7954 - Laserfiche WebLink

Home Browse Search

108 Decision-Making Process N > 250 No YES Larval Production YES Nursery No Genetic Habitat 11-1 # Risk YES I HIGH I LOW Recruitment NJ NO YES DO NOT STOCK YES NO Population YES Increasing _ MODDE ET AL. STOCK FIGURE 3.-Decision logic for determining whether or not to stock razorback sucker in the upper Colorado River basin. The population size (t) of 250 has been selected arbitrarily to achieve an effective population size of 50. Genetic risk is that to the wild population. Given the rational described above, we suggest that razorback sucker should not be stocked into the middle Green River, but rather habitat should be modified to improve the existing dynamics of the population. The demographic status of the Green River population most closely fits the criteria of the high or moderate risk categories suggested by Mace and Lande (1991). If the population continues to decline, spawning adults are rarely collected during the spawning season, and larval production and recruitment ceases, then the decision on whether to stock needs to be revisited. The maintenance of a refugium population would allow a genetic source of the original population that would support aug- mentation efforts. In contrast to the Green River, fish in the Colo- rado and San Juan rivers have populations smaller than 250 breeding individuals and no spawning or recruitment is known. These populations fit the very high risk category defined by Mace and Lande (1991), and stocking of razorback sucker, in accor- dance with habitat development, should proceed with care taken to match genetic and life history characteristics of local populations as closely as pos- sible. If the populations are functionally extinct, AGRs and broodstocks from the nearest neighbor population should be considered. In addition, sites historically occupied by razorback sucker in the upper Colorado River, such as the Gunnison and Uncompangre rivers or the Colorado River above Moab, also should be considered for reintroduction using the same criteria (Williamson and Wydoski 1994). Imprinting The use of artificial imprinting to attract razor- back suckers to target spawning sites will improve the probability of successful reintroduction pro- grams. The use of imprinting will expand manage- ment options following stocking and can be used to test the hypotheses that nonnative fishes and loss of wetland habitats are primary factors in causing the decline of this species. The hypotheses that non- native fish and loss of backwater nursery habitat are primary factors causing decline could be tested by introducing fish into a relatively predator free sys- tem at a site upstream from a good nursery habitat. The successful application of imprinting could also keep stocked fish separate from wild spawners, thereby reducing genetic risks to endemic popula- tions, and attract adult hatchery fish to spawning locations that would enable resulting larvae to col- onize new nursery sites. Based on evidence of homing to selected spawn- ing sites by native fishes in the Green and Yampa rivers (Holden and Stalnaker 1975; Tyus 1987; Tyus and Karp 1990), and the presence of water seeps at known spawning sites that might provide distinctive odors required for olfactory recognition (Tyus 1985), it is thought that natal imprinting may be involved in selection of spawning sites by razorback sucker (Wick et al. 1982; Tyus 1985, 1990; Tyus and Karp 1990). Scholz et al. (1992b) and Tilson et al. (1994) found elevated levels of thyroxine during two periods of imprinting (alevin-swim-up and smolt stages) for kokanee Oncorhynchus nerka (lacastrine sockeye salmon). Similar physiological conditions have been found for razorback sucker; peaks in whole-body thyroxine content of eggs and larvae from Green River adults and hatchery broodstock occurred at hatching and at swim-up (Scholz et al. 1992a, 1993). After swim-up thyroxine content de- clined to very low levels (Scholz et al. 1992a, 1993). If razorback suckers experience chemical imprint-