Use of Distance Sampling to Estimate Rainbow Trout Redd Abundances in the Henry's Fork of the Snake River, Idaho

Abstract

Distance sampling was used, as an alternative to a traditional census, to sample large-scale areas of the Henry's Fork of the Snake River, Idaho, for rainbow trout (Oncorhynchus my kiss) redds. Replicate transects perpendicular to flow were traversed by a combination of wading and snorkeling. Perpendicular distances from transects to detected redds were "sampled" and these data were analyzed using the computer program DISTANCE to estimate redd detectability and density. Distance sampling pruvided an unbiased approach to sampling large-scale areas in a river for redds and was robust to changes in detectability. As discharge in the Henry's Fork increased between sampling dates, detectability of redds decreased, and most observations were closer to the transects. The effective area sampled was smaller, but an increase in redd density was observed, indicating increased spawning activity and demonstrating the robustness of distance sampling to changes in detectability. Spawning activity in the Henry's Fork prior to the installation of the Buffalo River fish ladder in Octooer 1996 was concentrated in the area between Island Park Dam and the United States Geological Survey gauging station; only one redd was detected in Last Chance, and no spawning activity was detected in Box Canyon and Harriman State Park. Relatively little spawning activity was detected in the Henry's Fork in 1997, when spawning fish had access to the Buffalo River.