9692 (2) - Laserfiche WebLink

Home Browse Search

BIOLOGY OF YOUNG COLORADO SQUA WFISH 87 backwater temperatures were higher than the 21- 260C temperature range in which the hatchery fish were raised, and we expected Green River fish to grow faster. However, otoliths from 14 postlarvae collected in 1987 (mean, 38.2 mm T1.; range, 29- 45 mm) indicated that otolith-aged fish averaged about 13 d older than ages calculated from labo- ratory fish of the same size (R. Muth, Colorado State University, personal communication). This could be the result of slower growth by the wild fish, inadequate sample size, or invalid otolith ag- ing. However, growth equations for post1arvae should be further validated or refined by aging wild fish of different sizes before the equations are applied to management of this endangered fish. Habitat Use The timing of spawning, and thus of hatching, is critical to larval survival. Spawning occurred during the period of declining flows and increasing temperatures after spring peak runofl: This tem- poral adaptation (midsummer spawning associ- ated with declining flows) is associated with the reproductive biology of the species (Tyus 1986), whereby the length of exposure of eggs and larvae to cold temperatures of spring and to predation is reduced. It is perhaps critical that rapid growth is attained during the remaining short growth peri- od, and that the larvae are transported into suit- able feeding areas before their yolk supplies are depleted. Backwaters used by the young fish as nursery habitats were created by gradually de- creasing summer flows after spring runoff (snow- melt). Growth and Survival We expected greater growth of age-O fish in the lower Green River because the watler was 40C warmer than in the upper river. However, we found that age-O fish in the upper Green River were the same average annual size as, or were larger than, in the lower area. We assumed that other envi- ronmental factors (e.g., habitat availability and food abundance) may influence growth more in the Green River, at least in some years, than water temperatures alone. One plausible explanation is a difference in available habitat; annual sampling records indicated that there were about 150% more backwaters per kilometer in the upper than in the lower Green River. Backwaters in the lower Green River may yield a higher catch of young Colorado squawfish than in the upper Green River (Figure 2) and greater numbers may have resulted in less growth. Reduced growth at lower temperatures, as hypothesized by Kaeding and Osmundson (1988), was not detected in our study. Reduced abundance and growth of age-O fish in 1983 and 1984 were attributed to abnormally high summer flows from Flaming Gorge Dam. We were initially concerned that higher flows might have reduced our success in catching young fish in 1983 and 1984, and thus biased our results. However, spring sampling in 1984 produced few age-l Col- orado squawfish, indicating that the reduced standing crops of age-O fish recorded the previous autumn was accurate. However, low catches in autumn 1985 and 1987 may reflect sampling bias. Spring sampling produced more young in 1986 than in the preceding autumn for the upper and lower Green River, and more young were cap- tured in the lower river in spring 1988 than in the preceding autumn. Although it was impractical to count all back- waters in the Green River by boat because of the width of the river and limited personnel, we ob- served fewer backwaters during high-flow years (1982, 1983, and 1984) than during low- and av- erage-flow years (1979-1981, 1987-1988). These riverwide observations were similar to those not- ed in a Bureau of Reclamation study conducted during 1986-1988 on four reaches of the Green River. Aerial photographic mapping of shoreline habitats at different water levels revealed that backwaters were reduced in area when flows in the Green River increased above average historic summer flows (M. Pucherelli, U.S. Bureau of Rec- lamation, personal communication). We were unable to assess overwintering mor- tality of young Colorado squawfish adequately. Catches of age-1 fish in the spring (fish/lOO m2) were greater in half of our samples than catches of the same cohort the previous autumn. We as- sume that differences in habitats or in habitat use made the fish more vulnerable to capture with seines in the spring than the preceding autumn, at least during some years. Management and future recovery of the Colorado squawfish may require development of a better method for assessing re- cruitment during the early life stages if recruit- ment failure is determined to be a problem in the Green River. Age-1 fish were generally larger in the spring (mean, 45.2 mm) than in the previous autumn (42.3 mm), presumably a result of growth or a size-selective mortality. Thompson (1989) ob- served that age-O fish actively fed and survived in laboratory aquaria under simulated winter con- ditions (3-40C, 210 d), but they grew very little.