A GIS-based model was developed to identify potential amphibian road crossing locations statewide in Vermont. The purpose of the model is to provide a coarse screening tool for transportation planners to identify road segments that have a higher potential for spring amphibian migrations, where further review may be warranted. The focus was on species making spring migrations from overwintering to breeding habitat. A pilot study applied a similar model to one county in Vermont in 2019. That model’s results were reviewed and analyzed. Revisions were made and updated LiDAR-based land cover mapping was incorporated in extending the model statewide.
The model employs land cover mapping based on LiDAR and GIS modeling algorithms to identify potential breeding and nonbreeding habitat within 600 feet of roads and railroads. It also incorporates known vernal pool locations. A least-cost-distance approach was applied that made unfavorable habitats more “costly” to pass through by increasing the distance factor. The statewide model results were reviewed for 50 road segments where the presence or absence of actual amphibian crossings was already known. The model was then modified to obtain results that provided a balance between identifying too many and too few road segments as crossings. Approximately 7% of the 50-meter road and railroad segments were identified as potential crossings.
VTrans plans to use this new tool as a first-cut, coarse screening that indicates the potential for amphibian crossings and areas where additional research or field review may be warranted. Certain aspects of the model pose modeling challenges and warrant further research. In particular, intersections, irregular road geometry, small wetlands, and suitable wetland hydroperiods have proven problematic to model. VTrans will be applying the model to evaluate its effectiveness and to develop improvements.
amphibians
road mortality
crossing hotspots
Modeling