|
<br />Flood Routing Coefficients
<br />
<br />, The Muskingum method, which is a physically based rout-
<br />ing method, is recom~ended to perform the channel storage
<br />routing in the PMF detennination procedure because of the
<br />ease and flexibility of its application. Assuming a channel
<br />slope of 1%, a velocity of 4 ftlsec (fps) (1.2 mls) and 6.1 of 5
<br />min, an analytical method presented by Dooge et al. (1982)
<br />was ~sed to estimate the Muskingum storage routing coeffi-
<br />cient, x, Furthermore, Chow's (1964) rule was used to esti-
<br />mate the Muskingum reach travel time, k. Assumptions .were
<br />made to reflect minimum storage conditions (see Shalaby
<br />(1986) for details).
<br />
<br />Probable,Maximum Precipitation (PMP)
<br />
<br />According to Crippen and Bue (1977) ,and NWS (1977)
<br />maximum flood flows from small basins are generaIJy caused
<br />by intense, often short duration, storms over small areas.
<br />Because the shortest duration for the PMP available in "HMR
<br />No, 51" is 6 hr, a 6-hr PMP storm duration was selected for
<br />the assumed hypothetical 60 sq mi (155,5 km2) drainage basin
<br />for the Washington, D,C" area, For a given PMP duration,
<br />the PMP depth depends upon the selected'storm-area size;
<br />with the storm'area size being optimized by the HMR52 com-
<br />puter program. '
<br />
<br />Antecedent Storms
<br />
<br />The antecedent storms were assumed to result in com-
<br />pletely saturated soil moisture conditions, They are modeled
<br />indirectly by assuming completely saturated antecedent soil
<br />moisture conditions as represented by the constant miriimum
<br />values forthe <I>-index (see Table I),'
<br />
<br />, ,
<br />HMR52 Computer Program to Derive PMS from
<br />PMP Estimate
<br />
<br />The HMR52 computer prOgram computes the basin-av-
<br />erage precipitation for PMS in accordance with the criteria
<br />specified in "HMR No. 52." There are three basic types of
<br />data that are inputtoHMR52. The first 'i,draiIiage basin
<br />geometry data that includes the coordinates of points on the
<br />basin bOundary and a'scille factor' (scale of bOundary coor-
<br />dinatesm miles per coordinate unit), The secondtype of data
<br />is hydrometeotological data, which includes the prderred storm
<br />orientation from "HMR No, 52" and the PMP estimates from
<br />"HMR No, 51." The third Iype of data is the storm specifi,
<br />cation data that define the storm-area size. orientation. and
<br />location of the storm center. ,Calculation of a temporaf storm
<br />dist~ibution requir~s the 'desired time~ interval. &t~ ~d the
<br />ratio of the I-hr to 6-hr precipitation from "HMR No, 52."
<br />HMR52 'will optimize the storm-area size and orientation in
<br />order to produce the maximum basin-average precipitation.
<br />The user must provide the desired centering, although the
<br />centroid of the basin area is provided as a default option. The
<br />user must specify the time distribution for that storm, which
<br />produce a data file that contains this incremental basin. av-. -
<br />erage precipitation vah.~es for every sub~asin requested. The
<br />precipitation data file will subsequently be input to the HEC-
<br />'1 rainfall-runoff model for comptlt?tinn of the resulting fl.ood.
<br />The designer should then analyze the various storm variables
<br />and recompute the floods in order to determine the storm
<br />that produces the maximum runoff peak. The storm and run-
<br />off are defined as the PMS and PMF, respectively,
<br />
<br />HEC-l Procedure to Estimate PMF from PMS
<br />
<br />The most commonly used precipitation-runoff model to
<br />estimate the PMF from the PMS is the HEC-I computer
<br />
<br />INPUT TO HMR52
<br />
<br />1. f'MP: Oeph
<br />"..,.....
<br />IsiohyetalPderri
<br />l.oclItianofSlofmC<<rter
<br />Orien.btlonofS*:rnl
<br />~Sll..'
<br />T~RMIb1~
<br />~ .................
<br />
<br />3. ~8ak1; Ooiinag.Aru
<br />&..wr.inage~
<br />'x, Y Coonflllate$ br~" & Sub tirwagt. MIas ~
<br />
<br />HMR$2 COMPuTEs THE PMS .
<br />
<br />.0lITPUT ~M HMR52
<br />'. <.'AHyGCtlgraphbrEac::tl&lbbadn"
<br />
<br />INPUT TO HEC-1
<br />1.. HMR52~~~phs'
<br />
<br />2. '. Rat"~lrtfilpl~ jAntecedenl Soil MoIsture & l~
<br />.'-.
<br />3.' '.Uibc&m~'
<br />.t. ~ AoodRoU&.uP~
<br />
<br />i'"
<br />
<br />tEt;:-1 COMPlJ!ES ,lHE Pw= .
<br />OI.irPvrFROMtEc-1
<br />
<br />A Flood Hydmgraph Far E-" ~"
<br />
<br />n.. P-* at !he OuW. t!llI PMF
<br />
<br />fiG. 2." Flowchart oIHMR521HEC,1InputlOutpullor PMP/UH AI"
<br />proach .
<br />
<br />model. Although the HEC-I model is capable' of performing
<br />several tasks beyond obtaining a PMF. estimate, such as a
<br />dam-break simulation, and a flood damage" aIialysis;,the de-
<br />scription of HEC-l in this section will be limited to those
<br />elements that are needed to obtain a PMF estimate. ,The PMS
<br />, temporal rainfall distribution derived for each subbasin within
<br />the duinage' basin are input directly to the HEC-l model.
<br />The hydrologic input factors to' consider are as follows: (1)
<br />The rate of infiltratiOli; (2) the UH; and (3) flood routing,
<br />HEC-I provides several alternatives that may be used to de-
<br />fine the.'input forth~se fac.tors: However,. ther~:-are certain
<br />choices of these alternatives that are more appropriate for a
<br />PMF determination. Fig, 2 is;; flowchart of the HMR521HEC-
<br />1 input/output.' '
<br />
<br />RESULTS OF SENSITIVITY ANALYSIS'
<br />
<br />In order that the effect of a change in one factor on the PMF
<br />may be properly assessed, the results of the sensitivity analysis
<br />were transformed into sensitivity charts. Because the true value
<br />of the PMF is never known, a base point PMF was designated
<br />as a basis of comparison. The base point consists of the com-
<br />bination of values for the factors that yield the maximum PMF
<br />for each given landcover distribution under consideration. .
<br />
<br />,Guidelines on Reiationshlps between Factors and PMF
<br />
<br />A flowchart of the sensitivity analysis procedure is pre-
<br />sented in Fig, 3, The following hydrologic factors were held
<br />constant: (I) The size, of the drainage area; (2) the shape
<br />ot the drainage area; (3) the antecedent soil moisture; (4)
<br />the infiltration rate; (5) the shape of the UH; and (6) the
<br />flood routing coefficients. The land-cover distribution was
<br />varied (see Table 1), The PMP depth and the duration of
<br />the PMP were held constant, and the effect of the ante-
<br />
<br />330 I JOURNAL OF IRRIGATION ANO DRAINAGE'ENGINEERING I SEPTEMBER/OCTOBER 1995
<br />
|