Laserfiche WebLink
1 <br />entered into the analysis concurrently. Histograms for each of these habitat <br />variables were developed for each of the seven species, as well as for small <br />(< 100 mm) bluehead sucker Catostomus discobolus and flannelmouth sucker C. <br />latipinnis. <br />Data were analyzed for each river within years. Also, sections of the <br />two rivers considered concentration areas for age-0 Colorado squawfish were <br />analyzed separately (Colorado River, RM 0-70 [Valdez et al. 1982]; Green <br />River, RM 22 131 and RM 212-320 (Tyus et al. 1982]). <br />For the habitats from which Colorado squawfish were collected, <br />correlations between C/E for age-0 squawfish and C/E for each of the six <br />i sympatric species were estimated using the Pearson Correlation Coefficient. <br />All analyses were done using SYSTAT (Wilkinson 1988), a Statistical package <br />for use on microcomputers. <br />1 <br />Results <br />Analyses showed little discrimination between habitats (as defined by <br />depth, velocity and substrate) occupied by age-0 Colorado squawfish and those <br />used by most of the other species. Wilks' lambda, which reflects the degree <br /> of separation between the habitat use of Colorado squawfish and that of a <br /> sympatric species, ranged from a relatively low value of 0.219 for speckled <br /> dace in the Green River to 0.996 (essentially complete overlap) for red shiner <br /> in the Colorado River (Tables 1 and 2, Figures 1-4). Wilks' lambda exceeded <br /> <br /> 0.9 in all cases in both rivers for red shiner and in all cases except one for <br />sand shiner. It ranged from 0.497 to 0.995 for fathead minnow and it exceeded <br />0.9 in 60~ of the cases for this species. Channel catfish ranged from about <br />0.4 to 0.995, with three cases greater than 0.9, three cases between 0.8 and <br />t <br /> <br />