Differences in spatiotemporal patterns of vehicle collisions with wildlife and livestock

Road ecology research on animal-vehicle collisions (AVCs) in North America has tended to focus on collisions with wildlife species, especially large ungulates. Collisions with domestic livestock are not differentiated from collisions with wildlife in many agencies' data collection systems and in many scientific studies. This lack of differentiation between large domestic animal-vehicle collisions (DAVCs) and large wildlife-vehicle collisions (WVCs) limits our understanding of where, when, and how frequently DAVCs occur, and whether these patterns differ from those for WVCs. We used a 10-year, statewide collision data set for Montana to explore AVC patterns at multiple spatial and temporal scales and determine whether these patterns are different for DAVCs and WVCs.
Analyses of >25,000 AVC records indicated that WVCs exhibited two daily peak periods (around dawn and dusk) each day, while DAVCs exhibited only one prominent peak period (late evening/early night). Relative to WVCs, DAVCs were more likely in dark conditions and less likely in dusk, daylight, and dawn conditions. Both collision types were least common during winter and early spring, and most common during summer and fall, but DAVCs exhibited a more prominent late-fall peak. Locations of DAVC and WVC hotspots were not spatially consistent at either coarse or fine spatial scales. We observed large variation among Montana counties with respect to total numbers of WVCs and DAVCs recorded and the proportions of each, with DAVCs over-represented relative to WVCs in most eastern Montana counties and under-represented in most western Montana counties. Within two counties selected for finer-scale analysis, we found only a moderate degree of overlap between one-mile road segments with high WVC concentrations and those with high DAVC concentrations.
Our results highlight that DAVCs can be locally common even when regionally rare, and thus they may represent a high priority for management and mitigation measures to protect human safety in some areas. Although less common than WVCs, DAVCs are more dangerous to motorists on a per-collision basis and occur disproportionately on tribal lands in Montana. Mitigation measures that address collisions with domestic animals may be different than those for wildlife; for instance, tagging livestock with reflective materials and installing lighting along high-risk road segments could help to reduce DAVCs, but the effectiveness of these measures has not yet been demonstrated. We recommend that future highway mitigation research designs include collection of detailed information not only on AVC locations and timing, but also on the species of animals killed. Information on local grazing laws, status of roadside fencing, local livestock densities, proximity of public grazing allotments or private pastures to roads, and traditional routes used by livestock moving between home ranches and seasonal grazing locations could help to explain and predict DAVC hotspots.