7212 - Laserfiche WebLink

Home Browse Search

implicated in the decline or extirpation of native fish populations, direct evidence is difficult to obtain (Schoenkerr 1981). Other mechanisms (or more likely a combination of several) may also account for the increase in one fish species and corresponding decrease in another (Li 1979). Overlap in diet or ,-habitat use may infer direct interspecific competition, but a mutually used resource must be limiting before such competition becomes critical (Schoenkerr -'? Li 1979), and field documentation is difficult. 1981; Previous investigators (Holden 1979; Tyus et al. 1982) implicated backwater areas as the preferred habitat of Age-0 Colorado squawfish. Tyus et al. (1982) further indicated that small Colorado squawfish were distributed in the Green River from the mouth of the Yampa River to its confluence with the Colorado River during fall months, with relatively high levels of abundance in two areas of the 'river: between 50 to 145 and 400 to 470 km from its mouth. The obiectives of this study, conducted in areas of relatively high Colorado squawfish abundance, were: (1) to describe the micr?hsb' ark used-by Aye 0 Colorado squawfish and examine their diel patterns of use of these microhabitats; (2) to describe- patterns of habitat overlap in the Green River between Age-0 Colorado s uq awfish and otter fishes associated with them; and (3) to describe dietary overlap between Age-0 Colorado guawfish and these species. METHODS Fish were collected using 0.6-cm-mesh seines from predetermined microhabitats present at the study sites. Care was taken to include only one microhabitat in an individual seine haul. Microhabitat types were defined by Archer et al. (1980). The area sampled was quantified, and fishes were identified, grouped into size classes, enumerated and released alive after sampling was completed. Colorado squawfish were marked by fin-clipping before release. Mean water depth and water velocity were determined from measurements across a transect at the center of the area sampled. Water depth was recorded to the nearest 3 cm, and water velocity was measured with a Marsh-McBirney flow meter at 0.6 of the distance between the top and bottom of the water column and recorded to the nearest 3 cm/sec. Substrate type was determined according to values modified from Hynes (1970). Water. temperature was measured with a hand-held thermometer. Diel sampling was conducted using a systematic sampling design with a random start that sampled all two-hr- periods of the 24 hr day over 2 to 3 days. Several multivariate techniques are available for analysis of large numbers of independent (i.e. environmental ) variables (Green 1979 and references therein). However, discriminant analysis was used in this study since it is considered more robust than other techniques when derived axes are interpreted as environmental factors (Green and Vascotto 1978; Green 1979). The assumptions and mathematical formulae used by discriminant analysis were discussed by Klecka (1980). The values of environmental variables at each sample site were weighted by the catch per effort of each species. For example, if 10 individuals of a species were collected at a sample site, and a single individual was collected at a second site with equal effort, the values of the environmental variables at the first sample site were weighted 10 times more heavily than those of the second site. In this analysis, the dependent variables (i.e. groups) considered were Colorado squawfish, fathead minnow, red shiner, channel catfish and flannelmouth sucker. The environmental variables were sample time, habitat tune, mean water depth, mean water velocity, dominate substrate type and water temperature. Analysis was compl using SPSS: the Statistical Package for the Social Sciences (Nie et al. 1975). 45