Evaluating human–coyote encounters in an urban landscape using citizen science

Abstract Coyotes are ubiquitous in habitats across North America, including in urban areas. Reviews of human–coyote encounters are limited in scope and analysis and predominantly document encounters that tend to be negative, such as human–wildlife conflict, rather than benign experiences. The objective of our study was to use citizen science reports of human–coyote interactions entered into iNaturalist to better understand the range of first person accounts of human–coyote encounters in Madison, WI. We report 398 citizen science accounts of human–coyote encounters in the Madison area between October 2015 and March 2018. Most human–coyote encounters occurred during coyote breeding season and half of all encounters occurred in moderate development land cover. Estimated level of coyote aggressiveness varied significantly, with 90% of citizen scientists scoring estimated coyote aggression as a 0 and 7% scoring estimated aggression as a 1 on a 0–5 scale (with 0 being calm and 5 being aggressive). Our best performing model explaining the estimated distance between the human observer and a coyote (our proxy for a human–coyote encounter) included the variables distance to nearest paved road, biological season of the year relative to coyote life history, and time of day/night. We demonstrate that human–coyote interactions are regularly more benign than negative, with almost all first-hand reported human–coyote encounters being benign. We encourage public outreach focusing on practices that can foster benign encounters when educating the public to facilitate human–coyote coexistence.