8175 - Laserfiche WebLink

Home Browse Search

larger fish, because fish in the lowest quantiles were larger in spring while fish in the upper quantiles had not changed. A line that shifted upward with a slope greater than one was evidence for size-selective growth of larger fish. In this scenario, fish in the upper quantiles had grown more and were distributed over a greater range by spring than fish in the lower quantiles. Combinations of size-selective growth and mortality resulted in varying slopes, the angle of which depended on the intensity and relative dominance of the interacting selective mechanisms (Post and Evans 1989). We conducted Analysis of Covariance (ANCOV A) to detect significant differences in QQ plot regression slopes. Because of inherent variability in field collections, we considered relationships significant at a 0.10 alpha value. Associations among year class and physical characteristics Water temperature and flow data were taken from two USGS gage stations that correspond to study reaches. We used daily water temperatures (measured once daily) recorded at the gage, Green River at Green River, Utah (USGS station number 09351000) approximately 5251an below Flaming Gorge dam (lower Green River). We also used daily water temperatures recorded at the gage, Green River near Jensen, Utah (USGS station number 09261000) located approximately 174 Ian below Flaming Gorge Dam and 110 Ian below the confluence of the Green and Yampa Rivers (middle Green River) (Figure 1). Records from 1950 to 1959 represent the pre-dam era. Records from 1975 to 1993 represent the post-dam era (1982 was eliminated from the Green River post-dam records because of excessive missing data). We interpolated missing daily temperature records of other years (5 to 25% of annual records) by assuming no change if recorded temperatures before and after the missing date were the same and by assuming an even gradient of temperature change between recorded temperatures that were not the same. Temperature corresponding to fall and spring age-O and age-1 fish sampling between fall of 1987 and spring of 1995 was also recorded from the same gages to represent the middle and lower Green River areas, respectively. Growth ceases injuvenile and adult Colorado pikeminnow below approximately 13 oC (Black and Bulkley 1985; Osmundson 1987; Kaeding and Osmundson 1988). Using temperature preferendum data from Black and Bulkley (1985), Kaeding and Osmundson (1988) reported the relationship between growth and temperature as a non-symmetrical parabola with an optimum at approximately 25 oC and a steeper descending than ascending limb. Using these relationships, we calculated total annual degree-day accumulation (OD) by multiplying the degrees Celsius above 13 oC for each day by an index of growth rate linearly interpolated from the relationship reported by Kaeding and Osmundson (1988) and then summing these over the period of one water year (October 1 through September 30) and averaging years for the pre- and post-dam periods. The same DD calculation was also done for summer (July 1 through September 30) periods to represent the first growing season and winter (October 1 through March 31) periods to represent the first winter season corresponding to age-O and age-l sampling from 1987 to 1995. We examined associations among size and relative abundance of age-O fish in fall and that in the corresponding spring for age-l fish, physical flow and temperature characteristics and overwinter growth and mortality. Variables examined included mean total length of fish in fall and spring, relative abundance (CPE) in fall and spring, DD for winter and summer periods, peak 5