7198 - Laserfiche WebLink

Home Browse Search

i The original regression coefficients and exponents for the suspended sediment I oad as determ Ined i n the 1984 NPS study are shown i n Tabi e 3 al ong w ith the rev ised val ues based on the stati stica I correct 1 on. The original regression relationships were based on five years of USGS dally water and sediment discharge data from the Little Snake gaging station at Lily, twelve years of USGS water and sediment discharge data from the Yampa River at Maybel I, fifty-two water and sediment discharge measurements made at Mathers Hole by the NPS (O'Brien, 1984) in 1982 and 1983, and thirty-three water and sediment discharge measurements at Deerlodge by the USGS (El I Lott, et al., 1984) in 1983. The correction factor C is presented in the right hand column. This factor is a combination of the statistical bias correction factor described in the Procedure Section and an additional modification to reproduce the average val ue of the measured data for al I sampl i ng stations. The corrected coefficients and exponents shown In Table 3 were used in all the sediment budget computations. Calculations of the annual sediment load based on the modified regression relationships predicted identical annual loads for the Deer lodge and Mathers sampling sites (only 6 percent difference was determined). This conf arms the conclusions in O'Brien's report that the suspended sediment load from the Deerlodge Park reach is being transported through the Yampa ` Canyon. The sediment load In Deerlodge Park Is limited by the river's transport capacity because this Is an al I uv ial reach of river. The river tries to transport as much sediment as possible, but the sediment transport Is constrained by the f I ow conditions. The sediment load at Deerlodge, therefore, constitutes the sediment supply to the canyon and Mathers Hole. The Mathers Hole site probably has greater transport capacity than Deerlodge (because of its steeper slope, however, the sediment load Is I imited by the avai I abl a supply at Deerlodge. There are no major sources of sediment in (the canyon, therefore, the river cannot transport any more suspended jsediment load at Mathers than passes into the canyon from the Deerlodge Park reach. Mathers regression relationship predicts a sl ightl y greater load than Deer I odge (6 percent greater) so the Mathers relationship Is employed in this analysis to compare with the load predicted from the upstream gaging tation data at Maybell and Lily. Suspended sediment load data are used In the sediment budget analysis because al I the sediment records for the Yampa at May bel I ( 12 years) and the L ittl a Snake at L 11 y (5 years) consisted only of suspended I oad j measurements. The bed material in cobble bed reaches of the lower canyon is ' coarse. Its v iabi I ity as spawning substrate for the Colorado sq uaw f i sh Is dependent upon keeping the cobble interstlcles free of sand and gravel size sediment which is transported principally as bedload. (See O'Brien, 1984, for definitions of bedload, bed material, and suspended load). Although the bedl oad is critical to the maintenance of the sand free cobble substrate, It represents less than one percent of the total load. The unmeasured sand i load was estimated at two percent of the annual total load (O'Brien, 1984). y This three percent (bedl oad plus unmeasured sand load) of the average annual total load should be relatively constant over the period of record and its variation can be ref I ected by the variation In the suspended load. The analysis of the suspended load data, therefore, shoul d be interpreted as reflecting the total historic load in the system. The summati on of the Yampa (Maybel I ) and L ittl a Snake (L i I y) predi cted suspended sediment load was assumed to constitute the upstream sediment 9