Spatiotemporal drivers of rodenticide exposure in a mammalian forest carnivore.

Anticoagulant rodenticide (AR) use carries a risk to wild carnivores via secondary consumption of toxic prey, specifically via coagulopathy which can result in morbidity or mortality. Though proximity to urban development often correlates with increased exposure in non-target wildlife, outdoor use of AR products in more rural settings can also contribute to secondary exposures, though point sources are difficult to identify. For fishers (Pekania pennanti), we suspected that alternate-year pulsed resources (e.g., mast seeding), known to drive irruptions in small mammal abundance in densely forested systems, would influence AR exposure, especially where human densities are high. We further suspected AR exposure might increase over time independent of the masting and that fisher age and sex would influence AR exposure given differential patterns of space use. We quantified AR exposure in legally harvested fishers (n = 338) sampled across a gradient of forest coverage across New York State between 2018 and 2020. We detected between 1 and 5 AR compounds in 83 % of sampled fishers, with the rate of detection of at least one compound increasing by approximately 8 % per year. We modeled the total number of compounds detected and the probability of exposure to each of the three most-detected compounds as a function of fisher sex, age, and spatial covariates. Adult fishers (3-4 years-old) and males had the highest exposure. The odds of exposure to bromadiolone increased 4 % for a unit increase in the previous year beech seed production relative to the long-term mean-equal to 2.6× the long-term average, an impact that appears small relative to the high background rate of exposure owing to the apparently ubiquitous use of ARs across the landscape. A better understanding of patterns of AR use, as well as how these patterns interact with ecological processes, will be important to mitigate exposure of non-target wildlife.