In the mountainous region at the border of Tennessee and North Carolina, Interstate 40 winds through the steep and rocky Pigeon River Gorge. The busy highway divides the Pisgah and Cherokee National Forests and is in close proximity to the Great Smoky Mountains National Park. This ecologically diverse region is one of the largest contiguous, publicly protected landscapes in the eastern US, making it a top priority for research and mitigation.
We conducted research on a 28-mile section of I-40 to: (1) identify high incidences of wildlife-vehicle collisions (WVCs) of target species, bear, deer, and elk, and identify road and landscape attributes that influence WVCs; (2) index target species activity rates along the roadside, compare activity to WVCs, and identify road and landscape attributes that influence wildlife activity; and (3) evaluate wildlife use of existing highway structures.
We curated WVC data from multiples state agencies (2001–2021) and conducted our own weekly driving surveys (2018-2021). Through quartile ranking we identified 22 “hotspot” segments. Our top model indicated that all species WVCs increased as distance to “usable” road structures decreased (p<0.01). Black bear VCs increased as forest area increased (p<0.05) and white-tailed deer VCs increased as ridge area decreased (p<0.001). In addition, steep road cuts—developed during highway construction and ubiquitous throughout—appear to be funneling wildlife and influencing WVCs. We also found all WVC data combined (i.e., crash reports, carcass removals, and driving surveys 2018-2020) yielded 6.9 times greater locations than crash report data only, indicating WVCs are highly underreported.
We deployed 66 wildlife cameras within the forested highway right-of-way, logging 43,958 trap days (2019–2020) and obtaining 6,598 independent target species detections. We found no significant linear relationship (p>0.05) between black bear or white-tailed deer detection rates and number of VCs. Our top models showed that black bear detection rates (i.e., activity) were higher at sites with greater forest area (p<0.05) and more protected land (p<0.001), and white-tailed deer activity was higher in areas with less highly rugged terrain (p<0.001).
We deployed 42 cameras at 21 highway structures, logging 22,305 trap days (2019–2020). Only 64 confirmed crossing events were detected, excluding data from two land tunnels, which had some of the highest detection rates. Only eight of 19 structures (42%) had confirmed crossings by one or more target species, with only one (5%) with crossings of all target species.
Based on our results, we provided 20 detailed mitigation recommendations to our department of transportation partners in NC and TN for improvements to existing structures and creation of new structures that will increase highway permeability, reduce WVCs, and restore flow of regional plants and animals, thus improving ecosystem resiliency in the Southern Appalachians.