Factors influencing success of cutthroat trout translocations

Abstract

Native subspecies of cutthroat trout (Oncorhynchus clarki) in the western United States have experienced drastic declines in their distributions, often to <5% of their native range, due to habitat degradation, the introduction of nonnative salmonids, and overharvest. Of the 14 subspecies recognized (three are undescribed), one is extinct, four are listed as threatened or endangered under the Endangered Species Act, and conservation plans have been developed for most others. Cutthroat trout readily hybridize with other spring-spawning salmonids and are apparently displaced by fall-spawning species such as brook char (Salvelinus fontinalis) and brown trout (Salmo trutta), so many populations are restricted to remote headwater streams and lakes above fish movement barriers that prevent invasion by nonnative salmonids. Isolating barriers, while protecting cutthroat trout populations from nonnative salmonids, restrict them to areas that may be too small or have insufficient habitat to support a viable population. Establishing new cutthroat trout populations through translocation of genetically pure fish into fishless waters or those treated with toxicants to remove nonnative salmonids is one of the few management strategies for increasing their range. Unfortunately, success rates for establishing self-sustaining fish populations through translocation are generally less than 50%, and few recovery programs have reviewed their translocation attempts to determine the factors influencing translocation success in aquatic systems. Conservation of greenback cutthroat trout (O. c. stomias), one of the listed subspecies, and Rio Grande cutthroat trout (O. c. virginalis), a subspecies that has been petitioned for federal listing, has been ongoing for more than 25 years. More than 40 translocations have been conducted for each subspecies, so they represent two of the few cases where recovery efforts for any fish are sufficient to evaluate factors influencing success. My objective is to identify factors that promote establishment and persistence of cutthroat trout populations isolated in headwater streams by comparing attributes of translocations that successfully established a naturally reproducing cutthroat trout population to those that failed. In Chapter I, I analyzed unpublished data from natural resource agencies on reinvasion of nonnative salmonids, potential limiting habitat factors, and source of translocated fish to identify the attributes of the 14 of 37 translocations of greenback cutthroat trout that were successful, and the probable cause of failure for the other 23. Of the 23 that failed, 11 (48%) were reinvaded by nonnative salmonids, 10 (43%) apparently had unsuitable habitat, and 2 were depressed by other factors. Reinvasion occurred most often because of incomplete removed of nonnative salmonids in complex habitats or failed artificial barriers. Of those not reinvaded, success was highest in receiving waters with at least 2 ha of habitat that previously supported reproducing trout populations. In Chapter II, I report on a detailed analysis of the minimum habitat required for establishment and persistence of populations from field surveys of stream-scale habitat and map measurements of basin-scale habitat for 27 greenback and Rio Grande cutthroat trout translocations in Colorado and New Mexico. The best models developed from these data using polytomous logistic regression predict risk of translocation success or failure from stream-scale habitat attributes. These models indicate that cold summer water temperature, narrow stream width, and lack of deep pools limit populations of cutthroat trout. Cold summer temperatures are known to delay spawning and prolong egg incubation, which reduces the growth of fry and likely limits their overwinter survival. Furthermore, small streams with few deep pools may lack the space necessary to promote overwinter survival of a sufficient number of individuals to sustain a viable population. Models of basin-scale habitat were not as effective as stream-scale habitat for distinguishing between successful and unsuccessful translocations of cutthroat trout but indicate that watershed area is useful as a coarse filter for selecting potential translocation streams. Presumably, large watersheds (> 14.7 km2) encompass low elevation habitat that provides warmer summer temperatures, and would have relatively wide stream channels of sufficient length to provide an adequate number of deep pools. This research is one of the few attempts to determine the specific factors influencing translocation success for fishes, and it demonstrates that measuring attributes of local habitat over a whole watershed scale that matches the life history of the organism can be highly useful for identifying critical habitat factors. The models I developed from stream- and basin-scale habitat attributes will allow managers to choose future restoration sites with a high probability of translocation success and to identify whether populations in fragments of historical habitat are likely to persist. A multi-scale analysis of this type may also be applicable to other subspecies of cutthroat trout in central and southern Rocky Mountain streams (e.g., Colorado River cutthroat trout, O. c. pleuriticus) because similar habitat attributes probably limit their populations.