Assessing the spatio-temporal patterns of road kills on selected routes between Komani and Kokstad.

Abstract

Road kills are a challenge in the contemporary world and present potential negative impacts on society and the environment. Vehicle collisions with large animals can result in serious harm to motorists and their passengers. Those collisions that do not harm people still inevitably harm the animals involved. These harms can result in serious negative environmental impacts. The lack of attention given to the species composition of road kill perpetuates the challenges and impacts associated with animal-vehicle collisions (AVCs). Even if a species has a favourable conservation status, the impacts of AVCs can hinder local populations and ecosystem function for the areas those animals occupy. Although typically associated with road kill, wildlife is not always the category of animals in danger of animal related accidents. Other categories of biota such as domesticated animals (pets including cats and dogs) and livestock (animals owned for agricultural purposes including cows and sheep) have been known to fall victim to collisions, which can have impacts on the people who rely on them. In a given area, if the locations and times of a group of road kills is known, spatiotemporal patterns can be identified to explicate how and why road kills occur. Evaluating patterns and trends in road kill can enable opportunities to mitigate these incidences. The aim of this study was to assess the spatio-temporal patterns of road kills on selected routes between Komani and Kokstad in order to determine the underlying causes of animal-vehicle collisions and identify appropriate mitigation measures. The objectives were to; determine the species characterisation of road kills on the selected routes, to identify the spatio-temporal patterns, examine local factors contributing to AVCs, examine the impacts of road kill on local animal populations, and recommend suitable mitigation measures for a sustainable transport system. To achieve these objectives, a relevant road kill dataset was obtained from the Endangered Wildlife Trust, consisting of 249 road kills from 2007 to 2020, in the study area of interest. The road kill data was mapped in the study area using ArcMap 10.6 and ArcGIS Pro 3.0. The routes in the study area included the N2, N6, R56, R61, R63, R67, R349, and R396. Additionally, road infrastructure and climate data, retrieved from South African National Roads Agency SOC Ltd and South African Weather Services, respectively, were applied to identify any possible influencing factors for road kill. The road kill data was analysed using Optimised Hot Spot Analysis. A protected areas layer was incorporated to assess for a potential relationship between road kill incidents and proximity to protected areas. The climate and road infrastructure data were refined and evaluated in Microsoft Excel. The species composition comprised 35 different species of road killed animals; 249 specimens in total. A total of 154 road kills were unidentified species. The Optimised Hot Spot Analysis revealed 24 statistically significant hot spots in the study area. Of these 24 hot spots, 21 were in the 99% confidence interval and 3 were within the 95% confidence interval. Furthermore, 23 of these statistically significant hot spots were on the R56, and 1 on the N2, all of which were closer to Kokstad than Komani. These hot spots were also found across the provincial border between the Eastern Cape and KwaZulu-Natal. Proximity to protected areas, temperature, road width and a lack of appropriate road signage and speed limits were identified as potential influencing factors. In terms of temporal trends, recorded road kills were most frequent in the month of June and in the year 2019, for the 2007-2020 reporting period. Temperature data and correlation analysis revealed a weak relationship between road kill incidents and temperature. Correlation for road kill and road widths were found to be weak, overall, but require a more comprehensive analysis and extensive, available data. Spatio-temporal analysis and field observations, however, still provided important insights into road kill studies. The impacts of these AVCs were found to be a potential compromise of species population dispersal rates, gene flow, and ecosystem functioning. To mitigate these impacts, recommendations included implementing effective road signage, incorporating the challenges of AVCs in environmental and road policies, supervision of citizen scientists, and the development of an application that utilises a Geographic Information System (GIS) to predict and assess areas for probability of AVCs. This can help improve the judgement of drivers and reduce road kills.