While most carnivore populations are declining worldwide, some are successfully adapting to human-caused global changes. For example, coyotes (Canis latrans) have expanded their range across the United States and into many urban areas making it important to understand factors influencing broad-scale patterns of occurrence. In Illinois, coyote populations have substantially increased since the 1970’s, with coyotes becoming the apex predator in the state. We used citizen science data in the form of archery hunter observations from throughout the state of Illinois to evaluate factors affecting both detection and occupancy of coyotes and to estimate and compare coyote occupancy across Wildlife Management Units in Illinois. Our detection variables included date, hours spent hunting, and time period (AM or PM), which were provided by the hunter data. We also included temperature and precipitation. For occupancy, we chose five different habitat variables, which included forest patch density, forest patch index, grassland shape index, agriculture cover, and urbanization.
Results/Conclusions
Our statewide participant-level occupancy estimate was 58% greater than our naïve occupancy estimate, highlighting the importance of using modeling frameworks that account for imperfect detection when modeling cryptic species with low detection rates. Time period (AM or PM) had the largest effect on detection of coyotes, followed by the number of hours hunted (analogous to effort). In contrast, none of the landscape-cover covariates examined had a strong effect on coyote occupancy. While multiple mechanisms may explain this result, we suspect it is because the habitat covariates were measured at the county-level whereas participants effectively survey a much smaller area. Since scale affects the strength and direction of species-habitat relationships, this scale mismatch is likely an important limitation when using many sources of citizen scientist observations to infer species-habitat relationships.