Spatial Determinants of Animal Roadkill Occurrence and Hotspots, With Implications for Wildlife Conservation Along Nairobi-Mombasa Highway

Abstract

The continuous expansion of global road networks, driven by increasing human populations, poses an emerging threat to animal populations through vehicle collisions. This study identifies the causal factors and hotspots of animal roadkill along a 72 km stretch of the Nairobi–Mombasa Highway (from Kyumvi to Sultan-Hamud) and recommends appropriate mitigation measures to enhance wildlife conservation. Data on roadkill incidents were collected from 2013 to 2018 through road patrols conducted by field scouts. Environmental variables, including water points, corners, vegetation, culverts, migratory routes, settlements, slopes, land use, and fences, were gathered using Geographic Information System (GIS) techniques. A presence-absence binary logistic regression model within a GIS environment was employed to identify significant environmental drivers of roadkill and to map hotspots. A total of 408 roadkill incidents (0.944 roadkill km⁻¹ y⁻¹) were recorded, comprising 51 species from six animal groups. Wild herbivores (13 species) and carnivores (14 species) constituted the majority of the documented wildlife, with a density of 0.204 and 0.157 fatalities per kilometre of road per year, respectively. Roadkill incidents for all species combined were significantly higher in areas near migratory routes, settlements, culverts, and corners, while they were lower in grassland, dwarf shrub grassland, open shrub areas, and near fences. However, the roadkill of different animal groups was influenced by distinct sets of factors. Most roadkill hotspots were located between Konza Center and Salama Town, with birds having the shortest stretch and herbivores and carnivores having the longest stretches. Roadkill can be mitigated by establishing road signs, fences, speed bumps, overpasses, and underpasses, and maintaining a cleared vegetation zone along the road. These findings can assist wildlife managers and infrastructure engineers in incorporating wildlife welfare and the concerns and attitudes of local communities when planning and developing linear infrastructure projects.