Habitats associated with vehicle collisions with wild pigs

Context. Over the past few decades, the frequency of wild pig-vehicle collisions(WPVCs) and number of human fatalities associated with these accidents have increased with expanding populations of this species, particularly in regions outside its native distribution. Aims. To better understand this widespread and growing human safety threat, we quantified habitat attributes associated with 311 WPVC locations occurring between 1983 and 2012 at the Savannah River Site (SRS) in South Carolina, USA, to test the hypothesis that WPVCs occur more frequently in areas proximal to preferred habitats (i.e. riparian and bottomland hardwood habitats). Methods. At each collision site, we measured the distance to the nearest wetland and stream, as well as the composition of habitats within 100-m and 1699-m buffers. We then contrasted habitat attributes associated with collision sites with those from randomly selected locations along the same roads, to identify habitat characteristics contributing to a higher incidence of these accidents. Key results. WPVCs were non-randomly distributed across both spatial scales measured, with collisions occurring more frequently in areas of preferred habitat for this species. Specifically, collisions occurred in areas closer to streams and containing less pine forest than at random locations at both spatial scales evaluated. Conclusions. Similar to vehicle accidents with other ungulate species, our study suggested that vehicle collisions involving wild pigs are spatially clustered around preferred habitat types. Management efforts to reduce vehicle collisions with wild pigs should be focussed in areas where roadways bisect preferred habitats such as stream crossings and bottomland hardwood or other riparian habitats. Implications. These data will aid in the development of mitigation strategies to reduce the frequency and impacts of WPVCs in areas of high wild-pig densities. However, given the paucity of data on WPVC mitigation, additional research is needed to quantify the efficacy of various methods (e. g. signage, fencing, underpasses) at reducing the frequency and severity of collisions with this species.