7835 - Laserfiche WebLink

Home Browse Search

Simulated length distributions produced mean lengths by age similar to those calculated by adding mean growth increments (Fig. 3a), and indicate the expected range of variation in lengths by age (Fig. 3 b). Growth increments were log-normal within size classes, and simulations were conducted using log-transformed growth rates. When growth was assumed to be constant for fish Z 550 mm TL, the growth rate was steady above 550 mm TL as expected, but the variance in lengths for a given age was greater than when calculated growth increments were used (Fig. 3c). The 20 simulations of length distributions conducted using different random number sequences produced nearly identical length distributions by age. Simulations indicated there are broad ranges of ages for fish of given lengths, especially fish z 550 mm TL (Fig. 4a). Individuals may take 10-22 (mean = 15) years to reach 600 mm in length; 16-30 (mean = 25) years to reach 700 mm; 20-40 (mean = 32) years to reach 800 mm. Similar, but more variable ages were found for a given length when growth for fish z 550 mm TL was assumed constant (Fig. 4b). Survival Tests of Assumptions No significant differences (P > 0.05) were found among years in the length distributions of fish z 550 mm captured during 1991 through 1994 (N = 34, 41, 49, 34, respectively) in the upper reach. Similarly, no significant differences (P > 0.05) were found between length distributions for these years and for distributions for fish sampled in 1990 (N = 15) and 1995 (N = 44) under different sampling programs. The length distribution of fish captured in 1982 (N = 41) was significantly different from the distribution for 1992 (P = 0.028), but was not significantly different (P > 0.05) than distributions for all other years from 1990 through 1995. These results suggest that the length distribution for fish z 550 mm TL was essentially stable during this period. Catch rates, expressed as number of Colorado squawfish z 550 mm TL captured per net-set, were compared for the period 1991 through 1994 (number of sets = 139, 117, 121, 105, respectively). No significant differences were found among years for the mean number of fish per net-set (Kruskal-Wallis one-way ANOVA, df = 3, XZ = 1. 916, P = 0. 590), indicating no measurable change in abundance of fish a 55,0 mm TL. Differences in size distribution were found between Colorado squawfish (z 550 mm TL) captured with trammel nets and those captured using electrofishing. The only data set with comparable periods and regions of fish capture for both gear types was during 1994 in one section (RK 246-275) of the upper reach. The distribution of fish z 550 mm TL caught with electrofishing (N = 13) contained more large fish and was significantly different (P = 0.016) than the distribution for fish caught with trammel nets (N = 21). A bias toward larger fish is consistent with observations made by other investigators electrofishing other species (see Reynolds 1983). Although we could not test whether distributions of fish caught with A-11