris isochrones of interpolation between the plotted points. Areas between Creek for Cave from planimeteiing.To contrast,the time-area histogram determined was construcrcdfas shown in Fig. 23.13, by assumingthat the flow velocity was velocity obtainedby dividing f streamflow constant"u"ryrli"r", equalto the average into the channel length Z. This procedure simply places all points along each to the basin outlet. isochroneat equal distances

WatershedParameters
The watersheddrainage area (AREA), the impervious fraction of the watershed surfacedraining directly into the stream (A), the fractional streamand lake surface ground slopeof overlandflow perpendicularto the contours area(ETL), the average

572

CHAPTER23

MODELS SIMULATION CONTINUOUS FOR KENTUCKYAND PARAMETERNAME COMPARISONS STANFORDWATERSHEDMODELS

TABLE 2i,7

Parametername

Kentucky verston
NCTRI CTRI RMPF RGPMB AREA FIMP FWTR VINTMR BUZC SUZC LZC ETLF SUBWF GWETF SIAC BMIR BIVF

Stanford verston

descriPtion Parameter Integernumber of elementsin the time-area histograq Time-area histogramordinate data are to be printed (cfs) DischargevalueLelow which no synthesized a distant statton for data precipitation for factor Multiplication drainagearea (mi2) Watershed areathat is impervious Fraction of watershed area in lakes or swamps Fraction of watershed Maximum rate of vegetativeinterceptionfor a dry watershed(in'/hr) capacity storage surface lndex for estimating Ind,exfor estimatingsoil surfacestoragecapacityduring summers Index of moisture storagein soil profile abovewater table (in') parameterfor lower zone soil moisture Evapotranspiration flow from the basin Subsurface by phreatophytes Groundwaterevapotranspiration Factorvarying infiltration by season Index of infiltration rate Index of rate and quantity of water enteringinterflow Averagebasin ground slope (ftlft) Averageoverland flow distance(ft) coeff,cientfor overland flow Manning roughness coefficientfor flow over impervious areas Manning roughness constant Daily intelflow recession Streamflowrouting parameterfor 1owflows Streamflowrouting parameterfor flood flows Index capacity of existingchannel,bank-full (cfs) routing of flow for nonlinear Exponent Daily baseflow nonlinearrecessionadjustmentfactor Daily baseflow recessionconstant (in.) storage Indexof groundwater storageat beginningof year (in') depression and Depth of interception Cuirent equivalentdepth of moisturein the soil profile (in') Cunent value of BFNLR Interflow storage Control options for input, output, and internal branching Title of computer simulation output, alphanumericinput Mean daily diversion into or out of the basin (cfs)

Z C MINH K1 AREA ETL EPXM CX EDF LZSN K3 K24L K24EL EF CB CY

SS L NN NNU IRC KSC KSF CHCAP RFC

oFss
OFSL OFMN OFMNIS IFRC CSRX FSRX CHCAP EXQPV BFNLR BFRC GWS UZS LZS BENX IFS CONOPT

t<r/24 KK24
GWS UZS LZS

DKN

aoa
DIV

Qaa
DIV

(SS),and the nban length of overlandflow (r) for the Big Bordeauxcreek watershed and lengths measuredfrom 7.5-min series were determinedfrom-areas,elevations, parameter usGS topographicmaps (see Table 23.8 for BBC values).Other BBC : stream of the product the gri : half ETL from 0.0d5 (determined values utl : Fig '23'14 (determined from ftlft 0.088 length and channelwidth at the outlet), SS interseggridline each at contours 20-ft valuesbetween u, ih- -"un of'140 measured of l40lengths from Fig.23.14 as the average 183.2ft (determined tion), and L: points to the intersection gridline from lines measuredperpendicularto contour

FOR BIG BORDEAUXCREEK TABLE 23.8 SUMMARYOF INPUT PARAMETERS WATERSHEDS CREEK AND CAVE

Parameter name
TCONC TINC Z C

Description Time of concentration Routing interval Number of elementsin the time-area histogram Time-area histogramordinates

a^ BBC value(s) Units value(s)

AREA

^ ETL

ss
L

CHCAP IRC KK24 KSC KSF MINH K1 NN NNU EMiN EPXM CX EDF LZSN K3 K24L K24EL EF CB CY KV24 SGW UZS LZS GWS VOLUME EVCR

drainagearea Watershed Impervious fraction of the watershedsurface stream and lake surfacearea as fraction of Watershed 0.005 watershedarea 0.088 ft/ft Averageoverland flow ground sloPe ft 183.2 flow of overland Averagelength cfs 39 Bank-full flow in channelat gaugingstation 0.485 constant recession Daily interflow 0.977 \ constant Daily baseflow recession 0.989 low flows for parameter Streamflowrouting 0.989 Streamflowrouting parameterfor flood flows cfs 1.0 printed be to rate Minimum hourly flow 1.0 Precipitation adjustmentfactor for distant gauge 0.3'7 coefficientfor overlandflow Manning roughness 0.0i3 coefficientfor impervious areas Manning roughness 0.5 seasons by infiltration Factorfor varying 0.15 in./hr Maximum interceptionrate for dry watershed 0.80 index capacitY Surfacestorage 1.10 capacity moisturesiorage Soil surface ln. Il./d index storage profile moisture Soil 0.28 Soil evaporationparameter 0.0 Index of inflow to deepinactive groundwater 0.0 Fraction of watershedareain phreatophytes 1.0 rates infiltration for seasonal Factor allowing 0.'75 Factorcontrolling infiltration rates 3.0 Index controlling water enteringinterflow 1.0 Parameterfor allowing nonlinear recesslon in. 0.1 parameter volume storage Groundwater in. 0.0 Equivalentdepth of upper zone storage in. 7.0 storage zone of lower Equivalentdepth in. 0.2 moisture conditions Index of antecedent acre-ft 0.0 storage in swamP water Volume of 0.911 Jan. Monthly evaporationpan coefficients 0.911 Feb. 0.911 Mar. 0.911 Apr. 0.552 May 0.677 June 0.654 July 0.651 Aug. 0.642 Sep. 0.911 Oct. 0.911 Nov. 0.911 Dec.

min 105 min 15 7 0.129 0.158 0.221 , 0.151 0.126 0.145 0.070 mi2 9.22 0.0

60 i5
4

0.18 0.29 0.31 0.22

2.53 0.0 0.0 0.075 300.0 40 0.75 0.94 0.90 0.90 0.2 1.0 0.10 0.015 0.10 0.90 1.25 4:85 0.25 0.0 0.0 0.15 0.65 3.50 0.99

574

MODELS SIMULATION 23 CONTINUOUS CHAPTER 0.3

I

15-min isochrone

0.2

"15 rs 30 4s 60 90 105 Time(min)

-l

3li

Time.area

a5-min {

Ll--:Ja
68o -l \t

$.(-t

15 30 45 60 75 90 105 Time (min) 30-minisochrone

45-min isochrone

60-min isochrone N

--.1-^

t-\-75-minisochrone
90-min isochrone
n . , u miles r IJJJJ-JJJJJ 1 l

for Big BordeauxCreek' Figure 23.12 Time-areahistogramdevelopment The valuesrepresenttravel times (min.) to the outlet.

overlandflow distancescan also be estimatedas the nearestchannel).The average

resulting in an estimateof the bank-full BBC flood of 39 cfs. Another techniquefor rating curveif th CHCAP involvesthe useof the stream-gauging-station determining maps. from topographic gaugeheightfor bank-full flow is known or can be estimated the cross of from a hydraulic analysis thJCave Creekvalueof 40 cfs wasdetermined selected to is spctigl and profile of the main channel.As a rule of thumb, CHCAP )

.../

3l.ovo _

Time-arca histogtam Time (min) Contributing area (Vo)

0 -1 5 15-30 30-45 45-60

18.0 29.0 31.0 22.O

Figure 23.13 Derivation of the time-area histogramfor thecave creek (After Clarke'16) Dashedlines representisochrones. Watershed.

0
l r r

miles
r r r r t

1
r l

0 1000
f""t

h -w" -q
:h &"{

, TGrid interval:fm=tlZOft
\I

Figure 23.14 Grid overlay for determinationof the mean overland - slopeand distancefor Big BordeauxCreek.

576

CHAPTER23

MODELS SIMULATION CONTINUOUS

twice the value of producewell-definedoverbank and floodplain flow if Q exceeds is lessthan half of whenever assumed are flows Q low, shallow Similarly, bffCAp. CHCAP.

StreamflowRecessionand Routing Parameters

The rate at which the model allows water to passthrough the upper soil zonesto the channelsis controlled by the daily interflow recessionconstant(IRC)' A graphical semilogarithmictechniqueof hydrographanalysisdevelopedby Barnes is used to constantfor The determinationof the interflow recession estimatethis parameter:21 plot of is a semilogarithmic which Big BordeauxCreek is illustrated in Fig. 23.1.5,

I o.s E o.+
o

K 0.3
E o
d

V'L

p9

o.ro
q

lnterflow Llog Q = -0;725 Lt = 3.2 days K=0.485=IRC

& 0.04 0.03

o.os S o

1 4 1 6 1 8 2 0 2 2 2 4 2 6 2 8 3 0 1 3

(Ju1Y-August' 1969) Date Figure 23.15 Determination of the interflow recessionconstant (fnC) for th9 Big BordeauxCreekflood eventof July 20, 1969.The K: IRC, C' : QoK'andr: 1day. determinationof

MODEL STUDIES SIMULATION 2g.2 CONTINUOUS

577

!2
F

o b0

5
logQ,=1ogQ6+tlogK logK= LlogQlLt

d o o

f f i z o

' 3 " ' s1 9 1 1 3

1969) (July-August, Date Figure 23.16 Determination of the base flow recession constant (K"K24) for the Big Bordeaux Creek flood event of July 20, 1969' The determinationof K : KK24.

578

MODELS SIMULATION 23 CONTINUOUS CHAPTER

'

alternative method for determining Kl<24 using data collected several days after an severalToodpeaksis illustratedfor CaveCreek in Fig. 23.I7. The line represents points. envelopedrawn to the right of the data Two streamflowrouting parameteisare usedby the sinlulation model to route to the point of interest for the basin (in this caseto the stream inflow hydrographs garglng;tatiin ror compa;isonwith measuredflows). The first parameter 1_519t_lt (CHCAP), thechannelaapacity usedinroutingonlyif channelflowsarelessthanhalf during flood flows for channeland floodplainstorage and the second(KSF) accounts for Big _B-ordeaux data Respective capacity). (flows greaterthan twice the channel Figs. 23.18 and in shown hydrographs runoff Creek were determinedusing the two point on inflection the Because respectively. Z3.lg,representinglow and flood flows, of reversal a which at the time represents portion of eachhydrograph the recession KSF KSC and parameters the then flow directionihrough the channelbanks occurs, representdaily recessionconstantsfor the water in storagein the channel. The is derived from a hydrologicrouting equationused in determiningboth parameters inflow and outflow during average in the equatingl/K timei the difference technique in the channelreach.As outflow the of the routing period-withthe time rate of change for Big_Bordeaux parameters routing flow shownin the figures,the low flow and flood for Cave observed was result a similar Creek are identical. Table 23.8 showsthat Creek.

and Data Parameters Hydrologic

severalhydrologicparameters characteristics, watershed In addition to the described and an impressiveamount of hydrologicdata are required as input to the simulation of hydrologicdata for daily evaporation,hourly program. bue to the excessive'bulk pr"-lpitution, and daily streamflowfot a 4-yearperiod in BBC and a 10-yearperiod

4.0
o
d

3.0

1.0

2.0 3.0 4,0 5.0 24hr later(cfs) Discharge

Figure 23.17 Determinationof KK24 for CaveCreek.(After Clarke.l6)

STUDIES 579 MODEL SIMULATION 23.2 CONTINUOUS

0.36 days t = routinginterval= 15min = 0.0104 0.32 o?o ,, ',/ --# = - 0'28at inflection Point 0.264 K = - f foi , = - : 6 6=0e43dals --

x S C =K _ o . s t = 0 . 9 8 9 K + 0.5t ,r'
In{lection

E o.rt
o P0

pomr

6 0.26
E

! b0

8 0.22
0.20
0.18 0.16 0.14
' Date (October 1968)

Figure23.1sDeterminationoflowflowstreamflowroutingparamecreek low flow eventof october3, tei(rsc) for the Big Bordeaux 1968. in CC, the actual hydrologicdata valueshavebeen compiled but'are not includedin this text. , usedasinput only in the BBC applicationis the Oneusefulhydrologicparameter is which no printing of simulateddischarges below gauging siation the at flow hourly of desired.For the initial simulation of runoff from Big Bordeaux Creek, a MINH in to minimize the printed output. This can easilybe decreased 1.0 cfs was selected sirnulationtrials" after the uncertaintiesin someof the other parameters successive reduced. havebeen A precipitation-weightingadjustmentparameter(K1), representingthe longterm ratio of av"rage precipitation over the basin to averageprecipitation at the might ue greateror lessthan 1.0 if the gaugewere locatedat any precipitationgaugp, were within for both watersheds di*tun." frornthJstudy basin.The recordinggauges 1.0. to set were parameters K1 and both proxirtrity, close pan coefficients(EVCR) are usedby the simulationproivlonthly-evaporation data datato lake evaporation gram asmultlpters in convJrtinginput pan-evaporation evaporation nearest The rates)' potential e.'upottuntpiration for determining iand also 20 mi southat the Box Butte Reservoir, pan for Big BordeauxCieet is approximately

580

CHAPTER23

CONTINUOUS SIMULATION MODELS

t = routing interval = 15 min = 0.0104 days

do =-t'' at lnrlectlon Polnt

al K=o a9 =0.873days *n - ).) du/dt KSF= K-0.5t = 0.989 K + O.st

'? !) o E

5.0
,r'

Inflection pomt

4.0
d o

oo 6
J.U

20 21 22 23 24 25 (March 1969) Date Figure 23.19 Determination of flood flow streamflow routing parameter (KSF)for theBig Bordeaux Creekfloodflow event of March17,1969. providing records of daily and monthly pan-evaporation depthsfor the months between and including May and September. Estimatesof corresponding monthly lake evaporationamountsat Box Butte Reservoirwere obtained from maps and charts developed by Shaffer.22 Valueslisted in Table23.8 are assumed to apply to eachyear of the simulation even though the simulation program allows changes from year to year. Manning's roughness parameters for flow over soil and impervioussurfaces are both required as input to the program. For the Big BordeauxCreek area,the initial estimates for qverlandflow (NN) and impervioussurfaceflow (NNU) were 0.37 and 0.013,representing coefficients for dense shrubberyand forestlitter for overlandflow and smoothconcletefor impervioussurface flow. The later n is significantonly if the fraction of impervious area (A) is nonzero. The Cave Creek analysisincorporated NN : 0.1Qfor light turf and NNU : 0.015for concrete pavement.

t6

I7

18

t9

ModelCalibration
Severalof the following parameters are determined by trial and adjustmentuntil the comparisonbetweenthe simulatedand recordedstreamflows is satisfactory. Guidelines-forestablishing initial valuesexistfor only a few of the parameters, whereas most

STUDIES 581 MODEL SIMULATION 2g,2 CONTINUOUS The BBC datain Table23'8 represent ranges. from suggested are initially determined initial, unmodified estimates;those shown for CC are the optimal result of many repetitive runs. (EMIN) rangesbetween0'1 and One factor for varying infiltration by seasons and simu1.0 and has been shownby Briggs to be signiflcantin matchingmeasured parameter are for estimatingthis no guidelines Because lated winter peak flow rates.23 Bordeaux midvalue of 0.5 was selectedfor Big presently u*ilubl", the suggested Creek. requireinitial input valuesthat are Severalsoil moistureandrouting paraineters improve successively data.Calibrationprocedures from available difficult to estimate parameters For Big BordeauxCreek,the by trial and adjustment. the initial estimates as follows. are estimates initial and EPXM-the ma*imum interception rate (in./hr) for a dry watershed. trial valuesof 0.10, 0.15, and 0.20 fot crawford and Linsley2suggest moderateforest covers,and heavyforest covers,respectively. grasslands, for the moderateforestcoveralongBig ifr" O.ts-in./hr valuewas selected in the CaveCreek study' 0.10 used Clarke Creek. Bordeaux CX-an index of the surfacecapacityto store water as interceptionand from 0.10to 1.65,and normallyranges This parameter storage. l depression the midvalu" of o.go was selectedfor Big Bordeaux creek although a greaternumber might be indicative of the forest cover. Clarke independently.arrivedat a final, similar value (0.90) for CaveCreek' the capacity,representing moisturestorage EDF-an index of soil-surface due months warmer during capacityavailable additional moisturestorage to vegetation.Dependingon the soil type, the index ranggsfrom 0.45 to 2.00. Sandysoilsiimilar to thosein the BBC areareadily give up moisture capacity.lnitia,l valuesof 1.10 storage resultingin increased to vegetation, for the Big Bordeaux and cave and 1.25 were independentlyselected Creek areas. index (in.) approximatelyequalto LZSN-a soil-profile moisturestorage table and below the ground water the above the volume of water stored parameter,inversely runoff-volume major is a parameter surface.This flow. The LZSN groundwater and yields, interflow, related to the basin from soil, ranges yield the of specific the porosity and on index,depending an as used be can rainfall annual mean plus the half 2.0 to 20.0,and 4 in. a use of By the rainfall' seasonal experiencing initial estimatein areas Chadron, the precipitation, annual in. of 15.55 lg3I-Ig52 averageof : 11.78in. for Big Nebraskl, precipitation station givesan initial LZSN BordeauxCreek. A similar analysiswas usedfor CaveCreek' that controlsthe rateof evapotranspiparameter K3-a soil evaporation The parameterrangesfrom 0.2 to zone. soil lower the from losses ration cover.As an initial on the type and extentof the vegetative 0.9 depending which agreed selected, was 0.28 in BBC, coYer forest light the for estimate by Crawford and Linsley.2 Also, K3 is with the estimatis suggested and _ approximatelyequal to ihe fraction of the basin covered by forest

582

CHAPTER23

MODELS SIMULATION CONTINUOUS

for barren ground, grassdep-rooted vegetation.Recommendations2 lands,andheavyforestsare'respectively,0'20'0'23'and0'30;0'25was oPtimal for the CaveCreek studY' controllinl K24L - a Parameter from active groundwater stori It also basinboundarY' drainage groundwaterthat Percolatesto K}4LPatameter canbe estimatr are small thesel0sses to be zerobecause waterlevels,or it is often assumed runoff' and of rainfall comparedto the magnitudes K24FjL-'the friLctionof the total watershedover which evapotranspirato occul at the potential rate' tion from groundwaterstorageis assumed quantity of vegetation significant a unless zero d is assume This pararneter seekphreaticwater, that Plants table. water drawswater directly from the phreatophytes' called are alfalfa, or suchas cedar or cottonwood EF-afactorrangingfrom0.lto4.0thatrelatesinfiltrationratesto evaporationrates.Thisparametersimplyallowsamolerapidinfiltration A normal starting value of 1'0 was during raim"t seasons. rate recovery "nig Boideaux Creek, whereasthe CaveCreek value was set for selected much lower. on the soil CB-an index that controls the rate of infiltration, depending permeabilityandthevolumeofmoisturethatcanbestoredinthesoil.A Big midvalue oi O.ZS(0.3-1,.2) was selectedfor the sandy soils around BordeauxCreek;asmallervalueof0.65wasoptimalfortheCaveCreek study. moisture cY-an index controlling the time distributionand quantitiesof moderately a enteringinterflow' This index rangesfrom 0'55 \" 4':' and the waterfor Big BordeauxCreekbecause high valueof 3.0 was selected moderately a shedcontainsmany pine needfemats. clarke also selected high CY for CaveCreek. KY24-adailybaseflowrecessionadjustmentfaclorusedtoproducea An initial value of 1'0 for Big simulated"oruilin"u, baseflow recession. base flow recessionfor the the Bordeaux creek was selectedbecause hydrographinFig.23.16islinear.Lateradjustmentsmightberequiredin matchingsimulatedand recordedbaseflow recessions' SGW-agroundwaterStorageincrement(in.),reflectingthefluctuations for Big in storaqelolume. Usually, an initial estimate(0.10 was selected trials; simulation several after adjusted and Bordeaui creek) is made parameters four following the and Thib in. 3.90 SGW rangesfrom 0.10 to were usedonly in the BBC studY' UZS-thecurrentvolume(in.)ofsoilsurfacemoistureasinterception anddepressionstorage'Becausethesimulationbeginsonoctoberlofthe as zero first calibration yeai the parametermay initially be designated unlessprecipitation occutt during the last few daysof September'

23.2 CONTINUOUS SIMULATION MODEL STUDIES

583

LZS-the current volume (in.) of soil moisturestorage betweenthe land surface and the watertable.Sixty percentof LZSN, or 7.0 in., wasselected to initiate the Big BordeauxCreek iimulation. GWS-the current groundwaterslopeindex (in.). This index providesan indication of antecedent moistureconditions.Suggested initial valuesfall between0.15 and 0.25, or the value assigned to SGW can be used.A midrange0.20 in. was selected for Big BordeauxCreek. VOLUME-the volume (acre-ft) assignedto swamp storageand dry groundrecharge, accounting for the runoffrequired to recharge swamps in late summer.Since no swampswere visible on the USGS 7.5-min topographicmaps,an initial valueof 0.0 acre-ft was selected for Big Bordeaux Creek.

HydrologicData
In addition to the parametervaluesof Table23.8, the Stanfordmodelrequiresa large volume of hydrologic data for each water year of the simulation. The following description of requried input data for each water year illustrates the data that were compiledand reducedto necessary input form for 4 wateryearsbeginningon October 1, 1968andendingon September 30,1972.Only the input for BBC is described. The input includesthe following data for eachwater year: 1. The new year identificationentry containsthe water year and the recorded annual streamflow.The valuesof the annual streamflowfor Big Bordeaux Creek are listed in Table 23.9. 2. A descriptionofthe streamgaugingsite. 3. The title to be appliedto the ordinatefor graphicalplotsof the simulatedand recordedrunoff hydrograph;namely,the daily average flow rate (cfs). 4. Pan-evaporation data, read as 365 or 366 single entries containing daily evaporationamounts(in.). 5. The monthly evaporationpan coefficients,comprising the data listed in Table 23.8, beginningwith October. (average 6. Recorded daily streamflows flow for the day,cfs)readas 365or 366 entriesfor October I through September 30. 7. Hour$ rainfall data, read for each water year. Two entries per day, each containingan identificationofthe gaugeanddate,areusedto providehourly TABLE 23,9 ANNUAL BIGBORDEAUX CREEK STREAMFLOWS
Water years Recorded annual streamflow (acre-ft)

1968*1969 1969-1970 r970*1971 r97t-1972

434.0 465.4 296.4

584

23 CONTINUOUS SIMUI.ATIoN MoDELS cHAPTER depthsin inchesbefore noon on the first and after noon on the second.Valuesare precipitation.Because of the variablenumrequiredonly for half-daysexperiencing ber of possible rainfall values,a sentinelentry with the year setequalto 2001 is placed at the end of the data, indicatingthat all the precipitation data for the wateryear has beenread.

Version Outputfrom the Kentucky

are selected, a Dependingon which optimal input, output, and branchingparameters variety of output data are availablefrom the Kentucky version, including plotted graphs of measuredand synthesizeddaily streamflow rates. Options include the following: . 1. A table of synthesized average daily streamflowrates (cfs). daily flow rates. 2. A table of monthly and annual totals of synthesized monthly and annualtotals ofrunoffin equivalentinchesover 3. Synthesized the watershed. 4. Svnthesized monthlv and annual interflow amounts (in.) over the watershed. 5. Svnthesized monthlv and annual baseflow amounts(in.) over the water-

s[eo.
streamflowrunoff from the watershedfor the 6. The volume of synthesized entire water year (acre-ft). 7. A summationof all the recordeddaily streamflowrates(cfs)for eachmonth and year. 8. The recordedannual total of runoff (in.) over the watershed. 9. The recordedvolume of runoff from Novemberthrough March (in.) over the watershed. 10. The amountof synthesized snowfrom Novemberthrough March (in.) over the watershed. 11. The volume of the recordedannual streamflow(acre-ft). 12. The sum of the recorded precipitation for each month and for the year. (in.). 13. The synthesized monthly and annual totals of evapotranspiration 14. The monthly and annualrecordedlake evaporationamounts(in.). (in.). moisturestorage 15. End-of-the-monthlevelsof UZS, the current surface 16. End-of-the-month levels of LZS, the current soil moisture storage(in.). 17. End-of-the-month values of SGW the current groundwater storage fluctuation(in.). 18. An annual moisture balance (in.), which representsthe moisture not accountedfor within the program.This is illustratedin the CaveCreekoutput.

GaveCreekModelCalibration
Synthetic and actual flow rates at the Cave Creek gauging station are shown for a singleday in Fig. 23.20. Other typical output for portions of one water year of the an hour-byis presented in Tables23.10and23.ll. The former provides simulation 23.8. Note of MINH, Table value of the specified hour listing of all flow ratesin excess

SUMMARY ^ t20 Syr rthetic -r 100

585

a 8 0 {, o s 6 0 E 4 0

t/

l0

12 Time (hr)

I4

16

18

for and recorded hydrographs Figurp 23.20 Comparison of synthesized (After C1arke.t6.1 Cave Creek. MINH that the streamflowwas zero from Octoberthrough Decemberand exceeded only during two daysin January. in the 18 items of outpuf Table23.ll containsmostof the informationdescribed for the Kentuckyversion.The daily flows are followedby the syntheticand recorded monthly totals, monthly interflow and base flow amounts, monthly precipitation in'the totals, monthly actual and potential ET amounts,and end-of-month storages soil profile and groundwaterzones in inches. Of particular interest is the annual summaryin the lower right. Of the 37.5 in. of precipitation,23.8 in. went to ET, and the remaining2.1 in. recharged 11.6in. ran off or was discharged from storage, for during the year was 0.0844 in. the soil profile. Moisture not accounted

of ModelResponse to Parameter Changes Sensitivity

One interestingand useful aspectof simulation is the easewith which changesin watershedparameterscan be evaluated.Clarke testedthe sensitivity of KWM by rangeswhile holding varying severalof the parameters in Table23.8 over reasonable all other parameters constant.The resultsof his analysisfor CaveCreek on a typical were taken from day in March are summarizedin Table 23.12. Theseobservations graphssuchasFigs.23.2l and23.22,which illustratethe sensitivity of flood magnitude and timing to changesin Z and KSC. These results and the summary in Table 23.12 are applicableonly to the CC watershedand should not be viewed as generally applicable.

r Summary
precipitation recordsare available,one of the most effective If actual or synthesized flowsunderchanging meansof analyzinghistoricalflowsor evaluatingfuture possible land use patterns is through any of the continuousstreamflow simulation models described in this chapter.Hydrologicproblemsand applicationsthat can be analyzed

=
u a/) u r m
@ F S N

r <

Yi

rt

=
r
F O @ s
h N

z g
U'

z
t/) q

r'o

r s

r o

6 0

? o
r a
t!
@ F

s o

u
F

r @
@ @

r <

o
E

il

o t!
z

s
6
tr
l! a @ r

6 o N

- Y n u rlf
T

trul

o r a @
@

u
F

e ! q

oo
6-> z < d o
cc lL >ul
+ u llj

F qt

o o

@ r
F @

U)
L

u, r

N T + a i

o
L

m 6

U'
L

>z
Y Y F

O-r
= O

=
iE o U

=
J 4

=
E U r

=
J

o r

N Q

=
J

E> 9o 6H
) a

u u N T ll

c.l \ + m

N 6
D

z
CE

2
F

U
N T

ll

<)c r ( tu g < f =
o

:z u
u u

z x u J
u

=
x

o 6 @ o

+ei

= x =

< L

<o-

<o-

> = <o-

at N ut o

z >z U
u u E
o

(-)

u
h N N

u
4

U
L

@ o o

o
N

u U' E] k

O N

F S

+ o

N O @ F @ <

s o
O N

o r
N F

o @
@ N

r o @ o s(9

=
r

ri
N

qt
O

z.
U)

F @ @ +

+c t N

rtri

N @ o q i

z.
U) u \z
N

+ o
@ N

o r

bts O N

a N @ N F

c? r?

o r o o

I o o U u

N @ o @

<t=
6io F 5
4 F O N N
N

@
F

qq? < @

o r e @ + @

cq aj

N +

+ s 6i ci r D

r < f @ O N

oq\
N N @ -

\ a

o +

k =
o? e?
@ r o o d o i F @
@ o t s o O N

c!q o o
N -

qo?
@ h

< o
F

< o

< N O N

a q

9 O

r N

@ o

O F @ld)

ddi

=
CL

q q b s

o r

cqa < 6
6 6

r o

o
N

-tO

d -

o o

o < Ei+
@

u
F

o q O F

ctF @ h

< i o

qq o? + o
N F

u
r F @ q

o @ o m

d + u !z CE
J

v) r

o O N

U) r

\ q
F O

(tt r

<

tn
g

< f s

o? oj s e
N N

v)
L

@ t l

e? 6{ O F

CJ''

Y F

!z

rt+

F M

O N

F N N < f m

o @ @ o N N

qo{ rc9

t\

- h < f o

z
q)

r Y

d
F

=
iE - q 6 <
e N

il

O N

s
8
o
I (D N

z
v u u u

o r

@ o ri ct N N

N T

r +

o?F

- f O

E
u < L

=
x

>=
<o-

= >
< L

= =
<o-

x <o<o-

= =
x

!) 't (J

z. :z u
u r U
q

ul o

E61

'

H<
6 = > z o <
U
T
q

cn

=
Y F F Z.L U U YCn @ o

o A A
Q

a a

3-

a
u

? z
o a

- o

@ @

il>

F <

tr
lJ-

@ r o o

@ +

< m

o o

N N

F l

f (E

@ r

r @ o o

t s o

N F

O O

o o

r @

O N

m F a r

M N o o N O

r o r @
N F

o F @

O F @ F

6 e

:)

U) ]U TE l r q)

o @

@ F

ai 6i

r N N O

+ o

9 r : o
I > > a Ful
U Y
J

o o

o r

@ A

F F

o r

+ o

+ N

=E
O(r zlu

>- (-.) - U

3o
z.
u :z> -o 1 Z
F

eid

1k
-= { a OIE -r9 <= > a F
C\| IJJ

z.

o o

o ci et

?e
dt

E :z u r cc u
O

z . >< ir
E u

e o

o o

o o

o
o

o o

z. u>u < u
N O

+ 6

r @

N N

O N

a B u) t)

CD L)

u u

(N

H =
u

a = > z
o <
U' u

b z

z z

L)

tr

u v, a)

o
F f

U) u

U) u

v) v) U U

v, u

3 z

< =

z.

z.

z. z.

E =
F Z u Y, F L u Vt

m r r

+ o F O
@ F

@ 6

<t F

r
N

@ r

<f o

o @ N N

- a

+ =

6 N N

o h
N F O N

o o

o o O J

N O

O N O N O N

+ o
N O

+ I

? z
U

r + o
N

r r
F

o o o r O N

+ o
N O

F >

tr< =
cc r @ r
F N

+ N N O

@ F @

(Q o

F O < i e
O F

m + o o +
@ F

6 F

F O

F @ N F N N F O N

o
o
N

N O N S r N

r o-

o +

@ F

o + o $ + o

o
N

o o *

o m

o o

r @ d d
F

o r
@

r
E

6 N

E N + 6

r r

F @ @ N O F

o o

z.
N q F

o r
F O

@ h

o r o +

r
N

o + @ d6i

o + + a

:z z

-H 3o
z u v > z. z.
F

N
Q

r
o

F F O

N O F

o o

r o
N

r r
F O

^cn 6 -

'2
F

z .
u U u

x5
u z
Y

o +
o

e o o

@ N

o q
@ e o

HB
- o t 4 U

v) u)
N N t/) u F

u>u < o u

+ D
N N

r @ N N

=
N O

q
L

u
c) u

z z.
L

=
@

z.
u F

Ct) 7 (/j= N ;

zc z)

Q q

U) U a
p 4

z + > o ant

= d

z.
6

E
- X ; i - t

Hg esBE.g Es E B B - A '
ut E
rj'tr
d o O

U U E

6 d d 6

A22:EAfr22EEEA

t55t
A a

I g 3 .bE 83 3 t s . ; e I9E 0! h h'd boF EE

o
tll

u = - 6 U g o o o = o : : &&,&thl,E'ji/oZqMrh

o o
E E ! ! ! - -

I I I I93'ts

! E ! E

333E SKf r " , $$S$F

I I I I I

o o IU

AAAAAi!;EAAAA
ttErc o o 0 ) o ! ,r

(E

o) (t>

(u
o

|r
U)

tr
lll

.EQ,ae ! ilf H
o c (t E

t-

ti vj I I +6Nati i r i #; t !> Q 'Ai c |i N | c . l l -6 -6 - 6- 66- tr e ' o . A o o < -

E^,3E

ut
E. n q Y fuJ

r,J i dZZZZ-Z=a=

oH
> lrJ Y < 4 A
o o o o IE

9r. DE

8 6 z >

E (u
N I N

o o

c,

E
6 o

. 7

tu o
F

= Y

23.2 CONTINUOUS SIMUI-ATION MODEL STUDTES

591

6000 5000

4000

5 rooo
2000 1000 0
Time (hr)

Figure 23.21 Sensitivityof modelresponse to the length-of-overland-flow parameter.(After Clarke.16)

5000 ^ 4000

KSF = 0.85

5 :ooo
2000 1000 0
10

I
12
Time (hr)

\ \
S F = (t.99-

t4

16

18

20

22

Figwe 23.22 Sensitivity of model response to thechannel routingparameter. (After Clarke.16) usingcontinuous modelingincludewatershed yield studies, reservoirdesignand operation studies,sedimentyield estimatingfor reservoirdesignor analysisof impactsof erosion controls on water quality, water supply studiesfor municipal, industrial or agricultural demands, litigation over impactsof wellfieldsor direct diversions, determination of hydropower production potential, evaluationsof flows that will pass through critical in*stream or riparian habitat reachesof a stream,identification of flows that will be availablefor recreationalusesof a stream,and, amongnumerous other applications,analysisof water quantity and quality impacts of removingdams or rnaking other major upstreamchanges. The modelsdescribed in this chapter,and other similar continuoussimulation codes,are availablefrom the federal or state agencies that originatedthe code, or from numerouspublic outlets or vendorswho havebeen authorizedto distributethe software.

592

MODELS SIMULATION 23 CONTINUOUS CHAPTER

Problems
percentages the block diagram ofFig. 23.2 and plot approximate period. Reconstruct io show,for averageconditions,how the rain would be distributed(a) initially and (b) after 30 days. rain at afate of 3.0 in.i hr for 60 min' The 23.2. A sloping,concreteparking lot experiences ftlft. If the water detentionon the lot is ft deepand has a slopeof 0.000-1 tot is SOO for 1 ft overlandflow hydrograph complete zero at the start of the storm,calculatethe continue and interval routing 5-min Use a equations. the SWM-IV of width using computationsuntil all the detainedwater is discharged. 23.3. Calculatethe SWM-IV overlandflow time-to-equilibrium for the lot of Problem23'2 and compareit with the Kirpich time of concentrationfor the lot. Should thesebe equal? 23.4. Cornpare,by listing traits and capabilitesof each,the SWM-IV with its more sophisticated offspring HSP and HSPF. 23.5. Discussthe primary differencesamong the four versionsof the Stanfordwatershed in this chaPter. model described 23.6. Verify Eqs. 23.23 and 23.24by starting from Eqs. 13.4 and 13 '33' "typical" 23.7. Discuss the watershedbehavior that is depicted in Fig. 23.7. Is this a watershed? betweenthe two U.S. Departmentof Agriculture continuous 23.8. Comparethe differences that would be the applications simuiationmodels,USDAHL and SWRRB,and discuss best suitedto each. in 23.9. Review the differencesbetweenwater budget and simulation models discussed here described models simulation which of the continuous Chapter2l anddetermine could be usedto perform water budgetcalculations. 23.10. For what applicationsmight the following be best suited? API model USDAHL HSPF PRMS SWRRB four differby your instructor,describe 23.1r. For the continuoussimulationmodel selected ent types of problemsthat could be analyzedif you were given the ful1, calibrated model. h 3-in. rain in a 10-day in your localereceives that a 30-mi2rural watershed 23.r. Assume

REFERENCES
"ContinuousHydrographSynthesis with an and J. C. Monro, 1 . W. T. Sittner, C. E. Schauss, 1007'1022(1969)' 5(5), Res' API-Type Hydrologic Model," WaterResources "Digital Simulationin Hydrology:StanfordWaterL. N. H. diawford and R. K. Linsley,Jr., shedModel IV," Departmentof Civil Engineering,Stanford University, Tech. Rep. No.
J.

3 9 .J u l y 1 9 6 6 . "An Evaluationof Relationship and Watershed BetweenStreamflowPatterns L. D. James, Resources Water 36, No' Rep. Research OPSET," of the Use Through Characteristics Institute, University of Kentucky,Lexington, 1970'

Chapter24

Simulation Single-Event Models

r Prologue
The purposeof this chaPteris to: . Describehow storm event models are structuredand how they are used to for singlestorms. simulatedirect runoff hydrographs models(note . Describethe five most widely usedfederal agencysingle-event describedin are models simulation runoff urban popular single-event that 25). Chapter ' Provide a detailedcasestudy using one of the models,HEC- 1' . Introducethe emergingtechnologyof storm surgemodeling-the simulation of hydraulic surgesresulting from wind energy acting on the ocean surface. high-intensityrainfall events. by short-duration, floodsare caused Many severe and shortly following these during runoff simulates model watershed A single-event in the peak interested normally are models single-event Users of rain events. discrete flow rate, or the entire direct runoff hydrographif timing or volume of runoff is modelssimulatethe rainfall-runoff processand makeno special Single-event needed. for the rest of the hydrologic cycle. Few, if any, simulate soil to account effort occurring beinterflow, baseflow, or other processes moisture,evapotranspiration, rainfall events. discrete tween that are to studiesof watersheds in this chapterare applicable Modelsdescribed that are allowed,but for watersheds primarily rural in makeup.Urbanizedsubareas in described models continuous and single-event the principally urbanized, ire by created by surges induced is that flooding Coastal applicable. 2i aremore Chapter wind action on the ocean surfaceis modeledby a different class of single-event in Section24'3. described models,

SIMULATION EVENT 24.1 STORM

Event simulation model structurescloselyimitate the rainfall and runoff processes suchasunit-hydrograph approaches, Lumpedparameter in earlierchapters. developed methods,are generally incorporated even though some use distributed parameter

SIMULATION595 EVENT 24,1 STORM begins simulationstudies for implementingmost single-event Preparation techniques.. 24'1' inFig' illustrated as subbasins subdivisioninto homogeneous with a watershed a downstream in subbasln upstream Computationsproceed from the most remote direction, modelfor a typical basin(Fig. 24.I),the runoff hydrographs In any single-event and then routed and A, B, . . . , E are computedindependently, for eachof subbasins throughhydrographs design to obtain (called nodes) cornbinedat appropriatepoints storm the applies then storm; the for parameters out the basin.The modelieads input storm the from resulting hydrograph the Bf computes to the first upstreamsubbasin, A; combinesthe two comevent; repeatsthe hydrographcomputationfor subbasin by conventional hydrograph the routes into a singlehydrograph; puted'hydrographs it is combined R, where reservoir of end t[rough reachC to the upstream iechniques procedure the on through so and C; with the compuied hydrographfor subbasin inFig.24.l. detailed are most often determinedusing unitHydrographcomputationsfor subbasins The precipitation hyetographis 24.2. hydrogiaph proceduresas illustrated in Fig. leaving are abstracted, input iniiormly overthe subbasinarea,andptecipitationlosses the with 12) (see Chapter un "*..r, prelipitation hyetographthat is convoluted

Outlet (point of concentration)

reservoirsites,damagecenters' 1. Subdividebasinto accommodate divides,gaugingstations' subsurface and points, surface diversion features' land uses,soil types,geomorphologic precipitationstations, models: simulation in event sequence 2. Lomputation B' a. Computehydrographfor subbasin A. b. Computehydrographfor subbasin forA and B. c. Add hydrograPhs end of reservoirR' d. Routecomtined hydrographto upstream b. Compute hydrograph for subbasinC. D. f. Computehydrographfor subbasin at R. g. CombinethreehYdrograPhs h. Route combinedhydrographthroughreservoirR' i. Route reservoir outflow hydrograph to outlet' E' j. Computehydrographfor subbasin at outlet. k. Combinetwo hydrographs Figure 24.1 Typical watershed subdivision and computation sequence for event-simulation models.

Precioitation

J'VhYetograPh - ?l gH a
Time

s
/
Outlet
/ Gross precipitation Excess (net

Time

I c"-bt". I - -

\Z

hydrograph Total streamflow

modelof the rainfallevent-simulation Figure24.2 Typical lumpedparameter runoff process.

MODELS 597 AGENCY SINGLE.EVENT 24.2 FEDERAL prescribedunit hydrographto producea surfacerunoff hydrographfor the subbasin. lossesare divided among the loss componentson the basis of preThe abstracted flows and waters derived from groundwaterstorage Subsurface scribedparameters. which when combinedwith the runoff hydrograph, are transformedinto a subsurface hydrographat the subbasin streamflow the total surfacerunoff hydrographforms combined with another routed downstream, then be outlet. This hydrographcan is the only, or the final, this subbasin if output contributing hydrograph,or simply subbasin beine considered.

'

MODELS AGENCY SINGLE-EVENT 24.2 FEDERAL

by most in Figs.24.1and24.2 arerecognized depicted The rainfall-runoff processes techniques of the eventsimulationmodelsnamedin Table21.1. Specificcomputation river routing, reservoirrouting, and baseflow are comunit hydrographs, for losses, paredin Table24.1for five of the major federalagencyrainfall-runoff eventsimulaBrief descriptechniques. amongavailable tion models.All the modelsallow selection application of an tions of eachof thesemodelsare followedby an illustrative example near 250-mf watershed of the HEC-I model to a single storm occurring over a Lincoln, Nebraska.

Model SurveyRainfall-Runoff U.S.Geological

The USGSmodel can be usedin evaluatingshort streamflowrecordsand calculating basins.lThe programmonitorsthe daily moisture peak flow ratesfor natural drainage soil and can be used as a continuousstreamflowsimulation content of the subbasin its calibrationis model.The model is classifiedas an eventsimulationmodelbecause during a few discharges and evaporation, based on short-termrecords of rainfall, the USGS into has evolved and times several floods.It hasbeenmodified documented 25. in Chapter in detail described DR3M, urban continuoussimulation model, are parameters, which of 10 estimates of initial Input to the model consists simuprocedure matches that fitting optimization modified by the model through an lated and recordedflow rates. Other input includesdaily rainfall and evaporation, close-interval(5-60 min) rainfall and dischargedata, drainageareas,impervious areas,and baseflow rates for eachflood. hyetograph, Phillip's2infiltration equationis usedto determinea rainfall excess reservoir, a linear through routed then and outlet which is translatedto the subbasin 13. in Chapter described technique routing using the time-.areawatershed for relaThe USGSrainfall-runoff model can be usedto simulatestreamflows charlinear storage-outflow approximately tively short periodsfor small basinswith from Output ground is not significant. or frozen in regionswheresnowmelt acteristics storm runoff volumes, peak storm discharges, the model includes a table showing of parameters rainfall amounts,and an iteration-by-iterationprintout of magnitudes peak rates. flow and residualsin fitting volumesand

598

CHAPTER24

MODELS SINGLE-EVENT SIMULATION

TABLE 2II"1 HYDROLOGY PROCESSES AND OPTIONSUSED BY SEVERAL MODELS AGENCY RAINFALL-RUNOFFEVENTSIMULATIONS Table 21.1) Model code names (slee HEC-1 (Corps)

TR-20

Modeledcomoonents Infiltration and losses Holtan's equation Horton's equation Green-Ampt Phillip's equation SCS curve number method Exponentiallossrate Standardcapacitycurves Unit hydrograph Input Clark's Snyder's Two-parametergamma response SCS dimensionless unit hydrograph River routing Kinematic wave Full dynamic wave Muskingum Muskingum-Cunge Modified Puls Normal depth Variable storagecoefficient Att-kin method Translation only Reservoirrouting (Puls) Storage-indication Baseflow Input Constantvalue equation , Recession Snowmeltroutine

(scs)

USGS (USGS)

HYMO (ARS)

SWMM (EPA)

x
X

X X X X X

X X X Yes
No

X X
No Yes

for ProjectFormulation ComputerProgram Hydrology(TR-20)

A particular$ powerful hydrologicprocessand water surfaceprofile computerprogram was developed by CEIR, Inc.3and is known by the codename TR-20, which is Nuntber20. The ServiceTechnicalRelease an acronymfor the U.S. Soil Conservation model is a computerprogram of methodsusedby the Soil ConservationServiceas presented in lhe National Engine ering Handbook.a 'is The program recognized as an engineer-orientedrather than computerwith easeof use as a purpose.Input data oriented package,having been developed for easein use and interpretationby field engisheetsand output data are designed

MODELS 24.2 FEDERALAGENCY SINGLE-EVENT

599

neers,and the programcontainsa liberal numberof operationsthat are user-accomof machinetime. modating,even at the expense The TR-20 was designedto use soil and land-use information to determine for known stormsand to perform reservoirand channelrouting of runoff hydrographs model,with no provisionfor additional It is a single-event hyOrograpns. the generateO The programhasbeenusedin all events. or inflltration beiweendiscretestorm losses hazardstudies,for the designof flood and for flood insurance 50 statesby engineers planning. watershed rural and reservoirand channelprojects, and for urban usingthe SCS storm synthetic or historical Surfacerunoff is computedfrom an dimenstandard The losses' abstract 4 to in Chapter described curvenumberapproach each for hydrographs unit develop to is used hydrogiaphshownin Fig. 12.13 sionless the using constructed is hyetograph subareain the watershed.The ixcess rainfall to the incrementally applied then is and effectiverain and a given rainfall distribution storm' for th9 hydrograph runoff unit hydrographto obtain the subarea method to route As shown in Table 24.I, TR-20 usesthe storage-indication to the direct is added flow base The 13.2). (seeSection throughreservoirs hydrographs usesthe program The rates. flow total .onoffhyatographs at any time to producethe flows the routing hydrographs, flow logic depictel in fig. Z+.iby computingthe total those with hydrographs routed the combining andreservoirs, thiough streamchannels outlet' to the watershed from6ther tributaries,and routing the combinedhydrographs method convex the by hydrographs Prior to 1983, the model routed stream inflow by the modifiedatt-kinmethod(Sec(Section13.1),which has sincebeenreplaced tion 13.3).Asmanyas200channelreachesandggreservoirsorfloodwater-retarding in any singleapplicationof the model.To add to this canbe accommodated structures the capability,the programallowsthe concurrgntinput of up to 9 different stormsover area. watershed is facilitatedby determiningthe locationsof conSubdivisionof the watershed trol points. Control points are defined as stream locations correspondingto crossor sectionaldata, reservoirsites,damagecenters,diversionpoints, gaugingstations, requiredata Subarea desired. be may where hydrographdata tributary confluences ments include the drainagearea,the time of concentration,the reachlengths,strucin-Section 21.1, andeither routing coefflcientsfor eachreach ture data as described data along the channels.Whenevercross-sectionaldala are proor cfoss-sectional in addition to the peak flow vided, the model calculatesihe water surfaceelevations sizes are dictated by the Subarea rates and time of occurrence at each section. hazatd information, it is flood and routing locations of control points. To provide of the characteristics hydraulic the so that to defineenoughcontiol points necessary innormally TR-20 with Applications streamaie definedbetwJencontrof sections. The apart' more or 2 mi to feet hundred few between-a points spaced corporatecont*rol that contributerunoff to a control point are usually lessthan 5 mi2' ."rolting subarLas is no Common subareasizesfor structuresare less than 25 mi2 even though there program' the within limitation on reachlength or subareasize characteristics, to TR-20 includethe watershed Minimal input daia requirements distribution; and duration, depth, the including at leastone actual or syntheiicstorm the routing and structure; each for data the discharge,capaciiy, and elevation

600

CHAPTER24

MODELS SIMULATION SINGLE-EVENT

according data for eachreach.Input can be described coefficientsor cross-sectional to the following outline: characteristics. 1. Watershed a. The area(in mi2)contributingrunoff to eachreservoirand crosssection. (SeeChapter4.) b. Runoff curve number CN for eachsubarea. coded with eachsubarea, moisturecondition associated c. The antecedent as dry, normal, or wet. (hr). d. The time of concentrationfor eacfi subarea mainstream. subarea and e. The length of eachchannelrouting reach 2. Velocity-routingcoefficienttable. a. A table containingrouting coefficientsfor a rangeof velocities(ftlsec). This table is containedwithin the program and need only be enteredif the user desiresdifferent velocities. hydrograph. 3. Dimensionless a. This table is containedwithin the program and need only be enteredif the user desiresa different hydrograph. 4. Actual hydrograph. a. Actual hydrographscan be introduced at any point in the watershed. rates (cfs) spacedat equal Hydrographordinates are read as discharge time incrementsapart, up to a maximum of 300 entries. b. Baseflow rates (cfs)can also be specifiefl' 5. Baseflow. with a. In additionto the option of specifyingthe baseflow ratesassociated modified or can be specified a hydrographthat was input, the baseflow at any other control Point. 6. Stormdata. a. Stormsare numberedfrom 1 to 9 and are input as cumulativedepthsat equally spacedtime increments. b. As an alternativeto specifyingcumulativedepthsat varioustime increstorm can be input, and up to 9 stormscan be ments, a dimensionless by specifyingeachstorm depth and duration' synthesized data. 7. Streamcross-sectional a. Up to 200 crosssectionsmay be input for a singlerun. Cross-sectional versusflow area. data consistof up to 20 pairs of valuesof the discharge are detercoefficients routing b. If cross-sectionaldata are provided,the must specify a the user of suchdata, mined from them. In the absence routing coefficientfor eachreach. 8. Structure.data. rates(cfs) a. The rbservoirdataconsistof up to 20 pairsof outflow discharge versusstorage(acre-ft). b. A maximum of 99 structuresare allowed in a run. The desiredoutput from TR-20 must be specifiedby a set of input file control at eachcontrol point for eachstormcanbe printedby specifyHydrographs variables. ing the control point identification in the control cards. Any combination gf the following items can be producedat eachcontrol point:

MODELS 601' AGENCY SINGLE-EVENT 24,2 FEDERAL * . 2. 3. 4. rate, time of peak, and peak watet'surfaceelevations. The peak discharge rates in tabular form for the entire hydrograph' The discharge for t'heentire duration ofJunoff' Water surfaceelevations The volume of direct runoff, determinedfrom the area under the hydrograph. 5. Hydrographordinatesin any specifiedformat. 6. Summary tablescontainingwater balanceinformation'

from field surveys' by the computerprogramare determined Basicdata needed Rainfall frequency data are input from daia in the U.S. Weather Bureau TP-40 information for presentand future condiand area-flooded report.s Peak-discharge tions for severalreturn periodsare output by TR-20 in a form suitablefor direct use in an economicevaluationmodel.

for Hydrologic ComputerLanguage Problem-Oriented Modeling(HYMO)

who haveno convenfor use by hydrologists A unique computerlanguagedesigned and Hann.6 Williams by developed was experience tional computir programming that commands oJ sequence a forms user the Once the progru- has been compiled, watershed. any of subareas for hydrographs plot, add or synthesiz{ route, stoie, to transformrainfall data commandsare availableto usein any sequence Seventeen streamsand.reserthrough hydrographs these route and to into runoff hydrographs for individual yield irt tons sediment the computes voirs. The HYMO model also stormson the watershed. Watershedrunoff hydrographsare cornputed by HYMO using unit-hydroaccordingto can either be input or synthesized Unit hydrographs graph techniques. are the equations in Tetms 24.3. Fig. in shown unit hydrograph the dimensionless
1.0
. i l - I

q = q' " ( - \
\'p

ttr-'trtrtr-tt
I

t6 inflection point

s ls'o

q = qos(to-t)tx

tp

unit hydrographused in HYMO. Figure 24.3 Dimensionless (After Williams and Hann.6)

602

CHAPTER24

SINGLE-EVENT SIMULATION MODELS

4 : flow rate (ft3lsec)at time t q, : peak flow rate (ft3lsec) te = time to peak (hr) n : dimensionless shapeparameter q6 : flow rate at the inflection point (cfs) /o : time at the inflection point (hr) K : recession constant(hr) Once K and to and 4oare known, the entire hydrographcan be computedfrom the three segment equationsshownin Fig. 24.3. The peak flow rate is computedby the equation
u ^ - -

BAO

(24.r)

parameter,related to n as shown inFig.24.4 where B : a watershed A : watershed area (mi2) O: volume of runoff (in.), determined by HYMO from the SCS rainfall-runoff equationdescribed in Chapter4 t The duration of the unit hydrograph is equated with the selected time increment.The runoff Q for the unit hydrographwould of course be 1.0 in. The parametern in Fig. 24.4 is obtainedfrom Fig. 24.5. Parameters K and to for ungaugedwatersheds

500

50

10
0 2 4 6
n

Figure 24.4 Relation between dimensionless shape parametern and watershedParameterB. (After Williams and Hann.6)

MODELS SINGLE-EVENT 24.2 FEDEMLAGENCY L2

603

10

. K tp

' shape Figure 24.5 Relation betweendimensionless time to to parameter recession constant n and ratio of (AfterWilliams peak. andHann.6.1 located equationsbasedon 34 watersheds are determinedfrom regional regression ranging in in Texas,Oklahoma, Arkansas,Louisiana, Mississippi,and Tennessee, size from 0.5 to 25 mi2, or

i.'t sLp-o K : 27.oAo 231 (#,)"and

(24.2) (24.3)

! +"5rr-o''u( Ao tp : 4.63

)t'"

where SLP : the differencein elevation(ft), divided by floodplain distance(mi), betweenthe basin outlet and the most distant point on the divide LfW : the,basinlength/width ratio River routing is accomplishedin HYMO by a revised variable storqge The continuity equation,I - O : dS/dt, and the storage cofficient (VfC) method.T : equation,S KO, are combinedand discretizedaccordingto the methodsoutlined the the variability in K as the flow leaves in Chapter 13. The VSC methodrecognizes Relations area. and valley floodplain the confinesofthe streamchannelandinundates data, or by HYMO from the input cross-sectional betweenK and O are determined and if the floodplain equation HYMO will calculatethe relation using Manning's are also reach length and The bed slope coefficientsare specified. channelroughness part of the required input.

604

CHAPTER24

MODELS SIMULATION SINGLE-EVENT

method(seeChapter 13) is usedto route The Widelyadoptedstorage-indication curvemustbe determined The storage-outflow throughreservoirs. inflow hydrographs externallyby the userand is input to the prolram asa table containingpairedstorages and outflows,using storagedefinedas zero wheneverthe outflow is zero. for eachcommandare and the datarequirements comrnands The user-oriented as follows: 1. Sta;t: the time rainfall beginson the watershed. 2. Storehydrograph:the time incremenfto,be used,the lowestflow rate to be flow rates spacedat the stored, the watershed atea, and the successive specifledtime increment. 3. Develophydrograph:the desiredtime increment, the watershedatea, the channellength and maximum SCSrunoff curve numberCN, the watershed differencein elevation,and the cumulativerainfall beginningwith zero and at the end of eachtime incrementuntil the end of the storm. accumulated identificationnumber,numberof 4. Computethe rating curve:cross-sectional points in the crosssection,the maximum elevationof the crosssection,the main channel and left and right floodplain slopes,Manning's n fot each segment,and finally pairs of horizontal and vertical cpordinatesof the points describingthe crosssection. the numberof crosssectionsin the routing reach,the 5. Reachcomputations: time increment to be used in routing, the reach length, and the discharge rates for which the variable storagecoefficientis to be computed' 6. Print hydrograph:the idpntification nurnberof the cross sectionat which are to be printed. hydrographs at which the the identificationnumberof the crosssections 7. Plot hydrograph: hydrographs are to be plotted. 8. Add hydrographs:the identification numbers of the hydrographsto be added. 9. Route reservoir:the identificationriumbersof the locationsof .the outflow relation for the reservoir. andthe discharge-storage andinflow hydrographs, L0. Compute travel time: the reach identification number, reach length, and reach slope. 11. Sedimentyield: severalfactors describingsoil erodibility, cropping management,erosioncontrol practices,slopelength, and slopegradient. Output from HYMO includes the synthesizedor user-providedunit hydrothe river- bnd reservoir-routedhydrographs, graphs,the storm runoff hydrographs, Hydrographs and the sedimentyield for individual storms on each subwatershed. from the 34 test hydrographs computedby HYMO comparedcloselywith measured watersheds.

Model (SWMM) Storm Water Management

The EnvironmentalProtectionAgency model,SWMM,s is Hstedin Table21.1 in two to rainfall-runoff eventsimulationand urban runoff simulalocationscorresponding tion. The model is primarily an urban runoff simulation model and is describedin detail-in Chapter25.

MODELS 605 SINGLE-EVENT AGENCY 24.2 FEDERAL numefousmodiflcationsand Lilie mostothers,the SWMM modelhasundergone was a single-event improvementssinceits first releasein 1972.The initial version8 of urban storm modeling in continuous ri'se its allow model, and newer versionse,lo includesa new snowmelt The latestrelease waterflows and waterquality parameters. scour and routine, a new storm watei tto.ugt and treatment package'a sediment depositionroutine, and a revisedinfiltration simulation' androuting routinesarehydraulicratherthan hydrologic' swMM's hydrograph consisitingof single A distributed parameier approachis used for subcatchments as parking lots, city lots, and so on. Accumulatedrainfall on theseplots is first routed closed or overlandflow to gutter or storm drain inlets,whereit is then routed asopen Of the five channelflow to the receivingwatefsor to sometype of treatmentfacility' detail greatest modelscoripared in Table 24.l,the SWMM givesthe event-simulation in simulation,but cannotbe usedin large rural watershedsimulations. step using overland flow depths and flow rates are computedfor each time depth over a Manning's equation along with the continuity equation. The water a will increale without inducing an outflow until the depth reaches subcatchment subover the specifieddetentionrequirement.If and wheneverthe resulting depth outflow rate an Da, requirement, detention specified the than D., is largei c^atchment, is computedusing a modified Manning's equation
V:-J1-(D,-

r .49 .-

D,)2/3St/2

(24.4)

and

Q*:

vw(D, - D)

(24.s)

where V:thevelocity n : Manning's coefflcient S : the ground slope W : the width of the overlandflow rate Q. : the outflow discharge have been computed,they are After flow depthsand rates from all subcatchments to form the total combinedalong with the flow from the immediateupstreamguttef gutter. flow in eachsuccessive gutter and The PiPeflows are routed in the network wh of interest points to any step in the routir time each for ordinates from the storm runoff the until increments and length slope, gutter shape, ters of the pipes or channelsmust also be suppliedand are availttre ior Joef{cienti roughness able in most hydraulicstextbooks' includethe Other iniut requiredfor a typical simulationwith the SWMM model following: percent imsuchas the infiltration parameters, characteristics 1. Watershed pervious area, slope, area, detention storage depth' and Manning's coefficientsfor overlandflow' for the storm to be simulated' 2. The rainfall hyetograph

606

CHAPTER24

MODELS SINGLE-EVENT SIMULATION

and 3. The land-use data, averagemarket values of dwellings in subareas, populationsof subareas. 4. Characteristics of gutters such as the gutter geornetry,slope, roughness maximum allowabledepths,and linkageswith other connectcoefficients, ing inlets or gutters. 5 . Street cleaningfrequency. and their sizes. 6. Treatmentdevicesselected 7. Indexesfor costsof facilities. 8. Boundary conditionsin the receiving waters. 9. Storagefacilities, location, and volume. and invert elevations. 1.0. Inlet characteristics suchas surfaceelevations geometry, pipes slope,Manning's n, and 11. Characteristics of such as type, junction data. downstreamand upstream

(HEC-1) Package HEC-I Flood Hydrograph

a series HydrologicEngineeringCenterdeveloped The U.S. Army Corpsof Engineers universiof comprehensive computerprogramsascomputationalaidsfor consultants, ties, and federal, state, and local agencies(see Section 21.4). Programsfor flood analyprofile computations, reservoirsystem hydrograph computations, watersurface ses,monthly streamflowsynthesis, and reservoir systemoperation for flood control package, HEC-1, is described The single-event flood hydrograph comprisethe series.
here.tt

Two of The HEC-1 model consistsof a calling program and six subroutines. these subroutinesdeterminethe optimal unit hydrograph,loss rare, or streamflow The other hydrograph values. routing parameters by matchingrecordedand simulated subroutinesperform snowmelt computations, unit-hydrographcomputations,hydrographrouting and combiningcomputations,and hydrographbalancingcomputations. In addition to being capable of simulating the usual rainfall-runoff event processes, HEC-1 will also simulatemultiple floods for multiple basin development by numericallyintegrating plansandperform the economicanalysis of flood damages conditions. curvesfor existingand postdevelopment areasunder damage-frequency HEC- 1 underwentrevisionsin the early 1970sand againin the 1980s.Several features were added (e.g., SCS curve number method, hydraulic routing), and a in 1984. The 1985 releaseexpandedearlier microcomputerversion was developed versionsto include kinematic hydrographrouting, simulation of urban runoff, hyimpactsof drographanalysisfor flow over a dam or spillway,analysisof downstream and flood control sysdam failures,multistagepumpingplants for interior drainage, for PCs or Harris minicomputThe 1990versionof HEC- 1, available fem economics. ers, incorporatBsyet other improvements.It adds report-quality graphic and table capability, storageand retrieval of data from other programs,and new hydrologic procedures including the popular Green and Ampt infiltration equation(Chapter4) and the Muskingum-Cungeflood routing method (Chapter 13). In addition to the unit-hydrographtechniquesof the earlier versions, the by kinematic-waveoverland runoff modified HEC-I allows hydrographsyntheses for use in SWMM. The runoff can either be similar to those developed techniques, or uniformly addedalong the watercourse concentratedat the outlet of the subarea distributingthe inflow to the channelor gutter in linearly leogth through the subarea,

MODELS 24.2 FEDERALAGENCY SINGLE-EVENT

607

direction. The 1990 versionallows the use of increasingamountsin the downstream the Muskingum-Cungerouting method in a land surfacerunoff calculation mode. Precipitation can be directly input, or one of three synthetic storms (refer to A standardproject storm (SPS) is availablefor large Chapter 16) can be selected. (over 10 mi2) located east of 105' longitude, using proceduresdescribedin basins but the stormhasa 6-hr manuals.A 96-hr durationis synthesized, Corpsof Engineers peak during eachday. A secondtype of storm is the probablemaximum precipitation (PMP), using reports availablefor estimatesfrom National Weather Service hydrometeorologic is 24ht, and duration minimum (Chapter A these). 16 describes different locations of any synthesis method allows The third stormsup to 96 hr long may be analyzed. and duration desired the specify only duration from 5 min to 10 days.The userneed portion of the duration central the the depth around depth,and the programbalances usingthe blockedIDF methodof Section16.4. The later versionsof HEC-I include all the precipitation loss, syntheticunit hydrograph,and routing functions developedfor earlier versions.Additional loss methodsincludeboth the SCS curve numbermethodand Holtan's lossrate equation (an exponentialdecayfunction). the HEC-1 now includesTR-20 of the popularity of SCS techniques, Because duration ofthe SCS dimensionThe procedures and hydrographsynthesis. for losses 0.2 times the time to approximately lessunit hydrographis interpretedby HEC-1 as (this to 0.29 times the peak converts to g 0.25 times the time peak, but not exceedin lag time). For routing through streamsand reservoirs,the newestversion of HEC-1 inchannelroutcludesall previousmethods,and additionally performskinematic-wave shapes. ing for severalstandardgeometriccross-section models,HEC-1 is relatively compact In comparisonto other event-simulation and still able to executea variety of computationalproceduresin a singlecomputer there is no provirun. The model is applicableonly to single-stormanalysisbecause or no precipitation. periods of little sion for precipitation loss rate recovery during as shown in reaches routing and After dividing the watershedinto subareas of four methods: by one determined can be Fig. 24.6,the precipitation for a subarea precip(2) mean basin data, precipitation station (1) nonrecordingand/or recording precipitation distrihypothetical maximum itation, (3) standardproject or probable (Section Either 16.4). IDF data using butions,or (4) syntheticbalancedstormmethod actual depthsor net rain amountsmay be input, dependingon the user's choice of if The HEC-1 lossrate function is easily bypassed losses. for abstracting techniques the net rain is availablefor direct input. such The programlogic for HEC-1 is shownin Fig. 24.6.Hydrologic processes combining, hydrograph as the subarearunoff computation, routing computation, subtractingdivertedflow, balancing,comparing,or summarizingare specifiedin the illustratedinFig.24.L input usingthe sequence One lossrate in the HEC-1 modelis an exponentialdecayfunction that depends loss rate, in losses.The instantaneous on the rainfall intensity and the antecedent in./hr, is

L, = K'Pf

(24.6)

608

CHAPTER24

MODELS SIMULATION SINGLE-EVENT

READINPUTDATA; REFORMATDATAAND WRITE TO WORKINGFILE

REA,DANDPRINT JOB SPECIFICATION;

READ ANDPRINT DATA

COMPUTE RL]NOFF HYDROGRAPH

1 Figure 24.6 HEC-1 Program OperationsOverview.l

where

L, : P, : E : K' :

loss rate (in./hr) the instantaneous intensity of the rain (in./hr) the exponentof recession(rangeof 0'5-0'9) with time as lossesaccumulate acoefficient, decreasing K, : KoC-cuMlllo+ AK Q4.7)

where

Ko : the loss coefficient at the beginning of the storm (when value of 0'6 CUML = 0), an average loss(in.) from the beginningof the stormto time t CUML : the accumulated of 3.0 C : a coefficient,an average

If AK is zero, the loss rate coefficientK' becomesa parabolicfunction of the loss,CUML, and would thus plot asa straight-linefunction of CUML on accumulated graphpaperif K were plotted on the logarithmic scale.The straightsemilogarithmic in the lossrate coefficient ln Fig.24.7, showingthe decrease fine reLtion t aepiCteO much typically decrease rates loss during any storm.Because accumulate asthe losses above increased is K' tate loss the more rapidly during the initial minutesof a storm, the straightJineamountby an amountequal to AK, which in turn is madea function before theK' value is againequalto the th;t will accumulate of the amountof losses

MODELS 24.2 FEDERALAGENCY SINGLE-EVENT

609

0.2 cuMLl Ko

A,K=Oif CUML> CUMLy

(in.) loss Accumulated ' CUML K' with the of the lossrate coefficient Figure 24.7 Variation CUML. lossamount accumulated

loss,CUML,, is user-specifieda-nd-rs straight-linevalue,K. This initial accumulated rela6d to AK in sucha fashionthat the initial lossrateK' is 20 percenttimes CUMLI Ko are difficult to estimate,and greaterthan Ko (seeFig. 24.7).Initialloss coefficients to itandard purvesin Chapters4 and 23 are available determineinitial infiltration 26. For gaugedbasins,HEC- 1 allows the userto input rainfall and runoff data rateS, are optimizedto give a bestfit to the information from which the lossrate parameters provided. Estimatesof parametersfor ungaugedbasinsfall in the judgment realm lossrate function is availablein noted in Table 21.1. An alternativeto the described a HEC-1, which is an initial abstractionfollowed by constantloss rate, similar to a @index. The HEC-1 model provides separatecomputations of snowmelt in up to to be snowif the zone 10 elevationzones.The precipitationin any zoneis considered plus 2".Thp snowmeltis 32'F, usually temperatureis lessthan abase temperature, is the temperature whenever or energybudgetmethods computedby the degree-day considare usually zones The elevation "qou1to or greaterthan the basetemperature. ered in incrementsof 1000 ft althoughany equal incrementscan be used. unit hydrographsfor eaph subareacan be provided by the user, or clark's unit hydrograph (IUH) can be used. methodl3 o{ synthesizingan instantaneous Clark's method is more commonly recognizedas the time-area curve method of determined in Section12.6.The time-area histogram, described synthesis hydrograph design-storm a unit with from an isochronal map of the watershed,is convoluted in usingEq. 12.38,as illustratedin Fig. t2.l9.The methodsdescribed hyetograph reservoir linear through arc then usedto route the resultinghydrograph Section 1.3.2 coefficientK. Input data for Clark's storage usingEq. 12.35with a watershqd storage

610

24 SINGLE-EVENT MoDELS cHAPTER SIMULATIoN

method eonsists of the time-area curve ordinates,the time of concentrationfor the graph, Clark unit and the watershed storage coefficientK. If the time-area curve for the watershedunder considerationis not'available,the model-providesa synthetic time-arca curve at,the user's request. Because the Corps of Engineers commonlyusesSnyder'smethod(Chapter 12) in unit-hydrograph synthesis for large basins,the Snydertime lag from Eq. 11.5 and peaking Snyder's coefficientCofrom Eq.12.17 can be input, and Clark's parameters will be determinedby HEC-1 from the Snydercoefficients.The actual or synthetic time-area curve is still required. recession usingan exponentof Baseflow is treatedby HEC-1 as an exponential 0.1 in the following equation:

Or: #

(24.8)

where Q, : the flow rate at the beginningof the time increment Q2 : the flow rate at the end of the time increment R : the ratio of the base flow to the base flow 10 time incrementslater The base flow determinedfrom this equation is added to the direct runoff techniques. The startingpoint hydrograph ordinates determined from unit-hydrograph for the entire computationis the user-prescribed baseflow rateat the beginningof the simulation, which is normally the flow severaltime incrementsprior to any direct runoff. Ifthe initial baseflow rate is specifiedas zero, the computerprogram output containsonly direct runoff rates. "storageThe HEC-I package allows the user a choiceof severalhydrologicor All usethe and reservoirs. routing" techniques for routing floodsthroughriver reaches are derelation; some continuity equation and some form of the storage-outflow procedures included in the scribed in more detail in Chapter 13. The five routing programare the following: or storage-indication 1. Modified Puls-this methodis also calledthe storage for usewith normally reserved methodand is a level-pool-routing technique reservoirsor flat streams.The techniquewas describedin detail in Section I3.2. . 2. Muskingum-describedin detail in Section13.1. 3. Muskingum-Cunge-a blendedhydrologicand hydraulic routing method detailed in Section13.1. 4. Kinematicwave-described in Section13.3. lag method.The average 5. Straddle-stagger-also known as the progressive techniqpesimply averages a subsetof consecutiveinflow rates, and the averaged inflow value is lagged a specifiednumber of time incrementsto form the outflow rate. of datain a sequence Input to HEC- 1 is facilitatedby arrangingthree categories summarizedin Table 24.2. The compatiblewith the desiredcomputationsequence, individual input records are precededby a two-charactercode. The first character

MODELS 611 SINGLE.EVENT AGENCY 24.2 FEDERAL DATAFORHEC-111 OF INPUT TABLE24.2 SUBDIVISIONS Jobcontrol
I -, Job initialization V-, Variableoutput summary O , Optimization J -" Job type

andhydraulics Hydrology
K-, Job step control H-, Hydrographtransformation Q-, Hydrographdata B_, Basin data P _, Precipitation data L-, Loss (infiltration) data U_, Unit graph da-ta M-, MeIt data R_, Routing data data S-, Storage D_, Diversion data W-, Pump withdrawal data

andendof job Economics

E_, etc., Economics, data ZZ,End of Job

in Example24.1.

EXAMPLE 24.1 a severe In June 1963 the Oak Creek watershedshown in Fig. 24.8 experienced

watershedareaA-H within the boldface 1. For the remainingelongated over Subarea at eachof Points1- 8 usinga single border,usethe Junestormto simulatehydrographs run of HEC-I and comparepeak flows with recordedvaluesat Points 3 and 8'

SCS runoff equationaand a basin-widecompositecurve number of 73 (see Chapter4).

612

CHAPTER 24 SINGLE.EVENT SIMULATION MODELS

-o+fr; rsog

paralso

KA\
\: \%

, 1236

Scalein miles

;"'F4 d"fs;
Miles 2 Net rain (in) 7.8
+.-)

r--l
l Rafmond Nebraska I 190
\o

\ @ \8.

34
I

,,,G 'I-t
t).

dt\ -

B C D E F G H
f

33.4 26.9
. 27.3 9.2 28.3 17.0 5.0 28,0 82.9 258.0

65

4.1 2.8
A A

t.7 t;7 1.0 N.A.

TOTAL

Figure 24.8 Oak Creek watershed subarea map.anddata sheet,June 1963.

MODELS 613 SINGLE-EVENT AGENCY 24.2 FEDERAL The computation logic to simulate runoff for this storm consistsof the following22 steps: L. Computethe hydrographfor AteaA at Point 1. 2. Routethe A hydrographfrom Point 1 to Point 2. 3. Computethe hydrographfor Area B atPoint 2. 4. Combine the two hydrographsatPoint 2. 5. Route the combinedhydrographto Point 3. 6. Computethe hydrographfor Area C-at Point 3. at Point 3. 7. Combine the two hydrographs 8. Route the combinedhydrographto Point 4. 9. Computethe hydrographfor Atea D at Point 4. at Point 4. 10. Combinethe two hydrographs 1L. Routethe combinedhydrographto Point 5. 12. Computethe hydrographfor AreaE at Point 5. at Point 5. 13. Combinethe two hydrographs 14. Routethe combinedhydrographto Point 6. 1.5. Computethe hydrographfor Atea H at Point 6. at Point 6' 16. Combine the two hydrographs 17. Route the combinedhydrographto Point 7. 18. Computethe hydrographfor Area F at Point 7' at Point 7. 19. Combine the two hydrographs 20. Route the combineAhyJrographto Point 8. 21. Computethe hydrographfor Area G at Point 8' at Point 8. 22. Combinethe two hydrographs are simulatedby convoluting the net Runoff hydrographsfor subareas by Clark's method using synthesized hydrographs with unit storm hyetograph of 0.8 is applied for Oak vabte A Co (see I2). Chapter Snyder'i coiffiiients of the watershed. capacity retention high the moderately of Crlek because (11.5) usinga C' Eq. from found are subarea for each values lag time Subarea 2.0. valueof Hydrographstreamrouting is performedusing the Muskingumtechnique reachtraveltime, using (chapter 13) with x : 0.15 andK: the approximate velocity determinedas average A Ch6zy velocity. the avefage by length divided three If K exceeds is used. slope reach average root of the the square 100 times into shorter HEC-1 by subdivided is further the reach routing increments, lengthsto ensurecomputationalresolution. A sampleof the input and output for this job is shownasTable24.3.Each Only Steps 1-5 are in sequence. of the 22 computationalstepsare separated was not usedso function rate loss HEC-1 the that Note in sample. included the Note also that hyare equal. depths rain and excess end-of-perlod that the facilitated 2 was Point 1 to Point from hydrograph the A of routing drograph (AMSKK) l.2hr. of K-value a with each lengths reach equal using - three flow ratesfor eachof steps A summaryof HEC- 1 peak and time-averaged peak at Point 3 is 27,539 simulated the that Note given Table 24.4. in is l-22 27,500' The correspondof val:ue recorded the well with very agrees cfs, which -ing simulated peak flows at Point 8 arc 22,453and 21,600cfs, and observed -' - - - -respectively. Il *

o
N N

H
N

. E @ o i >

--

z x ci z
6 q
N

E E

o r F
F

7 z +

F'qEE . o co R E
L
N J : :

@ L F - = < t < / > E @ Ho

I I

U :

c o

ll o
F

> l x u , z
!l

= 5gs=

q
E

:
@

E
6 N

ct

c!

:z u
u
)l

J ; Z

gEcj

rt

< =

tsN
u q
g

- a o q * = ! e

N o

s^Niil;.1

_ ^o o o t : $ f l F _fr3ef lf

*-

z
F

3
F , i U N A

o o e F e v s E c c 3 V E V f f R c a 3V H v E v S B c c 3 Y ? .
I t Eo o o o < - +
-U_=

= tr
F N O t h @ F O O O F N O

F E

= 9 9 t r P 3

R F N R I R R R F R

B E

z.

226 S FtrE
e
o o Y
f,

z u e
J

F
F

s u
6

F E N

O N t s N q
N

z
o

t 3

t
' i

r
*

= u

h
P
=
<

EE
* *

tr

E A

: J
z

a ss cq ri 6i
4 d
=

N
o
F

dt c;
N

il F F

il

!!

;l
6

F 19 t - 6
F

EE

eirt

u
4 F

RE
]

z. u

=
r
q
F N

z u E r u
Y

6' u

=
z
m
a

$r

R =
m x E

c |r

=
E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E
q

d
o o
J

- - - - - Y Y - - - - - - - - - - - - Y - - - - = - - - - - - - -

F N 6 + o @ t s o o - N m + 6 @ F @ o o - N O < @ @ F O O - N O 6 h 6 n 6 6 b 6 6 6 @ @ @ @ 6 @ @ @ @ F F F F t s F F F F F @ @ @

<
>

E 1 6 E i 6 E t 6 E t 6 E F 6 E i 6 E b 6E F 6 E t d 6 ; 6 E b 6E i 6 6

=
o z
F F q F

h 6 @ @ F F @ @ O O O O F - N N O Q O O - F N N O O < < h s @ @ r - N N N N N d N O O O O O O O O O O O O e O O

i6 E;6 E i6 5;6 E

=
< o

z . z . z z z z . z z z z z z . z . z .z z z z zzzzz z z zzzzzz.z ------------------------------a-+ + + + < < + a + + + + + + + + + + 4 4 b h 6 6 A h b 4 h r a r a N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N

z. .

* * *

* * * * * * * * * *

+ * * * * * *

* * * * * * * *

* * * *

d
=

- s d s s 3 E 5 S R R S 3 = S S 3 ei fd F s 3 N 3 e g + e R 5 fr F h + F @ O - @ - e + - - 6 - F F - - @ b r N @ q N F F
- O @ N O 6 F + < N O O @ N o F h O N - - - N o o o o N N - F F

.
ll o q

s
<

a E a

= * @ E R q +N f i q B q s q q * E E q E e E e e E e e q E e E e e e ex q e q
U >< u

6i!P+

d F
F

;s H
>+
= N x (
T

aEs

q
o

BAA8AAt83888EB8ttt333AAEtt33EA=EA

A'

u 46

F E +
@ r { 8

z.

O O O F O o @ N F F o O F o O o o O O O O O O O a A Q Q A Q a Q A d ? q q e q < ? c q q q q q q q q q q q q q 9 - N f r + F @ F @ + o o N 6i

rr q O
J

O3=S3:=9=S9RFNRTRKRRRgESA
E

a E
F

nz.z
- L

z u z. z.
o H o

=
z.

- 6 - O - 6 6 6 0 0 0 0 0 0 0 0 0 0 0 0 0 < < h h @ @ r r @ @ o o E i 3 = = i < N ; 6 U E i i = = F n l 6 o + s ' A a N N N N d N N N O e e O e Q O O O O O O

ll

= +

E=

zz,zz

z zz

zzzzzz.

z .z z z . z . z . z , 2z z z z . z . z . z z z

zzz.

a - - - - - - - - - - - - - - - - - - - ? - - - - - - t - - - - o o m o o o o o o o o o o o o o o m o m q + + s s + + t $ + $ s +
N N N N N N N N N N N N N d N N N N N N N N N N N N N N

z
E

3
tr

=
n6;
Ylt
F

= ii

d d d d d d o d

d d d d d d o d e ; c j d d o o o

o o

FtrR R E 6S S S S S E S s B s S S S S H FE E S a o
z.
> = E o o o o o o o o o e i6 E ;6 E i6 F E6 E o o o e o o o o o d ;6E 6 + + 6 6 6 6 F N 6 6 6 d o O F F N N O S S + 6 6 a6 E Fi E

7 >
{ 6

z. z. z. z. z. z z z z z. z. z. z. z z z. z. z z z z z z. z. z - - - - - - - - - - - - - - - - - - - - - - - - n N o 6 N o N h o N 6 F r 6 o o 6 6 6 h 6 - - - N - - N N N N 6 N 6 N r h b o o o N N N N N N

* =

cj

ct

ei

ci

c;

<j

ci

ct

cj

ii

E 8 i l f rs s H h E a E E E S S A A E S S R F S R F F
2

= EBagBBaata=R===B=sBB33EEE Fi = H N N F K = 3 =S S EEEE=TEE=5! =S E
N

V
= = { O

z. z. z. z. z. z. z. z. z z. z. z. z. z. z z. z. z z. z. z. = = z z
= = = = = = = = = = = = = = = = = = = = = f - - - - - - - - - - - - - - - - - - - - - - - a = = = + + < + + + + + + + + + + + + { $ 9 e ! Q o o h o h d N N d N N d N N N N N N N N N N N N N N N N N N

o U

= z.
z.
o

z.
L

N F

z. o

E
F E

u
J

u
E

6 F

= < 9 d

ot + Gt ci 6i - N - m N 6 6 F o + O + O @ O @ O = - t s h + o N N F

r; r; Kj + ci + o o o @ O @ o O O @ + N - + O F h + N - F F

qt

gi

ot

ct

di

<t

q u

o E o

= =

@ t s o o F N O + h @ F @ O Q - N O S 9 @ r o +o { + + s + + N N N o o o 6 o d s N - o r i

+ +

t o

o'i tP +

nE ra o
Gi!P+

z.
6

= =

u F E

? Y x q E E frzz
o = = > o a

EE EEEEEEE EEEEEEEE=EFFEE=z. z. z. z. z. z. z. z z. z. z. z. z z z z. z z. z. z. z. z z z z.
- a ? - - ' t - - - - - a - a q q q = 9 q g g = g s s s s s < s < s < s d N N d N N N N N N N N N N N N N N N N N N N N N

ifE

r
e

8 E A o - 9
+!Pct hH!2
@ F o

u
{ c

=== =>>
'

=x
x > =
@

- z z.\ tr-g?Yx = * = 6 EZ>O> - = ' <

N

=
= J

c t d d d F +

o o r

3 $ B S g E R S N J H E F E E J = a s 3 f f i $ K 9 A A S 5 = 3 6 D
N N N N N N F F F

a . ar '+ t r r
a = Z =

a z. E V
E 5 o -

o'
E

o - N o + 6 @ F @ o Q - N A + E r

= - - -

=
u ^

+ v i Y

z.
E

>+

tr

E E E E E E E E E E E E F F = F IR =FFEEE=EE
z. z. z z z z z z z. z. z. z z z z z. z z. z. z z z z z z.
- - - - - - - - - - - - - - - - - - - - a - - - N

=
o

t d.6
*

$
t
T

v B
U A

N I
T

m o o o o o e e R R R R R R R R R N N F t t o N oN o N N N N N N N N N N N N N

=
J

-i

ct ci

ct

ct ct

ci

ei e; ci

et ci

ct

ci

c;

FFFREESASSFbBtsA6SESeSSssE P o
z.

EE S g E E E g E E E E E E E E E E F F E E F F E E
?
>

z z . z z . z . z . z . z . z . z . z . z . z zz 1 1 1 1 4 z
J

z zz
-' -

z
-

-J -'

3 n n K R R K K R K K FR .R S R K K R R R FR R R R

* * * * * * * * *
= E "i "i.i ci d d ei ct ci d ci a I ci d <j ct d d d d ct d cj

E 8 i l f iS A a h t s g A E D A g A A b S S R E S R f ,
z.

E EEEEEEEEEEFF=RFRFEEBE=EEE
7 U

= z
z.
dl

z.
o

= = = - - - - - 5 = = = D = > == = = = 1==== -= = ---

z z z z z' 1 = = 4 ) z. z. z. z z. z. = z. z. z z. z = z 1 1 ) 3 2 2 = = ) - t

---

---

---

z.

a
lo F

3 x x x x * N t N * x * * F * R * N N R R K R R K K
* x * * * * * *

u
J

E F

a 3 $ p F $ HEE 5 E F H= F E P
@ r U

+ s 3 n j6 '6 id jq ' @
i B ot E;\u M F 9

e R N R R E E S E S B H h S g S S S S $ s 9 + s g B
z
E
J L

E a

U =

6 ? Y x s E E

=gFFEEEF-9 I EEEE EEEE EEE EEgE EE

-

==
,,,=
U

r!6=

giH

E == > q< n
< F
- z z +

z. z. z. z z z z. z z. z z z. z z z z z 1 1 1 z < = z z z' = = = = = 5 = = 5 = = 5 - 5 -= i = = =r ? ? ? = ? - -r = - ? - - - - 5 = - - = = - = - - O

z>> - 3 5
;

*N X * * N * * X * X tN t* tfi * R x * fi N tx

;+ sHm
= N

E= EHii
\ o

"o-oNEH
- N

E FTVX = - t o
E > @ >
N

FEHffEPE$nHHFEE=EEE
q

a l

a = G V
o @ E 5 o -

o + 4 @ r @ o o F N 9 S P 9

= = E R F N , R X

R
U ^

= ES
F -

o <

z.

EgEEEEEEEEEEFF=FFFFREE=EE = -^
z .z z z z z 1 1 1 = 1 z = 4 = 4 = = < = = = = = <
= - = - i - i - - - - - - - - - - - - - - - - - n o :ZO u
N N N

Y Y

=
o

R R R R R R R R R R R R S IR R F K IR R R F S S N N

a >
E
F

=
@ E

z
u
F

s
U ' :

(5 o U F

6 9
u z. 2
@

< = , 6 @

u
Y

a
Y

@ E
@ F

x C
C

5 E
ts

O
F

ci

ct ct d

ci

ct

ci

ct d

ct d

ci

c; ci

c;

=
q @ U

ceeeEaeeeEeEEeeccaEEeeeEeEEeeeEE
EEEEEEEEEEEEEEEEgEEEEEEEECqEEqqEE

d
6 q

z ======F= seECqeeeeeeEeeeeeeEeaeEE a ---------E F N o + 4 @ F @ q o F N o + h @ r - o F S R S F R F R R B E S g 6 @ @ F l 6 @ @ @ @ @ @ @ ; e h h 6 h 6 6 h h 4 o

z.

EEEEEFFEEEEFFFFRRRFEEEgFEEEEgEEEEE
4, z z z z. z. z z z z z. z. z. z z z z. z. 1 1 1 z z z = 1 1 1 1 1 2 z z z
> = = r 5 ? -t 1 a -t -. -

a
q

E x x * * R f i N S t t N N S S N F t F K K K R K R K R R F R K K R K
* * * * * * * * * * * * * * * * * * * * * * * * * * + * * * * * * x

L <

? - s e f i s s i i * B $ = R i E E s B 6 R $D E EB s D R s
d r o 6 ; 6 o F c 6 m - N - 6 - s t a f N + F - + @ F @ o F o r o o N F F r o 6 o F F

o P + g 4 i 5

@ q

u x u

eE eeqEEeeEEeEeee EqNfrqqEfifrITqiEEEE

eE ceeeEeEeeeeeeEeEeEeee 3 =.AeEeeqee
Z 8 = N K t s B 9 S K =f
G

iqiqEEeqqqEt

eqeEeeeqeeE
R * K F N R R B E S A

, , x i l < _ 6 5 - i+ 6
J

tr i<

o
e

- N o + h @ F @ @ o = s P 3 5 9 = = = R F N

g 9 F

z
U

z. F z

EFgEEEEEEEEEEFE=RRRFFEE=gEEEEEEEEE e =s =
tl
J
4

U ^
L v

7 >

2 z z 1 1 1 1 1 z z z z' z.=z - ? z. - z. z z z. z. z z = =1 =1= 1 == - z. - =z = z5 z = z. = - = r = = = = - - = = = = = - - - - - - - = = = - - - - - i - - - - - - - - - - - - - -

2. E =

s s Rx * s s * N * * * 3 * * R R R R R R R R RR R R R R R S T R
N r a

s t

4. 4^
ts yE
u

:
R

=
E

ct cj

cj

e; ct

ct

ci

c; d

et ci

ct d

cj

ct ct d

ei d

t
f

a
z

F F F R E 5 S S S 3 S b E t sE ; s a s 3 E s I E F

E EEgEEEgEEEEEEEE=PFEEEEEEE
= = g z z .z . 1 1 z = z z .Z z .z = = = = < = = = = = - = = = = = = = = = ; = = = = i = = - = = - = - = - = = = K F R R R K R K K K R K RiI+AR N K K R R R R K R

= iE

6i

6i -

ci

cj

ci

ct d

ei d

ci

ci

cj

ct

E
z

E i l s s a H h a g E E 3 3 3 3 A E E t sR F S R * R

N q

EEEEFFEEEEFF=FRRFREEEgEEE
= = = = = = = < z- = = 4 = = = = = = = = = = = == = = = = = = = = = i ; = 4 4 4 4 = = - -= 1 = 1 = R S * * * x N f, * * * s * s tx $tR R R K R K R t I
i

z.

z .E o < 3 P
' u z,
o o
F . 6 f = >
* *

tr

< =

N
F

P>

$ R 6i ri di d +/ E E H H E il H d g E F H F F E E s 3
J F O F 6 + O N N F -

g g a s a E gs s s $ s s s s g B E K NRR3 6 s s
z.

==
L E

d r d
= N =

<

o <

o U
6

E EEEEgEEEEEEEEEEEEEFFFEEEE s
d P d
EC'I6 + @ =

z, z
F

=-= = = = = = = = == = = = = = == = = = = = == = 3 fi * R* t F s Nt x t s * sx t N* t * * * * NN

u < c

E F A
= F x

==
@

ffi85

z. o E

E
E

OEggF$FFF
r

6 F

u + + o = <

u z F z

U ^ = G

==
N

z.

z.
Y v O N

= = = = = = z. 4 = = = = = 2- z-.= = = = = = = = = = = = = = = = = = = = 4 = = = = 4 = = 1 = - = = = R R R R R R R N R R R R R R R R R R R R T S T S R

i a Y O u

5 o
F

=
J L J T

et ci

et cj

ci

ci

cj

ci

ct o

ci

ct d

ci

ct d

ci

R F R R B E S S $ s A b B g B s s a g S t D E g
zz.

==
z.z

E=gEEEEEEEEEEEEEEFFEEFFEE
z -z - z= z z. z. z z z z 1 1 = = Z. Z Z. z. z. z z z z z z = = - - - - - - - - - - - - - - - - - - - - - - ?

R R K R R R K K K R R K R K K R K K R K R K R R R

==
I L

O F

ci

ot r{.J
+ N

ot F -

Fi ct F

ct 6i ri

-.j

ci

6i e; Gi o

i EE3333EE33R=SRf, 6 i l f f if i A t s h f f g
z z

EEEEEEEFEEEEFFFFFRFFEEEgEEE
z.
F

== ========================= 3E xtRNx*tF*fiR*ttttstKRKKRRR * * * * * * * * * * * * * * * * * * * * * * * == 39

77

z. z z. z. z. z. z. z z. z z z. = 1 1 1 1 z z = z. z. z z z

@ F

z.
c F

I r

EnFEss$FEER=$ssFEEiEE
g
+

Fi d; ci O F O

ct

E E R N R R t E S g $ E E h S g g E $ e $ 5 s $ s e B
z. z.

d P d = N =

+.dE

>

g EEEEE=gEEEEEEEEEEEEEPFFEEE + d E E s
L -

= =
, , , = = + t u

9 = ! ! = N =

e
2 2 R
F J

=========== == =========,===== 33 xssNN*sNtsF*tK*NFsNstN*tF

* * * * * * * * * + * * * * * + * * * * * * * * * * *

? 7

z. z z z. z. z. z z z. z z. z. 1 1 1 1 1 1 = z z z z. z z

Vivx u E E
* O O

=q EFs
= R = @ P

E
E
oz. z =

E z z

=_ ==
@

=E{
==
; '
J J 4 (

E=E
+ N O F F @ L 6 l U q

cicic'di= B Eg

E$ g * F EF $ F HF E

t s N
@ @ @ N N N

o o u
F

F N o + 6 @ F @ o = = s P = 9 9 5 = S R F N R N K
U ^

9<
=F
o

= z
z.

z z

EE=EEEEEFEEEEEFF=FRRFEEE=gE = -^ :
v Y T

== ====-=====----=-=-=======
3 3

7 7

z z. z. z z z. z. z. z. z. z. z z. z 1 1 1 = 1 z z z z z z

d. 9.a
Y O

vi
b N

4 J

* R R R R R R R R R R R R R R R R R R R F F N S *

A^l

N
N

U Z t6 fl E P E c? .q ,,I.t' o U F rid O N

o > 1 4 . < d

3 f o .rl t
: :!

3 EE E
' = e

Y a

dd
6 t s (q\

==
E E +HJ
El: Z Y = o ii@ E - N

O o

= - F F

z.

z. z. < =
^ ; -i E tlJ 9

Y G EU < 6 J > oz @

5 *

z o z. u

O F

K 6

=
@

z.
u u r u
U

= z
z.

u F

v
q J

6 h N

4 @ N

L r N o N

< N

o
G

a eeeeeeeeeeeeEEeeeeeqeeeqeeEeeeeee
U

x U
q

z
E

F N o < 6 @ F @ o o F N O < o @ F @ O O - N O + D @ F O O F N O b b b h @ @ h b 6 @ @ @ @ @ @ @ @ F t s r F F F F F r t s @ o @ @

z.
=

= BBEBaaaBBaaaaBBBaaBEaaaaaBaaaaaaa
6 6 @ @ F F 6 6 O O O O F F N N O O O O F F N N m m + + 6 6 @ @ F F F N N N N N N N N O O O O O O O O O O O O O O O

=
<

z. z. z. z. z. z. z. z. z. z. z z z z z z z z z z z z z z z z z z z z z z z.
t a a a a -

N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N

d
^ E 5

ct d

o h N N o F O m O O @ @ O @ @ O @ F O @ q F F O @ O F N F o F @ O O O + N @ @ O + O @ 4 F F r + N F F F O 6 O N + h 6 O - O 6 @ + O N - -

ct ct et +

di qt ri

qt ct qt et ct -

<t F dt d

ci ct st qt -

ci et ct ri

ct ct d

e + q q e q e a a E r r a a E E E a E E t a a t a r fi u E l F f rE q = q f e
x U
q q

= a

flE+ = = g

o o o o o o o o o o o o o o o o o o o o o o o o o Q o o o o o o o

z.

O N F e T N - O A @ O N @ O O O o o o o o O o O O o O O O O O O O O - N N O 6 + O N F N - - O O O O o O O O O O O O O O O O O O O O

, , x

s=+ HEF
o
@ O O e N O + 6 @ F 6 O O - N O + 6 @ F O O F N O F E N N N N N N N N N N O O O O

o o a

= e
=
s
o o o o o o

^ ?.frtr
o <
=

z.
u

= z
z
a

5 = =

o o o e o e o o o o o o o o o o o o o o o o o o o o o o + + 6 6 @ @ F F @ 6 O O O O - - N N O o o N N N

e o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o F F N N O O + + 6 o @ N O O O O O O O O O O O O O

ll ! u ^ 2 '

aE+ =
-

=
< {
A

z z = z z z. z. z. z z. z. z. z. z. z. z. z. z. z. z. z. z. z. z. z. z. z. z. z, z z z z
a a a ? a a a 1 a a o o o o o o o o o o o o o m o o o m o o < < + + + + + + + + + + +

1 = < q ^
F V A

3 E
ts

N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N

= n a

j E
6

f;-

ci

ct ci

ci

ci

ct

ct d

ct ci

et d

c;

ci

cj

Q c;

s $ S s S A b E gB s s E s a a s a E a E FFRRS
z

EEgEEEEEEEEEEEFE=EFFEEEgEE
7
>

z . z z z . z . z .- z z z z z . z . z z - - j = - - - = - - -

- - )i - - -

z 1 1 1 1 = z
- - - - -

z z z
- ' - ' -

z
-

- -

E N P K F R K R K R K RR R K R R K F R R FK R R K i
* * * * * * *
N N F

*
rt

*
ct r

*
rt

*
6i -

*
-

*
-

*
ci

*
ci

*
d ci

= ;
E

ct
F F

di il
C <

oi F

ct 6i

ct d

ct d

i 3 a b t s f i3 6 D 3 3 3 A b 3 E R = S R F F a E s f fs
z

EE E E E F E E E E E F F F F F R F F E E = E E E E
z.

7 z z z z .z z 1 z z z = = = = = = = = = = < = = = =
> = = = = - - - - - 5 - i i - - - - - - - - - - - -

k F
q

3 x s t * t t N N s * t F * t x * t S R R K F R K K

* * * * * * * *
u z,
o

=
u

nRRFEE$$EHR;$sFEil
K R R R S E S S S S A D E B E E $ S+
< 6 @ F A < < q 9 F = m

F 6 0 @ O F O

t
@ L

=
F a

q!ni!!

= o

=
z

EEESgEEEEEEEEEEEE=EFFEEEg ? s --Ki
F a
U < E = F x

=
r z, Z N

z z 2 z. z z z. z z 1 z z z- = = = = = = = z z z z z: = = = = = = = = = = = J
- - - = - - - - i - a i - - - - -

i s i
o 6 r

* t N * t s * * N * S F S t * N S S * S N t S N S
* * * * * * * * * * * * * * * + * * * * * * x *

5u?6

> =
@ I

k z
o

= ;>
u o +

-*RNEFBSEus'pRxf HexnsE neEfr F

o o - \ o + h @ F @ o o = S p = = g = P P R F N R t R
U ^

d 5 i i6 ;N \ NE
@ts u +

24 =

z.

E=
trt

z
v v

EFEEEEEFEEEEEFF=FRRFEEE=EE = -^
4 z,zz.z.z=1=zz==z.=z==4z zz <z=4 = = = i = = = - - = = = = - = - = = - 5 5 - - - - :ZO r
N

*
J

E * n R R R R R R R R R f l t R R R F R R R R t * t F t

MODELING 625 24.3 STORM SURGE FLOWS ANDAVERAGE PEAK OF SIMULATED TABLE24.4 RUNOFF SUMMARY 1963 STORM AT POINTS 1 THROUGH 8 FORTHEJUNE,
Peak Hydrographat Routed to Hydrographat 2 Combined Routed to Hydrographat 2 Combined Routedto Hydrographat 2 Combined Routed to Hydrograph at 2 Combined Routedto Hydrograph at 2 Combined Routedto Hydrographat 2 Combined Routed to Hydrographat 2 Combined
I

6-hr

24-hr

72-hl

Area

2 2 2 3 3 3
4 A A

5 5 5 6 6 6 7 7 7 8 8 8

34475. 25622. 17310. 30092. 26759. 15400. 27539. 25912. 3362. 25925. 23911. 7 100. 239tt. 23911. 3349. 23911. 22949. 4113. 22949. 22453. 1684. 22453.

24048. 20848. 11207. 26453. 2406r. 10503. 25828. 2455r. 2382. 24606. 22858. 5604. 22874. 22874. 2478. 22874. 22024. 2764. 22924. 21595. 868. 21595.

69'74. 6974. 3100. 10074. 10070. 2995. 13056. 13052. 689. 13702. 13523. 1816. 14717. 14717. 7s0. 15268. t4984.
I tJ,

3382. 3382. 1503. 4885. 4885. 1452. 6337. 8409. 441. 8850. 8875. 1162. 10038. 10039. 480. 10518. 10482. 495.

33.40 33.40 26.90 60.30 60.30

zt.3v

t5t12. 14994. 228. 14995.

rc977.
10908. 146. ' 11054.

87.60 87.60 9.20 96.80 96.80 28.30 r25.10 l25.to 28.00 153.10 1 5 3r.0 17.00 170.10 170.10 5.00 1 7 5l . 0

MODELING 24.3 STORM SURGE

stormsbut alsofrom storm floodsfrom single-event Coastalareasnot only experience Severalcomby hurricanes. (short-termchanges in sealevel) normally caused surges Most are surges.l2 storm hurriCane-produced puter models are availableto analyZe the atmosfrom of momentumtransfer modelingthe physicalprocesses deterministic, analysis of adjusted to allow phereto the ocean.Parameters in thesemodelscan be (PMH) or probable hurricane maximum actual or hypothetical storms such as the projecthurricane(SPH). standard private consultants,and universitieshave developedsurge Federal agencies, surgemodelsare listed inTable 24.5. open-coast models.Severalnonproprietary, User's manualsare availablefor all the modelslisted. Storm surgemodels simulatethe effects of wind momentum on ocean water and wavehydrodyoceanography, This involvesprinciplesfrom meteorology, masses. geometry, and water-level about storm, namics,which-all operatefrom assumptions from surges damage conditions.Tide levelsare includedin mostof the modelsbecause peak level. surge entirely on concurrencewith the often depends The equationssolvedby the first three models in Table 24.5 are two-dimenof continuity,Eq. 13.40.The fourth of Eq. 13.51and the equatiotr sionalversions the continuity equation and solvesEq. 13.51 through a seriesof model disregards

626

MODELS SIMULATION 24 SINGLE.EVENT CHAPTER

MODELS SURGE TABLE24.5 STORM

Programname SPLASH

User/agencf
NWS

Data input radius to Atmosphericpressures, maximum wind, storm speed radius to Atmosphericpressures, maximum wind, storm speed radius'to Atmosphericpressures, maximum wind, storm speed, maximum wind speed,depth of shelf Wind field, pressuredifferences, radius to maximum wind, forward speed

equation Differential solutionmethod Finite-difference

coasts Applicable Mildly curved, Gulf and East coasts All coasts Gulf and Atlantic coasts

SSURGE FIA model

coE
FIA

Finite-difference Finite-difference

BATHYSTROPHIC

CERC

Finite-difference

All coasts

,NWS, National WeatherService;COE, U.S. Army Corps of Engineers; FIA, FederalInsuranceAdministration;CERC, Coastal Engineering ResearchCenter, Corps of Engrneers.

they treat time asa succession The modelsarenot truly dynamicbecause assumptions. end of eachtime stepusedin at the depths water of steadystates.Output is a file of the SPH, PMH, or any preinclude be simulated the simulation. Storms that can scribedwind field.

r summary
involvesthe useof singleBy far the largestnumberof hydrologicmodel applications in this chapteror the eventsimulationmodels,whetherthe generalversionsdescribed urban runoff models about to be presentedin Chapter 25. Data requirementsfor modelsare nominal-far lessthan for continuousmodelingstudies.In single-event the majority of cases,the data required for any subareaare easily obtained from readily availabletopographicand soils maps.Given the basin area,slope,soil types, of runoff hydrographs of peak flow ratesand shapes land use,and location,estimates can be obtainedfor given stormswithin a few ar severallocationsin the watershed hours' time usingthesemodels.They continueto be the primary tool usedby practicin analysisand designof stormwaterhandling facilities. ing engineers

PROBLEMS
for a river basin are as shownin the sketch.Preparea sche24.1. Six numberedsubareas connectrunoff components' matic diagramfor a modelstudyusingboxesas subbasin ing lines as channel routing links, circles as hydrograph combination nodes, and for this the computationsequence trianglesas reservoirrouting nodes.Then describe page). (See next on sketch basinin the samemannershownin Fig.24.1. in your locale using the HYMO model a unit hydrographfor a watershed 24.2. Synthesize from Snyder'smethodand unit hydrographs Comparewith corresponding equations. the SCSmethodin Chapter12.

PROBLEMS

627

t-r'\t-;

Sketch for Problem 24.1 a unit hydrographfor the entire 258 sq. to synthesize 24.3. Usethe HYMO modelequations in Fig. 24.8. mi Oak Creekwatershed 24.4. A watershedexperiences a l2-hr rainstormhaving a uniform intensity of 0.1 in./hr. Using E : 0.7, Ko : 0.6, C : 3.0, and AK : 0.0, calculatethe hourly lossratesl, model.Determinethe total andpercent by the HEC- 1 event-simulation asdetermined lossesfor the storm. 24.5. RepeatProblem 24.4 wing CUML1 : 0.5 in. 13.7to the outlet of the 30-mi reachusingthe in Pr-oblem 24,6. Routethe inflow hydrograph pairs of flows two time incremethod by lagging averaged HEC- 1 straddle-stagger outflow rates. ments (12 hr). Comparethe routed and measured in Problem 13.7through the reachby dividing the 30-mi 24.7, Routethe inflow hydrograph and treat the outflow from eachas inflow to the next in reach into three subreaches and comparethe f,nal outflows line. Lag flows one time increment in eachsubreach values. with the measured 24,8. Study Table 24.3 andFig.24.8, and then definethe following terms from Table24.3: AMSKK,X, TAREA, NP, STORM, TP, CP, TC, R, RAIN, ANdEXCESS' COMP Q. 24.9, Search the HEC-I printout in Table 24.3 to determinevalues (give units) of the following: a. The time incrementusedin the model run. B. b. Snyder'sCe,Eq. 12.17input for Subarea unit hydrograph. Subarea-A c. The pbak flow rate for the synthesized A. d. The total runoff (in.) from Subarea A. e. The peak outflow rate from Subarea f. The peak-to-peaktime lag in routing the outflow hydrographfrom Point 1 to Point 2,Fig.24.8. g. The percent attenuationcausedby the reachbetweenPoints,l and 2. A is neglected. peak outflow rate if Subarea h. The Subarea-B peak outflow rate. i. The simulatedSubarea-B

628

CHAPTER24

MODELS SIMULATION SINGLE-EVENT *Refer questions: to the HEC-1 output in Tables24.3 and24.4to answerthe following 24.10. "PRECIPITATION PATTERN" actual rainfall depths' or a. Are the valueslabeled

Points 1 and 2 of Oak Creek. of Branched 24.11. Describehow, for a given storm,you wciulddeterminethe effectiveness enumerAnswer^by 8. Point at flooding 24.8 to reduce "as oak reservoirat Point 9 in Fig. "runs" two of a maximum allowed are You ating your computationlogic illustrated. may be usedas long as you describeyour wlttr ilBc- t, und uny nui-rb", of subareas subareas. 24.12.AhydrologistwishestomodelawatershedusingthesCScurvenumbermethodfo runoff hydrographthrough a reservoir determinenet rain, and to route the watershed usingthestorage-indicationmethod.BaseflowistobeincorporatedaSaconstant model or modelswould valuethroughoutthe storm duration.Which event-simulation the task? accomPlish a hydrographusing the 24.13. The following data were preparedfor a Yl card for routing is 1'0 hr' methodof UgC-t' The time increment straddle-stagger
Value

Field

Field
A

0
I

2 3

5 6 7

0.0 0.0 0.0

_ I

by the straddle-stagger Usethis information to route the following inflow hydrograph is zero again' outflow until routing continue is zero. method. The initial outflow
4

Time (hr) Inflow (cfs)

0 0

1 0 3 0

2 6

3 0

5 9 0

6 3

' 0

7 0

r20

8 0

REFERENCES
t. 2.
J.

4.

Chapter25

UrbanRunoffSirnulation Models

r Prologue
The purposeof this chaPteris to: for simulation . Describenine of the most commonly usedcomputerpackages of urban rainfall-runoff processes' . Show that the modelsdo more than simulaterunoff; they allow the user to systems' and designcompleteurban stormwatermanagement analyze resultsobcomparing by precision . Demonstrate the'models;capabilitiesand tainedwhensimulatingtheSamewatershedwithdifferentmodels. . provide a ,,shopper's giid"" to commercialandpublic domainurban stormwater software. the volumeof runoff and also urbanization generallyhasthe effectof increasing Early attemptsto apply the runoff. of tendsto result in earlier ani greaterpeak rates and designwere successanalysis modelsof Chapter24 inurban system single-event total decreasing analogs' rainfall-runoff iutiy u""o*plished by seleciingmore intense area' pervious total the from zones 6y deductingthe impervious rainfaUabstiactions andby and in improvedchannels, oi traveloverthe land surfaces the speed increasing of approximations wave kinematic adding componentsto the codes to allow for stormwater urban and sewers in storm flow and hydrologicrouting of flow overla=nd retention or detentionPonds. urban watershed betweenthesemodels'predictionsand observed Discrepancies and continof single-event class whole haveresulteJin the dwelopmentof a response systems' in operating processes uous streamflow models of the oniq.t" -urbanized userto the allow also but process, Thesemodelsnot only simulatethe raintitt-runoff or to facilities management stormwater existingnetwork of interconnected analyzean ponds, detention sewers, storm (underground designnew componentsof the system ditches.streetinlet sizesand locations,etc')'

! = X ^ a

ts

E o

> E o - (!

U EY 8 :

cloq
G l \O

n
C.t

qq
O h ca +

iq\oqv'l

s t F L - q ' l <f, \n ca C..l t'-

=^ ' z
E E)N : = 5

ral c.l

0O<.\r)\Ocn

E - ^

h O

e { -

d . ? ) O N c n * i

odd SEXsE Nh
1.O
@ J

o > 5 = ^ ^
-

a.th

o cd \O

e{r) cJJ +:+

o\$cicio. r.ai-jio cn6.aa{$

o c) o

; 6 C O z
f - v

\r) ol

00 a.l

\o r)

c'{ \o e^l

o = t ^

5 =' e z

ho

o\

or)

cqNcqol$

o o o

l )

H"o
J

o a o tuq

d 8E E==tr=
H v r r A Y ^A r\ H v r ) n f ) n Y m Y F l - 5 m ; FA m

R.9 9 { X

.999 .9 X X X ; i X

gs - 9 iBe
Q 6' v . = - Y

a ul
T F IL

n n O\*
oo ci t'-

cl
oo

o?n
* O i

\qqqq

F-it-oj*f i n C { *

QE(u v o o oi5-

- b R

Eo
.=^
v -

s 5s HSRHX
cn Y * N - v l O O ne{ Y - Y - i O \ F - * i cnr)'am\o

t--

l--

O\

o r!(E R(U F

>v \r

q \ m s

co o1

Eo
(u

q n
Vcn cn v

\ori
6\O i

OCIOl S re.lcnt N ca
a

ot--F-m\o

o t,rJ
[ l h

7aa E tr !.! ulr

Fa1

-tr9 E o
';^
d]

F-O

+ + * n
.t)

\o
O
d

N6 hr 61 6l

e..l\o N<.r 6lF-0Oe.l C.) cn O 0O cq

i i n

.o r

S Z

s< c lt)
o| IT

a o

tr
V)

638

CHAPTER25

MODELS URBAN RUNOFFSIMULATION * 6 1.2

, / ),^
a
R / 9a . I U .

n R

o a
^ a

/
a
o

,t

jlo 2

)- nd
9 . {o4

a 8 . / 3

t , ' Q t3 4 'f1

.s&.' a 7
t.2

2 4 (cfs) peaks Observed

(a)

a 1.5
E

;p0
O r n

^0.12

Y o.os
'F 0.04 ; z 3

r L ^ r /

t \

ttT,*'

10 11

Drive basin'Long Island'N9wYork' for Woodoak Figure 25.3 RRL results (After and(c)thehydrographs' (b)volumes, peaks, 19,1966r(a) of October storm Stall andTerstrieP.e) ofthese Physicalunderstanding gutters,pipes,and openchannels' surfaces, many the of understanding present ii muctrgreaier than the flow phen-omena complexphenomenagov"erningrunofffromruralareassuchasantecedent "infiltratilon, soil moisture movement, transpiration, moisture conditions, evaporation,and so forth. provide a function for 5. A modification of the RRL method that would grasSedareacontributionstorunoffcouldbedevelopedintoavaluable in spiteof the This is believedto be possible iesign tdol for urban drainage. by the be could flexibility *unly "o*plexities involvei. Further ^offered storage' surface through runoff addiiional irovision for routing surface on an urban basin 6. The input data r"qoir"m"nt' foi o'" of the RRL methods for storm drainage a basin of evaluation for the engineering are reasonable than the data or_elaborate complex daia are no more The necessary design. design' drainage ;;dlly compiled foi a traditional storm

25.1 URBANSTORMWATER SYSTEMMODELS

l.) 3 1

639

E zoo
o

E U

$ roo
a a

3! 1 z oo" a t 2 a 1l

1.0

i.'
o

ot<

, +

,,,' ,9r iir.t,'ol

' 1a " 8

'J - ;'i^':ls

r-'f;3:'r ,"ni
0.5 1.0 Observed runoff (in.) (b) 1.5

100

200

300

Observed peaks (cfs) (a)

150

o bo

k 100

2 Time (hr)
(c)

Figure25.4 RRLresults for Echo ParkAvenue basin, LosAngeles, California, storm ofApril 18,1965: (a)peaks, (b) volumes, (c)thehydrogriphs. (AfterStall andTerstriep.e) 7. It appears that rainfall occursin greateramountsin the United Statesthan in Great Britain. This may accountfor the fact that the RRL method is successfuland widely used in Great Britain and yet suffers the abovedescribed breakdownsfor someof the basinsstudiedin the united States. 8. Better urban rainfall and runoff data are required for the proper testingof all mathematicalmodels.Research basinsthat do not havehydraulic pioblems,suchasundersized drainsor inadequate inlets, shouldbe selected and instrumented.

lllinois UrbanDrainage Area Simulator, ILLUDAS

As mentioned, the RRL method only simulatesrunoff from paved areas of the basin that are directly connectedto the storm drainagesystem.Grassedareasand nonconnected pavedareasare excludedfrom consideration. The ILLUDAS10model

640

MODELS SIMULATION RUNOFF 25 URBAN CHAPTER ofthe RRL methodbut also pavedareatechnique incorporaiesthe direct$ connected pavedareas', nolconnected grassed and ,""oggrir", and incorporatesrunoff from similar to the very subbasins.is the for areahydrogriphs Computationof grassed usedto same subbasin the shows 25.5 Figuie hydrographs. approachfbr pavedar-ea grassed contributing the is area shaded ihe illustrate pavedarearunofi inFig. ZS.Z. area,which is largely the front yards ofresidences.Rain falling on any not-directlyto run off instantly onto the surroundinggrassed connectedpavedarea is assumed areahydrologytakes over. Runoff from back and sideyards often area,and grassed street'The drainsgraluaily to a commonbick lot line andthenlaterally to the nearest consideration grassed areas such from traveltime requiredfor this virtually eliminates duringre1ative$shortintenseStormSnorma1lyusedfordrainagedesign. areahasbeenidentified,the curvein Fig' 25'5 can After the contributinggrassed time of be constructed.Travel times acrossthe grass strips are equivalentto the equilibrium from Izzatd's equation,2
t": 0.033KLq"o'67

(2s.r)

97 percentof4,, that is, reaches which is the time when the overlandflow dischrage ()\ )\ 4" : 0.00002311L of overlandflow (cfs/ft of width) at equilibrium where 4, -i = discharge to be 1'0 in ILLUDAS : raii supplyrate (in./hr), assumed (ft) L : length of overlandflow and 6:(0.00071 *c)S-o33

(2s.3)

where S : surfaceslope (ftlft) turf' c : coefficienthaving a value of 0.046 for bluegrass to be a straightline. The endpoint,as illustrated in The time-area curve is assumed the traveltime from the farthestpoint on the contributinggrassed Fig. 25.5represents storageis normally set at 0.20 in. but can be varied' In ILLUDAS, depression Infiltration is modeledusing Holtan's equation,lr atea,

f:a(s-F)'o+f"
where

(2s.4)

,f : infiltration rate at time / (in'/hr) turf a : avegetativefactor : 1'0 for bluegrass (storage at the soil porosity (in.) : mantle soil the in available storige S minus storageat the wilting point) of the wilting F ; water alreaJy storedin the soil at time /, in excess point(in.)(amountaccumulatedfrominfiltrationpriortotimet) s_F:Storagespaceremaininginthesoilmantleattime'(in') (in./hr), generallyequivalentto the f" : finallonstant infiltration rate hydraulic conductivity (in./hr) of the tightesthorizon saturated Presentin the soil Profile usedto computean If physicalpropertiesof the soil are known, the equationcan be_ the various between interrelation general innttiation curve. Figure 25.6 showsthe infiftratisn rates-andstoragefactorsinvolved'

25j
(a) Subbasln map (contributing grassedarea shaded)

SYSTEMMODELS URBAN STORMWATER

(c) Rainfall

641

(d) Runoff from supplemental paved area

(e) Losses

(b) Time versus grassed area curve

n
(f) Grassed area supply mte

92
a

cAoce't
GA: GAz GA
0 1 2 3 4 5 6 7 Travel time to inlet (min)

al

3 4 s 6 7 8 9 1 0 Time (min) (g) Hydrograph Qr= GAt (GASRT) + GAI(GASRI) Qz= GAz(GASR1) + GA1(GASR3) + GA2(GASR) Qz= GAz(GASRI) + "'+ GA1(GASR,) Qn=GAn(GASnr)

,? o

u F
I

8 10121416

Time from start of rainfall (min)

(After Terstriepand hydrographs. of grassed-area Figure 25.5 Elementsin the development Stall.lo)

642

CHAPTER25

MODELS URBAN RUNOFFSIMULATION

Infiltrat lon curve ial foinilial infiltration rate

S = Area cross-hatchedbelow curve equals total storage in soil (in.)

!.1 E

s-r'

\%
Final, constant -; " infiltration rate

Available smragem soil (in.)

Time(hr)

Figure 25.6 Diagram of inflltration^relations used in ILLUDAS, Eq. 25.4. (After Terstriepand Stall.'u) _ TabIe 25.4 provides an example computation of an infiltration curve for bluegrass on a silt loam soil in which soil moisture S of 6.95 in. is available. The equation is

f:r(6.9s-F)t++0.50

r r 55 )

Standardinfiltration curveshavebeen devisedfor use in ILLUDAS for soils having SCS hydrologicgroupsA, B, C, and D (Chapter4). Thesecurveswere synthesized LOAM SILT FOR CURVE OFINFILTRATION 25,4 COMPUTATION TABLE
Available srorage, Water stored, A F F (in.) (in.) 0.9s "1.0 1.0 1.0 1.0 1.0 1.0 0 0.95 1.95 2.95 3.95 4.95 5.95 6.95
lnfiltrationrate

Time Lt' (hr) 0.07 0.09 0.11 0.16 0.24 0.43 r.43 t (h0 0 0.07 0.16 0.27 0.43 0.67 1.10 2.53

s-F
(in.) 6.95 6.00 5.0 4.0 3.0 2.0 1.0 0

(S- D'+
15.0 12.3 9.5 7.0 4.65 2.64 1.0 0

t (in./hr)
15.5 12.8 10.0 7.5 5 .l 5 3.r4 1.50 0.50

'avg

(in./hr)
t+.1 I 1.4

8.7 6.3
A 1

2.3 o.'l

"Incremental rime, Lt : A'F/f*". Source: Tersftiep and Stall.ro

MODELS 643 SYSTEM STORMWATER 25.1 URBAN

from the Horton equation where /o : ,f" : e: k : / : f: f"+ (fo- f,)e-k, initial infiltration rate (in'lhr) ultimate infiltration rate baseof natural logarithms as k : 2 a shapefactor selected time from start of rainfall

(2s.6)

The curves This equationis solvedin ILLUDAS by the Newton-Raphsontechnique. inFig.25.7. are shown To accountfor wet versus dry conditions, II-LUDAS divides the antecedent ranges'shownin Table 25'5'Each moisturecondition (AMC) into four user-selected prior the storm day. Infiltration from to precipitation 5-day is based on the total specified' value AMC on the depending F;q.25.6 is varied, ILLUDAS allows the user to operatein two modes,analysisand design.For pipe andprovidesnominal stormsewer hydrographs mode,the modelgenerates design analysis In peak flows. pass the to surcharge, without adequate, diameters thatare throughoutthe basinnodesandlinks andthen mode,the modelg"n".ul"t hydrographs are too small' It also sumsthe volume of pipe diameters input alerts the user if iny by the it could not be accommodated because ai inlets up runoff water backed sewers. storm undersized

Time(hr) turf on Figure 25.7 Standardinfiltration curvesfor bluegrass four SCS soil types used in ILLUDAS. (After Terstriep and Stall.lo)

644

MODELS RUNOFF SIMULATION 25 URBAN CHAPTER CONDITIONS MOISTURE TABLE25.F ANTECEDENT LAWNS FORBLUEGRASS
AMC number
I

Description
Bone dry Rather dry Rather wet Saturated

Total rainfallduring storm 5 days preceding (in.)

2
J A

0 0-0.s 0 . 5 -I Over1

Source: Terstriep and Stall. lo

The ILLUDAS model requires estimationof severalinput parameters.Other The the sensitivityof ILLUDAS to variationsin parameters. evaluated studies12'13,14 study in Ref. 13 concludes: as the soil in AMC increases L. The sensitivityof the peak flows to changes group changes from D to A. the 2. The rangesof sensitivityto the soil groupsand AMC are approximately 3. XT;;rr" in the AMC from2to 3 and a changein the soil group from B to C are critical for large designreturn periods, and from C to D for small designreturn periods. 4. The ti'me to peak for various combinationsof soil group and AMC, for different return periods,remainsthe same. from 1 to 5. The peak flow and runoff volume increaseas the AMC changes is particularly important betweenAMC 2 and 3 in general 4. This increase and betweenAMC 3 and 4 for soil group A. The peak flow and the runoff from A to D for constantAMC. volumeincreaseas the soil group changes This increaseis particularly important between soil groups B and C in generaland soil groupsC and D for AMC 1. markedly for time inerementslarger than 5 min, 6. The peak flows decrease significantly for time incrementslarger and the pipe diametersdecrease than 10 min. 7. The time increment should not substantiallyexceedthe paved area inlet time.

Ovefflow Runoff Model (STORM) Storage,Treatment,

STORM is the Corps of Engineerscontinuoussimulationmodel of the quantity and quality of urban Stormwater resulting from single eventsor continuousdaily rainfall.ls It also simulatesdry weatherflow from domestic,commercial, or industrial Wet weather hydrographs,simulated from intermittent or continuous discharges. rainfall, can be used for a variety of hydrologicstudy purposes. hourly (hydrographs that alsoprovidewaterquality characpollutographs weather Wet or synthetic eventsand used in historical predicted individual for can be teristics) consistof The pollutographs receiving streams. runoff on impacts of of assessments pollutant concentrations. pollutants, and of rates, amounts runoff hourly

MODELS 645 SYSTEM STORMWATER 25.1 URBAN

RainfalVsnowmelt

Treatment

Figure25.8ConceptualviewofurbansystemasusedinSTORM.(AfterU'S.Army of Engineers'") CorPs in Fig. 25.8. Snowmeltis simulatedby the degreeThe model is conceptualized to aid in the selectionof day method (Chapter 14). Statisticalinlormation is output desiredcontrol of storm sbrage "upu"iti., und treatment rates required to achieve annualerosion' averannualrunoff, avefage *ut". rurr&f. Statistics,suchas average overflowfrom pollutant annual average and ageannualoverflowvolumefrom storige, are all Provided' storage, (to the interactionof precipitation,air temperature signal The model simulates erosion, dry weather flow' snowfall), runoff, pollutant accumulation,land Jurface or treatment system'The storage,treatmentiates, and overflowsfrom the storage prograncomputescontinuousorsingle-.event.runofffromrainfall' of the differencebetweenrainfall and depresnunotf is computedas a fractio-n on land use. Runoff in excessof the depends ,ion ,torug". The fiaction selected treatment'Runoff for subsequent capacityis divertedinto storage speciteOiTeatment overflow and is of both the treatment rate and stofagecapacitybecomes in excess diverted directly into the receiving waters'

No. 55 GR'55) SCSTechnicalRelease

for an The SCS TR-55 Procedures in described were areas urbanized mat recommends theseProcedures severalvendorshave1 Procedures,

646

CHAPTER25

MODELS URBAN RUNOFFSIMULATION

from through a numberof outlets.A public domainversionis available them avai{able the U.S. National TechnicalInformation Service.l6 initial studrenditionsof TR-55 should-perform Usersof the vendor-developed program would be one that TR-55 to verify the code.An ideal ies usinghand-checks notifies used, all assumptions states usesSCS sourcecode or has SCS endorsement, to adjustments for making options the user of range violations, and incorporates accountfor all or most of the following: 1. Changesin the 484 coefficient of Eq. 12.22 for steep, avefage,or flat watersheds. 2. Per centimperviousness. 3. Percentof channelthat is improved. 4. Pondingarea. range. length over width ratios that fall outsidethe assumed 5. Subarea 6. Slope. 7. Antecedentmoisture conditions' 8. Different storm distributions' 9. Proper lag time equation. that the duration for the derived 10. Recognitionof the SCS recommendation time of concentration' unit hydrographbe about 13 percentof the subarea 1,, greater than abstractions, that have initial 11. Allowance for watersheds S. 20 percentof the potential maximum retention, Whether by manual or computer operations,the SCS cautions that TR-55 hydrograph methods should not be used to perform final design if an error of Their adviceis to useTR-20, after ZS peicettt,inpredictedvolumecannotbe tolerated. is very complex wathershed urban if the -aking appropriateparameteradjustments, the useof regarding precautions Other of accuracyis required.lT or if a higher degree 15.2' the graphical and tabular methodsare identified in Section in among time parameters SCS Urban Time Relationships The relationships phenomobserved with reconciled SCShydrologicmethodshavenot beencompletely Several ena or time relationshipsin other modelsof urban rainfall-runoff processes. are effects, urban for adjustments formulations for lag time, with miscellaneous substantial have and literature, mentioned in Section ll.7 and elsewherein SCS impact on the shape of the hydrographs.The rational formula (Chapter 15), to that the time of concentration,deflnedas the time for rain falling illustrate, assumes at the most remote location to reachthe outlet, equalsthe time to peak of the urban hydrograph.lzzwd found this to hold approximatelytrue in observingrunoff hydrographsfrom pavedareas.2 The greatestdiscrepancyfound when comparing SCS and known urban time methis the prolongedtime basethat resultswhen SCSunit hydrograph relationships from runoff observed on is based shape The hydrograph odsin Chapter 12 areapplied. the dimenfor base the time 12.13, As shownin Fig. rural watersheds, undeveloped, unit hydrographis about 5.0 times the time to peak. It was shownin Chapsionless time must requiresthat the release of unit hydrographs ter I}thatlinear superposition /"' Recall of concentration, the time is equal the time baseof the IUH, which in turn

25j

SYSTEMMODELS URBAN STORMWATER

647

FOR A D-HR UNIT OF TIME RELATIONSHIPS TABLE 25.6 * COMPARISON BY SCS AND URBAN RUNOFFMETHODS HYDROGRAPH Rational/lzzardllUHModels of Urban Runoff Given: D: t" Time to peak: t, Releasetime : tc Solving: D:t" Time to peak = t Releasetime : tc Timebase:D+t,:2t,

UnitHydrograph SCSDimensionless
D : O.2 X time to peak Time to peak : Iag time + D/2 Lag time : 0.6 t" (Mockus Equation) Time base = 5.0 X time to peak D : 0.133 t" Time of peak : 0.666 r. Time base: 3.33 t, time : Time base - D : 3.20 t, Release

that the excess-rainfallreleasetime, t,, was definedin Chapter 11 as the time from rain to end of direct runoff. As shownin the Table 25.6, time relationend of excess unit hydrographof Fig. 12.13 giveprolongedrunoff shipsfor the SCS dimensionless durationscomparedto other urban runoff models.Only the time to peak is approxiUrban runoff modelsbasedon SCSdimdnsionmately equivalentin this comparison. less unit Mrograph proceduresmay result in longer time basesand hydrograph than other methods. recessions

Model(DR3M) RoutingRainfall-Runoff USGSDistributed

The U.S. GeologicalSurvey simulationmodel for urban rainfall-runoff applications tnodel for small watersheds originatedin tgZg as a lumped parametersingle-event to distributedparameter was expanded (deicribed in Chapter 24) and subsequently status,intendedprimarily for urban applicability.lsAlso, a soil moistureroutine was simulation. addedallowing quasicontinuous miles from a few acresto severalsquare The modelian be appliedto watersheds or It doesnot simulatesubsurface in size (an upper limit ol10 mi2 is recommended). interflow coniributionsto streamflow,and thesemust be externally addedif considered imPortant to the simulation. Routing of rainfall to channelsis by unsteadyoverland flow hydraulics,and by kinematic-wave routing hydrographsthrough channel reachesis accomplished methods(."t"ito Ctrupter13). The differential routing equationsare solvedby one of The usermay specify an explicit or implicit finitethree optional numericalmethods. differencealgorithm, or the method of characteristics' Time may be discretizedby the user in as small as l-min increments.The smallesttime increment is used by the program during any dayshaving short-time interval rainfall, calledunit days. Otherdaysare simulatedas24-hr intervals.Movement of surfacewater is simulatedonly during unit days.For the rain-free intervals, daily rainfall is input and usedto modify the soil moisturebalanceleadinginto the The formatforrain datais compatiblewith that of the U.S. Geological nextunit day(s). Input data Survey systlm, WATSTORE(Water Data Storageand Retrieval System). can also be obtainedfrom any local National WeatherServiceoffice. practically any basincanbe studiedby breakingit into severalsetsof four types (must be approximately of model ,"g*"ntr. Theseinclude overland flow segments

648

25 CHAPTER

MODELS SIMULATION URBANRUNOFF

Legend: Overland flow boundary Stream channel

oF1

Overland flow segment I

1 Reservoir R-ESI 1 CH1 Channel

Segment

Lateral inflow Rahfall.excess Rainfall excess oF1, OFz

Upstream inflow

oF1 oF2 cHl RES oF3 oF4 CHz

cH1 Rainfall excess Rainfall excess oF3, OF4

RESl

Watemhed outlet

for DR3M' of watershed Figure25.9 Segmentation rectangular),detentionstoragefacilities, channels,and nodes.This is illustrated in in the flgure may haveinflow from either lateral or upstream fig.2i.9. Eachsegment (or sources both, as occurs for segmentCHz). Rainfall is uniformly distributedover the overlandflow rectangles'Each has a Laminar flow is assumed and percentimperviousness. given length, slope,roughness, 13'56arefound 13.28and bandmforEqs. of Ioo".or&.rtheiesegrnents.Thevalues fromm: 3 and
n - :K

'

88So u

(t\ 1\

where56is the slope,r: is the kinematic viscosityof water, equalto 0.0000141ftlsec (for 50"i water), and K is a coefficientrelating the Reynold's numberN. to Darcy's 24.Thevalue frictionfactorfbyK: fN,.FlowoverroughsurfacesislaminarifK) K is related to rainfall intensity by (25.8) K: Ko+ 101 where K6 is fouhd from Table 25.7, andl is the rainfall intensity (in./hr). Channelsin Fig. 25.9 representeither natural or artificial gutters or storm pipes are allowed). Inflow to channels sewers(either op"n .hunntls oi nonpressure Nodesare used overlandflow (aslateral inflow), or nodes. comesfrom otherchannels, or whenthe user or reservoi-r, contributeto a channel whenmore than three Segments wishesto specify an input or baseflow hydrograph. describedin Section 13'3' Channel routing is by kinematic-wavetechniques, b and m. These arc parameters Input is the channei length, slope, and routing )

25.1 URBAN STORMWATER SYSTEMMODELS TABLE 25.7 ROUGHNESS COEFFICIENTS FOR OVERLANDAND CHANNELSEGMENTS Sudacetype Concreteasphalt Bare sand Graveledsurface Bare clay-loam soil (eroded) Spatsevegetation Short grassprairie Bluegrass sod
Source: Alley and Smith.r8

649

Laminarflow Ko

Turbulentflow Manning's n

24-r08 30-120 90-400 100-500 1,000-4,000 3,00010,000 7,000-40,000

0.01-0.013 0.01-0.016 0.012-0.03 0.012-0.033 0.053-0.13 0.10-0.20 -0.48 0.17

developed from the Manning equation and slope of the channel,respectively.The equationsare m : 1.67 and
D : -

l.4gsto/z
n

(25.e)

where ruis obtainedfrom Table 25.7 or similar information. The model adjustsboth b and m for various shapes,including circular and triangular (seeTable 13.1 for severalru values). If overbankflow is possible, a secondsetof b andmparameters can be input for all flows in excess of the channelcapacity. Reservoirinflow hydrographs are routed by either of two storage routing methods. If a linear reservoir model is appropriate,the storagecoefflcient K, relating outflow to storage by O : KS,is input. If the modifiedPuls methodis desired, a table of outflows and correspondingstoragelevels must be input. The model assumes an initial reservoirlevel equal to that corresponding to an outflow of 0.0 cfs. Pondingbehind culverts can be modeledas a reservoir if an S-O relation is input. This should include data points correspondingto roadway overflow to allow simulation of this common phenomenon. Excess rainfall (runoff) from perviousareasis developed from the precipitation input, minus severalabstractions. Infiltration is simulatedby

r:r(' .{of)

(2s.10)

where K is the hydraulic conductivity, P is the averagesuction head acrossthe capillary zone, andmo andm are the soil moisturecontentsbeforeand after wetting. The term SMS i$the soil moisture storase.The rate of excess rain is found from

r :
and

ir1<s

Qs.rr)
(2s.1,2)

, : r - ti f 1 > , s

where1 is the rate of rain suppliedto infiltration.

650

CHAPTER25

MODELS URBAN RUNOFFSIMULATION

on whetherthe areasare directly conRunoff from impervious areasdepends to flow connectedare assumed directly not Those system. drainage nected to the third of One inflow. as lateral added are they where pervious areas, immediatelyonto to the inflow is lateral rest the and is abstracted, areas connected the rain on direcdy gutter or channel. Soil moisture is accountedfor in a two-layer hypothetical storagezone. The amount in storageaffects the infiltration rate and allows continuoussoil moisture rain input) During unit days(dayswith short-duration rain events. between accounting Between zone. storage upper to the is addbd 25.10 Eq. from moisture all infiltraied During to SMS. added is rain the daily proportion of a user-specified unit days, rates pan evaporation input using from SMS, occurs evapotranspiration rainlessdays, which at event, rhin next the until process continues ttris a coefficient. multiplied by time infiltration is governedby the amount of soil moisture; acrossthe continentalU.S. and Applications of On:U havebeen documented measuredrunoff for almost and of computed Calibration Hawaii.le in Alaska and in peak flow estimatesand error median a reveal watersheds 37 400 storms over results obtainedfor highly best the with respectively, percent, 24 volume of 21 and to datearethat the model studies verification from Indications imperviouswatershedi. from stormshavingmagnifloods for simulated rates peak flow the may overestimate resultsfor smaller give better and facilities, control flood range of tudesin the design quality studies. runoff in used stormstypically

FHWAStorm SewerDesignModel (HYDRA)

of integrateddesigncomputerprogramscalledHYDRAIN,2othe As part of a package U.S. Federal Highway Administration developedthe HYDRA storm drain design The model is distributedundercontract model for useby federaland other engineers. with the FHWA through Milrans SoftwareCenterat the Civil EngineeringDepartHYDRA hasbeenlinked by commerment of the Universityof Florida at Gainesville. The program'sprimary useis analyzcial vendorsto an integratedCAD/GIS system. ing adequacyof existing storm drains or designingnew storm drains and inlets by the rational merhod dJscribedin Chapter 15 or by a modified rational method the hydrographas a Eapezoidhaving a volume equal to the calcuwhich represents lated net rain. by the modifiedrational method, Cornmercialversionsof HYDRA allow design SCS methods,or revised Santa Barbara hydrographmethods.2lHYDRA has one over other storm sewerdesignmodelsin that hydraulic gradelines through advantage can be checkedby hydraulic backwatercomputationsto determinetotal the systern Another or junction boxesare surcharged. andwhetherinlets,manholes, losses system storm ofthe capacity inlet the usefulfeatureis'that streetand gutter flowsthat exceed to the added and location sewersare routed by HYDRA to the next downstream at that Point' hYdrograPh

Model (SWMM) Storm Water Management

and appliedstormrunoff simulationmodel wasjointly develA very widely accepted oped by Meicalf und Eddy, Inc., the University of Florida, and Water Resources for useby the U.S. EnvironmentalProtectionAgency (EPA). This model Engineers22

SYSTEM MODELS 25.1 URBAN STORMWATER

651

rainfall is designei to simulatethe runoff of a drainagebasin for any predescribed pattern. The total watershedis broken into.a finite number of smaller units or subby their hydraulic or geometricproperties. that can readily be described catchments A flowchart for the processis shownin Fig. 25.10. The SWMM model hasthe capability of determining,for short-durationstorms of local floodsaswell asthe quantity of given intensity,the locationsand magnitudes and quality of storm water runoff at severallocationsboth in the systemand in the receiving waters. The original SWMM was an event-simulationmodel, and later versionsitkeep track of long-term water budgets. The fine detail in the designon the model allows the simulation of both water with urban runoff and combinedsewersysquantity and quality aspectsassociated here.Information obtainedfrom quantity described are aspects the water tems.Only for storm waterrunoff control, systems storm sewer to design would be used SWMM in regions where watersheds urban small limited to relatively model is Use of the documented. adequately are quality of water aspects in the differences seasonal

Subcatchments

Gutterflow 1. Overlandinput 2. Gutterinput 3. Flow (Manning's) 4. Depth(continuity)

Figure 25.10 Flowchart for SWMM Runoff Block hydrographiccompu-tation.({fter Metcalf and Eddy, Inc.22)

652

CHAPTER25

MODELS URBAN RUNiOFF SIMULATION

fhe simulation is facilitated by five main subroutineblocks. Each block has a devices specificfunction, and the resultsof eachblock are enteredonlworkingstorage part to be usedas of the input to other blocks. The main calling programof the modelis calledthe ExecutiveBlock. This block interfacingamongthe is the first and last to be usedand performs all the necessary other blocks. The Runoff Block usesManning's equationto route the uniform rainfall intensity over the overlandflow surfaces, through the small guttersand pipesof the sewer pipes, into the main sewer and out of the sewerpipesinto the receivingstreams. system pollutional graphs(pollutographs). provides time-dependent This block also the quality and package the TransportBlock, determines A third of subroutines, infiltration, and calculatesthe quantity of dry weather flow, calculatesthe system water quality of the flows in the system. is containedin for water quality determination A usefulpackage of subroutines specify or have the model the StorageBlock. The StorageBlock allows the user to processes treatment facility wastewater in an optional selectsizesof severaltreatment peak If this block simulates percentage flow. used, of the that receivesa user-selected passes as the sewage of the sewage the changes in the hydrographs and pollutographs processes. sequence of unit through the selected The earlier version22 allowed simulationof any reservoirfor which the outflow could be approximated aseither a weir or orifice, or if the waterwaspumpedfrom the reservoir. The newer,version23 allows input of 11 points of any storage-outflow relation and routes hydrographsthrough natural or artiflcial reservoirs,including backwaterareas behind culverts. Routing is by the modified Puls method, which assumes that the reservoiris small enoughthat the water surfaceis alwayslevel. Evaporationfrom reservoirsis simulatedby a monthly coefficient(suppliedby the user) multiplied by the surfacearea. completes the hydraulic calculationsfor overlandflows, in The Extran Block2a pipes and culverts.It solvesthe completehydrodynamicequations, channels, and in performs surcharging, dynarnicrouting, and providesall the depth,velocity, assesses grade and energy line information requested. by the user. areas,slopes, widths, and linkagesmustbe specified Subcatchment parts pervious impervious for and be supplied Manning's roughness coefficientscan of eachsubcatchment. model listed As indicated in Table 24.1, SWMM is the only event-simulation parameters If infiltration losses. that usesHorton's equationfor calculatingwatershed infiltration ASCE standard can specify the user for Horton's equationareunavailable, are compared for each time step determined capacitycurves.Infiltration amountsthus plus any surface subcatchment on the amountsof water existing with instantanEous is larger, it is set infiltration loss if the rainfall that occurredduring the time step,and of the maximal and consists for Horton's equation Input equalto the amountavailable. Eq. 4.1. The Green-Ampt k in recession constant minimal infiltration ratesand the equationis also usedin SWMM. Urban storm drainagecomponentsare modeledusing,Manning'sequationand the continuity equation.The hydraulic radius of the trapezoidalguttersand circular if pipesis calculatedfrom componentdimensions and flow depths.A pipe surcharges

MODELS 653 SYSTEM 25j URBAN STORMWATER it is full;provided that the inflow is greaterthan the outflow capacity.In this case,the surcharged amountwill be computedand storedin the Runoff and TransportBlocks at the head end of the pipe. The pipe will remain full until the stored water is completelydrained.Alternatively,the Extran Block canbe usedto conducta dynamic conditions. simulation of the systemunder pressure-flow Necessaryinputs in the model are the surfacearea, width of subcatchment, ground slope,Manning's roughness coefficient,infiltration rate, and detentiondepth. coefficient,invert slope, Channel descriptionsare the length, Manning's roughness diameter for pipes, or cross-sectionaldimensions.General data requirementsare processaccounts for all inflow, infiltration in Table25.8. A step-by-step summarized areas,providing a calculateddischarge losses, and flow from upstreamsubcatchment hydrographat the drainagebasin outlet. The following descriptionof the simulation the logic processincorporatedin early versionsof SWMM will aid in understanding of the model.2s hyetograph: to the specified 1. Rainfall is addedto the subcatchment'according Dt: D,+ &Lt where D, : the water depth after rainfall at time / D, = the water depth of the subcatchment R, : the intensity of rainfall in time interval At 2. Infiltration 1, is computed by Horton's exponential function, I, = * f" + (fo f")e-o', and subtractedfrom the water depth existing on the subcatchment (2s.r4) Dz: D, - I, A,t in Horton's equation(Eq. 4.1) where f",fr, k : coefficients Dt: the intermediate water depth after accounting for infiltration
FOR STORMWATER TABLE 25.8 GENERALDATA REQUIREMENTS MANAGEMENT MODEL (SWMM) population distribution, censustract data, Item 7. Define the StudyArea. Land use, topography, aerial photos,and area boundaries. Plans of the collection systemto definebranching,sizes,and slopes; Item 2. Define the System. types and generallocationsof inlet structures. Item3. Define the System Specialties.Flow diversions,regulators,and storagebasins. Maintenance.Street sweeping(descriptionand frequency),catchbasin Item 4. Define the System cleaning;ffouble spots (flooding). facility operatingdata; Item 5. Define the BaseFIow (DWF). Measureddirectly or through sewerage (composited BOD hourly variation and weekdayversusweekend;the DWF characteristics quantities,and quality). and SS results);industrial flows (locations,average period (6 months or longer) Item 6. Define the Storm Flow. Duly rainfall totals over an extended continuousrunoff encompassing the study events;continuousrainfall hyetographs, (BOD and SS) for the study hydrographs, and combinedflow quality measurements fully when and how taken). events;discreteor compositedsamplesas available(describe

(25.1,3)

654

CHAPTER25

MODELS URBAN RUNOFFSIMULATION

D, is larger than the specified 3. If the resultingwater depth of subcatchment detentiondepthDr, an butflow rateis computedusinga modifiedManning's equation.
tf Y _

!Ep,
n

- Do),/,s,r,

(2s.rs)
(2s.16)

and

Q-:

vw(D2 - Dd)

where V:thevelocity n : Manning's coefficibnt S : the ground slope IV : the width Q. : the outflow rate 4. The continuity equationis solvedto determinewater depthof the subcatchments resulting from rainfall, infiltration, and outflow' Thus

D,.L,:Dz-ftLt
whereA is the surfaceareaof the subcatchment.

o..

(2s.r7)

are comuntil computationsfor all subcatchments 5. Steps I-4 arerepeated pleted. 6. inflow (OJ to a gutter is computedas a summationof outflow from tribu(Q..,) and flow rate of immediate upstream gutters tary subcatchments (Qr,)

Q^:2Q-.,t2Qr.,

( 2s .18)

7. Theinflow is addedto raisethe existingwater depth of the gutter according to its geometry.Thus O',

Y.:Y,+7M

Q5.19)

where Yr, Y, : the water depth of the gutter A, : the mean water surfaceareabetweenY, and Y, 8. The outflow is calculatedfor the gutter using Manning's equation:
y :
1 ''A l -O '
fL

Pz/t51/z

(2s.20) (2s.2r)

and* where

Qr:

VA"

R : the hydraulic radius S; : the invert sloPe arcaaI Y1 A" : lhe cross-sectional

9. The continuity equationis solvedto determinethe waterdepth of the gutter resultine from the inflow and outflow. Thus

- oJf, Yt+^t: Y, t (Qr^

(25.22)
)

25,1 URBAN STORMWATER SYSTEMMODELS

655

t.0. Steps6-9 are repeateduntil all the gutters are finished. 11.The flows reachingthe point concernedare addedto producea hydrograph
coordinate alongthe time axis.

12. Theprocesses from 1 to 11 arerepeated in succeeding time periods until the

completehydrographis computed. Three general types of output are provided by SWMM. If waste treatment processes are simulatedor proposed, the capital,land, and operationand maintenance costs are printed. Plots of water quality constituents versustime form the secondtype of output. Thesepollutographs are producedfor severallocationsin the systemand in the receiving waters.Quality constituents handledby SWMM include suspended solids,settleable solids,BOD, nitrogen,phosphorus, and grease. The third type of outputis hydrologic.Hydrographs at any point, for example, the end of a gutter or inlet, are printed for designated time periods. The StatisticsBlock will provide frequencyanalysisof storm eventsfrom a continuoussimulation.

University of Cincinnati UrbanRunoffModel(UCURM)

The University of cincinnati urban runoff model (ucuRM) was developed by the Division of WaterResources, the Departmentof Civil Engineering,of the University of Cincinnati.26 A flowchartis reproduced in Fig. 25.11.The programconsists of three sections:(l) MAlN-infiltration and depression storage,and two subroutines, (2) GUTFL-gutter flow, and (3) PIROU-pipe rouring.It is similar to the EpA model and divides the drainagebasin into subcatchments whose flows are routed overland into gutters and sewerpipes. The rainfall is read in as a hyetograph.The infiltration and depression storageare summedand subtractedfrom the rainfall to give overlandflow. This is routed through the gutter system to storm water inlets and the pipe network. Starting at the upstreaminlet, the flows are calculatedin successive segments of the sewersystem,including discharges from inlets, to producethe total outflow. The drainage areais divided into small subcatchments with closely matched characteristics. The rainfall data are introduced and the infiltration is computedfor eachsubcatchment. Principal elementsof the modelingprocessfollow: 1. It is assumedthat runoff begins wheneverthe rainfall rate equals the infiltration rate and the mass of precipitation balancesinfiltration. The equations representing theseconditionsare

t.I /:-krn1
and

I |

Qs.zt1

t*.'#0

- e k): mi(r) +

+ + r) - ,rrlr]-1 {;trr ffiliv

(2s.24)

MODELS SIMUI-ATION RUNOFF 25 URBAN CHAPTER

,/ '1. Read Data RainfallintensitY 2. Infiltration constant storage caPacitidr 5. rj"p.o.ion

Infiltration (Horton's equation) Depression, supply storage

t
/ |

/-----ffi

Data 1 Slop 2. Roughness J. Lenl ;th 4. Widrh

il flow Overlan 1. Detention 2. Dis:harge

ReadData 1. Layout 2. Length Gutter flow input 1. Overland put 2. Gutterin.. o (continuitY) 3. Discharg Gutters '@

1. SloPe 2. Roughness 3. Diarieterorheight 4. Length

1. 2. 3. 4.

Flow routlng Gutter input Average velocitY (Manning's) Time offset AddoffsethYdrograPh and new qutter input

Pipes

Print hydrograPh

--16A \ -,/

Figure 25.11 UCURM model flowchart' (ATterPapadakisand Preul'26)

25j

URBANSTORMWATER SYSTEMMODELS

657

* where mi(I) : i(1) : k : "fo: f": DT : t :

the massprecipitateduntil time r (in.) the ordinatesof rainfall intensity curve the decayrate of infiltration (units/min) the initial infiltration capacity(in./hr) the ultimateinfiltration capacity(in./hr) the time incrementof rainfall intensity curve the time to intersectionof rainfall curve and infiltration curve x : arrincreinent of DT

The infiltration curve is computedfrom the equationsand t, I, andx arc stored. 2. Surfaceretentionis relatedto depression storage by an equationderivedby Linsleyet al..2s (25.25) s : (t _ f)e-@-F)/sd where 57 : P : F : i : f : s: the total depression storage(in.) the accumulated rainfall in storage(in.) the accumulated infiltratioh (in.) the rainfall intensity (in./hr) the infiltration (in./hr) the surfaceretention (in./hr)

The infiltration and surfaceretentionare subtracted from the rainfall intensity to yield the runoff. 3. The hydrographof the overlandflow is derivedby solving

- T+?:
where

,/,Dinlt. *ry ou(#)'f'''

(2s.26)

(25.27) D" : (0.009'79n'o'6ro6lo'6)fso'3 Dr.z : the detention storageat the beginning and end of time interval I (in./unit area) : rr, rz the overlandflow supplyat the beginningand end of time interval t (cfs/min) n : Manning's coefficient l: the lengthofoverland flow s : the slope (ftlft) (in./hr per unit area) 4 : discharge

4. For the initial time increment, q1 : 0 and Dr : 0 are substituted, D, is calculated,and q, is found from

q: yfnl ,,,,p,,,1r * o.o(S)'l','

L \D"/ )

es.z8)

where the symbolsare as previously defined.The determinedD" and q, becomeDt and qt. The overlandflow hydrographis derivedby repeating this cycle.
I

658

MODELS SIMULATION RUNOFF 25 UEBAN CHAPTER 5.Thegutterflowiscomputedusingthecontinuityequation

60 *6!7: At 6x

q,

(2s.2e)

where Z is the width of the water surface' Thetelm(ayla|risneglectedbecausethechangeinthedepthofthe to time. After integration,the equation gutterflow is very small with respect becomes (25.30) Q: q"L * Qo where Qs : L : q": Q: upstreamgutter contributions the length of the gutter (ft) the overlandflow from the hydrograph the flow from the gutter system

OF URBAN RUNOFFMODELSAND METHODS TABLE 25.9 COMPARISON

Degree of sophistication of surface flow routing Low Degree of sophistication of pipe flow routing Low Accurate modelingof surcharging No Flexibilityof modeling of storm draln components Low

Model

Surfaoe routing
Peak flows only Yes Yes

Pipe flow routing Peak flows only No In combination with surface In combination with surface Yes No Ye$

Quality routing No

Ratiorlal method Chicago Unit hydrograph STORM RRL MIT EPA-SWMM Cincinnati HSPF ILLUDAS
Sour ce;,}/ialifred.sftef

No No

Moderate

NA Low

NA No

NA Low

:"*
Yes Low Low Low

Yes

Yes Yes Yes

No

Moderate High High

No Yes Yes Yes No

Lowmoderate NA Moderate

No NA Yes

Low NA High

Yes Yes

Yes Yes

High Moderate

Low Moderate

No No

Low Low

(ucuRM)

Yes

Modetate

High

No

Low

I,age13nd Smit!'-8

25.2 URBAN RUNOFFMODELSCOMPARED

659

25.2 URBANRUNOFF MODELS COMPARED

Severalquantitative comparisonsof the RRL, SWMM, UCURM, ILLUDAS, and STORM models have been reported in the literature. A qualitative comparisonof severalwas preparedby Lager and Smiths and is shownas Table 25.9. Table 25.10 provides a bullit matrix showing componentsof most of the same models. Other comparisonsinvolve quantitative analysisof results of the models when applied to actual guagedstorm events.One of the first was an applicationby Heepsand Mein6 of three modelsto two urban catchments in Australia for a total of 20 storm events. A similar statisticalcomparisonof the samethree modelsapplied to 12 stormsover eachof three urban watersheds was performedby Marsaleket al.7Significantresults Another quantitativecomfrom theseindependent evaluations are summarizedhere. parisonis providedby Huber.27 The Heepsand Mein conclusions of model performanceare:6

1. The degree hasa significantinfluenceon the of subdivision of the catchment

peak dischargepredicted by each of the models.The RRL and SWMM methodsgive lower peaks and the UCURM gives higher peaks for finer subdivision. The UCURM containsseveraldeficiencies. The major ones,the effectsof predicted storages hydrographs, arelhat depression which canbe seenin the

Explicit modeling of insystem Treatmenl modeling storage No

Receiving model available

Degree of calibration/ verification required

Simulation period Individual storms Individual storms Individual storms Long term

Availability Nonproprietary

Documentation Good

Data requirements Low

NA NA NA

No

Usually not verified

Moderate High

NA No

No No

Nonproprietary Nonproprietary

Fair Fair

Moderate Moderate

No

Yes

No

Low

Nonproprietary

Good

Moderate

No NA Yes

NA NA Yes

No No Yes

Moderate Moderate Moderate

No No

No No

No Yes

Moderate High

Moderate

Individual storms Individual storms Individual or conttnuous storms Individual storms Individual storms or long-term Individual storms

Nonproprietary Proprietary Nonproprietary Nonproprietary Nonproprietary Nonproprietary

Good Fair Good

Moderate Moderate Extensive

Fair Fair

Extensive Extensive

Good

Extensive

uer6o:d lelnduoC elqElrE^E

o
f q) c

a J

eun-lEeu lo4uoc
u6;seg suollelnduroc o a a a o a o a
a a a a a

o a o o
a a

o a o

o
al,

ollrl osooqcuEc lBruolur uorlelnurs snonurluoc

6urnreceg uonelnursI1r;enb ralenir 6utntecag uollElnurs /v\ollJlE/v\

^Uleng srxrolsuoo^iqeq acueleq

luouleoJl roleMelsEM
o 3 o) o

a a a a a
a a a

^lrlpno suollgEar
rnocs pue uorteluor1jrpes 6urlnor r{1r1eng ,{Ulenb Jolel rulols fllenb reqleerirfuO sa!iloolaA sluud e6els sluud

a
J

tJJ F

o
a a

6
n

o o z 9)
F

a a a o a a o
a

a a o

a a

o a

z u z
(L

e6eroig suorlels6urdr-un3

a
a

E
o 3 (l) U)

suolsroAro puebur6reqcrng rvro;; ernsse.rd

pue uerlson loJluocMou uJEerlsu/v\op sloitlss ut Oupnorlrol3 a a
a

J UJ

o
I

a a a a
a

o o a o
o a

a o a o a o

o z
l (f,

suilols uoo/\laq acuelEqralEM sEale snotruoq uroJllJounH

a a a

a o

o a a

z
dl l t!

cc
c

uro,rl ssoJe snorrugour llounu

o E
o
G'

llau$ ous srlder6ole,{r.l lerolos}o }ndul

ill\Ol,lOt{}EO/tA ruO

o
z
U) (E

a o
a

a a
a o

o a
o

a a
f

a o
o a

o
!

sMollur lueuqclec eloruny\

o o
d
N

':
! E

u,l E

p
q

COMPARED 661 MODELS RUNOFF 25.2 URBAN are assignedfull when the rainfall intensity falls below the infiltration by infiltration. The are not depleted storages capacity,and that depression (difficult to obtain from intensity rainfall the of values useof instantaneous errors. volume cause can recordercharts) ' 3. The SWMM was the model with the best overall performancebut at the ' expense of large computerstorageand time requirements. 4. The RRL model predictedpoorly for stormsin which perviousrunoff was well for many other types of storms. significantbut performedreasonaf;ly of Stall and Terstriep'e those general, support in The results, models is the prediction of noncontinuous using problem with 5 . A major by use of the aggravated further problem is This antecedeniconditions. is virtuparameters the prediction of which for equation Horton infiltration ally impossible. usingthe samethree modelsfor three watersheds The Marsalek study results,T indicated that the SWMM model performed Maryland, in Illinois, Ontario, and both thesemodelswere more accuratethan and model slightly better than the RRL studied. watersheds the small the UCURM model for usedby Marsalekin the urban watersheds of provides descriptions TabIe25.11 timesto calculated and observed of comparisons Typical the runoff modelevaluation. provided inFig.25.12 are models three the for runoff volumes peak,peak flows,and in Kingston, Ontario. for the Calvin Park watershed to developthe qualitative comdescribed information the Marsaleket al. used conclusions:7 following at the and, arrived parisonin Table 25.12 modelsfor urbanrunoff, suchasRRL, SWMM, 1. Uncalibrateddeterministic betweenthe simulatedand good agreement yielded a fairly and UCURM, of small size. catchments urban typical on events runoff measured peak flows were within simulated percent the of 60 about 2. Onthe average, wasfound for the scatter same the About values. percent the measured of 20 betweenthe The agreement volumes. peak runoff and times to simulated calibramodel by improved further be could values and simulated measured tion. 3. Wherr comparing the entire simulated and measuredhydrographsusing wasfound goodfor SWMM, goodto fair the agreement statisticalmeasuies, caseof UCURM' the in fat and of RRL, in the case TABLE25.11DESCR|PT|oNoFTESTURBANWATERSHEDSUSEDFoRTHE
OF URBAN RUNOFFMODELS EVALUATION

Testurban drainage basin

Oakdale Calvin Park Gray Haven

Location
Chicago, IL Kingston, Ontario Baltimore, MD

Area (acres) 13.0 89.4

zt,J

lmoerviousness
4f .6

Land use Residential Residentialand institutional Residential

27 52

PART SIX

STATIST]CAL METHODS

Chapter26

and Statistics Probability

r Prologue
The purposeof this chapteris to: . Introducethe basictenetsof probability theory as appliedto random,hydroon the relativefrequency definition logic variables,with particular emphasis to the frequencyanalysisprofoundational of probability-a conceptthat is many other chaptersof this throughout in Chapter27 and cedurespresented text. . Describecommonprobability distributionsand showhow they are appliedto hydrologicphenomena. . Relate the fundamentalsof probabitity theory to hydrologic designcriteria in Chapter16, Section16.3. described . Acquaint the readerwith the theory behind linear regression and show how watershed predict how a study to in hydrology is used this powerful technique to past watershed of the responses examining by will respondto somechange in watersheds other similar of responses inputs or by statisticallyscrutinizing subject watershed. the for predictive equation order to developa ' Show how to transform many hydrologicvariablesthat havenonlinear relationships into new variatesthat can then be analyzedby performing linear on the transformedvariates. regression . Providethe theoreticaland practical foundationnecessary to fully capitalize the time-series 16 and in Chapter principles discussed on the hydrologicdesign 22. in Chapter procedures described modeling analysis*and commonly, the study of hydrologyis undertakenby readerswho lack the prerequisite backgroundin principles of statistics,probability theory, and frequency early in the review thesesubjects most hydrologycourses As a consequence, analysis. to methods on ;tatistical include chapters Practically all hydrology texts schedule. probability distribuprobability theory, principles of statistics, summarizethe basic time-seriesanalysis, tions, bivariate and multiple linear correlation and regression, at the end of this of this material placement the Thus,despite and frequencyanalysis. the material or will study background has this that the reader text, the authorsassume

672

AND STATISTICS 26 PROBABILIry CHAPTER with an understanding in Part Sif, prior to beginninga study of Part Three. Readers analysis,and the basicsof probability distribution of statisticalmethods,regression functions may wish to turn directly to Chaptet27.

ANALYSIS AND STATISTICAL VARIABLES 26.1 RANDOM

that isn't sufflciently exA random variable is one that demonstrates.variability havethis tendency,apMany hydrologicphenomena plained by physicalprocesses. or driven by some other pearing at times to be fully subjectto chancethemselves, of closelyrelated factor. In practice,hydrologistsoften analyzeproblemsas systems For example,precipitation is often connectedrandom and deterministicprocesses. of the complexityof understandstatisticallyas a randomvariablebecause evaluated processes are known to drive the precipitation that the atmospheric modeling ing and on the other hand, is viewed precipitation, from the results Runoff that system. of the majority arethe nucleus that analogs rainfall-runoff the using deterministically, this textbook. of Hydrology relies heavily on principles from probability theory, statistics,and generaWhole texts on frequencyanalysismethods,stochastic information analysis. available are analyses regional and variance, of and analysis regression of data, tion are of the principles.l'2Many hydrologicprocesses containingthorough descriptions sense. probabilislic in a only explained and interpreted can be that they complex so Hydrologic eventsappear as uncertaintiesof nature and are the result, it must be The inforwith randomor stochasticcomponents. of an underlyingprocess assumed, is containedin recordsof hydrologicobservamation to investigatetheseprocesses observed tions.Methods of statistical analysisprovidewaysto reduceand summarize data, to presentinformation in preciseand meaningfulform, to determinethe underphenomena, and to makepredictionsconcerning of the observed lying characteristics future behavior.Statisticalanalysisdealswith methodsfor drawinginferencesabout the population basedon examinationof samplevaluesfrom the population. These range,distributionwithin inferencesincludeinformation aboutthe central tendency, the range, variability around the central tendency,degreeof uncertainty, and frequencyof occurrenceof values. Statistical analysisinvolves two basic sets of problems, one descriptive, the otherinferentiat. The former is a straightforwardapplicationof statisticalmethods, little risk. The inferential problem,however, and representing requiringfew decisions of the methods requires an understanding risk, and some bearing decisions entails most common The predicting estimating. and in involved the dangers and employed when only the whole classof possibleoccurrences inferential problem is to describe portion population and the the is class The whole portion has been observed. of them a is the sample. observed understudyare continuousif they may take in the process The randomvariables on all values in the range of occurrence,including figures differing only by an infinitesimal amount; they are discreteif they are restrictedto specific,incremental values.Distribution of the variablesover the rangeof occurrenceis definedin terms of the frequencyor probabitity with which different valueshave occurredor might occur'"

26.2 coNcEPTSOF PROBABILITY 673

26.2 CONCEPTS OF PROBABILITY

The laws ofprobability underlieany study ofthe statisticalnature ofrepeatedobservations or trials. The probability of a singleevent, say Et, is definedas the relative number of occurrencesof the event after a long seriesof trials. Thus P(E ), the of the sameeventin N trials if N probability of eventEr, is nrfN for n, occurrences n, is thefrequency. and nr/N the is sufficiently large. The number of occurrences relatiw freQuenffobabilities and the rules governingtheir manipurationare known In the familiar coin-tossingexperiment,P(heads) : intuitively or from experience. P(tails) : |. Eachoutcomeof a singletoss(a trial) has a finite probability, and the is 1. Also, the outcomesaremutually outcomes sumof the probabilitiesof all possible exclusive;that is, if one occurs,saya head,then a tail cannotoccur.In two successive tests,there are four possibleoutcomes-HH, TT, HT, TH-each with a probability ofthe other one, probabilitiesfor eachtrial is independent of|. In this case,because trial) : L,x L : j. Again, the eachoutcomeare foundby P(first trial) X P(second sum of the probabilities of the possibleoutcomesis 1. Note that the probability of gettingexactlyoneheadand one tail duringthe experiment(without any regardto the is P(HT) + P(TH) : *. order) Summarizingthe rules of probability indicated by coin tossing,we find the following:

1. The probability of an event is nonnegative and never exceeds 1. 0<P(E)<1

( 26.r )

2. The sum of the probabilitiesof all possibleoutcomesin a singletrial is 1.

s
I

.1J

P(8,) : 1

(26.2)

andmutually exclusiaeevents 3. The probability of a numberof independent events. is the sum of the probabilitiesof the separate P ( 4 U E 2 ) : P ( E y )+ P ( E 2 )

(26.3)

The probability statement,P(Et U Et), signifies the probability of the "the probability of Et ot Er." union of two eventsand is read or in eventsoccurring simultaneously 4. The probability of two independent is the product of the individual probabilities. succession P(q ) E,) : P(E) x P(E) (26'4)

P(& a E ) is called probability of the intersectionof two eventsor ioint probability and is read "the probability of El and Er." or mutually Considerthe following exampleof eventsthat are not independent exclusive:An urban drainagecanal reachesflood stageeach summer with relative freqlrencyof 0.10; power failuresin industriesalongthe canal occur with probability

674

CHAPTER26

AND STATISTICS PROBABILITY

of a powerfailure for showsthat whenthereis a flood the ch'ances of 0.20; experience are statements probability The 0.40. to whateverreasonare raised P(power failure) : P(P) : 0'20 P ( f l o o d ) :P ( F ) : 0 ' 1 0 P(no powerfailure) : P(P) : 0'80 P(no flood) : P(F) : 0'90 P(power failure given that a flood occurs) : 0'40 is called a conditionalprobability. It signifiesthejoint occurrence The last statement of events and is usually written P(P I F).. Rules 3 and 4 no longer are strictly If Rule 3 applied,P(F u P) : P(F) + P(P) : 0.3. If the eventsreapplicable. theionditional probability P(P I F) would gqualthe marginal independent, maineO if the probability of either is not probability f (p). T'nor the eventsare independent i'conditionedby" or changed by knowledgethat the other has occurred'For independent events,the joint probabilitieswould be P(F)P):0.1 x0.2:0'02 P(FaF):o.tXo.8:0.08 P(FnP):0.9X0.2:0.18 The probability of a flood or a power failure during the summerwould be the sum of the flrst three joint probabilitiesabove. P(FUP):P(F nP)+ P(F.P)+p(F lP):e'23 of conditionalprobabilhowever,from the statement The eventsare dependent, : occurwith probability will power failure 0'1, a ity: When a flood o""u,. with P(F) : 0'1 x 0'4 : 0'04 : F) x P(P I 0".4, and true joint probability is P(F') : P(F) + P(P) P(F a P). The proUaUitityof the union is then P(F U P) P(F ) P) : 0.1 + 0.2 - 0.04 : 0.26' Note the contrast: P(F U P) : 0'30 for mutually exclusiveevents events P(r U P) : 0'28 for joint but independent : 0.26 otherwise P(F u P) The new, more generalrule for the union of probabilitiesis s. P(& u E,) : P(81) + P(E,) P(h . E2) and a sixth rule shouldbe addedfor conditional probabilities: E2) 6_ " ' P ( E , , l E P(Er r ) : -nw

P(FaF):o.gxo.8:0.72

(26.s)

{26.6)

26.2 CONCEPTSOF PROBABILITY

675

be exceeded.Becausethe probability of any single, exact value of a continuous "occur" can also mean the level will be reachedor exceeded. In the variableis 0.0, years' in 10 once average the on exceeded lhus or long run, the levelwould be reached return period* Z in yearsis definedas the average T : 1 P(Ft: (26.7)

and the following generalprobability relation.hold: in any year 1. The probability that F will be equalledor exceeded

P@):+
in any year 2. The probability that F will not be exceeded

(26.8)

P(F):l-P(F):l-+

(26.e)

in any of n succes3. The probability that F will not be equalledor exceeded slveyears / l\n

: P(F)': ( t - ;l p,(F)x &(Fl x . . . x P,(F) \ T l

(26.10)

at least 4. The probability R, called risk, that F will be equalledor exceeded years in n successive once

R : 1 - ( r - + l : 1- I P ( F ) F
WITH VARIOUSDEGREES TABLE 26,1 RETURNPERIODSASSOCIATED DESIGNLIFE OF RISKAND EXPECTED Expecteddesignlife (Years) Risk

(26.rr)

with variouslevelsof risk. Table 26.1showsreturn periods associated

e/.)
75 50 40 30 25 20 15 10 5 2
I

10 4.O2 2.00 7.74 3.43 10.3 4.44 14.5 6.12 7.46 " 17.9 22.9 9.47 31.3 12.8 48I . 19.5 98.0 39.5 248 99.5 498 198.4 6.69 t4.9 20.1 28.5 35.3 45.3 62.0 95.4 195.s 496 996

15

100 14.9 29.4 39.7 56.5 70.0 90.1 t23.6 190.3 390 990 1992 18.0 36.6 49.5 70.6 87.4 1r2.5 154.3 238 488 t238 2488 35.6 72.6 98.4 140.7 174.3 224.6 308 475 976 2475 : 4975 72.7 144.8 196.3 28r 348 449 616 950 1949 4950 9953

11.0 22,t 29.9 42.6 52.6 6'7.7 90.8 r42.9 292.9 743 1492

to denotethe recipro* The terms return period arldrecurrenceinterval are usedinterchangeably of exceedence. probability cal of-the.annual

676

CHAPTER 26 PRoBABILITANDSTATISTICS

EXAMPLE 26.1 If Zis the recurence interval for a flood with magnitudeQ*findtbe probability (risk) that the peak flow rate will equalor exceedQ" atleastoncein two consecutive years. Assumethe eventsare i4dependent. Solution. The solutionis easilyobtainedby substitution into Eq. 26.11.To assistin understapding the equations,an alternativederivation follows. The four possibleoutcomesfor the tyo years are: a.' nonexceedance in both years b.' exceedance in the first year only c.' exceedance in the secondyear only d: exceedance in both years Because thesefour representall possibleoutcomes, the probability of the union o f a , b , c , a n d d i s1 . 0 , o r f r o m E q . 2 6 . 2 , P ( a U b U c U d): l.0.Exceedanceinat least one year is satisfiedby b, c, or d, but not a. Thus the risk of at least one exceedance is P(b U c U d), which is the total less the probability of a. From Eqs.26.2and26.3,we find that ' z-yearris: k P(b U c U d):1P(a) From Eq. 26.3, we find that P(a) : P(Q < Q"inYear I) x P(Q I Q"inYear 2)

:(t-l-)ft-1-) r/ \ r/\
and

R i s k :r - P ( a ) - 1 - ( t - 1 - ) '
\

.r./

rr

EXAMPLE 26.2 What return period must a highway engineeruse in designinga critical underpass drain to accept only a 10 percent risk that flooding will occur in the next 5 years? Solution R : 1 - (' -;)" .10: 1-

Z : 48.1years

('-i)'
IT

26.3 PROBABILITY DISTRIBUTIONS

Randomvariables, either discreteor continuous, are characteizedby the distribution of probabilitiesattached to the specificvaluesthat the variablemay assume. A random variablethroughoutits rangeof occurrence is generallydesignated by a capital letter, and a specificvalueor outcomeof the randomprocessis designated by a small letter.

26.3 PROBABILIry DISTRIBUTIONS

677

^
b 0.2 5 ^ o.l

= 0.0s P(0) P(1= ) s.15 P(2)= 0.2s P(3)= g.2s

= s.15 P(4) = s.1s P(s)

P(6) = 9.63 P(7) = s.s2

Number of cloudY daYsPer week, x

per week' of cloudydays distribution Figure26.L Probability the value For example , P(X : x,) is the probability that random variableX takeson of the p(x,). distribution probability the shows 26.1 Figur" x,. A shoiter version is of number the because distribution a discrete ft is number of cloudy daysin a weet. had day a taken, were frequencies relative the which daysis exact;in ihe rlcord from to-be describedas cloudy or not. Observethat eachof the seveneventshas a finite probability and the sum is 1; that is,

) r(-t,) = t
Another important property of random variablesis the cumulativedistribution in X is lessthan or equal function, CDF, definedur ttr" ptotubility that any outcome to a stated,limiting value x. The cumulative diitribution function is denotedF(;r). Thus F(x):P(X=x)

(26.12)

monotonicallyfrom a lower limit of zeroto an upperbound andthe function increases CDF of the numberof cloudy daysin a week derivedfrom isthe 26.2 Figure of unity. fig. Z6JUy taking cumulativeprobabilities.The function showsthat the probability there thai the numberof cloudy daysin the week will be 5 or less.Conversely, is gOqo This complementary is a 10 percentprobability that it will be cloudy for 6 or ! days. where3 G(x), called prolability is sometimes cumula-tive Q6'13) r(x) = P(X> x) G(x):1continuous variables present a slightly different picture. Figure 26.3 is the were grouped The observations r""oid of annualstreamflows. histogramof an 85-year in eachinterval falling number the and cfs 0 to 900 from into nine intervalsranging is to plot the alternative convenient A ordinate. left the was plotted as frequeniy on record streamflow the for cDF The ordinate. right by the relative frequencyas shown districontinuous the increase, observations of number is shownin Fig. 26.4.As the broken the limit, In the intervals. the of size the reducing by bution will be developed curvesof Figs. 26.3 and'26.4 wilt appearas thosein Fig' 26'5' There ls a difference between the ordinates of Figs. 26'3 and 26.5a. Since the to reconstitute with probability, it is convenient relative frequencyis synonymous

.7'
CHAPTER 26 PROBABILITYAND STATISTICS

vl

3 o.a ^ -' .F o.o

II

P(x)

.l

0.4

U o.2

0 1 2 3 4 5 6 7

0.05 0.15 0.25 0.20 0.15 0.10 0.08 0.02

0.05 0.20 0.45 0.65 0.80 0.90 0.98 1.00

0 Figure 26.2

1 2 3 4 5 6 Numberof cloudy daysper week,.r

Cumulative distribution of cloudy days per week.

RIR

b
il lt a q

I o '5

o.top

3 4 5 6 7 Mean annualflow, -r (100 cfs)

Figure 26.3

Frequency distribution

of mean annual flows.

26.9 PROBABILITYDISTRIBUTIONS 679

'

Q . o r -'" Y
o q

0.6

o I o > 04

(J

0.0 0

3 4 5 6 7 flow, x (100cfs) Meanannual

Figure 26.4 Cumulative frequency distribution of mean annual flows.

probabilitydisFigure 26.5, Csntinuous (a) probabilitydensity function tributions: function. distribution and(b) cumulative

680

CHAPTER26

PROBABILITY AND STATISTICS

histograrfrso that the area in each interval representsprobability; the total area containedis thus unity. To do this, the ordinatein eachinterval, sayn/N for relative frequencyor probability, is divided by the interval width, Ax.-The ratio nfN Lx rs the literally the probability per unit length in the interval and thereforerepresents on densityof probability.The probability n/N inthe interval is fepresented the average as AF(x), or F(x + Lxlz) - F(x - L'xl2).We CDF (beforethelimitingprocess) then can define
l\x) :

.. --;liln Ax-o Ax

AF(x)

dF(x)
dx

(26.r4)

which is called the probability densityfunction, PDF.3This function is the density (or intensity) of probability at any point; f(x) dx is describedas the differential probability. For continuous variables, f(r) > 0, sincenegativeprobabilitieshaveno meaning. Also, the function has the property that f(x) dx: 1
()61\\

which again is the requirement that the probabilities of all outcomessum to 1. Furthermore,the probability that x will fall betweenthe limits a and b is written

p(a-x-b):lu,r*,o*

(26.16)

Note that the probability that x takeson a particular value,saya, is zero;that is,
fo

ft-l dx : o

(26.17)

that nnite proUalilities are definedonly as areasunder the PDF which emphasizes betweenflnite limits. The CDF can now be deflnedin terms of the PDF as P(-* < X < x) : P(X= x) : F(x) : I f(u) du

t'

(26.18)

J_*

whereu is usedas a dummy variableto avoidconfusionwith the limit of integration. The area under the CDF has no meaning,only the ordinates,or the difference in to Eq. 26.16, can be For example, P(*r3 X = x2),which is equivalent ordinates. evaluated as.F(x) - F(xt). in integral form, there are that cannotbe summarized For discrete.distributions to the propertiesgiven in Eqs. 26.15, corresponding analogous arithmetic statements datetaken from a contin26.16, and26.18.In particular, the distributionof sampled uousdistributionis a specialcaseof discretedistributionsand canbe givenin the form Thus of arithmeticsummations.s

),f(',):
P(a-X<b):

t
x1= b

(26.re)
(26.20)

).f(r)
r 7 -

26.4 MOMENTS OF DISTRIBUTIONS

k

681

P(x<d:2fQ,)

(26.2r)

in the sample,/(x) is the probability of xr for each For a finite numberof observations outcomein the samplespaceand thereforeP(x,) : P(xr) : P(xt) : . . . : I /N. Hence/(x) can be replaced with P(x,) in Eqs.26.19,26.20,and26.21. 26.3 DXAMPLE Table B.1 containsthe areabeneatha "standardnormal" bell-shapedPDF.Because the distributionis symmetrical,areasare providedonly on one sideof the center.Use the distribution to determinethe valuesof 1 . P ( 0< z ' 2 ) . 2.P(-2=z=2). 3. P(z > 2\. 4. P(z< -1) Solution 1. P ( 0 = z < 2 ) : . 4 7 7 2 . ) F r o m s y m m e t r y ,P ( - 2 - z < 0 ) : Since P ( 03 2 3 2 ) : . 4 7 7 2 . P(-Z = z < 2) : P(-2 3 z 30) + P(0 3 z 3 2), thenP(=2 = 7 < 2) : .4772+ .4772: .9544. the 3. This is the areaunderthe curvein the right tail beyondz : 2.0. Because area right of center (z : 0) is .5000, P(z = 2) : P(z > 0) - P(0 = z = 2 ) ,o r P ( z > 2 ) : . 5 0 0 0 - . 4 7 7 2 : . 0 2 2 8 . 3 ) ,P ( z = - 1 ) : P ( z = 0 ) - P ( - l = z < 0 ) ' B y 4. Fromthe s o l u t i o n t(o = z -1) : s y m m e t r y , P ( - 1 z = 0 ) - P ( 0 3 z s ! ) : . 3 4 1 3 , a n d P (= : .5000 .3413 .1587. rr

16.4MOMENTS OF DISTRIBUTIONS
The propertiesof many random variablescan be definedin terms of the momentsof the distribution. The moments representparametersthat usually have physical or geometripalsignificance.Readersshould recognizethe analogybetweenstatistical momentsand the momentsof'areasstudiedin solid mechanics. The rth moment of a distribution about the oriein is definedaso

tri : p::

x'f(x) dx

(26.22) (26.23)

2*,rr,r:12,,,

The first moment aboutthe origin is the mean,or as it is commonly known, the the distancefrom the origin to the centroid ofthe distribution average. It determines frequencyfunction. The prime is normally usedto signify momentstaken about the origin, but the mean is often written as p insteadof y'.

682

CHAPTER26

AND STATISTICS PROBABILITY

Moments can be definedabout axesother than the origin; the axis usedextensively in deflninghigher momentsis lhe mean or, as given above,the first moment about the origin. Thus

p,:
or tr,:

I
I

(x - rd'f\xl dx
- p)'

(26.24\
(26.25)

1<
;)-G,

Wheneverp', or p', are defined for r =l 1, . , the distribution/(x) is comto computemore than the first three rnoments; pletd defined.It seldomis necessary ieveral important distributionsrequire only two. The momentsare usedto specifythe of distributions that follow in the next parametersand descriptivecharacteristics are describedby combinadistributions of section.Becausevarious characteristics relationsare occasionfollowing the tions of the momentsaboutthe meanand origin, ally helpfull'a: Ft: 0 -p' Pz: FL
ttt : tJ"'z

- 3plr, * 2p'

(26.26) (26.27) (26.28)

CHARACTERISTICS 26.5 DISTRIBUTION

' , of probabilby the parameters of statisticaldistributionsare described Characteristics principal The moments. of the in terms ity functions, which in turn are expressed about orprobability grouping ofobservations the are centraltendency, characteristics skewand or observations; variate of the a central value;variability, the dispersion of asymmetryof the distribution.The theoreticalfunctions shownin ness,thedegree Fig.26.6 exhibit approximatelythe samegrouping about a central value,but/2 has a pronouncedright-skewwhilefi is much greatervariability thanfl, andf possesses symmetrical.

Symbol Gonvention
of statisticalprobof distributions,the usual sequence In introducingthe parameters are derivedfrom the distributionof sample lemswill be followed-that is, parameters of the population distribution' Summaof the parameters data and usedas estimates

Figure 26.6 Symmetrical and skewed probability distribution for continuousvariables.

CHARACTERISTICS 26.5 DISTRIBUTION

683

For example,the tion forms of integralsare usedto computemomentsfor samples. of the population best estimate as the it is used is designatedTand meanof sampledata parameters. population denote used to are Greek letters mean.By convention,

CentralTendency
The familiar arithmetic average,the mean, is the most used measureof central tendency.It is the first moment about the origin and is designated

r:lf*,
n i=t

(26.2e)

of the population mean p. The statistic 'Meansx is the best estimate mean-for example, the geometric mean the arithmetic other than : ,l> (Ilx,)-are alsoused. . . . : Two addi(x624 meani or harmonic x,)1/" T of the middle value is the which median, the are of central tendency tional measures in which the mode, and areas, into equal the distribution observeddata and divides variables continuous and in frequently most discretevariablesis the value occurring is the peak value of probability density.A11three are illustrated in Fig. 26.6.

Variability
deviation by the total rangeof valuesor by the average Dispersioncan be represented of statisticalimportanceis the meansquared aboutthe mean;however,the parameter deviation as measuredby the secondmoment about the mean. The parameteris by termed the varianceand is designated

c r : rn f

i=l

G , -p ) '

(26.30)

to is not known precisely and therefore it is necessary But the population mean /-{, computeinstead
n

s-:

Z\xt-xf

(26.3r)

- 1 in place of n in As the best estimateof o2, the quantity s2 is found using n of freedom loss of a degree the involves for this substitution Eq. 26.30.The reasoning text. of this scope by using 7 insteadof p, but a proof is beyondthe The squareroot of the varianceis a statisticknown as the standard deviation in the sameunits asthe variateand the (o or s), in which form variability is measured to interpretand manipulate.The coefficientof variation C", mean,and henceis easier useful in comparingrelative variability. definedas cf p, or sfi, is an expression

Skewness
A fully symmetricaldistribution would exhibit the property that all odd moments weightto either side would haveexcessive distribution,however, equalLero.A skewed of the centerand the odd momentswould exist. The third moment a is

o : ! n2 @ , - p ) '
i:l

(26.32)

684

CHAPTER26

AND STATISTICS PROBABILITY

The best-estimate of the third moment is computedby

u -

(n-r)(n-2)

Z-J

\xi-

x)'

(26.33)

The cofficient of skewness is the ratio afc3 and is estimatedby

n ---; - Q (-J

(26.34)

For syfnmetricaldistributions,the third moment is zero and C" : 0; for right skewness(i.e.,thelongtailtotherightside)C">0,andforleftskewnessC"<0.ThePDF is of forfi shownin Fig. 26.6has a right or positive skew.The property of skewness questionable from less than 50 sampledata statisticalvaluewhenit mustbe estimatdd points. EXAMPLE 26.4 Determinethe distributionparameters and comparethe distributionsof annualrainfall for the records shown inTable 26.2.

TABLE 26,2 ANNUAL RAINFALLFOR SELECTEDCITIES Annual rainfall(in.)

Year

Anniston, AL 48 49 )) 98 43 53 56 47 69 57 61 64 99 54 40. 47 58 44 44 64 44 51

LosAngeles, CA
9
lo

Richinond, VA.

t928 r927 t926 t925 r924 1923 1922 t92l 1920 1919 1918 t9t7 1916 1915 t914 t913 t9t2 191 I 1910 1909 1908 1907 1906

t9 9 8 6 15 20 11 o 18 8 23 t7 23 17 10 18 5
)4

43 44 38 31 47 49 52 31 51 40 4l 43
3 t

36
J+

38 36 37 43 34
JJ

l9 l5 21

49 47

FUNCTIONS DISTRIBUTION PROBABILITY 26.7 CONTINUOUS

685

Solution
Parameter
Mean, ;r Standarddeviation,s of variation,C, : s/V Coefficient C": a/s3 Coefficient of skewness,

Anniston
57.2 in. 15.5in. 0.27 1.69

LosAngeles
14.9 in. 5.9 in. 0.40 -0.16

Richmond

41.5 in. 6.7in. 0.16 0.16

Comments.(1) Anniston's record shows a high annual averageand a fair$ large right skew,caused variability. In particular,Anniston'sdistributionhas a pronounced valuesin this short period of record. (2) Los principally by two very large observed but a very largevariability and a slightly negative Angeleshasa small annualaverage uniform distribution:a relativelysmall variabil(3) the most has Richmond skewness. I I positive skewness. a slight ity and only

FUNCTIONS DISTRIBUTION 26.6 TYPESOF PROBABILITY

hydroMany standardtheoreticalprobability distributionshavebeenusedto describe not an is distribution theoretical that any processes. It should be emphasized logic approximates that a description process but only natural of the exactrepresentation and has proved usefui in describingthe observeddata. the under$ing phenomenon giving the PDF,mean,andvariance distributions, common summarizes Table26.3 {he presented in the table have experiencedwide The distributions functions. of the In textbookson statistics. standard in many and discussed derived and are application given. are distributions usbd most of the aspects follow, only material to the discreteprobability distributionsinTabIe26.3 The usesof binomial and Poisson eventsin which the outcomecanbe described generally random to those arerestricted and trials are independent the successive failure. Furthermore, or as a success either the In a sense, to trial.3'a trial from constant probability remains of success the techniques' or enumerating counting are distributions discrete common The binomial distributionis frequentlyusedto approximateother distributions, and vice versa.For example,with discretevalues,when n is large andp small (such the Poissondistribution.This is a thal np ( 5 preferably),the binomial approaches : (i useful in describringarrivals in is very and np) distribution single-parameter grows large, the binomial becomes and n queueingtheory. When p approaches I in the next section. described distribution normal from the indistinguishable

FUNCTIONS DISTRIBUTION PROBABILIry 26.7 CONTINUOUS

and the to be continuousrandom processes, Most hydrologicvariablesare assumed frequency as in sequences, fit historical to are used distributions continuous common for example(Chapter 27). Other applicationsare also important for continanalysis, The elementaryuniform distribution is the basis for computing distributions. uous

7
686 CHAPTER 26 PROBABILITYAND STATISTICS

^ ,-" ,
i l {
N H

+
'\-..7
a

,
C)N

n d l
d

, L t ,
I

.gb
> o
(UN

g b

G
,

B l
I

q ^

r
d

i
t v

B -'i.l
co. bl tl
H

.i l \ F t.x

+
x -:

Nl'\a T

,
tl
h

x el

a f f
tx o o

r-

dlJa , +' f

N I N t i - l xrnl -il

+ x J CO+

sr

\, {

B
l
a

+
l s l
-

-{.

o J o
I

Ir o z o ut a
U)

tr

(!

s \/ " x
rir

: " vr
x

" ' \/t :'

\/r "'

o x

8 vr '' R
\,r vt

8 vl ^
Vl

-\? .

^r
k

o ,? ,9,
vl x "' ' ' \/l

;
/\r
.,

M
d

u ;
6 ^

-r

l\t

IL

tr l
dl r

o o
o o

fr

a o
z o
E U'

o o
lt

o
E E

0)

o tu J o

(!

s
F

IL

a? (o
6|

UJ @

EE E ;F
= .E '.

EE a 6 .f

= i ' 5E

E "

E 3 E !
E E i 6 F[ E S EZ B i ' oO e F E

FUNCTIONS PROBABILIW DISTRIBUTION 26.7 CONTINUOUS

687

The wholebody of materialin the randomnumberssoimportant in simulationstudies. area of reliability and estimatingdependson derived distributionslike Student'sl, that follow concernthe more chi-squared,and the F distribution. The explanations The readeris referred common distributionsappliedin fitting hydrologicsequences. to standardtexts for more detailedtreatment.3-6

NormalDistribution
frequencyfunction, alsoknown be,ll-shaped The normal distributionis a symmetrical, many processes distributionor the natural law of errors. It describes as the Gaussian variations.The whole basisfor a large that are subjectto random and independent body of statisticsinvolving testing and quality control is the normal distribution. of hydrologic data, it has wide Although it often does not perfectly fit sequences data that do follow the with transformed for example, in dealing application, virture of the central limit by reliability in sample normal distributionand estimating theorem. The normal distribution has two parameters,the mean p, and the standard By a simple deviationa, for which 7 and s, derivedfrom sampledata, aresubstituted. only. function as a single-parameter be written the distribution can transformation, : : becomes PDF o dz,the Definingz \x t-i/o, dx

(26.3s)
and the CDF becomes F\z) :

-r*l-

e-"2/2du

(26.36)

The variablez is called the standardnormal variate; it is normally distributed with zero mean and unit standard deviation. Tables of areas under the standard normal curve, as given in Appendix B, TableB.1, serveall normal distributionsafter Given a cumulativeprobability,the deviatez is found of the variables. standardization in the table of areasand x is found from the inversetransform: x:p+za EXAMPLE 26.5 Assumethat the Richmond,Virginia, annualrainfall in Table 26.2 follows a noimal distribution.Usethe standardnormal transformationto find the rain depththat would havea recurrqnceinterval of 100 years. Solution. From example26.4,the meanis 41.5 in. and the standarddeviation is 6.7 in. This gives x: 4L5 + z(6.7) or x:7lzs

(26.37)

Equation 26.18 showsthat the areaunder the PDF to the right of z is the exceedence probabilprobability of the event.For the 100-yrevent ,F;q.26.7 givesthe exceedence Table 8.1 in From the figure accompanying tlI00:0.01. ity Pk): llT,:

688

CHAPTER26

PROBABILIry AND STATISTICS

Appendix B, F(z) : 0.5 - p(z) : 0.49, and z : 2.326 by interpolating the table' The expected100-yr rain depth is therefore x : 4t.5 + O.32O X 6.7 :57.1 in: The 100-yr event for a normal distribution is 2.326 standarddeviationsabove the mean. rt

Log-NormalDistribution
partly dueto the influence Many hydrologicvariablesexhibit a markedright skewness, other lower limit, and or some zero, greater than values phenomena having of natural will frequencies cases, In such range. the upper in theoreiically, beingunconstrained, distribunormal a follow may logarithms but their distribution, normal not follow the y : tion.7The PDF shownin Table 26.3 for the log-normal comesfrom substituting ln x in the normal. With p, andcy as the mean and standarddeviation,respectively, of the the following relations have been found to hold betweenthe characteristics y:r'7 variate transformed and the x variate untransformid

p:exp(p'y+412) oz:p,2lexp(d)_11 a : lexp(3fi) - 3 exp(fi) + 2lC3 C,:lexp(dr) - r1t'' C,:3C" * Cl

(26.38) (26.3e) (26.40) (26.4r) (26.42)

Also p, : lfl M, whereM is the median value and the geometricmean of the x's. opensthe extenthe,transformation usefulbecause The log-normal is especially Since both the distribution. normal of the uses applied and oI theoretical sive body only to comit is necessary distributions, two-parameter are log-normal normal and 26.38 and Eqs. solve x and variate untransformed of the variance mean-and pute the dislog-normal truncated or three-parameter on Information 26.39 simultaneously. literature.r'7 in the found be can tributions

lll) TYPe Gamma(and Pearson

The gammadistributionhaswide applicationin mathematicalstatisticsandhas been usedincreasinglyin hydrologicstudiesnow that computingfacilities make_iteasyto evaluate the gimma functioi insteadof relying on the painstakingmethod of using tablesof the incompletegammafunction that lead to the CDF, P(X < x). In greater useis a specialcaseof gamma: tbePearsonType/1L This distributionhasbeenwidely adoptedas the standardmethodfor flood frequencyanalysisin a form known as the log-pearson /11 in which the transform y :1og x is used to reduce skewness.8-r0 Aithough all three momentsare requiredto fit the distribution,it is extremelyflexible in that a zeroskewwill reducethe log-PearsonIII distribution to a log-normal and the pearsonType III to a normal. Tablesof the cumulativefunction are availableand variates A very important property of gamm-a will be explainedin a later section.lo'11 two such of sum the is that normals) aswell asnormal variates(includingtransformed syngenerating in is important variablesretains the samedistribution. This feature -thefie hy-drologic sequences.l''''

26J

DISTRIBUTION FUNCTIONS PROBABILITY CONTINUOUS

689

Distribution Gumbel'sExtremal
The theory of extreme valuesconsidersthe distribution of -the largest(qr smallest) The distribution of the nt observations occurringin eachgroup of repeatedsamples. extreme values taken from n, samples,with each samplehaving n2 observations, Gumbel was the first to dependson the distribution of the nrn, total observations. frequencies.la Chow has demonemploy extremevalue theory for analysisof flood log-normal with constantskewstratedthat the Gumbel distribution is essentiallya given 26.3 is in Table ness.ts The CDF of the densityfunction

P(X - x) : F(x) : exp{-expl-o(, - u)l}

(26.43)

a andu are given asfunctions a convenient form to evaluate the function. Parameters of the meanand standarddeviationin Table26.3.Tablesof the doubleexponentialare usually in terms of the reducedvariate,y - a(x - u).tuGumbel also has proposed (minimumannual) to fit instantaneous anotherextremevaluedistributionthat appears droughtflows.17'18

CDFsin Hydrology
Normal and Pearson distributionscan often be usedto describehydrologicvariables if the variableis the sum or mean of severalother random variables.The sum of a random variablesis approximatelynormally distributed.For numberof independent example, the annualrainfall is the sumof the daily rain totals,eachof which is viewed as a random variable. Other examplesinclude annual lake evaporation, annual pumpagefrom a well, annual flow in a stream,and mean monthly temperature. usedin approximatingthe distribuThe log-normal CDF hasbeen successfully tion of variablesthat are the product of powersof many other randomvariables.The the logarithm normally distributedbecause logarithm of the variableis approximately of productsis a sum of transformedvariables. Examples of variablesthat havebeenknown to follow a log-normal distribution include:

1. Annual seriesof peak flow rates. 2. Daily precipitationdepthsand stremflowvolumes(alsomonthly, seasonal,

and annual). rates. 3. Daily peak discharge 4. Annual precipitation and runoff (primarily in the westernUnited States). Earthquakemagnitudgs, 6. Intervalsbetweenearthquakes. 7. Yield stressin steel. 8. Sediment sizes in streamswhere fracturing and breakageof larger into smaller sizesis involved.
5.

The PearsonType III (a form of gamma) has been applied to a number of variablessuch as precipitation depths in the easternUnited Statesand cumulative watershed runoff at any point in time during a given storm event.The transformed the CDF for annualflood peaks.If Type III is most usedto approximate log-Pearson the skew coefficient C" of the variable is zero, the CDF reverts to a log-normal.

690

AND STATISTICS 26 PROBABILIry CHAPTER of concrete usedwith monthly precipitationdepthandyield strengths It hasalso.been members. A useful CDF for values of annual extremeis the Gurnbel or extreme value probability of distribution.The meanof the distributionhas a theoreticalexceedance streamshave peaks in natural years. Flood 2.33 T of interval 0.43 and a recurrence 2.33-year with means including disffibution, to this conformance exhibited strong variptraight-line fit for Gumbel a produces paper that Graph recurrenceintervals. is A sample extremes. annual of graphical tests for useful and ables is a available peak rates, discharge annual peak to applie! has been 27 The CDF shownin Fig. .2. wind velocities,drought magnitudesand intervals, maximum rainfall intensitiesof events. given durations,and other hydrologicextremesthat are independent

AND CORRELATION REGRESSION LINEAR 26.8 BIVARIATE

proceduresare widely used in hydrology and other sciCorrelation and regression The premiseof the methodsis that one variableis often conditionedby the ences.te value of another,or of severalothers,or the distribution of one may be conditioned by the value of another. Just as there are probability density functions (PDFs) for evaluatingthemarginal probability of a variable(seeSection26.2), so also are there PDFsforlhe conditional probabilities (also describedin Section 26.2) of variables. The conceptis illustrated in Fig. 26.7. For two variables,the bivariatedensityfuncfor eachvalueof x' The one tion,/(y li,), ptottedin the vertical on the frgure,changes : xr. Different distributionsmight occur shownappliesonly to variationsin y whenx for other valuesof x. A measureof the degreeof linear correlationbetweentwo variablesx and y is thelinear correlation cofficient, P*,y. Avalue of p',, : 0.0 indicatesa lack of linear

line Pylr regression

with conditional probability Figure 26.7 Bivariate regression function.

AND CORRELATION691 REGRESSION LINEAR 26.8 BIVARIATE correlation andp,,, : + 1.0 meansperfect correlation.The correlationcoefficientis found from
u - , , : -

cov(x, y)
(l*ay

cr.v
aroy

(26.44)

(seeEq. 26.30), and where o, ando, are the variancesof eachvariable,respectively, as defined cov(x, y) is the covariancesharedby the two variables,

y) : c,.,: cov(.r,

The samplecorrelationcoefficient,r : s,.rfs*s* is usedto estimatep',r. The sample covarianceis found from the squareroot of

f _f _Q'-

dvdx p)(y - p,)f(x,v)

(26'4s)

: s?,,

(26.46)

line shown on Fig. 26.7 is derived to passthrough the mean The regression so that for any given value of x, the mean value of y I x the distributions, of values line. The standarderror of the by the regression (read"y givenx ") can be estimated the conditional distributionsat through line drawn the y is depicted by x of estimate I a distanceof one standarddeviationfrom the mean.If the conditionaldistributionsat of the conditional are normal,it canbe shownthat the meanvalue,&y1,, all x-values distribution is related to the meansof .r and H or
a & , 1 , : l t ', * ( p 2,( * and the variance is tr,)

t26.47)

4t.
where

: #[' "-

, @- p)'1

(26.48)

a?: 40 - p')

(26.4e)

which is the variance of the residualsof the regression.Just as the mean of the distribution requiressubstitutionof the given value of x into Eq. 26.47, so also does to A the standard thevariance ,Eq.26.48. Whenthe valueof x in Fig. 26.7is setiequal error of the meanis
O-t=: ae -----7

(26.s0)

VN

betweeny and .r as the linear dependence Equation 25.47 is linear and expresses fixed valuesof x' Also, for y computed can be of value slrownin Fig. 26J.The mean predict values of y with less the can one is significant, if the correlationbetweenthem the fraction Eq. 26.49, from fact, y In alone. of distribution error than the marginal is regression the by or accounted explained of the original variance

o":t-*

(26.5r)

692

CHAPTER26

PROBABILIry AND STATISTICS

line is It can be seenalso from Eq. (26.47) that the slopeof the regression
cy

PA:

tl,yl,

lLy

r - rr^

(26.s2)

then p itself is the slope,where or, ifx andy are standardized, - p,r)/a, }rnt. (26.s3) p = (x - t*)/o, to coverhigher-order,multivariatedistributions. The bivariatecasecan be expanded

EQUATIONS REGRESSION 26.9 FITTING

ing value of x. The line to be fitted is -l Bx; Y,: a

(26.s4)

N 6 I

, ? ; xbo
J . F

Year
A1 /1

Jackson River

Cowpasture River

3[i
e o

43 44 46 47 48 49 50 51 52 r40 120 100 80 Lowest annualflow for 1 day (cfs) Virginia, 1941-1952 River at FallingSprings, Jackson 60
Mean = Standard deviation =

F r
; h

61 92 65 72 82 67 74 t 18 t24 r08 65 8884.7 21.7

58 81 70 63 68 58 74 105 134 108 93 85 83.1 23.2

B U

line: Iz : 4.94 + 0.923X; r : 0.86. Figure 26.8 Cross-correlationof low floWs.Regression

EQUATIONS 693 REGRESSION 26.9 FITTING The best estimates of o and B are sought. Thus to minimize

) 0, - f), : ) ly, (o + px,)1,

(26.ss)

valuesfrom Eq. 26.54,take values and!,ate the estimated wherey, are the observed follows: partial derivativesas

- (q+ n')r} *{> b, - (o* B';l'} tv, #{>

- n d- B ) " , : o

(26.s6) (26.s7)

resultin a and two equations After carrying out the differentiationsand summations, B, callednormal equations.

)y,

(26.s8) (26.se)

2*,y,-")xt-F2*?:o
yields SolvingEqs. 26,58 and 26.59 simultaneously e:-'i:y-BT 2v, F2^ n n

(26.60) (26.6r)

P: =7 - (>;y[
Recall the slopeis p(arf o), or as estimatedfrom sampledata B: rt;

(26.62)

equationis variancein the regression Also, the unexplained

4:4Q-p')

(26.63)

and is the squareroot of which is the standarddeviationof residuals(seeFig' 26'8) cailed the standard error of estimate.Thesecan be estimatedfrom

*-: 4 n s -l ( r-r,) z
s2": 2(y,-il'

(26.64) (26.6s)

wherey, and i, are as definedpreviously(seeEq' 26'55)' ivtanynyarotogicvariablesare linearly related,and after estimatingthe regresrangeof sion coeffici"ntr, p."di"tion of y can be madefor any value of x within the be should but performed often is x values.Extrapolationoutsidethe range observed

694

CHAPTER26

PROBABILIry AND STATISTICS

done wfth caution.Equation 26.48 showsthat the variancein the estimateof y for a givenx valuebecomes largewhenx is severalstandarddeviationsaboveor belowthe mean. EXAMPLE 26.6 Rivers are and Cowpasture The lowestannualflows for a l2-yr period on the Jackson tabulatedin Fig. 26.8. The stationsare upstramof the confluenceof the two rivers equationand the correlationbetween that form the JamesRiver. Find the regression low flows. Solution

: are2 x : 1016;) y : 997;2 x2 : 9 1 , 2 1 6 ; 2 The basicstatistics l " 88,777;and2 xy : 89,209. For the two-variableregressiona and B are found from Eqs.26;60and 26.61,. _ [(89,20e) (l0t6xee7)/(12)]: oQo? - (tor6)' 102) (9r,216)

o:

ee7 (0.e23x1016) : 4'9r i

is y : 4.91 + 0.923x. The regression 3. The correlation coefficientfrom Eq.26.62 is

(0.e23)(2r.7) :
23.2

0:86

s,, is 11.7,which is plotted error of estimate, 4. From F;q.26.64the standard rl Fig. 26.8. line in regression as limits aroundthe

for the Regression Coefficientof Determination

BeA regression equationreplaces(and extends)the data used in its development. some loss of causeit cannot reproduceall the basedata, the processresults in the information. This not only includeslossof information aboutparticular pairs of data, of the but alsoaboutthe variability of the data.The variancesf is a statisticalmeasure the spread valuesof y. The greaterthe valueof sl, the wider variability of the measured of information aboutthe variancein y that of points aroundthe mean.The percentage equationis called the cofficient of deteris retained,or explainedby, the regrdssion (differences between its value,the residualsor departures mination, Cr. To determine the y values)haveknown variance4, which represents unacactual and estimated equation.The explainedvariance would be the countedvariance in the regression o2, and the percentageretained (coefficient of determination)is difference, 4

(26.66)

EQUATIONS 26.9 FITTINGREGRESSION

695

Comparisonwith Eq. 26.49 revealsthat

Co= Pz

(26.67)

of d, explained ofthe correlationcoefficientp is the percentage Thus the square r2 is of determination coefficient the of data For any sample regression. by the the to equation regression good the fit of a r2 indicates . A large sl,rlslsl as estimated the percentage of large a to explain is able for or accounts the equation databecause the data. in variation
EXAMPLE 26.7

in Example26.6. Determinethe coefficientof determinationfor the regression r2, is 0.7396. Solution, From Eq. 26.67,the coefficientof determination, "accounts for" about 74 explainsor equationadequately Thus, the regression pefcent of the original information abouty containedin the raw data. Twentysix percentof the information is lost. I r For examto multiple linear r'bgressions. can be extended The bivariateexample variable andx1 andx2 ple, the linear model in three variables,with y the dependent variables,has the form the independent y:a*F$t*Fzxz The normal equationsare ) y : an-r Br) t' * FrZ *, (26.68)

(26.6e) (26.70) (26.7r)

)y"':

")

xtt FrZ*?+ Fr2*,*,

2 yr r : * ) xzI Fr 2 *r *, + F"2 *7
The squareof the standarderror of estimateis

s7:

2(y,-r,)'

(26.72)

where y, are the observedvaluesand y-,are predictedby Eq. (26.68). The multiple correlation coefficient is

n : ( t in Hydrology LinearTransformations

+)
^2\ | /2

s;/

(26.73)

Strongnonlinearbivariateand multivariatecorrelationsare also commonin hydrology, ind various mathematical models have been used to describethe relations. Piiabolic, exponential, hyperbolic, power, and other forms have provided better

696

CHAPTER26

AND STATISTICS PROBABILITY

graphicalfits than straight lines. Becauseof difficulties in the derivation of normal equationsusing least squaresfor thesemodels,many can be transformedto linear forms. The most familiar transformationis a linearizationof mtiltiplicative nonlinear relationsby using logarithms.For example,the equation
Y : g,xft1$z

(26.74) (26.7s)

becomes linear when logarithms are taken, or

log y : log * + Bllo$ x1 -f B"Iog x2

The log transformationprocedureresultsin a linear form when the logarithms in Eqs. 26.60 and 26.61.For are substituted of one or both setsof measurements by the data, logarithms example,if a bivariateparabolicform I : qXb is suggested allow use of the linear form log Y : log a -t b log X. The normal equationscan be usedby redefiningy : log Y,x : logX, e : log a, andF : b,thereby transforming cannow be performedon the logarithms, the equationto y : a * Bx. The regression and the estimateof a is found by taking the antilog of valuesof a and B determined, a. This transformationis possiblefor severalother nonlinearmodels,someof which are shown in Table 26.4. The variablesx and y must be nonnegative,with values preferably greaterthan 1.0 to avoid problemswith the log transformation.
OF NONLINEARFORMS TABLE 26,4 LINEARTRANSFORMATIONS

Equation Y=A+BX
Y = BeAx Y:AXB Y:ABx

Abscissa
X log X X

Ordinate
I

Eouationin linearform

log Y log Y log Y

lY): A + B[x] e)[x] tbc rl : locB + A(1og X] tbc rl : bc A + B[toc B)[X] A + (loC lloCrl : loC

Note.'Variables in brackets are the regressionvariates.

EXAMPLE 26.8 the regionalcorrelationis by Beard,20 In the following exhibit (Table 26.5) prepared deviationof flow logarithmswith the logarithmsof the drainage soughtof the standard areasize'andthe numberof rainy daysper year;X, is setequalto ( 1 t log s) to avoid equationand the multiple correlationcoefflcient. values.Find the regression negative Solution are 1. From Eqs.26.69,26.70, and26.7l, the parameters -0.49 a : 1.34; Fr : -0.013; Fz: equationis and the regression Xt : 1.34- 0.0I3X2- 0.49X3 log s : 0.34 - 0.013log(DA) - 0.49 log(days) or 2. The multiple correlation coefficient from Eqs' 26.72 and 26.73 is R : 0.56. ll

APPLICATIONS AND CORRELATION 26.10 REGRESSION DATA FOR 50 GAUGINGSTATIONS TABLE 26.5 LOGARITHMIC Xr:1 + logs Station number (1)
I

697

Xz: logDA

Per Year & : log numberof rainY-daYs

X2 (2)

x3 (3) 2.11 2.12 2.ll 2.04 2.07 2.04 2.09 1.99

x1 (4) 0.29 0.18 o.l7 0.44 0.38 0.3'7 0.30 0.35 0.16 0.11 0.32 0.34 0.25 0.43 0.2'l 0.25 0.52 0.18 0.39 0.40 0.25 0.23 0.54 0.51 0.45 0.63 0.45 0.59 0.46 0.32 0.96 0.12

Station numDer (5) 33 34 35 36 37 38 39 40 4I 42 43 44 46 4'7 48 49 50

x2 (6)
1.94
z.tJ

xs
(7\ 1.87 t.36 1.81 1.58 1.48 1.89

x1 (8) 0.20 0.58 0.64 0.37 o.27 0.54 0.63 0.78 0.46 0.44 0.24 -0.03 0.30 0.17 0.14 0.10 0.27 0.18 17.89 0.358 51.1527 52.7934 -r.640' 7 33.2598 34.4347 -r.r749 8.1635 6.4010 L7625

2 3 4 5
o

7 8
q

1.61 2.89 4.38 3.20 3.92 1.61 3.2r 3.65

3.23 z.rs

t0 11 t2
I J

t4 15 16 17 18 19 20 21 22
LJ

24 25 26 27 28 29 30 3l 32

2.08 4.33 1.60 2.09 2.00 2.82 2.00 2.40 2.09 3.69 2.18 2.19 2.09 2.17 |.91 4.48 1.9s 4.95 2.21. r.97 2.08 3.4r 4.82 l 88 r.93 r.78 L.74 4.39 3.23 2.01 3.58 2.04 1.64 1.78 1.7 6 4.58 3.26 1.93 1.81 4.29 1.23 1.89 1.48 3.44 |.97 2.lt

3.63 1.91 2.26 2.97 0.70 0.30 3.38 2.87 2.42 4.53 3.04 4.13 |.49 5.37 1.36 2.31 r4'7.55 2.951 503.7779 435.4200 68.3579

r.32 1.54
1.62 2.03 2.26 1.93 1.78 2.00 2.Or 1.95 2.tl 2.23 96.24 1.925 285.5627 284.0042 1.5585 r8'7.59r2 185.2428 2.3484

>X x
2 XX, 2 X2 X2/n 2 XX, 2 X2 X,ln

1.5585

2X2X'ln 2 xx1

-1.6407

Note: x = X - X. (AfterBeard.2o)

2 6 ' I o R E GR E S S |o N A N D C o RRELAT|oNAPPL|CAT|oNS

698

26 PROBABILIry AND STATISTICS CHAPTER desiredJtatisticalparameteras dependent variable,and the appropriatephysicaland The proceclimatic variableswithin the basinor regionas the independenlvariables. and dures are signiflcantly better than using relatively short historical sequences point-frequency analysis. Not only doesthe methodreducethe inherently large sampling errors but it furnishesa meansto estimateparametersat ungauged locations. There are limitations to the techniques of Section 26.9. First, the analyst assumesthe form of the model that can expressonly linear, or logarithmically linear, analdependence. Second,the independent variablesto be includedin the regression ysis are selected. variablesare And, third, the theory assumes that the independent indeedindependent or determined without error. Advancedstatistiand are observed cal methodsthat are beyondthe scopeof this text offer meansto overcomesomeof to satisfy them. Therefore,care theselimitations but in practiceit may be impossible must be exercised in selectingthe model and in interpretingresults. Accidental or casualcorrelationmay existbetweenvariablesthat are not funcbetweenhydrotionally correlated.For this reason,correlationshouldbe determined Because of the natural logic variablesonly when a physicalrelation can be presumed. dependence between many factors treated as independentvariables in hydrologic variableand eachof the independent studies,the correlationbetweenthe dependent variableswhen variablesis different from the relative effect of the sameindependent togetherin a multivariatemodel. One way to guard againstthis effect is by analyzed screeningthe variablesinitially by graphical methods.Another is to examine the results of the final regression equationto determinephysicalrelevance. significantvariaid in screening Alternatively,regression techniques themselves a procedurecanbe followedin which ables. When electroniccomputationis available, model, and the successive independent variableS are addedto the multiple regression in the multiple correlationcoefficient. relative effect of eachis judged by the increase it is useful otherwise Although statisticaltestscan be employedto judge significance, to specify that any variable remain in the regressionequation if it contributesor explains,say,1 or 5 percentofthe total variance,or ofR2. A frequentlyusedrqethod is to computethe partial correlation cofficients for each variable.This statistic represents the relative decrease in the varianceremaining( 1 - R') by the addition of the variablein question.If the varianceremainingwith the variableincludedin the regression is (I - Rz) : pz and the variance remaining after removal is (l - R'') : D'', then the partial regression correlation coefficient is Most PC spreadsheet software packageshave statistical routines for all the described here and many more. Most are extremelyflexible,requiringminianalyses mal instructions-and input data other than raw data. Specialmanipulationscan effect variables,bring one variableat a time an interchange ol dependent and independent into the regression equation, rearrange the independent variables in order of significance, and perform various statisticaltests.

\D'' - D')lD''.

Hydrologic Records Extending

frequently can be used to extend short records if significant Regression techniques correlation existsbetweenthe station of short record and a nearby station with a - lorigerreeord.Iq Example26.6,if the Jackson River recordswerecompletefromI94l

PROBLEMS 699 recordswere incompleteafter 1952,the cross-correlation to datebut the Cowpasture equationfor I could be usedto estimatethe missingyearsby solvingthe regression The reliability of suchmethodsdepends from 1953 on usingthe X flows as observed. on the correlationcoefficientand the length ofthe concurrentrecords.Ifthe concurrent record is too short or the correlation weak, the standarderror of the parameter to be estimatedcan be increasedand nothing is gained.The limiting value of crosscorrelation for estimatingmeansis approximatelyp : l/\/ n, where n is the length Thus any correlation above0.3 would improve the Cowof the concurrentrecord.21 pasturerecords.Estirnatesof other parameterswith larger standarderrors require for significantimprovement.Extendingor filling in deficient highercross-correlation for regional studiesin which every record usedshould be recordsoften is necessary adjustedto the samelength.

Hydrologic Variables Regionalized Predicting.

studiedvariousmethodsof regional flood analysisand found the Cruff and Rantzz2 techniquea better predictor than either the index-floodmethod multiple regression (Chapter27) or the fitting of theoreticalfrequencydistributionsto individual historitechniques to extendall recordsto a common cal records.They flrst usedregression base length. Next they extrapolatedby various methods to estimate the 50- and 100-yearflood events and with multiple correlation examined several dependent variables including the drainage areaA, the basin-shapefactor (the ratio of the parallel to the main diameterof a circle of sizeA to the length of the basin measured channel)Sa,channelslopeS, the annualprecipitationP, and others.They found only A and S, to be significant, which resulted in prediction equations of the form The Q, : cAS!,. These equationswere superior to those of the other techniques. multiple correlation coefficientwas as high as 0.954.It is interestingthat regression regionalvaluesof the techniques wereemployedin still a third way,that is, to estimate mean and standarddeviation after adjustingthe record length. Example26.8 illusanalysisto regionalizethe standarddeviationof trated the applicationof regression annual maximum flow logarithms as a function of the drainage area size and the number of rainy dayseachyear.

'

r summary
Statisticsis a diversesubject,and the treatmentin this chapterhasbeennothing more and practitionersmust return againand again than an introduction. Seriousstudents They will find that evaluatingnew developments to the theory in standardworks.23 of statismust claim a large shareof their time. Only certain aspects, and techniques principally the common distributionsand the tical hydrologyhave been presented, methodsfor analyzingfrequency of eventsobservedat a single point. In the next to common applicationsin hydrology. chapterthis information is extended

PROBLEMS
26.1. The probabilitiesof eventsE1 andE2 arc each.3. What is the probability that E1 or but not mutually exclusive,and E2 will occur when (a) the eventsare independent given (b) probability E2is .l? of Et, when the -

700

CHAPTER26

ANDSTATISTICS PROBABILITY

26.2. EventsA and B are independenteventshaving marginal probabilities of.4 and .5,
respectively.Determine for a single trial (a) the probability that both A and B will and (b) the probability that neither occurs. occur simultaneously, 26.3. The conditional probability, P(E, I E,r),of a power failure (given that a flood occurs) is .9, and the conditionalprobability,P(Ez I E), of a flood (given that a powerfailure occurs)is .2. If the joint probability, P (\ andE), of a power failure and a flood is .1, determinethe marginal probabilities,P(E) and P(E). (c) both 26.4. Describetwo random eventsthat are (a) mutually exclusive,(b) dependent, nor dependent. and (d)"neithermutually exclusive mutually exclusiveand dependent, cofferdamis to be built to protect the 5-yearconstructionactivity for a 26.5. A temporar;1 to withstand the 20-yearflood, major crossvalley dam. If the cofferdam is designed what is the probability that the structurewill be overtopped(a) in the flrst year, (b) in the third year exactly,(c) at leastonce in the 5-yearconstrucfionperiod, and (d) not at all during the 5-yearperiod? The to a frequencyanalysis. 26.6. A 33-yearrecord of peak annualflow rateswas subjected median value is defined as the midvalue in the table of rank-ordered magnitudes. Estimatethe following probabilities. the medianvaluein any , a. The probability that the annualpeak will equalor exceed singleyear. retlrrn period of the median value. b. The average c. The probability that the annual peak in 1993 will equal or exceedthe median value. d. The probability that the peak flow rate next year will be less than the median value. yearswill e. The probability that the peak flow rate in all of the next 10 successive value. be lessthan the median f. The probability that the peak flow rate will equal or exceedthe median value at years. least oncein l0 successive g. The probability that the peak flow rates in both of two consecutiveyears will equal or exceedthe median value. h. The probability that, for a2-yearperiod, the peak flow rate will equal or exceed the median value in the secondyear but not in the first' 26.7. What return period must an engineeruse in his or her designof a bridge openingif there is to be only a 50 percent risk that flooding will occur at least once in two years?Repeatfor a risk of 100 percent. successive 26.8. A temporary flood wall has been constructedto protect several homes in the floodplain. The wall was built to withstand any dischargeup to the 20-year flood magnitude.The,wall will be removed at the end of the 3-year period after all the homeshavebeen relocated.Determinethe probabilities of the following events: a. The wall will be overtoppedin any year. b. The wall will not be overtoppedduring the relocation operation. c. The wall will be overtoppedat leastonce before all the homesare relocated. d. The wall will be overtoppedexactly once before all the homesare relocated. for the flrst 2 years and then overtoppedin the third e. The wall will be adequate year. return periods(shownon the next page)are known 26.9. Waveheightsand their respective campsitewill accepta 25 percent Owners of a downstream for a 40-mi long reservoir. risk that a proiective wall will be overtoppedby wavesat least once in a 2}-yeat period. Determinethe minimum height of the protective wall.

PROBLEMS 701 Waveheight (ft) 10.0 8.5 7.4 5.0 3.5 period Return (years) 100 50 30 10 5

26.10. Assumethat the channel capacityof 12,000cfs near a private home was equaledor exceeded in 3 of the past 60 years.Find the following: a. The frequencyof the 12,000-cfsvalue. b. The probability that the home will be floodednext year. c. The return period of the 12,000-cfsvalue. d. The probability that the home will not be floodednext year. e. The probability of two consecutive, safeyears. f. The probability of a flood at leastonce in the next 20 years. g. The probability of a flood in the second, years. but not the first, of two consecutive h. The 20-yearflood risk. 26,11, The distribution of mean annual rainfall at 35 stations in the JamesRiver Basin, Virginia, is given in the following summary:

Interval(2-in. groupings) Numberof observations Interval (2-in. groupings) Numberofobservations

36 or 37 in. Z 44 or 45 in. 5

38 or 39 in. 4 46 or 47 in. 4

40 or 41 in. j 48 or 49 in. 2

42 or 43 in.

50 or 51 in. 2

(seeChapter27) andplot the frequencydistribution Computethe relative frequencies andthe cumulativedistribution.Estimatethe probability that the meanannualrainfall (a) will exceed40 in., (b) will exceed50 in., and (c) will be betweenthesevalues. 26.12. Write a simpleprogram to READ in N data points and compute the mean, standard deviation,and skewness coefficient. 26.13. A normally distributedrandom variablehas a mean of 4.0 and a standarddeviation of 2.0. Determinethe value of

"8

dx I f(x)

f@

26.14. For a standardnormal densitv the value of ' function. use Table B.1 to determine
fr+'o

J*-ro 26.15. A normal variableX has a meanof 5.0 and a standarddeviationof 1.0.Determinethe value of X that has a cumulativeprobability of 0.330. 26.16. If the mode of a PDF is considerablylarger than the median, would the skew most likely be positive or negative?

fG) dx

702

ANDSTATISTICS 26 PROBABILIW CHAPTER about the properties of a PDF by 26.17. tomplete the following mathematicalstatements insertingin the boxeson the left the correct item numberfrom the right. Assumethat X is a seriesof annual occuffencesfrom a normal distributibn. I r.Zerc a. I f(x) dx: J t " b.

:r dx f'_ro,
dx: '34 l-.o 'o'

2.Unity eachyear 3. Valuewith 5 percentchanceofexceedance 4. 0.68 5. Valueexpectedevery 50 r"urc on the average 6. P(X < mr) 'X'*r) 7.P(m1
8 . P ( m 1- X = m z )

d.

f,o"

dx:r

dx: .5 f-to,
f(x) dx : Z

I rt, dx: .02

l.

9. Median

10. Standarddeviation 26.18, The mean monthly temperaturefor Septemberat a weather station is found to be normally distributed.The mean is 65.5" F, the varianceis 39.3'F2, and the record is completefor 63 years.With the aid of TableB.1, find (a) the midrangewithin which two thirds of all future mean monthly valuesare expectedto fall, (b) the midrange within which 95 percentof all future valuesare expected,(c) the limit below which to be and (d) the valuesthat are expected 80 percentof all future valuesare expected, exceeded with a frequencyof oncein l0 yearsand oncein 100years.Verify the results by plotting the cumulativedistribution on normal probability paper. to be approximately basinis determined 26.L9. The total annualrunoff from a small drainage normal with a mean of 14.0 in. and a varianceof 9.0 in.2.Determinethe probability that the annualrunoff from the basinwill be lessthan I 1.0 in. in all three of the next years. three consecutive of 30,000cfs at a streamgaugingstationwasequaled 26.20. In the past60 years,a discharge return period (years) of this or exceeded only three times. Determine the average value. and havemarginal probabilitiesof .4 and .5, respec26.21. EventsA and B are independent tively. Determine the following for a single trial: a. The probability that both A and B occur. b. The probability that neither occurs. c. The probability that B, but notA, occurs. Existingrecordsrevealthe following information aboutEventsA and 4 whereA = a ?6il,

IongMerch'warmspellandB 1q

:!94!$99!.

PROBLEMS Year A : warm March? B : April flood? No Yes No No Yes No No Yes Yes Yes
No Yes

703
1n

Yes No

No Yes

Yes Yes

No No

On the basisof the 10-yearrecord, answerthe following: Prove. a. Are variablesA and B independent? Prove. b. Are variablesA andB mutually exolusive? c. Determinethe marginal probability of an April flood. d. Determinethe probability of having a cold March next year. e. Determinethe probability (onevalue)of havingboth a cold March and a floodfree April next year. f. If a long March warm spell hasjust endedtoday,what is the best estimateof the probability of a flood in April? 26.23. Two dependent eventsarc A : a flood will occur in Omaha next year and B : an ice-jam will form near Omahain the Missouri River next year. Useyour judgment to rank from largestto smallestthe following probabilities:P(A), P(A andB), P (A or B), P(A I B). 26.24. The probability of having a specifiedreturn period, [, is definedas: I or exceeded P(annual valuewill be equaled : /, \'-' exactly oncein a periodof r : I years) T,f \' Also, p (annualvalue will be equaledor exceeded _ exactly r times in a period of n years) pn_r(l _ p)r

the secondprobability shouldequal in parentheses, a. According to the descriptions the first when n and r are equal to what values? resultin the sameprobability for an annualvaluewhose b. Showthat both equations years.Discusss. frequencyis 33{ percentand the return period is Z.: /:3 below, find (a) the number b that will make the function 26.25. For the function described of x a probability densityfunction, and (b) the probability that a singlemeasurement will be lessthanl.
-

.l \^)

{i",,'

forx ( 0 for0<x=b forxlb

depth of precipitation in July. Betweenvaluesof 26.26. The random variabler represents -r : 0 and x : 30, the probability densityfunction has the equation/(-r) : x/40p',. In the past, the average July precipitation p,, was 30 in. a. Determine the probability that next July's precipitation will not exceed20 in. b. Determinethe singleprobability that the July precipitatjon will equal or exceed years. 30 in. in all of five consecutive depth of precipitation in July. Betweenvaluesof 26.27. The random variable-r represents x : 0 and .r : 30, the probability densityfunction has the equation/(.x) : x/1200. a. Determine the probability that next July's precipitation will not exceed20 in. 30 in. b. Determinethe probability that next July's precipitationwill equalor exceed

l---

704

ANDSTATISTICS 26 PROBABILITY CHAPTER 26.28.

26.29.

26.30. 26.31.. 26.32.

26.33. 26.34.

discharge Measured (1000 acre-ft) 12.2 10.4 10.6 1.2.6 14.2 13.0 14.0 t2.0 10.4 tl.4

recharge Estimated (1000acre-ft) t2.o 9.8 I 1.0

t3.z

14.6 14.0 14.0

I z,+

10.4 11.6

26.35. F i t a r e g r e s s i o n e q u a t i o n t o t h e d a t a i n P r o b l e m 2 6 ' 3 4 , t r e a t i l g d i s c h a r g e a s t h e
dependentvariable.computethestandarderrorofestimate.Estimatetheexpeoted if no of discharge when recharg"ir 13 Kti what would be the estimate discharge provided improvement relative the is what information were availatie on recharge? estimate? by the regression The program linear regression' 26.36. prepare a computerprogramfor simple,two-variable, variances' means' the (b) compute X' should(a) read in N pain oioUt"tuuiions, Y and constants'the regression the (c) find and andX, Y and standarddeviationsoiUott' coefficient. verify with the data in standarderror of estimal-, u"J,ir" correlation Problem26.34. of the mean annual rainfall with the 26.37. From the following observations of variation How iinea, predi.tion equationfor the catchment. altitude of the gauge,d";;;;; well correlatedare rainfall and altitude?

PROBLEMS Gauge number Mean annual rainfall(in.) Altitudeof gauge (1000ft)

A A

705

1 2 3
4

22 28 25 31
JZ J I

4.4
1 < < A

5 6 7 8 9 10 1l 12

36 35 36 46 4l 4I

5.6 5.6 5.8 6.0 6.6 6.6 6.8 7.0

26.38. Estimatethe expectedrainfall in Problem 26.31 for a gaugeto be installed at an

altitude of 5500 ft. equationY : A + estimates of A and B in the bivariateregression 26.39. The least-squares : aslogro)andXis defined Yis atransformation 3.0,where BXarcA: 2.0 andB : of a and b. the values determine y axb, by related Ify and r are as 1o916;r. defined rise from to for a stream (Z), deflnedasthe time 26.40. The time of rise of flood hydrographs (L) and length stream to the related is a storm, low waterto maximum depthfollowing Texas, in 11 watersheds given below for (S). information the From slope the average New Mexico, and Oklahoma, derive a functional relation of the form T,: aLbS'.

Watershed number
I

Ir

(min)

(1000 ft) 18.5 14.2 25.3 tt.7 9.7 8.1 21.'7 3.9 1.2
J.J J.)

(fv10oo ft)
7.93 19.0 t2.a 13.3 11.0 15.0 16.7 146.0 20.0 64.0 33.0

2 3
A

5 6 7 8 9 l0 11

150 90 60 60 100 75 90 30 30 45 50

26.4r, Repeat the exercise in Problem 26'40 by fitting the relation T,: dF", whete
F : L/{S

with t in mi and S in ftimi. Plot the results on log-log paper. 26.42. The squareof the linear correlationcoefficientis calledthe proportion of the variance Describethe meaningof this phraseby evaluatthat is "explainedby the regression." and what doesthe term givenin the text.What varianceis explained, ing the equations "explained" mean? pairs of valuesof normally distributedvariablesX and Y are ana26.43. Twenty measured lyzed, yielding valuesof X : 3O,V : 20, s' = 20, ands, : 0. Determinethe values

706

CHAPTER26

PROBABILITY ANDSTATISTICS

d a andb and the standarddeviationof residuals for a least-squares fit usingthe linear equationY:a+bX. 26.44. The least-squares estimates of A and B in the bivariateregres*sion equationy : A + BXareA:2.OandB : I.0, whereyis atransformation defined aslog1eLIf Iand X are relatedby Y : a(.b)',determine the valuesof a and b.

26.45. Given a table of ten valuesof mean annual floods and corresponding drainageareas
for a numberofdrainagebasins,statehow linear regression techniques would be used to determinethe coefficientand exponent(p and 4) in the equation Qzzz : pAq.

26.46. What choiceof transformedvariablesI and X would provide a linear transformation

for y : a/(x3 + b)? Also, if a regressionon these transformed variables yields I : 100 + 10X determinethe corresponding valuesof a and b. Would the linear transformationbe applicableto all possiblepairs and valuesofx and y? 26.47. Which measure of variation in a regression Y : a * bX is generallylargerin magnitude, the standarddeviation of I or the standarddeviation of residuals? For what condition would the two valuesbe equal?

REFERENCES
1. Ven T. Chow, "Statisticaland ProbabilityAnalysisof HydrologicData," Sec. 8-1, in Handbookof Applied Hydrology (V. T. Chow, ed.). New York: McGraw-Hill, 1964. 2. M. B. Fiering, "Information Analysis," in Water Supply and Waste Water Removal (G. M. Fair, J. C. Geyer,and D. A. Okun, eds.).New York: Wiley, 1966,Chap.4. 3. J. R. Benjamin and C. Cornell, Probability, Statisticsand Decisionfor Civil Engineers. New York: McGraw-Hill. 1969. 4. A. M. Mood and F. A. Graybill, Introduction to the Theory of Statistics, 2nd ed. New York: McGraw-Hill. 1963. 5. A. J. Duncan,Quality Control and Statistics.Homewood, IL: RichardD. Irwin, Inc., 1959. 6. P. G. HoeI, Introduction to MathematicalStatistics,3rd ed. New York: Wiley, 1962. 7. J. Aitchison and J. A. C. Brown, The Log-Normal Distributlon. New York: Cambridge UniversityPress,1957. 8. H. A. Foster,"Theoretical Frequency Curves,"Trans.ASCE 87,142-203(1924). 9. L. R. Beatd, Statistical Methods in Hydrology, Civil Works Investigations, U.S. Army Corps of Engineers,Sacramento District, 1962. 10. "A Uniform Techniquefor DeterminingFlood Flow Frequencies," Bull. No. 178, U.S. Geological Survey,1989. 11. "New Tablesof Percentage Points of the PearsonType III Distribution," Tech. Release No. 38, CentralTechnical of Agriculture,1968. Unit, U.S. Department 12. M. B. Fiering, StreamflowSynthesis. Cambridge,MA: Harvard University Press,1967. 13. F. E. Perkins,SimulationLecture Notes,SummerInstitute, "Applied MathematicalProgramming in WaterResources," University of Nebraska,1970. 14. E. J. Gumbel, "The Return Period of Flood Flows," Ann. Math. Statist. l2(2), L63190(June 1941). 15. Ven T. Chow, "The Log-Probability and Its EngineeringApplication," Proc. ASCE 80, 1-25(Nov. 1954). 16. "Probability Tables and Other Analysis of Extreme Value Data," Series22, National Bureauof Standards Applied Mathematics, 1953. 17. E. J. Gumbel, Statisticsof Extremes.New York: ColumbiaUniversity Press,1958.

Chapter27

Frequency Analysis

r Prologue
The purposeof this chapteris to: . Presentmethodsused in hydrology to evaluatethe recurrenceof particular magnitudes and durationsof random hydrologicvariables. . Elaborateon the definitionsof freqUency, reiurrence interval, return period, and risk analysis introduced Chapter 10, Section10.4. in . Illustrate the diverseapplicationsof frequencyanalysisin hydrology. . Teachseveral methodsfor conductingfrequencyanalyses, includingthe useof recurrence intervals for variables that allow estimation of frequencyfactors that follow conventionalprobability distribution functions. . Introduce methods of point and regional frequency analysis and describe regional USGSregression equationsthat have been 4doptedthroughoutthe U.S. for estimating flood flows for use in structure design and floodplain analysis. . Establishhow to estimatethe reliability of estimatesderived from point or regional frequencyanalyses. Explainthe widelyusedBulletin No. 17BLog-Pearson Type III procedures for performing uniform flood flow frequencyanalyses. Describehow variousfederal agencies apply frequencymethodsin designor analysisof water resources systems. In Chapter 26, probability and statistical characteristics of random variables were introduced,alongwith common distributionfunctions and principlesof regression and correlatiqn.The presentchapterprovidesapplicationsof theseprinciplesto common hydrologicvariables.

27.1 FREQUENCY ANALYSIS

presented in Chapter26 areusedmostfrequentlyin describing The statisticalmethods hydrologicdata suchas rainfall depthsand intensities,peak annual discharge, flood flows, low-flow durations,and the like. Frequencyanalysiswas introduced in Sec-

ANALYSIS 709 FREQUENCY 27.2 GRAPHICAL tion 10.4 and is defined as the inve or probabilitiesof magnit recurrencA wise, the frequencYof a hYdrologic discretevariablewill occur or some' in anY given Year'The lat exceeded freq-uenc probability or exceedance ir"qo"n"Y is a ProbabilitYand has as shownbYEc frequencY, ceedancs Two methods of frequencYa plotting techniqueto obtain the cum iu"to.t. The cumulativedistribution the probability of an eventequal to is used to obtain recurrencerntervz tioned whenworking with recordssl ofexpectedhydrologiceventsgreaterthantwicetherecordlength.

ANALYSIS FREQUENCY 27.2 GRAPHICAL

The frequencYof an event can be When annualmaximum valuesare imated as the meantime in Years'\ The t onceon the average' exceeded be shownto be x: where 7 = N = n : m: tuf,lj
m

(21.r)

the mean number of exceedances the number of future trials -oi of values the number values,with largestequal to I descending therunt t : 1' N : T' and If the mean number of exceedances
4 I _ -

n ) l
m

(21.2)

indicatingthattherecurfenceintervalisequaltothenumberofyearsofrecordplus 1, divided bY the rank of the event' They give different results as Severalpictting p*liion formulas are available'l for 10 yearsof record .The rangein recurrencerntervalsoutained .1, notedin Table2'7' plotting position formulasdo not account is illustratedin the right-hani column.Most formula ihat doesaccountfor samplesize for the samplesizeor length of record. One generalform wa, giuen by Gringorten' and has the
.r1 -

r -

n * l - 2 a
m - a

(27.3)

710

CHAPTER2TFREQUENCYANALYSIS
TABLE 27.1 PLOTTINGPOSITIONFORMULAS

Form:1 andn=10
Method California Hazen Beard Weibull Chegadayev Blom Tukey Solve for P (X > x\ m n 2m-l P

.10
.05 .067 .091 .06'l .061 .065

l0 20 14.9 11 14.9 t6.4 15.5

1 - (0.5)'/'
fn n * l

m-0.3 n+0.4

,m-i n+i
3m-l 3n*l

where n : the number of yearsof record the rank a parameterdependingon n as follows:

n a n a

10 0.448 60 0.440

20 0.443 70 0.440

30 0.442 80 0.440

40 0.441 90 0.439

50 0.440 100 0.439

in the Gringortenequation.If the distribuIn general,a : 0.4 is recommended tion is approximatelynormal, , : fi is used.A value of a : 0.44 is usedif the data follows a Gumbel distribution. order or decreasing in all cases is to arrangethe datain increasing The technique of magnitudeand to assignorder number m to the ranked values.The most efficient formula for computingplotting positionsfor unspecifieddistributions,l and the one now commonly usedfor most sampledata, is the Weibull equation P _ m n-fI (27.4)

Whenm is rankedfrom lowestto highest,P is an estimateof the probability of values being equalto or lessthan the rankedvalue,that is, P(X < x); whenthe rank is from in percentages, the value highestto lowest,P is P(X > x). For probabilitiesexpressed is IA\ml@ + 1). The probability that X: .x is zero for any continuousvariable.

ANALYSIS 27.4 REGIONALFREQUENCY

721

regionalvaluesof skewness regiond'lstudies. Methodsof "smoothing" and averaging havealsobeenproposed.ro'15 did Many techniques used in the past for generalizingregi6nal characteristics methodof extending The so-calledstation-year not rely on statisticalconsiderations. statisticalvalidity, especially rainfall recordshasprovedhelpful but has questionable areas.The method seriesor to stationsin nonhomogeneous if applied to dependent has been used to combine, say,two 25-yearrecords to obtain a single 50-year sequence. In practice, the analyst may have to use imagination and ingenuity to while remaining awareof actual and theoretical summarizeregional characteristics, considerations, lndex Flood Method The index-floodmethodusedin the pastby the U.S. GeologicalSurveyis an example The method usesstatistical successfully.t''tu of summarizingregional characteristics andgenerally It canbe supplemented them in graphicalsummaries. databut combines techimproved by using statisticalmethods,employing,for example,the regression niquesexplainedin Chapter26.The index method,as illustrated in Fig. 27.4, canbe outlined as follows. 1. Preparesingle-stationflood-frequencycurves for each station within the homogeneous region (Fig. 27.4a). taken from the curves at various 2. Compute the ratio of flood discharges frequencies to the mean annual flood for the samestation. 3. Compileratios for all stationsand find the medianratio for eachfrequency (Fi5.27.4b). 4. Plot the median ratios againstrecurrenceinterval to produce a regional curve (Fig. 27.4c). frequency involved are (1) a homogeneitytest to justify Two statistical considerations shortrecordsto placeall stations definitionof a region, and(2) a methodfor extending by Potterfor on the samebaseperiod. A somewhatsimilar techniquewas developed It relies on the graphical correlation of floods with the Bureau of Public Roads.rT physical and climatic variables and is thus a techniquethat refers in part to the in Chapter26 (seealso Chapter16). discussion

Regression Equations Regional U.S.G.S.

survey institutedaprocessof correlatingflood Early in the 1950s, the U.S. Geological Setsof regreswith drainagebasin characteristics. flow magnitudqs and frequencies floodshavebeendeveloped for the 2-,5-, l0-,25-,50-, and 100-year sionequations regionin every state.The work was for practically everyhydrologicallyhomogeneous largely inauguratedto developmethodsfor estimatingpeak flow rates for designof highway structuresat ungaugedbasins.Data from gaugedsites was evaluatedby modelsto the data. regional analysisto provide the best fit of regression gaugedata were consultedto Continuouswater stagerecordersand crest-stage Given the frequencycurves,a developfrequencycurvesfor all gaugedwatersheds.

6000 5000
o

4000
a

o po

3000 2000 1000 0

1.01 1.1

I ts
c") c..l
I

i'
2 5 1 0 2 (Yr) interval Recurrence
(al Recurrenceintervals (Yr)

50

100

station
1

1.1
^ ALt

1.5
n ?<

5
1 AA

10
1 q?

20
).55

50
3.03

(b)

3.0
2.8
I

2.2
6 o

2.0 1.8
l.o
' : I

po

7.4 t.2 1.0 0.8 0.6 0.4

1.01 1.1

t.52

10 (Yr) interval Recurrence

(cl

50

100

Figure27.4 Index-floodmethod of regional flood frequency flood frequencycurve, (b) ratios Q' (a) single-station an-alysis: (c) regional flood frequencycurve' and to Qr.rrfor 15 stations,

ANALYSIS 723 FREQUENCY 27.4 REGIONAL to predict the numberof correlationtestswere madeusingmultiple linear regression parameterssuch as drainage peak flows from various easily obtained independent duringthe snowmelt meantemperature aspect,elevation, area,basinslope,watershed and hundredsof other variables' season, Each study was reportedby state.The open file or water resourceinvestigation of the equations,comreports are availablefrom the USGS and include discussions copies 1n"ntron rangeof applicability,information on the reliability of the equations, of equationsfor estimatingfloods of all the gaugedbasin frequencycurves,and.sets Equationsfor all stateshavebeen compiledby the U.S. in ungaugid watersheds. Ceotogicit Survey into a PC software packagecalled NFF (National Flood FreHYDRAIN)' qu"n"y;, availablefrom tJreusGS (or FHWA as part of their package, was Regionalregression Figure 27.5 showsthe six regionsfor the stateof Texas. gaugedstationdata.As in many of by regionusingavailable independently conducted variableswere sethe reports, the Texasmanual revealsthat different independent fgr Region2 are: The equationsdeveloped lectedfor eachregion.18
Q, Qs Qro : : : 216 Aos74So'12s 184 322 Ao62oso

389 Ao6a6So'2ta 236 = 485 Ao 668510 Qr, : 555 Ao 6825'0'250 Qro : 628 Ao 6e4s0261 Qrco

(27.r2) (27.r3) (27.r4) (27.rs)

' (27.16)

(27.r7)

for given frequency,cfs where g : peak discharge A : drainagearea,squaremiles betweenpoints 10 and 85 percentof the slopeof the streambed S : average distancealons the main streamchannelfrom the mouth to the basin divide, feet per mile

EXAMPLE 27.6
Develop estimates of flood peaks for a 200-square-mile rural watershed near Dallas. The mean slope between the 10 and 85 percent points is 3.4 ft per mile. Solution. Dallas is in Region 2. Equations 2712-27.17 2 1 6 A o s 1 a s o l 2:s 5 , 2 7 0 c f s : cfs Os:322406205018o 10,770 : 1 5 ' 4 9 0c f s Q r o : 3 8 9A o ' 6 a 6 s o ' 2 1 a : 2 2 '300 cfs Q r r : 4 8 5A 0 6 6 8 5 0 2 3 6 : Q r o : 5 5 5 A o 6 8 2 s o 2 5 o 2 7 , 8 0 0c f s : ll Qrco : 628 Ao6e4so261 34,170 cfs Qr: give:

724

CHAPTER2T FREQUENCYANALYSTS

UNDEFINED

Pacoa

,'7I

N D R E

Figure 27.5 Hydrologicregionsin Texasfor 1976USGSregional regression equations.(From Ref. 18.).

ANALYSIS 27.4 REGIONALFREQUENCY

725

726

CHAPTER2TFREQUENCYANALYSIS

(NFD Program NationalFlood Frequency

floodlike Eqs.27.I2 through27:t7 for estimating equations Since1973,regression published, least once, at been have watersheds peak discharges for rural, unregulated in cooper1993 the USGS, In for every stateand the Commonwealthof PuertoRico. Manageation with the FederalHighwayAdministration and the FederalEmergency regression area and metropolitan ment Agency,compiled all of the current statewide equationsinto a microcomputerprogramtitled the National Flood Frequency(NFF) equationsfor estimatingflood-peak regression This program summarizes Program.re for estimatinga typical flood techniques It also addresses for all52 states. discharges probability peak discharge hydrographfor a given recurrenceinterval or exceedence prograr4 lists statewideregression The for unregulatedrural and urban watersheds. information and reference and providesmuch of the equationsfor rural watersheds for estimating equations input data neededto run the computerprogram.Regression least 13 statesare in at for severalmetropolitan areas urban flood-peak discharges also available. and the computerprogramare given.rT Information on computerspecifications Instructionsfor installing NFF on a personalcomputerand a descriptionof the NFF data base of regressionstatistics are also given. The program and the associated program is available as part of the Federal Highway Administration package, HYDRAIN, or by itself. Thoughthe USGSand FHWA do not distributeor servicethe software,information about vendorswho provide softwaresalesand servicecan be obtainedby contactingthe agencies.

from ChannelGeometry Flood Frequency

Stream channelsin alluvial systemsdeveloptheir width, depth, slope, and other that flow through hydrographs from the composite hydraulicgeometrycharacteristics that the shapeof some streamchannels,if their valleys. It has been demonstrated can be correlatedwith the mean annual properly evaluated by trained hydrologists, disflow, peak annual flow, bank-full flow, and the dominant, or channel=forming, similar to Eqs. 27.12-27.I7, havebeen successfully equations, charge.Regression standarderrors of estimate. with very reasonable derivedfor many perennialstreams for these studiesare normally obtained during low flow. The Measurements channelof interestis that channelbeing maintainedby the current flow regime.It is by the activechannel,limited laterally by the point barsand mostrecent characterized the most recentdepositions, geologicfloodplaindeposits. It is felt that theserepresent and are thereforeindicative of the width neededby the current flow and sediment the principle in determining the active, transport regipe. Figure 27.6 rll:ustrates the exampleas the width A-A . floodplain-buildingchannel,established'for Colorado,and in Nevadd,California, Kansas, havebeenconducted Suchstudies A USGSinvestigationof 53 gaugedstreamsin mountainregionsof Coloelsewhere. rado resultedin the following equations.2o
Qz = 0.666 Wr'eoaD*o Q, : 1.53 W1.682D-o2stAo'017
Qr6 : 2.38 W1 t43 2se 53o A.o D-o

(27.r8) (27.r9) (27.20)

ANALYSIS FREQUENCY 27.4 REGIONAL

Referenceline Low-flow water level

727

Distance, in feet

Figure27.6 Ref. 20.)

illustrating active channeldimensions'(From Typical streamcross-section,

Q25 Qro

: :

3.70 Wr'372D-o263Ao2rs 4.93 W127aD-02s6Ao'257

(27.2r) (27.22)

where Q W D A

: = : :

peakflow for the given frequency,cfs iop width of streamat bank-full condition, ft mean depth for bank-full flow, ft cross-sectionareaat bank-full flow, sq ft

The multiple correlation coefficientsfor theseequationsranged from 0-80 for the 42' 1 ranged_from errors,respectively, 50-yrflow to 0.89 for the2-yr event.Standard use in for tool yet another offer investigations of types These percentto 32.2percent. site-specific by floods evaluate to hydrologist the and allow p"uk flo*r, estimating conditionsversusmore uncertain regionalparameters'

RainfallCharacteristics Regional
The variation of rainfall frequencieswith duration was introduced in Chapter 2' can be usedto I analysis Regression thoseshownin ChaPter15, and the Many formulashavebeenusedin t a form with intensity(i) inverselypt of the form i : AIQ + B) to fit featuresof both the rainfall region A andB thereforeserveascharacteristic constants and the frequencyof occurrencein eacharea'

728

CHAPTER27

ANALYSIS FREQUENCY

EXAMPLE 27.7

intensity-duration-frequency Fit the following rainfall datato determinethe 10-year curve.

r : duration (min) i : intensity (in./hr) Ui

r0
5.9 o.l7

7.r o.t4

15 5.1 0.20

30 3.8 0.26

60
z-)

t20
1 A

0.43

0.71,

Solution in linear form as 1.. A model of the form i : AIG * B) can be expressed ,, of l.li versust yields l/i : 0.005t + 0'12, from which The regression A : 2 0 0 a n d B: 2 4 . 3. Thus the rainfall formula is i : 200/(t + 24). The correlation coefficient is -0.997. ll

lli:tlA+BlA.

Maximum averagerainfall depths have been publishedby the U.S. Weather for durationsbetween30 min and24hr and for recurrenceintervalsbetween Bureau22

depthrelation showninFig.27.7.
c
o F o

100

'a

9u

N
\\
3hr
mln

24ft

6hr

t 8 0
F

e 'ro
q

lhr

* 6 0 b - "

50

150

350

400

(mi2) Area use with duration frequency for curves Area-depth Figare 27,7 (U.S. Weather Bureau.) values.

FREQUENCY ANALYSIS 729 27.4 REGIONAL hedvily on the densityand location The accuracyof arearainfall data depends of the accumulation of gauges throughoutthe areaconsidered. The simpleaveraging givesno consideration in all gauges of the effectiveareaaround eachgaugeor to the of gauge in calculatingthe weightedaverage stormpattern.Two methodsare available records, the Thiessen polygon method and the isohyetal method. The Thiessen Perpendicmethodassumes a linear variation of rainfall betweeneachpair of gauges. ular bisectorsof the connectinglines form polygonsaround each gauge(or partial polygonswithin the areaboundary).If a sufficientnumberof gauges are availableto the weightingprocesscan be carried constructcontoursof rainfall depth (isohy.ets), out by using the average depth betweenisohyetsand the area includedbetweenthe Figure 27.8 showsboth schemes. isohyetsand the areaboundaries. An exampleof the effect of gaugelocation and density is shown inFig.27.9. stormrainFigure 27.9ashowsthe increase in variability betweenThiessen-weighted from the watershed falls and rainfall at a singlegaugeas the distanceof singlegauges Figure 27.9b showsthe effect of gaugedensityand total areaon the centerincreases. of precipitationpatternsoverlargeareas standarderror of the mean.Completestudies requiredetailedanalysis of depth-area-duration datathat dependon the masscurves in detail in other from a network of gauges. The methodis described of accumulation Figure 27.I0 depictsthe depth-area relation for the 24-hr storm references.23-25. taken at variousdurationsand the shownin Fig. 27.8.It also required observations successive determinationof average depthsby the isohyetalmethod.

(a)

(b)

(a) Thiessennetwork (24-hr total; Figure 27.8 Methods of determiningrainfall averages: = (b) precipitation 2.54 in.) and isohyetal map (24-hr total; averagebasin averagebasin precipitation = 2.50 in.), The arithmetic average over the basin = 39.10/15 * 2.61 in.

730

OHAPTER27

ANALYSIS FREQUENCY
H U.+

>.= O F

(n n

Distance from rain gauge to watershed center (mi) (a)

H 6
6 6 o

b g

9 R 6

500 (mi2) pergauge Area (b) Figure27.9 (a) Relationbetweenthe standarddeviationof the watershed

a4tlt,ililliT,i',l:##-:"::: :1111i:ifli'ffi ill"lH"rffi

cipitation as a function of the network density and drainageare for the Bureau') basin.(U.S. Weather Muskingum

STUDIES OF FREQUENCY 27.5 RELIABILIW

of theoretical statisticsis the central limit theorem.As a A significantdevelopment .on*-rqu"n"" of the law of large numbers,the central limit theorem statesthat for a populationwith finite varianceoz anda meanp, the distributionof samplemeansitrut ir, a numberof equally good meansfrom repeatedsamples-will be distributed as a normal disiribution with meanp, anda varianceequalto a2fn, where themselves or is the population standard deviation. This theorem does not limit the type of under$ing population distributionbut saysthat the distribution of the sampl" *9-uts The statisticalY n will approacha normal distribution as the samplesizeincreases. is the siandarddeviationof the distributionof meansand is called lhe standarderror

STUDIES 27.5 RELIABILIWOF FREQUENCY

731

9
o 00

<

500

1000
Area (mi2)

1500

2000 Figure27.10 Depth-area-duration curves for24-hr stormofFig.27'8. (FromRef' 23')

their of the mean. Listed in Table 27.6 are severalparametersof distributionsand are reliability, therefore and standard errors. It is apparent that standard errors, almost completelya function of the samplesize'

Limits Confidence
of a samplemeanbased It is possibleto placeconfidencelimits on the measurement underlyingpopulation' the of on the normal distribution of all meansand regardless of a normal variate observations two thirds of the As mentionedearlier,approximately two thirds Therefore, deviation' shouldfall betweentheiimits of + i and 1 standard +olfi. percent 95 The limits of all sample means should occur between the confidencelimits for the mean are ment requires knowledgeof the u only s2insteadof o2 is known and meat limits for a samPle confidence that are beyond the scope of this text. For more distributions of sampling use the testingand information in ine neta of inferential statistics-in particular, hypothesis statisticaldecisiontheory-the readermust turn to other sources. Approximateerror limits or control cufvescan be placedon freqlency curves. by Beardsinvolvesplacinglines aboveand below the fitted curve metdd proposed 27.6 TABLE
Measure
Mean Standarddeviation Coefficientof variation Coefficient of skewness

error Standard

"/{!_
a/\/2n

c,\/r + zcl/vzn
x6t"

- D/tn + l)(n - 2\(n + 3)

732

CHAPTER2T FREQUENCYANALYSIS TABLE27.7 ERROR LIMITSFOR FLOOD FREQUENCY CURVES Exceedancefrequency (7o,at 5ololevel) 99.9 99

Yearsof record (n) 5 10

I.)

50 0.76 0.57 0.48 0.42 0.35 0.31 0.28 0.24 o.2l 10 0.95 0.58 0.46 0.39 0.31 0.27 0.24 0.20 0,17 50

10 2.12 1.07 0.79

Q.64

0.1

20 30 40 50 '70 100

r.22 o.94 0.80 0.71 0.60 0.53 0.49 0.42 0.37 0.1

1.00 0.76 0.65 0.58 0.49 0.43 0.39 0.34 0.29

3.4r
1.65 1.19 0.97 0.74 0.61 0.54 0.44 0.36
99

0.50 0.42 0.36 0.30 0.25 90

4.41 2.11 1.52 t.23 0.93 0.77 0.67 0.s5 0.45

99.9

Exceedancefrequency (%, at 95% level)

Nate: Tabular valuesare multiples ofthe standarddeviation of the variate. Five percent error limits are addedto the flood value from the fitted curve at the sameexceedance frequency and the sum plotted. Ninety-five percent limits are subtractedfrom the flood value at the sameexceedance frequency. Log values are added or subtractedbefore antilogging and plotting.

to form a reliability band. Table27.7 showsthe factorsby which the standarddeviations of the variatemust be multiplied to mark off a 90 percentreliability band above and below the frequencycurve. The 5 percentlevel, for example,meansthat only 5 percentof future valuesshouldfall higherthan the limit, and,similarly,only 5 percent shouldfall under the 95 percentlimit. Nine of ten should fall within the band.
EXAMPLE 27.8

The maximum annual instantaneous flows from the Maury River near Lexington, Virginia, for a26-yearperiod are listed in Table 27.8. Plot the log-PearsonIII curve of best fit and determinethe magnitudeof the flood to be equaledor exceeded once in 5, 10, 50, and 100 years.Using Table 27.7, also plot the upper and lower confidencelimits.

Water (year) 1926 1927 1928 1,929 1930 t93l 1932 1933 1934

Discharge (cfs) 6,730. 9,150 6,310 10,000 15,000 2,950 8,650 I 1,100 6,360 1935 t936 1937 1938 1939 1940 t941 1942 t943 13,800 40,000 10,200 13,400 8,950 11,900 5,840 20:t00 12,300

Water (year) 1944 1945 1946 t947 1948 1949 1950 1951

Discharge (cfs) 6,680 6,540 5,560 7,700 8,630 14,500 23,700 15,100

STUDIES 733 OFFREQUENCY 27.5 RELIABILITY Solution 1. The statisticalcalculationsare summarizedas follows:

Arithmetic Mean 7 Variance s2 Skew coefficient C"

Log

11,606
53.87 x 106 " 2.4

4.001
0.051b 0.38

2. After forming an array and computingplotting positions,the data are plotted in Fig. 27.11. from TableB.2' III (Table27.9) aredeveloped 3. Plottingdatafor log-Pearson Confidencelimits are plotted in Fig. 27.II usingTable 27.7. r I
0.01 0.05 0.1 0.2 0.5 1 2 5 F t 0 g r o
d

Log-Pearson IIIFit

V
',;/

I bo 30

"

:l
t.,

? + o
d '"

5r)

360
9" 70
d

6 a o
o . 9 0 95 98 99 99.8 99.9 99.99
o

t
a

/a,

1 0 2 0 4 0 6 (1000 cfs) maximum discharge Annual

Figure 27.11 Maximum instantaneous annual flows, Maury River, Lexington,Virginia.

734

CHAPTER2T FREQUENCYANALYSIS TABLE 27,9

Chance

('/")
99 95 90 80 50 20 l0 4 2
I

/ (vr)
1.01 1.05 1.11 1.25 2 5 l0 25 50 100 200

(c" = 0.38) K
-2.044 - 1.530
\.zJ+

(t: 4.001) (sy= 0'227) y + Ksy: logQ 3.537 3.653 3.721 3.760 3.987 4.ft1 4.300 4.426 4.512 4.591, 4.666
1 44?

4,498
< )6n

0.5

-0.855 -0,062 0.818 1.315 t.874 2.251, 2.60r 2.930

9,705 15,380 19,950 26,690 32,5t0 39,030 46,360

SERIES DURATION ANALYSIS OF PARTIAL 27.6 FREQUENCY

In earlier examplesof frequency analysis,only the seriesof annual maximum or Theseextremes in the hydrologicrecord havebeen described. minimum occurrences and the manipanalysis frequency with is consistent constitutean annual series thal all floods data-say, the observed All probabilities of occurrence. ulation of annual of the Any subset series. a complete constitute or all the daily streamflows-would from a events annual maximum the partial selecting series.In completeseriesis a year the annual exceeds in one greatest event that the second record, it often happens maximum in some other year. Analysis of the annual seriesneglectssuch events. Although they generally contain the same number of events,the extreme values is usuallytermedthe without regardfor the period (i.e., year)of occurrence, analyzed partial duration series. In Table27.10 themaximumrainfall depthsthat occurredfor any 30-min period in the order areshown Maryland,1.945-1954, rainfallsat Baltimore, duringexcessive maximum The 11 series, a complete represent The observations 65 of occurrence. greatest 11 events the series. annual the represent and annual eventsare underlined partial duration the represent and asterisk by an are identified throughoutthe record series. The larger numbersoccur in both series,and hencerecurrenceintervalsfor the eventsare the same.The theoreticaldifferencesin recurrenceintervals less-frequent based on annual and partial duration series of the same length are shown in TabIe27.Il. The differencefor intervalsgreaterthan 10 years is negligible.The following exampleis illustrative. EXAMPLE 27.9 Performa frequencyanalysisof the 30-min Baltimore rainfall data in Table27.L0 as an annual and apafiial duration seriesand plot the results. Solution. SeeTable27.I2. The data are plottedin Fig' 27'1'2. r r

DURATION SERIES ANALYSIS OF PARTIAL 27.6 FREQUENCY MD,1945_1954 BALTIMORE, RAINFALL DEPTHS, TABLE i7.1O MMIMUM3O-MIN Storm num0er 1
z J
I

735

Year 1945

RF depth (in.) 0.38 0.47 0.39 0.76 0.56 0.35 0.43 0.40 0.36 0.62 0.55 0.88 0.47 0.36 1.15* 0.75 1.53* 0.51 0.88 2.04* 0.76 0,97 0.71 1.07* 0.94 r.20* 1.33* 0.65 0.47 0.84 0.68 0.63 0.47 0,52 0.49

Year 1953

Storm number 1
z
J A

RF depth (in.) 0.40 0.45 0.53 2.50* r.03 0.75 0.70 1.00*

5 6 7 8 9 1 2
J

5 6 7 8

5 6 7 8 9

0.55 0.63 0.69 t.27* 1.10* 0.88 0.97 0.59 0.46 0.50 0.55 t952 I 2 3 4 5 6 7 8 , 0.47 1.20* 0.93 0.70 0.57 0.46 0.48 1.30*

1 2 3 4 1955 1 2 3 4 5 6 'l 8 9 0 I 2

0.42 0.70 0.85 0-30 0.70 0.95 t.o2 0.50 0.65 0.55 0.52 0.45 0.54 0.60 0.80 0.95

1947

I z
J A

5 6 7 8

Nole: Underlineditems are the annual series.Asterisks identify the partial duration series.

BETWEEN THE TABLE27.1'I RELATION PARTIAL DURATION SERIES SERIES ANDTHEANNUAL

interval(yr) Recurrence Partialdurationseries

Annual series t.2 1.6 2.0 2.5 5.5 10.5

0.5 1.0 1.5 2.0 5.0 10.0

736

CHAPTER2T FREQUENCYANALYSIS TABLE27.12

Depth(in.)
Annual series
1 z
J i

Partial series

Recurrence interval (n + 1)/m 12 6

A J A A

5 6 7 8 9 10 l1

2.50 2.04 1.53 1.33 1.30 1.27 1.02 0.97 0.85 0.76 0.52

2.50 2.04 1.53 1.33 1.30 1.27 r.20 t.20 1.15 1.10 t.o7

1.7 1.5 1.3 1.2 1.1

The preceding example leads to considerationof the frequency analysis of rainfall depth or iniensity for various durations of rainfall. Design problems often require the estimationof expectedintensitiesfor a critical time period. Frequency of the rainfall record for.periodsother than the 30-min duration-for examanalysis ple, ihe maximum 5-, 10-, 20-, and 60-min occurrences-would yield a family of thesedata is to .oru"r similar to thoseof Fig. 27.10. The usual methodof presenting in intensitydata the aize conveftdepthin inchesto an intensityin in. /hr andto summ of the typical are curves duration-frequency curves as shown in Fig. 27.I3. These curves.join the frequency that of rainfall data.It shouldbe emphasized point analysis -o""u.t"tt""t from the samestorm; that is, they do not reprethat are not necessarily

2.5

/^

.i E 0) E t <

_x--

>-P

t
d

1.0

zAnnta

series

1.01

1.r1.21.1 3.5

2,

5 678910

20

interval (Yr) Recunence Figune 27.12 Difference in annual and partial duration series 1 1-year record of maximum 30-min durations' Baltimore, Maryland'

PROBLEMS 741 27.t Expand the computerprogram of Problem 26.12 to include the computationof the coefflcientofthe logarithmsofthe input data. mean,standarddeviation;and skewness order and Also, include a routine to sort the data by placing them in descending plotting positions.verify, usiiig the datain Ptoblem27.4. computethe corresponding 27.g, perform a completefrequencyanalysison one of the three 33-yearrecordsgiven in III and comparewith the normal type III or log-Pearson the tablebelow.Fit a Pearson or log-normal ofbest fit. Plot the dataand placecontrol curvesaroundthe theoretical 27.7. curve of best flt usins.Table

Year t928 t929 1930 r93l 1932 1933 1934 r935 1936 193'7 r938 1939 1940 t94l 1942 1943 1944 1945 t946 1947 1948 t949 1950 1951 1952 1953 1954 1955 1956 1957' 1958 1959 1960

River Trempeuleau Wl Dodge, (DA : 643 mi') Qp..r.(cfs) 3,700 1,700 3,360 1,650 3,600 11,000 2,5'70 4,490 7,180 1,780 3,170 6,400 3,120 2,890 5,680 5,060 2,040 8,120 4,570 5,4r0 4,840 1920 3,600 4,840 6,950 4,040 5,710 10,400 r7,400 713

Blow River Banff, Alberta, Canada (DA = 858 mi' ) (cfs) Opua,

James River VA Scottsville, (DA : 4570 mi2) Qp""r (cls)

10,200 7,590 9,280 6,610 9,850 11,000 9,490 6,940 7,'720 5,210 7;770 6,270 7,220 4,450 5,850 7,380 5,590 4,450 7,2r0 5,880 r0,320
4 )qo

75,600 44,700 45,800

2t,roo
31,400
5q 500

38,800 93,400 126,000 62,200 87,400 68,400 130,000 27,100 80,600 95,200 133,000 57,000 41,200 33,200 59,600 94,200 64900 54 snn 67,000 62,900 70,000 20,400 64,200 44,500 ,o ?no 64.200

10,080 8,570 5,460 9,180 10,120 8,680 9,060

5 160

r,r40
8,000 1,480

6,730 7,480 6,440

742

CHAPTER 27 FREQUENCY ANALYSIS

27.10. Compare results of Problem 27.9 wfih estimatesby Gumbel's extreme-valuedistribu'

tion for the 50- and 100-year events.

27.11. The pan-evaporation data (in.) for the'month of July at a site4nMissouri are
9.7 11.2 9.3 9.8 lt.7 8.8 9.2 8.7 It.2 tt.4 9.3 I1.5
11.3 I 1.8 9.3 10.9 l 1.5 8.9 10.4 10.2

Determinethe mean, standarddeviation,and coefficient of variation. What are the standarderrors of thesestatistics? Establishthe approximate95 percentconfldence limits of the mean. 27.12. On which type of plotting paper(probability,log-probability, rectangularcoordinate, log-log, semilog,extreme-value, none)would eachof the following plot as a straight line? a. Normal frequencydistribution. b. Gumbel frequencydistribution. c.Y=3X+4 d. Log-normal frequencydistribution. e. PearsonType III with a skew of zero. t' Q: 43Ao1' g. Log-PearsonType III with a skew of logarithmsequal to zero. h. PearsonType III with a skew of 3.0. 27.13. Determine the 50-yearpeak (cfs) for a log-PearsonType III distribution of annual peaksfor a major river if the skew coefficient of logarithms (base 10) is -0.1, the mean'oflogarithms(base10) is 3.0, and the standarddeviationof basel0logarithms is 1.0. 27.14. A 4Oaear record of rainfall indicatesthat the mean monthly precipitation during April is 3.85 in. with a standarddeviationof 0.92. The distribution is normal. With 95 percentconfidence, estimatethe limits within which (a) next April's precipitation is expectedto fall, and (b) the mean April precipitation for the next 40 years is expectedto fall. 27.15. Given a table of valuesof mean annual floods and corresponding drainageareasfor a number of basinsin a region, describehow regressionanalysiscould be used to determinethe coefficientsp and q in the relation Qz.tz: pAq. gaugelocationhasa range 27.16. The 80-yearrecord of annualprecipitationat a midwestern between14 in. in 1936and 42 in. in 1965.The recordha5a meanof 27.6 in. anda standarddeviationof 6.06 in. Assuminga normal distribution, (a) plot the frequency curve on probability paper,(b) determinethe probability of a droughtworsethan the 1936value,and (c) determine interval.ofthe 1965maximumdepthand the recurrence comparei&ith the apparentrecurrenceinterval. 27.17. A reservoirin the localeof Problem27.16 wlll overfill when the annualprecipitation exceeds 30 in. Determinethe probability that the reservoirwill overfill (a) next year, (b) at leastonce in three successive ybars. years,and (c) in each of three successive

PROBLEMS

743

27.18. *Using Eqs. 26.38 and 26.39 and log-probability paper, solve Problems27.16 and that the annual precipitation is log-normal. 27.17 assuming the return determine 27.19. Given the following valuesof peak flow ratesfor a small stre.am, period (years)for a flood of 100 cfs by flrst using annual peaksfor an annual series and then using all the data for a partial duration series.

Year

Peak (cfs) June I Aug.3 June7 July 2 May 18 June 3 July 4

Year 1968 1969 r970 t97r r972

Date
May 11 June 8 Sept.4 Aug. 8 May 9 Sept.8 May 4

Peak (cfs)

1963 t964 1965 1966 1967

90 300 60 80 r00 90 40

800 700 90 400 30 700 80

27,20. Recordedmaximum depths (in.) of precipitation for a 30-min duration at a single station are:

Year
1963 1964 1965 1966 1967

Date May 3 June 3 June7 June 2 June 1 Aug. 3 July 4

(in.) Depth 2.0 1.0 1.0 1.0 1.0 3.0 1.0

Year 1968 1969

Date
Aug.8 May 6 June 8 Sept.4 May 4 Sept.8 May 9

Depth (in.)

4.O 6.0 5.0 1.0 1.0 5.0 1.0

a. Determine the return period (years) for a depth of 2.0 in. using the california method with an annual series. b. RepeatPart (a) using a partial duration serles. to be c. Determinefrom the partial duration seriesthe depth of 30-min rain expectbd years. 8 (on the average) once every equaledor exceeded 27.21. For a 60-yearrecord of precipitation intensitiesand durations,a 30-min intensity of a total of 85 times. Al1 but 5 of the 60 years 2.50 in.lhr was equaledor exceeded experiencedone or more 30-min intensitiesequaling or exceedingthe 2.50-in./hr value.Use the Weibull formula to determinethe return period of this intensity using (a) a paltial seriesand (b) an annual series.

744

ANALYSIS 27 FREQUENCY CHAPTER 27,22. from the data given in the accompanyingtable of low flows, prepare a set of low-flow frequency curves for the daily, weekly, and monthly durations. (cfs)FORTHE MEANDISCHARGE LOWEST NUMBER OFCONSECUTIVE FOLLOWING VISTA, NEAR BUENA MAURY RIVER DAYS, VIRGINIA
Year 1-day 7-day 3O-day

1939 1940 194l 1942 1943 t944 1945 1946 1947 t948 1949 1950 195 I 1952 1953 1954 1955 t956 1957 1958 1959 1960 1961 1962 1,963 1964 1965

100.0 167.0 22.0 101.0 86.0 62.0 78.0 76.0 97.0 r54.0 136.0 113.0 95.0 115.0 85.0 68.0 83.0 64.0 62.0 88.0 76.0 83.0 99.0 90.0 60.0 51.0 64.0

103.0 171.0 59.4 127.0 93.9 65.9 80.7 78.6 102.0 176.0 138.0 125.0 95.3 116.0 86.1 70.0 96.1 66.3 64.1 92.6 80.9 91.7 103.0 95.0 60.6 54.1 68.7

r25.0 t94.0 69.1 1'73.0 103.0 77.4 90.3 87.1 123.0 215.0 163.0 139.0 101.0 r20.0 90.8 81.7 99.9 7r.7 75.8 107.0 117.0 103.0 152.0 105.0 70.8 62.0 76.2

27.23. For the 7-day low flows at Buena Vista given in Problem 27.22, attempt to fit a probability paper' prostraight-linefrequencycurve on log-normalor extreme-value ceedingas follows:From the original plot of the data, estimatethe lowestflow (say;4) flows (Q - q : at the high recurrenceintervals;subtractthis flow from all observed The Qr); and rcplot Qr versusthe original recurrenceintervals.Repeatif necessary. frequencydistribution. best fittiBg curve will be a three-parameter

PROBLEMS 745 the number of occurrencesof intensitiesof various 27.24. The following table surirmarizes for the given durations durationsfor a34-yearrecordof rainfall. Maximum intensities of the exceedances' made count a and storms were determinedfor all excessive -yearfrequencyand 5 on a expected exceedances of number the average for Interpolate plot the 5-yearintensity-duration-frequency curve'
were equaledor exceeded Numbei of times stated ihtensities Intensity(in./hr) Duration (min)

2.0
68

3.0 '73 51
LJ

4.0 48 26 11 3 1

5.0 2I l1
I

6.0
o

7.0 2
I

5 10 15 30 60 120

3
I

72 29
1 8 5

3s l7
6 l

7 2

analysisof over 200 flood recordsin Virginia led 27.25. The resultsof a multiple regression to the following regional flood frequencyequations: . Qt'z'v': 9'l3AeoeS2e3 Qzzt-v':20'84861S30e 3oo : 38'1A83oS Qs'v, 283 Qn', : 63'0A8o2S : 104A71e5266 Qzs-v, : l18A7esS21e Qso'v' for the given frequencyis in cfs, A is the drainageareain whefe the flood discharge betweenthe points that are 10 and in ftlmi (measured slope channei is the mi2, and S g5 percent of the total rivei miles upstreamof the gauging station to the drainage diviAe;. Devise a method for graphically portraying theseregional flood frequency relations.(Note that there are four factors,Q, T, A' and S') .25, flnd the predicted floods for the Using the regressionequationsin Problem 2'1 River, at cootes Store.Drainage atea : 215 mi2and chanNorth Fork, shenandoa-h : nel slope 44.3 ftlmi. equationsin Problem 27 '25 with the Comparethe predictions from the regression : by the frequencyanafsis inqxample27.8. Drainageatea 487 fii2 valuesestimated : and channelsloPe 2I.l ftltrl'l. storm rainfall over the city of Baltimore Referringto Fig. 2.6b, comparethe average on SeptemberI0, Ig57, computedby the isohyetalmethod, with the simpleaverage within the city. Neglectthe areato the south of total accumulationat the riin gauges isohYet' 1.0-in. of the Fit the forrnula i : AIQ + B) to the data derived in Problem 21.24 for the 5-year intensity-duration-frequency curve.

27.26.

27.27.

21.28.

27,29.

746

27 FREQUENCYANALYSIS CHAPTER 27.30. Deielop a regional flood index curve for the RappahannockRiver basin from the flood frequency data given in the following table:
(ft3/sec) FOR STATIONSlN THE PEAK FLOOD FREQUENCYDISCHARGES RIVER BASIN RAPPAHANNOCK

Station River Rappahannock near Warrenton, VA Rush River at VA Washington, Thornton River near Laurel Mills, VA Hazel River at fuxeyville, VA River Rappahannock at Remington,VA River Rappahannock at Kellys Ford, VA Mountain Run near Culpeper,VA Rapidan River near Ruckerwille, VA RobinsonRiver near Locust Dale, VA Rapidan Rivet near Culpeper,VA RappahannockRiver near Fredericksburg, VA

Drainage Typeof area(mi2) series 192 15.2 142 286 616 641 14.'7 111 180 456 | sqg
Annual Partial Annual Partial Annual Partial Annual Partial Annual Partial Annual Partial Annual Partial Annual Partial Annual Partial Annual Partial Annual Partial

2.33 (mean).
4,150 4,600 530 610 5,900 7,200 7300 8,300 11,000 12,000 12,300 14,000 '750 950 3,950 4,700 4,600 5,150 9,100 10,800 26,000 29.300

period in years Return 10 8,350 8,650 860 900 11,500 12,500 11,800 t2,400 14,500 15,200 19,000 20,000 t,750 1,900 7,100 7,700 7,000 7,300 16,400 17,600 39,900 42,000 9,000 9,20A r,290 1,310 19,900 20,500 17,200 18,000 18,100 18,900 26,800 27,500 3,350 3,550 11,600 12,000 9,800 10,100 26,900 27,600 55,000 57,500 25 14,000 14,000 2,100 2,100 34,000 34,000 25,000 25,500 24,500 25,000 42,000 42,000 6,000 6,000 21,000 21,000 !5,400 15,800 50,000 50,000 85,000 85,000 19,250 19,250 3,000 3,000 48,000 48,000 41,000 41,000 31,000 31,000 57,500 57,500 10,000 10,000 34,000 34,000 2r,5oo 21,500 78,000 78,000 117,000 117,000

27.31. From the information given in Problem 27.30, find the relation betweenthe mean shouldbe of annual flow and the drainagearea.(Note that the functional expression : rA".) the form Qz.zz the 30-yearflood for an 27.32. Using the resultsof Problems27.30 and27.3l, estima{e mi2. 540 arca of a drainage with watershed ungauged 27.33. Annual flood recordsfor a lO-yearperiod are given by:

Year Flood

1 300

2 700

3 200

4 400

5 1000

6 900

7 800

500

100

10 600

Mean : 550 cfs, median : 550 cfs, standarddeviation : 300 cfs. Use an annual seriesand the definition offrequency in a frequencyanalysisto determinethe magnitude of the 4-yearflood. Comparethis historicalvaiuewith the analytical4-yearflood obtained assuminsfloods follow a normal distribution.

PROBLEMS 747 a 30-min intensityof 2.50 anddurations, intensities recordof precipitation 27.34. For a 60-year a total of 85 times.All but 5 of the 60 yearsexperienced in.i hr wasequaled or exceeded the 2.50-in./hr value. Use the one or more 30-min intensitiesequalingor exceeding Kimball formula to determine the return period of this intensity using (a) a partial duration seriesand (b) an annualseries. to be approximately basinis determined 27.35. The total annualrunoff from a small drainage normal with a mean of 14.0 in. and a standarddeviation of 3 in. Determine the probabilitythat the annualrunofffrom the basinwill be lessthan 8.0 in. in the second yeafs. year only of the next three consecutive 27.36. Six yearsof peakrunoff ratesare given below.Assumethat the floodsfollow exactlya peak. the magnitudeof the 5O-year normal distributionand determine

Year Runoff (cfs)

1 200

2 800

3 500

4 600

5 400

6 500

21.37. Annual floodsfor a streamare normally distributedwith a mean of 30,000 cfs and a flood returnperiodT,of a32,000-cfs the average of I x 106cfs2. Determine variance in the stream. 27.38. Annual floodsfor a streamhavea normal frequencydistribution.The 2-yearflood is 40,000cfs and the l0-year flood is 52,820cfs.Determinethe magnitudeof the 25-year flood. 27.39. The 80-yearrecord of annual precipitation at Linclon, Nebraska,yields a range of valuesbetween10 and 50 in. with a meanannualvalueof 25.00 in. and a standard a sum of many random annualprecipitationrepresents deviationof 5.30 in. Because that annual variables(i.e., depth of precipitation for eachday of the year), assume precipitation is normally distributed. a. In 1936 the precipitation at Lincoln was a mere 14 in. Determinethe probability that the annual precipitation will be 14 in. or lessnext year. this amountwould be equaledor b. In 1965Lincoln received42 in. On the average, exceeded once in how many years? c. Comparethe theoreticaland apparentreturn periods of the record-highvalue of 50.00in. 27.40. Annual floods (cfs) at a particular site on a river follow a zero-skew log-Pearson Type III distributions.If the mean of logarithms(base 10) of annual floods is 2.946 and the standarddeviationof base-10 logarithmsis 1.000,determinethe magnitude of the 50-yearflood. 27,41,. Annual floods (cfs) at a particular site on a river follow a zero-skew log-Pearson 1.733and annualfloodsis Type Illdistribution. Ifthemeanoflogarithms(base10)of the standarddeviationof base-l0logarithms is L.420,determinetfe magnitudeof the 10O-year flood. of flood magnitudes record for a drainagebasin gives 10- and 5O-year 27.42 The 100-year 12,500 and22,000 cfs. Determinethe magnitudeof the mean annual flood if (a) the flood peaksfollow the index-flood curve of Fig. 27.4c atd (b) the flood peaksfollow distribution. a Gumbel extreme-value

748

CHAPTER27

FREQUENCYANALYSIS *The following parameters were computed for a stream: 27.43.

Period ofrecord : 1960-1984, inelusive. Mean annual flood : 7000 cfs Standarddeviationof annual floods : 1000 cfs Skew coefficientof annual floods : 2.0 Mean qf logarithms(base l0) of annual floods : 3.52 Standarddeviationof logarithms : 0.50 Coefficientof skew of logarithms : -2I Determine the magnitudeof the 25-yearflood by assumingthat the peaks follow a (d) log-PearsonType III distribution, (b) Gumbel distribution, and (c) log-normal distribution. 27.44. Peak annual dischargerates in the Elkhorn river at Waterloo,Nebraska,yield the following statistics: Period of record : 1930-1969, inclusive Mean flood : 16,900cfs Standarddeviation : 17,600cfs Skew of annual floods : 0.8 Mean of logarithms(base 10) : 4.0923 Standarddeviationof logarithms : 0.3045 Skew of loearithms : 2.5 Determinethe 100-yearflood magnitudeusing the uniform techniqueadoptedby Council for all federal evaluations. the U.S. Water Resources b. Determinethe'100-yearflood magnitudeassumingthat the floods follow a twoparametergamma distribution. Type III variableX has a mean of 4.0, a standarddeviationof 2.0, and a 27.45. A Pearson of I:f.(X)dX. coefficientofskewof 0.0.Determinethevalue(foursignificantfigures) . 27.46, A timber railroad bridge in Hydrologic Region2 of Texasat Milepost 738.04 on the railroad systemshownin the sketchis to be replacedwith a new concretestructure. The 50- and 100-yearflood magnitudesare neededto establishthe low chord and Determinethe designflow ratesusingthe USGS respectively. elevations, embankment Regression Equations.The drainagearea is 0.43 sq. mi, and the streambedslopeis 62 ftoer mi.

REFERENCES
t . M. A. Benson,"Plotting Positionsand Economicsof EngineeringPlanning,"Proc. ASCE
J. Hyd. Div. 88057-71(Nov. 1962). Res'68(3), L I. Gringorten,"A Plotting Rule for ExtremeProbability Paper,"J. Geoplrys. 1 813-814(Feb .9 6 3 ) .

ucENoftp.*

l l * *l* --)
.J$
^ " : 1- ' .

,
i

t \

I
I

Slrip

\l Sketch for Problem 27.46

750

CHAPTER2TFREQUENCYANALYSIS 3. VerfT. Chow, "A General Formula for Hydrologic FrequencyAnalysis," Trans. Am. Geophy s. Union 32, 231-237 (1951). 4. Water ResourcesCouncil, Hydrology Committee, "Guidelines:for Determining Flood Frequency,"Bulletin 17B, (Revised)U.S. Water Resources Council, Washington, D.C., Sept.,1981. 5. L. R. Beard, Statistical Methods in Hydrology, Civil Works Investigations, U.S. Army Corps of Engineers,Sacramento District, 1962. 6. A. Hazen, Flood Flows. New York: Wiley, 1930. 7. V. T. Chow, "Statistical and Probability Analyqisof Hydrologic Data," in Handbookof Applied Hydrology.New York: McGraw-Hill, 1964. 8. P. Victorov, "Effect of Period of Record on Flood Prediction," Proc. ASCE J. Hyd. Div. 97(Nov.1971). 9. L. R. Beard, Statistical Methodsin Hydrology, Civil Works Investigations; Sacramento District, U.S. Army Corps of Engineers,1962. 10. M. A. Benson and N.C. Matalas, "Synthetic Hydrology Based on Regional Statistical " Wat Parameters, er Re sources Re s. 3(4)(1967). 11. N. C. Matalas,"MathematicalAssessment of SyntheticHydrology," WaterResources Res. 12. Yet T. Chow, "Statistical and Probability Analysis of Hydrologic Data," Sec. 8-I, in Handbookof Applied Hydrology (V. T. Chow, ed.). New York: McGraw-Hill, 1964. 13. T. Dalrymple, "Flood-Frequency Analysis,"Manual of Hydrology,Part 3, U.S. Geological Paper1543=A. Washington, D.C.: U.S. Government Printing Offlce, $urveyWater-Supply 1960. 14. G. M. Fair,J. C. Geyer,andD. A. Okun,Water andWasteWater Engineerlng. New York: Wiley, 1966. 15. "Monthly Stream Simulation," Hydrologic EngineeringCenter, ComputerProgram23C-L267, Sacramento District, U.S. Army Corps of Engineers,July 1967. 16. R.W.CruffandS.E.Rantz,"AComparisonofMethodsUsedinFloodFrequencyStudies for CoastalBasinsin California," Flood Hydrology,U.S.G.S.Water Supply Paper 1580: Washington, D.C.: U.S. GovernmentPrinting Office, 1965. 17. W. D. Potter, "Peak Ratesof Runoff from Small Watersheds," Hydraulic Design Series No. 2, Bureau of Public Roads,Washington,D.C.: U.S. GovernmentPrinting Offlce, Apr. 1961. 18. U.S. GeologicalSurvey, "Technique for Estimating the Magnitude and Frequencyof Floods in Texas," WaterResources Investigations Report 77 -110, 7977. t9. M. E. Jennings,W. O. Thomas, Jr., and H. C. Riggs, "Nationwide Summary of U.S. GeologicalSurvey's RegionalRegression Equationsfor Estimating Magnitude and Frequencyof Floodsat Ungauged Sites,"U.S.G.S. WRI 93-1, Reston, VA, 1993. 20. U.S. Geological Survey, "Selected Streamflow Characteristicsas Related to Channel Geometry of Perennial Streamsin Colorado," Open-File Report 12-160, Water ResourcesDivision, Lakewood,Colorado,May 1972. 2 r . E. W. Steel,l{ater Supplyand Sewerage,4th ed. New York: McGraw-Hill, 1960. 22. D. M. Hershfield,"Rainfall FrequencyAtlas of the United States,"Tech.PaperNo. 40, U.S. Weather Bureau.1961. 23. HydrologyHandbook, ASCE Manual of Practice,No. 28, 1949. 24. W. G. Knisel, Jr.,and R. W Baird, in Al?SPrecipitation Facilities and RelatedStudies. Washington, D.C.: U.S. Departmentof Agriculture, Agricultural Research Service,1971, Chap.14, , \ R. K. Linsley, Jr., M. A, Kohler, and J. L. H. Paulhus,Applied llydrology. New York: McGraw-Hill, 1949.

3(4)(re67).

Appendixes
A APPENDIX:
FACTORS ANDCONVERSION PROPERTIES. CONSTANTS. TABLEA.1 WATER
atm Heat of vaporizationof water at 1-.0 Gasconstants(R) 540 callg: 970 Btu/Ib R : 0.0821(atm)(liter)/(g-mol)(K) R : 1.987g-ca1/(g-mol)(K) R : 1.987Btu/(lb-mo1)('R) Specificheat of air Accelerationof gravity (standard) g : 32.17ftlsecz:980.6 cm/sec2 Cp : 0.238call(g)("C) Density of dry air at OoCand 760 mm Hg 0.001293g/cm3 Heat of fusion of water 19.7 callg: l44BtulIb Conversionfactors per squaremile : 0.03719 inch I second-foot-day

persquare *t" 1inch ofrunofr

: 33:3 :::::r1;?t-0"t,
: 2,323,200cubic feet

e = 2.71828 log e = 0.43429 ln 10 : 2.30259 Metricequivalents foot : 0.3048 meter mile : 1.609 kilometers acre : 0.4047 hectare 4047 squaremeters I squaremile (mi') : 259 hectares 2.59 squarekilometers (km2) 1 acre foot (acre-ft) : 1233 cubic meters I million cubic feet (mcf ) : 28,320 cubic meters I cubic foot per second(cfs) : 6.6rta, cubic metersper second 1.699 cubic metersPer minute I acre-in. per hour : 1.008cubic feet per second(cfs) " 1 second-foolday (cfsd) = 2447 cubic meters I million gallons (mg) = 3785 cubic meters 3.785 million liters 1 million gallons per day (mgd) = 694.4 gallons per minute (gpm) 2,629 cubic meters per :ninute 3785 cubic metersper day

per second t3jL":'"i'ff1":* n'"' I cubic root : ? I horsepower : 3Jr-f,f*:llld,p".,..ond

752

A APPENDIX OF WATER TABLE A,2* PROPERTIES

Traditional U.S.Units Temperature Specific gravity ('F)

iz

Unit weight 0b/ft3)

Heat of vaporization (Btu/lb)

Kinematic viscositY (ft'lsec)

Vaporpressure
psl in.Hg

40 50 60 70 80 90 100

0.99987 0.99999 0.99975 099907 0.99802 0.99669 0.99510 0.99318

62.416 62.423 62.408 62.366 62.300 62.217 62.11,8 61.998

t073 1066 1059 r054 r049 l044 1039 1033

1,.93x L67 x - 1.41 X 1.21x 1.06 x 0.929x x 0.828 X 0.741

10*5 10-s 10-5 105 10-s l}-s l0-5 10-5

6.11 0.09 8.36 0.12 12.19 0.18 17.51 0.26 24.79 0.36 34.61 0.51 47.68 0.70 64.88 0.95

0.18 0.25 0.36 0.52 0.74 1.03 1.42 r.94

Sl Units Temperature Specific Density gravity (o/cm1 fC) 0 5 10 15 20 25

JU

Heat of vaporization (cal/g)

Kinematic viscosity (cs)

Vaporpressure (mmHg) 4.58 6.54 9.20 12.78 17.53 23.76 31.83 42.18 55.34 92.56 149.46 233.79 355.28 525.89 760.00 (mb)
(g/cm')

35 40 50 60 70 80 90 100

0.99987 0.99999 0.99973 0.99913 0.99824 0.99708 0.99568 0.99407 0.99225 0.98807 0.98323 0.97780 0.97182 0.96534 0.95839

0.99984 0.99996 0.99970 0.99910 0.9982r 0.99705 0.99565 0.99404 0.99222 0.98804 0.98320 0.97777 0.97179 0.96531 0.95836

597.3 594.5 591.7 588.9 586.0 583.2 580.4 577.6 574.7 569.0 563.2 551.4 545.3 539.1

1..790 .t.520 1.310 1.140 1.000 0.893 0.801 0.723 0.658 0.554 0,4'14 0.4r3 0.365 0.326 0.294

6.23 6.11 8.89 8.72 L:2.27 r2.s1 17.38 17.04 23.83 23.37 32.30 31.67 43.2.7 42.43 5'7.34 56.24 75.23 73.78 123.40 125.83 199.26 203.19 311.69 317.84 473.67 483.01 70r.13 714.95 1013.25 1033.23

B APPENDIX

753

B APPENDIX:
TABLE 8.1 AREAS UNDERTHE NORMALCURVE

=1,f:*"*'"t'a, ,<,t
01 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 r.0 1.1 r.2 r.3 r.4 1.5 r.6 r.7 1.8 r.9 2.0 2.1 2.2
z-5

.02 .0080 .0478 .0871 .1255 .1628 .1985 .2324 .2642 .2939 .3212 .3461 .3686 .3888 .4066 .4222 .4357 .44't4 .4573 .4656 .4' 126 .4783 .4830 .4868 .4898 .4922 .4941 .4956 .4967 .4976 .4983 .4987 .4991 .4994 .4996 .4997

.03 .0120 .0517 .0910 .1293 .1664 .2019 .2357 .2673 .2967 .3238 .3485 .3708 .390'7 .4082 .4236 .4370 .4485 .4582 .4664 .4732 .4788 .4834 .487r .490r .4925 .4943 .4957 .4968 .49'77 .4983 .4988 .499r .4994 .4996 .4997

.04 .0159 .0557 .0948 .1331 .1700 .2054 .2389 .2704 .2995 .3264 .3508 .3729 .3925 .4099 .4251 .4382 .4495 .459r .467r .4738 .4793 .4838 .4875 .4904 .4927 .4945 .4959. .4969 .4977 .4984 .4988 .4992 .4994 .4996 .4997

.05 .0199 .0596 .0987 .1368 .1736 .2088 .2422 .2734 .3023 .3289 .3531 .3749 .3944 .4115 .4265 .4394 .4505 .4599 .4678 .4744 .4798 .4842 .48"18 .4906 .4929 .4946 .4960 .4970 .49'78 .4984 .4989 .4992 .4994 .4996 .4997 .0239 .0636 .1026 .1406 .r772 .2123 .2454 .2'764 .3051 .33t5 .3s54 .3770 .3962 .4t3t .4279 .4406 .4515 .4608 .4686 .4750 .4803 .4846 .4881 .4909 .4931 .4948 .4961 .4971 .4979 .4985 .4989 .4992 .4994 .4996 .4997

.o7
.0279 .0675 .1064 .1443 .1808 .21,57 .2486 .2794 .3078 .3340 .3577 .3790 .3980 .4147 .4292 .4418 .4525 .4616 .4693 .4756 .4808 .4850 .4884 .4911 .4932 .4949 .4962 .49' 72 .4980 .4985 .4989 .4992 .4995 .4996 .4997

08 .0319 .07t4 .1103 .1480 .1844 .2190 .2518 .2823 .3106 .3365 .3599 .3810 .3997 .4162 .4306 .4430 .4535 .4625 .4699 .4762 .4812 .4854 .4887 .4913 .4934 .4951 .4963 .4973 .4980 .4986 .4990 .4993 .4995 .4996 .4998

.09 .0359 .0753 .rr4r .1517 .1879 .2224 .2549 .2852 .3133 .3389 .3621 .3830 .4015 .4177 .4319 .444r .4545 .4633 .4606 .4767 .4817 .4857 .4890 .4916 .4936 .4952 .4964 .4974 .498r .4986 .4990 .4993 .4995 .4997 .4998

2.4 2.5 2.6 2.7 2.8 2.9 3.0 3.1

J.Z 5.5

3.4

.0000 .0398 .0793 .1179 .t554 .1915 .2257 .2580 .2881 .3159 .3413 .3643 .3849 .4032 .4192 .4332 .4452 .4554 .4641 .4713 .4772 .4821 .486r .4893 .4918 .4938 .4953 .4965 .4974 .498t .4987 .4990 .4993 .4995 .4997

.0040 .0438 .0832 .1217 .1591 .1950 .2291, .2611 .2910 .3186 .3438 .3665 .3869 .4049 .4207 .4345 .4463 .4564 .4649 .4719 .4778 .4826 .4865 .4896 .4920 .4940 .4955 .4966 .49:15 .4982 .4987 .4991 .499? .4995 .499't

4.0

.499968

London: CambridgeUniversity Press,1957 (for Mathematical Statistics. Source:AfrerC. E. Weatherburn, z : 0 to z : 3,1); C. H. Richardson,An Intoduction to Statistical Analysis. Orlando, FL: Harcourt Brace Jovanovich.1994(for z = 3.2to z: 3.4); A. H. Bowker and G. J. Lieberman,EngineeringStatistics. EaglwooGCliffs,Nf: Prentice-Hall, 1959 (for z : 4.0 and 5.0)'

u?

o F o\ t -

o\ 6 $ o J

a.. < +

cn xt F *

F$

c{ r) \o S

d h +

r) J n +

$ Ri + r+

N F a. st

co cr, o sf

cn N N $

F. $ i *

o\ \c) c) <

o 6 o\ o

o oo ; N o\ 6 m,cn

h + r o

-l \O \o o

r) r n co

0\ 6 + o

n O + co

c..l o

co c{ o.l o

ar 6 cn n

n !. Q cn I

q\ $ o c{ h

\o q) al

!! \O \ al

o r-\o d

\o F. n cn

- + n rO O n a\ + q e q oi * * * \ - q q "? n i q i
+ cq o s o o o o cn .n ci o o o

cO al O\ r) (j

O cn h ri

\O rn cO cr\ S cO O : t rr oo o F. -i
cn ca o

o\ N n ! S l-.. Q N + Sp N q q oq oq \ I q q -t q 9 n =q n cl
o cQ o N ol N ot ol N N al N e.l

O\ F- N F- rr $ \a

1t E.l Q \O

o
lf,

C.t $ ri o J J

st ; *

ci oi -

r O

+ O

6 d O

r o\ 6

di ri d ;
Fq cN Ft-q N rl,ot F- t-q q ol at

;
t-. \o (\l 6l

.d ;
c.l \o c.l cl \O r) ol 6l

6i 6i 6i 6i 6i c.i 6i 6i ..i $i 6i ci 6i Gi oi 6i ..i t..i c.i c.i 6i 6i .'i c'i

O + cl N O ci q al o\ d q N FO 6! ol ca o\ o.l O\ F.. -': 6l o \o -': ol \O S -1 al c-t -: N 0o O .': c.l l= \9.o !9 $ ! c q q q a] ol N e.:l O Q O\ 0o r) !1 t> O, q\ I q] n S - q q q q oc q oq \ i i : : * * # r * \n \ :

V) O c) N c-.1 i oo + J F + : O\ O\ O\ oO oO cO t--

s + F-

\Q \Q \O n or O \o e{ oo * t-- \O \O r) rn

o\ rf

ci h .+

ro +

O\ * r) d \o ca cO N

Ci

O. t' :f, n o'!a) r i O

oo $ cl d

o o c

O n O + co O .i + + + c.l O - oi = e.l 6 v) \o F.- co o\ o ; * - i cl cj q q c\ cl el cl Q c.l q

: * * * i * d i i i i i * * i * d * ; i d * H *

S O\ m F o\ O -' O o\ @ cn oQ cQ tt- q\ n Cl a'l N N co 6 o d S + o cn cn ei cl - Q Q o\ e1 rl q q q q crl c.l q q dl q c.l c'l 6) a] cl q

0)
G

e
o
N

N d a'-o\ o* 6l +\DF 6* cn rt F- o\o n!-{ !! A:t n n v? 9 9 9 9 9 q \ \ \ \ \ \ \ \ t t ? T I T oq s n i n n v? v-1

O O O O\ O\

t'.. r) $

o] O\ Ft o

O r) O r) \o

O\ N u) oo q, a a Q 9p \O $ o sf r \o 0oo\QQ

Q \O N

z
F

o o
O (L

c) J (t,

\O O * 5 d\

) -

d) tf

F \O ) r) * - - O O. \O n + c{ - O\ \O $ m C9 I !o !! - - n "1 { { { "1 I q - q .t i c] ct q al -1 -1 -.: - -'1 -l q c q ? I c r l r r l r r l l l l r l l l l l l l l l l l l l l l

i \O n \E \o \O : U U :U * oo \o : U \O : 5 O tr: U :U U r) O\ .i c.i F tr: 9: 9 U al ri a:9 r) \o F: F F t*-l tr-: 0\ 6\ a-: 6 tt- h N O ;i oo @ -1,: -i N cn 0a -l @ cn oo -i + < -i ,ai + r) 0o oo -i -i * A oo ^ \O !A 0o -l t.A j F!Q i \O A 0o A \o !? 0o A !! !! oo A Q !A oo A

\O oo O -

O O. t-- t

c{ oo $

O \a O !a O +

oo N \O g\ 6 - q\

O .i

F-

l
N t m A

U)

o (E

C)

o (L

3
lu TL r

ro (\l

\O 6 \o : U

\O $ A

I o i

l 6l $

| r- -

| o

| *

I -

| -{ !.

| I

| F

l m F

| \o o

I cl

z a

CE lrJ (L (f
IL

F- 0Q S \O O a.l S \O 0o O - c.l S G oo O c.l + rr- O. ; + \o or' cat st t- O\ q q - \ i \ \ "9 "q "e * q q q q q C q I -: - - - - ci c'l oi q cj cl q' r | | r I r I r | | | | | | | I | | | l ' l I I I | | | | |

!n o\ s

o\ n

r) o

Q \o -

r) oo o vl

(o (\l ro

r) oo ro o6 6.O a o I I

* * rcj |

\O i< o |

cl \o f.o I

O 5 F o I

o\ O ol V O\ ;,h dj * + oo od - O\ O\ o o o d o I I I t l

+ oo O\ o I

O c-t c) -.j I

\O n O .i I

6 o\ O -j |

- A \9 c! co \o o d i - c.l a{ .j .i -i .j | I I I

O oo c.l -j I

l-i o i I 6

cn rl m .j |

6 cq .j |

ci e'I $ ..i |

oo !D t .i |

! q\ + i |

{ oI n ..j l q

!Q I \o rl \n oo i S n n O -'i -.i..i .j , l I $ 9o or \O O t'- v) cl

IJJ l

tt* O + \O 6 .i

v
c! o lrl
J C > o

O C O O C O C C O O O -

$ co O !a C.l + \o o\ aj b O + 6 o oo + O, r} q a 1 q n q q q \ i \ \ "c * q q q q g -1 -1 c'l dt dt n n n \? \ oq og - - c ^ l c l o l c { o l

O o\ O\ ol F- r) \O O at- (-'. O t\

\O 6 co q, * H

Q 6 \Q

\Q O h oo

t t

l l

l l l

: E V ( /)o

6 .E...

d 6 i 6 i 6 i o l c t c . l a l e { 6 . . { q . . t -

q q oq \ I n =q q ol - q q o9 \ 9 !.1 n (.l q -1 q q T \ 9 ! 1 I I : ? O O O C O O O C O O

ro

g$ i5 AA$EqEXB$$Fn q3 R5i E E3 E5 E E ' '

X:.i :l:l " . iX -':'j: .'-i -':'j -j -i -: -l J a : 1 . . 1 . ' lX : : - -j: ': a -' -j -':r - 'r - 'r 6

3E *f : iF I 3 E E i 5 E E E g E E E g E R = i t n E : F
; f : :': : : -' -.r' : -i' r i - - .-i j - j .

|o

EtA$[Fg3$$*ilx:s:ssi5353$5S5
s a q E E r s: s EFil R sE:835n$gasFRri -' R Y : = - d o c; c; d o ci o ci d : f :l'i r
: : : : : ! i : i
* l * * i * i

r.c)
A,I

RKlITE=:::==9===Iii35:A35ii=
g.
(d

c)

o C) J 0) o O (L

s BilRHFhHililr+nssIEAFIpA$F= ; & ;d U # # dt cb dt "e ob oq "e "C e oq e \ \ \ \ \ I : : : X : :

d i o o i o c i c - 3 3 3 3 ; i i c d d b l c o o d o o i c

o) o c)
N

z
ro

0)

q $oE : E E E E a : l = g E 5 : A = n Fn R X :cfi * i 8fl d dH o c ig : 33 33: ; =; = = d:j:j : J ;i J J J J o o c;

r
o

(r

(\

s s $ ; s s F FR PR r fH i re E E - rri$9-{S c [ c cq oq i -q e! i oq +r + i 3 : : 3 3 d e i ; e d : b b b c ; o+ T Tt t I I I I | | t't
o

r I r | | | | I r , ' T T T T T T

+ 3 s -E si x : ' ;' nn RgR=gg333 i ;I d , , X d . i -;i ' - i. , -i ii - _ i - i

::'i::',

c, 6 r 66n eq e
I I | |

i J f | I I I | | I | | I I r r T l ' r r I | | I I I I I |

HX n N= E E ' XR R , X
-i-

e {

(o
lo

(\

,: e ? ? i = $ e E F * E a a a E $ A q q l E l 5 : n iiiTTTTTTTiTiiiTTTTTTTTTTTTTTT

s
O q o ul

O)

? $ * : F s s h $ 8 3$i q ln !8 ef E iA EE f rlS;1
F<

+TtTTtTtTiiTiTTiiiTTTTTTTTTTTT

o
F

=.9
j o v

c)

=t : : : : : : 3 : I : : l : : : * t f ri l I ? ? ? ? ? ? ? ? T T T Tr r I | | I | | I | | I I I I | |

: 3 R . .ir x i:

3
q

755

759 Open channelflow, 131 Orographicprecipitation, 19 Overland flow, 561 Pan, classA, 129 Parshallflume, 131 Partial duration series,734 Partially penetratingwells, 473 Peakflow,3lI-343 Penmanmethod"97 Permeability,438 Piezometrichead,435 Point precipitation, 27 101 Potentialevapotranspiration, Potentialtheory, 463 Precipitablewater, 16 7, 15-39, 164 Precipitation, Probability, 67I-699 Probablemaximum flood. 373 Probablemaximum precipitation (PMP), 34,392 Rainfall. 281.727 127 Rain gauges, Rationalmethod,312 Recession, 574 equation,692 Regression Relativehumidity, 16 135 Remotesensing, 245 Reservoirs, Reynoldsnumber,437 Laboratory (RRL) Road Research Method,631 Routing,234-257 Runoff, 153-169, 302 Runoff curve number,73 Saltwaterintrusion. 474 Service SCS:seeSoil Conservation SCSAtt-Kin Tr-20 Method,254 SCSmethod,73 SCSTR-55 Method, 320,445 SCSTP-149Method,331 198-201 S-hydrograph, Simulationmodels,548-59 1, 594-628,630 Snow,265-305 Snowmelt,271-284 Snyder'smethod,207 Service(SCS) Soil Conservation 73 number, runoff curve 211 method, unit hydrograph 165 55, Soil moisture,25, Specificyrelds,432 Standarddeviation,683 Standardproject storm (SPS),395 Model IV Stanford Watershed

(swM-IV),550
67| -699, 7O8-739 Statisticalanalysis, Steadyflow routing, 252 methods,508 Stochastic. 402 Storm drainage, Storm Water ManagementModel (swMM), 604,650 111-118, l7 l, 574 Streamflow, and Reservoir StreamflowSynthesis Model (SSARR), Regulation 565 446 Streamlines, 448 Surfaceof seepage, runoff. 177 Surface

swMM,650

Syntheticunit hydrographs,205-227, 335 Temperatureindexes,291 467 Theis' nonequilibrium approach, Thiessenmethod. 30 20 Thunderstorms, 182-185 Time of concentration, 535-544 Time series, 444 Transmissivity, 95- 100 Transpiration, Unconfined aquifer, 443, 461 Uniform flow field,463 188-227 Unit hydrographs, U.S. GeologicalSurvey Index-Flood Method.336 Survey U,S. Geological Rainfall-Runoff Model, 597

University of Cincinnati Urban Runoff Model (UCURM), 655-658 flow,442,467 Unsteady flow routing, 252 Unsteady 309 Urban drainage, Urban runoff models.309-351 114 Velocity measurement, Velocity potential, 440

Water budget,86, 293, 5'l.I Water equivalent,268 : Water t'apor, 16 WatershedHydrology Model

(usDAHL),563

t53,570 Watersheds, Weirs,131 Wellfields,.465 Well function, 468 Wells,460-473