Date Thesis Awarded

4-2019

Document Type

Honors Thesis

Degree Name

Bachelors of Science (BS)

Department

Interdisciplinary Studies

Advisor

Matthias Leu

Committee Members

Daniel Cristol

Robert Rose

Andrew Fisher

Abstract

ABSTRACT: Wildlife vehicle collisions are a significant problem for both humans and wildlife. Millions of dollars are spent annually on repairs and medical expenses; meanwhile, 1.23 million deer are killed annually nationwide. Information on the spatial distribution of, and contributing factors for, deer-vehicle collisions in Virginia is lacking. We used georeferenced deer vehicle collision (DVC) police records to map year-round high-risk areas for white-tailed deer (Odocoileus virginianus) on Virginia roads. We also classified high-risk, medium high-risk, medium low-risk, and low-risk areas for DVCs across the commonwealth based on quantile means and coefficients of variations. Using these risk classifications, we developed ordered linear regression models to identify environmental and road-related factors that contribute to the level of risk. Our spatial analysis revealed several hotspots along primary roadways, secondary roadways, and tertiary roadways. Our modelling revealed eighteen variables with appreciable effect on risk-level, including sinuosity, deciduous-agriculture edge length, proportion of urban development, and proportion of agriculture. Our research provides crucial information to policy makers and regulatory agencies in Virginia for the placement and development of DVC mitigation programs.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Available for download on Sunday, May 02, 2021

Share

COinS