7872 - Laserfiche WebLink

Home Browse Search

1 i~ 1 ii that although some of these components had obvious conceptual importance, data were not available to support their inclusion in a mechanistic model. Virtually no data were available on the relationship between water discharge and availability of prey for larvae, so this component was excluded from the final model. The starvation component was also omitted despite its important role in recruitment of other fishes (Houde 1987). Vulnerability to starvation is primarily a function of body size of fish larvae (Miller et al. 1988). Although Colorado squawfish larvae are not extremely large, they have demonstrated considerable starvation resistance. Bestgen (1996) starved laboratory-reared larvae for 17.5 d and reported that 57% recovered and survived when offered food. Thus, we concluded that direct starvation mortality was probably not a major factor in recruitment dynamics of age-0 Colorado squawfish. Lastly, we reduced the number of potential predator species in the model from three to one: red shiner. Green sunfish Lepomis cyanellus and channel catfish Ictalurus punctatus were omitted because they are relatively rare in Colorado squawfish nursery habitats compared to red shiner (Haines and Tyus 1990). Although red shiner is not typically perceived to be an important predator of other fish, recent studies have suggested that they may consume significant numbers offish larvae. Ruppert et al. (1993) reported that 15% of red shiners collected in backwaters near the confluence of the Yampa and Green rivers contained 1 to 9 fish larvae in their gut. Because fish larvae are rapidly digested and remain recognizable in the gut of predators for only 1-2 hours (Rice et al. 1987), results of Ruppert et al. (1993) suggest that predation rates on larvae by red shiner were high. Colorado squawfish larvae do not commonly occur at the localities sampled by Ruppert et al. (1993) and none were identified in gut contents, but during the same collection 7