9565 - Laserfiche WebLink

Home Browse Search

1368 TABLE 7.-Average annual transition probabilities (proba- bilities of movement into or out of a reach using only nonzero values) for average-size Colorado pikeminnow (499,5 mm TL) captured in the Green River basin, 2000-2003. A positive net value indicates that more fish moved into the reach than out of it, a negative value that more fish moved out of the reach than into it. See Table 6 for descriptions of river reaches. Movement Reach Into Out of Net Yampa River 0.012 0,027 -0,015 White River 0,027 0.035 -0,008 Middle Green River 0,054 0,021 0,033 Desolation-Gray Canyon 0.066 0,050 0,016 Lower Green River 0.D15 0.121 -0,106 distance movements just across the boundary of one reach to another (e.g., from the White River mouth to the middle Green River) and thus may not necessarily reflect a large total distance moved. The average transition probabilities (only nonzero values were used) for a fish of average TL moving from the Yampa River (0.027) and the middle Green River reach (0.021) were low, reflecting a relatively high propensity of fish to remain in those reaches (Table 7). The values from the White River (0.035), the Desolation-Gray Canyon reach (0.050), and the lower Green River reach (0.121) were higher. The mean values of '" into a particular reach from all other reaches were lower for tributaries at the extremities of the Green River basin (Yampa River = 0.012, White River = 0.027) and the lower Green River reach (0.015) than for main-stem Green River sites (Desolation-Gray Canyon = 0.066, middle Green River = 0.054) in the middle of the study area. The negative net values of '" for the Y ampa and lower Green River reaches suggest a net movement of fish from those areas. The positive net values for the middle Green River and Desolation-Gray Canyon river reaches suggest movement into those areas during each year of this study; the near-zero net value for the White River suggests equal numbers of fish entering or leaving that area. The relatively high and positive values of '" for small fish in the lowermost two sections of the Green River reflected upstream movement. The estimates of '" for fish moving from the lower Green River to the Desolation-Gray Canyon reach (0.121), from Desola- tion-Gray Canyon to the middle Green River reach (0.078), and from the White River to Desolation-Gray Canyon (0.080) were the largest detected in this study (Table 6). Estimates for upstream movements (i.e., to the White and Yampa rivers) from downstream areas were lower. The highest downstream values of '" detected were for fish moving from the middle Green BESTGEN ET AL. OJ "0 ..c E OJ -i 2 1.8 1.6 1.4 1.2 I 0,8 0,6 0.4 0,2 o 1991 1999 1993 1995 2001 2003 1997 Year FiGURE 7.-Estimated finite rate of population change (lambda) from ISMP data collected at 10 sites in the Green River basin from 1991 to 2003, Error bars represent 95% confidence intervals, Values of lambda greater than I indicate an expanding population, values less than 1 a declining population, and values of I a stable population, River to Desolation-Gray Canyon (0.045) and from the White River to Desolation-Gray Canyon (0.080). Finite Population Rates of Change We attempted to fit models that estimated A as a function of time for the ISMP reaches, the middle and lower Green River reaches, and the Yampa and White rivers. Parameter estimates were imprecise or models did not converge when A was estimated as a function of p, S, river, and time. Therefore, we dropped the river term and simply estimated a riverwide A over time with p and S held constant, which allowed valid estimates for the first and last intervals as well. We did not use the TL covariate because fish were likely to grow substantially over capture-recapture intervals during the 1991-2003 period, which may have biased size effects over time. Estimates of A for the period 1991-2003 suggest a long-term increase in the abundance of Colorado pikeminnow in the Green River basin up to year 2000 because A was greater than 1 in every annual interval except 1996-1997 (0.80) and 1998-1999 (0.99) (Figure 7). The confidence intervals for the estimates were typically wide. For all three annual intervals between the years of abundance estimation sampling (2000-2003) A was less than 1, and the upper confidence limits did not include 1 in any periods. ISMP CPUE Trends, 1991-2003 The CPUE indices for Colorado pikeminnow at standard ISMP reaches in the middle and lower Green River and the White and Yampa rivers were variable but generally increased until 2000 (Figure 8). White River CPUE indices were particularly variable, as they