9633 - Laserfiche WebLink

Home Browse Search

Colorado Pikeminnow Distribution 465 Figure 5. Graph showing the mean catch per unit ef- fort (CPUE) in sampled reaches between 1990 and 1995, based on the ISMP protocoL Data are for pop- ulations in mid-September. Letters in parentheses be- neath reach designation indicate channeVvalley reach classification. "- III Qi :!: III ~ CT (fJ 8- t:: =B w :t:: c: :J 8. .c ~ fa III :!: 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 0 (F) needs (Hilborn and Mangel 1997). We found that it was necessary to make many assumptions in developing the model, because experimental and field data about key biological and physical processes are unavailable. Future monitoring and research might focus on resolving these key uncertainties. The simulation model integrates hydrologic, geomorphic, and biologic factors and was written using Stella Research Software (Richmond 2001). The model accounts for larval input, main stem transport, and entry and retention in backwaters (Figure 6). Larvae are input at the upstream end of the study area and routed downstream, and the proportion swept and retained in backwaters is estimated as a function of hydrologic and geomorphic characteristics, as well as the estimated rate of fish growth. Backwater population in each reach is calculated. Other biological processes are largely ig- nored. Model Description Hydrology. The model is structured such that a cohort of larvae enters the Green River each day from the Yampa River, and these fish drift downstream through ten reaches (Figure 6). We assumed that larvae drift at the mean flow velocity. The predicted time it takes larvae to drift through each reach is deter- mined by dividing reach length by the estimated mean velocity. Mean daily discharge is used to compute mean ve- locity. We used data for U.S. Geological Survey gaging stations (Figure 1). Discharge of the Yampa River was for II II - W&I o ~ hSl o Mean1990 Mean1991 Mean1992 Mean1993 Mean1994 Mllan1995 (R) (F) Reaches the gaging station at Deerlodge Park, Colorado, or was estimated as the sum of measurements near Maybell, Colorado, and of the Little Snake River near . Lily, Colorado. Discharge in reaches A, B, and C was esti- mated as the sum of the Yampa River and the Green River near Greendale, Utah. Mean daily discharge measured at Jensen, Utah, was used for reaches D to H. Further downstream, Green River discharge is estimated as the sum of the flow at Jensen, the Duchesne River near Randlett, Utah, and the White River near Watson, Utah. The velocity in each reach was estimated from sta- tistical relations between discharge and mean velocity at the gages (Table 2). These relationships are of the form v == a(2b (1) where v is mean velocity, (2 is discharge, and a and bare coefficients (Leopold and Maddock 1953). Although more sophisticated methods could, be used to estimate velocity, we do not believe such methods are appropriate because of other large uncertainties in other model pa- rameters. We assumed that the relationship for Deer- lodge Park applied to the Yampa River, the Jensen relationship applied to the Green River in reaches A-H, and a relationship calculated for the abandoned gage near Ouray applied in reaches I and J. Geomorphology. The model predicts the number of larvae transported into backwaters in each reach in each time step. Since little is known about this process, we parameterized this movement as a function of hydraulics and channel geomorphology. We assumed that the