As coyotes (Canis latrans) continue to expand their geographic range, understanding what variables, such as habitat coverage or anthropogenic disturbances, explain their spatial patterns is necessary to inform upon the breadth of their ecological role and effect on community interactions. Here we integrate synthetic and empirical research to obtain a comprehensive view of what variables have been determined to best estimate coyote occupancy. We performed a literature search in the ISI Web of Science database to explore the variables influencing coyote occupancy. We classified variables into three different groups (anthropogenic, ecological, or study design) and pooled grouped variables into several representative classifications. Then, we compared the total number of variables explored in each study, the number of variables belonging to each category found to be significant, and the significant variables of each category classification. In addition, we contribute to the current body of literature by performing the first occupancy study of Great Lakes coyotes in a rural, forested ecosystem. Across 11 weeks, we used remotely-triggered cameras deployed across 59 locations at the University of Michigan Biological Station to estimate coyote occupancy and the effect 8 variables had on the parameter.
Results/Conclusions
We found that the variables both explored and explaining coyote occupancy across 18 different studies were heterogeneous, with 25 of 138 different variables recorded as significant. The number of covariates examined in a single study ranged from 2 to 30, with an average of 7.56 covariates per study (SE 1.56). Coyote occupancy was best estimated by ecological variables (15 significant variables), most numerously by prey-related variables, and appeared less sensitive to anthropogenic variables (9 significant variables). The most common ecological-based classification considered by studies was vegetation, while the most commonly considered anthropogenic classification was urbanization. Empirically we found little support for the variables we considered at the station, as our null model of occupancy ranked highly in AIC model comparisons (ψ = 0.74 ± SE 0.08). Coyotes are a ubiquitous species interacting with other carnivores, wild and domestic prey species, and humans. As landscapes continue to change and human pressures on ecosystems grow, assessing how coyotes respond to altering environments remains essential for understanding their dynamic role across the landscape.