Predicting species assemblages at wildlife crossing structures using multivariate regression of principal coordinates

Area of Science:

• Wildlife ecology
• Conservation biology
• Transportation ecology

Background:

• Roads pose significant threats to wildlife populations globally.
• Wildlife crossing structures (WCSs) and fencing are mitigation strategies to reduce wildlife-road mortality.
• Understanding WCS effectiveness is crucial for wildlife management and infrastructure planning.

Purpose of the Study:

• To develop a predictive model for mammal community composition and usage of WCSs.
• To identify key factors influencing the utilization of WCSs by various mammal species.
• To assess the effectiveness of WCSs at the community level, not just for target species.

Main Methods:

• Utilized camera trap data to analyze mammal community composition at ocelot-focused WCSs.
• Developed a predictive model incorporating spatial, temporal, structural, environmental, and anthropogenic variables.
• Evaluated model performance in predicting total detections, successful crossings, and failed crossings.

Main Results:

• Structural and anthropogenic characteristics were more significant predictors of WCS use in the first year than environmental factors.
• The model showed reasonable accuracy in predicting total mammal detections at WCSs.
• Predicting successful and failed crossings was less accurate, potentially due to micro-scale factors like noise and microclimate.

Conclusions:

• The developed model offers a generalized framework for predicting WCS use by mammal communities.
• Initial WCS effectiveness is strongly influenced by physical and human-related features.
• Further research into finer-scale environmental dynamics is needed to improve predictions of crossing success.