A spatiotemporal analysis of ungulate–vehicle collision hotspots in response to road construction and realignment

Although roads are central to human society, they have many negative environmental impacts and create risk for traveling motorists. Our aim was to evaluate the spatiotemporal evolution of ungulate–vehicle collision (UVC) hotspots in response to major road construction. We examined two different locations and scales in the province of Alberta, Canada: (1) a highway bypass adjacent to a large city with 4.5 km of wildlife mitigation measures (wildlife fencing and two underpasses) and (2) 55 km of rural highway that was converted from a two-lane to a four-lane divided highway. Using government police collision and carcass data (2000–2021), before-after and control-impact analyses were used to assess changes in UVC rates. Our approach is novel in that we tested the paired use of a clustering method known as kernel density estimation plus and a spatiotemporal stepwise modification of this method to monitor UVC hotspots. By monitoring UVCs over space and time, we could identify stable vs. ephemeral UVC hotspots, a fence-end effect, and a barrier effect due to traffic volume, and we could explore hotspot stability before and after construction. The wildlife mitigation measures along the highway bypass resulted in 86% fewer UVCs compared to an unmitigated highway. At a larger scale, however, net benefits were affected by road density. The construction of a four-lane divided highway with no wildlife mitigation measures and an increase in the posted speed limit resulted in a slight increase in UVCs and the reemergence of the majority of historical UVC hotspots. Our analysis highlighted the need to incorporate wildlife considerations at a variety of scales throughout the transportation planning and mitigation evaluation process.