PREFACE xiii
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 to Environmental Problems 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
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 GreenAMPT Model 64 HugginsMonkeModel 67 Holtan Model 68 Recoveryof Infiltration Capacity 69 Temporal and Spatial Variability of Infiltration Capacity SCS Runoff Curve Number Procedure 73 76 @Index
70
103
115
oONTENTS Vii
PART TWO AND fT'IEASUREMENTS MONITORING 121 HYDROLOGIG 7 CHAPTER DataSources Hydrologic
7.1 7.2 7.3 7.4
123
I24
144
153
153 and Watersheds, DrainageBasins Catchments, 155 Affecting Runoff Basin Characteristics RudimentaryPrecipitationRunoffRelationships 164 166 StreamflowFrequencyAnalysis 168 Forecasting StreamflQw
viii
ooNTENTS lI.4 11.5 11.6 Il.7 BaseFlow Separation I77 HydrographTime Relationships Time of Concentration I82 BasinLae Time I82 181
234
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
359
CONTENTS
IX
HydrologicDesignFrequency Criteria DesignStorms 373 Critical EventMethods 391 Airport DrainageDesign 400 Designof Urban Storm Drain Systems FloodplainAnalysis 409
365
402
427
Introduction 427 Groundwater FlowGeneral Properties 429 Subsurface Distribution of Water 429 GeologicConsiderations 430 Fluctuationsin GroundwaterLevel 433 GroundwaterSurface Water Relations 433
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
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
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
481
505 5O7
HydrologicSimulation 508 Groundwater Simulation 509 Hydrologic Simulation Protocol 524 Corps of EngineersSimulation Models
526
548
594
597
CONTENTS Xi
630
Urban StormwaterSystemModels 63I Urban Runoff Models Compared 659 VendorDevelopedUrbanStormwaterSoftware 663
669 671
2 6 .r0
RandomVariablesand StatisticalAnalysis 672 Concepts Probability of 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 Correlation Applications 697 and
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
Preface
New federalthrusts,the growis Watermanagement taking on new dimensions. public sentimentregardingenvironmentalprotecing list of global iisues,and strong tion havebeen the principal driving forces. and development manageIn the early yearsof the 20th century,waterresources on water supplyand flood control' Today,these ment were focuied almostexclusivd issuesare still important, but protecting the environment,ensuring safe drinking and compete equally for recriatioinal experiences water, and providing aesthetic for public is pressing an environmentallyconscious attentionand funds.Furthermore, practices,with fewer structural compomanagement greaterreliance on improved to ients, to solve this nuiion'. water problems.The notion of continually striving replacedby one of husbandingthis preciousnatural provide more water has been resource. There is a growing constituencyfor allocatingwater for thebenefitof fish and wildlife, for protectionof marshesand estuary areas' and for other natural system uses.But estimatingthe quantitiesof water neededfor environmentalprotectionand is for maintainingand/or restoringnatural systems difficult, and there are still many of are ,putt", and our understanding the complexinteracunknowns.Scilntific data is of an scales rudimentary.Indeed,this is a critical issue' tions inherentin ecosystems sincethe quantitiesof water involvedin environmentalprotection can be substantial of and competitionfor thesewatersfrom traditional water usersis keen' The nations decisionsregardingnatural systemsdecisions that are the world are facing major urgencyassociladenwith significantectnomic and social impacts.Thus there is_an and of ecologicsystems of their hydrologic a atedwith developing betterunderstanding components. Water policies of the future must thereforetake on broader dimensions'More and mustbe placedon regionalplanningand management, regionalinstituemphasis mustbe practicedat, Watermanagement be devised. this tions to accommodate muJt Land useand wateruseplanningmustbe more andbetween,all levelsof government. 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 applyingwaterin all waterusingsectors; for and analyzingand designingwater managementsystemsincorporating technical, economic, environmental, social, legal, and political elements. knowledgeof hydrologicprinciplesis A a requisitefor dealingwith suchibsues. This fourth edition hasbeendesigned meet the contemporaryneedsof water to scientistsand engineers. is organizedto accommodate It 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 problemsolvingtechniques includethe use of spreadsheets appropriate.New feaas turesof eachchapterincludean introductory statement contentsand,at the concluof sion 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 recognizedfor their helpful commentsand reviews,parare ticularly the following reviewers. Gert Aron, ThePennsylvania StateUniversity JohnW. Bird, Universityof NevadaReno IstvanBogardi, Universityof Nebraska RonaldA. Chadderton, VillanovaUniversity RichardN. Downer,Universityof Vermont Bruce E. Larock, Universityof CalifurniaDavis Frank D. Masch,Universityof TexasSan Antonio Philip L. Thompson, FederalHighwayAdministration A specialnote of thanks is due to Dr. John W. Knapp, President the Virginia of Military Institute,coauthorof previouseditionsof this book, for his pastcontributions andvaluableguidance. WarrenViessman. Jr. Gary L. Lewis
PARTONE
L.
Chapter1
lntroduction
I Prologue
The purPoseof this chaPteris to: . , . . Define hydrology. . earth science' ,.1L Give a brief niJiory of the evolution of this important Statethe fundamentalequationofhydrology' decision trow ffiofogic principle, "urib" appliedto supplement Demonstrate management' for support systems water and environmental
1.2 A BRIEFHISTORY
focusedtheir i Ancient philosophers flows Production of surfacewater oc"ur."n"e of water in variousstag to from the seato the atmosPhere t faulty'l early speculationwas often reservoirsth of large subterranean is interestingto note, however'tha
*suoeriornumbersindicatereferencesattheendofthechapter.
CHAPTER1
INTRODUCTION
Greek aqueducts both conveyance on crosssectionand velocity.This knowledgewas lost to the Romans,and the proper relation betweenarea,velocity, and rate of flow remainedunknownuntil Leonardoda Vinci rediscovered duringihe it 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 be the predecessor modern to of notionsof 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 ancienttimes,it is only fair to state in that practical applicationof varioustryorotogic principleswas often carried out with considerable success. example,about4000 s.c. u du was constructed For 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 Palissyindependently 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 hydrologicvariables. of The adventof what might be calledthe "modern" science hydrology of is usually considered beginwith the studiesof suchpioneersasPerrault,Mariotte, to and Halley in the seventeenth century.r'a Perraultobtainedmeasurements rainfall in the Seine of 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 rate of evaporation the the of 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 be drawn reggrding to the hydrologic phenomena being studied. brth numerousadvances hydraulic theory in zometer,the Pitot tube, Bernoulli's theorem, ples.8 perimental hydrology flourished.Significant ydrology and in the measurement surface of water. Such significantcontributionsas HagenPoiseuille'scapillary flow equation, Darcy's law of flow in porous media, und th" DupuitThiem well formula were evolved'elrThe beginningof systematicstream guoling can also be traced to this period' Although the basis for modern hydrology wui tirrr established the ninein ,)
of teenth century, much of the effort was empirical in nature. The fundamentals In or not yet beenwell established widely recognized. the early physicalhydtotogyhad of years of tle twJntieth ""niury the inadequacies many earlier empirical formulabegan to known. As a result, interestedgovernmentalagencies tions becamewell From about 1930to 1950,rational programsof hydrologicresearch' their own develop analysis began to ieplace empiricism.3 Sherman's unit hydrograph, Horton's to infiltration theory, und Th"it's nonequilibriumapproach well hydraulicsare outof the great progressmade'r2'o 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 of continueddevelopnient highspeeddigital computershavemadepossible, both mathematicalmanipulationsthat would iense,extensive a practical and an economic havebeen overwhelmingin the past. For a more compiehensivi historical treatment, the reader is referred to the and their coworkers'1'2'4'5'15 works of Meinzer,Jonls, Biswas,
CHAPTER1
INTRODUCTION
FiguresI'I,I.2, and 1.3 showa hydrologic budgetfor the coterminous United States, conceptualized a hydrologiccycle,andthe distributionof a precipitationinput, respectively. Thesefiguresillustrate the components the water cycle with which a of hydrologist concerned. a practicalsense, is In somehydrologicregionis dealtwith and a budgetfor that region is established. Suchregionsmay be topographicallydefined (watersheds river basinsare examples), and politically specified(e.g couniy or city limits), or chosenon some other grounds. Watersheds drainagi tasins are the or easiest deal with sincethey sharply define surfacewater boundaries. to Thesetopographically determinedareasare drainedby a river/streamor systemof connecting rivers/streamssuch that all outflow is discharged through a single outlet. Unfortunately,it is often necessary deal with regions that are not well suitedto tracking to hydrologiccomponents. theseareas, hydrologist For the will find hydrologicbudgeting somewhatof a challenge. The primary input in a hydrologicbudgetis precipitation.Figures 1.11.3 illustrate this. Someof the precipitation (e.g.,rain, snow,hail) may be interceptedby trees,grass,other vegetation, and structuralobjectsand will eventuallyreturn to the atmosphere evaporation. by Onceprecipitationreaches ground,someofit may fill the (becomedepression depressions storage), part may penetralethe ground (infiltraie) to replenishsoil moisture and groundwaterreservoirs,and some may become surface runoffthat is, flow over the earth's surfaceto a definedchannelsuchas a stream. Figure 1'3 showsthe dispositionofinfiltration, depression storage, surface and runoff.
and vegetation
'1.'
*rd{i;
" 1 ) t t l l ,
p\
ffi
1vtivv
P P P P P P
t t
E' evaporation;P' Figure 1.2 The hydrologiccycle: ?, transpiration; 'surface groundwater flow; and I' runoff; G, R, p.Sqipi*i"tt inflltration.
'l,r\
t SSeamflow (hyclrograPh)
CHAPTER1
INTRODUCTION
water enteringthe ground may take severalpaths.Somemay be directly evaptransferfrom the soil to the surfaceis maintained.This can easily oratedif adequate a high groundwater table (free water surface) is within the limits of occur where usingsoil moistureor groundwaVegetation capillary transportto the ground surface. by can also transmit infiltrated water to the atmosphere a processknown tei directly transpiration.Infiltrated water may likewise replenishsoil moisture deficiencies as and enter storageprovided in groundwaterreservoirs,which in turn maintain dry weather streamflow.Important bodies of groundwaterare usually flowing so that 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 : runoff ottiut"ty reaches surface.Surface an and rivers,and finally reaches ocean.Along the course major streams like), flowsto evaporationand infiltration can also occur. of a stream, The foregoing discussion suggeststhat the hydrologic cycle, while simple in complex.Pathstaken by particlesof water precipiconcept,is actually exceedingly are numerousand varied before the seais reached.The time scale tated in any arca minutes,days,or years. may be on the order of seconds, basedon theprocessesillusgeneralhydrologicequationcan be developed A 1.2 and 1.3. ConsiderFig. 1.4. In it, the hydrologicvariablesP, E, T, trated in Figs. R, G, and l are as definedin Fig. 1.2. Subscriptss and g are introduced to denote R, respectivd. For example, vectorsoriginatingaboveand belowthe earth's surface,
["
Earth's surface Surface channels
R2
signifies groundwaterflow that is effluent to a surface streamoand E, represents areas.Letter S stands evaporationfrom surfacewaterbodiesor other surfacestorage as specified A hasa lower boundarybelow regionunderconsideration The for storage. which water will not be found. The upper boundary is the earth's surface.Vertical boundsare arbitrarily set asprojectionsof the peripheryof the region.Remembering in, that the water budgetis a balancebetweeninflows, outflows,and changes storage, whereall values statements, be translatedinto the following mathematical Fig. 1.4can are given in units of volume per unit time: 1. Hydrologicbudgetabovethe surface P+R1 RrIRrE"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 droppedfrom Eq. 1.3 sothat letterswithout subscripts are valuesof surfaceflow, undergroundflow, evaporation, to total precipitation and net the transpiration,and storage; hydrologicbudgetfor a regioncan be written simply as p_RG_E_T:LS (1.4)
whereterms For This is the basicequationof hydrology. a simplifiedhydrologicsystem G, E, and Z do not apply, Eq. 1.4 reducesto pR:AS (1.5)
of Equation 1.4 is applicableto exercises any degreeof complexity and is therefore all hydrologicproblems. 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 Precipitationis measured rain or snowgauges various devicessuchas weirs, flumes,velocity meters, using flows can be measured of locatedin the rivers and streams the area.Under goodconditions and depthgauges percent or more accufate,but large floods cannot be these measurementsare 95 Soil directly by current methodsand dataon sucheventsare sorelyneeded. measured usingneutronprobesand gravimetricmethods;infiltration moisturecan be measured can be deterrnined locally by infiltrometers or estimated through the use of of precipitationrunoff data. Areal estimates soil moistureand infiltration are generare The extentandrate of movementof groundwater usually illy very crude,however. are dataon quantitiesof groundwater determine, and adequate difficult to exceedingly for of the geologyof aregion is essential groundwater 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 Most estimates evapotranspiration the presentstateof developmentiof science. or masstransfermethods, empirpans,energybudgets, obtainedby usingevaporation ical relations.A predicamentinherent in the analysisof large drainagebasinsis the fact that rates of evaporation,transpiration,and groundwatermovementare often to assumed be highly heterogeneous. that it can The hydrologicequationis a usefultool; the readershouldunderstand the magnitude and time distribution of be employed in various ways to estimate hydrologicvariables.An introductory exampleis given here,and otherswill be found throushout the book. EXAMPLE 1.1 The average 20 wabrshedreceived in. of precipitation. In a given year,a 10,000mi2 measured the river drainingthe areawasfound to be 700 cfs (cubicfeet in rate of flow and of per second). Make a rough estimate the combinedamountsof waterevaporated the region during the year of record. transpiredfrom Solution. Beginningwith the basic hydrologicequation P  R  G  E  Z : A S can and and sinceevaporation transpiration be combined, ET:PftGAS (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 drainageareais quite large (measured hundredsof squaremiles),a presumpdivide is probadivide (boundary)follows the surface tion that the groundwater In bly reasonable. this casethe G componentmay be consideredzero. The 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 wellwateredregionswith balancedwithdrawalsand good is potentials.In arid areaswheregroundwater beingmined (AS consisrecharge Neverin supposition many cases. it tently negative), would be an unreasonable qualified by is the theless, assumption made here for illustrative purposesand sayingthat pastrecordsof waterlevelsin the areahaverevealedan approximate Hydrology is not an exact science,and reaconstancyin groundwaterstorage. are sonablewellfoundedassumptions required if practical problemsare to be solved. Using the simplificationsjust outlined, the working relation reducesto ET:P_R
'I1 1.6 HYDROLOGIC DATA which canbe solveddirectly.First, change into inchesper yearsothat the units R are compatible:
_ ft3
r \ t
1
. ,
sec
sec
area (m nl
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, : 20  0.95 : 19.05in./yr. ET The amountof evapotranspiration the year in questionis estimatedto for be 19.05in. This is admittedlya crudeapproximation could serveasa useful but guide for water resources planning. ll
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 of meterologicdata.Many other federal,state,and local agencies other organizations also compile hydrologicdata. For a completelisting of theseorganizationssee Refs.3 and 17.
'
PROBLEMS
13
r summary
distribution' movethe Hydrology is the scienceof water. It embraces occurrence, an sense, account"nt, urii propertiesof the watersof the earth. In a mathematical of inputs,outputs,and waterStofages a regionsothat a history ing may be madeof the of water movementfor the region can be estimated' the hydrologic After reading this chapter you should be able to understand You shouldalso budgetand make a simpleu".ouniing of water transportin a region' be used to facilitate have gained an undersiandingof trow hydrologic analysescan for processes water resourcessystems' designand management
PROBLEMS
1.1.. Onehalf inch of runoff resultsfrom a stormon a drainagearea
of 50 mi2.Convertthis amount to acrefeetand cubic meters. surfacearea of t.2. Assume you afe dealing with a vertical walled reservoir having a will it take to occurs:How many hours of 1.0 m3/sec 500,000 m' and that anlnflow raise the reservoirlevel bY 30 cm? time is 15 acreft and 1.3. consider that the storageexistingin a river reachat a reference tie reach is 500 cfs and the outflow from the reachis at the sametime the inflow to 650cfs.onehourlater,theinflowis550cfsandtheoutflowis630cfs.Findthe meters' changein storageduring the hour in acrefeetand in cubic walled reservoir was t.4. During a24hr time period, the inflow to a 500acre vertical or fall in the sameinterval, evaporationwas 1 in. was there a rise 100 cfs. During in inchesand centimeters' How muchwasit? Give the answer waterelevation? surface areais 3000 acres' 1.5. The annualevaporationfrom a lake is 50 in. If the lake's surface daity evaporationrate in acrefeetand in centimeters? what would beiire time requiredto raise 1.6. A flow of 10 cfs entersa 1mi2vertical walledreservoir.Find the levelbY 6 in. the reservoir Iftheaverageannualrunoffis5l02cfsand areaof4511mi2. t.7. Adrainagebasinhasan for losses the areain the averalerainfall is 42.5 in.,estimatethe evaportranspiration you think this estimateis? 1 year. Iiow reliable do Determinethe storage in 1.8. The storage a reachof a river is 16.0acreft at a given time. of inflow and outflow during the hour are 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 600acfe the feetper second;(b) determine 3day rate Determinethe average ofrainfau in cubic volumeofrainfallinacrefeet;and(c)determinethe3dayvolumeofrainfallininches of equivalentdepth over the 600acrearea' 100 acreft/day'Deter1.10. The evaporationrate from the surfaceof a 3650acrelake is (feet) in the lake during a 365dayyearifthe inflow to the lake mine the depthchange is25.2cfs.1s the changein lake depth an increaseor a decrease? if acrefeet the drainage to 1.11. One and onehalf inchesof runoff areequivalent how many : 43,560 ft"') areais 25mi2?lNote: I acte rate of how many cubic feet L.12. onehalf inch of rain per day is equivalentto an average many metersper second? p". ,".ona if the areais 500 acrei?How
14
cHAPTER INTRoDUCTIoN 1
REFERENCES
1. P. B. Jones,G. D. Walker,R. W. Harden,and L. L. McDaniels,"The Development the of Scienceof Hydrology," Circ. No. 6003, 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: McGrawHill,1964. 4. O. E. Meinzer,Hydrology,Vol. 9 of Physicsof the Earth.New York: McGrawHlll, 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 JewelpieceDiscovery of the Millennium Bull. 16(6), 10941096(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 9. G. H. L. Hagen, "Ueber die Bewegungdes Wassers engen cylindrischenRohren," gendorf Ann. Phys. Chem.16, 423 442(1839). Pog s V. 10. Henri Darcy, Les fontaines publiques de la ville de Dijon. Paris:. Dalmont, 1856. des 11. J. Dupuit, Etudesthdoriqueset practiques sur le mouvement eaux dans les canauxs ed. dtcouvertset d travers les terrainspermdables,2nd Paris:Dunod, 1863. 12. L. K. Sherman,"Stream Flow from Rainfall by the UnitGraph 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. 519524(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: GlobalProblemwith Local Roots,"Environ. Sci. A 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 estimatingareal precipitation amountsfor specific for storm eventsand for maximum precipitationgenerating 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. many In 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 the amountof rainfall, soil moistureconditions,topography, on 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 very small comparedto quantitiesof is 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 a hydrologist for to be acquainted with waysfor evaluatingthe atmospheric watervapor contentand to understand thermodynamiceffects of atmospheric the 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 overthe United Statescomeslargelyfrom oceanicevaporation, watervaporbeing circulation system. primary atmospheric over the continentby the transporated humidity are relatedbasicallyto condiof Measures watervapor or atmospheric occurring over a level surfaceof pure water. tions of evaporationand condensation Considera ilosed systemcontaining approximatelyequal volumesof water and air If maintainedat the sametemperature. the initial condition of the air is dry, evaporaA tion takesplace and the quantity of water vapor in the air increases. measurement proceeds,pressurein the of pressurein the airspacewill reveal that as evaporation of in of because an increase partial pressure the watervapor (vapor airipaceincreases pressure the overlying air equalsthe of preJsure). Evaporationcontinuesuntil vapor leaving(evaporatof of surfacevapor pressure measure the excess water molecules [a At this point, evaporationceases, ing from) the water surfaceover thosereturning]. is of and if the temperatures the air spaceand water are equal,the airspace saidto be of closed,the equilibriumwould not saturated.If the containerhad beenopeninstead Somecomhavebeenreached,and all the water would eventuallyhaveevaporated. absohumidity are vapor pressure, moistureor of monly usedmeasures atmospheric humidity, and dew point temlute humidity, specifichumidity, mixing ratio, relative Perature.
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
2.2 PRECIPITATION
17
0;7
VCT NJ 0.8
0.9 1.0
Bromsville
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 not sufficient condition for precipitation. Contibut nental air masses usuallyvery dry sothat mostprecipitationis derivedfrom moist are maritime air that originates overthe oceans. North America about50 percentof the In evaporated water is taken up by continental air and movesback againto the sea.
18
2 CHAPTER
PRECIPITATION
Formationof PreciPitation
to are Two processes considered be capableof supportingthe growth of dropletsof (dropletsfrom about 500 to 4000 p'min diameter)to overcomeair sufficient mass fall and consequently to the earth asprecipitation.Theseare known as the resistance process' processand the coalescence ice crystal their process one by which the small cloud dropletsincrease is The c^oalescence with other droplets through collision. Water droplets may be size due to contact to as considered falling bodiesthat are subjected both gravitationaland air resistance at equilibrium (terminal velocities) are proportional to the effects. Fall velocities more quickly squareof the radius of the droplet; thus the larger dropletswill descend ones. As a result, smaller droplets are often overtakenby larger than the smaller droplets,and the resulting collisionstend to unite the drops, producingincreasingly largir particles.Very large drops (order of 7 mm in diameter)break up into smallof processand producesomewhat a chain effect. dropletsthat repeatitre coalescence large raindropsmay be producedto generatesignificant In this *unn"r, sufficiently precipitation. This processis ionsidered to be particularly important in tropical 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 with otherparticles.This is considet mechanism. The artificial inducementof precipitation has been studied extensively,and nuclei supthat condensation thesestudiesare continuing.It has been demonstrated induceprecipitation.The ability of humansto ensurethe producplied to cloudscan location or timing has not yet been iion of precipitation or to control its geographic attained,however' with the prospects problemsare associated Many legal as well as technological ..rainmakiig" processes. interesthereis the impacton hydrologicestimates that Of of partially controlled artificial precipitation might have. Many uncontrolled oi onty as naturally occurringLydrologicvariablesare considered statisticalvariatesthat are or distributedwith a random component.If the distribueither randomty distrlUuted tion or time seiiesof the variablecanbe modeled,an inferenceasto the frequencyof occurrenceof significanthydrologiceventsof a given magnitude(suchas precipitation) can be made.If, however,artificial controls are usedand if the effectsof these may prove to be totally unreliable cannot be reliably predicted,frequencyanalyses tools.
TyPes Precipitation
and of Dynamic or adiabaticcoolingis the primary cause condensation is responsible transport of air massesis a for most rainfall. Thus it can be seen that vertical requirementfor precipitation.Precipitationmay be classifiedaccordingto the condi
'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 typical of the tropicsand is 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 moistureladenair becomesunstable;and pronouncedvertical currents are developed. Dynamic cooling takes place, causing condensation precipitation.Convectiveprecipitationmay be in the form of light and are showersor stormsof extremelyhigh intensity (thunderstorms a typical example). 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 developed usein producingisohyetal of is with Cyclonic Precipitation Cyclonicprecipitation associated the movement differregions.Thesepressure regionsto lowpressure from highpressure 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 lowpressure area. Frontal precipitation results from the lifting of warm air over cold air at the contact zone between air masseshaving colder If are different characteristics. the air masses moving so that warm air replaces 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.
20
CHAPTER PRECIPITATION 2
Thunderstorms
convectivestorms,which are Many areasof the United Statesare subjected severe to generallyidentifiedasthunderstorms of because their electricalnature.Thesestorms, 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 surface heatingor orographiceffects.Thereare threeprimary stages the life history 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 strongupdrafts that often reachaltitudesof over 25,000ft. Vertical by wind speeds upperlevelsare often as greatas 35 mph. As indicatedinFig.2.3a, at there is considerable horizontal inflow of air (entrainment)during the cumulusstage. of This is an important elementin the development the storm, as additional moisture 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 the water dropletsincreaseasthe stageprogresses. duration ofthe cumulusstage is approximately i015 min. and the The strong updrafts and entrainmentsupport increasedcondensation in development waterdropletsand ice crystals.Firrally, whenthe particlesincrease of 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 Downdraft speeds over20 mph are 70 mptrin the early periodsof the maturestage.
El
o o F
(a)
(b)
(c)
cell. (Departstages a thunderstorm of 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 downdraft velocity. Gusty surfacewinds move outwardfrom the region the of rainfall. Heavyprecipitationis often derivedduring this preiod, which is usuallyon the order of 1530 min. In the final or dissipating stage, downdraftbecomes the predominant until all the air within the cell is descending and being dynamically heated.Since the updraft ceases, mechanismfor condensation the 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 79).
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 k> 6
ts
EU)
q . 
$,,\ g( J: qi
F
O r +
! H
5 >
!2
i < F "
. E

,E( ; \ * \ i \
F h i 
,/
'r
"
9999qa9
h
\ IILItrILIL,]
R Y F =
H S
au)
: R
t J
oooo t s
h+6Nio o
ooooo *odio
Ntr
o L ?1 ^
CHAPTER2
PRECIPITATION
 3
'
/'?:;"
E [u Y.s.
"@l5
Ashd 1.56
I o R.H.R. i 3.50
1.55 upl.
^ 1<
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.
25
CHAPTER2
PRECIPITATION
(a/
\
I I
(c)
Figure 2.7 The runoff process:(a) dispositionof precipitation, (b) componentsof infiltration, and (c) dispositionof overland flow supply.
27 PRECIPITATION 2.4 POINT ofhighintensity, shortdo characteristics not filter out (1) fluctuationscharacteristic practices.6 this basis,small On (2) the effectsof land management duration storms;or 100 mi2. A large basin is one in which basinsmay vary from lessthan an acre up to channelstorageeffectively filters out the high frequenciesof imposedprecipitation and effectsof landusepractices.
2.4 POINTPRECIPITATION
at Precipitation eventsare recordedby gauges specificlocations.The resulting data 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 characteristics small urban er other watersheds. detail in Chapter 16. Point rainfall data are usedto deriveintensitydurationfrequencycurvessuch as those shown in Fig. 2.8. Such curves are used in the rational method for urban in storm drainagedesign(Chapter 25); thek constructionis discussed Chapter27 'ln the applyingthe rational method,a rainfall intensityis usedwhich represents average 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).
\
00yr frequency 50yr frequency I 20yr frequency f, 10yrfrequency
7 , 5Yr,frequencl
\ \
t\
B.
a
\ \
\1
tl=.t
120
Duration (min)
curves Figure 2.8 Typical intensitydurationfrequency for Baltimore, Maryland, and vicinity.
28
CHAPTER2
PRECIPITATION
It is occasionallynecessary estimatepoint rainfall at a given location from to recordedvaluesat surroundingsites.This can be doneto completemissingrecordsor precipitation to be used at the point of interest. The to determinea representative National WeatherServicehasdeveloped procedurefor this which hasbeenverified a 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 a weightedaverage the as of Table 2.L The estimated points.The weightsarereciprocalsof the sumsof the squares AX and AY; of other five 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 POINT RAINFALLFROM DATA AT NEARBY OF GAUGES
t1)
Point
(2) Rainfall
(in.)
(3) AX
(4) AY
(5) (D')
(6) wx103
(7) PxWx103
B C D E F Sums
4 1 3 3 7 0
2 6 2 3 2
20 37 1,3 18 8
,TT3
_l
29 PRECIPITATION 2.5 AREAL to precipitation to one with no records.This is considered be the most logical proceresult will alwaysbe lessthan the The dure for handlingthis unusualcase.7 estimated greatestand greaterthan the smallestsurroundingprecipitation. For specialeffects suchas mountain influences,an adjustmentprocedurecan be applied.
2.5 AREALPRECIPITATION
it For most hydrologicanalyses, is important to know the areal distributionof precipitation. Usually, averagedepths for representativeportions of the watershedare determinedand used for this pwpose. The most direct approachis to use the arithare metic average gaugedquantities.This procedureis satisfactoryif gauges uniof is formily distributedand the topography flat. Other commonly usedmethodsare the at method.The reliability of rainfall measured one isohyetalmethodand the Thiessen gaugein representing average depth over a surroundingareais a function of (1) the area,(2) the sizeof the the distancefrom the gaugeto the centerof the representative (e.g.,stormeventversus (4) area,(3) topography, the natureofthe rainfall ofconcern For meanmonthly), and (5) local stormpatterncharacteristics.8 more information on errorsof estimation,the readershouldconsultRefs.7 and 8. Chapter27 alsocontains a discussion areal variability of precipitation. of Figures 2.10 and 2.11 illustrate how the measuredrainfall at a single gauge in with change ( 1) the relativeposition rainfall over a watershed relatesto the average of the gaugein the watershedand (2) the time period over which the averageis In calculated. the first caseit is clearthat the more centralthe gaugelocation,the more area, providing for will match the average a representative closely its observations that areal averages not that the regionis not too large.Figure 2.11 shows, surprisingly,
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
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 ) {
(a)
(b)
average estimatewatershed Figure 2.L1 Errors resultingfrom use of a singlegaugeto Service).(aiWatershedareais 5'45 mi2 and the ConserVation (time period effect, Soil gaugeisontheboundary.(b)Watershedareais5.45Inzandthegaugeisonthe boundarv. to conform more closely over long time periods, in this case one year, may be expected storm event' This suggests to a single guog" uu"ruge than those for an individual with both space and time that the Oerlgnof guuging networks should be tempered considerations.
lsohyetalMethod
are the isohyetal of The two principal methodsfor determiningareal averages rainfall basedon interpolation method and the Thiessenmethod. The isohyetal method is in surveyingand between gauges.It closely resemblesthe calculation of contours to plot the^rain gauge upptng."Tf" first step in developingan isohye?1..up is (Fig' 2'I2)' Next' an locationson a suitablemap and to reJord the rainfall amounts at selectedincreinterpolation betweengaugesis performed and rainfall amounts to thenconnected form mentsare plotted.tdenticaldepthsfrom eachinterpolationare of the weightedaverage is isohyets(lines of equal rainfall depth)' The areafaverage isohyets'The isohyetal depthsbetweenisohyets,that is, the meanvaluebetweenthe precipitation over an method is the most accurateapproachfor determiningaverage attentionto topographic area,but its proper oserequir#a skilled analystand careful the represenand other tactori that impact on areal variability. Figure 2. 13 illustrates tationofamajorstormeventinNorthCarolinabyanisohyetalmap.
ThiessenMethod
In method' this is arealrainfallaverages theThiessen of method calculating Another ascenters' using subareas raingauges into is the procedure area subdivided polygonal Thiessen depth' average the in estimiting watershed as are Thesubareas used weights for suitable is not in as are diagrams constructed shown Fig. i.t+. fnis procedure
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 Figure 2.12 Construction an isohyetal and gauges; (c) plot isohyets' between (b) pdt values; interpolate network is fixed of mountainousareasbecause orographicinfluences'The Thiessen are if any gauges mustbe reconstructed io, u glu"o gaugeconfiguation, and polygons relocated.
AccuracY
locatior\of the Irrespectiveof the methodusedfor estimatingareal precipitation,the tothe point of applicationof the estimate guu* orra in derivingthe estimaterelative may be lbcattiesovertical distances In mustbe takeninto consideration. mountainous are horizontal spacings For gentlelandscapbs, ft" irnpottant than horizontal ones.
32
CHAPTER PRECIPITATION 2
a . !.1 a.l  o !
oo>
I
I
I ti
a ( h w t
o 'rr
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
eH
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.
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.
34
CHAPTER 2
PRECIPITATION
TABLE 2.2 DATA AND THIESSEN POLYGON FOR EXAMPLE 2,1. CALCULATION (1) GaugeNo.
(3)
Precip.in 1.56
(4)
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 computedas 3.45 inchesof rainfall. The use of a is (Table 2.2) facilitates computations and aids in organizing spreadsheet data. lr
) {)
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
$ zoo
E
9?
sr o u 9 9 t
4 4 0 20 10 l0
1oo
20 30 40 50 6Q 70 80
Figure 2.17 Seasonal variation, depthareaduration relations; percentage to be applied to 200 ni224 hr probable maximum precipitation values for August in Zone 6. (U.S. Department of Commerce,National WeatherService.)
intensity.e earlierdesignation "maximumpossible An of 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 may occurin combination that with snowmelt. In both of these cases,annual probable maximums might be less important than seasonal maximums.Figures2.16 and,2.r7 display 24hr pMp for the eastern half of the United Statesfor 200ffi2 watersheds during the month of August (similar figuresare availablefrom the National WeatherService).
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, muchbf the and contentof this book is devotedto the conceptualization individual components of of the various hydrologicprocesses and to synthesizing thesecomponentsinto holistic representations hydrologicevents. of
PROBLEMS 37
r Summary
for precipitation is the sourceof fresh waterreplenishment the planetEarth. Too much In prosperityand disaster. betweenthese canmeanthe differencebetween or too little with a frequency are the normal precipitation eventsthat are experienced extremes position and topographicfeatures' and intensity relatedmainly to geographic 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 valuesofprecipitation for a region shed It shouldbe recognizedihataverage models. uses, some light on the quantiiy of water that might be made availablefor various are of while a i<nowledge tne timeaistribution and timedispositionof precipitation 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 2.3. compute the meanannualprecipitationfor the watershed the following figureusing The the arithmetic mean, the'itri"*"n polygon method, and the isohyetalmethod.
38
CHAPTER2 PRECIPITATION gaugereadings gauges for AK, 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 the figurefor Problem2.3 in using the arithmetic mean and the Thiessen polygonmethod.The gaugereadingsfor gauges AK, 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. noon on August and 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
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. noon on August 11. Would you considerthis to be a period of and 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 cover 4000 acres.The annualrainfall is 37 in., the mean to 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 m3. in 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.)
02
a i
46 68
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, McGrawHill, 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. Papers133. Washington, DC: U.S. GovernmentPrinting Office. Ven Te Chow (ed.), Handbook of Applied Hydrology. New York, McGrawHill,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 6,12,24,and48 Hours,"Hydrometeorological of Rept.No. 33, U.S. WeatherBureau,Washington, D.C., 1967.
Chapter3
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 estimatingthe quantitiesof waterinterceptedand to storedin depressions during precipitation events. Figure 1.3indicatesthe pathsthat precipitatedwatermay take asit reaches earth. the The first encountersare with interceptingsurfacessuch as trees,plants, grass,and structures. Waterin excess interceptioncapacitythen beginsto filI surfacedepresof sions.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 which by the grossprecipitation input becomestransformedinto net precipitation.
3.1 INTERCEPTION
Part ofthe stormprecipitationthat occursisintercepted vegetation otherforms by and ofcover on the drainage area.Interception be defined that segment can as ofthe gross precipitationinput which wets and adheres aboveground to 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 becomepart of the throughfall. The modifying effect to 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 a function of ( 1) the stormcharac (2) the is ter, species, and density age, ofprevailingplantsandtrees,and (3) the season ofthe year. Usually about 1020 percentof the precipitationthat falls during the growing season
41 3.1 INTERCEPTION is interceptedand returned to the hydrologiccycle by evaporation.Water lossesby closedforest standsas much as interceptionare especiallypronouncedunderdense has 25 percentof the total annualprecipitation.Schomaker reportedthat the average is annualinterceptionlossby Douglasfir standsin westernOregonand Washington about24 percent.l A lOyearoldloblolly 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 sprucefirstand, a moderatelywellstockedwhite and gray birch stand,and an improved pasture.r Lull indicatesthat oak or aspenleavesmay retain as much as 100 drops of For tree, interception storageon the order of 0.06 in' of wdtter.z a welldeveloped retentionof about precipitationcould thereforebe expected the basisof an average on CATCH PRECIPITATION OFSTANDARD TABLE WEEKLYAVERAGE 3,1 IN LOCATED RAIN WEATHER BUREAU.WPE GAUGES U,S.
A SPRUCEFIR 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'
Sprucefir 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
Sprucefir 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
10 2
J
t6 t7 0 I6 13 1l 9.5
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 rNrER;;;;rr;.r,o* s
"ro*ou=
20 drops per leaf. For light showers(where gross precipitation P < 0.01 in.) 10t percentinterceptionmight occur, whereas showers for whereP > 0.04 in., losses in the rangeof 1040 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.l6Potentialitorm interception losses be estimated usins2'3,6 can bv 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 horizontal projection:, to of this area E : the amountof waterevaporated hour duringthe precipitationperiod ' per (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 accountfor the rainfall to amountTe
L;:S(1 eP/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. or mm. in It is important to recognizethat forms of vegetationother than trees can aiso interceptlarge quantitiesof water. Grasses, crops,and shrubsoften haveleafareato groundarearatios that are similar to thosefor forests.Table3.2 summarizes some observations that havebeen madeon crops during growing seasons on a variety and 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 longterm models,but for heavy rainfalls during individual storm events,accountingfor interceptionmay be unnecessary. is important for the It 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 INTERCEPTION OF 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 5060.1
s7 l s 7 f 17 l )
31 17
)A
Pdor to harvest
t4t9
"Valuesroundedto nearest percent.Data for table were obtainedfrom Refs.2,4, and 5, 'Grass heightsvary up to about 36 in.
44
CHAPTER3
but Equations3.1 and 3.2 canbe usedto estimatetotal interceptionlosses, for to ofindividual storins,it is necessary dealwith the areal variability detailedanalyses however. are for Generalequations estimatingsuchlosses not available, 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 Using the following equationsdeveloped Horton6for interceptionby ash and oak trees, estimatethe interception loss beneaththesetrees for a storm having a total precipitation 1.5 in. of Solution 1. For ashtrees.
3.2 THROUGHFALL
for A numberof relationships estimatingthroughfall for a variety of foresttypeshave tt been developed.n Deiermining factors for throughfall quantities include canopy 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 tions for throughfall mustincludemeasures canopysurface variables.An example of a throughfall relationship for an easternUnited States hardwoodforest follows.l2 For the growing season Tn: 0.901P 0.O3ln 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
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 considering by 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 P" permits a determinaof tion of V. The mannerin which Vvarieswith time must still be estimated depression if 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 this manner, the total volume would be in 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
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_ekP")
(3.8)
u : (Sokekg#
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:eor.(if)
(3.11)
i  f
(if)*Gf)e."" i  f
(3.r2)
o.125
o.25
0.315
0.50(turf)
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
lPF)tS, " =   e I I
i,t
tl
o b!
B
o E
{
.: 40
9 d J U
o o0
s2
o
6zo t
!)
I
E
50
100
150
200
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  ekP")
i * f
:(i
i  f
 gkP"
(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
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 all depressions full before overlandflow that in the figure: the flrst one assumes with a mean depth of 0.25 in., the figure begins.For a turf area having depressions showsthat for P  F valueslessthan 0.25 in., thereis no overlandflow supply,while for P  F valuesgreater than 0.25 in., the overlandflow supply is equal to i  f . will be greaterthanzero. Tholin For the exponentialmodel (Model 2), c always a relation betweenthosepreviouslymentionedis that and Kiefer haverecommended A of likely more representative fully developedurban areas.15 cumulative normal in and is also described Fig. 3.3 probability curve was selected this representation for (Model 3). are storage deductions usually madefrom the first part of the storm 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 estimatingdepression for to is often considered be negligible.Some studiesof experimentaland other watersheds. 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 and 0.0625in. forpavements.ls valuesof 0.25 in. inpervious urban areas by Viessmanyieldedthe information shownin four small imperviousdrainageareas storageloss is highly correlatedwith slope.This is 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
Time (min)
Figure3.5 Depression storage intensity versus for animpervious time area.(After Turner.l7)
easilyunderstood, sincea given depression hold its maximum volumeif horizonwill tally oriented. Using very limited data from a small, pavedstreetsection, Turner devised curvesshownin Fig. 3.5.17 the Other sources datarelatedto surface of storase are available the literature.2,18,1e in
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 a given foresi standor ground for 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 be negligible. to
PROBI.EMS
3.1. UsingFig. 3.2, estimatethe volume of depression storage a 3acrepaveddrainage for area' Statethe volume in cubic feet and cubic meters.Convert it to equivalentdep;h over the area in in. and cm.
50
CHAPTER3
of 3.2. Estimatethe percentage the total volume of rainfall that is indicatedas depression in storage Fig. 3.2. 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 estimates you havethe most confldence? water budgetbe most affectedby interception? of 3.5. Using Fig. 3.4, estimatethe percentage rainfall that would be lost to depression for storage a l0acre parking lot havinga mean slopeof 1 percent.Repeatfor a slope 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 mm and in. 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. mean depth of depression 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 rainfallrunoff 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: McGrawHill,1964. New York: McGrawHill' 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, 19531954," J. S. Afr. Foresty Assoc'27,7285(1956)' "Rainfall Interception," Monthly WeatherRev.47,603623(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 8. D. M. Gray (ed.),Hand.book the Principles of Hydrology.National Research WaterInformation Center,Inc., 1973. Canada,Port Washington: and J. L. Thames,Hydrology and the 9. K. N. Brooks, P. F. Folliott, H. M. Gregersen, Ames, IA: Iowa StateUniversity Press/Ames,1991. 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 9 7 5 . , "Throughfall in PlantedStandsof Fourth SouthernPines 11. F. A. Roth, II, and M. Chang, Bulletin 17' 880885(1981) in Species EastTexas,"WaterResources
59rs92(r937).
REFERENCES 51 "canopy and Litter Interceptionby Hardwoodsof Eastern J. D. Helvey and J. H. Patric, Resour. Res.l, 193206(1965)' United States,"Water New York: R. K. Linsley, Jr., M. A' Kohler, and J. L. H. Paulhus,Apptied Hydrology' McGrawHill, 1949. "A Linear Model for synthesizingHydrographs Small Drainage for warren viessman,Jr., at paperpresented the FortyeighthAnnual Meetingof the AmericanGeophysical Areas," D.C., APr. 1967. Union, Washington, "The Hydrology of Urban Runoff," Trans. ASCE 125, A. L. Tholin una C. J. Kiefer, ( 1 3 0 8 1 3 7 91 9 6 0 ) . ..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 how they of 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, determines availability,if any,of the precipitation it the 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 surfaceinterface,its rate of downwardmovement the is controlled by the transmissioncharacteristics the underlying soil profile. The of volume of storageavailablebelow ground is also a factor affecting infiltration rates. Considerable research infiltration hastakenplace,but considering infinite on the combinationsof soil and other factors existing in nature, no perfectly quantified generalrelation exists.
4.1 MEASURING INFILTRATION Commonly usedmethodsfor determininginfiltration capacityare hydrographanalyas Infiltrometersareusuallyclassified rainfall simulators sesandinfiltrometer studies. In the former, arlificalrainfall is simulatedover a small test plot or flooding devices. ofrainfall andrunoff, with considerfrom observations and the inhltration calculated storageand surfacedetention.lFlooding infiltrometers are ation given to depression usually rings or iubes insertedin the ground. Water is applied and maintainedat a required' made of the rate of replenishment constantlevel and observations over havethe advantage infiltration basedon hydrographanalyses Estimatesof relating more directly to prevailing conditions of precipitation and infiltrometers of field. However,they areno betterthan the precisionwith which rainfall andrunoff are Of measured. partitular importancein suchstudiesis the areal variability of rainfall. Several meth;ds have been developedand are in use. Reference1 gives a good descriptionof thesemethods.
54
CHAPTER4
INFILTRATION
9 aoo
Ei o
. 300 .
d
2oo
0.1
0.2
0.3
0.4
0.5
conducFigure 4.1 Typical capillary suctionrelative (AfterMein andLarson.e.) tivitymoisture content relation. Capillary conductivity is the volume rate of flow of water through the soil under a gradient of unity (dependent soil moisture content). Relative conductivity is the on 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 wetting front is virtually at the soil surface. the 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 CALCULATION INFILTRATION
55
Moisturecontent.d
I
proflledevelopment a constant with 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 lefthandside 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 contentat the land surface. Figure 4.3 indicatesthat until saturationis reached the surface, infiltration rate is constantand equalto the rainfall applicationrate at the the surface. Point 4, apoint that coriesponds the time at which saturationoccursat At to 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, a given soil type, on the initial moisture content and the rainfall for 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. Fig. 4.1 it can be seenthat the plot of capillary suctionversussoil In 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 drying history of the soil. A numberof approaches the hysteresis 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 farranging. Becauseof this, no simpleinfiltration model can accuratelyportray all the conditionsencountered the fleld. The searchhas thus beenfor modelsthat can in of be called upon to give acceptable estimates the rates at which infiltration occurs 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 continuous in simulationprocesses is affectedeven though runoff doesnot occur. The secondcaseis one in which the rainfall rate exceeds saturated the 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 caseshaverelevance hydrologicmodeling,particularly when it is continuousover 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)
f, 0
,r*"
58
CHAPTER4
INFILTRATION
the infiltration capacity(depth/time) at sometime / inf capacity the a constantrepresenting rate of decrease d final or equilibrium capacity the initial infiltration capacity
It indicatesthat if the rainfall supply exceedsthe infiltration capacity,infiltration in tendsto decrease an exponentialmanner.Although simplein form, difficulties in determininguseful valuesfor/. and fr restrict the useofthis equation.The areaunder the the curve for any time interval represents depth of water infiltrated during that 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 / / versus asshownin Fig. 4.5,we canestimate 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 be multiplied by a suitablecover factor for other types of cover complexes. range 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 inchesuniformly distributedover a watershed in at calculatedby picking valuesof DRH discharge equal time incrementsthrough the 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 24hr period
For most cases differencebetweenthe original rainfall and the direct runoff the can be consideredas infiltrated water. Exceptionsmay occur in areasof excessive drainageor tracts of intensiveinterceptionpotential.The calculatedvalue subsurface as of infiltration can then be assumed distributedaccordingto an equationof the form of Eq. 4.1 or it may be uniformly spreadover the stormperiod. Choiceof the method on employeddepends the accuracyrequirementsand size of the watershed. To circumvent some of the problems associatedwith the use of Horton's Fig. 4.5. Notethat where Consider can infiltration model,someadjustments be made.6 the the infiltration capacitycurve is abovethe hyetograph, actual rate of infiltration
59
o  1
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 l, = U.UUUUJTI . ) 9 ( l
,4 f=0.53+2.4
0691ty
l
e
l
fle
rOa"
9l rras!
o\
r\$4"
iro$dts"
Infiltration capacity curve (/.
10
70 80
100
t20
140
160
t80
Figure 4.6 (a) Typical infiltration curve. (b) Infiltration capacity and mass conditionsof turf areas.fAfter A. L. Tholin and curvesfor normal antecedent "The Hydrology of Urban Runoff," Proc. ASCE J. Sanitary Clint J. Kiefer, Ens. Div. S4(SA2),56 (Mar. 1959).1
60
CHAPTER4
INFILTRATION
Soilgroup
High (sandysoils) Intermediate(loaps, clay, silt) Low (clays,clay loam)
f, (in./hr)
f' (mm/h)
Closegrowingcrops
Row crops
is equal to that of the rainfall intensity, adjustedfor interception,evaporation,and other losses. Consequently, actuafinfiltration is given by the
(4.3)
where/(r) is the actual infiltration into the soil and i(l) is the rainfall intensity.Thus the infiltration rate at any time is equal to the lesserof the infiltration capacity, f,(t) or the rainfall intensity. Commonly,the typical valuesof foandf" are greaterthan the prevailingrainfall intensitiesduring a storm. Thus, when Eq.4.l is solvedforS as a function of time alone, it shpwsa decrease infiltration capacity even when rainfall intensitiesare in much lessthanfo. Accordingly,a reductionin infiltration capacityis maderegardless of the amount of water that entersthe soil. To adjust for this deficiency,the integratedform of Horton's equationmay be used,
: F(tp)
lr'' ,
o, : f,tp +
(t  a' *,,1
(4.4)
whereF is the cumulativeinfiltration at time to, as shownin Fig. 4.7.rnthe figure, it is assumed that the actual infiltration has been equal to$. As previouslynoted, this is not usually the case,and the tnie cumulativeinfiltration must be determined. This
61
0
0
,u, l,' o,
(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 expression solvedfor r' This equation,
F:fJrt (lek'n)
(4.6)
It explicitly for to,but an iterativesolutioncan be obtained. should cannotbe solved time r. Thus that the time to is lessthan or equal to the actual elapsed be understood that given as infiltration capacity shownin Fig. 4.7 is equalto or exceeds the available a described,fbecomes functionof the actual by Eq. 4.1. By makingthe adjustments in amount of water infiltrated and not just a variable with time as is assumed the original Horton equation. In selectinga model for use in inflltration calculations,it is important to know shortcomings; a its limitations. In somecases model can be adjustedto accommodate are if in othercases, its assumptions not realisticfor the natureof the useproposed, 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 When thesemodelsare designed continuvarioushydrolqgicmodelingprocesses. of is ous simulation,the approach to calculatethe appropriatecomponents the hydroof overtime. A discussion how infiltration could 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 the actualamountof infiltration that hasoccurredup to that time, a value on infiltration capacity,fo, availableover the next time stepis calculated of the average using
(4.7)
'v  [f, tt
(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 NewtonRaphson horizontal and When the valueof tp > I6f k, the Horton curve is approximately : f". Once this point hasbeenreached, there is no further needfor iteration since fo on f is constantand equal to f" and no longerdependent F.
EXAMPLE 4.I
ft Givenan initial inflltrationcapacityfiof 3.0 in./hr anda time constant of 0.29hr', if the ultimate infiltration capacity is derive an infiltration capacity vs. time curve 0.55 in./hr. For the first ten hours, estimatethe total volume of water infiltrated in 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")ek, Substitutingthe appropriatevaluesinto the equationyields ,f = 0.55 + (3.0 0.55)so'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,
63
Time (hr) 0.00 0.10 0.25 0.50 1.00 2.00 3.00 4.00
Time (hr) 5.00 6.00 7.00 8.00 9.00 10.00 15.00 20.00
0.58 0.56
r.32
4.1 can To find the volumeof waterinfiltrated duringthe flrst 10hours,Eq' be integratedover the range of 010 v : Y: I J [0.5s + (3.0 0.55)eo2e' ldt
\ ;
l? 6 .:
2.0
\ \
15
1.0
"'0 0.5
L
8
10
12
t4
18
64 4.4
CHAPTER4
INFILTRATION
MODEL GREENAMPT
in The GreenAmpt infiltration model,originallyproposed 1911,has had a resurgence inte1gs1.3'61t approach based Darcy's law (seeChapter18).In its on This is of 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 GreenAmpt model to a steadyrainfall input.eThey the alsodeveloped procedurefor determining valueof the capillary suctionparamea the ter usedin the model. In 1978,Chu demonstrated applicability of the model for useunder conditionsof unsteadyrainfall.lo As a result of theseand other efforts, the GreenAmpt 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 with a uniform initial watercontent(seeFig. 4.9). Wateris assumed soil 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
I
Ho
I
I
I I
Figure 4.9 Definition sketch for GreenAmpt model.
65
f   , \ r p : K.(,
t * jt' )
F /
(4.11)
#: *,(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 This form of the GreenAmpt equationis more convenient usein watershed infiltrationto the the cumulative processes than Eq. 4.l}beciuse it relates modeling a of this equation assumes ponded time at which infiltration began.The derivation equal to the infiltration capacityat all surfaceso that the actualrate of infiltration is a the times.Using Eq. 4.13, we can determine cumulativeinfiltration at any time, in All the parameters the equation modeling. featuredesiiablefor continuoussystems The determinaare physicalpropertiesof the soilwater systemand are measurable. s is often difficult, however,particution of suitablevaluesfor the capillary suction be larly for relationssuchas that ihown for a claytype soil in Fig. 4.10. It can there is a wide variation of capillary observedfrom the figure that for this curve content.3 with soil moisture suction The MeinLarson formulation using the GreenAmpt model incorporatestwo The stages.3,6 first stagedealswith prediction of the volume of water that infiltrates before the surfacebecomessaturated.The secondstageis one in which infiltration usingthe GreenAmpt equation.In the widely usedstormwater capacityis calculated of the ,nunug"*"nt model, the irodif,ed GreenAmpt model of infiltration is one Computationsare madeusing optiois that can be employedto estimateinfiltration.6
v)
Moisture content,0
66
CHAPTER4
INFILTRATION
F. :7:
IM?
i/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 the surface saturates calculations and {, for infiltrationthenproceed usingEq.4.I4. Notethat by substituting/for in Eq.4.l4 i and rearranging, equationtakesthe sameform as Eq. 4.11. the 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 achieved the surface,Bq.4.l1 showsthat the infiltration is at 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 GreenAmpt 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 NewtonRaphsonroutine 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
67
MODEL 4.5 HUGGINSMONKE by problem introducing the have investigators circumvented timedependency Several by proposed equation the following as soilmoisture the depenO"niuutluUfe.2'1013 and Huggins Monkeis an examPle:2
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 process'At the in the iteration for variableF mustbe catculated eachtime increment
s  itr"
exceedsthe inflltration capacity,the rate zone," which determinesthe soil moistu (1) as evaluated follows:10 wherethe moist the field capacity(amountof waterheld in z drained),the drainagerate is considered soil is s to the infiltration rate when the constan and (3) if the watercontentis be drainagerate is comPutedas
(4.r7)
pore volume where P, : the unsaturated G:maximumgravitationalwater,thatis,thetotalporosityminusthefield caPacitY Datafromsprinklinginfiltrometerstudiesofvariouswatershedsofinterestare in the usedto estimate coefficients Eq' 4'16''
5.6
EVAPOTRANSPIRATION
101
of a masstransferequationthat has often been employedfor this purpose.However, Linsley and coworkers indicate that there is some question as to the adequate The losses.27 equation is 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. Hg) vaporpressures (mph) wind speeds the mean temperature("F) of the layer betweenthe lower level zt and the upper level z2
is in It is assumed Eq. 5.25 that the atmosphere adiabatic and the wind speedand moistureare distributedlogarithmically in a vertical direction. In view of the sm4ll to differencesbetweenwind and vapor pressure be expectedat two levelsso closely evaporation, and sincethesegradientsare directly relatedto the soughtafter 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 ClassA pan records.This is not theoreticallycorrect because albedo(amountof of incoming radiation reflectedback to the atmosphere) vegetatedareas and soils should be As rangesas high as 45 percent.28 a result, potential evapotranspiration somewhatlessthan free water surfaceevaporation.Errors in estimatingfree water from pan recordsare such,however,asto make an adjustmentfor evapotranspiration value. potential evapotranspiration questionable of by developed the AgriAn 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 , " ( # l x
where
(s.26)
potential (in./day) ET : evapotranspiration : growth index of crop in percentage maturity of GI usually 1.0I.2 for short grasses' K : ratio of G1 to pan evaporation, height,and 1'62.0 for forest 1,.2L6 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
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 firstyear 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
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
z5J
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 41172' Sept' 1970' Asiicultural'Watersheds," Aiter H. N. Holtan et a1.2e
The GI curveshave been develoPed for evapotranspiration severalcrops (Fig. : (Fig. 5.6).Equation5.26is used daily rate hydrolq USDAHL74 model of watershed Represental evapotranspiration. late daily Table5.5.
104
The PenmanMethod
Both the energybudgetand masstransportmethodsfor estimatingevapotranspiration (ET)have limitations dueto the difficulties encountered estimatingparameters in and in making other required assumptions. circumventsomeof theseproblems,penTo man developed method to combinethe masstransport and energybudgettheories. a This widely usedmethodis one of the more reliableapproaches estimatingETrates to 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 net radiation) (mm/day) of E : daily evapoiation(mm) ET : the evapotranspiration consumptive for a given period (mm/day) or use The variablesE and.Fl are calculatedusing the following equations:
( s .28) 0.0098ar) : the saturationvapor pressureat mean ak temperature(mm Hg) where eo e6 : the saturationvapor pressure meandew point (actualvapor pressure at 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 0.905) (5.29) + where R : the mean monthly extraterrestrialradiation (mm HrO evaporated per dav) : the estimatedpercentage reflecting surface of B = a temperaturedependent 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 vaporpressures, of sunshine duration,net radiation,wind speed, and mean temperature. Unfortunately,regular measurements theseparameters often unof are availableat 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
"C 50 40
r04 86
68
313 =
ts
: o
o B t ^
14 1 0
303
ro? z6J   0.8  I I I t273
ct
!d 6 6 F.
+ o
o
o F
t r 1 n
20 10
s0l. zzV 
0.2
0.4 0 6
of Value A (mmHg/'F) Figure 5.8 TemperatureversusA relation for use *i?rt ,n" PenmanLquation. (After Criddle'23)
that using to thereis evidence supportanothervalue'it appears Thus,unless unity.3234 satisfactoryresults for surfaces a value of 1 for the reduction coefficient may give of free water evaporation having varied vegetal covers.Accordingly, aty estimate by an appropriatereduction could be used ro estimaieEZ, providin[ it is modified coefficient.
EXAMPLE 5.4
ET, giventhe following data: ': using the PenmanMethod, Eqs.5.21 to 5.29, estimate : 30 degrees c' c, 20 degrees temperatureof 1l temperatureat warer r"tru"" month is : (48 mi/day)' the relative humidity : +O p"r""nt, windvelocity i mph r is given ut 0'07' and S is found to be 0'75' north, 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'": : 31'83 X 0'4 12'73' a relativehumidity of 40 percent,e,t Then,usingEq. 5.28' + .83 E : 0.35(31  12.73)(1 0.0098x 48) 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, + H: 16'5(I  0'07)(0'18 0's5 x 0'75)  17.01(0.5 0.092x 12j30)(0.10 + 0'90 x 0'75) 6 = 6.04 mm/daY H
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 tFO\dcl\fh!nv1<.cqolO
F
f U)
J
tr
oN catf,r +s+coNooo\o
: : * * * d d i * 
nv?qnnqqnv.tnv.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.l9qqvln
>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
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
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
3. UsingEq. 5.27, Er : 0.0 x 6.04 + 0.27 x 9.83)/(1+ 0.27) ET : 6.85 mm/day is Thus the estimated evapotranspiration 6.85 mm/day. ll
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 estimates especiallyimportant. They generally to A numberof approaches estimatingEThavebeendeveloped. analyttheoretical,basedon the physicsofthe process; fall into the following classes: Equaical, basedon energyor water budgets;and empirical, basedon observations. are tions usedin making ET calculations usually of the type illustrated by Eqs. 5.1, 5.8,5.10,5.19,5.22,and5.26.
1oB
PROBLEMS
5.1. An 8000mi2watershed received20 in. of precipitationin a 1yearperiod. 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 of 1 in. The total monthly precipitaloss tion 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:
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 5day period for a lake with a 250aqe 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 the river at the of basinoutlet is measured an average 650 cfs. Assumingthat the change storage at of in for the system essentiallyzero, estimatethe EZlossesfor the areain inchesand cm is 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 the if meanmonthly temperatures May, June,July, and August are 62,71,16, and 75'F for 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: 19752000,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. "WaterLossInvestigations,"Vol. 1, Lake Hefner Studies,U.S. GeologicSurveyProfessionalPaperNo. 269 (1954). (Reprintof U.S. Geological SurveyCitc.229, 1952.) 5 . N. N. Gunaji, "EvaporationInvestigations ElephantButte Reservoir,New Mexico," ^lnl. at Assoc.Sci.Hydrol. Pub. 18, 308325(1968). 6. I. S. Bowen, "The Ratio of Heat Lossesby Conduction and by Evaporationfrom Any Water Surface,"Phys.Rev. 27, 7797 81(1926). 7 . E. R. Anderson,L. J. Anderson,andJ. J. Marciano, "A Reviewof Evaporation Theory and Development Instrumentation,'o of 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," Meteor,Soc.Q. "I. 75(No. 236),335350(Oct. 1949). R. 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 Commerce, of 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: McGrawHill, 1982. 1 5 . H. L. Penman,"Natural Evaporationfrom Open Water,Bare Soil, and Grass,"Proc. R. Soc.LondonSer.A 193(1032),120145(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 Evaporation for Suppression," Eng.(June10, 1963). Chem. 1 9 . D. R. Maidment(ed.),Handbookof Hydrology. New York: McGrawHill ,1993. 20. H. F. Blaney,"Water and Our Crops," inWater, the Yearbook Agriculture. Washington, of D.C.: U.S. Department Agriculture,1955. of
110
AND 5 CHAPTER EVAPORATION TRANSPIRATION "Monthly Consumptive Use Requirementsfor Irrigated Crops," Proc. ZI. H. F. Blaney, J. ASCE, Irrigation DrainageDiv. 85(IR1), 112(Mar' 1959)' 22. yen Te Chow (ed.), Handbook of Applied Hydrology. New York: McGrawHill ' 1964. "Methods of Computing ConsumptiveUse of Water." Proc. ASCE J. 23. W. D; Criddle, Irrigation DrainageDiv. 84(IR1), 127(Jan' 1958)' ,;Estimating Evaporation," Trans. Am. Geoplrys. union 37(l), 4324. H. L. penman, "Reductionof Transpiration," Geophy. Res.66(10), 33093312(Oct. J. 25. N. J. Roberts, 1961). "Research control of Phreatophytes," Proc. Ninth Annual water conon 26. E. H. Hughes, ference,New Mexico StateUniversity,Las Cruces,Mat. 1964' New York: Hydrology Engineers. for 21. R. K. Linsley,Jr., M. A. Kohler, and J. L. H. Paulhus, McGrawHill, 1958. "National WeatherServiceRiver ForecastSysLaboratory, 28. Staff, Hydrologic Research tem ForecastProcedures,"NWS HYDRO 14. U.S. Departmentof Commerce,Washington. D.C..Dec. 1972. "usDAHL74 Revised 29. H. N. Holtan, G. J. Stiltner,w. H. Henson,and N. C. Lopez, ARS Modelof WatershedHydrology," Tech.Bull. No. 1518.U.S.DepartmentofAgriculD.C., 1975. ture, Washington, "EvapotranspirationReview of ReH. N. J. Rosenberg, E. Hart, and K. w. Brown, Lincoln, 1968' search,"MP 20, Agricultural ExperimentStation,University of Nebraska, R. H. McCuen, HydrologicAnalysis and Design.Englewoodcliffs, New Jersey:Prentice 31. Hall, 1989. "Correlation of ClimatologicalData with WaterR'equire3 2 . W. O. Pruitt and F. J. Lourence, University of California Water ScienceEng. Paper9001. Davis, June ments of Crops." 1968. "Potential Evaporation:The cornbination concept and Its Experi33. C. H. M. Van Bavel, Res.2' 455467(1966)' mental Verification," WaterResources "Discussionof Paperby H. L. Penman,'Estimating Evaporation,"' Trans' 34. H. F. Blaney, Union37,4648(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, EPA6001284109a(NTIS PB84198423).Cincinnati, oH: EnvironmentalProtection Agency, Nov. 1981. Model User's R. 37. L.A. Roesner, P. Shubinski,and J. A. Aldrich, StormWaterManagement (NTIS PB84198431). I, Manual, versionIII: Addendum Extran, EPA600/284109b Cincinnati, OH: EnvironmentalProtection Agency,Nov' 1981'
5o(1es6).
6 Chapter
Streamflow
r Prologue
The purposeof this chapteris to: . Introduce the conceptof streamflow. . Describethe characteristics a hydrograph. of ' Presentapproaches measuringstreamflow. to 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 supply,and management. knowledge development, waterresources with concerned quantity and quality of streamflowsis a requisite for: municipal, industrial, of the flood control; reservoir designand agriculiural, and other water supply endeavors; recreation;navigation;fish hydroelectricpower generation;waterbased operation; management;drainage; the managementof natural systemssuch as und .ildlit" treatment. wetlands;and water and wastewater by is generated precipitationduring stormeventsand by groundwaStreamflow ter entering surface channels.During dry periods, streamflowsare sustainedby reservoirsare below streamchannelsWhere groundwater groundwater discharges. to in arid regionsstreams cease flow during protractedprecipitationoften the case periods. Relations between precipitation and streamflow are complex, being free influJnced by the factors discussedin the foregoing chapters.As a result, many to approaches relating theseimportant hydrologicvariableshavebeen developed.l3 in Severalof them are discussed detail in Part Three of the text' of streamflow are based on the use of flowmeasuring Field measurements of devicessuchas weirs and flumes and on the measurement channelcrosssections velocities(seeChapter 8). along with streamflow
BASINEFFECTS DRAINAGE
in The quality and quantity of streamflowgenerated a drainagebasin are affectedby Accordingly,it is important and climatic features.ae physical,vegetative, the bisin's ofthe soils,rocks,plants,topography, havea good understanding hydrologist that the
J12
CHAPTER STREAMFLOW 6 of that landusepatterns,and otherbasincharacteristics influencethe sequence events precipitation and runoff. It should be pointed out, however,that while separating natural basin featuresare very important elementsin the runoff process,landuse parking lots, agricultural ieatures createdby humans (e.g., housingdevelopments, practicescan patterns)may,in somecases, the dominantones.Land management be suchasin retardingerosion,and they can alsobe detrimentalwhenthey be beneficial, In natural hydrologicprocesses. Chapter10,the principal basin function to accelerate of concernto the hydrologistare discussed. characteristics
Stormperiod hydrograph
\%
. Continuous hydrograph
6.4
113
' bI) 6 6
o bo ri
3000
Discharge (secft)
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 crosssectional is discharge normally estimatedby measuring devicesare Where flowmeasuring into areato translatethis measurement discharge. rating curve to translatethis into used,it is customaryto observethe head and use a calculations cannot be made, discharge When direct flow measurements discharge. chemical tracers, electrical are often facilitated by use of velocityarea relations, 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 gauging siie and this record is convertedinto discharge one or more of several for this purpose(seeChap"tttoAr. Rating curves,tables, and formulas are used rating curve' The instrumentsand ter 11).Figure 6.2 rho*r a typical stagedischarge to mustbe carefully adapted the to recordings discharge methodsusedto convertstage gauging site so as to ensurethe natural or artiflcial conditions encounteredat the reliability of conversions.'
115
and thesecan be translatedinto velocity by conductivity changes salt injection sense keeping track of time. Electrical methods knowing the spacingof electrodesand oxygen voltagegeneration, electromagnetic includethe useof hot wire anemometers, Streamflowconditionsvary widely and thus no polarography, waves.a and supersonic specificapproachto velocity determinationis universally suitable.The best method 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 by sectionis closelyapproximated the average the velocitiesoccurringat depthsof (seeFig.6.3).11'12 percentof the total sectiondepthrespectively 20percentand 80 on the orderof 0.5 feet, singlemeterreadingsat about Where depthsarevery shallow, the 50 percentpoint havebeen shownto yield good results.t' Velocity measurements permit estimatingchannel and geometricdefinitionsof streamchannelcrosssections havebeen made. velocity and depth measurements flows at locationswhere
116
CHAPTER6 STREAMFLOW
Water surface
6.1 for crosssection Example Figure 6.4 Channel seriesof geometricshapes(seeFig. 6.4). AL eachvertical location along the crossvelocity of The section,the meanvelocity is estimatedfrom measurements. average of to areabetweentwo verticalsis considered be equalto the average the flow for the betweentwo vertifor eachof the borderingverticals.The discharge meanvelocities velocity for the sectionmultiplied by the areaof the section. cals is thus the average total flow for the are discharges then summedto provide an estimated The individual to at that location. Note that it is important to have enoughmeasurements channel is The the crosssection. procedure illustratedin Example6.1. characterize
EXAMPLE 6.1
at Calculatethe discharge the sectiongiven in Fig. 6.4. Data from field observations are shownin Tables6.1 and6.2.
TABLE 6.1 DATA FOR EXAMPLE6.1
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 THESLOPEAREA 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
A2 A3
A4
A5 A6 A7
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 summedto yield an estimated376.13 cfs of are flow being deliveredby the full channelwidth. r I
(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 highwater period. For a of more completediscussion this and other streamflowdeterminationmethods,the
118
CHAPTER6
STREAMFLOW
4 references the end ofthe chaptershouldbe consulted.References and 10 give an at excellent overview of techniques and include a valuable list of references on streamflow.
r Summary
Streamflowis the result of stormperiodprecipitation, snowmelt,and groundwater discharge.l3 is a primary sourceof water for a host of instreamand offstreamuses. It The graphicalrepresentation streamflowis the hydrograph,a plot of flow versus of time at a prescribedlocation alongthe water courseof interest.As illustratedby Fig. is 1.3, the end product of many hydrologicmodelingprocesses a hydrographwhich is 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 RaquetteRiver (Fig. 6.2). What would you estimatethe discharge be in cfs and in be? m3/sec? the gaugeheighthad been9 ft, what would the discharge Which of the If 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 the sectiongiven in Fig. 6.4 if the depth measurements at 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 the sectiongiven in Fig. 6.4 if the velocitieswere: 0, 2.3, at and 2.6,3.1,2.9,2.7,2.5, and0 fps respective$, the depthsof Problem6.5 applied. 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: McGrawHill, 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, "GeologyPart II," Handbook of Applied Hydrology. New York: McGrawHill. 1964. 8 . R. E. Horton, "Drainage Basin Characteristics,"Trans.Am. Geoplrys.Union l3r 35O36r(1932\. 9 . W. B. Langbeinet al., "TopographicCharacteristics DrainageBasins,"U.S. Geological of Survey,Water Supply Paper,968c, 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:PrenticeHall, Inc., 1951. 12. C. H. Pierce,"Investigationof Methodsand EquipmentUsedin StreamGauging,"Water Supply Paper 868A, U.S. GeologicalSurvey,Washington,D.C.: GovernmentPrinting Office,1941. 1 3 . D. R. Maidment (ed.),Handbookof Hydrology.New York: McGrawHill , 1993.
PART TWO
Chapter7
DataSources Hydrologic
r Prologue
The purposeof this chaPteris to: . Describethe principal sourcesof data for hydrologicinvestigations. and modeling. forecasting, Data on hydrologicvariablesare fundamentalto analyses, publications of state and federal agencies, Such data *uy 1" found in numerous Severalofthe most significant and otheroganizations. universities, institutes, research in this chapter'13 briefly sourcesof hydrologicdata are described
DATA 7 . 1 GENERALCLIMATOLOGICAL
solarradiation,wind, and of The most readily availablesources data on temperature, publishedby the EnvironrnentalData humidity are ilimatological Data bulletins Service of the Nationai Oceanic and Atmospheric Administration (NOAA), and by Monthly Summaryof Solar RadiationData publtshed the National Climatic Data in cooperationwith the World MeteorologCenter.The EnvironmentalData Service, ical Organization(WMO), also publishesMonthly Climatic Data for the World A 1968 publication of the Environmental SciencesService Administration, entitled humidity, evapowind, temperature, Climalic Atlasof the UnitedStates,summarizes seriesof maps. In additionto these ration, precipitation, and solar radiation on a and agriculfederal sourcesof data, stateenvironmental,geologic,water resources, universitiesalso publish a variety of shouldbe consulted.Most state tural agencies hydrologicdata through their researchcentersand extensionprograms.
E # = ,
124
CHAPTER7
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 with recordingauges. stationlocationmap is alsoincluded.This stationsequipped publication is availablefrom the EnvironmentalData Service.Another publication, periods.Data are regionsfor 10year by is Records, issued geographic World Weather listed by country or areaname, station name,latitude and longitude,and elevation. pressure, and tempersealevel 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 sitespecificcontext.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.
REFERENCES 125
r summary
Numerous in Climatic and other data are keystones hydrologicmodelingprocesses. to sources of data exist and may be accessed support model developmentand verification, statisticalanalyses, and specialstudies.
PROBLEM
7.1 Developa list ofdata sources your stateorlocality by visiting the library or through in 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 HYDRO22, National Oceanic and Atmospheric Administration, Washington, D.C., May 1975. , D. R. Maidment (ed.), Handbookof fudrology. New York: McGrawHill, 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 frequency,and in an appropriateform if they are to be of value. tity, with adequate are A variety of instruments usedto obtain and transmitthe data.They are the subject of this chapter.
8.1 INTRODUCTION
problemanalyses, research, supportsarealinvestigations, Hydrologicinstrumentation are A planning,and environmentalpolicymakingand analysis. hostof measurements 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 instrumentation, ratherfrom using of utility, not somuchbecause lack of adequate availableinstrumentsto measurethe wrong thing or in too limiting a fashion.Today it is widely recognizedthat it is important not only to selectappropriateinstruments of but to selectthem in the contextof datanetworksthat meetthe needs moderntimes. More will be said about this in Chapter9, In Section8.2, instrumentsfor measuring hydrologicvariablesand waysin which they can be usedjointly to createa complete of representation a functioning hydrologicsystemare discussed.
'
Precipitation
for Gauges measuringrainfall and snowfall may be recording or nonrecording.The 8in. gauge. Servicestandard gauge the U.S. Weather is mostcommonnonrecording interval but often this is daily. The gaugeis The gaugemay be read at any desirable calibrated so that a measuringstick, when inserted, showsthe equivalentrainfall depth. Such gaugesare useful when only periodic volumes are required, but they cannotbe usedto indicate the time distribution of rainfall' the continuouslysense ratb of rainfall and its time of occurRecordinggauges type or the tipping are rence.Thesegauges usually either ofthe weighingrecording gauges run for a period of 1 week, at which time usually buckettype. Weighingtype with Problem2.5 is typical of the The figure associated their chartsmustbe changed. depth versustime is the product, and this recordedoutput. A masscurve of rainfall curve can be translatedinto an intensitytime graph by calculating the ratios of rainfall to time for whatevertime stepis desired.Tipping bucketgauges, 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 located below the rainfall entry port of the gauge. When it fills to the 0.01in. bucketinto position.Thesetwo small buckets incrementit tips over,bringing a second are placed on a swivel and the bucketstip back and forth as they fiIl. Each time a an bucketspills it produces indication on a strip chart or other recordingform. In this way a record of rainfall depth versustime (intensity)is the outcome.For rain gauges to record snow accumulations,some modifications must be made. Usually these involveproviding a melting agentso that the snow can be convertedinto measurable water. Figure 8.la is the diagram of a selfreportingrain gaugingstation.The tipping a bucketmechanismgenerates digital input signal whenever1 mm of rainfall drains 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 periods can be calculated accordingly' Figreceived 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 and is measured the tipping bucket,which operates mechanism compensating in Fig. 8.1 can easilybe incorporated of station's transmitter.Gauges the type shown that can be used in a variety of forecastingand into realtime monitoring systems operatingmodes.
"
128
CHAPTER INSTRUMENTATION 8
Antenna
Antenna mast
Antenna
Antenna mast
Temperatffe compensation overflow mechanism Glycometh collecting section Tipping bucket Drain holes (4)
Funnel assembly
Tipping bucket
(a)
(b)
Figure 8.1 Selfreporting (a) rain and (b) snow stations. (Courtesy of SierraMisco, Inc., EnvironmentalProducts,Berkeley,CA.)
..
129
but variety of pan types havebeen developed the U.S. WeatherBureauClassA pan havebeenusedto Pan is the standardin the United States.4 evaporationobservations from wellwatered 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 gauges, datacan be recorded goodwatersupply.a in the caseof precipitation As and transmittedcontinuouslyto a central receiving station. are measurements often made using lysimeters.ThesedeEvapotranspiration vices are containersplaced in the field and filled with soil, on which sometype of growth is maintained.The object is to study soilwaterplant relationsin vegetative a natural surrounding.The main featureof a weighinglysimeteris a block of undisin turbed soil, usually weighing about 50 tons, encased a steel shell that is 10 ft by 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 changes soil water level. The scalesare set to counterbalance 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 longterm meaevents.s precipitation suchas between surements
130
CHAPTER INSTRUMENTATIoN 8
Temperaturemeasurements madeusing standardthermometers various are of types,while hurnidity is measured using a psychrometer. psychrometer A consistsof two thermometers, one called a wet bulb, the other a dry bulb. Upon ventilation the thermometers measuredifferently, and this differenceis called the wetbulb depression.By usingappropriate tables,dewpoint, vaporpressure, relativehumidity can and , 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
Main housing
Signal cable
Ground level
Transmitter
Figure 8.3 Selfreporting weather station. (Courtesy of SierraMisco, 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 forecasting,quantitativedeterminations oncoming storms,determinationof wind effect on tidal areas,and establishinga data basefor irrigation.
OpenChannelFlow
of Measurements open channel(natural and created)flow are made using standard measuringdevicessuchas flumes and weirs, and they are also madeby calibrating of special control sectionsalong rivers and streamssuchthat measurements depth devicesare designedso Flowmeasuring (Jtage)of flow can be related to discharge. that sensingsomeparametersuchas depth automaticallytranslatesthe observation of When a control sectionis used,observations crossinto units of flow (discharge). flow velocitiesmust and average sectionalareafor variousdepthsmust be obtained, In so for be ascertained variousstages that a sectionrating curve can be established. U.S. Bureau of Reclamation,the the United States,the U.S. GeologicalSurvey,the havedoneextensive Service,andthe U.S. Army Corpsof Engineers Soil Conservation instrumentsand proceduresfor flow measuringand havebeen active in developing rates of flow.2'6 ascertaining devices.When they are properly inWeirs Weirs are common watermeasuring simple and accuratemeansfor gauging stalled and maintainedthey can be a very The most often usedweir types are the rectangularweir and the Vnotch discharge. weir (Flg. 8.4). To be effective,weirs usually require a fall of about0.5 ft or more in the channelin which they are placed.Basically,a weir is an overflow structureplaced acrossan open channel.For a weir of specificsize and shapewith freeflow steadystateconditionsand a proper weirtopool relation, only one depth of watercan exist by Flow rate is determined measuring in the upstreampool for a particular discharge. part of the weir to the watersurface from the crestof the overflow 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 8.5 depictsone of thesedevices. that can be installedin a channelsection.Figure (1) it can operatewith a relatively small head flume has severalmajor advantages: to loss;(2) it is fairly insensitive the approachvelocity; (3) it can be usedevenundet conditions;and (4) its flow velocity is usually sufficientto precludesedisubmerged by The Parshallflume was developed the late Ralph ment depositsin the structure.2 flume. The constrictedthroat of the L. Parshill and it is a particular form of venturi Thus,asin the case a flume produces differential headthat canbe relatedto discharge. the (head)is all that is requiredto determine rate ofdepth .ofthe weir, an observation are generally best suitedto gauging of flow at the control point. Weirs and flumes weirs can be installed althoughlargebroadcrested and openchannels, small streams For major rivers, other measuringapat dam sites as part of overflow structures. proachessuchas developingfield ratings at a specifiedcontrol sectionmust be relied 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 depth,crosssectional of 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 depthsensing 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
t x t xf,tngle
can area,discharge be calculatedas the product of meanvelocity the crosssectional can area.If observations be madefor a rangeof depths,a rating and crosssectional of developed the control sectionsothat only measurements depthwill for curvecanbe to estimaterate of flow at somelater time. Additional information on this be needed proceduremay be found in Refs.2 and 6.
Figure 8.6 Recorderhouseand stilling well for a streamgaugingstation. (U.S. Bureau of Reclamation.)
134
CHAPTER INSTRUMENTATIoN 8
Depth (stage) Measurements Most depth measurements made using a float are and cable arrangement a stilling well or a bubbler gauge.In the first insiance,a in 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 selfreporting stilling well liquidlevel station. Data from this station can be transmitted to un! central location for analysisand/orother use.A bubblertypeinstallation 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
Transmitter
Hydraulic damping device Level sensor Counter weight Mounting brackets
lnlet tubes
Figure 8.7 Selfreportingstilling well liquidlevelstation. (Courtesy of SierraMisco,Inc., Environmental products. Berkeley,CA.)
8.4 REMOTE SENSING 135 corresponding pressure change. This resultsin a fluidlevelchangein the manometer connected the gassupplyand this in turn is usedto reflect stagevariationover time. to 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.
136
CHAPTER INSTRUMENTATIoN 8
I
L. I
Utah
Anzona
q)
B .H
20
40
60
"&*" rq*
s
Taylor'sFerry gauge
,&
Scale miles of
Lake Havasu
Cibola gauge
Colorado nver
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
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 s> ,, Y q F .t
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!!
o g o 4 =
O F , Y o E
I E t
Ee
L
+e
d
8:
at)
142
CHAPTER INSTRUMENTATIoN 8 With the advancement satellite technology,the use of satellitesas remote of platformshasspread.Currently available sensor can sensors operatein a multitude of 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 improved to permit the attainmentof are greater radiometric and geographicresolution, and as computer imageenhancing techniques becomemore sophisticated, is certain that this powerful watermanageit ment 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),812(May/June 1987). 6. R. K. Linsley, Jr., M. A. Kohler, and J. L. H. Paulhus,Applied Hydrology. NeW York: McGrawHill, 1982. 7. R. J. C. Burnashand T. M. Twedt, "EventReporting Instrumentationfor Real:TimeFlash Flood Warning," American Meteorological Society, Preprints, Conference on Flash Floods:Hydrometeorological Aspects, May 1978. 8. D. E. Colton and R. J. C. Burnash,"A FlashFloodWarningSystem,"American MeteoAspects, rologicalSociety,Preprints,Conference Flash Floods:Hydrometeorological on 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 RainfallRunoff Model," International Symposium on RainfallRunoff Modeling,Universityof Mississippi, May 1821, 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, June2426, 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 ThirtySecond in Needsfor,tgricuttrire, Forestry,and Natiiral iesources,o' Proceedings, Institute, Agiicultural R'searcih'Institute. Bethesda,MD: Agricultural Research tuteeting, 1 9 8 3p p . 7 5  8 5 . , *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' 915' MD: American Water Resources eds). Bethesda, 16. G.K.Moore,..AnlntroductiontosatelliteHydrology,''hsate.tliteHydrology(M. MD: American WaterResqurces Deutsch,D. R. Wiesnet,and A. Rango,eds.).Bethesda, t74L. 1981,PP. Association,
ChapterI
Networks Monitoring
Prologue
The purposeof this chapteris to: . Describeelementsof systems monitoring hydrologicvariables. for . Indicate the importanceof realtime and continuousrecording of hydrologic events. for Information (data)is the requisitefoundationfor designing schemes manipulating (managing) hydrologicsystems, evaluatingthe efficacyof actionstaken to correct for 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 how these systerns of 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 the stateof the system of lies beingoperated. The key to meetingsuchrequirements in the productsof carefully designedand managedmonitoring networks.Developingsuchnetworksis no small task, however,as the numberof variablesthat must be observedmay be very large, and the instruments measure to them costlyto install and operate,andthe datastorage managementrequirementsextensive.Accordingly, a monitoring network's design must beginwith a thoroughunderstanding its purposeso that the degreeof resoluof for but tion providedby its observations adequate, not excessive, the task at hand. is of within the constraints what A goodrule is to keepthe networkassimpleaspossible, must be accomplished.
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 the following obje"by tives:planning,realtime system operating,enforcingregulatoryprograms,and environmental policymaking. In the flnal analysis, ultimate purposeof monitoring is the to enhancedecisionmaking, whetherit be for development, management, regulation, or research aims.
Timeand SpaceVariability
In general, monitoring networks are designedto have both spatial and temporal dimensions. Although monitoringa specificpoint locationmay be all that is necessary undersomecircumstances, is more commonthat what is happening a regional it in settingis of importance.The temporal aspectis similar. While a snapshotat some point in time may suffice for some purposes, time varianceof conditions to be the tracked is usually critical for effective analysesandlor decisionmaking.Both the shorttermand longterm variabilities of many targefsof monitoring must be ascertained. For example,water quality in a streamcan changerapidly with time, while changes lake levels,suchas thoseexperienced the GreatLakes in the 1980s,are in in the result of longtermhydrologicvariability. Spatial variability must also be represented a monitoring network: for examin ple, 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 be monitored,an understanding what the data to of obtainedby the systemwill be used for, and a knowledgeof the level of detail in collecting the data that must be exercised spaceand time. in
DataRequirements
The amountand type of datato be generated a monitoringsystem by 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 data must be tightly relatedto monitoring of
146
CHAPTER9
MONITORING NETWORKS
:
Telephone
Hardwire
+ + , . .
lnfrared + + xgnr
Water quality monitoring
(Courtesy SierraMisco, Environof Inc., Figure9.1 A telemetry monitoring system. mental Products. Berkeley. CA. insteadof the to objectives.Furthermore,it might be necessary monitor surrogates condition to be tracked.' For example,if lake eutrophicationis the issue,phosphorus If is flre&sureS; this approach taken, andchlorophyllconcentrations might be surrogate is selectionof appropriatesurrogates very important and the foregoingcomments aboutdata formats and so on are also applicable.Hydrologic,waterquality, land use and treatment, topographic,soils, vegetativecover, meteorologic,and many other in typesof datamaybe requiredin combinationor separately a monitoringplan. It is 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 a plannedsystem activitiesdesigned producea quality product(datain this case) of is that meetsthe needsof the user. Quality assurance a plannedsystemof activities designed guarantee to that the quality control programis being carried out properly. A quality managementplan should be part of the overall monitoring program to and should be preparedwhen the monitoring program is being developed ensure that the data collectedwill be of a satisfactorynature for the monitoring program's objectives.3
9,4
NETWORKS HYDROLOGICALMETEOROLOGICAL
147
To useror
148
CHAPTER9
MONITORING NETWORKS
Water level
Weather station
Computer
,/: :
Modem Data Collection
(Courtesy SierraMisco, Figure9.3 Computer in a realtime use telemetry system. of Inc., Environmental Products, Berkeley, CA.)
networks may be designedto monitor physiographic, climatic, hydrologic,biologic, and chemicalfeatures, combinations these,in a region or river basin.They must or of have gaugedensitiesand distributionsthat are sufficientto permit interpolationbetweengaugesitesin a mannerpermitting valid conclusions be drawn for the entire to areacoveredby the network. Typically, measureinents madeof suchvariablesas are 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 hydrologicalmeteorological 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 essential better understanding system is to interof actions and to the designand testing of hydrologicmodels.It is also the meansby which a determinationcan be made of the effectiveness measures of taken to alter performance. watershed
REFERENCES
t . S. J. Nix and P. E. Black (eds.),Proceedings the Symposiury. Monitoring, Modeling, of on
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. 6978. J. J. Lawrenceand A. S. Y. Chau, "Quality Assurancefor EnvironmentalMonitoring," in Ref. I, pp. 165176. 4. H. E. Postand T, J. Grizzard, "The Monitoring of StreamHydrology and Quality Using Microcomputers," Ref. 1, pp. 199208. in 5 . P. J. Gabrielsen A. J. Carmeli, "Operation of a HydrologicMeteorologicMonitoring and Networkin a Severe Winter Environment," Ref, l, pp. lI3122. in
PARTTHREE .
C h a p t e r1 0
r Prologue
The purposeof this chaPteris to: . ' . . . the Expandon definitionsof termsfrequentlyusedin describing runoff process. Presentconceptsof surfacerunoff and drainagebasin discharge. Introduce elementsof drainagebasin geomorphology' characteristics. of Describequantitativemeasures watershed 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, and flsh and wildlife management. recreation, ity control, waterbased precipitationandrunoffis influencedby variousstormand The relationbetween of Because thesecomplexitiesand the frequentpaucity of adebasin characteristics. to quaterunoff data, many approximateformulashavebeendeveloped relaterainfall the whereas were usuallycrudeempirical statements, andrunoff. The earliesfofihese descriptiveequationsbasedon physicalprocesses. trend now is to develop
154
CHAPTER RUNOFF THE 1O AND 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 quantitiesexceeding soil or vegetation in the 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 definedswaleor channel,usually beforeflowing a more than a few hundredfeet. The lines separating land surfaceinto watersheds the are called divides.Thesenormally follow ridges and moundsand can be delineated using contourmaps,field surveys, stereograph pairs of aerial photographs idenor to tify gradient directions.
Area Contributing
In the majority of hydrologicanalyses, magnitudeof total surface the areacontributing direct runoff to somepoint of interestis needed. Because variationsin topography, of the true surface areacannotbe easilymeasured. horizontalprojectionofland area The 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 blockedareascan exit or only by seepage evaporation,or by transpirationif vegetated. sufficientrainfall or If 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 the watershed may contributeto the peak surface of 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. will be shownlater (Chapter12) unit hydrograph As 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.
155
Subdivisionof contributing from noncontributingareashas traditionally been subjectively accomplished from siteinspection, topographic and soilsmaps,and aerial photos.Soils having good drainageclassifications, dark tones or colors on aerial or photos, can often be consideredas passiveareas.Other signs of noncontributing areaswould includepresehce wetlands,grassed of areas,rooftops (unlessconnected to the drainage), terraces, erosioncontrol structures, stockwateringponds,and flood control dams. Boughtontdeveloped quantitativemethod of determiningthe proportion of a a watershed that contributessurfacerunoff in different storms,and at different times duringthe samestorm.by analyzingrainfall andrunoff records.His logic is asfollows: 1. Watersheds be idealizedas a group of "surfacestorage can 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 Boughton by 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 the rainfallrunoff graph.Both the capacityof the cell and in its area as a percentage the watershed estimated. of are 5. After subtracting contributionto runoff from the smallestcapacitycell, the 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 contributing,whicheveroccurs first. is When Boughton applied the procedureto a test watershed,r was found that it runoff occurredfrom the entire watershed only 3 of 30 eventsin the l5year study on period.In abouttwothirds of the runoff events, discharge occurredonly from the cell with the smallestsurfacestoragecapacity.
156
StreamPatterns
Wind, ice, and water act on land surfaces createseveraltypes of drainagepatterns to 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, freezethaw activity, and sedimenttransport.
157
Typical streampatterns shownin Fig. 10.1.The most common,dendritic, are is characterizedby numeroussmall tributariesjoining at right angles into higherorder 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 few hundredfeet. a Trellis patterns characterizedby are long main streams intercepted numerous by shorterrightangletributaries.They are commonin the Appalachians (EasternUnited States)and are also seenin the Rocky Mountains along the foothills, Multibasin patterns,also called derangedsystems, occur in low gradient swampy areaswith numeroussurfacedepressions normally haveonly a few tributaries.Theseoccur and 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 rarely straightexcepton steepslopes homogeare in neous 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 other forms when sediment or 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 sediment the 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 sedimenthungry 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.
1 58
CHAPTER10
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, alsomodify the land drainage can forms by creatingvalleys,deltas,and other geomorphicfeatures. Streamsare classified beingyoung,mature,or old on the basisof their ability as 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 old whenthe channels as in its system havebecomeaggraded. The flow velocitiesof old streamsare low dueto gentleslopes that prevail.Wide meander belts,broad flood plains,anddeltaformation are alsocharacteristic old streams. of The lower reaches the Mississippi, of 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.3a 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 modelprototypestudiesof aircraft, dams,and turbomachinery. Suchstudiesinvolveconsiderations geometricas of well as dynamic similarity. In the samemannerthat inferencesas to the operationof a prototype can sometimes drawn from a geometricallysimilar model, inferences be canalsobe drawnaboutthe operationof one drainage areaon the basisof information obtained from a similar one. Perfect similarity will never be realized if natural drainagesystems compared,but striking similarities have been observedwhich are can often be put to practical use.
159
;)
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 because numberis a dimensionless locations in the two systemsthat have a similar geometry; that is, secondorder streams, thirdorder streams,and so forth. of Streamlengthsare determinedby the measurement theh projectionsonto a If horizontal plane. Topographicmaps are useful for obtaining suchmeasurements. possible Z, the meanlenglh of a streamsegment of order a is definedas L,, then it is to determine Z, using2
tr
2!i, L,,
N,
(10.1)
of whereN, is the number of streamsegments streamorder u. Another measurerelated to stream length is the distanceL"o from a point of the intereston the main streamto a point on the primary channelthat is nearest center 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 development a drainagebasin in Of particular significance the physiographic 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 to the intersectionof this flow path with an established channel, then basin of the first order is the basicelementof a larger drainagesystem, drainage overlandflow length can be determinedfor thesefirstorderbasins. 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 andtoaverage
160
15), the use of the longestoverland flow length is prescribed,measuredfrom the upstreamend of the firstorder 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. example, quantity of discharge For the from any drainage basinis obviously a function of the areal extentof that basin. Correlationshavebeenobserved between average the area, of basinsof order Au, u, andthe average length of streamsegments, Thesevariablesare often relatedby 2,. an exponentialfunction. For example,studiesof sevenstreamsin the MarylandVirginia areaby Hack haveproducedthe relationship6
L : I.4Ao6
(10.2)
where L = the streamlength measured miles to the drainagedivide in A : the drainage area(miz) Hack's observations indicatethat as the drainagebasinincreases size,it becomes in 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: a measureof flow suchas mean annual runoff where Q A : the size of the contributine drainaeeatea
(10.3)
Valuesof c and m are determined regression by analysis(seeChapter26); Fig. 10.3 illustratesa relation of this form.
r u o
8 ( ( ) 6
= a
./
10
20
200
Figure 10.3 Runoffdrainage arca correlation for five Maryland streams (1933 storm
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 segments a drainagebasin (total frequency is defined as the summation of all of number of segments all orders)divided by the drainagearea.
Channeland BasinGradients
havea very strongeffecton the surface basinand its channels of The slopes a drainage streamchannelprofiles exhibit the characteristic runoff pro""r, of thairegion. Most direction.Figure 10.4illustratesthis slopeproceedingin a downstream 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 asparameters describedrainaget 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 maximum to from the centerof massof rainfall excess the peak rate of runoff ) and discharge. In additionto the slopeofthe streamchannel,the overall land slopeofthe basin factor.A quantitativerelation betweenvalley wall slopes is an important topographic method A has and streamcnannetstopes been derivedby Strahler.3 commonly used has been presentedby Horton'SThe method of determiningthe slopesof a basin map of the drainagearea grid over a topographic a involvessuperimposing transparent b"t*""n its intersectionswith the drainage in question.nacfr grlJnne is measu."d needed' of divide; the numberof intersections eachgrid line with a contourline is also then be made using A determinationof the land slopecan nsec0, S:t (10.4)
162
uContour'
Figure 10.5 Determinationof meanland slope:numberof vertical intersections: 72; tumber of horizontal intersections= 120; total : length of vertical grid segments 103,900ft; total length of : horizontalgrid segments 101,200 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 the horizontal and bv 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 obtain to an estimate the meanland slope.This procedure illustratedin Fig. 10.5. of is
AreaElevationRelation
How the areawithin a drainagebasinis distributedbetweencontours(Fig. 10.6)is of interestfor comparingdrainagebasinsand gaining insight into the storageand flow characteristics the basin.For suchstudies.an areadistribution curve suchas that of shown in Fig. 10.7 is used.The curve can be obtained by planimeteringthe areas
163
l l \ ' / t r 1 l l \ / / / / r r \ f t /
,ffi__.w;i.i
.Figure 10.6 Topographic of WendyRun drainage area map lines. 20,40,and60ftcontour 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 elevations of from the areaelevationcurvesas the elevationat 50 percent, 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 Percentage area
Figure 10.7
1 64
CHAPTER10
the form
o:($)rrPr)
1902.
I
(10.5)
1898r
l .
a 903
^ 4 0
190r l
 J 6 .E
I t28 a Slope
8 3 6
. ) J +
,77 X 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.)
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 be used with care. For rough approximationsin this type of computation should preliminary planning studies,suchmethodsare frequently helpful, however.Equaprecipsuchasantecedent Eq. 10.5areimprovedif otherparameters tionsresembiing are characteristics included. Suchrelations and storm itation, soil moisture,season, or techniques graphicalmethods'Linsley using multiple regression can be described correlatreatmentof methodsfor developing and coworkerspresenta very complete tions involving severalvariables.e Soil moisturerelationsnormally havea soil moistureindex as the independent soil of variable,sincedirect measurements actualantecedent moistureare not generat insertedare groundwater,flow the beginning ally practical. Indexesthat havebeen Groundwatervalues precipitation, and basin evaporation.lo of the storm, antecedent shouldbe weightedto reflect the effectsof precipitation occurring within a few days from previousrains may affect results. of the storm becausesoil moisture changes employedto estimatesoil moistureamounts, can measurements be Panevaporation since eviporation is related to soil moisture depletion.tl Antecedentprecipitation precipitationis readily indexes(API) haveprobably receivedthe widestusebecause deficiencyof the basin. measuredand relatesdirectly to moisture precipitation index is A typical antecedent (10.6) Po: aP6'l bP, f cP., where
p,." po, :
P.:
fi"HTlJ:;fJil:rr:Hifl3*ff"e
Coefficientsa, b, andc are found This index links annualrainfall and runoff values.r2 to by trial and error or other fitting techniques producethe best correlationbetween 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 wherethe r subscript P refersto precipitationwhich occurredthat many daysprior to b to the given storm,and the constants (lessthan unity) are assumed be a function In by of t. Valuesfor the coefficientscan be determined correlationtechniques' daily to evaluationof the index, b, is considered be related to / by b,: K' (10.8) normallyrepoftedin the range0.850.98.The initial constant whereK is a recession
166
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 : I. t 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. addedshortcomingis that many formulas fit only a specific An watershedand have little generalutility. Empirical equationsdemandgreat caution and an understanding their origin. of EXAMPLE 10.1 PrecipitationdepthsP, for a l4day 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 1.5 l t,24 t.22 1.10 1.09 1.28 t.l5 1.04 1.54 1,39 .25
167
C a o a ao
12 months
x.
0.1 10
RecurrenceInteryal (Yr) 100
for dataconsolidated FiveRivfrequency Figure 10.9 Lowflow (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 atedwith proposeddesigns anticipatedoperatingschemes. of usuallydiiectedtowardstudies miximum (flood) andminimum (drought)flows.ra'1s Figure 10.9 illustrates a typical drought frequency analysis.Unfortunately,many existingrunoff records are shortterm; 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. Timeseries analysesare particularly pertinent to the problem of estimating trends, cycles, and fluctuationsin hydrologic data. They also permit derivation of by generatingprocesses which syntheticrecordsof runoff can be developed.
168
CHAPTER RUNoFFAND oAToHMENT 10 THE 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 Chapter27. in
r Summary
Runoff is probablythe most complexyet most important hydrologicprocess underto stand.It has attractedthe attention and focus of engineersand scientistsand comprisesthe greatestpercentage far of most hydrology textbooksand publications. by The conceptsintroducedhere will be more fully developed the next six chapters, in as well as in significantportions of Parts Five and Six.
PROBLEMS
REFERENCES 169
Figure for Problem 10.3 from one or two other hydrologytexts, or L0.4. Using any dictionary,plus indexes glossaries 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, "GeologyPart II," in Handhok of Applied Hydrology. New York: 3. A. N. Strahler, McGrawHill, 1964. "TopographicCharacteristics DrainageBasins,"U.S. Geological of 4. W. B. L4ngbeinet al., Survey,Water Supply Paper,968c(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. s," Characteristic Trans.Am. Geoplty Union 33(1952). "Discussionof Paper,Flood Flow characteristicsby c. S. Jarvis," ?ans. R. E. Horton, 8. ASCE 89(1926)'. 9. R. K. Linsley, Jr., M. A. Kohler, and J. L. H. Paulhus,AppliedHydrology,2nd Ed' New York: McGrawHill, 1975. 10. ven Te chow (ed.), Handbook of Applied Hydrology. New York: McGrawHill,1964. 11. R.K.Linsley,Jr.,andWC.Ackerman,"MethodofPredictingtheRunofffromRainfall," Trans.ASCE 107(1942). 12. S. S. Butler, EngineeringHydrology. EnglewoodCliffs, NJ: PrenticeHall, 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 that the reader can rel9le them so of to the quantltative assessments runoff presentedin subsequdnfchapters. . 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 three chaptersare dedicated the in numerousdevelopments hydrographanalyses, the and expands conceptsintrosubjectin 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 drographrouting methods,are presented Chapter 13.
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
Processes PrecipitationStreamflow
to Duriirg a given rainfall, water is continually being abstracted saturatethe upper levelsof the soil surface;however,this saturationor infiltration is only one of many Rainfall is also interceptedby trees, plants, 4nd roof continuous abstractions.46 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, to the watershed outlet. Figure 11.2aillustratesthe distribution of a prolonged flow uniform rainfall. Although such an event is not the norm, the conceptis useful for storagewould be distributed. showingthe mannerin which detentionand depression possesses certain amountof baseflow a In general,the channelof a watershed during most of the year. This flow comesfrom groundwateror spring contributions as and may be considered the normal daytodayflow. Dischargefrom precipitation excessthat is, after abstractions are deducted from the original rainfallthe constitutes direct runoff hydrograph(DRH). Arrival of direct ninoff at the outlet enoughtime continues, for accounts an initial rise in the DRH. As precipitationexcess the qlapqgs progressively for distant areasto add to the outlet flow. Consequently,
179
Depressionstorage
fi
Time (b)
Figure 11.2 (a) Distribution of a uniform storm rainfall for condition of is no interceptionloss.Note that all water storedin depressions ultimately is to evaporated infiltrated while somedetentionstorage also subjected or (b) theselosses. Equilibrium discharge hydrograph.
duration of rainfall dictatesthe proportionatearea of the watershedamplifying the peak, and the intensity of rainfall during this period of time determines resulting the 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 (runoff)] is reached.The condition of iqrliUri"t" finflow (rainfall) equals outflow Extended maximum storageand equiiibrium is seldom if ever attained in nature' its duration negateany rainfall lnuy o"Iu., but viriations in intensity throughout possibility of u IRH of the theoretical shapefor constantrainfall intensity' Anormalsingle.peakDRHgenerallypossessestheshapeshownbycurveBin peak magnitudeof this in Fig. 11.2b rather tian iy the "ur',oJ Fig. 11.2a.The time to ofthe rainfall' and the size,slope, onih" intensity and duration hyirograph depends for once peak flow has beenreached a capacityof the watershed. shape,and storage its source of supply coming given isolatedra'instorm,itt" ORg beginsto descend, suchas detentionand channel within the watershed largely from water accumulated storage. by processes involvedin forming the DRH can be better understood visualizing theprecipitationexcessaspartiallydisposedofimmediate$bysurfacerunoffwhile later from boundariesand is released a portion remainsheld within the watershed integratedeffectsof the duration Thus the shapeand timing of the DRH are storage. factorsaswell asthe effect of intensityofrainfall and other hydrometeorological and storagecapacity' upon the factorsof the watershed the physiogiaphic
how the hydrographcan be subdividedinto its com\It is important to understand ofprecipitation andwatershed shape and nenrpar15 to look at the effecton hydrograph 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), lrydrographis generallytaken to by applicationof a rating curve.Hereafter,the term hYdrograPh' indicate a discharge a pgriod Figure t t.gi lttustrai"* ih" hydrographof a permanentstreamduring lrydrographbecausegroundbetween precipitation events,known i" i bot" flow of the base water sustainsthe flow. Four general conditions cause modification HortonTusing the following sets of 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.
(e)
(0
fo)
(h)
Figure 11.3 Effectsof stormandbasincharacteristics hydrograph on 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 > f. In this example,interflow and groundwater i flow are zero,assoil
176
CHAPTER11 HYDROGRAPHS
End ofrainfall
9C
Groundwater flow
Time
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 be typical of a large storm event. Figures 1l.jeh illustrate how hydrograph shape can be modified by areal Minor variations in rainfall and rainfall intensity and by watershedconfiguration.8 are linked to variationsin stormintensity.In fluctuationsshownin thesehydrographs Fig. 11.3eonly the delayingeffectspertinent to a storm over the upstreamsectionof Figures11'3g of the Figure 11.3fshows reverse this condition. the areaareindicated. effects of basin geometry. and h depict the comparative interflow and channelprecipitation are grouped In most hydrographanalyses, Channelprecipitation begins treated independently. with surfacerunoff rather than endswith the storm.Its distributionwith respectto time with inceptionof rainfall and is highly iorrelated with the stormpattern.The relative volume contributiontendsto increasesomewhatas the storm proceeds,since stream levels rise and the water and areaoccupiedby streams The fraction of watershed areatendsto increase. surface on the order of 5 percent or less,so the percentage lakes is generallysmall, usually of rrnoff relatedto channelprecipitation is usually minor during important storms. rate up to by Distribution of interflow is commonly characterized a slowly increasing period, followed by a gradual recessionthat terminatesat the the end of the storm intersectionofthe surfaceflow hydrographand base flow hydrograph.Figure 11.4 illustrates the approximatenature of the componentsof channelprecipitation and interflow. The baseflow componentis composedof the water that percolatesdownward as reservoirand then flowsto surfacestreams groundthe until it reaches groundwater The groundwaterhydrographmay or may not show an increase water discharge.
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. pointed out in Fig. 11.3,the relative magnitudeof eachcomponent As for a given storm is determinedby a combinationof many factors.Hydrographsare analyzed provideknowledgeof the way precipitationand watershed to characteristics interact to form them. The degreeof hydrographseparation requireddepends the on 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 separate overlappingparts of consecutive to the surfacerunoff hydrographs.
11.4 BASEFLOWSEPARATION
Severaltechniques usedto separate hydrograph's are a surface and groundwater flows. Most are basedon analyses groundwater recession depletioncurves.If there is or of no addedinflow to the groundwater reservoir,and if all groundwater discharge from the upstreamareais interceptedat the streamgauging point of interest,then groundwater discharge can be described either e'to by er: eoK' or er: eoK' where eo : q, : K: e: a specifiedinitial discharge the discharge any time / after flow 46 at 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 approximately For most watersheds, are stable,sincethey closelyfit watershed geology.Nevertheless, recession the 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
4 s  Q t  Ko log"
(rr.2)
This equation determinesthe quantity of water releasedfrom groundwaterstorage betweenthe times of occurrenceof the two discharges interest,or it can be usedto of calculatethe volume of water still in storageat a time some chosenvalue of flow occurs.To get the latter, 4, is setequalto zero and qsbecomes reference the discharge. Figure 11.5ais a plot of Eqs. 11.1and II.2 andprovides additionaldefinition.
178
CHAPTER11
HYDROGRAPHS
qo
.. ,?
K
\r3
t (a.)
?r
I
I
I
q d
I
/
I I
qo
Qt
t1
by Groundwaterdepletioncurvescan be analyzed variousmethodsto evaluate Data from a streamgaugi.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 intervals.Flows at the beginningof eachinterval are analogous 4e,whereasthose to at the end are analogous q1.Next, selectseveraltime intervalsand plot correspondvalues 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 Points taken from true groundwater recessionperiods should approximately 4.
,/
o
I
A
,(
.5r"2'
/(^lI
 s l o p e o 0 A= f
inEq11.1
9/ .,
fi:
/#
/
Figure 11.6 Graphical rnethod for determiningrecessionconstant K. (U.S. Departmentof Agriculture, Soil ConservationService.)
describe straightline. a Because : eo : 0 when4o: 0, a straightline canbefitted h 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) as a curveon arithmeticpaper,Fig. 11.5a. or
Techniques Separation
r Severalmethodsfor baseflow separationare used when the actual amount of base flow is unknown. During large storms,the maximum rate of discharge only slightly is affectedby baseflow, and inaccuracies separationmay not be important. in The simplest baseflow separation technique to draw a horizontalline from the is 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 daysafter the peak, is connectedto point C by a N 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 backcalculateall base flow to the left of Point fl
180
cHAPTER HYDRocRApHS 11
9p F
Time
Flgure 11.7 Illustration some of 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. fourth A 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).
is of B. All thesemethodsare approximatesincethe separation hydrographs partly a subjectiveprocedure. techniques determinesurface to Figure 11.8illustratestwo graphicalseparation the . runoff and groundwaterflow components Line AD represents simpleprocedure 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
(rr.4)
182
CHAPTER11
HYDROGRAPHS
11.6TIMEOF CONCENTRATION
The most common definition of time of concentrationoriginatesfrom consideration the of overlandflow. If a uniform rain is appliedto atract, the portions nearest outlet contribute runoff at the outlet almost immediately.As rain continues,the depth of growsand discharge ratesincrease throughout.Runoff contribuexcess the surface on to tions from variouspoints upstreamarrive at later times, addingthemselves continuing 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 either the excessrainfall releasetime or the wave travel time because time for is runoff to arrive at the outlet from the most remotepoint after rain ceases assumed 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 Fig. 12.2).The reasoning usedin this definitionis that direct runoff ceases the point of inflection. For a small tract of land experiencing uniform rain, the entire areacontributes an at approximately sametime that the runoff reaches equilibrium.This givesrise the the to yet anotherdefinition of time of concentration.If rain abruptly ceased, direct releasetime t,. On the basis runoff would continueonly as long as the excessrainfall 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 tions are presented Chapters12, 15,16, and25.
11.7 BASIN LAG TIME The relative timing of rainfall and runoff must be known if drainageareashaving subbasins to be modeledor if continuoussimulation is desired.A basic measure are 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 to the peak rate of flow. Time lag is characterized the ratio of a flow length to a 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.
183
Remarks
Developedfrom SCS data for sevenrural basins in Tennessee with welldefinedchanneland steepslopes(37o to 1O7o); overlandflow on for 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. 677I).t4
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?
rainfall intensity,in/h retardancecoefficient length of flow path, ft slopeof flow path, ftlft
Developedin laboratory experiments Bureau of by 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 intensitydurationfrequency curve gives direct graphical solution for t". Equation developed SCS from agricultural waby tersheddata; 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.
r" = 1.8(1.1 C)Losofso333 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
',' : 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
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 original equation has been widely used and modified in His variouswaysby other investigators. Eagleson proposedan equationfor lag time has on sewered drainageareashavinga minimum sizeof 147 acres.2z early investigaAn tion (1936) on small drainageareas(24 acres)was conducted Horner in his by classical work on urbandrainage St. Louis,Missouri.23 in 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 the drainagearea. of Snyder'sstudybasedon data from the AppalachianMountain regionproduced the following equationfor lag time:z1
t,1: c,(L""L)o3
(1 .s) 1
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 be constantfor to 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 equation the
t " ', : 
L
(l.5 /n)R2/3 t/2 s
(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 refinementof the Snyderapproach. also a It indirectly includesthe important effect of the storm pattern.
PROBLEMS185 of Linsley and Ackerman give examples applicationof the following modified 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 the syntheticunit hydrographdiscussions Chapter 12.
r Summary
is the Understanding structureof hydrographS important to many designand water the hydrographrepresents portion of the hydrologiccycle supply applications.The for ratesof flow in streams setting needin orderto determine mostoften that engineers and establishing designingflood protection measures, 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 estimates thesevolumesare important to designof dams,reservoirs, ' numerousother structures. including the time components, of After graspingthe fundamentals hydrograph presentedin this chapter,the reader should be well preparedfor the relationships presented throughquantitativedevelopments hydrographtheory and applications of 12 through 16 and in Part Five. out Chapters
PROBLEMS
to and usetwo different techniques separate 11.1. Referto Fig. 11.1.Replotthis hydrograph the baseflow. for 11.2. Obtain streamflowdata for a water courseof interest.Plot the hydrograph a major the event and separate baseflow. runoff 11.3. For the event of Problem I1.2, tabtlate the precipitation causingthe surfacerunoff the and determine duration of runoffproducing.rain. Estimatethe time of concentraComparethis with tion and useEq. I 1.4 to estimatethe time baseof the hydrograph. 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 b. Determinethe hour of cessation the direct runoff usinga semilog 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
8 9 l0 11 t2
IJ
t4 15
(b) hyetograph, beginningofdirectrunoff, (c) cessationtimeofdirectrunoff' (d) base fllw separationassumingthat additional contributions to base flow are negligible of during ihe period of rise, and (e) crest segment the hydrograph' of by your instructor,obtain measures the watershed assigned 11.6. For an urban watershed of length, and slope,and compareestimates the time of concentrationusing the area, in Kirpich, USBR, FAA, and SCSLag equations Table 11.1'
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.4I116' Service, AgriculturalResearch 3. R. K' Linsley, M. A. Kohler, and J' L. H' Paulhus,Applied Hydrology' New York: McGrawHill, 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, Soil Sci.Soc.Am. 5,399417(1940). 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,,'in EngineeringHandbook, Sec. 4, U.s. Department of Agriculture, Soil 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. s," Characteristic Trans,Am. Geophy Union 33(1952).
REFERENCES 187 "Streamflow 12. L, K. Sherman, from Rainfall by the UnitGraphMethod," Eng.NewsRec. 108(1932). 1 3 . D. F. Kilber, "Desktop methods for urban stormwatercalculation," Ch. 4 in Urban Stormwater Hydrology, Water ResourcesMonograph No. 7, American Geophysical Union, Washington, C., 1982. D. 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. 129146, December1946. 16. FederalAviation Administration,"Circular on Airport Drainage,"ReportA/C 050532058, 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, 447454(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), 140205(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 discharge from rainfall events and 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 1932. He in defineda unit graph as follows:2 produces1in.depth given oneday drainage area, Ifa given rainfall a ofrunoffoverthe thehydrograph showing rates whichtherunoffoccurred be consideredunit at can a the graphfor that watershed. baseflow) Thus, a unit hydrographis the hydrographof direct runoff (excludes net rain (the total runoff after abstracfor any stormthat produces exactly 1.0inch of tions).Sucha stormwould not be expected occur,but Sherman'sassumption that is to the ordinatesof a unit hydrographare t.O/P times the ordinatesof the direct runoff hydrographfor an equalduration 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 time. The storm duration,X, that producedthe of has unithydrographmustbe specified because watershed a differentunit hydrograph a
189
is for eachpossiblestorm duration. An Xhour unit lrydrograp,h defined as a direct runoff hydrograph havinga 1.0in. volumeand resultingfrom anXhour stormhaving would havea 1.0in. volume a net rain rate of 1,/Xin.lhr. Azhr unit hydrograph producedby a 2hqstorm,and a 1dayunit hydrograph would be producedby a storm having 1.0 in. of eicessrain uniformly producedduring a 24hr period. The valueX is often a fraction. Figure 12.1illustratesa2ltr,l2hr, and24hrwthydrograph for a given watershed.
t (b)
24hr
T (c)
)z nl
Figure 12.1 Illustration of 2br, I2hr, and 24ht unit hydrographsfor the same watershed(Note: a : b : c : 1' X A).
190
CHAPTER UNIT 12 HYDROGRAPHS rainfall By Sherman'sassumption, applicationof an Xhourunit 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 3hr storm producing2.0 in, of net rain would haverunoff rates2 times the valuesof the 3hr unit hydrograph. Onehalf inch in 3 hr would produceflowshalf the magnitudeof the 3hr in unit hydrograph. This principle of proportional flows is expanded Section 12.3 and 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 generate to direct runoff hydrographs any storm with durationsthat are multiple integersof the U.H. duration. The constructionof unit hydrographs stormswith other than integermultifor ples of the derived duration is facilitated by a method known as the Slrydrograph The procedure,as explainedin Section 12.4, developed Morgan and Hulinghorst.3 by employs a unit hydrographto form an Shydrographresulting from a continuous appliedrainfall. The need to alter duration of a unit hydrographled to studiesof the unit rainfall. The concept of shortestpossiblestorm durationthe instantaneous instantaneous unit hydrograph(IIJH) is tracedto Clark6and can also be used(Secfor tion 12.5) is constructingunit hydrographs other than the derived duration. The previousdiscussion assumes the analysthasrunoff and rainfall datafor that deriving a unit hydrographfor the subject watershed.The application of unit hyand drographtheory to ungauged receivedearly attentionby Snydera also watersheds of by Taylor and Schwartz,5 who tried to relate aspects the unit hydrographto watershed characteristics. a result, a full set of synthetic unithydrographmethods As emerged. numberof theseare presented Section12.6. A in
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 be to employed.An upper limit of watershedsize of appro5imately1000 mi2 is overcautious, althoughgeneralstormsover suchareasare not unrealisticand somestudiesof areasup to 2000 mi2 have used the unithydrographtechnique.The lower limit of watershed extentdepends numerousother factorsand cannotbe preciselydefined. on A generalrule of thumb is to assume about 1000acres.Fortunately,other hydrologic techniques help resolveunit hydrographs watersheds for outsidethis range. The preliminary screeningof suitable storms for unithydrographformation shouldmeet more restrictive criteria before further analysis: 1. Duration of rainfall event should be approximately1030 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 a single to duration by meansof Shydrographs IUH procedures. or 4. Direct runoff ordinatesfor eachhydrographshouldbe reducedso that each eventrepresents1 in. of direct runoff. 5. The final unit hydrograph a specificdurationfor the watershed obtained of is 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 of equal duration storms.Extremerainfall set intensityis not reflectedin the determination. intensestormsare needed, study of If a recordsshouldbe madeto ascertain their influenceupon the discharge hydrograph by comparingpeaks obtainedutilizing the derived unit hydrographand actual hydrographsfrom intensestorms. Essentialstepsin developing unit hydrographfor an isolatedstorm are: a l. Analyzethe streamflowhydrographto permit separationof surfacerunoff from groundwaterflow, accomplished the methodsdeveloped Secby in tion 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 HYDROGRAPHS GHAPTER UNIT 4. Finally, the effective duration of the runoffproducing 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 transposed from one basinto anotherundercertain circumstances. exampleillustratesthe 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
I
I
*zTotal
o Time (hr)
1l
t2
500
o FA
/^
3 4 5 6
t'# j
7 8 Time(hr) Dfuect runoffduration
Baseflow separatlon
193
the hydrograph rise curve startsan appreciable and endswherethe recession intersects baseflow curve.The importantpoint hereis to be consistent the in methodologyfrom storm to storm. 2. The depth of direct runoff over the watershedis calculatedusing > (DR x Ar) _ 2447 cfshr : 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 1.415.Table12.1outlines computation the unithydrograph of by the ordinates. 4. Determinethe duration of effectiverainfall (rainfall that actuallyproduces surfacerunoff). As statedpreviously,the unit hydrographstorm duration
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
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
0 0
1 100
2 250
3 200
4 100
5 50
6 0
Develophourly ordinatesof the total hydrographresultingfrom a 4hr designstorm amounts: havingthe following excess
1 Q.5
0.5
1.0
4 1.0
Solution. The 4hr duration of the designstorm is an integermultiple of the can duration.Thus,the total hydrograph be foundby addingthe unit hydrograph rain, asshowninFig. l2'3c. of of contributions two 2hr increments endtoend a has The first 2hr stormsegment 1.0 in. of net rain and thus reproduces unit 2.0in. of netrain (in 2 hr);thus 2hr stormsegmenthas The hydrograph. second its ordinatesare twice those of a 2hr unit hydrograph.The total hydrograph,
12.3 F
195
1.0
r>
h
u.) 0
2 3 4 . 5 6 (a) 2hourunit storn excess
300 200
i90 zoo
1oo '6 0
1.0
ir
'6 400
:i
Figure 12.3 Example 12.2 deivation of total runoff hydrographusing a 2hr 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 the of watershed not influencedby previousstormsthat is, one can superimpose is hydrographs offset in time and the flows will be directly additive. The simplestway to developcompositedirect runoff hydrographs multiplehourstormsis in a spreadfor sheet.Care must be taken, however,in visually confirming, as in ExampleI2.2, that the start and end points of runoff from eachcontributingXhr incrementof rain are properly selected. commonerror is to lag eachadditional contributinghydrograph A 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 2hr lunit hydrograph Table 12.1, determinethe direct runoff in hydrographfor a 4hr. storm havingthe following excess rain amounts:
o.7
Solution 1. Tabulatethe unit hydrograph intervalsof the selected at time interval, A/, as shown inTable 12.2.
TABLE 12.2 UNIT HYDROGRAPH APPLICATION EXAMPLE12.3 OF Effective rainfall (in.)
Time (h0 0
I
Contrib. of first 2hrrain U HX 1 . 4 ' 0 I1.9 l19 463 531 321 181 91 49.4 6.9 0
uH x 2.4'
2
J
4 5 6 7 8 9 10 1l t2
0 11.9 119 483 734 11 1 5 1091 641 359 163 84.7 I 1.8 0
12.gUNTHYDRoGRAPHAPPLCAToNSBYLAGGNGMETHoDS197
X 2. Determine the correct UH multiplier for eachXhr interval. Because is storm producea total net 2 hrs for this example,the first two hours of the rain of 1.4 inches. Similar$, the last two hours of the storm produce 2.4 inchesof net rain. 3. Determinethe correct start and end times for eachof the two hydrographs and tabulatethe contributionof the l.4inch and 2.8inch rains at the Xhr storm startedat / : 3 hrs, the appropriatelag times. Because second : 3 hrs as shown inTable I2'2. runoff for thisstormcannotbegin until / the 4. Add the contributionsat eachtime to determine total runoff hydrographs for the 4hr storm. and 5. Checkthe tabular solutionby plotting eachof the two hydrographs sum rr at the ordinates each/, as showninFig.I2'4' In addition to using a given Xhr UH for determiningthe runoff hydrographfor a given storm, tagging of ttt" Xhr UH can be used to developother duration unit is hyirographs.The pioCedure the sameas applyingthe Xhr UH to 1.0 in. of net rain in f nouri. As earlier, Y must be an integermultiple of X. For example,if a 1hr unit resultingfrom a 2hr a for is hydrograph available a given watershed, unit hydrograph siormis tbtained by plotting two Lhr unit hydrographs,with the secondunit hyin drographlagged t hr, adding ordinates,and dividing by 2' This is demonstrated unit hydrograph'Thus nigl ti.S, riliere the dashedline rept"sentsthe resulting2ht the t in. of rainfall containedin the original 1hr duration has been distributedover a Zhr period.
Time units
\_
198
CHAPTER UNITHYDRoGRAPHS 12
Time
Figure12.5 Unithydrographlaggingprocedurero develop another hydrograph. unit Modificationsof the original unithydrograph duration can be madeso that two 1hr unit hydrographs usedto form a 2hr unit hydrograph;two 2hr unit hydroare graphsresult in a 4fu diagram,and so on. Care must be taken not to mix durations in the lagging procedure, since errors are introduced; a lhr and a}hr unit hydrograph do not representa 3hr 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 SHYDROGRAPHMETHOD
199
*l
Dhr
Time
(a)
*l
l
discharge the Shydrograph of occursat a time equalto D hourslessthan the time base of the initial unit hydrographas shown inFig. 12.6a. To constructa pictorial 2hr unit hydrograph,simply lag the first Shydrograph by a secondShydrograph time interval equalto the desiredduration.The difference a in Shydrograph 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 Dhr Shydrograph; this Shydrograph hr, andmultiply the lag f difference in Shydrographordinatesby D/t. Accuracy of the graphical procbdure dependson the scaleschosento plot the hydrographs.Tabular solution of the Shydrograph methodis also employed, hydrograph but tabulationsmustbe at intervals of the original unit.hydrographduration. EXAMPLE 12.4 Given the following 2hr unit hydrograph,use Shydrograph procedures construct to a 3hr unit hydrograph.
Time (hr) 0 (cfs)
o 0
 100
2 250
3 200
4 100
5 50
6 0
Solution. The 2hr unit hydrograph the runoff from a 2hr stormof 0.5 is in./hr. The Shydrograph formedfrom a net rain rate of 0.5 in./hr lasting is indefinitely shown Fig. 12.6a. ordinates foundby adding 2hr as in the Its are (UH) runoff ratesfrom eachcontributing unithydrograph 2hr block of rain:
200
CHAPTER12
UNIT HYDROGRAPHS
/,/
90
'{.3
u0 Figure 12.7 Shydrograph. Time (min.)
Time (h0 0 I
z
2nd Zhr
3rd2hr
3
4
5 6 7 8
0 100 250
as To find a 3.hrhydrograph, Scurveis laggedby 3 hr and subtracted shown the in Fig. 12.6b.This results in a hydrographfrom a 3hr storm of 0.5 in./hr, or 1.5 in. total. Thus the ordinatesneed to be divided by 1.5 to producethe 3hr unit hydrograph:
201
Time (hr)
0 1 2 3 4 5 6 7
Shydrograph
Lagged Shydrograph
Difference
(r2.4)
The Qo  Q" dtfferences dividedby I Lt to convertfrom a stormwith 1Al inches are in Al hours to one with 1.0 in. in At hours,which is the definition of a Arhour unit graph. As Ar approaches zpro, Eq. 12.4 becomes
: a 0,(ruH) ! d ts I
(12.s)
202
Figure 12.8 Unithy&ographdescriptionof the runoff process. Unit hydrograph; a sequence lmin (a) (b) of (c) storms; superposition runoffhydrographs each for of of (After Schaake.e) the lmin storms. which showsthat the flow at time I is proportional to the slopeof the Shydrograph at time r. In applications, slopeis approximatedbyLQ/A,I, and the IUH ordinates the can be estimatedfrom pairs of closelyspacedpoints of the Shydrograph. Ifan IUH is supplied,the aboveprocesscan be reversed,and any Xhour unit graph can be found by averaging IUH florvsat Xhr intervals, or
{r2.6)
Use of this approximate equation is allowed for small X values and permits direct calculationof a unit graph from an IUH, bypassing normal Shydrograph the procedure.
203
EXAMPLE 12.5 determinethe IUH, and then use it to Given the following 1.0 in./hr Shydrograph, a estimate 1hr UH.
3.5 750 4.0 800
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
JU
luHf
luHr_1
0 1 2 3
A
5 6
The readershouldverify that the 1hr UH obtainedthrough use of the IUH is t t[e approximately sameas that obtainedby lagging the Shydrograph hr, subtractthe differenceto a 1hr UH. ing, and converting rainfall The ordinatesof the IUH representthe relative effect of antecedent the runoff rate at any instant of time. By plotting the IUH with time intensitieson increasingto the left rather than to the right (seeFig. I2.9), andthen superimposing 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 past) is easily observed.In other words, the runoff rate at any time is time into
204
12 CHAPTER UNITHYDROGRAPHS
oo=[;f@xi(tr)dr
unit Figure 12.9 Calculationof runoff rates with the instantaneous hyof average the antecedent is a weighted The runoff rateat arrytime drograph. unit imageof the instantaneous hyThe timereversed rainfall intensities. (After Schaake'v) furtction. the represents weighting drograph of computedas a weighted average the previousrainfall intensities.Therefore,the of.the rainfall pattern computedrunoff hydrographis the weighted,moving average image of the unit hydrograph'e and itre weighting irtt"iiott is the timereversed the Statedmathematically, runoff rate at any time is given by the convolution integral
Q(A:
(r2.7)
ll:
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) time t  r at
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.
GammaDistribution
The shapesof hydrographsoften closely match a twoparametergamma 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 valuer corresponding the apexl to 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 F@ + 1), and variance p2(a + I). Many of the syntheticunit 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 greater in detail later) is thit it guarantees unit areaunder the curve. It can convenientlybe a used to synthesize entire hydrographif the calculatedpeak flow rate an eo and its associated time to are known. This usesa proceduredeveloped Aron und'White.ro by If time r is substituted x in Eq. 12.g, the time to peak tois aB. At this point, for the function/(r) equalsthe peak flow rate e,, or
o,:ffi=ffrr*1
(r2.e)
206
CHAPTER12
UNIT HYDROGRAPHS
1.0
B1
0
x
Figure 12.10 Gamma function shapesfor various shape and scale parameter values. factor A. whereC, A is the unit volumeof runoff from a basinwith arca The conversion = 1.008 is selected make @(a)dimensionless. to C, The function f(a) is shownby Aron and white to be relatedto a by1l (12.10) + a : 0.045+ 0.5d + 5.6Q2 0303 well in the range 1 reasonably Collins showsthat this can be approximated
byt'
( a ( 8
(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 subsequntly Severalof the methodsdescribed be estimated. can foundfrom Eq.12.9, and a from Eq.12.10 ot l2.Il. The unit hydrograph now to. Substitutingalo by be constructed calculatingQ at any convenientmultiple, a, of for x in Eq. 12.8 gives the flow at t : atp as
Qoto: QraoeQo)"
(r2.r3)
which can be solvedfor all the flow rates of the hydrograph. EXAMPLE 12.6 is watershed 1720cfs and of The peak flow rate for the unit hydrograph a 36,000acre the the initiation of runoff. Use Eq. 12.8 to synthesize rest of o".oi, 12 hr following the hydrograph.
207
O(ctu)
0 tt25 1720 876 9 0
0 6 l2
)4
60 120
sufficient intermediatepoints shouldbe generated define the entire shapeof to the hydrograph. tl
Snyder's Method
one techniqueemployedby the corps of Engineersl3 many othersis and 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 obtained,rig.lz.rt,and cleckel to seeif it contains is 1 in. of direct runoff.
208
CHAPTER12
UNIT HYDROGRAPHS
that a Time to Peak Snyder'smethodof synthesizing unit hydrographassumes the from Eq. 11.5.Its locationis lag, estimated peak flow rate occursat the watershed rate are both as established shownon Fig. 12.11.The lag time and peak discharge For characteristics. the lag time, the watershed correlatedwith variousphysiographic and C, is variablesL and L"o for Eq. 11.5 are estimatedfrom map measurements, TableL2.3summaor for developed the locale,usingSnyder'sestimates other sources. rizes a variety of C, valuesfor variousregions. that lag time is a constantfor a particular watershedthat'is, It is assumed uninfluencedby variations in rainfall intensitiesor similar factors. The use of L"o shape,andC, takescare of wide variationsin topography, for accounts the watershed 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)
Average Co
0.6/\4" 0.3
1.2 o;7 0.4 0.8 0.30.9 1.3
1,4
0.51.0
0.8
12.6 SYNTHETIC HYDROGRAPHS UNIT 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 unithydrograph development a function of lag time is
l 
'
tt
5.5
(r2.1s)
where /, : duration of the unit rainfall excess (hr) t1 : the lag time from the certtroidof unit rainfall excess the peak of thb to unit hydrograph This synthetictechnique alwaysresultsin an initial unithydrograph duration equalto in fi/5.5.However, sincechanges lag time occur with changes duration of tlie unit in hydrograph,the following equation was developed allow lag time and peak disto chargeadjustments other unithydrograph for durations. tm:h+0.25(t*t) where t1p: tt : to: t, the adjustedlag time (hr) the original lag time (hr) the desiredunithydrograph 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 inflow of 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 Fig. 12.12) gives additional assistance plotting time widths for points on the hyAs drographlocatedat 50 and 75 percentofpeak discharge.l3 a generalrule ofthumb, proportionedeachsideof the pegk ttr" til width at l/so and IV^ ordinatesshouldbe in a ratio of I:2 with the short time sideon the left of the syntheticunithydrograph Eq. peak. As noted earlier,for smallerwatersheds, I2.I4 givesunrealisticvaluesfor by ihe bur" time. If this occurs,a value can be estimated multiplying total time to the modified basedon the amount and peak by a value of from 3 to 5. This ratio can be boundaries. ii." tut" of depletionof storagewater within the watershed by The envelopecurvesin Fig. I2.I2 ate defined 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 handfittingand planimetering or by curvefitting. to techniques fit a Pearson regression Hudlow and Clarkla used leastsquares III (gamma)probability densityfunction (referto Chapter26) throughthe seven type points. This function has an areaof 1.0 and a shapesimilar SnyaeruoitnyAtograph
1000 800 400
N
,2
9 200
oo
1oo ? 8 0 E 6 0 9" 40
) \
4l
'.,
s
*
t *tS
08,
F\
\, \ \ $ \
40 60 80100
10 8 6
0.2
I 0.4 0.60.8
810
Figure L2.12 Unit hydrographwidth at 50 and 75 percent of peak flow' o, a, observedvalue of Wso. observedvalue of lfi5.
,lll::i
eGtp')/b
nr(;
(r2.20)
where a and b are shapeand scaleparameters.Hudlow and Clark presenta trialanderror solution to the leastsquares normal equations,using Newton's method, to developestirnatesof a and b. The application of Snyder's syntheticunithydrographmethod 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 the use of unit by hydrographs the region under study which havethe proper lag timerainfall durain tion ratio; that is, t, : ttf 5.5. If anotherrainfall durationis selected, variations C of and C, can be expected.
SCS Method
A methoddeveloped the Soil Conservation by Servicefor constructingsyntheticunit hydrographs basedon a dimensionless is 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
212
CHAPTER UNIT 12 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 Fig. 12.13 are listed in Table 12.4. The peak flow to for the hydrographis developed approximatingthe unit hydrographas a triangular by 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 shouldbe reduced(causing and 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 SCS OF DIMENSIONLESS UNIT HYDROGRAPH OF FIGURE 2.13 1
Q/Q, Q/Qp
(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
12.6 SYNTHETIC HYDROGRAPHS UNIT 213. The average is 0.6/",where/" is the time of concentration, lag 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":L1tpD
(r2.24)
The dimensionless unit hydrograph,Fig. 1.2.1.3,has point of inflection at approxia mately 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, it shouldnot exceed0.25t, or 0.I7t". pnce but for the 0.133r"hourunit hydrographis developed, unit hydrographs other durdtions can be developed using Shydrograph IUH procedures. or 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 Chapter4 in
or The lag from Eq. 12.27is adjustedfor imperviousness improved watercourses, or both, if the watershedis in an urban area.The multiple to be applied to thq lag is M : 1. P(6.8 X 103 + 3.4 x 104CN 4.3 x 10?CN, 2.2 x 108CN3) (12.28) where CN is the curve number for urbanized conditions, and P can be either the percentage that imperviousor the percentage the main watercourse is hydraulically of 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 nP _ 4 8 4 x 7 0 Y
9.5
att Qo:3J60 cfsoccurring :9hr 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: c 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 f s . 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 size for applicationof this method Gray.16 approximateupper limit of watershed 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\( \s1
\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 hydrograph unit 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 the unit hydrograph. of can be develof characteristics the watershed Correlationswith physiographic opedto get the valuesofboth Poand y'. parameAs 4qexample, storage the 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) + Z) : (N  I + z)(N 2 + z)... (t + Z)/f(r + Z), : f(N
where N equalsthe integer.""rl q, the function is approximated by
 #fu.
215
K 1 6
F re o ' E o
6
1.0
2.0
t PR
Figure 12.14 Dimensionlessgraph and fitted twoparameter gamma distribution for Watershed (After Gray.r6) 5.
='(ft)'
957o confidence belts
al> ": lH l0
e ttr x t9!
o l H
tlr
for Ppll'
Ttta
b
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 watershedsNebraska, in Iowa, Missouri, Illinois,andWisconsin. 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 CHAPTER UNITHYDROGRAPHS
\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
factor y' andperiodof risePn.(After P*f of Figure12.16 Relation storage Gray.r6; The solutionproceedsas follows: watershed' 1. DetermineL, 5", and A for the ungauged P*, Determineparameters T', and q. 2. Ppfy'. a. With Lf\/5", useFig. I2.I5 to select With P*/y', use Fig. 12.16 to obtain Pa. Compute y' as the ratio b. Pol@*ly'). Substitute7' obtained in Step 2b into the equation Q :  + y', and c. 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 substituting by 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.xAnizx640Tf nu'
I
n'rnln
x 43'560 ;
. ^ ft2
x V o : 2 q , x 0 . 2 5 x P 'R 6 0 
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^/4hr storm.
b. Convert the dimensionless graph ordinatesto the unithydrograph 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 solutionof Gray'smethodfor a Missouri the watershed. plot representing derivedhydrograph shownin Fig. l2.r7.Dkect A the is runoff commeirces the centroid of rainfall. Thus it is necessary add D/2 or Po/B at to 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
(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 2.Vo:0.25 x 24.9 60 X > q,:373.52 q' 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
where
L : total distance (ft) along the main channel from the point being boundary to considered the upstreamwatershed
12.6 SYNTHETC HYDROGRAPHS UNIT 219 : main channel slope (ftlft) defined by H/0.8L, where 11 is the S 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) = I4zrr width of the unit hydrographat75%oof p. (min) 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 onethirdof the calculated width placedto the left of the peak and twothirds to the right.
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 recognizes that the discharge any point in time is a function of The technique of the translation and storagecharacteristics the watershed.The translation is obtained by estimatingthe overland and channeltravel time of runoff, which is then effectsof a watershed. by combinedwith an estimateof the delaycaused the storage mouth rainfall from its point of falling to the watershed The translationof excess This is a histogramof is accomplished using the timearea curve for the watershed. lines . as incrementalrunoff versustime, constructed shownin Fig. 12.19 The dashed in Fig. I2.I9a subdividethe basin into severalareas.Each line identifiesthe locus of "times" are pointshavingequaltraveltimesto the outlpt. The isochrones drawnequal to apart, and sufficient zonesare selected fully definethe timearea relation. The timearea graph of Fig. l2.I9b is a form of unit hydrograph.The area to beneaththe curve integrates 1.0 unit of rain depth over the total areaA, and it has \ are a translationhydrographshapeif sufflcient subareas delineated. at If one unit of net rain is placedon the watershed t : 0, the runoff from At 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 outlet. the
Lt
I
I
J
^1
.l
A2
J
^3
Lt
A
As
Time interval
ft) of Figure L2.19 Development timearea histogramfor use with Clark's method: (a) isochronesspacedLt apart (shown as dashedlines) and (b) timearea histogram.
221
The impact of watershed storageon the translationhydrographis incorporated by routing the timearea histogramthrough a hypotheticallinear reservoirlocatedat the watershedoutlet, having a retardance coefficient K equivalentto that of the watershed. the simplest For form of reservoir,the storage at time f is linearly related S, 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 the lag time of the watershed. by From continuity,the inflow, storage, and.oq,{flow the reservoirare relatedby for
I,Q,:#:U#
Q'
(r2.36)
If the differentialis discretized LQ/LI, andif Q, andQ, are the flowsat t andt  l, to 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,
where
and
cr: #+T;
 Lt 2K "  ,t  z K + L t
The IUH is found from Eq. 12.38by solvingfor Q, at the end of eachsuccessive time interval. EXAMPLE 12.9 Given the following 15min timearea curve, find the IUH for the 1000acrewatershed.Then determinethe 15min syntheticunit hydrograph.The storagecoefficient K is 30 min.
Solution. From Eqs. 12.39and 12.40,Cs : 0.4 and Ct : 0.6.Routing most is easily accomplished a tableauas follows: in
222
CHAPTER12
UNIT HYDROGRAPHS
Time (hr) 0
I (acrein./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
due to the magnitudeof K for long recession The IUH has a characteristically hr, the flow becomes0.6 times the previous this example.Note that after 1.25 in to flow andcontinues decayat this rateindefinite$.As discussed Section11.5, time, which is release the excessrunoff the time baseof the IUH shouldequal Clark's method often producesproone definition of time of concentration. of longedrunoff because this shortcoming. The 15min unit hydrographis found using Eq. 12.6, ot
(h0
0 0.25 0.50 0.75 1.00 t.25 1.50 :
(efs)
0 160 576 1146 848 509 :
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 approximatelythe inflection point on the recession Fig. t2.2, when inflow to the of the estimate the Kvalue is the ratio If dataare available, channelceases. hydrograph at this particular point on the hyof the flow rate to the slope of the hydrograph drograph.
Nash'sSyntheticIUH
One of the earliest formulations of the IUH was developedby Nash.l8Instead of characteizingrunoff as translationfollowedby storage a singlelinear reservoiras in 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 time. Once the IUH is developed; can be used lag it to synthesizeany other hydrograph by application of the convolution integral, Eq. I2.7, or from the approximate methods discussed Section12.3. in The derivation of Nash's equationfor IUH beginsfrom continuity at the first reservoir:
(r2.4r)
KQr, at time
(r2.42)
eu:
Integrationfrom / : 0* to time t gives
'  kI o '
(r2.43)
(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
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, : nr nt/K
, (12.49)
e,,:
(12.50)
#1n1rt/K
rain falling by Because outflow from the nth reservoirwas caused 1 in. of excess the an Eq. instantaneously, 12.50describes IUH.
for EStimatiOn of K and n Valuesof K and n arc needed applicationof Nash's IUH. By integration,the centroid of the distribution (Eq. 12.50) occurs at t : nK.  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
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. time can be determinedfrom equationssuchas centroidal distance.Thus, if the lag reducing the number of those in Section 11.6, the product nK can be established, trials. to haveattemptedto relateNash'sK andn parameters basin Someinvestigators relations developed Rao et al.1e techniques. usingregression and storm characteristics for urban areasgreaterthan 5 mi2:
K :
and
Lt 
(r2.s3) (r2.s4)
is wherenK : tlandA is areain squaremiles,D is net rain duration in hours,Pn", the net (effective)rain depth in inches,and / is the ratio of imperviousto total area.
225
(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 steep, for flat, or sewered basins. an urbanareais fully sewered, calculated from Eq. 12.55 If the C, is decreased percent. If sparselysewered,a 10 percent increaseis made. If the 10 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. plotting ft., For the smallerof 35 percentor 0.670is placedleft of the peak.For W15,45 percentis placedto the left, or O.424To 0.67pwas usedfor W'r6. is the time from beginning if I of runoff of the unit rainfall to the peak time. Severalinvestigatorshave suggested that Snyder's C, and slope S are correlated.a'S original CUHP procedurewas alteredto recognizethis relation, making The the adjustments C, unnecessary. the modifiedversion,Eqs. 12.57and 12.58are in For bypassed, thetime to peakratherthanlag time is usedin Snyder's 11.5,where and Eq. 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
CHAPTER12
UNIT HYDROGRAPHS
0.18
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
Ec,.2l
20
Eq.3 80
100
40 60 Percentimpervious,l, (a)
12
Equations of curve: P = aI?+ blo+ c
10
Eq.
\t<
b 0 012
+216
06
zr ,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 (b) relationbetweenpeakingparameand ter 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 can The peakrate andtime of the unit hydrograph be developed 12.6t. After four additional points definedby Eqs. 12.59 and t2.60 areplotted, and the rest of the 5min hydrographcan be fitted to provide a total area representing 1.0 in. of direct runoff. A hand fit is applied, or the mathematical curvefitting 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
6P,)
02.63)
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 in procedures described Chapte 23,24, and25. Thesemodelsare as unit hydrograph and simply computerprogramsthat perform the unit hydrographsyntheses convoluthe in tion stepsdescribed this chapter.Any softwareuser shouldunderstand origin, hydrographmethod for applicability,and parameterestimationprocedures eachunit usesof the computermodelswill resultfrom a thorough The seiected. most successful in described this chapter. familiarity with the processes
PROBLEMS
a 12.1. Given the following storm pattern and assuming triangular unit hydrographfor one determinethe compositehydrograph. 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 1.S in. net rain in .t time unit for Time units 1 2 3 Flow (cfs) r00 98 220 4 4.5
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
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 singlepeaked runoffevent. Separate baseflow and deterthe mine 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 Shydrograph problem 12.6to find a 3 timeunit of 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 the unit hydrograph. of 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 a2hr pnrthydrograph the if 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
10
11
12
I P.M
800
cease? a. At what times did direct runoff begin and Determinethe @index (in'/hr) for the basin' b. ordinates(cfs)for eachtime.listed' c. Derive the 2hr'unitnydrograpir time) for the basin. release concJnffation(excess d. Estimatethe tirne of e.Atwhattimewoulddirectsurfacerunoffceaseiftherainfallof2.T5in./hrhad than 2? and begunat 9 .q.'l\a. had lastedfor 8 hr rather .ur" (cfs) and the direct runoff (in.) for a uniform urg" 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
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 cfshr' the c. Determine { indexfor the basin' basin by tabulating time in hours and d. Derive " , nt ltii ttyJrog'api' for the in discharge cfs. time of the basin? release e. What is the excess the direct tt" the derived2hr unit hydrograptrtodetermine f. For the ,"*"'t"'it, beganat 1 p.u. and ralnfall runoff rate (.fr) ;;; ;.*, on a day when excess(netj 't"i at 5 P'M' iti"tt ity of 2 in'/hr for 4 hr' ceasingabruptly "t " continued 2.6.A5hrunithydrographfora425}acre.basinisshownintheaccompanyingsketch. Thegivenhyd.og.uphactuallyappearedasadirectrunoffhydrographfromthebasin' of 5 hr,beginning
by caused rainf;ffi;;;;ii"'"iriv
att = 0. time of the basin' release a. Determinethe excess index for the basin' b. Determinethe @ contributingto direct runoff 4 hr after c. what n".""*"J" "'i,ti" JJr"g" t"r* was rain began(r = 4)? as shownin the sketch'Do not scale to d. Use your response part c to determineQp' Qp from the drawing' t = 3 andr : 5' Why did the hydrograph e. Note that rain continuedto fall between : 3 andt = 5, rathet than continue to rise during those form a ptateaut"i*""n t 2 hours?
230
CHAPTER12
UNIT HYDROGRAPHS
QO
e 40o
! :oo $ zoo
100
Time, t
f. Usethe given 5hr unit hydrograph determine direct runoff rate (cfs)at 7 p.r'r. to the on a day when rain fell at an intensity of 0.60 in./hr from I p.rrr. 11 p.vr. to 12.17. The 2hr unit hydrographfor a basin is given by the following tabie:
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 a2hr period and abruptly increases to a rate of 6.5 in./hr for a second2hr period. convert theseactual intensitiesto netrain intensities usinga {index of 0.5 in./hr. Constructa tablethatproper$ lags and amplifiesthe 2hr unit hydrograph, and determine hourly ordinates(cfs)of the direct runoff for the storm.The deriveddirect runoff hydrographshouldbegin and end with zero discharge values. 12.18. Given the following 2hr unit hydrograph for a drainage basin, determine hourly ordinatesof the 4hr unit hydrograph:
0 0
1 50
2 300
3 400
4 200
5 s0
6 0
12.19. Usethe following 4hr unit hydrograph a basinto determine peak discharge for the rate (cfs)resulting from a net rain of 3.0 in./hr for a 4hr duration foliowed immediately by 2.0 in.lhr for a 4hr duration.
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 for a > 1. Also solvefor the centroidaldistance taking the flrst moment aboutthe y axis. 12.23. Accordingto the rational method(seeChapter15) of estimatingpeakflow from small areasn peak rate for a storm with uniform continuingintensity is equal to the net the For what conditions,if any, would rain rate and occurs at the time of concentration. ' of Eqs. 12.63 and 12.17result in agreement the peak magnitudeand time, estimated by CUHP, with those of the rational method?Discuss. L2.24. Describetwo methodsthat could be usedto constructa 2hr unit hydrographusing a lhr unit hydrographfor a basin. rates (cfs)from a 2.48n12drainagebasin 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. 3 p.lrl.was a constant 100 cfs. to
Time Discharge(cfs)
8 e.u. 100
9 100
10 300
11 450
12 300
100
a. At what time did direct runoff begin? b. Determinethe grossand net rain depths(inches). c. Derive a 2hr unit hydrograph for the basin by tabulating time in hours and in discharge cubic feet per second. d. Derive a 4hr unit hydrographfor the basin. e. Derive a lhr unit hydrographfor the basin. 12,26. Given below is a 3hr unit hydrographfor a watershed.The {index is 1.5 in./hr. 3hr rainfall ratesof 2.5, Desiredis the DRH for an 18hr 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 2hr unit hydrographto determinethe peak directrunoffdischarge rate (cfs)resulting from a net rain of 2.0 in./hr for 5 hr'
0 0
1 50
2 200
3 300
4 200
5 1s0
6 100
7 0
12.28, The ordinatefor a 5hr 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 UNITHYDROGRAPHS 12 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 2hr 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 2fu 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 part a? in 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 12hrunit hydrograph, and convert it to a 6hr unit hydrograph.Then apply the 6hr unit hydrographto determinethe total hydrograph(including 400 cfs baseflow) for a24hr storm having four 6hr 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 triangularshaped 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 a 12.34. Which of the techniques synthesizing unit hydrographrequiresthe leastcompuWhich probably requiresthe the tationai effort in developing entire unit hydrograph? most?
REFERENCES
Division, 1948. 1. W. D. Mitchell, "Unit Hydrographsin Illinois," illinois Waterways "StreamFlowfrom Rainfall by the UnitGraphMethod," Eng.NewsRec. 2. L.K. Sherman, 108,501505(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. Characteristic Trans. Am. Geoplry Union 33' 235246(19 52). s," ASCE Trans.ll0,14191446(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, 19661970. 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 of Sanitary Engineeringand WaterResources, JohnsHopkins University,Baltimore, 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, "DiscussionFitting a Gamma Distribution over a Synthetic Unit HyBull. 19(2) (Apr. 1983). drograph,"WaterResources Engin' 13. "FloodHydrographAnalysisand Computations,"U.S. Army Corpsof Engtneets, D.C.: U.S. Government neering and Design Manuals,Emlll021405. Washington, PrintingOffice,Aug. 1959. "Hydrograph Synthesisby Digital Computer," Proc. 14. M. D. Hudlow and R. A. Clark, ASCEJ. Hyd. Div. (May 1969). in Characteristics SyntheticHydrographAnal15. V. Mockus, "Use of Storm and Watershed Service,1957. ysisandApplication,"U.S.Department Agriculture,Soil Conservation of for 16. D. M. Gray,"synthetic Unit Hydrographs Small DrainageAreas," Proc' ASCEJ. Hyd. Div. 87(HY4) (July 1961). for for 17. W. H. EspeyandD. G. Altman, "Nomographs TenminuteUnit Hydrographs Small Urban Witersheds,"EnvironmentalProtectionAgency,Rept.EPA600/978035,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 the hydrographs movedownstream through the systems. as . Distinguishbetweenthe two major classifications hydrograph of routing techniques. . Familiarizethe readerwith procedures determiningwhen to apply eachof for the variousrouting methods. water Flood forecasting, reservoirdesign,watershedsimulation,and comprehensive planning generallyutilize someform of routing technique.Routingis used resources a to predict the temporal and spatial variationsof a flood wave as it traverses river 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 dynamicsofflow than doesthe hydrologicrouting technique. the Applications of hydrologicrouting techniques problemsof flood prediction. to of evaluations flood control measures, assessments the effectsof urbanization of and 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 presented separate are in sectionsof this chapter.
235
Three of the most popular hydrologic river routing techniques described subseare in quent paragraphs,
Method Muskingum
Storagein a stableriver reachcan be expected dependprimarily on the discharge to of into and out of a reach and on hydraulic characteristics the channelsection.The as2 storagewithin the reach at a given time can be expressed
s:2;y7^n+(1  X)On1
a
(r3.2)
of at Constants andn reflectthe stagedischarge a characteristics control sections each characteristics the section. of end of the reach,and b and m mirror the sta$evolume 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 the amountof translationthat is, the time lagand by As by the amount of attenuationor reduction in peak discharge. can be noted from 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. i in change the river reachduringthe routing interval from Eq. 1 3 . 3s The storage
(13.s)
( 13.6)
and substitutingthis into Eq. 13.4 resultsin the Muskingum routing equation
in which
co
(r3.7)
(13.8)
KKX+0.54t
(13.e)
Note that K and Ar must havethe sametime units and aiso that the three coefficients sum to 1.0. if Theoretical stability of the nunrericalmethod is accomplished Al falls be X). The theoreticalvalue of K is the time required tweenthe limits 2KX and2K(I for an elemental(kinematic) waveto traversethe reach.It is approximatelythe titne interval betweeninflow and outflow peaks, if data are available.If not, the wave volocity as for velocity can be estimated variouschannelshapes a function of average uniform flow canbe estiinated for steady flow rate Q. Velocity Vfor any representative by either thi Manning or Chzyequation: The approximatewavevelocities for differare ent channel'shapes given in Table 13.1. FOR WAVE 13.1 KINEMATIC VELOCITIES VARIOUS TABLE '
CHANNELSHAPES Channelshape Wide rectangular Triangular Wide parabolic Manningequation
tV !v
lJrl 9 '
Ch6zyequation
3tl 2 l
tv Zrl
6 '
RIVER 13.1 HYDROLOGIC ROUTING 237 by Since1, and Irareknown for every time increment,routing is accomplished incrementsusing each02as Ol for the next time time solvingEq. 13.6for successive incremeni. Example 13.1 illustratesthis rowbyrow 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' Rowbyrowcomputation ll given in TabIe13.2. routing of Determination of Muskingum K andX Values K andX for Muskingum 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)
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
4l 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
t,823 3,2'72
4 114
26,9'70
)\ 117 )'t 171 t7 1))
29' 290
238
CHAPTER13
HYDROGRAPH ROUTING
XI + (l  X)O are flrst graphed severalselected for valuesof Xas shownin Example 13.2.Because andXI + (1  X)O are assumed be linearlyrelatedvia Eq. 13.3, S to the accepted value of X is that which gives the best linear plot (the narrowestloop). After plotting, the valuefor K is determined the reciprocalof the slopethrough the as narrowest loop, sincefrom Eq. 13.3
K :
n+0x)o
(1 .10) 3
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, + (l  X)O in Table 13.3. A preferablemethod would be to plot S, XI (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
(1O3cfsdays)
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
25 326 27 28 29 30 3l 4r 2
J
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
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
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 work with shorterreaches the river or to to of employ a more sophisticated approach. I I it Muskingum Crest Segment Routing Sometirnes is desirableto solve for a 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, + Ctlnt + 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
Ont : Colnt + ClIn2 * C2O,2
(r3.r2)
By repeatedsubstitutions the rightsideoutflow tetm On2,On2,. . . can eachbe for eliminated and On can be expressed a function only of the flrst n inflow rates or, as 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)
240
CHAPTER13
HYDROGRAPH ROUTING
Using data from Example 13.1 to find the outflow rate on 326, we obtain Kr : C6: 0.0477 Kz: CoC2 Cr : 0.0477(0.524) 0.428:0.4530 + + : Kz : KzCz: 0.4530(0.524) 0.2374 Krt: KrcG : 0.0013
Thus the outflow on 326 is calculated O11: 0.0477 X (47,g90)+ 0.4530 X as : (57,740)+ . . . + 0.0013(4260) 51,469cfs.
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 TR20, the SCS storm event simulation computer program describedin Chapter 24. Newer versions of TR20 usethe attkin methoddescribed Section13.3. in Analysis of Fig. 13.2 producesthe working equation for the convex routing method.Because areasunderboth curvesare equal,and because peak outflow the the
Figure 13.2 Geometricrelationsusedin the scs convexrouting method.(After U.S. Soil Conservation Service.3)
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: Ot + C,(11 O)
(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 Because is one limb of the triangle in the inset to Fig. 13.2, Lt _O" O, IrO, 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 w r_ L t v
(r3.r7)
routing method. is Determination of K and C, Proof that K from Fig. 13.2is a constant left to the by reader.It is a storageparameterwith time units and can be approximated the K twice the MuskingumX. from the Muskingummethod.Similarly, C, is approximately from Table 13.1) will provide The reach length divided by wavevelocity (estimated of another estimateof K, or actual measurements reach travel time can be used. by Equation 13.l'l canthenbe solvedfor C,. The valuerecommended the SCS,in the is of absence other estimates, C,:
(13.18)
steadydischarge,andV * 1.7 approxiwhere Vis the velocity for a representative (speed)of a kinematic wavetravelingthrough the reach.The units matesthe celerity (fps). of V in Eq. 13.18are feet per second The after the C, valueis estimated. accomplished Routingby Eq. 13.14is easily lessof the time to peak of the inflow hydrographto interval Ar shouldbe onefifth or assurea sufficient number of calculatedoutflow rates to deflne the hydrograph.As so with all routing methods,the time interval shouldbe selected that one point falls peak and other locationsof rapid change. at or near the
242
13 CHAPTER
ROUTING HYDROGRAPH
the Unlike otherrouting methods, convexmethodequationfor O, is independent be used to forecast outflow from a reach without of 1r. Thus the proceduie can tnowing the concurrent inflow. This provides a method for early calculation and to warninf for floods.Flow recorderscanbe linked through microprocessors warning flood potentials at least one full routingtime systemsthat calculate downstream interval aheadof the flood. to The procedurecan be reversed find the inflow hydrographfor a given outflow cumulativemasscurve of inflow to the reachinsteadof hydrograph,or it can route a itself. the hydrograPh
Method MuskingumCunge
been the Severalattemptsto overcome limitations of the Muskingummethodhavelot complexity or difficulties in physically totally ,u"""riful becauseof computational The Muskingum parametersare best derived interpretingthe routing parameteri.a's are not easilyrelatedto channelcharacteristics. and fiom streamflowmeasurements blendedthe accuracyof the diffusion wave method (seeSection 13.3) Cunge6 with the simplicity of the Muskingum method, resulting in one of the most recomgives as for mendedtechniques generaluse.It is classified a hydrologicmethod,yet it methods' resultscomparablewith hydraulic Cungeshowedthat ihe finitedifferenceform of the Muskingum equationbefor comesthe diffusion waveequationif the parameters both methodsare appropriI3'3, the Muskingumequationis atelyrelated'From Eqs. 13.1 and
K+ln+(lx)ol:to
dt
( 1 1e) 3.
(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,
(r3.22)
co
Lt/K  2X 2(rx)+Lt/K
(r3.23)
243
cr:
4 v2 
LtlK + 2x
2(rx)+Lt/K 2(Ix)cLt/Lx
" '
(r3.24) (r3.2s)
2(Ix)+Lt/K
./r .\ L Aal?
the K Because : A,xfc,it represents time for a waveto travelthe routing reachlength at velocity t. Congeshowsthat the velocity c is the celerity of a kinematic Ax, moving (Table 13'1). described wavepreviously translation X : 0.5 and c Lt/A,x: 1.0, the routing equationproduces When When Ax : 0 (zeroreachlength),no translationor attenuation without attenuation. occurs. by the If previousflood data are available, routing parameterc canbe extracted can also be obtained calculations.Estimatesof the parameters 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 4o : discharge unit width (cfs/fO,normally determined 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 m comesfrom the uniform flow equation coefficient Q: bA
(13.28)
o Q:
A
*g:
A
*y
(r3.2e)
Q*4:o
6x At
(13.30)
givesEq. I3.2t if c : mv.If dischargedata ate available,m canbe estimatedfrom are nq. tZig. Valuesfor commonshapechannels given in Table 13'1' 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 I3.2i rs solvedfor c, the valueX is derivedfrom Eq. 13.26,andE6' 13.23 to 13.25are solvedusing K : Lx/c. to When using this rnethod,the valuesof Ax and Ar shouldbe selected assure time to peak of inflow detailsare proper$ routed.Nominally' the that the flood wave is broken into 5 or 10 time incrementsAt. To give both temporal and spatialresolution, the total reachlength l, can be dividedinto severalincrementsof Ax length, and outflow from eachis treated as inflow to the next.
244
CHAPTER13
HYDROGRAPH ROUTING
EXAMPLE 13.3 Usethe MuskingumCungemethodto route the hydrographfrom Example13.2.Use areaat Q : 59,960is 5996 ft2, So: 0.0001,Lx : 545 mi, flow crosssectional : 59,960is 60 ft, and Ar : 1.0 day (asin Example13.2). width at Q Solution. From the inflow, the peak rate of 59,960cfs gives
So:T
o^
59.960
l0fps
":;1,
L* K :
't
7:
545(5280) ff: 172,800 sec C": 0.1765 Ct: 0'7647 Cz: 0'294I
and
Date, t
CoQ'*L*
Ct Qti*ro*
Cz Qlutno*
Qli*.*
r9r0 (3r7) 4440 6900 11,010 16,050 16,050 16p20 29,580 44,360 48,750 44,880 35,730 (329) 20,9s0
316 17 18 t9 20 2l 22
L3
24 25 26 27 328
 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 MuskingumCunge example. The value C, is alwayspositive,and negativevaluesof C, arenot particuin this condition should larly troublesome. Although C0is negative this example, valuesof Coare avoided be avoided practice.As seenfrom Eq. 13.23, negative in
when
^,
Other Methods
,* rr
(13.31)
includingthe working havebeendeveloped, river routing procedures Other hydrolSgic method. method,Tatummethod,and multiple storage R&D method,straddlestagger They all appear as options in HECI, the U.S. Army Corps of Engineer's event in simulation and routing model described Chapter24.
(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 Eq. 13.32 by Flow through a free outlet discharge where' I H C .r : : : : pipe (ft') the crosssectionalareaof 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 Storagevaluesfor variouspool elevations a reservoirare readily determined from topopool areasmeasured of computations volumesconfinedbetweenvarious graphic maps. Since storageand outflow both dependonly on pool elevation,the relation (Eq. 13.32) can curve and the outflowelevation resulting storageelivation graph. Storagein a reservoirdepends easily be combinedto form a storageoutflow on to only on the outflow, contrasted the dependence the inflow and outflow in river routing(Eq. 13.3). "surchargestorage"or the storage ,S For convenience, is often defined as the is spillwaycrest.Normally the overflowrate is zero when,S zero. abovethe emergency and ifthe slopeofthe line relation is found to be linear, Ifthe graphedstorageoutflow
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.lusingr: 0'0inEq. 13.3' showninFig. Muskingumriverrouting specialcaseof only while the inflow exceeds rate Note alsothat the,outflow in Fig. tg.t is increasing that assumptions the inflow immewith the is the outflow.This observation conJistent only andthat the outflow depends overthe entirepool surface diatelygoesinto storage on this storage. time by Routing through a linear reservoiris easily accomplished first dividing s2 : Ko2into Eq. I3'4 and into a numberof equal incrementsand then substituting increment' solving for oz, wtrictris the only remainingunknown for eachtime reservoir,the storageoutflow through anonlinear Ro'oa To route an emergency the outflow relation and the continiity equation,Eq. L3'4,are combinedto determine I3'4 canbe rewritten as and storageat the end of'eachtime inciement A/. Equation
( r , + r n +.1 *  o , )
:'+
t on+t
(r3.34)
side'Pairs in which the only unknown for any time incrementis the tgt_*:n the right that satisfy Eq. 13.34 and checked b" generated of trial valuesof S"*, andO,*1"ould "uru" fot confirmation.Rather than resort to this trial procein the storageoutflow outflow curve are replottedas and dure, a valrieof At is selected points on the storage indication" curve shownin Fig. 13.3' This graph allowsa direct determithe ';storage has been a nation of the outflow O,11 o11ce value of the ordinate Zl,*rfLt I On+r from Eq. f :.:4' ftre secondunknown' S,*t, can be read from the SO calculated 13.3)orfoundfromEq'13'34' curve(whichcouldilsobeplottedonthegraphinFig.
t200 I
/ / {,
: .il< 600
0 4 0 Outflow (cfs)
247
using of integration Eq. 13.34with Fig. 13.3is illustrated This rowbyrownumerical : t hr in ExampleI3.4. Ar
EXAMPLE 13.4
inflow hydrographand the 2S/Lt + O curve ofFig. 13.3 Giventhe tria4gularshaped find the outflow hydrograph for the reservoir assumingit to be completely full at the of beginning the storm.(SeeTable 13.4.)
o 90
o Time (hr)
In selecting routing period Ar, generallyat leastfive points on the rising limb a number of of the inflow hydrographare employedin the calculations.An increased the accuracy,sinceas A/  0 points on the rising limb, that is, a small A/, improves the the numerical integrationapproaches true limit of the function being integrated, 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 problem the reservoir is assumedfull at the commencement inflow. Therefore, there is no availablestorage.
TABLE 13.4 ROUTINGTABLE (1) Time (hr) 0 I z 3
+
(2)
(3)
ln
'no.
(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)
(cfshr)
5 6 '7 8 9 t0 ll t2
1 2 3 4 5 6 7 8 9 10 ll 1 t
5 18
3Z AA
52
)6
780 979
63 65 65 64 62 58 54
12.5 46 96 164 250 361 458 506 510 467 404 344 288
248
ROUTING 13 CHAPTER HYDROGRAPH 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 endoftimeinterval o, of 5 cfs, which is recordedin column 7 , The corresponding to columns6 and7 and recordedin column 8. Moving s,, storage, is calculatedfrom n : 2 usings' be found fof 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
tions. Hydraulicfoutingtechniquesarehelpfulinsolvingriverroutingproblems,oversolution the simultaneous land flow, or sheetnow. uyaraulic routing proceedsfrom 1'negeneralforms of the combinationfor of of expression, continoiti unao*"ntuir, fl'ow equations' rivers are called the spatically varied unsteady
13.3 HYDRAULIC ROUTING 249 RIVER 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 presentation, generalform of the spatially varied of a unsteady flow equations will be presented first.
Equationof Continuity
The equation of continuity statesthat inflow minus outflow equals the changein storage. relatethis conceptto a river sectionundera condition ofrainfall or lateral To inflow, consideran elementof length Ax and unit width into the page as shown in Fig. 13.4. The total inflow is
+ o(vt*+)(,.#?) o'
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
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 vr " 6 x v  6 x At :
( 13 .3e)
e d{x + v d x * *d tY
whereA is the width times the depth,y.
l : 0
(13.40)
Momentum Equation
In accordance of with Newton's second of motion, the change momentumper unit law of time on a body is equal to the resultantof all externalforces acting on the body. The following derivationof the momentumequationof spatiallyvariedunsteadyflow (1) is presentedsubjectto the following assumptions: the flow is unidirectional and is velocity uniform acrossthe flow section;(2) the pressure hydrostatic;(3) the slope of the river bottom is relatively small; (4) the Manning formula may be used to evaluate friction lossdueto shearat the channelwall; (5) lateral inflow entersthe the stream with no velocity componentin the direction of flow; and (6) the value of I represents spatial and time variationsof lateral inflow. the Forcesactingon an elementof lengthAx andunit width are shownin Fig. i 3.4b. The forcesF1andF, represent as hydrostaticforceson the elementand are expressed
f,  atvA) F,: ylyA ;zl axl   .f^ a(la)Axl rz: lL,!^A Z l
(13.41)
(r3.42)
wherey is the distancefrom the watersurface the centroidofthe area.The resultant to 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 )
251
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)
(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(ss)
x A
(13.50)
( 1 3 . s) 1
that expressions can be Equations13.50and 13.51form a set of simultaneouS solvedfor V and y subjectto the appropriateboundary conditions.
Ev
VAV
IAV
(r3.s2)
Friction slope
Bed slope
Convective acceleration
Temporai accelelation
252
are flow routing in open channels classifiedas of The three typesof analysis unsteady depending kinematic,diffusion (also called noninertia), anddynamicwaveanalyses, differ not only by on which terms in F;q. 13.52 are retained. The three techniques regarding flow including different terms of Eq. 13.52 but also in the assumptions 13.5 showstheseassumpconditions for satisfyingthe momentum equation.Table tions.
HYDRAULIC USED 13.5 ASSUMPTIONS IN VARIOUS 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 (stagedischarge but nonuniform flow has constantdischarge varying water surfaceslopesuchas that 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 inbetween(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 is approxi(nated assumingsteadyuniform flow, and routing is accomplished combiningthe continuity equationwith any form of friction lossequation.Typically, either the Manning equationor Ch6zyequationis used.The Chdzyequationis
V : C\/RS
( 13.s 3)
( 13.s 4)
where C andn arefriction coefficients,S is the friction slope,and R is the hydraulic if radius (areadividedby wettedperimeter).Both give velocity in feet per second area and wetted qerimeterare input using squarefeet and feet units. *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)
is which, after taking derivatives, 6Q = 6*4*t AA Multiplying Eq. 13.55by dQ/dA gives
: ma : M V A
(r3.57)
, ,a Q mv dx
+ 9t : o a
( 13.s 8)
. dQ * 9 : o  x
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 that c is constantand assume Most applications rating curve,which varieswith stage.
9 * , 9 : dd{x9 dx At
(13.61)
accounts for the The lefthand 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. (resultingin small d) is why kinematiowaveanalysis valid whenbed slopesare steep the channells extremelywide (resultingin small 4)' For flat bed slopes, or when the hydraulic diffusion coefflcientis particularly important' Numerical solutionsof Eq. 13.6I ate presentedby severalinvestigators'12ra Thesenormally involve substitutionof the relation
ao
A
 B3 v +
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 in which the conveyance is Q/\/Srfromeither equation.Diffusion wavesapply to a wider range of problems than kinematic formulations, but their use may not be it warrantedbecause requiresabout the sameeffort as dynamic routing. for accounts all terms inBq.13.52, includingthe The third type of waveanalysis "dynamic" or It nonuniform,unsteady, and inertia components. is referredto as the "full dynamic" formulation. Dynamic wave solutionsare far more complicatedbut for are often necessary analysisof flow along very flat slopes,flow into large reservoirs, highly unsteadydambreakflood waves,or reversing(e.g.,tidal) flows. These on conditions are often encountered coastalplains. As a generalrule, full dynamic 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 )
SCSAttKin TR20Method
method(Section13.1)with themodifiedattkin the In 1983,the SCSreplaced convex (attematronkinematic)methodas the agency'spreferredchannelrouting method.ls The 1.964SCS TR20 (Chapter 24) singleeventsimulation model used the convex modified to route by the attkin method. method but was subsequently indication and kinematicwavemethods. The procedureis a blend of the storage Figure 13.5 showsthe twostep processof simulating attenuationfirst by meansof
o PP
RIVER r3.3 HYDRAULIC ROUTING 255 routing and then translatingthe wavein time by the kinematic wavemethod storage in to accountfor the fact that routed flow ratesnot only decrease magnitudebut also require time to traversethe length of the routing reach.The storagerouting portion providesattenuationwith instantaneous translation,and the kinematic waverouting providestranslationand distortionbut doesnot attenuate peak. Both are needed the 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 interval 12.This samevolume would translateand distort downstream kinematic actibn, flowing out of the reachduring interval /r. and selectionof routing coefficients,the Through its theoretical development attkin methodequationssatisfythe physicalpropagationand timing of the peak flow (areasunder the three hydrographs of rate first. Conservation massis also assured of 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 spondingto the maximum storagein the reachduring passage the flood. Becausecelerity changeswith storage,the other flows of the storagerouted hydrographare translated,pachby a different celerity,to their respectivefinal times and values. by The storageindication routing is accomplished substitutionof the relation 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 crosssectional areathroughoutL is relativelyconstant,the ofrouiing reach,andifthe crosssectional 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 Input to the methodrequiresselectionof a reachlength and estimates b andm for 13.1and Eq. 13.27,m canbe shownto be for usein Eq. 13.59.As discussed Table velocity (under bankfull conditions)with wave celerity, or a factor relating average
(13.70)
256
CHAPTER13
ROUTING HYDROGRAPH
the The larger ru becomes, shorterthe travel time, A value of m I 1.0 would incorby flow velocity.Studies SCSresulted rectly makethe celerity slowerthan the average Signif,canterrors resultedfor m values in a recommendationof ! for general use. greater than2.0. Equation 13.67 is appropriatefor cross sectionshaving a single but ihannel with regulaishape.Complexcrosssectionsare more diffcult to evaluate, table for the stream'rs la valuescan be developid from a rating attenuationof the peak flow increasesdue to As the coefficient b decreases, by in storage the reach.The value b canbe estimated reducedvelocity and increased fitting the linear form of Eq. 13.67 (referlo plotting Q andA on loglogpaper and faatei6.+1. The slope*oUa be m andtheinterceptut 4_: 1 would be b. The SCS for nomographs estimatingb and m'" 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
= minimum acceptablereach length = minimum recommended reach length MainTime Increment Hours
mV ft/sec
0.5 0.7
2
I
0.5
n?5
0.3
0.2'
3 /
5 7 10 T2 15
0.1 0.05
t00 reach length for Figure 13.6 SCS nomographfor determining atlkin method of routing. (After U.S. Soil ConservationService, "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 a reachlength between accumulate'Thus routed, this incrementattime^errorwould but a length greater thal c. A/ is recommended' c Lt andc Lt/2 is acceptable' Figure13.6providesth"rangeofminimumacc"eptableandminimumrecommended routing reachlengths.
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 a1d hydraulic d e p t e r 2 4 ) , p r o v i lplied severalchoices.In I the models can be found in the routing procedures' suggestions g]]:i,:Ytditg .T" At 1iterarure.,6,17 sel ln The readerwho will engage rlver or when eachof the methodsshouldbe applied' reservoirroutingisencouragedtodwelopthecomparisonrequestedinProblem13.25 before leavingthis chaPter'
PROBLEMS
techniques' betweenhydrologic and hydraulic routing 13.1. Discussthe main differences 13.2.
t+}i ;q":: ;;:..ii+l q 191 "'*ifi:'i I :il# 1iT .'SX,ffff ffi?#t r" :, Iit:t?;;i,Y Yllli 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;
v\v p:.'i"l' the or time *u,'::9,":1i: 4_ of the Siif#ffi"#ru:=#;"i"ei*inetlme oerro.; i'w 12' v2' .xg 9z lT"jll outflow, and storage at the beginning ;;: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
258
ROUTING, 13 CHAPTER HYDROGRAPH 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
Assumethat the outflow hydrographat a section 3mi 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 discharges both endsof.a30mi river reach: 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
Time June1417
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
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
Noon 6 p.tvt. Midnlght 6 e.u. Noon
emergencyspillway of a certain 13.10. The outflow rate (cts) and storage(cfshr) for an units of hours'Use
: Sl3,wherethe number3 has reservoirare linearly t"fui"a Uy d = I tt + sr  or Ltl2to determine the continuity equationss2 + 02 Ltlz this and the following inflow event: itr" p""f. outflow rad fr;m the reservoir for Time (h0 0
L
I (cfs)
o
(cfs)
(cfshr)
4 6 8
260
CHAPTER13 HYDROGRAPH ROUTING ' 13.11. A simple reservoir has a linear storageindication curve defined by the equation
" : 2* .
whereAr is equalto 10 hr. If s at 8 e.u. is 0 cfshr,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 verticalwalled 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: Verticalwalled 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 acreft and cfsdays, determinethe valuesfor reservoirstorage corresponding S 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 dischargestcirage curve (cfs versuscfsdays)and the storageindication 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 500ftwide ideal emergency spillway (C = 3.0) locatedat the 1160ft elevation.Storageareaelevation 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
261
4.0
A <
table. a.Findthel5minunithydrographbySnyder'smethod.
the 30 and 1 of for nrst min 063 l i"li:l! ff;ru#iHfllSiii.?om 8in. rain
.. il"i:i.,i[T;iitf;,ersus
curvesprovloeo' outflow and storage to the bottom it the reservoirassuming is full e. Route,n" uJ#f,iu.;;;n";;"dh elevation980' of the sPillwaY height of water^inthe reservor' Indicate maximum f. plu'"d to obtain 5 ft of freeboard? "f th;;;;;; g. At what "tJ;;;;iJ'n"'op Total storage
104(ft3)
Elevation (ft)
zto
590 1240
the reservoirinitially empty' 13.16. RepeatProblem 13'15 with l3,IT.AfloodhydrographistoberoutedtytheMuskingummethodthroughalOmireach roml 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 HYDROGRAPH 13 ROUTING 13.18. RepeatProblem 13.6aby dividing the 3mi reach into two subreaches with equal K valuesof 5.5 hr. Comparethe results. 13.19. Discussthe problemsassociated with the useof a reservoirrouting techniquesuchas the storageindication method in routing a flood through a river reach. 13.20. VerifyEq. 13.51. t3.21. Precipitationbeganat noon on June14 and caused flood hydrograph a stream.As a in the storm passed,the following streamflow data at cross sectionsA and B were obtained:
Time June1417
6 .q,.N{. Noon 6 p.u. Midnight 6 .r.lr. Noon 6 p.lr. Midnight 6 a.Ira. Noon 6 p.rra. Midnight
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 SectionA (continuecomputations at 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)
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 discharges both endsof a 30mi river reach: 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 andptot tt ree outflow hydrographs valuesofX equalto the computed 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 section3mi 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 b. Plot the inflow and outflow hydrographs a singlegraph' : 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 retained and consultedfrequently' into a list or table, ihis taUleshouldbe
264
REFERENCES
1. Graeff, "Trait6 d'hydraulique." Paris,1883,pp. 438443. 2. Y. T. Chow, Open ChannelHydraulics. New York: McGrawHill , 1959. 3. U. S. Soil ConservationServiie, National EngineeringHandbook, Notice NEH 4102. 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 9 5 1 . , "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, 20s230(1967). "KinematicFlood Routing,"Trans.ASCE L0(3) (1967). 7 . D. L. Brakensiek, "Water Studies,"Bureauof ReclamationManual, Yol. 8. U. S. Departmentof the Interior, D.C.: U. S. GovernmentPrinting Offrce, 1941. IV Sec. 6.10. Washington, "Numerical Techniques Spatially VariedUnsteadyFlow," University for 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, Res. for WaterRouting,"WaterResources AGU II(2) (Apr. 1975). "WavePropagation Rivers," HEL Ser. 8' No. 1. in 12. J. A. Harder and L. V. Armacost, Hydraulic Engineering Laboratory, College of Engineering,University of California, 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" DiSilvio, "Flood Wave Modiflcation Along Channels," Proc. ASCE J. Hyd. Div. 14. G. 9s(HY7) (1e69). "Simplifled DamBreach 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 DamBreak 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 . Indicatethe importanceof snowmeltto water supplyand management cold regions. . Describemethodsfor measuringsnowfall and describingits waterproducing 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 suppliin the high Llevations the ColoradoRockiesis derivedfrom snowfall.' water high proportions are also likely in the Sierrasof California and numerous 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 understand natureand distributionof snowfallandthe mechanisms snowmeltprocess. Snowmelt usually beginsin the spring. The runoff derived is normally out of such phase with the periodsof gieatestwaterneed;therefore,variouscontrol schemes to reseivoirshavJ been developed minimize this problem. An additional as storage is point of significanc.e that $omeof the greatestfloods result from combinedlargeon rainstormsand snowmelt.Streamflowforecastingis highly dependent adeicale of knowledgeof the extentand characteristics snowfleldswithin the watershed. 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 the rate of snowmelt.Early resultscan be be obtainedby speeding melt process' by or the snowmeltperiod can be extended delayed retarding it. The annual whereas snowfall distribution in the United Statesis shownin Fig. 14'1'
&
v /
S:.
I O
) \
,4
00 co
I @
()
!1 a !? o
E
q
{)^ > j
5E
p P
.=.Y
o
t
tu5
267
factorsis as much a prerequisite of understanding meteorological An adequate The in consideringthe snowmeltpro""rs as it is in dealingwith evapotranspiration. of water vapor on suppliesmoisturefor both snowfall and condensation atmosphere of the snowpack,regulatesthe exchange energy within a watershed,and is a controlling factor in snowmeltrates. and geologic, topographic, vegetaAs in the rainfallrunoffprocess, geographic, runoff process. tive factors also are operativein the snow accumulationsnowmelt For rainfallrunoff relations, point rainfall measuresare used in estimating areal and time distributions over the basin. A similar approachis taken in snow Mathematical hydrologyalthoughthe pointarealrelationsare usuallymore cornplex. of snowmeltat a given equationscan be usedto determinethe various components of location. Adequatemeasures depth and other snowpackproperties can also be in obtainedat specificlocations.The use of thesemeasurements estimatingamount , 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 basinwidehydrologicestimates'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.215 listed at the end of this chapter. approaches
268
14 CHAPTER SNOW 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 are described Horton.t6 it by 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 samepull of the snow lessgravity. The meltwaterdirectly entersthe soil the 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 infiltration capacity. the A slush layer forms and water infiltrates the soil at the infiltration capacity rate. Excesswater acts in a manner analogous surfacerunoff but at a much decreased to overlandflow rate. As warm weathercontinues, melt process maintainedand accelerated the is until the snowcover dissipated. is
269 CHABACTERISTICS SNOW AND 14.4 POINT AREAL 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 to aging,the densityof snow increases valueson the order of 50 percentor grPater.  A snowcourseincludesa seriesof samplinglocations,normally not fewer than about50 100ft apartin a geometric 10 in number.ttThe variousstationsare spaced in pattern designed advance.Points are permanentlymarked so that the samelocations will be surveyedeachyearvery 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 and private enterprises also ensnow surveys.Various states,federal agencies, in of gagedin this type of activity. Sources snowsurvey dataaresummarized Ref. 14.
(14.1) (r4.2)
or where
D t b 
Pu:
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
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 not uniformly distributed,weighting coefficients a good approximation.For stations used of basedon the percentage the basin areaportrayedby a gaugeare sometimes group. in determiningN, for the method.In Another system usedin estimatingareal snowfallis the isopercental as precipitation is expressed a percentage this approach,the storm or annual station on of the normal annualtotal. Isopercentallines are drawn and can be superimposed the (NAP) to producenew isohyetsrepresenting a normal annualprecipitation map storm of interest.A NAP map indicatesthe generalnature of the basin'stopographic from this pattern.The advanthe map shows deviations effects,while the isopercental map directly is that relatively consistageof this methodover preparingan isohyetal tent storm pattern featuresof the NAP can be taken into considerationas well as observedindividual storm variations.
Areal Snowcover
Estimatesof the areal distribution of snowfall are very helpful in making hydrologic on A forecasts. knowledgeof actualarealextentof snowcover the groundat any given and in making seasonalvolumetric time is also*applied in hydrograph synthesis are of forecasts ofthe runoff. Observations snowcover generallyobtainedby ground photography. Between snowcoversurveys,approximaand and air reconnaissance data. hydrometeorological are tions of the extentof the snowcover basedon available patternswithin a given basin are normally relatively uniform Snowcoverdepletion from data gathered indexes often be developed can from yearto year; thus snowcover stations. at a few representative
27'I
EnergyBudgetConsiderations
as If snowmeltis considered a heat transferprocess,an energybudgetequationcan of the be written to determine heatequivalent the snowmelt.Suchan equationis of the form2 H^: H,r + H," + H" + H" + Hs + He + Hq (14.3)
where H*: the heat equivalentof snowmelt H4 : rrlt longwave 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(gcal/cm2).Therefore,203.2langleysarc neededto get 1 in. of runoff from a snowpackof 100 percentthermal quality. If the the term H. represents combinedtotal heat input in langleys,an equationfor snowmelt 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; exceeds for ripe snowpacks 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 M : 0.75in. rl
Turbulent Exchange is The quantity of heat transferredto a snowpackby convectionand condensation has Suchan approach been equations. from turbulentexchange commonlydetermined mustbe madein and vapor pressure of widely used,sincemeasurements temperature and wind velocity gradithe turbulent zonewherevertical watervapor, temperature, several ents are controlled by the action of eddies.In the following two subsections and practical equationsfor estimatingcondensation convectionmelt are given. Here to a combinedtheoreticalequationis presented acquaintthe readerwith the theory of 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 horizontal area vapor through a unit dqldT = the vertical gradient of the property
e : the property z : lhe elevation A: an exchange coefficient Property 4 must be unaffectedby the vertical transport.Propertiespertinent to this discussion the air temperature, are water vapor, and wind velocity.Theoretically,the potential temperatureshouldbe used,but air temperatures measured normal disat of 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 neutral or unstableatmospheric conditions.The power of 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 law exponent the
to If 3, is made equal to I, q, assumed be the property value at this height, and the subscript droppedfor the second level,Eq. 14.6becomes
q : qtz\/"
(r4.7)
at The magnitu of q is takenasthe differencein valuesof 4 measured the level de is For example,if T : 38'F at height z, and temperature the 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)zod/"
(14.8)
(r4.e)
from Thus,eddyexchange elevation is determined ofthe property at a specified e 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
related the prope"t r.:O*r,i to inversely  )n,Orr, dqld, h result: the in Substitution Eq. 14.8for dq/dz gives following
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 Substituting 41 in Eq. I4.I2 gives
a: \;)a'"'
Using the power law equation(14.7), we find that q, : q;1/"
and D, : 1171/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. specifictheoreticalequationsfor condensation developing melt. The property to be tlansported consider caseof the condensation the First, the coefficientexpresses transfer in this caseis water vapor, and sincethe exchange it to air maqs, is necessary determinethe moisturecontentof the air mass.This of an by can be accomplished using the specifichumidity, which gives the weight of the containedin a unit weight of moist air. Equation 14.18 can be usedto water vapor calculatespecifichumidity: q : 0.622e (14.18)
where e : the vapor pressure of . Po: the total pressure the moist air
PROCESS 275 14.5 THESNOWMELT Inserting this expression in Eq. 14.17 for q"yields2
(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 height.The constantmay be madeindependent density,and thereforeof elevation, "introducing a factor to compensatedirectly for the densityelevationrelation' by this, and the equationcan be adjustedby to serves accomplish Atmosphericpressure muttipiying Uy tne ratio pf po, wherepis the pressureat the snowfieldelevationand tevetpr"rrur". Introducingthis ^ratioinBq. 14.20and a new constantk1, po is the sea gives' which is related to sealevel pressure,
{r4.2r)
importanceinEq. 14.17is airtemperature. Forconvectionmelt,thepropertyof into ttreimat units, the specific heat of air crmust be To convert air temperature H" introduced. Putting these values in Eq. I4.l7,heat transfer by eddy exchange convertsto2
(14.22)
cgs Sincethe latent heatof fusion is 80 callg, convectivesnowmeltM" in gramsin the is system givenby H"l8O, or / 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 developed Light.le Widely used,its individual convectionand condensation in componentsare discussed following sections.The combined form of the Light equationis
D :
where
pk',
80 rn(alz) ln(b/zo)
+ ulc,r ( e  6.IDryf 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 levelsat which the wind velocity,temperature, 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 to Heat for snowmeltis transferredfrom the atmosphere the snowpack convection. processis relatedto temperatureand wind velocity. The amountof snowmeltby this The following equationcan be usedto estimatethe 6hr depth of snowmeltin inches 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 coefficient of 0.00184 X 100'0000156'has valuefor the exchange theoretical in given by Light.le In this relation h, the elevation feet, is usedto reflect the change is in barometric pressuredue to the difference in altitude. The expression said to of representconditions for an open, level snowfieldwhere measurements wind and abovethe snow.Values are temperature madeat heightsof 50 and 10 ft, respectively, vary of the exprtssion1000000156' from 1.0 at sea level to 0.70 at 10,000ft of elevation.The actual valuesof K are normally lessthan the theoreticalflgure due to such factors as forest cover. Empirical 6hr K values have been reported in the literature.20 due for Anderson and Crawfordls give an expression the houily sno.wmelt to 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 at 4 ft, with wind gaugedat 15 ft. The corresponding are valueof c is reportedas 0.0002.
Condensation
snow.A total yield of around 8.5 in. of snowmeltincluding the condensate thus is derived. A water vapor supply at the snow surfaceis formed by the turbulent exchange process; consquently, masstransferequationsimilar to thosepresented evapoa for ration studiesfits the melt process. equationfor hourly snowmeltfrom condensaAn tion takesthe formrs
u : W n { r "6 . 1 1 )
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 4 and 15 ft, respectively.ls at A similar expression for 6hr snowmelt(D) is given as but
D:KrV(eo6.11)
(14.30)
where the theoretical value of K, is said by Light to equal 0.00579 if wind and
(14.3r)
where E : the hourly evaporationin inches es : the saturationvapor pressure over the snow k : an empirical constant
278
HYDROLOGY 14 CHAPTER SNOW 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 longwave very important source of heat energy for snowmelt. Under clear skies, the most in significantvariables radiationmelt are insolation,reflectivity or albedoof the snow, and air temperature.Humidity effects, while existent, are usually not important. in When cloud cover exists,striking changes the amount of radiation from an open snowfield are in evidence.The general nature of these effects is illustrated in as radiation exchange a function of cloud .2 Fig. 14.3 Combinedshort and longwave 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 longwave in Forest canopiesalso exhibit important characteristics regulating radiative from thoseexhibitedby the cloudcover, Theseeffectsdiffer somewhat heatexchange. Cloudsand treesboth limit insoradiation is concerned. especiallywhereshortwave insolation lation, but cloudsarevery reflective,while a largeamountof the intercepted the by is absorbed the forest.Consequently, forestis warmedand part of the incident radiation; an additional energydirectly transferredto snow irl the form of longwave 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 northfacingslopes for inclinesin the northern hemisphere, example. is lessthan that for southexposure Solar energy provides an important sourceof heat for snowmelt. Above the the earth's atmosphere, thermal equivalentof solarradiation normal to the radiation 3.97 x 103 Btu/cm2).The path is 1.97 lingleys/min (1 langleyis approximately actual amount of radiation reachingthe snowpackis modified by many factorssuch and ofcloudiness,topography, vegetalcover.The importanceofvegetal asthe degree has cover in influencingsnowmelt,long recognized, promptedmany forest managesnowmelt.t'to'"''o to ment schemes regulate 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 1010langley/(minK')] T : the temperature(K)
(r4.32)
crurar,.ier,iiTth
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
r.)
^ r _ t "n"n t : t lr I
> 10
\
.= rt \
M,= M^
Mrt
4
\M,
o.5
0.0 0.2 0.4 .'
(a)
0.5
(l) ,*^=l,r*
I
H o.o
0.5
'1
Mf
Mr, + Mr1
11.0
0.0
Mrt
? 'qTF r
n 7s7Q.  3.30
0.6 0.8
0.4
(b)
melt in the forestwith clear  Figure 14.4 Daily radiation skies:(a) duringspring,May 20; and (b) duringwinter, 15. February (AfterU.S.Army Corpsof Engineers.2) Becausesnow radiates as a blackbody, the amount of radiation is related to its temperature(Planck's law), and total energyradiated is accordingto Stefan'slaw. Longwaveradiation by a snowpackis determinedin a complexfashionthrough the interactionsof temperature,forest cover, and cloud conditions'
SNOWMELT PROCESS 281' 14.5 THE Direct solar shortwave radiation receivedat the snow surfaceis not all transferredto sensible heat.Part of the radiationis reflectedand thus lost for melt purposes. Shortwavereflection is known as albedo and ranges from about 40 percent for melting snow late in the season approximately80 percent for newly fallen snow. to This Valuesas high as 90 percenthavealsobeenrepofted in severalcases.22 property 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 shortwave radiation not reflectedand availablefor snowmelt may becomelongwave radiation or be conductedwithin the snowpack.Some heat may also be absorbed the ground with no resultantmelt if the ground is frozen. by An expression hourly shortwave for 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, longwave and its surroundings.Snowmelt from net positive longwave radiation follows Eq. 14.33.If the net longwaveradiation is negative (back radiation), there is art equivalentheat loss from the snowpack. An approximate methodof estimating12hr 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 halfday in clear weather (0 m : thedegree cloudiness for clearweather,1.0 for completeovercast) of Suggested valuesfor Do are 0.35 in. (March), 0.42 in. (April), 0.48 in. (May), and 0.53 in. (June)within latitudes 4048".2o
Rainfall
Heat derivedfrom rainfall is generallysmall, sinceduring thoseperiodswhenrainfall of occurson a snowpack, temperature the rain is probablyquite low. Nevertheless, the 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 T : the webbulb temperatureassumed be that of the rain 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 one degree. by where
282
(r4.36)
where Md : the daily snowmelt(in.) , P d : the daily rainfall (in.) air To : the meandaily air temperature('F) of saturated taken at the 10ft
level23
Conduction
are of Major sources heatenergyto the snowpack radiation, convection,and condensation. 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 entire melt season, ground conduction. 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 by ground conductioncan be expressed the relation2 dT Hn: Kwhere 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 ground to the snow is insignificant.20 amountof snowmeltfrom ground conduchas been estimatedat approxirnately0.02 in.lday'23 tlon during a snowmelt season Groundconductiondoesact to providemoistureto the soil; thus,whenotherfavorable of conditionsfor snowmeltoccur, a more rapid development runoff can be expected. physicsof snowmelt.The mannerin which heat This sectionhasemphasizedthe I4.3I and 14.27 Equations was to canbe provided initiatethemelt process discussed. L4.33I4.37 inclusivecan be usedto estimatethe melt at a given point. The task of in computingrunoff from snowmeltin a basin cannot be approached sucha simple 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 ft above sea lel'el at a latitude of 44" N: air a ripe snowpacklocated at 10,000 temperature50' F; mean wind velocity, 10 mph; relative humidity, 65Vo;avenge reading,48oF. Estimate rainfall intensity,0.03 in./hr for I2hr; wet bulb psychrometer radiation, and warm rain for convection,condensation, the snowmeltin in. of water for the 12 hr period.
April 23 100 80 9 6 0
Eo
<2
Apnl24
Aprii 25
1600 2000 2400 0400 0800 1200 1600 2000 2400 0400 0800 1200
 Shortwaveradiation Longwave radiation ffi
nm
lt ii
l,Incident shortwave radiation (t)
G *
40
Absorbedshortwave radiation
Downwad
H . n E z v
l 
20
=
o
_40
E E o.o4 On?
.,
6
E 0.02
(Each bs reDresents mem value
E o.ot
0
17001800hr 2400*0100hr' 09001000Itr
no condensation
no condensation 
no convgctiotr or 
meltllcondensationmeltmelt t t t t l E Radiation melt (Mr) Convrction condensation 6 mett (Mce)  Radiationmelt total (Hourlyconvectioncotrde$ation is melt added of subtacbdftom totalhouly to radiation to arnveattotalhouly melt
computed melt) I Totalcomputedmelt
_o*
0 0.05 Apil23lNet 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
(14.27)
(14.30)
(r4.34)
( 14.3s )
Condition Snowpack
The mannerin which runoff from eitherrainfall or snowmeltis affectedby conditions Variousviewson prevalentwithin the snowpack of primary interestto a hydrologist. is of storagecharacteristics a snowpackhave been advanced.These range from the
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 important to baseany runoff considerations a knowledgeof the character becomes condition, at of a snowpack the time of study.Winter runoff is relatedto a snowpack's in whereas the spring,onceactivemelt begins,little or no delayin the transportof melt or rainfall through the snowpackoccurs. For drainagebasins in mountainousareas,snowpackstorageeffects may be into relatively uniform areas.Normally, by approximated subdividingthe watershed by this will be accomplished usingelevationzones.Snowpackat the lowestlevelsmay a be conditionedto transmit readily rain or meltwater,whereasin higher elevations be very the liquid water deficit may prevail. At uppermostelevations, snowpackmay dry and cold and thus in a condition for the optimxm storageof water. The storage of measurements the 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 pack.In this case,storage relateddirectly to a by fashioned assuming homogeneous liquid water defici1and cold content of the pack. Time delay is a function of the the inflow rate. It is consideredthat the snowpackstoragepotential must be entfuely permits but In beforerunoff begins. reality this is not the case, the assumption satisfied snowpack an analysisto be made. As melt proceeds,the storagepotential of any diminishes. to is beforerunoff commences considered be the sum of of Storage a snowpack to of waterrequilementto raisethe temperature the snowpack 0'C (cold the equivalent If capacityof the snowpack. the cold content contentlV")and the HqJidwaterholding to temperature 0oC,it may be is given in inchesof water neededto bring a snowpack by2 represented
w"T" *r=ffi
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 to temperature OoC thus given by to The time L in hoursneeded raisethe snowpack , "  r 6 o 1 i + * 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.)
( where ".1 : the amount of water stored(in.) T _ the percent deficiencyin liquid water of the snowpack JP
(14.40)
286
CHAPTER14
SNOW HYDROLOGY
(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 sinceit is not available the runoff until This is also known as "permanent" storage, componenttransitory storage has the snowpack finally melted.An additional storage while movingthroughit to becomerunoff. Until S,is that waterstoredin the snowpack as initiation of runoff, the transitory storagein inchescan be expressed
^ __T__ ",: D ( i + m )
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 for /, in hours.Assumingthat I4z" very small comparedwith Wo,the depth of the snowpackis given by
O:%
P'
(r4.4s)
Then of with p" the density the snowpack. wo (r4.46) nv in the Beforethe runoff commences, total waterS storedin the snowpack, inches, givenby is 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)
term the runofffromthesnowpack, onlysignificant the After establishing active with the overallbasinlag and is in Eq. 14.49 /,, andthis is usuallysmallcompared
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 balance a snowpack Figure 14.6 Water can be neglected.With increasedsnowmelt and runoff, additional increments of water prevlouslywithheld by snowblockageto drainageoutlets and other factorsare A at Adequatequantifi;ation of this cannotbe accomplished present'24 deep released. 5oC, could storeabout 4 in. of ft, havinga meantemperatureof 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 Will condensation the snowpackoccur, basedon the vapor pressure? is the cold content of one sq ft of surfacearea of the snowpack? what 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
288
: Temperature 22oF: 5.6oC in. W": 0.20x l2Ox 5.6/160: 0.84 is its is d. The snowpack not ripe since temperature belowfreezing. I I
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 measurements watershedvalues,indexescan be used to record both areal and temporal aspectsof basin values. Indexes serve to permit (1) readily obtainable cannotbe which themselves to observations depict hydrologicvariablesor processes and (2) simplificationof computationalmethodsby allowing indieasily measured, valuesin time and to or vidual observations groupsof observations replacewatershed of space.The adequacy an index is basedon (1) the ability ofthe index to describe it the , adequately physicalprocess represents(2) the randomvariability of the obseris (3) the degreeto which the point observation typical of actual conditions, vation, and basin means.2 (4) the nature of variability beiweenthe point measurement and Indexesmay be equationsor simplecoefficients,and variable or constant. thermal budgetstudiesare The types of data requiredto make comprehensive ones.As or in normally unavailable wh^ole part for wut"tJh"dt otherthan experimental a result, a hydrologistmust make the best use of information at hand. The most humidity, commonly availabledata aredaily maximum and minimum temperatures, of measurements thesedata, andfew and wind velocity.Lessprevalentare continuous datacan Hourly cloudiness stationsrecord solarradiation or the durationof sunshine. be sometimes obtainedfrom local airport weatherstations. A completelygeneralindex for reliably describingsnowmeltrunoff 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 indexesthat havebeen used successfully snowmeltinvestigations. VARIABLES BUDGET THERMAL USED INDEXES TODESCRIBE 14,1 SOME TABLE
component Thermal budget
radiatron Absorbed shortwave Longwaveradiation' Convectiveheat exchange Heat of condensation Index Duration of sunshinedata Diurnal temperaturerange Air temperaturefor heavy forestedareas radiation shouldbe estimated For open areaslongwave (7"  T)V, where f is air temperature, T6 the snow surface temperatureor basetemperature,and Vthe wind speed of (e"  e")V,whereeoand e, are vapor pressures air and snow surfaceot a basevalue, and V is wind speed
" Figure 14.7 iilustrates an approximatelinear relation betweenmelt and longwaveradiation usedby the U.S, Army Corps of Engineers for index purposes.
289
Y:a+>b$t
where I : a : b, : 4 : the snowmeltrunoff aregression constant the regression coefficients individual indexes
( 14.s1)
Variousindexes usableto represent termsof Eq. 14.51are selected a standard the and regression performedto determine andb,.It shouldbe notedthat everyterm analysis a 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
Longwave radiation melt in forest or with low overcast sky in open Ma= 1ZL n:{r'1 ,, toTf, oT!) {referto Kelvin scale)
E 'c o
n v.+ /
1L O
r l t l M, = longwaveradiationmelt, 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
1 A
10
20
30 40 60 50 Meanair temperature, Zo("F) 270 2'75 280 285 Absolutetemperature, (K) Z1 290 300
260
265
Figure 14.7 Longwave radiation melt, with linear approximation. (After U.S. Army Corps of Engineers.2)
290
t.6
dataof Equationderivedfrom combined 1954 1955 and (T G RO = 0.00238 + 0.0245 ^ '77) 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 "yx
ri
5 0.6
0.4
,/,
aa,
'24.
0.2
/ a o
0.2 a .
1.0
1.2
Figure 14.8 Observedversusestimatedrunoff for (X) snowmelt 1954 and (O) 1955. RO = the daily generated runoff (in.) depth over a snowcoveredarea; G : the by daily net allwaveradiation absorbed snow in the open (langleys); 7u* : 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 devekiped the Corps of Engineers2 the partly forestedBoiseRiver basin above by Twin Springs,Idaho, was
(r4.s2)
area where Q = the daily snowmeltrunoff (in.) over the snowcovered in G = an estimatedvalue of the daily allwaveradiation exchange the 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 suitableindexes snowmelt,a hydrologist In attemptingto develop the approachmost closely resemblingthe thermal budget of the area, within the limitations of availabledata.
Temperature Indexes
The atmospheric temperature an extremelyusefulparameterin snowmeltdetermiis nation. 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 degreedayrelations in snowmeltcomputations. I A degreeday is defrnedas a deviation of 1ofrom a given datum temperature consistentlyover a 24hr period. In snowmeltcomputations, referencetemperathe ture 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 the24hr period, there will be24 degree for eachdegree hr departure above32'F. In this examplethere would be 264 degree for the day of observation. hr Variouswaysof estimatingthe meantemperature haveenabledinvestigators to take severalapproaches. methodis simplyto average maximumandminimum one the daily temperatures. Basesother than 32"F are also used.Regardless the particular of attackemployed, degree a hour or degree is an index to the amountof heatpresent day for snowmeltor other purposesand has proved useful in pointsnowmeltand 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 degreeday factorthe ratio of runoff or snowmeltto accumulated degreedaysthat producedthe runoff or melt.
2.8
1 A
; 1 A o
B
6 rz
34
38
42
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
and between factor has been found to vary seasonally Unfortunately,the degreeday values should be used with caution. Pointbasins;therefore, single representative rangefrom 0.015to 0.20 in. per degree basins factorsfor snowcovered degreeday 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 degreeday and snowcover at meltingtemperatures. in./degreedayunderconditionsof continuous for index equations springtimesnowmeltfor clear 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) rainfree When rain is falling, heattransferby convectionand periods and(2) periodsof rain.2a radiation is condensation of prime importance.Solarradiationis slight, andlongwave rainfree periods When can readily be determinedfrom theoretical considerations. prevail,both solarand terrestrialradiationbecomesignificantand may require direct evaluation. Convection and condensationare usually less critical during rainless intervals.The equationsare summarizedas follows:2 1. Equations periodswith rainfall. for For open (coverbelow 10 percent)or partly forested(coverfrom 10 to a. 60 percent)watersheds,
+ M : (0.029+ 0.0084fto 0.007P)(7" 32) + 0.09 (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 10ft level ("F) wind velocity at 50ft level (mph) the average t ^.  the basin constant,which includesforest and topographic of exposure the areato average effects,and represents from about 1.0 for clear wind. Valuesof k decrease plains areasto about 0.2 for denseforests ,, Equationsfor rainfree periods. a. For heavyforestedareas, , M:0'074(0.537'"+ O.47fi)
(14.ss)
(14'56)
+ 0.7870+ 0.0297" M : k(0.0084u)(0'22r'" areas, c. For partly forested M : k'(I  rx0.00401,)(l a) r) + 0.78r) + F(0'029 + k(0.0084o)(0.227'"
Q4.57)
)
14.6 SNOWMELT RUNOFF DETERMINATIONS 293 d. For open areas, M : k'(0.00s084)(1 a) + (1 'N)(0.0zIzT:  0.84) + N(0.02gr') + k(0.00S ao)Q.227'" 0.78Ti) + (14.58)
where M, a, k : as previouslydescribed T'" : the difference betweenthe t0ft air and the snow surface("F) temperatures T'o: the difference between the 10ft dewpoint and ('F) snowsurface temperatures I, : the observedor estimatedinsolation (langleys) a : the observed estimated or meansnowsurface albedo k' : the basin shortwaveradiation melt factor (varies , from 0.9 to 1.1),which is relatedto mean exposure open areascomparedto an unshielded of horizontal surface F : the mean basin forestcanopy cover (decimal fraction) Ti = the difference between the cloudbase and (oF) snowsurf,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
A , M : (0.029+ 0.0084ftrrA.007P)(7, 32) + 0.09 + M _ (0.029 0.0084 0.5 x 20 + 0.007x 0.3)(42 32) + 0.09 + x
I'tl 
b . M = (0<074 0.007P)(7" 32) + 0.05 + M * (0.074 0.007 0.3x53 32) + 0.05 + x M _ 1.65 in./day r:
I.24 in.lday
294
CHAPTER14
SNOW HYDROIOGY
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,1P" 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 the changein availablesoil moisture 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. 14.6214.64 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 z^v ^ o
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
./
The expression  (W,  Wr1represents snowmeltM; thercforc, P", the R:P,,+MQ"^QrL" (14.67) If reliable estimates the terms in Eq. 14.67 canbe secured,the basin discharge of R is computable.
ElevationBand 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 each on band during a prescribedinterval of time; and usethe weightedsum of thesecontributionsto provide an estimateof the excess wateravailable runoff. For eachband, for it is assumedthat snowmelt, rainfall, and lossesare uniform over the band. The subareas considered be either snowcovered snowfreeand melting or not are to or
296
melting. For eachband, snowmeltis computedusing equationsof the type preSented or basedon expectations historic information, and losses earlier,rainfall is estimated havebeen in are estimatedas described Chapterb3 through 5. Ohce theseestimates made for each band, the following equation servesto provide a weighted value of excess water availablefor runoff from the basin.
)
M _
)a,
where : snowmelt water availablefor runoff (cmlday), Pr is the rainfall on the 'band, M 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 15 of Table I4.2, estimatethe amount of excess method(Eq. 14.68). usingthe elevationband available runoff from the watershed for 14,5 14.2 DATA FOR EXAMPLE TABLE
(1) Elevation band no.
I
(2)
Subarea slze sq km
(3)
Snowmelt cmid
(5)
(6)
(7)
( 3 )+ ( 4 ) + ( 5 )
(6) x (2)
z
J A
5 6 Totals
Solution. The solution for the numeratoris the sum bf the productsgiven in is Table 14.2;the solutionfor the denominator the sum column 7 of spreadsheet of the subareas siven in column 2 of the table. : for The excesswater availabie runoff : 3269.6511316 2.5I cmlday'
II
Recessions Hydrograph
in curveshavebeen discussed Chapter 11 and take the generalform Recession Q: ' where Q : the dischargeat time t Qo : the initial rate of flow k:__afecessionconstant Qoeo'
(r4.6e)
, : 
297
o u
Time
hydrograph' of Figure 14.11 Separation a snowmelt permit evaluationof the amountof runoff Studiesof daily streamflowby hydrographs of the derived from snowmelt.The t""hniqutnred is essentiallyone of separation illustratesthe proceFigure 14.11(not to scaleandoversimplified) daily hydrographs. fit peaks,respectively, snowmdlt dure. Assumeihat the first, second,and succeeding in time, at a point A the backward days.If the ultimate recessioncurve is extended from recessions curveft'om Hydrograph2 will intersectit. The areabetween recession is the melt attributedto Day Hydrograph1 and HydiogtipiZ (showncrosshatched) their can 1. In like unn"r, a seriesofsnowmelt hydrographs be studiedto determine featuresas th height to By individual melt components. observingsuchhydrograph 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 snowhydrologyareoftwotypes'The ofrunoffhydrographsassociated syntheses of The secondkind is the development flow distribution first is a shorttermforecist. or a ' for a comPletemelt season forecastingis very helpful in prel of controls,while the synthesis part lating designfloods. To forecasta snowfieldand streamflowneedbe k parameters to ing, it is necessary havethe reliablepredictionof variousmeteorological can initial conditions.Known historicparameters be in additionto a knowledgeof the satparameters or flows whereasassumed generated usedfor reconstructinghistoric Figure 14.12 displayssomecommon hydrometeorologic isfy designflood syntlieses. data. severalfactors(not of great conceln where In snowmelthydrographsyntheses, First, a drainagebasin with snow"utlfully considered. only rainfall exists)mustbe system,since the areal extent of the as a homogeneous cover cannot be accepted
298
CHAPTER14
SNOW HYDROLOGY
6 5 ^ ! 9 > > v
B 5
(h
Shortwaveradiationngtesj.^, .
i hsolation obsened by USWB. of Boise Cily' lD 2. Basin albedo of snow surface ^  = ,  L Absorbed shonwa\e ralialion compukd b) lomula 1ab. ll(
AbsorbedShortwavefadiation
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" entiredrainageonly that portion withsnowcover' from bare areaswhile o""o., during the snowcoverperiod, contributionscan come may in may producecombinedrunoff' The natureof losses suchcases other expanses differ greatly for nonsnowoverlayedand coveredlocations' subThe altitude is an exceedinglypertinent factor in the hydrology of tracts due to a generalreducjected to snowfall.Ratesof .no*tnitid"crease with elevation tion in temperaturewith height. orographic effects and the temperatureelevation snowcover relations tend to raise the amount of precipitation with altitude' Greater melt rates' As a result' precipitation and reduced of depth occursbecause increased is apthe basinwide melt and coverarea increase with height as the snowline
14.6 SNOWMELTRUNOFFDETERMINATIONS
299
proached,then diminish with elevationover the higher placesnormally completely snowcovereduntil late in the season. snowpack A exhibitsanotherimportant trait in relation to rainstorms.In the spring, relatively little runoff occurs from snowfree 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 first dividesthe basin into a seriesof elevationzoneswheie The the snowdepth,precipitationlosses, melt are assumed and uniform. A second method considers watershedas a unit, so adjustmentsare made to accountfor the areal the extentof the snowcover,varying melt rates,precipitation, and other factors. To synthesize snowmelthydrograph,information on the precipitation losses, a snowmelt, and time distribution of the runoff are needed.Snowmelt is generally estimatedby index methodsfor forecasting,but in design flood synthesis heat the budget approach, the most used. Precipitation is determinedfrom gaugingsand is historicor generated data.Losses definid in two wayswheresnowmeltis involved. are For rainonsnowhydrographs the water is considered lossif delayedvery long all a 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, deeppercolation,or or permanentlyretained in the snowpackis consideredto be lost. Assessing time the distribution of runoff from snowcovered areasis commonly done with unit hydrographsor storagerouting techniques. rainonsnow events,normal rainfalltype For unit hydrographs applied;for the distribution of strictly snowmeltexcess, are special longtailedunit graphsare employed.Storage routing techniques widely exeicised are 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. errorsin estimates baseflow arenot generallyof any Big of 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, is customaryto it separate surfaceand subsurface the componentsand route them independently. Storagerouting has beenusedextensively routing floodsthrough reservoirs for or river reaches. is also applicablein preparingrunoff hydrographs. snowmelt It In runoff estimates, rainfall and meltwaterare treatedasinputs to be routedthrough the the basin, using storagetimes selectedfrom the hydrologic characteristics the of watershed. Two basichydrologicrouting approaches relatedto the assumption are of ( 1) reservoirtypestorage (2) storage or 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 additional system An usesa multiple storage,
300
CHAPTER14
,:
Rn
o 6 U
il
Time (hr)
storagerouting' Figure 14.13 Example of multiplestagereservoirtype routreservoirtypestorage iti. ngur" illustratesih" ut" of multiplestorage to runoff in u ttrunn". analogous unit ing for"evaluatingtime distribution of tiOrog.uptr. (Af1erU.S. Army Corpsof Engineers'2)
is routed throrrgh two or In reservoirtype storagescheduling.2 this method inflow 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' use of singlestage that the. clark has suggested drologic characteristics. to irp", in time permits Jo=mputations be^^simplified.27 rfter translatirrg has snowmeltrunoff hydrographs method for synthesizing e most practicJd graphsdiffers primarily in ,,nir hrrrrrnoranh The characterof snowmeltunit 12' that of rainstorm unit plots' As disclssed in chapter time base length fiom storm events'In single isolated rainstorm unit hydrograph, ort"n are derived from snowmeltrunoff,ratesofwaterexcessaresmallandapproximate$continuous'Asa is result, the use of Shydrographs indicated'2 utility, since it allows (1) adiustri"titoA has considerable The Shydrogrupfr 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 veragingseveralhydrographs get a unit
312
HYDROLOGY AND 15 CHAPTER URBAN SMALLWATERSHED havehad wide application;however, of Both categories peakflow determination in two relatively major difficulties are normally encountered applyingthe techniques. First, the rainfallrunoff formulas,suchas the rational formula, aie difficult to apply unless the return periods for rainfall and runoff are assumedto be equal. Also, of estimates coefficientsrequiredby theseformulas are subjectiveand havereceived 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 empiriccorrelativQ and severalof the the urban designscene, ilue to their simplicity,persistin dominating most popular forms are briefly describedto acquaintthe reader with methodsand are assuption*. Urban runoff simulationtechniques describedin Chapter25.
RationalFormula
The rational formula for estimatingpeak runoff rates was introducedin the United Sincethen it has becomethe most widely used by States Emil Kuichlingin 1889.18 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 C _ the runoff coefficient(assumed be dimensionless) 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 cI : the average rain intensity (in./hr) for a storm with duratiofl: t, 1.0 because acrein./hr is to The runoff coefficientcan be assumed be dimensionless for stormsof 510year return periods 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 of the watershed contributingto the outflow at the point of design.That condition is met after the elapsedtime t", the time of concentration,which usually is taken as At the time for a waveto flow from the most remotepart of the watershed. this time, the runoff rate matchesthe net rain rate. Figure 15.1 graphically illustrates the relation. The IDF curve is the rainfall intensitydurationfrequencyrelation 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 FOR 15.2 PEAKFLOWFOHMULAS URBAN TABLE15,1 ryPICALC COEFFICIENTS 5FOR TO 1OYEAR FREQUENCY DESIGN
Descriptionof area Business Downtown areas ' Neighborhoodareas Residential Singlefamily 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,27Vo Steep,TVo Lawns; heavy soil: Flat,2Vo Average,27Vo Steep,TVo Flunoffcoefficients
0.700.95 0.500.70 0.300.50 0.400.60 0.600.75 0.250.40 0.500.70 0.500.80 0.600.90 0.100.25 0.200.35 0.200.40 0.100.30 0.700.95 0.800.95 0.700.85 0.750.85 0.750.9s 0.050.10 0.100.15 0.150.20 0.130.17 o.r80.22 0.250.35
Time (min)
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, resultingintensityl from IDF curvesis too the small, and the rational flow rate p is underestimated. Rational Method Applications Most applications the rationalformulain deterof mining peak flow rates utilize the following steps:, 1. Estimatethe time of concentrationof the drainagearea. 2. Estimate runoff coefficient, the Table 15.1. 3. Selecta return period T, and find the intensity of rain that will be equaled or exceeded, the average, once every I years.To produceequilibrium on 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 shortenedstorm duration. the One of the principal assumptions the rational method is that the predicted of 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
130
Time (min)
prediction.Another assumption, onethat hasreceivedclosescrutinyby investigaand is tors,re'2o the constancyof the runoff coefficient during the progressof individual from a list stormsand also from storm to storm. The"coefficientis usually selected infiltration capacityof the drainagesurand basedon the degreeof imperviousness face. BecauseC : I,.rf I,the coefficientmust vary if it is to accountfor antecedent moisture,nonuniform rainfall, and the numerousconditions that causeabstractions rainfalls. In practice,a composite,weightedaverand attenuationof floodproducing agerunoff coefficientis computedfor the various surfaceconditions.Times of conof centration are determinedfrom the hydraulic characteristics the principal flow into two parts, overland flow and flow in defined path, which typically is divided channels;the times of flow in eachsegmentare addedto obtain /". Another assumptionwith the rational method is that the equation is most moisture conditions that exist for frequent storms,in the applicableto antecedent of interval,representative stormstraditionally used rangeof the 2 to 10yrrecurrence Becausemore severe,less frequent for design of residential storm drain systems. moisture conditions,the rational coefficient is stormsoften have wetter antecedent by increased multiplying it by a frequencyfactor.The commonly usedmultipliers for lessfrequent stormsare:
period (yrs) Return
Multiplier
2to 25
)U
100
EXAMPLE 15.I runoff ratesfor the area design and 5Oyear to Usethe rational.method find the 10year shownin Fig. 15.2arc applicable. 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 =
316
2. Runoff coefflcient: c : [(3 x 0.3) + (4 x 0.7)]lQ + 4) :0.53 for 10yrevent C : 1.2(0.53): 0.64 for 50yr event 3. Rainfall intensityfrom Fig. 15.2: Irc : 4'2 in'/ht : 1so 5'3 in'/hr 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 dependent the soil type, antecedent use, slope, amount of urban development,rainfall intensity, surface and channel of and roughness, durationof storm.Tablesand graphsgenerallyallow determination 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 j 2(0.001CN1 tso't{(P+ 48)0 e1s0 I)/zfo C : j .Z(t1i)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 a The rational formula is a simplemodelto express complexhydrologicsystem. the by resultsimplying acceptance designmethodcontinuesto be usedin practicewith results. ers, officials, and the public. The methodis easyto apply and givesconsistent in the methoddemonstrates clearterms From the standpointof planning,for example, times of because increases surfaces runoff from developed the effectsof development: increase. and runoff coefficients concentrationdecrease is the For storm drainagesystems, designer normally askedto estimatethe peak at flow rate that might be equalledor exceeded leastonce in a given numberof years using the rational (described the frequency see Section 10.4). For designs as equal to the frequencyof is assumed formula, the frequencyof the peak runoff event the rain event(an eventbeing deflnedas somerain depthin a given duration).Studies Figure 15.4 showscumulativelognormal probabilhaveexploredthis assumption.z2 of ity functions (Chapter 26) fitted to observations rainfall and runoff on a 47acre of imperviousness 0.44. The Surface area in Baltimore, Maryland, with an average and rainfall sequence the the observed to data are partial seriesfitted independently to correspond runoff doesnot necessarily Thus the largest runoff sequence. observed betweenany runoff lack of correspondence the largestranked rainfall, and a similar
s:
./
!2 k
..t
tt
't'2'
" i/
oo
y"
o
o'
2
a a
a o
cd
5 1020
Recurrenceinterval (Years)
in and the rainfall that producedit holds for the ranked position of the observations yearrainfall frequencyof In the arraysof the two i"putut" sequences. Fig. 15.4, the 5 a the of4'0 cfs/acre; ratio indicates runoff io 6.5in./hr corresponds a runofffrequency eachcloselylogare coefficientof 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 in that the rainfall distributions Figs. 15.1and 15.4havesimilarproperties. IDF from many different 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 intensitydurationfrequencyrelation. by to The most critical (highestpeak) runoff eventis often assumed be caused 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 testbecause assumption 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 generally warranted.Considerabl" are 3udg"nt is riquireAin 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, compositeweighted a C factoris computedfor the drainage areaaboveeachpoint. The time of concentration is composed an inlet time (the overlandand any channelflow timesto the first inlet) of 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 flow full foi this to calculation.(see Fig. 15.6.) Nomographs also are available solvethe Manning to 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 1020 min is more uiual.
d o
10n
""
4U
Fo F
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
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
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 not the summationof the individual dischapges, is 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 continuously. shouldbe noted that the designat eachpoint downstream 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. referencesto illustrate the application of the rational duced from standarddesign method to an urban sYstem.2a EXAMPLE 15.2 of Basedon the storm sewerarrangement Fig. 15.7a,determinethe outfall discharge. Assume that C : 0.3 for residentialareasand C = 0.6 for businesstracts. Use a a 5yearfrequencyrainfall from Fig. 15.7b andassume minimum 20min inlet time. 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 and for subsequent in drawingfinal designplans.(see Table 15.3.) lI 'orational" that thepeak in Modified Rational Method Therationalmethodis truly flow rate is simply set equal to the net rain rate after sufflcient time occurs for the to entire watershed contributerunoff. This resultsfor any storm equallingor exceed
1 4 5 6 7 8 9 10 t1 T2 t3 14 15 t6 t7
o)
I
u\q
_ 90lt
t
Legend
l
numbers Sewerlinewith manhole Storminlets Gutterflowline
MH11 r ,\e / 
\
a 5
N
\
J
2:iyr averagefrequencY
l l r l ! l l
\ {
10yr averagefrequencY
r
Ij'ff:::i::::?
K:t.
10 20
30
40
50 60 70 80 Duration(min)
90
Figure 15.7 Sample storm drainage problem: (a) typical storm rainfall seier design plan anO (U) intensitydurationfrequency for Davenport, Iowa. (After Ref' 24.) 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
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 adols U to (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
s n + sq q E q 3 N qE 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!
+ 3 3 : 3 : $ .3i J A 3 A A+ : : :
n q c ' i ' 1  ' 1 n
C Sq
O + N d N C ] +
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 FORMULAS URBAN FLOW 15,2 PEAK a In the modified rational method., full hydrographis developqdrather than the peak flow rate, using the following reasoning.If the storm simply estimating duration exceedslhetime of concentration,the runoff rate would dse to the rational At formula peakvalue,then stayconstantuntil net rain ceases. that point, runoffrates from the basin.Ifthe rainfallexcess rain is released to zero as excess would decrease releasetime (seeChapter 11) is equal to the time of concentration,the hydrograph rising to the peak al t : t",femaining flat trapezoldshape would havean approximate D until/: therainduration,D,andthenfallingalongastraightlineuntilt:I t". packagesfor urban hydrology incorporate the modified rational Many software method for hydrographanalysis.the method is approximateand shouldnot be apover 50 acresin size. plied to watersheds
SCSTR55Method
for procedures estimatingrunoff volServicedeveloped The U.S. Soil Conservation They are known collectively as from urban areas.zs ume and peak ratei of discharge TR55und indiuidually as thegraphical method,chart method, andtabular method. to in The threemethodsadjustrural procedures NEH426 urbanconditionsby increasing the curve number CN foi impervious areas and reducing the lag time /1 for and imlperviousness channel improvements.Allowances are also made for various the The SCS designed first two slopes,and times of concentration. shapes, watershed complete for synthesizing methodsto be usedfor estimatingpeak flows,and the third up (usedfor small watersheds to The tabular method and chart method hydrographs. to help explainthe evolution here are wererevisedin 1986,21but described 2000 acres) for of the methods.All three were developed use with 24hr storms'Use with other storm durationsis not advised. up watersheds, to 20 mr2 for The graphicalmethodwasdeveloped homogeneous by represented the runoff curve in size, on which the land use and soil type may be number. As shownin Chapter4, the runoff curve number is simply a third variable 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, TypeII storm distributions expeof the 24hr thunderstorm Chapter f 6l. A TypeII storm distributionis typical from rienced in all staiei exceptthe Pacific Coaststates.Figure 15.8 was developed in modeldescribed Chapter ofine SCSTR20 eventsimulation numerousapplications time of concentrationin hours is enteredinto 24. To apply Fig. 15.8, the watershed per rate in cfs/mi2of watershed inch of net ttre grapir'to prJdu." the peak discharge from the 24hr gross rain during tk Z+nr period. The 24hr net rain is estimated in amountusingthe scS curve number approachdescribed chapter 4. Fig. 15.9.Oncethe composite using the effectof urbanizationcanbe estimated for curvenumber(CN) hasbeenestimated the previousarea,a modifiedcurvenumber by is determined enteringFig. 15.9 with the value of the percentimperviousareaon to r"ading vertically to the curve corresponding the CN for the modified watershed, to determinethe modified the pervious watershed,and then reading horizontally compositerunoff curve numberthat would be usedin determiningthe net rain depth for the urbanizedwatershed' of Useof the 1975graphicalmethodis restrictedby the assumptions the tabular caseof the tabularmethod.This method.The methodii a ompositeof resultsfor one
324
CHAPTER15
URBANNNO SUNUWATERSHEDHYDROLOGY
o P,O R
E o
200
100 0.1
0.2
2.5
(cfs/mi2lin.) runoff versus Figure 15.8 Peakdischarge of time of con(AfterU.S.SoilConsercentration for 24hr,TypeIIstorm /. distribution. vation Service.25) restricts its applicationsto runoff volumes greaterthan about 1.5 in. (if the curve numberis lessthan 60). Time of concentration shouldrangebetween and 2.0hr, 0.1 and the initial abstractionshould not exceedabout 25 percent of the precipitation. The chart method allows determinationof peak flows for 24hr TypeII storms over watersheds having a fixed length/width relation and no ponding areas.Three chartsareusedfor flat, moderate, steep or slopes approximately of I,4, or 16percent. Tablesof adjustmentsfor intermediateslopesare provided in the technicalrelease. Severalmicrocomputersoftwarepackages urbanhydrologyhavebeendevelfor oped.28 Over twothirdsarebased SCSprocedures, cautionshouldbe applied on but
80
L) e .E
7n
u 6 0
l0
20
30
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 FOR FORMULAS URBAN FLOW 15,2 PEAK that the commercialprogramsfully imitate TR55 or other SCS handin assuming would carry would includeall three methods, An book methods. ideal TR55 package would incorpoand would stateall assumptions limitations, and SCS endorsement, of percentage for rate all SiS adjustments peak coefflcient,percentimperviousness, and slope. length/widthratio variations, channelimproved,pondingor swampyareas, Its use shoutAako be cautionedfor other than 24hr stormshaving a TypeII SCS not adheringto these limitations would not be qualified as distribution. Packages TR55 procedures. is A significantproblem in someof the commercial softwarepackages the use for producehydrographs of a trianlularshaped unit hydrographfor convolutionto to conceptualizethe stormsofvarious durations.The SCS used a triangular shape use but peak flow rate of a curvilinear unit hydrograph, has never endorsed of other tabulatedhydrg; in than either the curvilinear shapediscussed Section I2.5 ot the graphsgiven in the TR55 manual.For further reading,the SCS publisheda guide2e in for the useof the 1975TR55 intendedto clarify procedures the original technical release.
SCSTR55Method Prevailing
for is ratherthan the 1975version, recommended use. The 1986editionof TR55,27 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. TR55 computerProgram.
Tablet5.4 andFig.15'10 Ratherthanrelyingtotally on Fig 15.9,the newTR55uses instancesindicated in the table' to provide urban runoff curve numbersfor certain For the new graphicalmethod,an urban curvenumberand the 24hr designrain from the SCSrunoff 1, ih"n un initial abstraction is determined depth are estimated, peak flow is found from linear interpoequation(Chapter4) or from Table 15.5.The on lation of the curvesin Figs. 15.ll,15.I2, !5.13, or 15.I4, depending the rainfall ratio falls outsidethe curves,the distributiontype (Fig. rcn).If the computedI"f P nearestcurvi should be used. If the watershedcontains a pefcentageof ponds or swampyareas,the peak flow is multiplied by a reductioncoefficientfrom Table 15'6' EXAMPLE 15.3 concentration,CN: 75 A 1280acreurbanTennesseewatershedhasa6.0hrtimeof tainis 6'0 in' The25year,24ht of from Table15.4,and5 percent the areais ponded. Find the 25yearpeak discharge.
TABLE 15.4 RUNOFFCURVE NUMBERSFOR URBANAREAS (see Sec. 4.9 foT other values)
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 50757o) 49 Good condition (grasscover > 757o) 39 Impervious areas Pavedparking lots, roofs, driveways,etc. (excluding right ofway) 98 Streetsand roads Paved;curbs and storm sewers(excluding rightofway) 98 Paved;open ditches(including right ofway) 83 Gravel (including righrofway) 76 Dirt (including rightofway) 72 Westerndeserturban areas Natural desertlandscaping(pervious areas only)' 63 Artificial desertlandscaping(impervious weed barrier, desertshrub with 12in. sandor gravel mulch and basin borders) 96 Urban districts Commercial and business 89 85 Industrial 81 72 Residentialdistricts by average size lot 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 conditionsmaybecomputedusingFig. 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. : Source U.S. Soil Conservation 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 TR55 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
328
300
u o P0 R
I
tnn
3 S
100 R o 60
40 01
0.6
0.8 l
(q*) for SCS TypeI 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
(q,) Figure 15.12 Unit peakdischarge for SCSTypeIA rainfall distribution.(After 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
(q,) for SCSTypeII rainfall distribution. (After Figure 15.13 Unit peak discharge Service.) U.S. Soil Conservation
*
d E
E
,nn
= 5 roo
80 60 40 0.1
0.2
0.4
0.6 0.8 1
810
(q,) Figure 15.14 Unit peak discharge for SCSTypeIII rainfall distribution.(After 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
From Solution. From Fig. 16.17, the TypeII storm appliesto Tennessee. q,: 96 csm/in. 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 simulationmodel TR20 shouldbe usedif the watershed very'complexor a higher 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 use presentand developed conditionsof a watershed, the sameprogedure for estimating[. Both the graphicalandtabular methodsare derivedfrom TR20 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 Hydrographs the subwatersubwatersheds. dividedinto a numberof homogeneous shedscan be routed and added.
15,3
331
and is not detailedhere. in The tabularmethodis described the technicalrelease are employed: In using the method, the following steps and 1. Subdividedthe watershedinto areasthat are relatively homogeneous reaches' haveconvenientrouting in 2. Determinedrainageareaof eachsubarea squaremiles' in hours. The procedurefor estimatingI is 3. Estimate T"for eachsubarea outlinedinTR55. 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 7. Determin the 24hr rainfall for the selected from CN andrainfall (Chapter4)' 8. Calculatetotal runoff in inchescomputed from Table 15.5. 9. Find I,fot eachsubarea selectone of the hydrographs 10. Usingthe ratio of I,f P andT,for eachsubarea, in tabulated TR55. 11. Multiply the hydrographordinates(csm/in.) by the area (mi2) and runoff subarea. (in.) of eachrespective 12. Route and combinethe hydrographs. The SCS recommendsthat TR20, 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 havebeenincorporated SCSin a computerprogram.Copies The TR55procedures are availablefrom the U.S. National TechnicalInformation Service.
'
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
.E
rno
'z ..\"\
L'//
l
'//.2
2\
8 0 & 7 0 3 a o 50 40 30
W
L
& .Mt/,,
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
tPI49 peak Figure 15.15 ratesof discharge small watersheds, for TypeI storms:24hr 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.)
LU
I,il
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
TypeII storms: Figure 15.16 TP149 peak rates of dischargefor small watersheds, "A 24hr 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 TypeI and TypeII stormsusingFigs. 15.15and and that only stormtype changes all other conditionsare equal. 15.16.Assume with CN : T0lesultsin on Solution. A 4in. rain over200 acres a watershed : II Qo : 52 cfs for a TypeI 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
( 1s .3)
A in where qois the peak discharge m3/sec, is the drainageareain sq. km., Q is the 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 HEC19SCS 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' Hec19, Hydrology' FHWAIP95, 1995.
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 and 0.5 0.1 /" shouldbe about samefor any of the main channels, watershed more if has than one main channel . no channelor reservoirrouting is allowed . no storagefacility on main channel ' watershedareain storage ponds and lakes shouldbe lessthan 5 percent . . . '
bYs)
SyntheticUnitHydrograph PeakRateFormulas
Peak flow rates from small watersheds can also be determinedusing the synthetic unithydrographtechniquesdescribedin Chapter 12. A storm having a duration definedby Eq. 12.22wirl produce,accordingto Snyder'smethodof synthesizing unit hydrographs, peak discharge 1.0 in. of net rain given by Eq. 12.17,or a for
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 the SCS methodfor constructing to 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 time plus half the stormduration. lag 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.
Popular dischargearea 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 tudesfor atange of frequencies a statebystate in discussed Chapters26 and 27 and techniques USGSusing the multipl regression formulas are availablefrom the illustrated for Virginii in Problem 27.25. Similar for equations the Softwarecontainingall the USGSregression USGSfor other states. and Federal Highway United Statesis availablefrom the U.S. GeologicalSurvey Administratioh as part of the HYDRAIN softwarepackage'
(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 indexflood curve (see Fig. the 26.4c).The usqof suchcurvesin urban hydrologyis limited because USGS data smallerthan 10 mi2. method seldomincludeswatersheds indexflood network for the in later, areapplicablefor watersheds the described equations, The USGSregression rangeand larger. 110 mi2
'
CyprusCreekFormula
includingestimates Extreme$ flat areasposeparticular difficulties to the hydrologist, runoff rates.Flooding in theseareastends!o runoff volgme,and,peak of infiltration, Flow velocitiesare low, and water standson the surface be shallow and widespread. by long periodsof time. Theseareasare often distinguished networksof for relative)y ihannels that have been constructedto store and eventually disstraight drainage rain. chargethe excess peak flow to a developed procedure3s calculatethe'instantaneous The Scs basedon first calculatingthe capacityof canalsthat would be flatland areas to limit flatxeaflooding for the designstormto a duration that would prevent 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
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
ui dr : ; < r t
338
CHAPTER15
URBANAND SMALLWATERSHEDHYDROLOGY
.Duratio4
o o
Iime
Figure 15.18 Illustration of relation befweenCyplys Creek flow. (After Soil ConservationService'35) stJntaneous
to duration was} hours,considered be the maximum allowabletime for The selected to called the CyprusCreekformula, wasdeveloped inundationof crops.An equation, flow rate, called the 24hr removal rate. The equation, determinethe canal design curve based on rainfall depth,contributingdrainage area,and the SCS composite number is
Q: CA5/6
( 1s .10)
channelcapacity fot 24ht 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 24hr peak flow rate is Once Eq. 15.10is solvedfor the given frequency,the instantaneous from 1 to about areas drainage is from Fig. 15.19.The procedure limited to obtained rate peakinstantaneous to the 24hr that 200 squaremiles.It is suggested ratios of the than or equalto 1.0.For flatland areas canaliemoval rateUetirniLO b values greater that the peak flows 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 methodssuchas TR20, TR55' 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
:
N
o
N
0 ) d L r
: r I
3rX
340
CHAPTER15
URBANAND SMALLWATERSHEDHYDROLOGY
s40
o !l
o 9 ^ ^ i 4rl
0
Percentage Area Servedby Storm Sewers of
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 50yr flow rate fron a 1.0 sq mi and drainage areathat hasa CN : 80,is 50 percent stormsewered, hasa50yr,24hr rainfall depthof 12.0inches. of Solution. From Fig. 4.I4, thedirectrunoff for 12 inches rain is 9.45in. The drainage coefficient,C, is found from Eq. 15.11,
: + C : 16.39 (14.75X9.45)155.t
The 24hr removalrate is found from Eq. 15.10' : '0)5/6 155'8cfs 0 : 155'8(1 rate From Fig. 15.19the ratio of instantaneous to removalrate is 2.0, giving a From Fig. 15.20,it is existed. designflow rate of 311.6cfs if no stormsewers by shouldbe increased 35 percentfor 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
ez: l3.2Ao.zt(13 BDFlo.azpnotz es : 10.6Ao.rz(13 BDF)o.3eReo18 t0(13 BDnlotuRQ?dn Qto : 9.5rAo ts(13 BDF)o'z+P9o'to Qt5 : 8.68Ao ts(13 BDFlo'zzR03o" Qso: 8.o4Ao ts(13 BOrTot'RQ?r!& Qno: 7.70Ao : A0'16(13 BDF)'*RQ1i& Qsoo 7.47
(Ls.r2) (1 .1 ) s 3
(15.14)
(1s.1s) (1s.16)
(15.17) (1 .1 ) s 8
from data at 199of the 269 original sites.The other siteswere Thesewere developed detentionstorageor missingdata. All theseequaof the presence.of because deleted coefficientsof determinationabove0.90. tions have used values and of the Figure 15.22shows correspondence estimated observed Eq. 15.15.Forty percentof the valuesfall within one standarddeviation in devedping line. Graphsfor other recurrenceintervalsare similar to the 1Oyear 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,
400
10 200
t2 0
: discussthe Rework Example 15.2 basedon a C : 0.2 and C 0'4' Compare and at effect of C on the discharge the outfall' AwatershedhasareaA.Startingwithatriangular.shapedunithydrographwithabase : 484A/to' lengthof 2.67t, and a heightof [0, deriveEq. 15'9 (seealsoEq' t2'25)' Qo the derivation' State and carr units of eachterm usedin
UsingtheSCSdimensionlessunithydrographdescribedinChapter12,determinethe to peak peaidischargefor a net storm of 101n.in 2 hr on a 400acrebasinwith a time ^of + nt and i lag time of 3 hr. Comparewith Eq' 12'17' with a 100min time of concentrationreceivesrainfall at a A 10.00mi2watershed rate of 2.75 in.lhr for a period of 200 min' 1f (cfs)from the watershed C : 0'4' a. Determinethe peak discharge rate lcfs) 150 min after the beginningof rainfall. b. Estimatethe discharge rate from the watershed 40 min after the beginning of 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)afterbeianingof rainfall
354
CHAPTER15
1s.13. A storm gutter receivesdrainagefrom both sides.On the left it drains a rectangular
600acreareaof t" : 60 min. On the right it drains a relatively steep300acrearea of t" : 10 min. The f index on both sidesis 0.5 in.ihr. Use the intensitydurationfrequency curves in Fig. 15.7 to determinethe peak discharge(cf$ with a25year recurrenceinterval for (a) the 600acreareaalone, (b) the 300acrearea alone,and (c) the combinedareaassuming that the proportion of the 600acreareacontributing 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 10hr 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 1.0mi2parking 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 50yearflood for a20mi2 basin at the northwestcorner of Nebraska. that Fig. 26.4 appbes. Use the indexflood 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 indexflood method to determinethe 10 and 50yearpeaksfor a 6400acre
drainagebasin near Lincoln, Nebraska.Assumethat Fig. 26.4 applies. 15.20. For the drainagebasin in Problem 15,19 determinethe probability that the 20year peak will be equaledor exceeded leastonce (a) next year and (b) in a 4yr. period. at Referto Section26.1.
15.24. You are asked to determinethe magnitudeof the S0yearflood 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 the Oak Creek Watershed Lincoln, Nebraska,are shownin Fig. 24.8. of at The watershed a SCScurve numberof CN : 75 which may be usedto determine has the direct runoff for any storm.Assumethat IDF curvesin Fig. 27.13 applyat Lincoln. Treat the entire watershed a singlebasinand determine 50yearflood magnitude the as 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 indexflood method. Figure 26.4 applies.
REFERENCES 355
15.26. RepeatProblem 15.25 with SubareaI excluded.compare the results with Prob
event of the at lem 15.25 and comment on the effectiveness Point 8 for the 5Oyear flood easilystorethe 100year BranchedOak Reservoirat Point 9. (This reservoirwill from Area I.) 1s.27. RepeatProblem 15.25 for SubareaA' 15.28. RepeatProblem 15.25 for SubareaI' is 15.29. Describecompletelyhow the magnitudeof the 30yearflood for a watershed determined by the USGS indexflood 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 be expected a 6.2in.rain. unit peak flow for the SCS dimensionless hydrographin Ch. l2,_determine 15.31 Using the for the piak discharge a net storm of 10 in. in Zhr on a 400actebasinwith a time to 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 100year are flood magnitudes neededto establishthe low chord and embankmentelevations, respectively.Determine the designflow rates using the scs TP149 method. The bridge drainsthe zone marked,about45 acres.The moderatelyslopedbasinlies in a rainfall rypen stormregion,the curve numberis 70, and the 24hr 50 and 100year g.4" respectively' depthsare 8.6" and
Bidge27l.32
method.The problem 15.32usingthe FHWA HEC19 peakflow SCSdesign 15.33. Repeat from the relationships can be determined is 0.2 hrs. Valuesof 1o time of concentration in , in Fis. 414. Provide the answers both metric and English units.
REFERENCES
"A Critique of Current Methods in Hydrologic Systems 1 . J. Amorocho and W. E. Hart, Investigations Trans.Am. Geophys.Union 45(2),301321(Jwe 1964)' ," ..NonlinearInstantaneous UnitHydrographTheory," ASCE J. Hyd. Div. 2. f. p. jingh, 347(Mar' 1964). 90(HY2), Par I, 313
356
HYDROLOGY 15 ANDSMALLWATERSHED CHAPTER URBAN "ContinuousHydrographSynthesis with an and 3. W T. Sittner, C. E. Schauss, J. C. Monro, 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), 124127(Iuly 1965). J. 6. S. L. S. Jacoby,"A MathematicalModel for Nonlinear Hydrologic Systems," Geophy. 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, "synthesisof the Inlet Hydrograph,"Tech.Rep. 3, Storm Drainage Jr., 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. Sept.1968, A79A103. "Characterization, 14. S. R. Weible,R. B. Weidner,A. G. Christianson,and R. J. Anderson, of Treatment,and Disposal of Urban Storm Water," in Proceedings the Third Interna(S. International AssociationonWater Pollution Researcft H. Jenkins, 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 Water ManagementModel," Environmental Protection Agency, Vol. 1, Inc., 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 150532058.
r14L2r(r9s7).
r971.
Chapter16
Design Hydrologic
Prologue
of The PurPose this chaPteris to: for usedin the United States designing . Introducethe hydrologist procedures to of structuresfor safe and effectivepassage flood flows' , 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. piotto studying 26 to Readersare encouraged reviewthe materialin Chapters and27 in presented this designprocedures rater discharge PredictingPeak minor and ma for use in designing aspects of engineering hydrology' I small crodsroadculverts, levees,dt akPort drainage structuresto the lumped together'with major structr design information. GenerallY,a h dischargefor a designfrequencY,a dischargehYdrograPhfor a design rates,lowflow frequencYanalysis, are often conductedas part of a designproiect'
360
DESIGN 16 CHAPTER HYDROLOGIC Mostdesignsinvolvinghydrologicanalysesuseadesignfloodthatsimulates are futirre eventorlmitates ime historicalevent.If streamflowrecords somesevere storm records from available are unavailable,designflood hydrographs synthesized are of procedures Chapters2, 12,and 15. Only in rare cases usingthe rainfallrunoff in small watersheds' particularly for streamflowrecords adequate complex designs, in and the empiriccoirelativemethodsdiscussed Chapter 15 are analyses Regional in Methodspresented chapsites' 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 levels,rnethodsfor this chapter.included are discussions;f data needs,frequency for designstorms,and hazard assessments floodplainsand dams' synthesizing
361
are of Major Structure Design Hydrologicdesignaspects maior structures considculvert, or urban drainage crossroad erably more complexthan thoseof a small dam, system.A designstorm hydrographfor a'large dam still is required but it is put to greater use. The designstorm hydrographis routed to determinethe adequacyof The economic spillwaysand outlets operatedin conjunction with reservoir storage. possibilities dictatesthe final designand ofthe spillwaysizefrom the various selection is a function of the degreeof protection providedfor downstreamlife and property, and policy andconstructionstandards, reservoiroperational project economy,agency largely basedon critical eventmethodspreMajor structuredesignis requirements. in sented Section16.5. in presented Water Resource System Design Most information and techniques protection aspectof small and large structhis chapterare directedtoward the flood for to tures.Needless say,a major structureis designed more than just flood protecfor irrigation, power, water supply, tion; it is multipurposeand may provide storage to The proper allocationof storage theseuses navigation,and lowflow augmentation. history in terms of the frequency of requiresan understanding the entire streamflow and yearly flows, as well monthly, seasonal, of occurrenceof low flows and average presentedin Part Five to provide a as the historical and designfloods. Material is hydrologistwith the tools to developcomplete streamflowhistories for a complex multipurposesysteminvolving various combinationsof minor and major structures, practlces. projects, and management water development
Methods FlowBased
or For designlocationswhererecordsof streamflows are available, whereflows from to can be transposed the designlocation, a designflood magnitudecan anotherbasin be estimateddirectly from the streamflows by any of the following methods: ' 1. Frequencyanalysisof flood flows at the designlocation or from a similar basin in the region. from regresnormally developed 2. Useof regionalflood frequencyequations, gaugedflood data. sion analysis(seeChapter 26) of 3. Examination of the stream and floodplain for signs of highest historical of floods and estimationof the flow rates using measurements the crosssectionand slopeof the stream.
Methods PrecipitationBased
for or recordsare unavailable inadequate streamflow.estimaWhere streamgauging floods can be estimatedby evaluatingthe precipitation that would protion, design ducethe flood, and then convertingthe frecipitation into runoffby any ofthe rainfallin runoff methodsdescribed Chapters1015 or 2l27. Typical methodsinclude: 1. Design using the greateststorm of record at the site, by converting the precipitation to runoff. in historicalstormfrom anothersimilar watershed of 2. Transposition a severe 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 flood flow rate is desiredin all cases, flowbased the the preferred over conversionof precipitation to runoff. Due to the relatively longer period of time and greaternumberof locationsat which precipitation amountshave been recorded, precipitationbasedmethods are used in the majority of designs, especiallywith small and very large basins.Flowbasedmethodsare typically used in the midrangeof basin sizes.
FrequencyBased Methods
most often proceed Regardless whetherflow or precipitation dataareused,designs of by selectinga minimum acceptable recurrenceinterval and using proceduresfrom Chapter27 to determine corresponding the worst condition storm or flood that could be equalledor exceeded recurrenceinterval. Criteria for selecting during the selected designrecurrenceintervals are summarized Section 16.3. Resultsfrom frequency in analysisof flood flow data normally provide reliable estimatesof 2, 5, 10, and 25yearflows.Extrapolationbeyondthe rangeof the period of flow recordsis allowed, but is lessreliable.
RiskBased Methods
Recenttrends in designof minor (and major) structuresare toward the use of ecoThe risk methodselects the nomic risk analyses rather than frequgncybased 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 eachof severalsizes. for The optimal sizeis that with the smallestsum.Structuresselected risk analysisare by normally constrained sizesequal to or larger than thoseresultingfrom traditional to frequencybased 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 critical eventtechniques by 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 rainfallrunoff model,including snowmeltrunoff if pertinent. 3. Examining the flood plain and stream to identify palaeofloodevidences such as highwater marks, boulder marks on trees or banks, debris lines, historical accountsby local residents, geologicor geomorphologic or evidences. 4. In somecases, critical eventmethodinvolvesestimatingthe magnitude the of the 500yr eventby various frequencyor approximatemethods.Often, suchas in mappingfloodplains,the 500yr flood is estimatedas a multiple of the 100yr event, ranging from 1.5 to 2.5. Due to lack of longerterm records, frequencybased estimatesare seldom attemptedfor recurrence intervals exceeding 500 years.
Physiographic Data
The hydrologic study for any structurerequires a reliable topographicmap. United StatesGeologicalSurveytopographic mapsusually are available. The mappingof the United Statesis almost completewith 15minute quadrangles, and many of these areasaremappedby 7.5minutequadrangles. County mapsand aerialphotoscan also be usedto advantage making preliminary studiesof the watershed. in
364
drainage Based on an area map, a careful investigation of the watershed's information can be obtainedfrom USGS maps behaviormust be made. Additional and erosive that depict predominantrock formations.Soil types and the inflltration districtsor univerfrom U.S. Soil Conservation of characteristics soilscanbe secured divisions. sity extension of an The drainageareascontributing to large dams require stricter analysis The possibility of a minor structures. in designing area,shydrotogyitranis necessary for large uniformly intenserainfall over the entire basin is an unrealisticassumption shouldthus The watersheds. influenceof temporaland spatialvariationsof the rainfall "worst possible" rainfall values are the estimated For major dams, be considered. which is then usedin reservoir hydrograph, generallyconvertedto a designdischarge and storage, reservoii and spillway size, surcharge routing calculationsto propoition maintain power requirementsor sustaineddownany 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 the pi"ta reconnaisiance alwaysincreases understanding structurebeing design"a. insignificantthe structule might be. matter how an area,shydrology*no
Data Hydrologic
datafor the regionunder one difficulty in hydrologicdesignis that of gettingadequate by issued publishedreports from previously data study.ConsiOerabie canbeacquired is a list of federal agencies and/or universities.The following agencies governmental hYdrologicdata: 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, intergovernments, of Additional dlta often canbe procuredfrom departments state and regional and local agencies' statecommissions,
MeteorologicData
and Atmospheric The National WeatherService, couchedin the National Oceanic sourceof meteorologicdata publishedin a variety Administration, is the primary Figure 16.2 of forms, including their Hyirometeorologic Report (HMR) series. showstheapplicablereportsforvariousgeographicandtopographicregionsofthe
365
Figure16.2HydrometeorologicalreportseriescoverageofconterminousUnitedStates. of (U.S.Bureau Reclamation') collect and analyze state,and local agencies United States.lNumerousother federal inspect, or regulatelarge who design, precipitation information"tp""iurivtr'rore
"*"8l;.H[1;:1;ce
in the region,maximum amount ttre meteoroioii" .huru"t"ristics of Jtott of total storm ;;r;r of precipitation,frequencies itable moisturein the atmospheJo snowmeltfor stormsover the of and influence. durationsbf U;t, depthsf*;;;i"", of major mountain chains,topography ", f;;;iii'*gion, region.tn ,o1n" areassuch has a very distinct impact on precipitation'
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 frequencyof a culvert,which undercases excessive trafflc. the Servicerecommends use of a2lyear frequencyfor The Soil Conservation minor urban drainagedesignif there is no potential loss of life or risk of extensive damagesuch as firstfloor elevationsof homes.A 100yearfrequencyis commonly property damagemay occur.t recommended when extensive
DESIGNFREQUENCIES TABLE 16.1 MINORSTRUCTURE
Typeof minor structure Highway crossroad drainage" ADT' 0400 4001700 ADT 17005000 ADT ADT 5000Airfields Railroads Stormdrainage
Levees Drainage ditches
= Frequency 1/7, 0.10 0.100.04 0.04 o.o2 0.20 0.040.02 0.500.10 0.500.02 0.200.02
Large Dams
of duringthe design the original structureandduring Damsrequirehydrologicanalysis when are periodic safetyevaluations. Significanteconomicand humanlosses possible from storage. quantitiesof water are rapidly released large the Initial heightsof retardedwater behind the dam, disregarding total volume water, can produce destructiveflood wavesfor a considerabledistance of stored Basedon two criteria, the TaskForceon SpillwayDesignFloodsrecomdownstream. in of the classification large damsas li,sted Table 16.2.The type of construcmerided has not been included in this grouping, althoughit affects the extentof failure tion resulting from overtopping. of haveadopteddefinitionsfor hydraulic elements Many of the federalagencies The following list is usedby the Soil ConservationService: 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 ically controlled, to regulatethe discharge excess to Theprincipal spillwayis the ungatedspillwaydesigned conveythe water for rates established the structure. from the retarding pool at release to spillway of a dam is the spillwaydesigned conveywater The emergency of in excess that impoundedfor flood control or other beneficialpurposes.
367
DAMS FORIARGE CRITERIA TABLE16,2 DESIGN danger lmpoundment Potential Category (1)
Major; failure cannot be tolerated Storage (acreft)o (2)
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 501O0year recurrence interval
>50,000
lntermediate
100050.000 40100
Possiblebut small
<1000
<50
None
and future potential oBased on consideration of height of dam above tailwater, stoarags volume, and length of damage reach, present of floodplain' 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
Emergency spillway
Qo
zones' and poollevels storage 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 minimum designdimensions the emergencyspillway' 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' 100mph by shownin Tablq 16.3, atebasedon waverunup caused storm winds with The fetch wind velocities.Minimal freeboardpertainsto wind velocitiesof 50 mph. 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 DESIGNFREQUENCY 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 of this table.sDesign frequency_requirements consultedfor full interpretation use of the structures.The SCS classifles selectedto fit the planned or foreseeable grouPs:u into structures three 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, generally ft The physicalsizeof a small dam can rangeto over 100^ in heightbut acreft of storageat the emeris restrictedto structuresretarding lessthan 25,000 are gency spillway crest. Small dams generally receive.special attention if they could causethe loss of life' Many ion.i.o"i"d in populatedareaswhere dam failure the by havebeencaused dam or leveefailure. When this possibilityexists, flood deaths of the probablemaximum precipidesignstorm for small damsis lstablishedby use maximization of the tatio'n, PMP. The PMP is generally defined as the reasonable maximum storm. Other definitions 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 conditionsthat are reasonably and hydrologic tion of critical meteorologic possiblein the region. T'he pMp Las a low, but unknown, probability of depthat the designlocation the It occurrence. is neither maximumobserved immune to exceedance' 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
Earth
Vegetated
irrigation only
0.sDrJ 0.75DL
!D 0 5
O.5DL
None
0.75DL
D6 r25
SingIe or multiples
0.5(Pso + Ploo)
D r 100
0.5(Prs+ Pso)
rD 0 5
(b)
100
(c)
All
Proo
I Product of reservoir storagevolume V, (acrefeet) 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 establish maximumpossibleamountof precipitablewaterthat could to the of Figure 16.4providesestimates precipbe 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 6hr storms.These and similar publishedcharts for other durations are helpful in selectingthe PMP for any region in the United States. the deiign frequenliesfor principal spillwaysfor small SCS Class aob, or c damsare providedin Table 16.5.Theseare basedon 6hr rainfall depthsfor ( 1) the for (Fig. 16.7) and(2) the PMP (Fig. 16.6).Designstormdepths 1OOyear frequency all watershedshaving a time of concentration less than 6hr are establishedin are with greatertime of concentration,adjustments Table 16.5.For thosewatersheds madeto the 6hr storm depth to accountfor the gteateramountsof direct runoff in a in longerperiod of time. Theseadjustmentsare discussed Section 16.4.
DEPTHSOF PRECIPITABLE WATER IN A COLUMN OF MILLIBARS AIR OF GIVEN HEIGHT ABOVE 1OOO
Assuming saturation with a pseudoadiabatic 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
Figure 16.4 Diagram for precipitable water determination from 1,000 to 700 millibars.(U.S. Bureauof Reclamation.)
372
DEPTHS OF PRECIPITABLE WATER IN A COLUMN OF AIR OF GIVEN HEIGHT ABOVE 1OOO MILLIBARS lapse Assuming saturation a pseudoadiabatic with
rate for the indicated surface temDeratwes Adapted from the US. Weather Bureau Hydrometeorolgical Report No 23 EMPEMTre
l8 26
66 70
74
200
Figure 16.5 Diagram fbr precipitable water determination from 800 to 200 (U.S. Bureauof Reclamation.) millibars.From 1281. 103D1908.
16.4 DESIGNSTORMS
373
MaiorStructures
Duration
The length of storm usedby the SCS in designingemergencyand freeboardhydrographsior small damsis of 6hr 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 trationof 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 24hr durations for all urban watershedstudies' but for Durationsof approximately6hr or lessare satisfactory small watersheds, depthsfor periodsof up to 10 days' the lengthsof stormiln large areasrequire storm valuesare availablefor durationsof from 2 to I0 daysfor locations Frequencybased Similar data are also availablefor other selectedareas within the United States.to outsidethe United States.Generally,however,designcriteria for large damsrequire of estimates storm depthsthat do not havefrequencylevelsassigned.
Durationof RainfallExcebs
and Initial rainfall during most stormsinfiltrates or is otherwiseabstracted, the durarainTf,is lessthan the actual rain duration by an amount.equalto the tion of excess time that initial abslractionsoccur. Excessrain duration 7s can be estimatedfor a 6hr storm as a function of the curve number CN and precipitation P from
374
i 7t?
"d, \
tv
,o
Bureau, Figure 16.6 The 10mi2or less PMP for 6hr duration (in.). (U.S. Weather NOAA.)
storm by Fig. 16.8. This family of curves was developed the Sclrlt yh:I" P is the zero A CNof 100 represents una O""pttr CNis a losi parameterdefined1lhaptgr {. lossessothatZg:6hrforCN:l00.Table16.6isusedtofindthedurationof is the excess rain for any storm duration greater than 6 hr. The rainfall ratio (T'iUle 16.7) divided by the total precipitation 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"
\ \
5P
85' 6u
Depth
available.
386
0.5
0.3
v.z
0.1
3 4 Time(hr)
country. For more preciseinformation on boundaries a statehavingmore than one in 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 24hr 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 threegeographical in zonesshownin Fig. 16.6.The corps of Engineers usesa distribution curve similar to Fig. 16.18for 6hr SpS analyses. Triangular Distribution The simplest design storm distribution is a triangular shape. Because depth,P, andduration,D, of rain are alreadyestablished, peak the the intensity,lu,, is 2PfD, foundby solvingfor the height of the triangularhyetograih as shownin Fig. r6.2L.The only remaining decision the time to the peak,40. is The ratio to/D has been investigated a large number of storms at locationsin California, for Illinois, Massachusetts, New Jersey,and North carolina. values range from about 0.3 to 0.5.17 Once the triangle is constructed, intensitiesat regulaiintervals may the be graphicallyor analytically determined input to the rainfallrunoff rnodelbeing for usedfor design. Blocked IDF Distributions A frequentlyusedprocedurefor developing design a 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
llffil rvnerrl
rFooo
V v
(b)
for zonesI' Figure 16.19 SCS 24hr rainfall distributions:(a) 24hr rainfall distributions for scs rainfall distributions'(After Ref' 16') boundaries IA, II, and III and (b) approximate
388
CHAPTER16
;
E tr ;
o.7o
o6n nsn 0.40
E o.3o
o.20 0.10
Time (hr)
Figure 16.20 Distribution of 6hr PMP for any area west of the 105' meridian. (After Ref. 6.)
Time, hr Figure 16.21 Triangular hyetograph. design durationfrequency curveto find therain intensities A/, 2 At,3 L,t,etc.,increments for 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 equalprobability storms with durations A,t,2Lt,3A,t,4Lt,etc.,aredetermined of 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 directions from the center block. The balanced method, for working out in both that exampli, assumes a Arhr stormwith intensityia,from the IDF curvecould occur, with equal probability, during the middle of the Dhr designstorm. This intensity is Next, the rain depth for plotted as the middle block of the designstorm hyetograph. obtained from the IDF curve. Its distribution is assumedto be a duration 2Lt is twobar histogramwith the first half matchingthe intensity of the Arhr storm; the secondhalf intensity is calculatedby spreadingthe rest of the rain depth for the 2 for is Alhr durationunifoimly over the secondAr interval.The process repeated rain a of with durations 3A/, 4Lt, . .. , up to D. The goalis to develop depthsfor storms suchthat a storm of any duration,centeredat the middle of the storm hyetograph will blockedIDF hyetograph, havea total rain depthmatchingthe rain depthfrom the 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 The usual approach major structureanalysisis to selecta design(usually 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 400 mir. Thesewere identifiedby iluff in his analysis midwesternstormpatterns.ll The letters H andL representareaswith high and low precipitation depths,respecas tively. The typical isohyetalpatternfor SPSstormshasbeenestablished generally 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
t t Scale: miles
lsohyet
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 shownin Fig. 16.24.The percentages as shownfor isohyets B, . . . , G A, are multiplied by the 96hr SPS depth to give an elliptical pattern with the desired average depth.Similarmapsfor 24.,48, or 72hr stormscanbe obtained simplyby modifyingthe 96hrpercentages Fig. 16.24.Thisis accomplished of usingthe depthareaduration curves Fig. I6.24.For example, a24hr stormis used, in if first notethat theA isohyetof Fig. 16.24encloses areaof 16nrr?. an From Fig. 16,25thecorresponding SPSpercentage a24hr stormis 116 percentrather than the 140percentvalue for used with a 96hr storm. Thereforethe pattern percentages vary with the selected designstorm duration. An additional aid for constructingdesignstorm distributionsover smallermid(up westernl8 watersheds to 400 mi') is presented Table 16.10.The ratio of maxiin mum 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
391
: miles
Figure 16.24 GeneralizedSPS isohyetal pattern for a 96hr storm' The pattern may be orientedin any direction and may correspondto the depthby arearelation represented a 96hr storm.
lsohyet
A
Area (mi2)
B C D E F G
(After Ref. 19.)
known. Ratios for 50, 100, and 200nr2 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.
392
16 CHAPTER
\
+ 72hperiod
\\\
s=
,*Nr"
\\ \ \
o I
\ \ \
60
140
Figure 16.25 SPS depthareaduration curves by 24hr 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 frequencybased paleohydrology.
MaximumPrecipitation Probable
Probablemaximum precipitation depthsfor drainagebasinsin the United Statesare identified in Fig. 16.2. providedin the respectiveNational WeatherServiceHMRs2o reasonstorm considered most severe The probablemaximum stormis deflnedasthe probablemaximumflood is customarilyobtained to ably possible occur.The resulting of and by usingunit hydrographs rainfall estimates the PMP preparedby the National rily'eather (seeFigs. 16.6 and 16.26). Service2l
EVENT METHODS 393 16.5 CRTTTCAL ON POINT MEANRAINFALL 400 miz TO OF TABLE16.10 RATIO MAXIMUM
Mean rainfall(in.) Rainfallperiod (hr)
1.0
1.5
2.O
2.5
4.0
5.0
0.5 1 2 3 6 l2 18
)4
48
0.5
I
L
3 6 12 l8
JA
48
0.5
I
2
J
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
that suggests the by PMP advocated Hershfield2l methodto estimate A proposed 24hr PMP at a point be computedby the equation
PMP24:P*KS,
(16.1)
the 24hr probablemaximum precipitation the meanof the 24hr annualmaximumsover the period of record a constant equalto 15 the standarddeviationof the 24hr 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 510 percent.
394
CHAPTER16
HYDROLOGIC DESIGN
zo.t
,/
\\       " . . 20.1(3)
ptvtp(in.). (1) Alexandria, Figure16.26 Twentyfourhour 2000mi2 LA, (3) (2) June 1317,1886. Eautaw, April 1518,1900. Elba, March AL, AL, (4) (5) l1 16,1929. Yankeetown, September , 1950. Altapass, NC, FL, 3=7 (AfterRef. (6) July1317, 1916. Jefferson, September 1013,1878. 18.) OH, 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 peak dischargeof the PMF the without overtopping.In addition, the dam and appurtenant featuresmust accommodateat leastthe first 80 percentof the PMF volumewithout failure. For embankment dams,failure is assumed occur if overtoppinglevelsare reached. to
395
$ zoo
q? 100
500 300
s 0
ii JU
20 10 5 3 2
I
= {6gtrQs9ae
o 048 11
Nh
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) unusualflood discharges. Congo at Inga, Congo.(2) Tigris at Samarra,Iraq. (3) (4) Caroni at Guri, Venezuela. Tigris at Eski Mosul, Iraq. (5) Jhelumat Mangla, Pakistan. (6) Diyala at DerbendiKhan, Iraq. (7) GreaterZab at Bekhme,Iraq. (8) Surinameat (9) Brokopondo, Suriname. Lesser 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), is the drainage area (sq km) and the A coefficients from Table 1611 usingFigure 16.29. are
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
/:fr>y%^"1",
ndia
? 1oo
; 6 0 s 4 0
& 2 0 10 6 4 2
Fr
g
Ihra]l, TX imethport, I A 
ffi"*"[,,"1,
anlalca
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
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 ro; + \i cr ;  ; d d
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
q *E5o ii:E
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* =' E 9 
 . 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
V:
i:::\
including storm Figure 16.36 Children's artist rendering of urban undergroundsysteT, A 1976by David Macaulay.Reprintedwith perdriins. 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 for normally accounted in urban storm drain designare: Two items many 1, Infil,tration. The ability of the soil to infiltrate water dependson given of characteristics the soil as noted in chapter 3. The rangeof values continuous in the following table is typical of variousbare soils after t hr of rainfall' RATEs rvircnr NFILTRATIoN
Soilgroup
High (sandy,oPenstructured) Intermediate(loam) structured) Low (clay,close
409
such to 2. Retention This is usually assumed be 0.10 in. for pervioussurfaces surfaces' as lawns and normal urban pervious Developmentof hydrologicparametersfor designof storm sewerpipes, street gurrers, detentionbasinsis by the rational method(or modified rational rnethodor or are hydrographs adequate, iee Chapter25) when peak flow ratesand approximate synthesismethodswhen greater unit hydrographand kinematic wave hydrograph The latter usually involveuseof public domainor vendordeveloped detail is needed. stormwaterdesignsoftware.The hydrologicaspectsof computerizedhydrologicdetext' in signtools are deiailedin Chapter25. In addition to the material presented this modified rational method, ILLUDAS' descriptionsof usesof the rational method, TRsj, SWMM, DR3M, and other tools in designingurban storrrtdrainagefacilities Additionally, in are addressed numerousurban drainagedesigntextsand handbooks. and county engineer'soffices have of many statedepartments transportationor city tocattyapplicabledrainagedesignmanuals.As well, the American Society developed a i'model" drainagedesignmanual for local adaptaof Civil Engineeishas developed of tion, availableby contacting ASCE in New York. Finally, the discussion urhan "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 methof floodplainsof the nation's waterways ment hasbeenconductingstudies inthat causes prop"tty and pieventing overdevelopment ods of protecting life und flooding. Hydrology is a key ingredientin widespread water levelsand more creased these studies for identifying potential flow rates, studying effects of dams, open volumes and on water control structures hydrographs determining and.other channels, and by be safelystoredand conveyed the waterways 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 floodplainswith mappingof identify flood hazard areasand to provide occupants tolorvcostflood insurance'The NFIP requireslocal the floodproneareasand access programsthat preventdevelgou"rn"nt5, to adoptand implementflood management of opmentsin excess national standards' Since the inception of the National Flood InsuranceProgram, flood hazatd The areas have been mapped in over 18,000 communities in the United States. and has sinceconvertedto a maintenance programcost over $t.O Uittion to complete occur'Each of these effort of updating and expandingthe maps as developments detailedevaluationof peak flow rates for studieshas required eithei approximateor has a range of recurrenceintervals. The 100yeardischarge,called the baseflood, portion ofthe floodplainoccupiedby the baseflood The in beendetermined all cases.
4'10
has been mapped,allowing communitiesto determinewhether a property is in the 100yr floodplain,and in many cases, what water surfaceelevationwould be experiencedat the property during the baseflood. Figure 16.37illustrates typical NFIP mappingand floodplain the management procedure.Surveyedvalley and channelcrosssections used in determiningthe are 100yr flow depth, allowing the hydrologistto delineatethe lateral extentof flooding duringthe 100yrflood. Then afloodwaywidth is generallydetermined that portion as of the floodplain that is reservedin order to dischargethe 100yearflood 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 relocating by the floodway or mitigating the water surfaceincreasedue to the development. The is that portion of the floodplain outsidethe floodwayin which developflood fringe ment is allowed,up to a point of full encroachment buildings,roadbeds, by berms,and so forth. As much as sevento ten percentof the total land area of the United States lies within the 100year 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 500year frequencydischarge streams. for 2. The 10, 50, 100,and 500year flood elevations riverine,coastal, for and lacustrine floodplains.
',100_year" Floodplain
Channel
411
width' the for Figure 16.38 Procedure determining floodway 3.The100and500yearmappedfloodplaindelineationsatscalesranging to from 1:4800 1:24,000' 4. The 100yearfloodwaydata 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 information' tabusurcharge datalncluding velocity, floodwaywidth, and dati, and profi'lesof the 10' 50' 100' and lated flood insurance'zone flooding' for 500yearflood elevationversusstreamdistances riverine of provide delineations the 1002. Flood InsuranceRateMaps (FIRM maps) and500yearfloodplains,basefloodelevations'coastalhighhazardareas' andinsuranceriskzonesonaplanimetricbaseatascalebetweenl:4800 and 1:24,000. 3.FloodBoundaryHazardMapsprovidedelineationsofthel00and500floodplainand channelcrosssections locationsof surveyed yearfloodplains, floodwayon a of and delineations the 100year usedin hydraulic analyses, 1:4800and 1:24'000' baseat a scalebetween planimetricor topographic
NFIP studiesare basedon Flood flow frequency estimatesfor gaugedlocationsin Annual peak Type III (seechapter 27; analysisof streamflowrecords. logpearson by recommended FEMA' flows and historical data arcfitted accordingto procedures
412
CHAPTER16
HYDROLOGIG DESIGN
through For ungauged locations,flood flow frequency'estimates developed are regionalfrequencyanalysis throughrainfallrunoff modeling.Equationspublished or to by the U.S. Geological of Surveyrelatepeak dfscharges variousfrequencies various shape,andland use.These drainage basincharacteristics suchassize,slope,elevation, equationsare developedusing multiple regressiontechniques(see Chapter 26) at gaugedsitesthroughoutthe region. (Chapter24) use syntheticrainfall hyetoRainfallrunoff modelingtechniques graphs.Stormeventmodels, such as the Corps HECI and SCS TR20 packages, employdesignstormsof particular frequencies and then mathematicallysimulatethe physical runoff process.The resulting peak dischargeis assumed have the same to frequencyas the rainfall.
U.S.Flood Hazards
Despiteconsiderable effort and expenditurein identificationof floodplainsand flood damage hazardareas, continueto resultin severe dam failuresand other catastrophies life, property, and the environment.Floods from hurricanes,intenserainstorms, to 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 eastcoast 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 these floods have produced flood dam4gesin billions of dollars, disCollectively, 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 have occurred in floodplains. of investmentsin flood damageable billions value that Table 16.17lists the breakdown,by state,of estimated1990 development will be in harm's way.
SUMMARY 413
1OO MOREDEATHSIN THE OR 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 coastNew 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 Rainfallriver floods Rainfallriver floods Hurricane tidal flood Hurricane tidal flood Rainfallriver floods Rainfallriver floods Hurricane tidal flood Hurricane tidal floo<l Rainfallriver flood Hurricane tidal and river flood Rainfallriver floqd Rainfallriver flood Hurricane tide and waves Hurricane tidal flood Dam failure Hurricane tide and waves Hurricane tidal flood Rainfallriver flood Rainfall, snowmeltriver fl oods Rainfallriver flood Hurricane tidal and river flood Hurricane rainfallriver floods Hurricane tide and river floods Hurricane rainfallriver 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
Figure 16.39 Number by state of major hurricanesin the United States, NaI 8991989.(Soarce.' tional Hurricane Center, National WeatherService, NOAA.)
Figure 16.40 Average in annual flood damages the U.S., 191685. (Source:National WeatherService,NOAA.)
rg2o 25
30
35 40
45
50
55
60
65
70
75
80 8s
$50 MILLION TABLE 16.16 FLOODSRESULTINGIN DAMAGESEXCEEDING IN THE UNITEDSTATES Damage($ millions)
Year
Streamor place
Cause
Rainfallriver flood Dam failure Hurricane tidal floods Rainfall and dam failure Rainfallriver flood Rainfallriver flood Hurricane rainfallriver floods Rainfallriver flood Hurricanetidal and river floods Rainfallriver floods Rainfallriver flood Rainfallriver flood Hurricane tide and waves Rainfallriver flood Rainfallriver flood Rainfall snowmelt fl ood Rainfallriver flood Hurricanetidal and river floods Rainfallriver floods Rainfallriver floods Rainfallriver floods Huriicane tidal and river floodb Rainfallriver floods Rainfallriver floods Hurricane tidal and,rivet floods Rainfallriver floods Hurricanetidal and river floods Rainfallriver floods Rainfallriver floods Rainfallriver floods Rainfallriver floods Snowmeltfloods Rainfallriver floods Hurricane tidal floods Huiricane tidal and river floods Rainfallriver floods Rainfallriver floods Rainfallriver floods Rainfallriver floods Hurricanetidal and river floods Hurricane tidal floods Hurricanetidal and river floods Rainfallriver floods Rainfallriver floods Rainfallriver floods Hurricanetidal and river floods Rainfallsnowmelt river flood Rainfallriver floods Rainfallriver floods Hurricane tidal flood
River t844 Upper MississiPPi PA 1889 Johnstown, 1900 Galveston,TX and 1903 Passaic DelawareRivers 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 1935 Susquehanna DelawareRivers r936 NortheasternUnited States t936 Ohio River basin 1937 Ohio River basin 1 9 3 8 New England sffeams 1938 California streams t942 MidAtlantic. 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 I 1 9 5 KansasRiver basin 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
N.A, 50 284 50 36
zzl
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 VALUE RISKFROM PROPERry TABLE 16.17 ESTIMATED ON BASED 1990COSTS RANKED DECREASING IN 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. j6.
4;
GEoLocy sotLs,AND 17 cHAprER GR.,NDWATER, 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 may be encountered this region: hygroscopicwater, which is adsorbed 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 soilwatetzone 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 nearground surface region and the nearwatertableregion through which infiltrating fluids must pass. from the watertable (Fig.I7 '2) to 3. Capiltary zone. Acapillary zoneextends in by a height determined the capillary rise that can be generated the soil. 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.
Waterthat can be drainedfrom a soil by gravity is known as the specificyield. as It is expressed the ratio of the volumeof waterthat can be drainedby gravity to the 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 determiningthe specificyield.12 for
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
where AS : the changein storagevolume specificyield of the aquifer S, : the average the AV : the volumeof the aquiferlying between original watertable andthe 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 causing compression, theioad on the aquiferincreases, othermeans, a alsocauses small the hydrostaticpressure that somewateris forced out. Decreasing an additional releaseof water. For confined expansion,which in turn produces in aquifers,water yield is expiessed terms of a storagecofficient S", definedas the per volumeof wateian aquifei takesin or releases unit surfaceareaof aquiferper unit of Figure 17.2 illtstrates the classifications changein head normal to the surface. aquifers. ratesofyield, thereare strataexhibitingsatisfactory In additionto waterbearing stratathat may contain large quantitiesof and impermeable also nonwaterbearing rates are not high enoughto permit effectivedevelopwater but whosetransmission ment. An aquifugeis a formation impermeableand devoidof water; an aquicludeis an imPerviousstratum. In the following three chapters,the mechanicsof groundwaterflow and the presentedall elementsof groundwatermodelingwill be introduced.The techniques physical systemto be modeled.Before a numerical of dependon a=knowledge the a model can be developed, conciptual framework must be devised.This framework must take into accountthe region's topographyand geology;the types of aquifers, lithological variations,and characteristics; lateral extent,boundaries, their thickness, areas,their rates of dischargeand dischar_ge the nature and extent of rechargeand table'' recharge;and the elevationof the water
Topography
to it operates, is essential know something how a groundwatersystem To understand compiledshowingall surface map shouldbe A aboutthe region's surface. topographic water bodies,including streami, iakis, and artificial channelsand/or ponds,as well as land surfacecontours.Furthermore,an inventory of pumping wells, observation suchas identifying types of wells, and explorationwells shouldbe madefor purposes watertable locationsandrates;and determining soilsand rocks,pinpointingdischarge elevations.
GeologY Subsurface
governsthe occurrenceand movement The geologicstructureof a groundwater'basin of the grirndwater withinlt. Specifically,the number and types of waterbearing formations, their vertical dimensions,interconnections,hydraulic properties, and Once the before the systemcan be analyzed.u outcrop patternsmust be understood and lower conditions have been identified, contour maps of the upper subsuriace
SUMMARY 433 boundariesof aquifers,watertable contourmaps,and mapsof aquifercharacteristics can be prepared.Welldrillers logs, experimentaltest wells, and other geophysical explorationmethodscan be usedto obtakr the neededgeologic data.se'13'14
r Summary
The importance of groundwaterto the health and wellbeing 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 this resource clear.Quantity and quality dimenwatersupply.The needto husband sionsare both important.
4g4
CHNPTENTZ
Groundwaterprotectionandmanagementpracticesmustbebasedonanunder. groundwateris^distributed standing of groundwatersources,the rianner in which of topogiaphic,and soil characteristics the region' 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, GegpeF Pirltder'.Groundwater e. F. Wood, Ra'ymond p"rru.u, diUiu G' Gray' and E. CoitaminationfromHazardousWastes'EnglewoodCliffs'NJ:PrenticeHall'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:PrenticeHall'19'19' York: McGrawHill' 1993' of llydrology' D. R. Maidment (ed.), ii"aLook ,,Niurnerical New Basins,"InternaModeling of Groundwater J. Boonstraand N. A. O" nlJa"r, The Netherlands'1981'' tional Institute for Land Reclamationand Improvement' New York: Wiley' 1965' 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, Printing Office' 1984' Supply PaperNo. 22a2' iishiniton, D'C: U'S' Government DisposalPracproteciion ig"""y, "The Report to Congress:^Waste 1 1 . U.S. Environmentat summary'U'S' EPA' PB 265364' tices and Their Eft'ectson Groundwater,"Executive 1977. States,"U'S' Geological tz. o. E. Meinzer, "The occurrence of Groundwaterin the united 1923' Survey,WaterSupplyPaperNo' 489' "A GeneralizedGraphical Method Evaluating for 1 3 . H. H. Cooper, Jr. and i. n. lacoU, Trans'Am' Geophys'Union Formation constants una so*u.izing wellField History," 27, 526534(1946). "Outline of Methodsfor EstimatingCrounlw11e1fupplies"' U'S' Geolog1 4 . O. E. Meinzer, D C:' 19?2ical Survey,WaterSupplyPaper638C, 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 Basinwide 1959. project california,,, po.to. oi Eni thesis,university of california, Berkeley,
1B Chapter
of Mechanics Flow
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 Water locatedin pore spaces a saturatedmedium is under pressure(calledpore in by cin be determined insertinga piezometer the mediumat a point pressure), which it If LocationA (Fig. 18.1)is considered, can be seenthat pore pressure of interest. is given by P:h"l where p : the pore pressure(gaugepressure) from the point to the water table h o : the headmeasured the specificweight of water v: positive or negative,dependingon whetherthe pressure Pore pressureis considered the above(poJitive)or below (negative) point under consideration. headis measured the If an arbitrary datum is established, total head or piezometric head abovethe 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 sures a porous or tension vacuum(negative,pore of Iermh"is the pore pressure A while hu denotes pressure) B. at
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 crosssectional porous material K : the hydraulic conductivity of the material the Q: the flow across control areaA In Eq. 18.3 h : z * where h  the piezometrichead the elevationabovea datum p : the hydrostaticpressure C _ an arbitrary constant in If the specificdischarge = QIA is substituted Eq. 18.3, S 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 Of specialinterestis the fact that the Darcy equationis analogous Ohm's law
' : (*)u
where i:thecurrent R : the resistance E : the voltage
(18.7)
as current and velocity are analogous, areK andI f R, andE anddhldx.The similarity 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 10s ft2lsec. gives4 : I.2/86,400 : 1.39 X 105' The the velocity4 into units of ftlsec thesevaluesin the meangrain diameterin ft : 0.08/12 : 0.0067.Substituting equation,we obtain
Nn:
r.39x10sx0.0067
1 . 2 1X 1 0  s 0.0077
18.4 PERMEABILITY
into The hydraulic conductivity K is an important parametrthat is often separated medium, the other to the fluid. The product one related to the two components,
k:Cdz
(18.8)
can be written
X:4
lL
(18.e)
of Dimensions intrinsic permeability areL2. Sincevaluesof k given asft2 or cm2 known as the darcy has beenwidely adopted. are extremelysmall, a unit of measure 1 darcy : 0.987 x 10s cm2 or 1.062x l01r f* hydraulic conductivity are reportedin the literature. Severalwaysof expressing hasdefinedthe standardcoefficientof permeabilityK" as Survey The U.S. Geological  the nunber of gallons per day of water passingthrough 1 ft2 of medium under a
446
CHAPTER18
OF FLOW MECHANICS
#o;+frat:o
(18.s2)
must be a constant.A seriesof is The total differential dry' equal to zero,and ry' of curves {(r, y) equalto a succession constantscan be drawn and will be tangentat 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 the sectionis designated Q, it is apparentthat
n: r,'
f*t
udy
( 1 8.s 3)
Q : and
I d,t' J,t,^
(18.s4)
Q:QtQ,
(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 The curvesin Fig. 18.2 designated @rand Q,, calledeqaipotentiallines, ate 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
(18.s6)
(18.s7)
dy dx
u
t)
(18.58)
and sf The lineq Thusequipotential arenormalto flowlines. rystem flowlines equiponet. tentiallinesformsa flow
447
'
is One significantpoint of differencebetween{ and ry'functions that equipotenis irrotational. For twodimensional flow the tial lines exist only when the flow condition of irrotationality is said to exist when the z componentof vorticity (. is
ZA[O; Ot
: t,: (x,4) o
'
dQ
(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 Once either streamlines equipotentiallines in a flow domain are determined, of the other is automaticallyknown because the relations in Eq. 18.48.Thus r
and
(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 satisfies conditions of continuity and irrotationality simultaneously. Equations presentedin this section have been limited to the case of twodimensional flow. Extension to three dimensionswould be obtained in a similar fashion.
surfaces). Impervious boundariesmay be artificial objects zuch as.concretedams, rock boundary In strata,oi soil stratathat arehighiy impervious. Fig' 18.3the impervious
448
CHAPTER18
OF FLOW MECHANICS
AB represents sucha limit. Sinceflow cannotcrossan imperviousboundary,velocity In components normal to it vanishand the imperviousboundaryis a streamline. other : 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(hy) The definition of a velocity potential statesthat (18.64)
o :  * \ v *' r )* . ' (z /
in for Substituting pressure Eq. 18.65yields
(18.65)
o:*l
and
*r]*t
(18.66) (18.67)
and surfaceBC, often termed a reservoir boundary, is an equipotentialline. CD The free surfaceor line of seepage in Fig. 18.3 is seento be a boundary zones.Since flow doesnot occur acrossthis unsaturated betweenthe saturatedand along this free surfacemust be boundary,it is obviouslyalso a streamline.Pressure 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 DE The surfaceof seepage of Fig. 18.3 represents location at which water downstreamface of the dam and trickles toward point E. The seepsthrough the is The surfaceof seepage neither a flowline nor an pressurealongDE is atmospheric. line. equipotential
ImPervious laYer
449
18.9 FLOWNETS
of Flow nets, or graphicalrepresentations families of streamlinesand equipotential 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 that eachunit b.ounded a pair of streamlinesand equipotentiallines is approximately 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 the literature and are not discussed in usingthe geometryof it After a flow net hasbeenconstructed, can be analyzed the net and by applylngDarcy's law. that the hydraulic that Remembering h : p/y + z, we find thatFig' 18'4 shows 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 whereLm represents crosssectional plane of the diagram. If the flow net is constructedin an orthogonalmanner and
h Ll
Equalpotential lines (0 = constant) Lm
r d " Qz
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
sections If wherc his the total headloss over the n spaces. the flow is divided into m per unit width of the medium is by the flowlines,then the discharge
o:iu:Y
(r8.72)
can when the medium's hydraulic conductivityis known, the discharge be computed 18.72 and a knowledgeof flow net geometry' using Eq. the Where the flow net has a iree surfaceor line of seepage, entranceand exil of discussion these given in Fig. 18.3 are useful. A more comprehensive conditions 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 to De'Wiestmay be placed in three separatecategories.a the first classification lines do not intersectat right angles'Sucha situationoften and equipotential flowlines with a flowline; PointA in Fig. 18.5isanelamnle. occurswhen a boundarycoincides has a discontinuityalongthe boundary that abruptly The secondclassification such the changes slope of the streamline.In Fig. 18.6 PointsA, B, and c represent B it is zero' At discontinuities. PointsA and c the velocity is infinite, while at Point direction insidethe flow in of discontinuitymeasured a counterclockwise If the angle the than 180o, velocity is zero; if larger than 180o'it is inflnite' The angle field is less examPle. atA is 270",for net' The third categoiyincludesthe casewherea sourceor sink existsin the flow of the flow net Under these circumstancesthe velocity is infinite, since squares
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. in a practical senseand are discussed representsinks and sources
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 crosssectional 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 andB is shownon the figure to be equalto both Lh" andA,hu, A and sincethere can be only a singlevalue,  Lh": Ah6 in Introducingtheseexpressions Eq. 18.73produces
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 K2 sideof the boundarywill no longerbe squares the equipotentialline spacing 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)
18.11ANISOTROPY
on In many cases hydraulic conductivity is dependent the direction of flow within a 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 representative average
book.35'r2
THEORY 18.12DUPUIT'S
free surfacecan be analyzed Groundwaterflow problemsin which one boundaryis a flow. This theory is foundedon two on the basis of Dupuit's theory of unconfined is only slightly madeby ptptl in tgO:.t' First, if the line of seepage assumptions equlp:tenand, correspondingly, horizontal r""y b" considered inclined, streamlines tl" slopesof the line of seepage utfd tial lines will be essentiallyvertical. Second, at" known to be satisfactorily hydraulic gradient *r "fit. Whel fleld conditioos to Dupuit's theory by represented theseusruption*, the resultsobtainedaccording "o'pur"veryfavorablywiththosearrivedatbymorerigoroustechniques' into a mathemdtFigure 18.8 is u,"fol in translatingthe foregoingassumptions 'ateadx dy th" figut" which has a base 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  K  a h. d Y Qo: *h expansionas The outflow at x = dx is obtainedby a Taylor's series K*.h  " + o . . dv _ _ *(**n ,_ Qor: dr 6x\ Ax,. aY\+ ' ' _ /
(18.7e)
(18.80)
'
454
CHAPTER18
MECHANICS OF FLOW
6x/
tt*.
or
(r8.
wheredx anddy are considered in fixed lengths.A similar consideration the y tion yields
r,o,:ryy*W)
i
, (18
; )
Assuminsthat thereis no movementin the vertical direction.theseare the components the inflow and outflow.Furthermore,still dealingwith steadyflow, of I changein storagemust be zero. As a result,
o8
or
(18.
Consequently,. Laplace's equation for accordingto Dupuit' s assumptions, functionh2 must be satisfied.la In the particular casewhererecharge occurring as a resultofinfiltrated is reachingthe water table, a simple adjustmentmay be made to Eq. 18.85. If recharge intensity(dimensionallyLTt) is specified R, thenthe total recharge as 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 onedimensional 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 any position x along at imperviousboundary.For the onedimensional 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 d2h2 ;;: 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)
h2=2f*+nt
(18,e5)
456
18 .CHAPTER
MECHANICSOF FLOW
(often calledDupuit' s parabola)' It This is the equationof a free surface. is a parabola 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 = L o l'  n ? ) 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
K(h'Lh7) :ry
r
* QoL
RL 2
1)^::
K(h? h'r) z L
PROBLEMS
457
K(h'^ h?\ . ,) 2L R 0 . 0 1 7 . 5 : 0.075gpd/ft2 _ o : 0.075(1000 soo)+ 8(50, 40r) 2000 0.075x 500 , 8 x 9 0 0 2000 3 7 . 5+ 3 . 6 4Ll gpdlftz T I
R(.
r summary
Understanding movementof groundwaterrequiresa knowledgeof the time and the space dependencyof the flow, nature of the porous medium and fluid, and the boundariesof the flow system. particular, groundwaterdevelopment In and management dependon understanding storage propertiesofthe associated the soils and rocks ' and the ability of thesesubsurface materialsto transmit water. Fundamentalto the mechanics groundwater of flow is Darcy's law (Eq. 18.3).Usingthis equationalong with a knowledgeof the hydraulic conductivity K, estimates flow can be had. The of 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 temperature 65"F and the velocity is 0.8 ftlday. the is 18.4. A laboratory test of a soil gives a standardcoeff,cientofpermeability of 3.8 x 102 gpdlft2.If the prevailingfield temperature 60"F,find the field coefficientof permeis ability. " 18.5. ReworkProblem 18.4 assuming is 3.8 x r02 gpd/ft2and the temperature 65'F. K" is 18.6. Given the well and flow net datain the following flgure, find the discharge usinga flow net solution.The well is fully penetrating; : 2.87 X 104 ftlsec, a: 180 ft, K b = 43 ft, and c : 50 ft. 18.7. ReworkProblem 18.6assuming : 8.2 x 105 mlsec,a: 85 m, b : 2l m, and K c:26m.
458
CHAPTER18
MECHANICS OF FLOW
Stagnation flowline
Axis of
5 4 3 2 I
symmetry 1 2 3 /
K: 1 8 . 8 . R e w o r k P r o b l e 1 8 . 6 a s s u m i n g8 ' 4 X 1 0  5m l s e c , a : 1 0 0 m ,b : 2 2 m , a n d m 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 103 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: McGrawHill, 1970. William C. Walton, GroundwaterResource New York: Wiley, 1965. R. J. M. DeWiest,Geohydrology. New York: McGrawHill, 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
. Presentmethodsfor calculatingconfined and unconfinedsteadyradial flow toward a well. . Describeproceduresfor dealingwith unsteadygroundwaterflow conditions. . Describea method for estimatingflow to an infiltration gallery. Groundwater collectedprimarily by wells, althoughinfiltration galleriesare someis times usedwherethe circumstances appropriate.tOutflows from natural springs are are also amenable collection,but once thesewatersexit the ground, they become to 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 applyingDarcy's law. , by
19.1 FLOW TO WELLS A well system be considered composed three elementsthe well structure, can as of 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 locationssomedistancefrom the well at in order to replenishthis withdrawal. Because flow resistance offered by the soil. of 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 steadystate conditions are established.
461
Impervious
by of The hydraulic characteristics an aquifer (which are described the storage permeability) can be determinedby laboratory or field tests' coefficient and aquifer "o111*onlyusedfield methodsare thp applicationof tracers,the use The three 111ort of A tests.3 discussion aquiferper!o.r. uni of field permeameterr, aquiferperformance for of given here alongwith the development flow equations wells.2'a's mancetestsis testsmay be either equilibrium or nonequilibriumtests.In Aquifer performance mustbe stabilizedfor a flow equationto be an equilibrium testthe coneof depression tesithe derivationincludesa condition that steadystate derived.For a nonequilibrium tests Adolph Thiem publishedthe first performance conditionshavenot beenreached. in 1906.6 basedon equilibriumconditions
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
(re.2)
Jt
462
Q n f1 _ 2
and
( 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 of estimates Q or Krcanbe obtained usingEq.19.4 or tr9.5,even ho,reasonable at if the heighth, is measured the well peripherywhere1L: r*, the radius of the well boring. EXAMPLE 19.1 an An 18in. well fully penetrates unconfinedaquifer of 100ft depth.Two observa: 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 = rwQ35l1'00) log(r2/r') hz: 10021=79ft
ht=
Q :
K(h? h?)
o : 2nxmXr!x "d
Integrating,we obtain h"h, O:2rrK'm:# " ln\r2/h)
(1e.6)
(re.7)
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 steadystate 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')
464
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 number of elementsthat approachlittle squares shape.Once the net is complete, Uy if can be analyzed consideringthe net geomtry and using Darcy's law in the manner of Todd.3 EXAMPLE 19.3 to Find the discharge the well of Fig. 19.3by using an applicableflow net. Consider as the aquiferto be 35 ft thick, 4 : i.65 x 1d4 fps, and other dimensions shown' Solution. UsingEq. 18'72,we find that
n : * *n'
whereh:35*25:60ft m : 2 X 5 : 1 0 , n:I4 3.65X104x60x10
19.5 WELL FIELDS When more than one unit in a well fleld is pumped,there is a compositeeffect on the is This consequence illustratedby Fig. 19.4in which the conesof free water surface. are depression seento overlap.The drawdownat a given location is equalto the sum 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 'n If the drawdownat a given point is designated m, andsubscripts particular well(e.g ., mt fefersto the drawdown 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 wellfieldgeometryand 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
466
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 softenedbecause inEq. 19.10is not very sensitiveto is Q are 16.Equation19.11shouldbe usedonly wheredrawdowns relativd small. for For flow in a confinedaquifer the expression combineddrawdownbecomes
n
h o  h : 2 =Q:, ,n'o'
z1rKm
f1
(r9.r2)
Equationsfor well flow covering a variety of particular wellfield patternsare reported in the literature.3'7 Those given here are applicablefor steadyflow in a homogeneous isotropic medium.
19.6THEMETHOD IMAGES OF
Some groundwaterflow problems subjectedto boundary conditions negating the fitting these into infinite systems direct useof radial flow equations be transformed can equations applyingthe methodof images.2'8'e by When a streamis locatednear a pumped well and the streamand aquifer are interconnected, drawdowncurve of a pumpedwell may be affectedas shownin the 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 boundaries supplanted Streamsare replaced recharge by wells while impermeable for by pumpedimagewells.Computations the caseof a well and imperviousboundary directly follow the procedures outlined under the sectionon well fields.For the well to discharge. and streamsystem, recharge the imagewell is considered havea negative 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
467
468
negligible,and changes aquifer storage in 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
(1e.13)
in which Q : (constant) pumping rate (L3Tr units), Z : aquifer transmissivity (LtT 'units), and a is a dimensionless variabledefinedby
, ,s" u : r4tT
(re.r4)
where r is the radial distancefrom the pumpingwell to an observation well, S. is the aquiferstorativity(dimensionless), r is time. The integralin Eq. 19.13is usually and known asthewellfunctionof u andiscommonlywritten asW(u).It maybe evaluated from the infinite series
(1e.1s)
(1e.i6)
The basic assumptiops employedin the Theis equation are essentiallythe sameas thosein Eq.I9.7 exceptfor the nonsteadystate 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:
(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 comparing loglogplot of a versus a by I4z(z) known as i type curve, with a log1ogplot of the observeddatar2ft versuss. In plotting type curves,W(u) ands are ordinates, and rt ft areabscissas. two curves The z are superimposed movedaboutuntil segments and coincide.In this operationthe axes must remain parallel. A coincidentpoint is then selected the matchedcurvesand on 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 VALUES U OF TABLE19.1 VALUES W(U}FORVARIOUS 'L0 5.0 4.0 3,0 2.0 XI x 10r x 102 x 103 x 104 x 105 x 106 x 107 x 108 x 10e x 1010 x 1011 x 10t2 x 1013 x 1014 x 1015 0.219 t.82 4.04 6,33 8.63 10.94 t3.24 15.54 17.84 20.15
aa A<
469'
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
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*Supply Paper 887'
values for data. Inserting these values in Eqs. 19.13 and 1,9.1'4and,rearranging, found' transmissibilityI and storagecoefficientS" can be "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 and the expression Zbecomes 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 with rothe time corresponding zerodrawdown. of manipulation Eq. 19'13.
470
, = t4
r\too'
= 4e,8oo sd/ft
tl
I I
I
for Usingthe following data, find the formation constants an aquiferusinga graphical 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
't I  , I W(u)vs. u
E 0.3
5 6 7 891x105
I rrzldav)
equation' to solution Theis's ' 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 114.6 540
^ 1.9 :
91,860gpdift
uT
1.87r'1t
7"andstorage of in usingthedatagiv.en Fig. 19.6,findthe_coefficienttransmissibility : 300ft' Q: 1000gprn andr S, coeffrcient for an aquiflr, given Solution.FindthevalueofAhfromthegraph,5.3ft.ThenbyEq.9.20 2640 264 x31000 T 
T: ff:
s":
O.3Tto r'
472
Note from Fig. 19.6 that ts : 2.6 min. Converting to days, we find q6il)ti\l becomes
, 0 
and
s":
(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 104, and
gpm, 0 = 3oo
473
AQUIFERS 1e.8LEAKY
havedealtwith free aquifersor thosecopfinedbe#een imperThe foregoinganalyses t"ulity, many casesexist wherein the confining strata are not comvious strata.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_)
474
X+
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. is It thereforeapparentthat the computedparaboladoesnot accuratelydescribethe real water table. The differences,however,are small except near the point of outflow, providing the initial assumptions satisfled. The calculateddischarge approximates are 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 103ftlsec,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(3.25 103)(20 20  3 x 3)/50 x x : 0.072 cfs, the flow into one foot of gallery I r \
19.11SALTWATER INTRUSION
The contaminationof fresh groundwater the intrusion of salt water often presents by a seriousquality problem. Islands and coastalregions are particularly vulnerable. Aquifers locatedinland sometimes contaiq highly salinewaters as well. Freshwater
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 whenlongtermwithdrawls will of Lasting supplies groundwater be assured period. The potentialof a groundduringthe corresponding arebalanceJby recharge by water basin can be assessed employingthe water budgetequation,
)r)o:As
where the inflow ) l includesall forms of recharge,the total outflow ) O includes duringthe accounting in the everykind ofdischarge,and AS represents change storage forms of recharge and dischargeare those listed in 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 the competentto evaluate consequences imposing 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 drafts, however,can depleteundergroundwater supplies a Excessive encountered. The mining of waterwill ultimately is not feasible. point whereeconomicdevelopment depletethe entire supply. AND OF TONT"TSRECHARGE DISCHARGE 19,2 SdUE TABLE
Recharge from streams, Seepage Ponds,lakes inflows Subsurface Infiltrated precipitation Water rechargedartificallY Discharge to Seepage lakes, streams,spnngs outflows Subsurface Evapotranspiration Pumping or other artificial meansof collection
i_
476
r Summary
primarily through the constructionof is The collection of groundwater accomplished influence the numerical estimation"of their performance. wells, and many Tactors to Some situation, ur" amenable solutionthrough the utilization of relatively simple applicationof the hyOthers dependupon sophisticated mathematicalexpressions. of undervariousConditions nonuniformity of aquifermaterials drodynamicequations and a variety oiboundary conditions.The readeris cautionednot to be misledby the and simplicity oi ,o" of thqsolutionspresented to observethat many of theserl\ situations.(. are not applicableto all groundwaterflow to speciaiconditions and ,) of water through the ground is of a different magnr'tudd The rate of movement than that through natural or artificial channelsor conduits.Typical flow ratesrange of the from 5 ft/day ti afew feet per year. Theselow ratesof flow exacerbate impact groundwatersourcesand complicate cleanup since natural contaminani spills on flushing from the site may take many yearsto occur' in The methodsdescribed this chapterfor estimatingflows to collection devices principlesof fluid flow embodiedin Darcy's law. Applicafions arebasedmainly on the in arelimited to flowsin the laminar range,but undermostconditionsencountered the the describing mechanics Examplesof the ure of equations field, Darcy's law applies. of flow to wells andinfiltration gilleries were given in this chapter.Both steadystate as and unsteadvflow conditions were addressed well'
PROBLEMS
coefflcient of 19.1. A 12in. 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 l2in. 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 ft thick. The steadystate aquifer is composedof medium sand and is 100 in an observationwell 75 ft awayis 14 ft, and the rate is 1200 gpm. The drawdown well 500 ft awayis l.2ft. Find Kin gallonsper day drawdownin a secondobservation 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:8.0x10e u:8.2 X 10e
'
PROBLEMS 477
19.6. ' Determinethe permeabilityof an artesianaquiferbeingpumpedby a fully penetrating pumping of well, The aquifercomposed medium sandis 130ft thick. The steadystate is 1300 gpm. The drawdownin an observbtionwell 6{ft awayis 12 ft, and in a rate secondwell 500 ft awayis 1.2 ft. Find Ky in gpdlft2. 1g.7. Considera confined aquifer with a coefficientof transmissibilityT : 700 ft3ldayft. : Atr : 5 minthe drawdown 5.1ft; at50min,s = 20.0ft; at 100min,s = 26'2ft. of The observationwell is 60 ft from the pumping well. Find the discharge the well. pumpedat atate of 300 gpm is confined and pumping 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 1o4
fcl(dar(fo r : 650
a : 8.0 x 10e u : 8.2 X 10e u : 8.6 x 10e
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 18in. 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 of 700 gpm. The well fully penetrates aquifer. Watpr levelsin observationwells 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+
478
DEVICES ANDCOLLECTION 19 CHAPTER WELLS "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
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 A241n.diameterwell penetrates full depthof an unconfinedaquifer.The original belowthe watertable and a bedrockaquifugewere located50 and 150ft, respectively, land surface.After pumping al arate of 1700 gpm continuously for 1920 days. and the original water levels in equilibrium drawdownconditions were established, 100 ft from the center of the pumpedwell were observationwells located 1000 and lowered 10 and 20 ft, respectively. a. Determinethe field permeability (gpdlftz) of the aquifer. b. For the samewell, zero drawdownoccurredoutsidea circle with a 10,000ftradius 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 100ft depth of a saturatedunconflned aquifer. The using drawdownat the well casingis 40 ft whenequilibriumconditionsare established of a constantdischarge 50 gpm. What is the drawdownwhenequilibriumis established of using a constantdischarge 66 gpm? by a long rainless period, the flow in Wahoo Creek decreases 8 cfs from 19.16. After an 8 Memphisdownstream mi to Ashland.The streampenetrates unconfinedaquifer, 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 towardtheLincoln wellfield at 0.00095. Computethe transmissivityof the aquifer knowing Q : TIt, where1 is the slopeand I is the length. 19.17. The timedrawdown 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
Drawdown (ft)
Time
(h0
9.8 12.2 t4.7 16.3 18.4 21.0 24.4
Drawdown (ft)
level of the water table for an unconfined 19.18. Over a 100mi2surfacearea, the average
of aquiferhas dropped I 0 ft because the removalof 128,000 areaftof waterfrom the coefficientfor the aquifer.The specificyield is 0.2 and aquifer.Determinethe storage the porosityis 0.22. 19.19. Over a 100mi2 surface area, the averagelevel of the piezometric surface for a confined aquifer in the Denver area has declined 400 ft as a result of longte1m pumping.Determinethe amountof the water (acreft) 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 104, andQ 280 gpm. 19.21. Find the permeabilityof an artesianaquiferbeingpumpedby a fully penetratingwell. pumpof The aquiier is 130 ft thick and is composed medium sand.The steadystate ft awayis 12 ft' and ing raie is 1300 gpm. The drawdownin an observationwell 65 well 500 ft awayit is 1.2 ft. Find Kyin gpdlft2. in a second 19.22, An 18 in. well fully penetratesan unconflnedaquifer 100 ft deep.Two observation 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 : 0.001,find the drawdownin the obser: 100meters,and the storage coefficient r vation well for a time period of (a) t hr, and (b) 24 hours' = 0.0028mzls, 19.24. A well is being pumpedat a constantrate of 0.003 m3/s.Given thatT = 0.001, and the time sincepumplng beganis 12 hours,find the storage"oiin"i"nt well for a radial distanceof (a) 150m, and (b) 500 m' the drawdownin an observation
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
r Prologue
of The PurPose this chaPteris to: ' Describethe featureS largescalegroundwatersystems. of to . Introduce the principlesof finite differenceapproaches modelingregional groundwatersystems. . Iilustrate the application of groundwatermodeling techniquesto the Upper Big Blue basin in Nebraska. flow of The analytical methodsdescribedso far havebeen applicablemainly to the regionalgroundwarcr rc individual wells. In this chapter,the conceptsof analyzing are are water systems introduced.Suchanalyses requisitesfor the wise development, management,andoperationofexpansivegroundwaterresources' mustbe dealt with jointly in most quality aspects Given that waier quantity and modelsmrrstoften prol"tt"t, t"gional groundwater decisionmaking waterresources of aspects groundThe to be designed includeboth of tfuse dimensions.l2s fluid flow are complexand in water modelsare presented this chapter.Solutetransportmodels. this book, but a biief introduction to them is given at the end of the scopeof beyond and role of the chapter.It is important for the readerto understandthe importance thesewater qualityorientedmodels'
482
CHAPTER20
SYSTEMS MODELINGREGIONALGROUNDWATER
Approximate equations numerically resulting in a matrix equation that may be solved using a computer
t_
model' a Figure 20.1 Logic diagramfor develqping mathematical OH') Worthington' WellAssociation, Water (Courtesy theNational of
a the Figure 20.1 characterizes procedurefor developing deterministicmathematical model. A conceptual model is formulated based on a knowledge of the of of characteristics the region and an understanding the mechanicsof groundwater flow. The next stepis to translatethe conceptualmodel into a mathematicalmodel, accompanied by usuallyrepresented a partial differential equationor setof equations 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 quality and/or heattransIf dimensionality(one,two, or threedimensional). water describing are fer considerations to be incorporatedin the model,additionalequations of energyare involvedand conservation of conservation massfor the chemicalspecies required.Typically usedrelationsare Fick's law for chemicaldiffusion and Fourier's 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 to or suchasfinite differences dnite elements represent applicable the regionis divided for In tial equations. using a finite differenceapproach, example, as into grid elementsund th" continuousvariablesare represented discretevariables at the nodal points. In this manner,the governingdifferential equationis replacedby that a finite nub"r of algebraicexpressions canbe solvedin an iterativeway.Models of this type find wide applicationin the estimationof sitespecificaquifer behavior. They havepro\iento be effectiveunderirregularboundaryconditions,wherethereare and heterogeneities, where highly variablepumping or rechargerates are expected'l Severil types of groundwater models and their applications are summarizedin Figure 20.2. region' A numberof stepsmustbe followedin modelinga targetedgroundwater They may be Figure 20.3 is illustritive. The f,rst step is to define the boundaries. ptiysical,suchas an imperviouslayer, or arbitrary, suchas the choiceof a politically,
Applications Watersupply
Regionai aquifer analysis Nearwell performance Groundwatelsurface water interactions Dewatering operauons Seawaterintrusion Land fills 'Geothermal Thermal storage Land subsidence
Waste injection
Figure20.2Typesofgroundwatermodelsandtypicalapplications.(Courtesyof OH') Worthington, WellAssociation, theNationalWater by Next, the region is divided into discreteelements or otherwise,definedsubregion. grid (seeFigute 20'4)' a superimposing rectangularor polygonal ' (s. and z) and the once thegrid is determined, controlling aquiferparameters If grid element. solutetransportis includedin the the initial conditionsare setfor each suchai hydrodynamicdispersionpropertiesmust also model, additional parameters hive beenmet, the modelcanbe operated Atter atl of thesespecifications be specifigd. comparisonsof and its output comparedwith recordedhistory (history matching). with counterpartmodelpredictionspermit recordedvaluesof headand other features and computeddata are considered to parameteradjustments be madeuntil observed by the modelerto be in closeagreement. to analyzea Upon completion of thelodel's calibration, it can be applied options. The model's prediction of the and/or development variety of management to decisionmaking outcomesof these alternative strategiescan be a valuable aid include: the problems that can be addressed Examplesof the types oI processes. of use; the impact on an aquifer of utilty of an aquifer to suppoit various levels storageof varying naturaiand artificial rechargerates; the effects on underground contamrate of movementof subsurface and spacing, pumpingrate; the weli location, intrusion. inants;and saltwater caution while numerical groundwatermodelshavemuch to recommendthem, Prickett and interpretedappropriately. to must be exercised enJurethat they are used three waysin notesthat overkill, inappropriatepiediction, and misinterpretationare To avoidthesepitfalls, both the modeler #"t, can Uemisused.s which groundwater
484
CHAPTER20
MODELINGREGIONALGROUNDWATER SYSTEMS
illiiT3:':,",*itlffi,ffi:ff tlSl
upon which the model was and user must understandthe underlying assumptions founded,its limitations, andits sources errors.Usedwisely,modelscanbe powerful of decisionmakingaids. Used inappropriately,they can lead to erroneousand sometimes damagingproposals.
20.2 FINITEDIFFERENCE.METHODS
Digital simulation requires an adequatemathematical description of the physical processes be modeled.For groundwaterflow this descriptionconsistsof a partial to 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 The model can simulateyears of physical activity in a span of seconds, that the
(a)
A.r Ly ..,V ^t ,
/_/  1"
', / 1 . 1 , / 4Finitedifference
/
/
srld DlocK
(b) (b) boundaries. FinitedifFigure 20.4 (a) Map view of aquifershowingwell field and At' is the spacing t i"r""n"" grid{oi aquifir study,wGre Ax is the spacTqii lhu *ection' nodes;open in tt" yat"rtion, and f is itre aquifer thicknesg.Solid dots: blockcenter worthingof the National Waterwell Association, circles:sourcesink nooo tcoo*sy ton. OH.)
486
CHAPTER20
SYSTEMS MODELINGREGIONALGROUNDWATER
of consequences proposedactionscan be evaluatedbefore decisionsinvolving construction or social changeare implemented.The expectationis that the model runs * will lead to wiser and more costeffective'decisions. of The finitedifferencemethodis basedon the subdivision an aquiferinto a grid with zonesof the aquifer.The equationthat must and the analysisof flows associated and be solvedis derivedfrom continuity considerationq Darcy's law for groundwater motion. This yieldsthe followirigpartial differential equation(a versionof Eq. 18.41), aquifer.Note that the equationpresented flow through an areally extensive describing here describes twodimensionalcase: the
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 the aquifer(I]lT) of sourceand sink term (L/T)
(20.r)
In the aboveequation,vertical flow velocitiesare consideredto be negligible are in everywhere the aquifer.The following assumptions implict in the derivation:the hydraulicconducfluid densityis constantin time and space; flow is twodimensional; within the aquifer; flow obeysDarcy's law; and the specificyield of tivity is uniform the aquifer is constantin spaceand time. Equation 20.1 is nonlinear for unconfined variable. transmissivityis a function of headand thus the dependent aquifersbecause to integrateEq.20.I, initial valuesof head, transmissivity,saturated In order and sinksmust thickness the aquifer,andthe amountsof waterproducedby sources of for every point in the region of the integration.The specificyield and be identified location of geometric boundaries must also be defined. Unfortunately, analytic It solutionsto Eq. 2O.l are impossibleto obtain exceptfor the most trivial cases. is necessaryto resort to numerical integration techniquesto obtain the desired thus
answers.t'toto
to Application of finitedifference techniques groundwaterflow problems reor quiresthat the region of concernbe divided into many small subregions elements valuesof all the variablesin (Fig. 20.5). For each of theseelements,characteristic which to Thesevaluesare assigned the centersof the elements, Eq. 20.I are specified. nodes.The headsin adjacentnodesare relatedthrough a finitedifference are called equation,which is derivedfrom Eq. 20.1. Thesedifferenceequationscan be derived The or by an appropriateTaylor's seriesexpansion by massbalanceconsiderations.8 to equationscan then be solvedsimultaneously yield the headsat resulting algebraic eachnode for eachtime stepconsidered. that the simulation methodspresentedin this chapter It should be understood pointed toward the analysisof regional rather than localized groundwaterprobare the lems suchas the prediction of the drawdownat a particular well. In suchcases, Here we,are in Chapter 19 are usually the most appropriate. methodsdiscussed that might occur over a large area with waterlevel or headchanges mainly concerned prescribedwaterusepractices. due to
t, I
o m,1
J
{
eAx'
Node
Subregion
I Ay
l,,i
I
\ \ l, nd Boundary ofregion of integration a m,n
BoundaryConditions
In order to integrateEq.20.l,the governingboundary conditionsmust bespecified' in here.Otherswere presented ChapTwo typesof bJundary condition aie discussed ters 18 and 19. chosen Where the region of integrationis limited by a political or arbitrarily boundarycondition'10 constantgradient boundary,it is oftei the policy to employ a change In this "ur", unar.umption is madethat the gradientof the watertable will not llvel may rise or fall. Where streamswith alongthe boundaryeventhoughthe water streamboundarycondiare system encountered, to interconnections the groundwater mathematicallyas are expressed boundaries tions are employed.Constantgradient
a h l \
(20.2)
*:81x'!) g\x, y): a constantspecifiedat the location x, y throughoutthe period of 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 constanthead the Darcy equation:ro 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 lefthandsideof the elementalregiondepicted.The node i  I, j lies outsidethe regionof integrationand may An that no information aboutit is available. assumption thus it may be assumed the across boundary be madeto circumventthis problem.It is that the transmissivity is uniform and equal to Tt,t. In finitedifferenceform the head changeterm in Eq. 18.26 can be statedas
ah _hi.jhil,j
6x
A,x
hi*r,j 
(20.s)
(20.6)
f,,,u#'LY Q;r/",i:
(20.7)
Boundary where 6 h= r . dx
At the beginningof each time step, a new volumetric flux is calculatedalong each by boundary.This is accomplished usingthe headsand transmissivconstantgradient previoustime interval. . ities computedin the boundariesin groundtreatedas constanthead are sometimes Surfacestreams body wherethe waterlevel in the surface is waterproblems.The assumption adequate In during the time period of the modelingprocess. remain unchanged is expected to are flows,and henceheads, significantlyaffectedby however,surface many instances, They may then groundwatersystem. to withdrawalsor recharges the interconnected the To system. accommodate of water supplyfor the groundwater. be a limited source This expression linkage, a leakageterm may be applied.to watergroundwater 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 When Eq. 20.8 is used,the streamis considered coverthe entire areareprenode. After eachtime step the leakagefrom the streamto the sentedby the related are aquiferis calculatedand streamflows depletedaccordingly.If the streamflowat a particular node becomeszero, the model can be madeto note that the streamis dry and break the hydraulic connection atthat point.l0
490
CHAPTER20
w
IlDconstantatt>0
Confined aquifer
l+Ay+l (After groundwater case. flow Figure20.7 Gridnotation a onedimensional for McWhorter Sunada.8) and suchasthat Let us considera onedimensional flow in a confinedaquifersystem illustratedby Fig. 20.7.lt is assumed that the flow is unsteadyand that the flowlines On are parallel and not time dependent. this basis,a unit width of the aquifercan be 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 (Qrt,) from I element  1 to element minusthe outflow{Qt,*r)fromelementlto element * I I I 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 thicknessand that it is homogeneous isotropic. Thus the valuesof K, ,S, and Ay 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
(20.e)
Usingthis notation ln the wherethe subscript represents total numberof grid elements. and Fig. 20.7, we can seethat the flow from elementi  1 to I is
. Oiri: UOO' ;i '
(20.10)
where i : the elementnumber time n : the selected It in as Equation20.10is recognized Darcy's equation. is assumed this representationthat the head generatingthe flow at time n is the differencebetweenthe averageheadsat the two adjacent elementsdivided by the distancebetweentheir as exactness Ay diminishesto zero. centers(nodes).This approximationapproaches
area of flow and is in The ireaA appearing Eq. 20.10 is the crosssectional 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  " i  t h,! 7"i '
Ay
(20.11)
for where 7 : Kb. A similar expression the flow from element ito i * 1 may be obtained:
(20.r2)
the Equations20.11 and 20.12 represent inflow and outflow from elementi. Considering 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
fr:'.(ry)
' (zo.tz)
1zo.r+i
(20.1s)
which is known as the explicit or forward difference form of the finitedifference as equationif n is designated the current value of time. If, on the other hand, n is definedas / + Ar, thin the equationis the implicit or backwarddifferenceequation. The Eachof theseforms hasits own solutiontechniques.8 explicit solutionto Eq. 20.15 here. will be discussed By letting n : t inF;q.2O.l and rearranging,one obtains
(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
here, case discussed an unstable condition. Irt the onedimensionalhomogeneous if stability is assured
<1 s/,vY t
TLt
(20.r7)
The equationshowsthat the choiceof time and spaceincrementsis not independent. an Satisfactionof Eq. 20.17 doesnot guarantee accurateapproximation,however;it for provides a stablesolution.8 only EXAMPLE 2O.T Refer to the onedimensionalflow 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 Using the : 0.02. As shown in the figure, there are five elements. 0.5 m/day and S 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 a v s 6
was obtainedusingthe relation The valueof Zused in the aboveexpression T:Kb: Z:0.5X2.0:1.0 a To be on the safeside,we shall choose time stepof 0.1 days,althoughany stability. value lessthan 0.16 would have assured
8m
493 20.3 FINITEELEMENTMETHODS ^' 2."For the first time step,t : 0.1, we can calculateh'a* and corresponding headsfor the other elementsusing Eq. 20.16.Thus
: h,;",
: h2,
For element3
#(h2,
+ + hg\ nZ'lt
#]
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 Finiteelementmodelersmust understand partial differential equationsand the calculus of variations.6 The finiteelementapproachoffers someadvantages over the finitedifference technique.Often, a smaller nodal grid is required, unA tnir offers economies in computer effort. The finiteelementapproachcan also accommodate one condition that the finitedifference 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 finitedifference approach cannot produce a solution, whereasthe finiteelement approach can. TLe finiteelementtechniquecan be used to simulate transient aquifeiperformance. A detaileddiscussion the finiteelementtechniqueis beyondthe scopeof this boor<, of but there are many good references the interestedreader.6'rsts.zi for ,
./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 on water use,and what when waterlevel declineswould posean economicconstraint would haveon thdrate of decline. and/or management impactsfuture developments represento The model developed explorethesefeatureswas a twodimensional 20.1, alongwith the tation of flow through an areally extensiveaquifer.loEquation appropriate boundary conditions, constituted the model. The region shown in was divided into a finitedifferencegrid and, after substitutionof the nodal n1g.Z0.S to ualu"t of Z and S, the model was operatedto predict waterlevelchanges the year Calibraof development. 2020 for variouspolicies of rechargeand for severallevels The model was operatedover usinghistoricdata. tion of the model was accomplished the period 19531972 using the known distribution of wells and the averagenet pupug" per well to establisha match betweenobservedand estimatedwaterlevel the simulation of future trends proceeded. "ttung"i. Once this was accomplished, process. in achieved the matching Figures20.II and20.12showthe correspondence waterlevelsin the study that it On the basisof the model studies, wasdetermined was limited to the l912level. It areawould continueto declineevenif development groundwasevere wasfurther predictedthat someparts of the areawould experience by employingartificial ter shortageiby the year 2000. It was found, however,that the To rechargefrethods, permanentgroundwatersuppliescould be assured. assess modeled.Both of these 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, was the cancellationof iiulut"d. The grosseffect of introducingthe rechargewells the effects of the proportionatenumber of pumping wells. Figure 2OI3 showsthe plan computedwaterlevelchangesat one location under a graduateddevelopment with no rechargeand.then (projected on the basisof the 1972rate of development) The continual for wlth graauateddevelopment eachof the three rechargeschemes. (curve 1) clear$ showsthe nature of downwardtrend in waier level with no recharge depictingthe threeartificial the problemin the UpperBig Bluebasin.The othercUrves such an approachis taken. ,""hurg" options show that stability can be achievedif it While the costs of implementing artificial rechargemight be excessive, is water table problem, short of apparentthat any longterrn solution to the declining sourceof water. reducinguse, would require a supplemental Operationof the modelprovidedusefulinsightsinto the natureof the watertable that problem and suggested irrigators shouldbe making someimportant water managementdecisionsabout their future mode of operation..^ ^In is The modeling of groundwatersystems complex.lo25 structuring models usuallybe made.Thesehave must simplifying assumptions suchasthat just discussed, to do with aquifer pbrameterssuch as transmissivity,specificyield (for unconfined Furthermore,the boundary aquifers),and storagecoefficient (for confined systems)' conditions are normally approximationsof what occurs in the physical system.and strataare someaboutthe uniformity of materialsin varioussubsurface assumptions to cannotbe expected yield models times crude.This doesnot meanthat groundwater must be cautiousabouthow useful results.It doesimply that the usersof the models aquifermodel suchasthat they interpretthe output.For example,an areally extensive to problem can be expected give by developed Huntoon for analyzingthe Blue River
o E
ZusI
E.l(
F \ $
o
a
H H
()
o
e
a
:d65'
\ o
\=e 1t "
rO
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
trends'(AfterHuntoon'10)' waterlevel and Figure20.1.2 Measured computed reliable information about waterleveltrends for various conflgurationsof developan ment. It shouldnot, on the other hand, be considered accuratepredictivetool for at monitoringthe waterlevelchange somespecificpoint in the regionof concern.This type of information could be derived only from a more detailed modeling of the the lolafif surrounding point. The informationprovidedby the Blue River modelwas what the future might hold for severaldevelopment to targeted showlocal landowners 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 unlessfuture development restrictedand supplemental 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 waterlevel 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. Scheme (After Huntoon.lo)
500
CHAPTER20
MODELINGREGIONALGROUNDWATER SYSTEMS
options.A,model suchas this, carefully usedand properly interpreted,can thus add a powerful dimensionto decisionmaking processes.
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 a changein storage of constituentis given by
,dc Pndt(e l q, t aph)c : ec. * q,c,
(2o.re)
(b)
model'(After of Figure 20.L4 Schematic the GelharWilson and Novotny Chesters.2a) where c: c; : c, : c: concentration coflcefltrationof the natural recharge concentrationof the artificial recharge of a firstorderrate constantfor degradation the contaminant
may be that dispersionis negligible.This assumption The GW model assumes objective of the model iS to estimateregionalaverage made on the basis that the The concentratiolrs.2e model also provides for the determinationof hydraulic and of Theseare measures the lag that occursin the times for the system. soluteresponse constituentinputs to the system.Gelhar and Wilson moue"ni of both water and of that the response an aquiferto a specificinput can be likened to that of a assume wellmixed linear reservoir.Their studiesshowedthat the model's determinationof of leavingan aquifer is representative the average the concentrationof constituents that the model in the aquifer.On this basis,,itappears concentrationof the constituent
502
CHAPTER20
MODELINGREGIONALGROUNDWATER SYSTEMS
to stream, is well suitedto estimatingthe quality of groundwater discharging a surface providingihe aquifer is narrow relative to the length along which discharge occurs.
r summary
in is Groundwater a regionalaquifer system constantlyin motion. The amountstored evapotranspiration, flow to at any time is affectedby artificial and naturalrecharge, 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 longterm and extensive water withdrawalsfor irrigation is a good example.On the other hand, by inadvertently, humanintervention. waterlevelshavebeenmadeto rise, sometimes for Leaky irrigation canalsin central Nebraskawere at one time responsible groundwaterlevel risesin somefarming locationsof a magnitudesufficientto jeopardizeuse occur, they are of the land. Once major problemsof depletionor overreplenishment policy for groundwater has management not easilydealt with. In general,a safeyield 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 tial equationo setof equations, or describes hydraulicrelationswithin the aquifer. 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, is necessary rely on numericalmethodsfor modeling.Under it to suchcircumstances, approximatesolutionis obtainedby replacingthe basicdifferan ential equationswith another set of equationsthat can be solved iteratively on a computer.Both finite differenceand finite elementmethodsare applicable. The finite differenceapproachdescribed this chapterreplacesthe governing in Thesecanbe solvedon partial differential equations with a setof algebraicequations. at the computerto producea set of water table elevations a finite numberof locations 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 valuable adjuncts to decisionmakingproare Models 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 ftldayands : 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,GroundWater 1981. Well Association, 2. C.A.AppelandJ.D.Bredehoeft,"statusofGroundwaterModelingintheU.S.Geological ey Survey," U.S, Geol. Surv Circular 737(1976). "Utilization of Numerical Groundwa3. Y. Bachmat,B. Andres,D. Holta, and S. Sebastian, ter Models for Water ResourceManagement,"U.S. Environmental Protection Agency ReportEPA600/878012. "Contribution of GroundwaterModeling to Planning," J. Hydrol' 43(Oct. 4. J.E. Moore,
r979).
5. T. A. Prickett,
t2rr28(r979). r7(2),
6. R. A. Freezeand J. A. Cherry, Groundwater.EnglewoodCliffs, NJ: PrenticeHall,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 "PredictedWaterLevelDeclinesfor Alternative GroundwaterDevelop10. P. W. Huntoon, Rep.No. 6, Conservation in the Upper Big Blue River Basin,Nebraska,"Resource ments and Survey Div., University of Nebraska,Lincoln, 1974. "The Numerical Solution of Parabolic and and H. H. Rachford, Jr., 11. D. W Peacemen Indust.Afpl. Math. J.3,284I(I955)' Soc. Elliptic DiffetentialEquations," "Application of th Digital Computer for Aquifer 12. G. F. Pinder and J. D. Bredehoeft, 1968). 4(4), 1069 1093( Res. WaterResources Evaluation," Hydrol13. I. Remson,G. M. Hornberger,and F. J. Molz, NumericalMethodsin Subsurface ogy. New York: Witey, 1971. Techniques GroundwaDigital Computer for 14. T. A. Prickett and C. G. Lonnquist,Selected Evaluation,Illinois StateWater Survey Bull. No. 55, 1971. ter Resource
PARTFIVE
MODELING HYDROLOGIC
Chapter21
r Prologue
The purposeof this chapteris to: . Introduce the types and classes hydrologicmodels. of . 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 23, 24, and25. 22, Chapters ratesandvolumesof flow at any point of interestalonga stream Information regarding in is necessary the analysisand designof many types of water projects. Although many streamshavebeen gaugedto provide continuousrecords of streamflow,planinformastreamflow facedwith little or no available are nersand engineers sometimes must rely on synthesisand simulatlon as tools to generateartificial flow tion and regardingstructuresizes,the effects of for sequences use in rationalizing decisions water supplies,water quality, and the effects of control measures, land use, flood natural or inducedwatershedor climatic changes. making in both the designand operationof largescale The problemsof decision have and canals,aqueducts, water supplysystems reservoirs, of systems flood control to suchas simulation and synthesis mathematicalapproaches resultedin a need for project. Simulationis definedas the mathematicaldescriptionof the investigate total 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 or hypotheticalstorm; or simply fllling in the missingvaluesin a 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 To cHAprER rNTRoDUcroN HyDRoLocrc MoDELING streamdowsynthesis stochasticmethods.Chapters2325provide detailsregardby ing determinislic continuousmodels, singleeventmodels,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 theseparameters. overviewof synthesis An of techniques presented Chapter22. is in
Code name
Model name
I 10 10 10 3 5 10
r990 1973 I 965 t972 r972 t971. t972 r974 t970 I9'71 1972 1978 1979
Rainfallrunoff eventsimulation modelsChapter HECI Flood HydrographPackage Corps Computer Program for Project Hydrology scs USGS RainfallRunoff Model USGS Hydrologic Model Computer Language ARS Storm Water ManasementModel EPA
10 1 5 z 3 5 5 I 5 5
SIMULATION509 21.1 HYDROLOGIC trials Theseare normally validatedby repeated inputs that arejudginent pafameters. are primarily eventsimulationmodelsbut with the models.The urban runoff models modelsare deferred of the havebeenisolatedin Table2 1. 1 because descriptions urban 25. to ChaPter streamflowsimulationmodelsshown Severalof the major eventand continuous 23 and24.TheStanfordandHEC1 models in Table2!.1aredescribedin chapters Table 2I't are briefly For are emphasized. further referince, most modelslisted in "Models and Methmodels,in the publication alongwith about 100 other described, Fleming's text plesentscomplete Studies."1 ods Applicable to Corps of Engineers models'2 descriptionsof the SSARR, S'[iM, HSP, USDAHL' and other
'
510
CHAPTER21
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 Xhour 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 being analyzedthroughuseof are distributedparameterGeographical Information System(GIS) technologies. BlackBox vs. Structurelmitating 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 a blackbox in fashion In hydrology, blackbox models may sometimestransform "plants" into "medicine" even though the reasonsfor successare not clearly understood.For example, modelthat accepts sequence a a ofnumbers,reduces eachby 20 percent,and outputsthe resultsmight be entirely adequate predictingthe attenuationof a flood for waveas it travelsthrough a reachof a given stream.In contrast,a structureimitating modelwould be designed useaccepted to principlesof fluid mechanics and hydraulics to facilitate the transformation. Stochastic vs. Deterministic Models Many stochastic processes approxiare mated 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 historical inflows are of repetitive. Deterministic methods of modeling hydrologic behavior of a watershedhave becomepopular. Deterministic simulation describes behaviorof the hydrologic the cycle in terms of ma[hematicalrelations outlining the interactionsof variousphases of the hydrologicclcle. Frequently, modelsare structuredto simulatea streamflow the 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 fit the known period of data,additional to periods of streamflowcan be generated. Event'Based vs. Contlnuous Models Hydrologicsystems be investigated can in greaterdetail if the time frame of simulationis shortened. Many shorttermhydrologic modelscould be classifiedas eventsimulation modelsas contrastedwith seauential
'
511
or coniinuousmodels.An exampleof the former is the Corps of Engineerssinglemodel eventmodel, HEC1,3 and an exampleof the latter is the Stanfordwatershed which is normally operatedto simulatethree' by developed Crawford and Linsley,a might use four, five, oi *or" yearsof streamflow.A typical event simulation model perhapseven I min' a time incrementof t hr or have arisen water Budget vs. Predictive Models sevelal model classiflcations One important comparison of that distingtish betweenthe purposes the model types. precipito the model proposes predict future conditionsusing synthesized is whether tation and watershedconditions or model is definedas a model or set of in of inflows, outflows,and changes that simulationmodel studier advised that affirm the b use the Parameters example,meteorologicdat shed.For watershed'A water application amounts might be known for a given agricultural (ET) formula the budgetmodelwould be uied to determine correctevapotranspiration in the continuity equation parametersby testing a range of valuesuntil a balance or monthbyoccursfor all time increments.This is often performedon a daybyday the water budget model' month basis. once the ET parametersare derived from conditions'or farming predictive simulationsof diffirent crop patterns,meteorologic in relationships the model that the practicescould be performedwith the satisfaction corroboratehistorical water budget (precipitation outputsare measured dt require the simultaneous studies such as ET, infil ondarY Processes spatial distribution of water applications'
of Limitations Simulation
systems'some Becausesimulation entails a mathematicalabstractionof realworld systembehaviorcan^occur.The extent to which the degreeof *i,,"p,",entation of "depends on many factors' The test of a developed model and systemoutputs vary that the behavioris consisby demonstrating of simulationmodelconsists u"iifi"uiion the physicalsystem' tent with the known behaviorof resources Even verified simulation models have limitations in usesfor water of models will allow performanceassessments planhing and analysis.Simulation options,particularly optibut specificschemes cannotbe usedefficiently to generate once a nearoptimal plan is formulated^bysome mal plans, for stated objectives. effectivefor testing a othertechnique, limited numberof simulationruns are normally combinationsof decisionvariablesusing ranand improving the plan by modifying for Techniques generatingoptimal plans are dom or systeriaticsamplingtechniques. in described Section21'3' proceAnother limitation of simulationmodelsinvolveschangingthe operating beingmodeled'Programming of the system duresfor potentialor existingcomponents for example,requires u "o*pot", to handle reservoir storageand releaseprocesses' reprogrammingis rerules, and considerable large portions to define the operatin! qoir"a if other operatingproceduresare to be investigated.
t
512
oHAPTER INTRoDUoTIoNHYDRoLoGIc 21 To MoDELING A'fourth limitation of simulationmodelsis the potential overreliance sophison ticated output when hydrologicand economicinputs are inadequate. The techniques of operational hydrology can be used to obviate data inadequacies, these also but require input. Controversyover the use of syntheticdata centerson the questionof whetheroperationalhydrologyprovidesbetter information than that containedin the input.
Utilityof Simulation
Computersimulation of hydrologicprocesses severalimportant advantages has that shouldbe recognizedwheneverconsidering merits of a simulation approachto a the 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 systems of suchas the designof hydraulic structuresor the investigation of streambank stability.However,for the analysis complexwaterresourcesystems of 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 hand are frequently by achieved simulation.In addition, the systemcan be nondestructively by tested;prdposedmodificationsof the designs systemelementscan be testedfor feasibility or of comparedwith alternatives; manyproposals be studiedin a shorttime period. and can An often overlookedadvantageof simulation includes the insight gained by gathering,organizing,and processing data, and by mentally and mathematically the 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 secondphase model conceptualiTation, is 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) elements, formulating the comprehensive mathematics the techniques, (3) of and translatingthe proposedformulation into a working computerprogramthat interconnects all the subsystems algorithms. and 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 (2) modifying the algorithms as necessary to improve the actual or a similir system; the simulaaccuracy of the model; and (3) putting the model to work by carrying out tion exPeriments.
Model Protocoi
from recommendations adapted modelstudies, Five axiomsfor performingsuccessful by Friedrich,5are: 1. Evaluatethe data beforebeginning. t Document assumPtions. of 3. Plan and control the sequence computerruns' of 4. Insist on reasonableness output. 5 . Document,document,document. An Examining and evaluatingthe basic data are essential. annotated,bibliographic program \ record of the data ,orrr."* shouldbe maintained.It is alwaysgood adviceto of the numerical 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 aheadof the rainfall? Are can inconsistencies identify errors. Didthe runoff occur characters Do waterlevelsgraduallyvaried,or are there discontinuities? alphabetical be interpretedas missingor appearin th! data?Will blank valuesin the data sets For hyzeros?Will zeros in the data sets result in overflows (division by zeto)? the limits recommended fall drographrouting, doesthe time interval selected between of for staUltityand convergence the numerical method usedto solvethe differential equations? ^ of Assumptionsare also important to the success a simulation' Assumptions and additional assumpthe were madeby the p.og.u"iwhen developing mod91, deviation the tions aremadeby userJ.Fo, "*uple, a programthat calculates standard the sample size must be from an unbiased estimating equation assumesthat = 30 is often considered sufficiently lafge to validate ihe estimate.A value of N 24hr storm being used' as minimal. For a TP149 (Chapter 15) application, is a readingand by assumed the method?No computerprogram shouldbe usedprior to and becomingawareof the madeUy ttre programmer undershndingthe assumptions implicit in the hydrologicprocessthat was programmed' assumptions 'ih" lor cpst of simulationcan reiult in unnecessary runs and may enticeusers the information originally bstantively to purposeofruns) and working the plan can
r alproximate time and in a single useasa guideduringa simulationproject. combining severalinvestigations Some of the modelsavailable run is anotherway to conductarrefficient simulation. allowthis.Forexample,TR20(seeChapter24)allowsthegenerationofflood
514
21 CHAPTER
'
to from severalstormsat once.It is often desirable generatethe2,5, hydrographs location. at flood discharge a singlewatershed 10, 25, 50,100, and 500year is able to generatefar more output than the hydrologistcan The computer Most modelsincorporateoptionsallowingthe userto specifyoutputquantity. analyze. shouldbe madeof which specific In additionto controlling outpul, a predetermination performed. A tabulation of key output data can be developedto analyseswill be compile and evaluatetrends (and make coursgcorrections)after eachrun. Because many opportunitiesexist in deterministichydrologyis about 80 percent acbounting, water budgelbalances.If the total rechargeto an aquifer is simulation for assessing less than the total outflow and withdrawls, but simulatedwater tables are rising, a shouldbe made.Writing important conclusions checkof input and model parameters runs helpsdocumentthe study and guiderevisions 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 made,andreasons each generalunderstanding the work plan followed,decisions of of made,provide samples the input stateassumptions should run. The documentation the statehow sensitive resultsare and output,explaininput preparationrequirements, and documentreasonsoutofrangeparameand 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 code. the usedin the code and in preparingthe input. Include all assumptions List publishedor known limitations or rangesof the applicability of the model. the 7. Characterize uncertaintiesin the model; describesensitivitytests. 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.
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 unnecessary. routing of or the far rlgtriif either storage distribuiion were deemed by flow could be accompdbhed simple lumped parameter channelflo* o, overland of for flow equations discretesegments or techniques, solutionsof the unsteadystate algorithms and the channelcould be used.In other words,the selectionof techniques dependson the d9ere9of refinementdesiredas output to fepresenteachcomponent is justifled and also on knowledgl of the system.A distributedparameterapproach in described of Components modelsare information is adequate. only when available 2225. ChaPters
516
CHAPTER 21
3. Designstorm abstractions: evapotranspiration, infiltration, depression, and interception losses.composite curve numbers,infiltr_ationcapacitiesand parameters, indexes. @ 4. Landusepractices,soil types, surfaceand subsurface divides. 5. Waterusesites 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 variousreaches, hydrhulicor field data for or from which thesecoefficientscould be estimated. 3. Channeland valley crosssectional 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 or near the watershed. in 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 irrigation practices,including water useefficiencies. and 6. Crop types and water consumptionrequirements. 7. Pastconservationpracticessuchas terracing,insfallation of irrigation return pits, and conservationtillage. 8. Stock wateringpractices. 9. Presenceand types of phreatophytes stream valleys and along ditch in 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; stagedischarge, velocitydischarge, depthdischarge 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 wateruseallocationsdivertedfor 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 verificationof flood routing techniques be usedand neces3. 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. Elevationstoragecurves. 3. Elevationarea curves. 4. Normal, minimal, other pool levels' loss 5. Evaporationand seepage data or estimates' 6. Sediment,dead storagerequirements' 7. Reservoireconomiclife. 8. Flood control operatingpolicies and rule curves' weir and outlet equations,controls. 9. Outflow characteristics, recreationbenefit functions' 10. Reservoirbased L1. Costsversusreservoirstoragecapacities' and benefits. of 12. Purposes eachreservoirand beneficiaries
Assessments Nonmodeling
can be the After researching availabledata and information, the needfor simulation model can be If assessed. a decisionis made to proceed,the appropriatesimulation planned,and data prepared' a selected, sequence Transformationof raw data into usabieform doesnot alwaysrequire a simulacan assessments for tion model.Much of the usualinformationneeded waterresources in microcomputer be preparedby hand or by using analytical proceduresavailable format.Typicalnonmodelinganalysesincludethefollowing: 1. Identify wateruser groups and all basin sites for hydropower producirrigation, flood damagereduction from tion, reservoirbasedrecreation,
518
CHAPTER 21
reservoircapacity,industrial and municipal water supply,diversions,and flow augmentation. t Compile annualand seasonal and streamflows flood recordsat eachgauged site for the period of record at eachsite. 3. Determine the fraction of the allocation to each consumptiveuse that is assumed return to the streamat eachuser site in the basin. to of 4. Perform frequencyanalyses annual streamflowand flood valuesat each gaugesite in the basin. losses. and seepage Determinefor eachreservoirsite the eVaporation meanprobabilitiesto be usedin the firm and secondaryyield analy6. 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 waterusesite in the basin. 9. Determine existing and proposed hydropower plant capacitiesand load factors. to 10. Identify any minimal allowablestreamflows be maintainedfor flow augreachin the basin. mentation at eachflow augmentation 11. Specify any maximal or minimal constraintson any of the annual or seaor sonal water allocations,storagecapacities, target yields. or minimal dead storage,active stor12. Specify any constraintson maximal age, flood control storage,or total storagecapacitiesat any or all of the 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 benefitloss functions for a variety of allocationsto domestic, commercial,industrial, and diversionuses. 17. Determine shortrun 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. Developflooddamagereduction beneflt functions at eachpotential flood site. This analysisrequiresrecords of historical and/or simulated damage floods, channel storagecapacities,and flood control reservoir operating policies. 20. Developreservoirbased recreationbenefitfunctions at eachrecreationsite in the basin.
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 and development management groundwatersystems. poroui mediumwerederivedin Chapter18 andmodelingof regional flow of waterin a in was discirssed Chapter 20. This section deals primarily with techniques systems used in solving the hydrodynamicequationsof motion and continuity, followed by of (2) of brief discussions (i) typical input requirements, techniques calibrating and (3) the sensitivity of groundwatermodels to parameter verifying the models, and modelis also of changeslAnexample the calibrationand applicationof a groundwater 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 to the codeby inputting the system beingsiudiedis described the system (aquifer and flow field parameters,initial and boundary known internai operandi conditions, and water use and flow stressesapplied in time to all or parts of the system).Codes have emerged in four general categories:groundwaterflow particle tracking codes,andaquifer testdata analysis 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 rechargedischarge any two points From Darcy's law, the flow passing at stresses cells or line segments. differential. The codes are used to model both can be calculated from the head confined and unconfinedaquifers.Eachcan be structuredto model regional flow, or and flow in proximity of a singG well or wellfield. Steadystate transientconditions and be barriers,full or partially penetratingstreams Boundaiiescan canbe evaluated. By heador constantgradientperimeters. application lakes,leaky zones,or constant after solving can velocitiesof groundwater be determined the of Darcy's 1aw, seepage for the head differentials. and velocitiesareknown, the advection,dispersion, seepage When groundwater can be modeled.Solutetransportmodelsbuild on in changes concentrationof iolutes groundwaterflow modelsby the addition of advection,dispersion,and/or chemical due or If reactionequations. the chemical,dispersion, dilution concentrationchanges particle tracking codesmodel transport by m groundwaterflow are not important, methodthan solutetransportmodelsto track the path advectionand providean easier moveunderthe influenceof headdifferentials.Aquifer that 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
Acronym code for PLASM MODFLOW AQUIFEM1 GWFLOW GWSIMII GWFL3D MODRET SUTRA RANDOMWALK MT3D AT123D MOC HST3D FLOWPATH PATH3D MODPATH WHPA TECTYPE PUMPTEST THCVFIT TGUESS
Description
Groundwater flow models Twodimensionalfi nite difference Threedimensionalflnite difference T!woand threedimensionalfinite element Package of1 analytical solutions Storageand movementmodel' Threedimensionalfinite difference from retention ponds Seepage Solutetransooftmodels Dissolvedsubstance transport model Twodimensionaltransientmodel Threedimensionalsolutetransport Analytical solution package Twodimensionalsolutetransport 3D heat and solutetransport model Particletrackingmodels Twodimensionalsteadystate Threedimensional transient solutions Threedimensional transient solutions Analyticai solution package Aquifertest analyses Pump and slug test by curve matching Pumping and slug test Pumping and slug test c Specifi capacitydetermination
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, groundwater and modelingreliesonfinitedffirence 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 aquifer and water systemparameters input of are and other parametersare calculated.Approximate solutionsof simultaneous linear and nonlinearequationsare found by making initial estimates the solution values, of testingthe estimates the equationsof motion and continuity, adjustingthe values, in andfinally accepting minor violationsin the basicprinciplesor making further adjustments ofthe parameters an orderly and convergingfashion. in
521
r '
or of Theorderly solutionof finite differenceanalogs the steadystate unsteadypartial difierential equation of motion for flow of groundwaterin a confined stare methods'An eady relaxaquifer or an unconfinedaquifer is obtainedby rela_xation For by of the equationis discussed Jacob.T twodimensionalproblems, ation solution and by (ADI) methoddeveloped Peaceman the iterativealternatingdirectionimplicit is often adoPted. Rachfords prickett and Lonnquisteused the ADI techniqueto calculate fluctuations in at watertable elevations all nodesin an aquifermodelby proceedingthrough time in from a known initial state.Their modelis computationallyefficient small increments and readily appliedand is particularly attractivefor usewith problemsinvolvingtime are variablesind nu"tous nodes.The primary aquiferparameters the permeability constantover the aquifer plan, result in a which, if assumed and storagecoefficient, and homogenJous isotropigcondition.For thosefamiliar with relaxationmethods'the (SOR)methodshavehad application overrelaxation Gaussseidelandthe successive equations. in solvingdifference as modelsmay be classified spatialand temporal' Spatial system Input to groundwater thicknessdata over the initial oi projectedwater table maps,saturated input inciudes contourmaps,transmissivitymaps,regionalvariationsin storage ."gion, land surface coefficients,locationsand typei of wells and canals,locationsand types of aquifer actualor net pumpage both lateral and vertical, a nodecoordinatesystem, boundaries percolationand rechargerates for precipitation and other appliedwaters,logs rates, of drilled wells, geologicstratigraphy,and soil types and cropping patterns. Timedependenidatarequirementsfor aquifer models involve principally the using a rangeof time incrementsfor suchvariablesas formulation of ti" schedules, groundhydrographs, canaland streamflow precipitationhyetographs, pumpingrates, variablessuchasthe timing of added rates,and development waterevapotranspiration can apply only Becauseeachtemporal schedule wells or other systemcomponents. requirementsare also particular iubset of node positions,the timedependent to a spatial. aquifer modelsrequire reliable estiIn addition to the listed input parameters, of percentages waters in the land phasethat actually percolateto the matesof the aquifer being modeled.Theseestimatescan be basedon knowledgeof the physical involved in unsaturatedflow through porous medium but are most often processes of that are modifiedduringthe calibrationphase the obtainedasjudgmentparameters stated,the lateral movementand the changesin piezometeror simulation. Simply (withdrawalratesor stresses watertable leneli are easilymodeledif the nodebynode by are aie known. The latter parameters governed the complexmoverates) recharge ment of water in the unsaturatedsoil zone and by the random precipitation and lies systems use consumptive patternsof the region.The art of modelinggroundwater to evaluatetheseparameters' in the ability
DataRequirements
Calibration
involvedin parameter modelcalibrationremovessomeof the guesswork Groundwater knowledgeof basedon available of Severalcombinations parameters, determination. physical syqtem,are testedin the model during a period for which records are the
I L,
522
21 CHAPTER INTRODUCTIoN To HYDRoLoGIc MoDELING availabie. Simulatedresultsare then comparedwith historical events.After structuring the model,calibrationis achieved operatingthe model during the studyperiod by by imposinghistoricalprecipitationamounts, evaporation and evapcanaldiversions, otranspirationrates, streamflowsand stream levels, pumping rates during known periods, and other stresses the aquifer. Calibration is achievedafter the flow, on storage, and other parameters lirnits to produce havebeenadjustedwithin reasonable the best imitation of recordedevents.
CaseExample
modelingin A typical finitedifferencestudyinvolvingsurface 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 the by severalcanalsand laterals,over 1200wells withdraw waterfrom the aquiferbetween the PlatteRiver and the Gothenburg DawsonCounty canals.The aquiferrecharge and and withdrawal amountsas percentages precipitation, snowfall, pumped water, of deliveredcanal water, evaporation,and evapotranspiration were estimatedusing a mix of judgment andphysicalprocessevaluations. The resultingsetthat producedthe wells showninFig.2l.2 bestcomparison with recorded events the six observation at is summarized Table 21.3. Samplesof the comparison in betweenrecordedand simulatedwaterlevelsin the DawsonCounty study during aZyearcalibrationperiod are shownin Figs. 21.3 and2l.4. The Prickettand Lonnquistmodelwasappliedin the DawsonCounty study.The by storage coefflcientfor this unconfinedaquiferwas established calibrationtrials as 0.25 and the adopted permeability was 61 mlday. Other trials were made using variouscombinations S andK, with S rangingfrom 0.10to 0.30andwith Kranging of between4I and I02 mlday. As with most unconfined aquifer models,water table elevations were most sensitive fluctuationsin the storasecoefflcient.Fisure 21.5 is to
tt'iorX
\F 'q;
SIMULATION 21.2 GROUNDWATER ALLOCATIONS FOR CRITERIA WATER RECHARGE TABLE21.3 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 atall 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
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 waterlevels observation and Figure2L.3. Simulated recorded 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 Position in Fig. 21.2. After vefification, the Dawson County model was applied to investigatethe Included among the schemes schemes. shortterminfluenceof 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
o a
d
0.999
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 varying and with changes constant Figure 21.5. Waterlevel Well storage coefficients. 4 Fig. 21.2;K : 6Im/day;I : 97;J : 42. contributesto the waterbalancqof the aquiferbut is not the the existingcanal system from precipitationandfrom the dominantfactor\n the shortrun. The naturalrecharge Platte River accountfor the longterm water table stability in the region.
525
useof ulation ofmanyyearsof prototypebehavior.This is the featurethat makesthe of eventhe largest, thesetooii so attractiveand hoids suchpotential for the analysis so well It mostcomplexsystems. is alsothe prinipal featurethat makesthis approach suitedto water resourcessystemplanning and analysis' models, Apart from the useof cot1entionalhandmethodsand someelementary however,as judgment.This is changing, planninghastraditionally beena practiceof that permit the analysisof large numbersof alternaquantititive tools are developed is elementof the process, not ruled out but is tives and plans.Judgment,anessential to those in the planning 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, alternativesfor lesscost than can provide detailedinformation aboutmore planning principally at of uny oth", approachavailable.Development thesetools has occurred universitiesand federal agencies.
526
CHAPTER21
alternativesowing to time and cost limitaof for preliminary screening development currenttime and sizelimitations evolves, of tions.Unlessa new generation computers by do not allow screening simulating all iilternativesunlesssubstantialsacrificesin followedby detailedsimularealismare made.For the present,preliminary screening to tion appears be the most effectivemeansfor arriving at optimal waterdevelopment and management Plans.
Purpose
Simulatesthe precipitation runoff river process any comPlex in
July 1966
many subbasilsof a river basin using gaugedata and weightings (includedin HEC1).
Iuly 1966
August 1966
August 1966
Computessubbasin interception/infiltration, unit baseflow, and hydrographs runoff hydrograph(included in HEC1). Estimatesbestfitvaluesfor unit graph and lossrate Parameters from given precipitation and runoff (included in subbasin HEC1). at Combinesrunoff from subbasins and routes confluences througha river hydrographs network using hydrologicrouting methods(includedin HEC1).
527
Name
Streamflow Routing OptiffIization
Dateof latestversion
July 1966
Purpose
Estimatesbestfit valuesfor hydrologic streamflow routing parameterswith given upstream' downstream,and local inflow (includedin HECl). hydrographs gravitY and Computes seepage, pressue flow, pumping and for overtopping discharges Pond areasbehind leveesor other flow obstructions.Main river elevation and ponding area elevationareacapacitydata 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.
November1978
JuJy1976
hYdraulics River/reservoir
HEC2, Water SurfaceProfiles August 1979
supercritical. AnalYzes for allowableencroachment a given rise in water surface. Gradually Varied UnsteadyFlow Profiles Iarnary L976 Simulatesonedimensional, 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.
June 1976
534
CHAPTER21
PROBLEMS
in are 21.1. Simulation and synthesis treatedseparately Chapters'22and 23. List the most of distinguishingcharacteristics eachmethod and give an exampleof each. 21.2. Listatleastthreereasonsmanyofthedevelopedmodelsoftherainfallrunoffprocess 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 PaperH748, National TechnicalInformation Engineers Urban Studies,"Miscellaneous Service,Aug.1974. in 2. GeorgeFleming, ComputerSimulation Techniques Hydrology. New York: American Elsevier,1975. "HECI Flood HydrographPackage,"Users and ProU.S. Army Corps of Engineers, grammers Manuals,HEC Program723X6L20I0, 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 WaterModelsScientific 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 8 . D. W. Peaceman N. H. Rachford, Indust.Appl. Math.3' (1955). J. DifferentialEquations," Soc. "selectedDigital ComputerTechniques Groundfor 9 . T. A. Prickett and C. G. Lonnquist, Ilinois StateWaterSurveyBull. No. 55,197I. Evaluation," waterResource 1 0 . D. R. Maidment, (ed.), Handbookof Hydrology.New York: McGrawHill, 1993. "Digital Simulationof ConjunctiveUseof 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: PrenticeHall,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 2530' on draulicsDivision Symposium Engineering 1993.
Chapter22
Time":?J5: Hydrologic
r Prologue
The purposeof this chaPteris to: . Show how time seriesanalysisis used for generatingsynthetic hydrologic records. of aspects hydrologic . Give definitionsof termsusedto describe stochastic the series. including masscurve analy. Introduce fundamentalsof streamflowsynthesis and sesequences, serialdependent sis, random generationof sequences, quences havingprescribedfrequencydistributions' Timeseriesanalysisof hydrologicvariableshas becomea practical methodolof ogy for generatingsyntheticsequences precipitation or steamflowvaluesthat can uiusedlor u tung" of applicationsfrom filling in missingdata in a gaugedrecord to, extendingmonthiy streamflowrecordsl, and from analyzinglongterm reliability of floodsor snowmeltrunoff quantito or yields ofwatersheds2 reservoirst'o forecasting Thesesynthetic hydrologytechniques iies from syntheticprecipitation sequences.5 widein augmentttre simulationtools described Chapter21. Both have,experienced generationof a Synthesisinvolves the spiead use by hydrologistsand engineers.6 or ,"qu"n"" of valuesfor somehydrologicvariable(daily, monthly, seasonal, annual). for are The techniques most often applied to produce streamflowsequences use in generaterainfall sereservoir designor operation studiesbut can also be used to be quences that can subsequently input to simulation models' of to If historioalflows iould be considered be representative all possiblefuture during its lifetime, there would be little variationsthat someproject will experience for The hiitorical record is seldomadequate predicting needfor synthetichyOroiogy. patternis unlikely to recur, future eventswith certainty,however.The exacthistorical of sequences dry years (or wet years) may not have been as severeas they may beiome, and the singlehistorical record gives the planner limited knowledgeof the magnitudeof risks involved.
536
CHAPTER22
particularly if Syhthesisenableshydrologiststo deal with data inadequaci.es, Short historical records of hydrologic record lengths are not sufficiently extensive. hydrologicsynto variables*"fu u, streamfloware extended longer sequenceslrsing scienceknown as operational hydrology'1 of thesisand other techniques the broad either preservethe statisticalcharacterof the historThesenew, syntheticsequences ical records or follow a prescribedprobability distribution, or both. When coupled provide hydrologistswith imthe with computer simulation techniques, 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.
MassCurveAnalysis
by was devised Ripplla to investechniques One of the earliestand simplestsynthesis that the future His analysisassumes tigate reservoirptoragecapacityrequirements. in inflowsto a reseivoirwill be a duplicateof the historicalrecord repeated its entirety to many times end fo end as is necessary span the useful life of the reservoir. as Sufficient storageis then selectedto hold surpluswaters for releaseduring critical 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.1517 produce identicalto pastflows.Randomgenerationand Markov modelingtechniques of, that seqUences are difJerentfrom, although still representative historical flows.
EXAMPLE 22.1 Streamflowspast a proposedreservoir site during a 5year period of record were, acreft.use Rippl's 8000,and 12,000 10,000,6000, 14,000, in respectively, eachyear method to determinethe size of reservoir neededto provide a yield of ,nur. Crrru" 9000 acreft in eachof the next 10 years. is of solution. A loyear sequence syntheticflows,usingRippl's assumptions, equal to the historical record repeated shown in Table Z2.l.Inflows are set twice. rl FOR 22.1 STREAMFLOWS EXAMPLE22'1 TABLE
Flows (thousandsof acreft) 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
90,000 80,000
^
I
Eigarc 22.1 Mass curve for ExamPle 22.1: inflow;  cumulativedtaft.
cumulative
538
RandomGeneratiOn
one method of generatingsequences future flows is a simplerandom rearrangeof ment of past records.If the streamis ungauged and recordsare not available, a probability distributioncan be selected and a sequence future flows that follow the of distribution and haveprescribedstatisticalmomentsis generated. Wheneverhistorical flows are available,a reasonable sequence future flows of can be synthesized first consultinga table of randomnumbers,selecting number, by a matchingthis with the rankinfile 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 next flow, and so on. Randomnumbershavingno the corresponding flows are neglected and the next randomnumberis selected. TableB.3 in Appendix B is a table of uniformly distributedrandom numbers(eachsuccessive numberhas an equalprobability of taking on any of the possible values). illustrate To the use of Table B.3 in the random generation process, the first three yearsof a syntheticflow sequence could be generated selecting 53rd, 74th, and23rdfrom by the the list of past flows. Alternatively, the flows in 1953, 1974, and 1923 could also be selected the new randomsequence. as Most computershave random number generationcapabilitiesin their system libraries. Rather than storing large tables of numberssuchas Table B.3, successive random integersare usually generated the computer. by EXA]I/IPLE22.2 Annualflowsin Crooked Creekwere 19,000, 14,000, 21,000, 8000,11,000, 23,000, 1 0 , 0 0 0 , a n d 9 0 0 0 a c r e  f t , r e s p e c t i v e l y , f o,r2 ,e a 4s 5 , 6 , 7 , a n d 8 . G e n e r a t e a l y 3,r , 5year sequence annual flows, O,, by matching five random numberswith year of 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 8year sequence, the next random number, 2, places so the 14,000cfs flow in Year 2 in the first position of the synthetic 5year sequence.rI
TABLE 22.2 DEVELOPMENT OF s.YEAR SYNTHETIC SEQUENCE Randomdigit
I o
Q(acreft)
2 3
I
5 6
2 5 8 I
I
544
CHAPTER22
'
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 Given the calculatedstatistics, simulationof daily flows could be structured 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 5. Add the differencebetweenthe total monthly flow generated the given given monthly flow, and repeatthe simulation. 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 for rnaintained.When generatingstreamflowsequences an entire system, preserbetweenstationsbecomesa significantfactor. vation of crosscorrelation are runoff sequences employedto determinethe capacSyntheticallygenerated flow magniities of reservoirsto satisfy specifieddemands.Individually generated of sequences flows. Hydrolotudesare uncertain, as are the syntheticallygenerated severalequally likd sequences probabilitiesof flowsby generating gistscanestimate of recurrences certain values.Herein lies one of the most of flowsand then evaluating useful applicationsof Markov generatingtechniques.
i Summary
as Hydrologicmodelingis often presented comprisingonly the deterministicmodels in described Chapters2I,23,24, and 25. The fully of the rainfallrunoff process equippedhydrologistincorporatesthe synthetichydrology models describedin this A chapterin the analysisand designof water resotrces systems. growing numberof basis of synthetichydrology and timeprojects are constructedor operatedon the sedesanalysiseachyear.
PROBLEMS
545
PROBLEMS
record in Example22'2 and detet22.1. plot cumulativeinflows versustime for the Syear yield of neededto provide a mine by mass curve analysisthe size of the r_eservoir What is,the maximumyield possible? 12,000acreft in eachoithe next 24years. of sequence synthetic 'r, t use the annual rainfall trom Table 26.2to generatea lOyear assumption' annual rain depthsfor Richmond using Rippl's masscurve curve methods'Use 22.3. RepeatProblem 22.2 wingrandom generationrather than mass digits of numberJfrom taUl'n.: and matchthesewith the last two i*oOlglt.unaom in the yearnumbers Problem26.32.
22.4. RepeatProblem22.2wingrandomgenerationtogenefateal0yearsyntheticsestandard
tr<
and qu"n"" of annual rain defths that has.a normal CDF with a mean 1 equal to that of ihe annual rain data from Problem 26'32' deviation follow a logPearson Repeat Problem 22.4 assumingthat the annual rain depths statisiicsusethe mean,standarddeviation,and skewofthe Type III distribution.For logarithmsof annualrain at Richmond' (c) Pearson using (a) normal distribution, (Uitognbrmal distribution' and distribution.
CanyouconvertthesimulationprobleminExampe22.4toalog_normaldistribution given in the example? simulation?What difficulties aie encounteredwith the data duration of rainfall data, flt a distributionto the time betweenstorms,and month. Preparea computerpfogram data coveringthe selected 20 yearsof recorded of storms' the to randomly generate times betweenstormsand the durations
22.9. Flowsduring6yearsofrecordwereusedinsynthesizingthemasscurveshownonthe
following page. a.UseRippl'sassumptionandthegraphtodeterminethemissingmagnitudeofthe flow for the 12th Year. yield of 2000 acreb Determinethe reservoircapacityrequired to allow an annual acteftlYr' for ftlyr. RePeat 500 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. is given below' use random of 22.LI. A sequence uniformly distributedrandom numbers annual rain depths that will follow a of generationto generatei 5_y"u, sequence in'' a standarddeviation F"u..on fype iU distribution and will havea mean of 25'8 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 the is 10,000 acreft', standarddeviationis 1000 acreft' : distribution. The mean is 0.50. Start with Q1 coefficient skewis 0.6, and the lagoneserialcorrelation of randomly selected 10,000and find nu" oiJuu.kovgenerated flows if a sequence and50 years' return periodsgives2, 100, 10,2,
546
CHAPTER22
25,000
20,000
9l B
15,000
10.000
101112
REFERENCES
1 . U.S. Army Corps of Engineers,*HEC4 Monthly Streamflow Sirnulation," Hydrologic EngineeringCenter, 197l. 2. R. M. Hirsch, "synthetic Hydrology and Water Supply Reliability," Water Resources Research, 15, no. 6, L979. v. "The Valueof Stochastic StreamflowModelsin Overyear J. R. M. Vogel,andJ. R. Stedinger, v. Research, 25, no. 9, 1988. 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. Research, 12, no. 5, 1976. 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," Hydraulics J. 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
r Prologue
The purposeof this chaPteris to: , Introduce and describecomputercodesavailablefor performing contrnuous simulation of surfacerunoff and streamflow' model. Presentin detail how one of the programsthe Stanfordwatershed "otnponentsof the hydrologiccycle' simulatesthe miscellan"ou, .Showthemajorsimilaritiesanddifferencesoftheleadingmodels. from are . Providea detailedcasestudyof how the modelparameters developed 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. with tools for estiin Simulation modelsdescribed this chapterprovide hydrologists in accounting tittt" for precipitation'direct runoff ' by mating streamflow continuously interflow, deeppercolation' baseflow' and streaminfiltration, evaportranspiration, simulationmodelstrack flow. During rainfree intervalsbetweenstormi, continuous deeppercolation,and baseflow, to evaporation, of the storage water.andits depletion until the next rain or snow eventoccurs' ThemodelsarebasedonthephysicalprocessesdescribedinChaptersl_14.As
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 model, version IV (swMIV)' is presented 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 casestudiesof independent and comparestwo samemanner. Section 23.2 ptesents were {etermined and how the Stanford model studies,showinghow the parameters modelswere calibratedand applied to the problemsbeing assessed'
Figure 23.1 Schematic diagram of APItype hYdrologic model. (After Sittner et al'r)
550
CHAPTER23
StanfordWatershedModel lV (SWMV)
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 subsequently) the original program. The domain FORTRAN version (discussed of hydrograph a as SWMIV hasbeenwidely accepted a tool to synthesize continuous parameterapproachis A hourly or daily streamflowsat a watershed,outlet. lumped used and data requirementsare much less than for alternative distributedmodels. data,and a variety of watershed Hour$ and daily precipitationdata,daily evaporation parameters input. are of The relationsand linkage of the variouscomponents SWMIV are shownin 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 Userassigned zonesshownin Fig. 23.2. condiThree zonesof moistureregulatesoil moistureprofiles and groundwater for is in encountered smaller watersheds accounted tions. The rapid runoff response in the upper zone,while both upper and lower zonbscontrol suchfactorsas overland for The lower zone is responsible longerflow, infiltration, and groundwaterstorage. term inflltration and groundwaterstoragethat is later releasedas base flow to the flow, and is stream.The total streamflow a combinationof overlandflow, groundwater interflow. of Model Structure The SWMIV is madeup of a sequence computationroutines in for eachprocess the hydrologiccycle (interception,infiltration, routing, and so on). of Separatediscussions each componentare provided in the following paragraphs. Actual calculationsproceedfrom processto processas ilfustrated by the arrows in or Fig.23.2. All the moisturethat was originally storedin the watershed wasinput as 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 is simulatedand balancedbeforethe programproceeds the next time interval. The simulationterminateswhen no additional data are input.
551
(ET) is assumed occur at to Evapotranspiration In SWMIV evapotranspiration "upper" storage zone' The upper and the the potential rate from interceptionstorage soils.The lower soil zone surface and the zonesimulates depressions highly permeable storagezone' , 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 frominterception arrdlpper zone storage ttotug". Evapotranspiration groundwater to is assumed be the lake evaporationrate Jet equal to the potential rate, Eo, which calcuiatedasthe productof a pan coefficienttimes the input valuesof the evaporation to wateris assumed occur at a rate equal of pan data.The evaporation any intercepted has when the interceptionstorage and ceases rate io the potential evapotranspiration been depleted. alsooccursat the potentialrate. The from streamand lake surfaces Evaporation surfaceareaof streamsand lakes(ETL) defined by total volgmeis governed the total to and as the ratio of the total stream lake areain the watershed the total watershed groundwaterstoragealso occurs at the potential rate from area.Evapotranspiration and is caiculatedin a similar fashion using a surfacearea equal to a factor K24EL the area.Thus the parameterK24EL represents fraction multiplied by the watershed from the groundwater evapotranspiration of tfre total watershedarea over which at set will occur.Most investigators this parameter avalue equalto the fraction storage Its phreatophytes. value is normally small but can area coveredby of the watershed be large, for eiample, in an agriultural area that has many acresof subirrigated alfalfa. the is If interceptionstorage depleted, modelwill attemptto satisfythe potential at for ET by drawingfrom the upper zone storage the potential rate. Once the upper lower zonebut not at the potential rate; zone sto;ageis dJpleted,ET ociurs from the the ET ratb from the lower zone is alwayslessthan Eo. When interception and the entersas EoinFig23.3, do uppef zone storage not satisfythe potential,any excess
/'\
/r \.
\
t., /
/
l*              \
model IV flowchart. (After Crawford and Linsley.2) Figure 23.2 Stanford watershed
553
F".""I
(: "iPrf) @
Channel inflow
@
Channel inflow Channel inflow
zi
\
k^*ff^." \     ;   J
Simulated streamflow
tt ,'
t
"7
554
CHAPTER23 CONTINUOUS SIMULATION MODELS
25
50
75
100
E:Eo_% Q3.2) 2r The variable r is the evapotranspiration opportunity, deflnedas the maximum water amountavailable ET at a particular location during a prescribedtime period. This for 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 vary over a watershed to 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
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 Fig. 23.4.This wateris assumed infiltrate into the lower and groundwater by zones.The areaenclosed the trapezoidis given by the equations storage the in the first row of Table23.3.If themoisturesupply7 exceeds maximum 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 as a stateofbeing transported interflow at any instantis calledthe interflow detentionor detainedinterflow. to 3. Any remainingmoisturesupplied,AD in Fig. 23.4, contributes increasing Equations for this the surface detention during the time increment. area are included in Table 23.3 for various valuesof 7. triangularshaped
TABLE 23,3 EQUATIONSFOR THE SHADEDAREAS IN FIG. 23.3
Component
Net infiltration Increasein interflow detention Increasein surfacedetention of Percentage increaseddetention to assigned interflow
Searce; After Crawford and Linsley.2
x<b
b<i<cb
x>cb
b 2
x
zb
t
b 2 i2 2cb
ut,  ts
2'  c b
2rb
/ r \
xt
/ , t \
r oo( :) r
r oo( "* r
_ r tr )
too2 _ j ilb
c  l
556
ThEquantity of net infiltration is contrblledlargelyby the maximuminfiltration because the capacityb, while the pbrameterc significaltly affectshydrographshapes increment. The 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 infilffation and depends the soil permeabilityand the volume of moisturethat can 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 CC rangefrom 1.0 to 5.0. of If the soil moistureratio is lessthan 1.0, the variable b is found from
u "
cB
2ALzs/LzsN)
(23.4)
. 0 :
,nrr"tvttt*,
*n
rr?5)
by Theseequations were developed Crawfordand Linsley from num0rousttials using a When the soil moistureratio reaches value SWMIV in many different watersheds. The parameterc 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 23.5 and23.6.Midrangevalues CB : 1.0and CC : 1.0wereusedin developing 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 as'depicted Fig. 23.2.The portion that streamflowof enterthe upper zone storage
0.0
0.2
1.0 1.2
r.4
ratt" zone moisture (!ZL'J Lower soil b Figure 23.5 Variationin patarneter for variousvaluesofthe soil moisture ratio. (After Crawfordand Linsley.2)
0.2
0.4 0.6
0.8
1.0
1.2
1.4
1.6
/ Lzs)
Figure 23.6 Variation in parameterc for various values of the soil moisture ratio. (After Crawford and Linsley.2)
o o I
E
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 Figure 23.7 Typical SWMIV response moisture supply 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" zone moisture (ffi) soil Upper Figure 23.8 Delayedinfiltration as a function ofupper zone soilmoisture ratio. (After Crawford and Linsley.z)
558
CHAPTER23
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
e: roolr'(ffi)(ri)"'' ]
where UZI is determinedfrom
(23.7)
r J Z r r : r r l # k  r . o ]r . o +
The curve is definedto the right of the inflection point by
(23.8)
f/
t.0_)r,',l
(23.e)
rrzr2:rrlffiz.+] .o or
(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 transpiredis proportionedto the surface or 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
(23.rr)
where CB is an index that controls the rate of infiltration. It rangesfrom 0.3 to 1.2 depending the soil permeability and on the volume of moisturethat can be stored on 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 topographyand cover and is alwaysconsidered be 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 SWMIV 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
559
The parametersLZSN and CB must also be estimatedat the beginningof a simulation study.The combinationthat will most satisfactorilyreproduceboth longto and shorttermhistorical responses hydrologic inputs can be determinedby the 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 and betweensynthesized recordedstreamflows satisfacuntil agreement CB LZSN is adjusted; is do tory. If the annualwaterbudgets not balance, 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 from F;q.23.I1.Thepercentage net infiltrationthat reaches tion rateis determined groundwaterstorage depends the soil moisture ratio LZS ILZSN as shownin Fig. on P, 23.9.If this ratio is lessthan 1.0,the percentage is found from
P,:1oo[#(#t,"',
is andif LZS/LZSN is greaterthan 1.0,the percentage
(23.r2)
+: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 technique delcribedin Section11.4andillustrated Fig. 11.8.In modifledform was 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 rate discharge to the 1 Secti,on 1.4), whereeachconstantis the ratio of the groundwater
560
CHAPTER23
Inflection 9oo
oF \ i ltt
0.5
1.0
1.5
2.5
*u" moisture (;R) Soil of Figure 23.9 Percentage infiltratedwaterthat reaches and (After Crawford Linsley'2) storage. groundwater (K n groundwaterdischargerute 24 hr earlier. Thus the recessionconstantKK24 hasvalues The variableGWS in Eq.23.15 usingt : I day. Eq. 11.1)is determined Its on the longterminflows to groundwaterstorage. value on any given that depend day'svalue,adjusted previous as (e.i., the lth day)is calculated 97 percenLofthe day : 0'97 (GWS'1 * inflow to groundor storage, GWS; foi any inflow to groundwater during daYl)' water storage 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 when KV is zero, E,q.23.15reduces Eq. 11.1and ratesas time passes. recession follows the linear semilogrelation If the usual dry season recessilon groundwater the are slofaggs being rate KKz4is too largefor wet periods(whengroundwater recession sothat the KV ihe from the streams) parameter is handadjusted by recharged seepage +KVaGWS) will reduce the effective rate to some desiredvalue during term i0 periods.Table 23.5illustratesthis computationby showingeffectiverecesrecharge to 1'0' sion rates for variouscombinationsof KK24 and GWS when KV is set equal lost to deepor that is either The fraction of activeor deepgroundwaterstorage or is diverted as flow acrossthe drainage groundwaterstorage Gi;.b.D inactive to basin boundary is input ur pu.u*tt"r K24L. This fraction is the total inflow FOR RATES RECESSION 23.5 EFFECTIVE TABLE
OF VARIOUSCOMBINATIONS KK24 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
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 Averagevalues of lengths, slopes,and roughnesses overland flow in the calculatethe Manning and continuity equationsare usedin SWMIV 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 SWMIV 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 , : # y,,(?)"'[, * o,o(r2)']"'
where e : the overland flow dischargerate (cfs per ft of width) D : the averagedetention storagedurrng the time interval
(23.18)
but that occurs after rain ceases, the The equationalso appliesduring the recession Typical overland flow roughnesscoefficientsafe to ratio blD" is assumed be 1.0. 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 endofintervalsurfacedetentionD, 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 Manning's for overlandflow
TABLE;3.6
Watershed cover Smooth asphalt Asphalt or concretePaving Packedclay Light turf Denseturf DenseshrubberYand forest litter
Source: After Crawfordand Linsley.2
(23.20)
(D: + D)/2' EquaThe discharge is found from Eq' 23.18usinga valueo! D . @ of overland flow using easily tions 23.17 2i.20 allow the completedetermination overlandflow' length, slope,and roughness found basinwidevaluesof the average same is Interflow The watertemporarilydetainedasinterflow storage treatedin the inflow to interflow detention was fashion as overland flow detention storage.The similar constant usinga daily recession is in defined Fig. 23.4.Theoutflow simulated constantIRC is the recession The interflow discharge. to that definedfor groundwater 24 at ratio of the interflow discharge any time to the interflow discharge hr average is outflow from detentionstorage earlier.For each15min 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. applied to each time value continuouslychangeswhen the continuity equation is accordingto continuity, on the interval. The endofinteival 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 routing technique translatethe channelinflow to the watershed drologicwatershed as in outlei. Clark's IUH timear"a to"ihod described Section 12.6is adoptedalmost l2.Inplaceofthenetrainhyetograph,theStanfordmodelviews presentedinchapter "inflow" hyetograph. This inflow is the sum of all channelinflow componentsas an whereit is next routed then translatedin time through the channelto the basinoutlet, by to system accountfor the attenuationcaused storage through an equivalentstorage (linear in the sensethat Roulingthrough the linear reservoir in thechannelsystem. is accomto storageis assumed be directly proportional to the outflow' Eq. 12.35) plishedfrom (23.23) I or:7KS1(1 01) )
MODELS 563 SIMUIATION 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
K  Ltlz :v*6*
(23.24)
from watershed parameters Examplesof the determinationof K and othernecessary in Section23.2. data areincluded Applications of the SWMIV Applicationsof the model typically beginwith data for a three to sixyearcalibration period for which rainfall and runoff data arc of adjustments severalparameters Thesedata are usedto allow successive available. of recordedhydrographs the streamflowagree.If sufficient until the simulatedand data areavailable,a secondperiod of record may be reservedfor use as a control to checkthe accuracyof the paiametersderivedfrom a calibration with the first half of the data. in modelwasoriginally developed 1959and hasunderThe Stanfordwatershed translatedthe Crawfordand Linssincethat time. James3 goneseveralmodifications ley model from ALGOL to FORTRAN. Several modifications of the FORTRAN Included amongtheseare the versionhave evolvedfrom a variety of investigations. model (KWM),4'5 the Kentucky selfcalibrating_version Kentucky watershed (OpSETi,4the Ohio State University version, the Texas version,6the Hydrocomp nonproprietary SimulationProgram (HSP) written in PL/1, and EPAproduced, HSP called HSPF, and the National WeatherServicerunoff FORTRAN u"rrion of of forecastingmodel. Brief descriptions severalof theseare includedbelow.
564
CHAPTER 23
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 temperature,and the potential snowmeltper day in the zone air snowpack. Precipitation falling during a snowmelt day also contributes to the snowmeltequation. Service Among otheruses, modelhasbeenappliedby the Soil Conservation the Figure 23.10 showsthe results of in preparing environmentalimpact statements. and widely applying the 1974 version to annual runoff'from four widely separated In diversifiedARS experimentalwatersheds. addition to the runoff, the model computes 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 designed relatively small rural watersheds,
$i
E U
Figure 23.10 Chart showing the accuracy of USDAHL74 model for estimatingthe cumulativecomputedrunoff as comparedwith the cumulao OH: A W 11, tive measured runoff at four watersheds. W97, Coshocton, Hastings,NE; I W3, Ft. Lauderadale,FL; x WG, Riesel, TX (After Holtan andLopez.l)
565
566
MODELS 23 SIMUIATION CHAPTER CONTINUOUS duintutt data can be input at any numberof stationsin the basin.The part that or will run off is divided into the baseflow, subsurface interflow, and surfacerunoff. The division is based on indices and on the intensity of the direct runoff. Each and componentis simply delayedaccordingto different processes, all are then comrunoff is then outflow hydrograph.This subarea bined to producethe final subbasin routed through stream channelsand reservoirsto be combined with other subarea hydrographs, of which becomepart of the output. all by Routingsthrough channelsand reservoirsare accomplished the sametechallowances and of nique.This requiresan assumption shortstreamreaches, occasional in for backwatereffectsare necessary the channelrouting process.Streamflowsare on synthesized the basisof rainfall and snowmeltrunoff. Snowmeltcanbe determined 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 watershedrunoff initial reservoir elevationsand outflows, flow among the three processes, viding drainageareas,bounds on usablestorageand allowabledischargefrom reservoirs, total computationperiods,routing intervals,and other specialinstructionsto control plots, prints, and other inputoutputalternatives. This model was one of the earliestcontinuousstreamflowsimulation models and has its primary strengthin its verified using a lumped parameterrepresentation indicatedby testsconductedin severallarge drainagebasinsincluding the accuracy ColumbiaRiver basin and the Mekong River basin.
29j
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 kinematicwave ratesand other waterparameters relatedto variabledischarge are quality constituents so that the coupling of quantity and quality of the runoff is accomplished. radiation,wind, and Inputs foisimulating waterquality includethe temperature, of the factors under study which form the basis for humidity and observedvalues calibration.At lest two yearsof data are preferablefor calibration; however,calibrawith lessthan one year of data. Other requiredinput includes tion hasbeenachieved and potential in hourly precipitation 5, 15, or 30min or greater time increments or water quality simulation is desired,the temperature, if evapoffanspiration; snow radiation, wind, and humidity factorsare needed' 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, and total dissolvedsolids,algaecounts,phosphorus' dissolvedoxygen remperarures, BOD, colipH, carbonaceous nitr;te, nitrite, ammonia,iotal nitrogen, phosphate, the usual daily, monthly, and annualwater budgets, metals,and forms, conservative snow depths,and endofperiodmoistureequivalents' Typical simulation periods in HSP applicationsrange from 20 to 50 years. of Hourbyhour data are not viewed as an exact sequence future flows. Rather, the of probability of occurrences rangesin the factorsof data arcusedfor analysisof the the model is functioning with a purposesimilar interest.When usedin this unir"r, in described Chapter22. to that of someof the operationalhydrologytechniques
ModelingSystem(PRMS) USGSPrecipitation'Runoff
their urban stormeventmodel, DR3M (seeChapter 25), the U.S' After developing severalother computercodesto model continuoushyGeologicalSurv"y developed for pRUS performssimulationof daily streamflows a variety drologicpro""rr"r. fne from the USGSin is available It of precipitation,climate,and land usecombinations.
I
L.
568
CHAPTER23
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 approach utilized in PRMS.The smallestsubdivision A lumpedparameter to the study watershedis a hydrologic responseunit that is assumed. behaveas a 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' kinematicwave Hydrographrouting in channelsand reservoirsis accomplished.by methods,respectively. and storageindication
ExpertSystems
of analysisare leadingto developmenl expert system(ES) Recenttrends in systems use in planning and designof water that rely on artificial intelligencefor techniques
, .oo^,,r^^o r^io;id7[mrfian.m hcciw?i' iiFiYmqfinn nhfqined frnm extencive
C*.')
' /READTNPUTDATA/
INITIALIZE PARAMETERS
ACE
COMPUTEPEAK RATE TRANSMISSIONLOSSES. OUTIET SBDT,M}{T YIEIO ENO ROUTE SEDIMENT TO BASIN
570
CHAPTER23
interviJws of one or more experts in some field. The computer can then make "decisions"in muchthe sameway asthe,experts, applyingtheir judgment and experito othersthrough the expert systemmodel. ence and making theseavailable simulation,incorStreamflowmodels,especiallythosethat perform continuous judgment. Developers and usersof parameters require considerable that porateinput have accumulated decades of experience in assigning ahe watershed models and judgment can be extractedby an Their experience coefflcientsand parameters. hundredsof questionsto build an expert systemmode'. interview processinvolving about uncertainties but direct answers alsoaddresses The modelnot only incorporates promise.ls modelingtechniqueshow considerable with this each.Early applicdLtions have the potential to be In addition to streamflowsimulation, expert systems of of complexriver basin systems dams,reseruseful in the designand management Operationsfor such flood control structures. voirs,powerplants,diversioncanals,and systemsinvolve independentand collective decisionsby dozens of professionals' contact with numerousother conTheseexpertsare normally in radio or telephone from theseteams,the could be developed If trollers and decisionmakers. ES data A prime incentiveof implementingexpert exists. potential for improvedmanagement profesinvolvescapturinginsightsof experienced systems iystemsin waterresources positions. sionalsbefore they retire or move into other
571
minimum of 10 yearsof continuousrunoff recordsin order to establish the existingrainfallrunoff 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 runoff data were 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 Drainageareasof 2.53 mi2for the CC watershe and9 .22 mi2for the BBC watershed for maps.Input parameters the were found from U.S. GeologicalSurveyquadrangle 23.7. Numerical Stanfordand Kentuckyversionsare comparedand definedin Table names,aretabulatedfor both watersheds usingthe Stanfordversionparameter values, in in Table 23.8. Each parameteris described detail in the following sections.
TimeArea HistogramData
The timearea histograms for the BBC and CC watershedsare developedin for Traveltimes and times of concentration the Figs.23.12 and23.Ii, respectively. larger found from the Kirpich equation(for watersheds 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) : the slopebetweenthe two points S the In developing timearea histogramfor Big BordeauxCreek, 30 points and times (min) were plottedas shownin Fig. 23.12.The dashed correspondingi.auet by (lines of equal travel time to the basin outlet) were constructed linear isochrones paris of isochroneswere interpolation between the plotted points. Areas between from planimeteiing.To contrast,the timearea histogramfor CaveCreek 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 isochroneat equal distances the basin outlet.
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
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 timearea histograq Timearea histogramordinate data are to be printed (cfs) DischargevalueLelow which no synthesized factor for precipitation data for a distant statton 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 Groundwaterevapotranspiration phreatophytes 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,bankfull (cfs) routing of Exponent flow for nonlinear Daily baseflow nonlinearrecessionadjustmentfactor Daily baseflow recessionconstant (in.) storage Indexof groundwater storageat beginningof year (in') and depression 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)
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 7.5min series were determinedfromareas,elevations, parameter usGS topographicmaps (see Table 23.8 for BBC values).Other BBC gri : 0.0d5 (determinedfrom ETL : half the product of the stream values utl from Fig '23'14 length and channelwidth at the outlet), SS : 0.088 ftlft (determined 20ft contoursat eachgridline intersegvaluesbetween u, ih "un of'140 measured of from Fig.23.14 as the average l40lengths 183.2ft (determined tion), and L: lines from gridline intersection points to the measuredperpendicularto contour
FOR BIG BORDEAUXCREEK TABLE 23.8 SUMMARYOF INPUT PARAMETERS CREEK WATERSHEDS AND CAVE
Parameter name
TCONC TINC Z C
Description Time of concentration Routing interval Number of elementsin the timearea histogram Timearea histogramordinates
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 of overlandflow Averagelength cfs 39 Bankfull flow in channelat gaugingstation 0.485 constant recession Daily interflow 0.977 \ constant Daily baseflow recession 0.989 parameterfor low flows Streamflowrouting 0.989 Streamflowrouting parameterfor flood flows cfs 1.0 rate to be printed 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 infiltration by seasons Factorfor varying 0.15 in./hr Maximum interceptionrate for dry watershed 0.80 capacitYindex Surfacestorage 1.10 capacity moisturesiorage Soil surface ln. Il./d profile moisture storageindex Soil 0.28 Soil evaporationparameter 0.0 Index of inflow to deepinactive groundwater 0.0 Fraction of watershedareain phreatophytes 1.0 infiltration rates for seasonal Factor allowing 0.'75 Factorcontrolling infiltration rates 3.0 Index controlling water enteringinterflow 1.0 Parameterfor allowing nonlinear recesslon in. 0.1 storagevolume parameter Groundwater in. 0.0 Equivalentdepth of upper zone storage in. 7.0 of lower zone storage Equivalentdepth in. 0.2 moisture conditions Index of antecedent acreft 0.0 water in swamPstorage 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
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
15min isochrone
0.2
3li
Time.area
a5min {
Ll:Ja
68o l \t
$.(t
60min isochrone N
.1^
t\75minisochrone
90min isochrone
n . , u miles r IJJJJJJJJJ 1 l
for Figure 23.12 Timeareahistogramdevelopment Big BordeauxCreek' The valuesrepresenttravel times (min.) to the outlet.
resulting in an estimateof the bankfull BBC flood of 39 cfs. Another techniquefor rating curveif th CHCAP involvesthe useof the streamgaugingstation determining maps. from topographic gaugeheightfor bankfull flow is known or can be estimated the cross of from a hydraulic analysis thJCave Creekvalueof 40 cfs wasdetermined to spctigl and profile of the main channel.As a rule of thumb, CHCAP is selected )
.../
3l.ovo _
Figure 23.13 Derivation of the timearea 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 r v a l : f m = t l Z O f t inte
\I
Figure 23.14 Grid overlay for determinationof the mean overland  slopeand distancefor Big BordeauxCreek.
576
CHAPTER23
twice the value of producewelldefinedoverbank and floodplain flow if Q exceeds wheneverQ is lessthan half of Similarly, low, shallow flows are assumed bffCAp. CHCAP.
I o.s E o.+
o
K 0.3
E o
d
V'L
p9
o.ro
q
S o.os o
1 4 1 6 1 8 2 0 2 2 2 4 2 6 2 8 3 0 1 3
(Ju1YAugust' 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
577
!2
F
o b0
5
logQ,=1ogQ6+tlogK logK= LlogQlLt
d o o
f f i z o
1969) (JulyAugust, 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
'
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 inflow hydrographs the point of interest for the basin (in this caseto the stream garglng;tatiin ror compa;isonwith measuredflows). The first parameter 1_519t_lt (CHCAP), thechannelaapacity usedinroutingonlyif channelflowsarelessthanhalf during flood flows for and the second(KSF) accounts channeland floodplainstorage capacity).Respectivedata for Big _Bordeaux (flows greaterthan twice the channel shown in Figs. 23.18 and Creek were determinedusing the two runoff hydrographs the Because inflection point on respectively. Z3.lg,representinglow and flood flows, the represents time at which a reversalof portion of eachhydrograph the recession KSC and KSF flow directionihrough the channelbanks occurs,then the parameters water in storagein the channel. The representdaily recessionconstantsfor the is equationused in determiningboth parameters derived from a hydrologicrouting inflow and outflow during in the average equatingl/K timei the difference technique of the outflow in the channelreach.As the routing periodwiththe time rate of change for shownin the figures,the low flow and flood flow routing parameters Big_Bordeaux a similar result was observedfor Cave Creek are identical. Table 23.8 showsthat Creek.
4.0
o
d
3.0
1.0
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 flow at the gaugingsiation below which no printing of simulateddischarges hourly of desired.For the initial simulation of runoff from Big Bordeaux Creek, a MINH in to 1.0 cfs was selected minimize the printed output. This can easilybe decreased sirnulationtrials" after the uncertaintiesin someof the other parameters successive reduced. havebeen A precipitationweightingadjustmentparameter(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 di*tun." frornthJstudy basin.The recordinggauges both watersheds were set to 1.0. proxirtrity, and both K1 parameters close pan ivlonthlyevaporation coefficients(EVCR) are usedby the simulationprodata datato lake evaporation in convJrtinginput panevaporation gram asmultlpters evaporation rates)'The nearest potentiale.'upottuntpiration for determining iand also 20 pan for Big BordeauxCieet is approximately mi southat the Box Butte Reservoir,
580
CHAPTER23
'? !) 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 flood flow streamflow of routing parameter (KSF)for theBig Bordeaux Creekfloodflow event of March17,1969. providing records of daily and monthly panevaporation depthsfor the months between and including May and September. Estimatesof corresponding monthly lake evaporationamountsat Box Butte Reservoirwere obtained from maps and charts developed Shaffer.22 by Valueslisted in Table23.8 are assumed apply to eachyear to of the simulation even though the simulation program allows changes from year to year. Manning's roughness parameters flow over soil and impervioussurfaces for are both required as input to the program. For the Big BordeauxCreek area,the initial estimates qverlandflow (NN) and impervioussurfaceflow (NNU) were 0.37 and for 0.013,representing coefficients dense for 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 determined trial and adjustmentuntil the are by comparisonbetweenthe simulatedand recordedstreamflows satisfactory. is Guidelinesforestablishing initial valuesexistfor only a few of the parameters, whereas most
MODEL SIMULATION STUDIES 581 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 no Because guidelines estimatingthis 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 the initial estimates trial and adjustment. initial estimatesare as follows. and EPXMthe 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 ifr" O.tsin./hr valuewas selected the moderateforestcoveralongBig Creek. Clarke used0.10 in the CaveCreek study' Bordeaux CXan 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 EDFan index of soilsurface during warmer months due 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 LZSNa soilprofile moisturestorage abovethe water table and below the ground the volume of water stored surface.This parameteris a major runoffvolume parameter,inversely related to the basin yields, interflow, and groundwaterflow. The LZSN from on index,depending porosity andthe specificyield of the soil, ranges plus half the meanannualrainfall can be usedas an 2.0 to 20.0,and 4 in. rainfall' By the use of a initial estimatein areasexperiencingseasonal 15.55 in. of annual precipitation, the Chadron, lg3IIg52 averageof : 11.78in. for Big Nebraskl, precipitation station givesan initial LZSN BordeauxCreek. A similar analysiswas usedfor CaveCreek' that controlsthe rateof evapotranspiparameter K3a soil evaporation from the lower soil zone.The parameterrangesfrom 0.2 to losses ration cover.As an initial on 0.9 depending the type and extentof the vegetative which agreed for the light forestcoYerin BBC, 0.28 was selected, estimate by with the estimatis suggested Crawford and Linsley.2 Also, K3 is and _ approximatelyequal to ihe fraction of the basin covered by forest
582
CHAPTER23
grassfor deprooted vegetation.Recommendations2 barren ground, 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 basinboundarY' also drainage groundwaterthat Percolatesto K}4LPatameter canbe estimatr are thesel0sses small to waterlevels,or it is often assumed be zerobecause and runoff' of comparedto the magnitudes rainfall K24FjL'the friLctionof the total watershedover which evapotranspirarate' to tion from groundwaterstorageis assumed occul at the potential zerounlessa significantquantity of vegetation d is This pararneter assume water, drawswater directly from the watertable.Plantsthat seekphreatic or alfalfa, are called phreatophytes' suchas cedar or cottonwood EFafactorrangingfrom0.lto4.0thatrelatesinfiltrationratesto evaporationrates.Thisparametersimplyallowsamolerapidinfiltration 1'0 was A rate recoveryduring raim"t seasons. normal starting value of "nig Boideaux Creek, whereasthe CaveCreek value was set for selected much lower. on the soil CBan index that controls the rate of infiltration, depending permeabilityandthevolumeofmoisturethatcanbestoredinthesoil.A Big midvalue oi O.ZS(0.31,.2) was selectedfor the sandy soils around BordeauxCreek;asmallervalueof0.65wasoptimalfortheCaveCreek study. moisture cYan index controlling the time distributionand quantitiesof a moderately enteringinterflow' This index rangesfrom 0'55 \" 4':' and the waterfor high valueof 3.0 was selected Big BordeauxCreekbecause a moderately shedcontainsmany pine needfemats. clarke also selected high CY for CaveCreek. KY24adailybaseflowrecessionadjustmentfaclorusedtoproducea for Big An simulated"oruilin"u, baseflow recession. initial value of 1'0 the base flow recessionfor the Bordeaux creek was selectedbecause hydrographinFig.23.16islinear.Lateradjustmentsmightberequiredin matchingsimulatedand recordedbaseflow recessions' SGWagroundwaterStorageincrement(in.),reflectingthefluctuations for Big in storaqelolume. Usually, an initial estimate(0.10 was selected and adjusted after several simulation trials; Bordeaui creek) is made parameters SGW rangesfrom 0.10 to 3.90 in. Thib and the following four studY' were usedonly in the BBC UZSthecurrentvolume(in.)ofsoilsurfacemoistureasinterception anddepressionstorage'Becausethesimulationbeginsonoctoberlofthe as zero first calibration yeai the parametermay initially be designated unlessprecipitation occutt during the last few daysof September'
583
LZSthe 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. GWSthe current groundwaterslopeindex (in.). This index providesan indication of antecedent moistureconditions.Suggested initial valuesfall between0.15 and 0.25, or the value assigned SGW can be used.A to midrange0.20 in. was selected Big BordeauxCreek. for VOLUMEthe volume (acreft) assignedto swamp storageand dry groundrecharge, accounting the runoffrequired to recharge for swamps in late summer.Since no swampswere visible on the USGS 7.5min topographicmaps,an initial valueof 0.0 acreft was selected Big Bordeaux for 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. Panevaporation 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 BORDEAUX BIG CREEK STREAMFLOWS
Water years Recorded annual streamflow (acreft)
584
23 SIMUI.ATIoN MoDELS cHAPTER CONTINUOUS depthsin inchesbefore noon on the first and after noon on the second.Valuesare precipitation.Because the variablenumof requiredonly for halfdaysexperiencing 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.
s[eo.
streamflowrunoff from the watershedfor the 6. The volume of synthesized entire water year (acreft). 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(acreft). 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. Endofthemonthlevelsof UZS, the current surface 16. Endofthemonth levels of LZS, the current soil moisture storage(in.). 17. Endofthemonth 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 is in simulation presented Tables23.10and23.ll. The former provides hourbyof MINH, Table23.8. Note value of hour listing of all flow ratesin excess the specified
585
a 8 0 {, o s 6 0 E 4 0
t/
l0
12 Time (hr)
I4
16
18
for and hydrographs Figurp 23.20 Comparison synthesized recorded of (After C1arke.t6.1 Cave Creek. MINH that the streamflowwas zero from Octoberthrough Decemberand exceeded only during two daysin January. in Table23.ll containsmostof the informationdescribed the 18 items of outpuf 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 endofmonth 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 11.6in. ran off or was discharged from storage, the remaining2.1 in. recharged for the soil profile. Moisture not accounted during the year was 0.0844 in.
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 this chapter.Hydrologicproblemsand applicationsthat can be analyzed in
=
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
Or
= 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
: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 @ @ +
+ct 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
9o r :
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
+ Oo F
@ 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 v > z. z.
u
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 sEBBA' E
ut E
rj'tr
d o O
U U E
6 d d 6
A22:EAfr22EEEA
t55t
A a
u =  6 U g o o o = o : : &&,&thl,E'ji/oZqMrh
o o
E E ! ! !  
I I I I93'ts
! E ! E
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
E^,3E
ut
E. n q Y fuJ
r,J i dZZZZZ=a=
oH
> lrJ YA < 4
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
591
6000 5000
4000
5 rooo
2000 1000 0
Time (hr)
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 model of response thechannel to 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 this chapter,and other similar continuoussimulation in codes,are availablefrom the federal or state agencies that originatedthe code, or from numerouspublic outlets or vendorswho havebeen authorizedto distributethe software.
592
Problems
percentages the period. Reconstruct block diagram ofFig. 23.2 and plot approximate io show,for averageconditions,how the rain would be distributed(a) initially and (b) after 30 days. rain 23.2. A sloping,concreteparking lot experiences at afate of 3.0 in.i hr for 60 min' The ftlft. If the water detentionon the lot is ft tot is SOO deepand has a slopeof 0.0001 for completeoverlandflow hydrograph 1 ft zero at the start of the storm,calculatethe the SWMIV equations.Use a 5min routing interval and continue of width using computationsuntil all the detainedwater is discharged. 23.3. Calculatethe SWMIV overlandflow timetoequilibrium 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 SWMIV with its more sophisticated offspring HSP and HSPF. 23.5. Discussthe primary differencesamong the four versionsof the Stanfordwatershed in model described this chaPter. 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 simuiationmodels,USDAHL and SWRRB,and discuss applications best suitedto each. in 23.9. Review the differencesbetweenwater budget and simulation models discussed here simulationmodelsdescribed 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 by 23.1r. For the continuoussimulationmodel selected your instructor,describefour different types of problemsthat could be analyzedif you were given the ful1, calibrated model. h in that 23.r. Assume a 30mi2rural watershed your localereceives 3in. rain in a 10day
REFERENCES
"ContinuousHydrographSynthesis with an and 1 . W. T. Sittner, C. E. Schauss, J. C. Monro, Res'5(5), 1007'1022(1969)' APIType 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, the Use of OPSET," ResearchRep. No' 36, Water Resources Through Characteristics Institute, University of Kentucky,Lexington, 1970'
Chapter24
r Prologue
The purposeof this chaPteris to: . Describehow storm event models are structuredand how they are used to for simulatedirect runoff hydrographs singlestorms. models(note . Describethe five most widely usedfederal agencysingleevent popular singleeventurban runoff simulation models are describedin that Chapter25). ' Provide a detailed