Understanding the influences of human-coyote interactions and coyote distributions within northeast Ohio
(Book - Regular Print)

Thesis,MSES,Green Mountain College,2019

Includes bibliographical references.

With growing urban and suburban areas across the nation there is an increasing need to understand human-wildlife interactions. The Coyote (Canis latrans) is an adaptable and opportunistic species that can thrive in anthropogenic environments. This study was conducted with two goals in mind. First, to understand the influences of human-coyote interactions using public reports, and second, to use these interactions along with wildlife cameras to predict where future human-coyote interactions may occur. Using public reports, our research objectives were to discern when coyotes are most actively encountering humans and to also determine what activities trigger human-coyote interactions. Over three years, 261 reports were collected from the public documenting interactions in northeast Ohio. Reports were categorized according to variables such as date, time, season, and activity type. Chi-squared and logistic regression analyses were performed using R 3.3.1 (R Core Team, Vienna, Austria). Significant results were seen within the categorical variables of month, time of day, activity type and coyote biological season. We saw significant results with the time of day and activity type with interactions when reports were differentiated between basic observations and reports of concern from reporting individuals. From these results we were able to understand: 1) the likelihood of when the general public and coyotes are having an interaction that leads to a public report, 2) any activity type which involved the presence of a domestic dog was 4.7 times more likely to have an interaction of concern. The second half of this project used the locations of public reports and wildlife cameras to create species distribution models based on land cover variables. Using this date, several modeling methods were performed (machine learning, occupancy, and frequency regression). However, an interpretive model was not crated due to limitations and biases of the data and model type. Instead, this chapter focuses on the methods used and the benefits and pitfalls of each of these methods. The findings of this study are an important step in understanding factors that may lead to interactions, offer insights on coyote management within urban settings, and also provide valuable information on ways to improve future reporting and modeling attempts.