7835 - Laserfiche WebLink

Home Browse Search

year may have resulted in abnormally high condition factors. Slope and intercept coefficients were derived by regressing log-transformed mass as a function of length. Condition of each fish was calculated by using month-specific constants. For instance, the condition of a fish captured in April was relative to that of the average fish captured in April, not relative to fish captured in other months. This allowed us to make condition comparisons between areas that might otherwise be biased by the differences in proportions of fish caught in various months. Bias could result if condition factors were calculated using constants from one weight-length regression derived from all spring or all annual captures. For example, if fish in June had a higher condition factor than those in April or May, average condition within a reach where there was a higher proportion of June-captured fish would likely be higher than other reaches where proportions of June-captured fish were lower. After condition factors for individual fish were calculated, data were partitioned by reach (upper and lower) and by 100-mm length classes. Again, fish not weighed with an electronic balance and those captured in June 1994 were excluded. We used a T-test to test for differences between lower and upper reach fish (within 100-mm length classes), and ANOVA and the Tukey-Kramer multiple comparison test to compare condition among length classes within reaches. Relative Abundance of Forage Fish Electrofishing and trammel-netting catch rates were used as an index of relative abundance of larger-sized (> 100 mm TL) forage fish. Relative abundance among strata of small minnows (Notropis, Pimephales etc.), the likely forage for young Colorado squawfish, was not investigated. For electrofishing surveys, two or more 0.8-km sample reaches were randomly selected within each stratum. To determine number of sample reaches for each stratum, total miles within the stratum was rounded up to the next 10-mile increment and divided by 10. Thus, a 14-mile- long stratum had two sample reaches; a 28-mile-long stratum, three. Reach locations within strata were determined by using a random numbers table to select the river-kilometer (to the nearest 0.1 km) of the upstream end of each reach. Electrofishing gear was the same as described before. Within each 0.8-km-subreach, both shorelines were electrofished in a downstream direction. Total shocking time for each shoreline was recorded in seconds by an automatic counter on the VVP. Each shoreline within each subreach was treated as a separate sample, resulting, for example, in four electrofishing samples for a stratum with two 0.8-km subreaches. Species and total length of each captured fish was recorded; fish were then released at the lower end of the subreach. Sampling was conducted from 20 April through 7 July, 1993. Catch-per-unit-effort (CPUE) was calculated by dividing the total number of fish (by species) by total shocking time for each electrofishing sample; values for all samples within a stratum were averaged to provide mean CPUE for that stratum. B-6