Modelling of road-kill hotspots in steppe landscape in Turkey

Roads and traffic have various negative ecological and environmental consequences on wildlife. Their impacts on animal populations are not limited to deaths. The main goal of the study is to analyze the spatial patterns of wildlife–vehicle collisions (WVC) at the Ankara–Kırıkkale highway. More specifically, we identified the road- and landscape-related features that may promote the WVC likelihood and locate the WVC hotspots. The hotspots of WVC were generated by CrimeStat 3 software. Variables related to the WVC locations were statistically analyzed, and then, they were modelled with logit regression approach. A total of 76 medium and large mammal were recorded from five species. The number of high-, medium-, and low-density clusters in the hotspots map that contains all the accidents are one, two, and four, respectively. For all species, the model shows that the probability of a fatal accident increases as the speed limit increases and the road slope decreases. To spatially identify where these events are concentrated provides valuable information for reducing the accidents and creating the correct strategies to ensure the safe migratory movements of wild animals.