FLOOD09697 - Laserfiche WebLink

Home Browse Search

meters to urbanization characteristics. The equa- tions used were of the form: S[ =a +bCr+cLr . (3.24) in which S[ = the volume of the inlerception storage capacity (Equation 3.2) Cr = the percentage impervious cover on the walershed (Chapter II) Lr = the characteristic impervious length fac- tor (Chapler II). The coefficients a, b, and c are determined for each watershed parameler from each of the three storms. For exampie, for S(; S( = a + bCr + cLr (3.25) ( SI2 = a + bCr + cLr " SI3 = a + bCr + cLr " Sj for each storm evenl (Equation 3.25) is de- termined by calibration and the three equations are solved for the Ihree unknowns a, b, and c. A similar set of equations was solved for Ihe depression storage capacily, Sd' the initial infillration rale, fo, and the equilibrium infiltration capacity rate, fc' Each of these watershed parameters were thus expressed as a function of the two urban parameters, percenlage impervious cover, Cf, and characteristic impervious length factor, Lr. From these relationships, values of the model parameters Sj. Sd, fo, and fc were calcu- laled as needed for parlicular values of the urban para- melers (which characterize Ihe degree of urbaniza- tion). A fifth watershed parameter, the hydrograph rise time, tr, was estimated as a function of the drain- age area. The five equations thus eSlablished were: S[ = fo = Sd = fc = I, 0.272- 0.203Cr + 0.22Lr . 0.793 .OA5ICr -O.040Lr . 0.113 + O.072Cr + O.l68Lr 0.277 - 0.247Cr - 0.168Lr 0.144 A . (3.26) (3.27) (3.28) (3.29) (3.30) The above equalions apply to the total area of Mill Creek, Big Cottonwood, and little Cottonwood Creeks combined, but it also was necessary to derive similar equations for the Ihree individual walersheds. A major problem, hOYfever, was the lack of storm run~ offhydrographs for individual walersheds. Since the available runoff records on the Jordan River inlegrate the runoff from the tluee areas of concern, it was necessary to separate the total hydrograph into com- ponents which could be reasonably assumed 10 apply to the three drainage areas of in terest. The above parameters for the entire area were used 10 calcuiate a combined runoff hydrograph from Big and little Cottonwood Creeks for one slorm. lhis hydro graph was then subtracled from Ihe lolal recorded hydrograph to isolate Ihe hydrograph for the third walershed (Mill Creek) for the chosen storm (say, slorm number one). Then Ihe model was cali- brated to match this hydro graph, and a set of water- shed parameters thus was determined Mill Creek for storm number one. Using the lolal watershed para. melers for Big Cottonwood Creek and those just de- termined for Mill Creek, a combined hydrograph for these two walersheds was computed. By subtracting this hydrograph from the tolal hydrograph, Ihe liltle Cottonwood Creek hydrograph was isolated and used to estimate watershed parameters for the little Cotton- wood Creek. Finally, using the Mill Creek and the lil- tie Cottonwood Creek parameters for the respeclive areas, a combined hydrograph was calculated and sub- tracted from the lotal walershed hydrograph. The re- sulting hydrograph was assumed to be the Big Cotton- wood Creek and was used to determine values for the watershed parameters for that subwatershed. lhis procedure was repealed for the second and third storms, except that the order of subbasin selection was altered to prevent a bias from the order in which the storms were selected. The above procedure was followed to estimate individual runoff hydrographs for Ihree storms cor- responding 10 each of the three watersheds within the sludyarea. For each runoff event, Ihe values of the watershed parameters S(. Sd' fo' and fc were deter- mined from Ihe model calibration procedure. The sels of equations of the form given by Equation 3.25 then were solved for each parameter, and thus the co- efficients a, b, and c were evaluated to produce equa- tions for each of the three watersheds similar 10 those of Equations 3.26 through 3.29. The data covered a fairly broad spectrum of values for Cr and Lr; for ex. ample. Cf varied between 10 and 50 percent. A fourlh storm event (May 23, 1968) was used to lest the equa. tions. Figure 3.11 gives a comparison of the observed and computed tolal discharge rates on the Jordan River at stations 1705 and 1710 for this storm. Flow from Mill Creek at this time (station 1700) was negii. gible and not included in the calculations. Obviously, the results would have been better had the individual walershed outpuls been gaged, but the melhod pro- vides flood peak estimales for various levels of urbanl- zation. These equations could be used 10 estimate stream flow under all conditions of urbanization. Watershed dala and precipitation data (U. S. Army Corps of Engineers, 1969a) for Ihe desired relurn per- iods were used in Ihe model to graph the peak dis- charge resulting from specified degrees of urbanlza- tion by frequency as shown in Figures 3.12, 3.13, and 38