Date Approved

8-24-2026

Graduate Degree Type

Thesis

Degree Name

Biology (M.S.)

Degree Program

Biology

First Advisor

Jennifer Moore

Second Advisor

Paul Keenlance

Third Advisor

Eric McCluskey

Academic Year

2025/2026

Abstract

Predation by mesopredators is widely recognized as a major threat to wood turtles (Glyptemys insculpta), yet whether spatial variation in predator density corresponds with differences in adult abundance remains largely unknown. We evaluated the relationship between raccoon (Procyon lotor) density and adult wood turtle abundance across 37 riparian site-year combinations in northern Wisconsin and Michigan's Lower Peninsula. Raccoon density was estimated from standardized camera-trap surveys using the Random Encounter Model, while adult wood turtle abundance was estimated using hierarchical negative binomial N-mixture models that accounted for imperfect detection. Candidate abundance models evaluated the influence of raccoon density, stream width, stream velocity, watershed, and state on adult wood turtle abundance. Contrary to our hypothesis, raccoon density received little support as a predictor of adult wood turtle abundance. The highest-ranked abundance model included linear and quadratic effects of stream velocity (QAICc weight = 0.37), whereas the null model ranked second and models incorporating raccoon density received substantially less support. Although raccoon density varied considerably among watersheds and study sites, this variation did not correspond to differences in estimated abundance among sites. Collectively, these findings indicate that habitat characteristics, particularly stream velocity, received greater support as predictors of adult wood turtle abundance than predator density and suggest that raccoon density alone may not adequately represent predation pressure at landscape scales. This study provides one of the first landscape-scale evaluations of the relationship between raccoon density and wood turtle abundance in the Upper Great Lakes region, advancing our understanding of wood turtle predator–prey dynamics while providing a framework for integrating predator density estimates into hierarchical abundance models and informing future conservation and management of this threatened species.

Comments

Grand Valley State University, United States Fish and Wildlife Service, USDA Forest Service, Wisconsin Department of Natural Resources

Available for download on Wednesday, August 25, 2027

Included in

Biology Commons

Share

COinS