Metabarcoding reveals striking dietary variation in a reintroduced mesocarnivore

Abstract Carnivore populations are often limited by prey availability, so knowledge of their foraging ecology is required for successful recovery efforts. Limited availability of energy-efficient, or optimal, prey can be a key determinant of the conservation status of mesocarnivores—thus investigating diet habits during reintroduction efforts can provide insights into the trajectory of a reestablishing population. The reintroduction of fishers (Pekania pennanti) in Washington state, United States, offers an opportunity to assess diet variation across reintroduction areas using high-resolution DNA metabarcoding methods. We assessed Fisher diet in 2 reintroduction sites in the Cascade Mountains of Washington (the South Cascades and the North Cascades) where fishers performed differently across multiple metrics used to measure initial reintroduction success. We had the rare opportunity to test correlations between metabarcoding relative read abundance and prey biomass consumption while fishers were fed a known diet while in captivity. We found a moderately positive, though highly variable, correlation (r = 0.47) between approximate biomass consumption of food items and relative read abundance within individual diet samples; however, we detected a strong, though statistically insignificant, correlation between relative read abundance and prey biomass when looking at the relative importance of prey groups in aggregated diet profiles (r = 0.72). We collected 300 Fisher scats and gastrointestinal tracts using telemetry and scat detection dog teams to characterize diet profiles. We found strong, positive correlations between relative read abundance and frequency of occurrence metabarcoding summaries. Hares and rabbits (Lepus americanus and Sylvilagus floridanus) were the most prevalent prey in both populations (North Cascades = 17%, South Cascades = 68%), but there were significant compositional differences between diets of the reintroduced populations. South Cascades fishers had limited prey diversity (n = 19 taxa), whereas 71 taxa were represented in the North Cascades Fisher diet. The diet composition of fishers in the South Cascades was akin to populations in the central and eastern United States, where range expansions are facilitated by access to optimally sized (>250 g) prey. In contrast, the diverse diet of North Cascades fishers was more similar to populations at the southernmost extent of their range (Oregon and California), where optimally sized prey is scarce and population recovery is stunted. The surprisingly high levels of dietary variation among adjacent reintroduction sites highlight the utility of diet studies in evaluating whether translocated populations are obtaining adequate resources to support population establishment and thus informing subsequent management interventions.