Density of Fishers and the Efficacy of Relative Abundance Indices and Small-Scale Occupancy Estimation to Detect a Population Decline on the Hoopa Valley Indian Reservation, California

We used a mark–resight design to calculate density estimates of fisher (Martes pennanti), a candidate for listing under the United States Endangered Species Act, on the Hoopa Valley Indian Reservation in northwestern California, USA in order to determine population status in 1998 and 2005. Our density estimation results and simultaneous population-monitoring data provided a post hoc opportunity to evaluate the relative efficacy of 3 classical indexing techniques (catch-per-unit-effort, frequency of detection at camera stations, and frequency of detection at track-plate stations) and small-scale occupancy estimation to accurately detect population change. We calculated densities (and 95% CI) of 52 (43–64) and 14 (13–16) fishers/100 km in 1998 and 2005, respectively. We detected a decline in the relative abundance of fishers between 1998 and 2005 using catch-per-unit-effort indices (x ! 10.18, P " 0.007), but not in magnitude similar to our density estimates. We detected an increase (x 1⁄4 4.23, P 1⁄4 0.040) and no difference (x 1⁄4 1.38, P 1⁄4 0.240) in the relative abundance of fishers between surveys using frequency of detection indices at camera stations and at track-plate stations, respectively. Occupancy estimates did not differ between 1998 and 2005. We speculate changes in prey habitat, increases in predation, disease, or some combination of these potential causes, were responsible for the population decline. Our results reinforce the importance of careful thought given to the study goals and potential limitations of any technique. For populations deemed valuable (e.g., at risk or sensitive), we suggest managers consider adopting more defensible, large-scale occupancy estimation or mark–recapture methods to monitor changes in population sizes. ! 2011 The Wildlife Society.