<br />".
<br />Jl ,
<br />:~, " .
<br />j,'." ,
<br />"
<br />
<br />" ".
<br />
<br />. ,
<br />. ,
<br />",
<br />
<br />.'
<br />,
<br />
<br />1:,' ,
<br />
<br />"',.
<br />'~.,' ,
<br />'/. ,{'
<br />,t,
<br />.. ,
<br />
<br />, .
<br />, ,
<br />,.,.
<br />
<br />"
<br />,',
<br />
<br />
<br />'1t~,: .
<br />\~ '
<br />'!\
<br />
<br />.~.:.' ,."
<br />
<br />',".,'"
<br />
<br />, '.
<br />
<br />'.
<br />'"
<br />
<br />" ..
<br />'. .
<br />
<br />I;::
<br />
<br />"
<br />
<br />" '.
<br />. '"
<br />
<br />t:,.
<br />
<br />'MARr28-2000 TUE 10:48 AM
<br />
<br />" ~ I
<br />
<br />".
<br />" ,I
<br />
<br />SUMMARY AND CONCLUSION
<br />
<br />. , '.
<br />
<br />, .
<br />
<br />A data sd of 20 stonn even~.: 'wa. "ocd In ~hio
<br />study t.o select design values of the' HEel loss par am-
<br />ctora Sl'RTL and CNSTL in order to route the proha-
<br />ble maximum flood, PMF, through the Englowood
<br />wnterahed, Ohio. The parameter STRTL reprc~ents
<br />tho initial volumo or water lost due to abstractions
<br />and incomplete saturati<ln of the soil prior to the
<br />~torm, The parameter CNSTL represents II coutinu-
<br />ous'loss rate and depends on the 'watershed only.
<br />Because ' few stonnll' provided uniform precipitation
<br />clWeralle ova' till. whele'wato..h.di 0 o"bhMin (Br:\tl-
<br />ford) was used to estimate the loss parameters. This
<br />,approach waa feasible because the' topography and
<br />. the soil coverage of the subbasins were very similar.
<br />. Optimization of STRTL and CNSTL- from each storm
<br />event 'was conducted 'by:minhuizing the residual.
<br />, botween t.he simulated and' observed surface runoff
<br />hydrographs. The optimization approach used in this
<br />study consisted of performing a'systematic sweep of
<br />the parameter space. :which rea~,~bly 8!L~ured that
<br />the III0b..l, op,timum w..l9C8ted....~" I: ',; " ,
<br />: 'STRTLi optima varied betweell.'O.O'and 8,4 Inche.
<br />> ~ith an II.v!,ragc or'f.o, inch :and 'a '.tandard error of
<br />. 0.18 inch. CNSTL varied betweeD,O,02 and 0.26
<br />, . in~hlbour, and it foljo~iKI"a' ~Ot:l!1al probability distri.
<br />bution witll a mean of about 0,1 iuchlhour and a stan.
<br />dard error of 0.016 'inchlhour. Because CNSTL is an
<br />invariant quantity tb'at',does 'not"depcnd on ~torm
<br />events, its variation is' attributed to' stochastic andlor
<br />, , sampling errors. .The ab~oDce of, corrclation between
<br />,optimum CNSTL val.ues and tot~'rainfall, total loss,
<br />and runoff duration supported tho selection of the
<br />mean CNSTI. as a design value for PMF routing.
<br />, ,The routing of the.PMF through the Englewood
<br />watershed revealed that the bydrograph at the outlet
<br />is not sensitive to STRTL. However, this wns not the
<br />case for the Bradford 'subbasin. 'The insonsitivity of
<br />tho PMF to STaTL was due to the presence of a daln
<br />Ilt the outlet of the watershed that caused the buildup
<br />of water (backwater) in the watershed, 11ler~by mllsk.
<br />illg the storago effects of STRTL: However, no (Inm
<br />was prasont at the o1.ltlet of the Bradford subbo.in,
<br />and storage effects tpefe were, dictated by STI,rr.
<br />only.
<br />The PMl<' was highly. affected by eNST). varia-
<br />tilll1S, 'l'he peak PMl<' increased by about 27 percent
<br />whlln the design CNSTL'wlLS decreased tn 0,05 inch/
<br />hour, and t1ccrensed by about' 18 percsllt when the
<br />design CNSTLwas increased to O,15'incMlour,
<br />I"or ench oftha 20 storms, the surface runorrhydro-
<br />graph was scpnrnted' frnm the totnl runoff using
<br />n graphical procedure. after assuming that the base~
<br />now recedes at an exponential'rate, The recession
<br />. , '.'
<br />".
<br />
<br />JAWRA
<br />
<br />, "
<br />
<br />FElX NO.
<br />
<br />P. 10
<br />
<br />llf'lur"dC'l
<br />
<br />constant, k, wns foulld [0 vary beLween 2,1.5 lIl1 I ,Ill.:;
<br />daY5, and to follow a Ilormnl probability di,'tl'iJl\llh'li
<br />with n mean vnl".. of 33,1 rlnYR nl1,l " ~1,11I\<l',,',l "I'IW
<br />nf 1.6 "\lays. No correlation was found belw",'" \; \"nl-
<br />ues and tho corresponding CNSTL v"lueG, Tlti., "\":N-
<br />vaLioll alol\g wilh the fact that the dislrihlll.I<'1\ "r k
<br />was narrow. indicates that the graphical pract.,lmc is
<br />"reliable means for scp",'nl.ing hasellow,
<br />
<br />ACKNOWL.l':l>GMEN'fS
<br />
<br />The nuthor is growful to hilS Ml'1illt.ars tb('Ril; prlvillor, Ill' PI,'wn
<br />liucl1t..t@.lt,oftho tJntvt:I"IJi'ly o(Cjnc.lllnAU, ror vnll1.n.h'j~ tl1'oIl,'f<l\lfl.
<br />This work was flllpport.t'd in part.. by t.he Mi~\mi Cnnr:tr"':' I\"'~' ni~.
<br />\.riel, Ohio. Hnwev~tl it dnC8110t nc!c("l';~nrily fPnec.l t.hf' ".ic'w,<: nf"t1uo
<br />Miami COnliQ'rYanc:y District, t'lnd no offieii'\l cntloTfll"nlf'n l r..h r"lld ~
<br />inferred.
<br />
<br />l.I'rElt!ITlJnE CITRD
<br />
<br />l~llrd. Y.. 1974. Nonlinct'lt PaTameter EIlt.i1nntlOtl. Arnllrrnll" flre!11
<br />New Yotk, Ne., Veri,.
<br />UampfI, U. S'I 1930. Th,,'f(tnltlI.\H'a ftrPIl.lihnr.l.I:l~1ilh~~~m'\ f'l'il'~'"
<br />floe TraIJS"etioDIJ, AOU1 20;721~1215.
<br />Batu,ll. C., 1988. NonHncmr Dis.crctc Flood J:~vrnl :\l,",'ll'h.
<br />Z. !u:lcument or Stnullrlicnl NonlioP'JlIrily. .Jonmall)f J1."ln.lngy
<br />99m-89.
<br />Bate8, n: C. and L. R. Townl('y, J988. Nonlin{,l'lr Di~;r;r('t" F"lo(lll
<br />Ji:v~nt Models, 1. Dl1ycsil)n EstinHltion of fJlu'ump.1tou, .!lI'Jn,:tl of
<br />Hydrology !l9:6]-7G.
<br />Bouredol. M. C.. ]998. Unl. !lyd,.graph. nori.od fn',o lh" N...
<br />MndG'l. J. or Lb.e American WnLcr Ue5Qurccs ^,,,oc:iiltiQn atlS;-
<br />177,
<br />Fon,tain., T, A.. 1995. Rninral1-Jtunorr Model Acc:ur~C'\' ror aa.
<br />~"~111C Flood. J. o(Hj'dTnulie En~in"'''l'inr:, ASCJo: 1?1 ~1{..J7(,
<br />Ford, D. 't., E. C. Mnrris, and h. [). FElldmnn, 19aO. Co..,'': Iln~nJli.
<br />ftC-erR' Experll'lnec Wit.h Automu\.ie Ci\1ihr"lion llr ,I Prf"dril:\.
<br />tion-nullofC Medel. l,t: Wnter anrl ndolr.d L;,",l l:"o,;(;'Ufcl'
<br />SystelTls. P(!Tgamozl l't('5S, New York, New York.
<br />llaU. A. J.. 1971. BaneOow ItceeAilicl1\:I "nd Dl'l!ll:.nOW lh"~'i"lirl\ph
<br />Separation, fJfC'Jbkm. In: lI'ydf('Jln~l>" SympUl;lUlll, In:<(i['IU'l!\ Clf
<br />Engin.nen. C:mbp.rrn. AUl\lralia. P)), 159.170.
<br />Hayte-r, A. ,I., 1996. Probilbilily nnd St.nth.lie" 1'nT l'~n.~;ld"'n ;\Ild
<br />Hei13Jn.i&tll. PWS PubU&hinJ; COn1.pilrIY, New Yotll, 1\',."" '~ofk.
<br />p.048,
<br />IIoi'tQn, n. F.. !!l:'3. The I{r)l~ of [nnlLrat.i<lR in tIll\: Ir)flrrll".!i... (').:1...
<br />Eos 'l'tnnuactiamr. GU li:-f4G-4GO.
<br />Kuc1.cra. G.. 1982;'1, ImpTOvl1.rJ InroTC'l"Icc Clr (~~ll.rl1T'" ,.1 MaiM
<br />P"ratnaLel'6, IWporL I/HJ82, Ucp.lrLmcnt or Civil b''l'lr:l~cnng,
<br />Monash UniV('Tl'ity, C~il)'t()n, Aus~}ji\.
<br />I{uc1.crn, G., J082b. 01\ the RclatTnMhlp nc,iwet'n ill.. Hdi,t1'ililYrlf
<br />PUllul('tl"r ESt.inlllt,(o$ nn'" Hydrologic Tin-H.' Serif''' 1l:1f \ t 'c,,;~1 in
<br />Ci'\Hbrolion. WlltOt" Re5OUl"ct,'S r<l"'l'Ipnrc:h JR: I.i('")-1I1'1.
<br />KUC7.Crl'1 0., 1990. ElI,iimOltil)}l nrnunnff-RcluLing Morkl P~'.r;'lnt'l"n
<br />tTsil'g Incompnt.iblc 8to1'\'n DilL". Jom'l1al of liydll"ll"I'" J U H.
<br />00.
<br />Lastlon. L. S., A. n. Wnrrt'n. A. Jain, llntl M. H:i"ltnr-r. I' 1','11, 1~(!!'lb'tl
<br />lInG 'l>l'it.ing of a GC1'lcl'31i1.cd H~c1u<:1'!(l Gnu~i('nt {OrJ.I!, fi,l' :-\t1ntin.
<br />rar Progl'nmrning. ~rr:ml1iaciion!'. in Ma~h~Jtutti('~l :l(lll\\ :11'(' t 34.
<br />60,
<br />
<br />212
<br />
<br />JoUI~NAL 01' THE AM~11ICAN WATER rlESOUACI!" M' ... ,,-,,:..
<br />
<br />,--
<br />
|