7758 - Laserfiche WebLink

Home Browse Search

INSTREAM FLOWS TO ASSIST THE RECOVERY OF ENDANGERED FISHES 7 1970. However, accurate annual population esti- mates, based on recovery of fish tagged in the earlier study, are biased by differential tag reten- tion, although the population clearly has re- mained "much less than 1,000" (Kenneth P. Burn- ham, Colorado State University, 1 June 1993 letter to Tim Modde, U.S. Fish and Wildlife Serv- ice, Vernal, Utah). A few young razorback sucker have been collected in the Green River in recent years (e.g., three fish <415 mm in 1993; Tim Modde, personal communication), and the age structure of the few razorback sucker collected annually on the Colorado River has declined in recent years (Chuck McAda, U.S. Fish and Wild- life Service, Grand Junction, Colorado, personal communication). Thus, some recruitment of adult cohorts may be occurring in the Green and Colo- rado rivers, perhaps related to higher flows. Whether stable or declining, the population of razorback sucker in the Green-Yampa system probably has not exceeded more than 1,000 fish in the last 2 decades. Because most of the very few razorback sucker captured in the Upper Colorado River Basin are older fish, I conclude, as did Tyus (1991a), that very little recruitment of adult ra- zorback sucker has occurred since the 1960's. Razorback sucker have been observed spawn- ing or in spawning condition (ripe) during the rising limb of the spring runoff at temperatures 5-100 C below (McAda and Wydoski 1980; Tyus 1987) the experimentally observed optimum range (20-22° C) for reproduction (Inslee 1982; Hamman 1985; Marsh 1985). Razorback sucker were commonly (50 or more per year) collected in the 15-mile reach of the Colorado River in the early 1970's, mostly in a gravel pit connected to the river near Grand Junction, Colorado (McAda and Wydoski 1980; Valdez and Wick 1983). That gravel pit washed out in the 1984 spring flood of record, and only incidental captures were made subsequently (Osmundson and Kaeding 1991). However, in spring 1993, 67 razorback sucker were taken from another gravel pit (Etter Pond). One 20-year-old fish was collected, but the rest were 9 years old, corresponding to spawn during the 1984 flood, when the pond was last connected to the river (Chuck McAda, personal communication). In addition to their propensity to inhabit man- made gravel pits that are at least ephemerally connected to the river, razorback sucker are most often captured in low velocity habitats in the chan- nel (Fig. 2) and wetland ponds connected to the channel (McAda and Wydoski 1980; Tyus et al. 1987). Bulkley and Pimentel (1983) showed that razorback sucker preferred temperatures of 22-25° C in shuttle box experiments. In the pota- mon reaches of the Upper Colorado River Basin, shallow, backwater, and wetland habitats are typi- cally closer to the preferred temperatures than is the river channel, especially in the upstream reaches, where razorback sucker are most com- monly found. Indeed, Wick et al. (1983) showed that backwaters flooded by spring runoff on the Yampa River were significantly warmer than the channel, thereby offering more degree-days for maturation of spawning condition. Naturally functioning backwaters (i.e., not influenced by erratic, regulated flows) also contain food sources, such as zooplankton, invertebrates associated with macrophytes, and microbially rich detritus, needed to mediate growth of razorback sucker (Wick et al. 1982; Wick 1991). The reproductive bottleneck that is preventing recruitment of razorback sucker in the Upper Colorado River Basin is unknown. Clearly, these suckers prefer lacustrine-like environments, ow- ing to their proclivity for low velocity habitats, especially flooded gravel pits and wetlands, dur- ing high flows. River flow regulation, wetland revetments, diversion dams (which limit migra- tory pathways; see Fig. 1), and presence of abun- dant native and nonnative predators (also dis- cussed below with regard to similar influences on squawfish) may prohibit the fish from using back- waters and seasonally flooded wetlands in a man- ner that will allow recruitment to occur annually. Indeed, in Lake Mohave on the Lower Colorado River, where a large population of razorback sucker has persisted for many years but did not recruit in spite of apparent spawning success each year, the recruitment bottleneck was attributed to predation of larvae and early juveniles by nonna- tive minnows and sunfish (Marsh and Langhorst 1988; Marsh and Minckley 1989; Papoulias and Minckley 1992). The recruitment bottleneck for razorback sucker in the Upper Colorado River Basin very likely relates to the current paucity of low velocity, warm, food-rich, and nonpredator- dominated habitats during spring and summer. Instream flow recommendations are based pre- dominantly on ecological knowledge of Colorado River squawfish, which are the most abundant and best known of the endangered big river en- demics. Squawfish occur most abundantly in the potamon reaches of the Yampa, Green, White,