Predicting reintroduction outcomes: assessing the feasibility of reintroducing African wild dog to a small protected area

Large mammalian carnivores have experienced significant contractions in population sizes and geographical ranges. The conservation of large carnivores is crucially important, particularly due to their vulnerability to extinction and their functional significance and ability to structure ecosystems. Due to an expanding human footprint, large carnivores are increasingly subject to modified and spatially constrained habitats. A growing debate exists as to how to conserve and coexist with large carnivores in an anthropogenically induced environment. Reintroduction, as a conservation tool to restore locally extirpated large carnivores to portions of their former ranges is increasingly being applied. However, in South Africa, habitat to support large carnivores remain small and non-contiguous. Food is a fundamental ecological requirement to sustain reintroduced large carnivores. Therefore, an understanding of large predator foraging patterns can be informative in the context of how the predator species influences and utilises a novel ecosystem. We investigated the foraging behaviour of reintroduced African wild dog (Lycaon pictus) at five small protected areas in KwaZulu-Natal, South Africa. Wild dog utilised 16 prey species, albeit they primarily used nyala (Tragelaphus angasii) and impala (Aepyceros melampus) which collectively form 75 % of their diet. Only nyala was significantly preferred, suggesting that this prey species is used in greater proportion to their abundance. As wild dogs are social cooperative hunters, we tested whether wild dog pack size was correlated to prey mass selection. There was no evidence to suggest that larger packs use larger prey. However, the mean wild dog pack size in our sample sites, was relatively smaller than those encountered elsewhere. Furthermore, wild dog have been shown to modify their hunting behaviour in the presence of wildlife-proof fencing, by using fences to aid in the capture of larger prey species than would innately occur. We compared the prey mass of wild dog kills in relation to proximity of these hard boundaries. Despite the affinity towards kills occurring within 200 m of fences, the upward bias caused by fences on prey mass selection was inconsistent across sample sites. The relatively small size of wild dogs makes them particularly vulnerable to competition. As the energetic output of wild dog is high, interspecific competition can increase foraging costs. The reintroduction of large carnivores to small artificially induced systems may be a contentious issue as resources available to support large carnivores are expected to be relatively more finite. We compared both the potential inter- and intraspecific dietary niche dimensions of an intact large carnivore guild in context of a wild dog reintroduction. We determined cheetah (Acinonyx jubatus), leopard (Panthera pardus), and lion (Panthera leo) prey composition, diet breadth, overlap, prey preference and predicted the density of an intact large carnivore guild in a novel landscape. Further, we compared the foraging behaviour of these large carnivores to that of wild dog. Our findings suggest that large African carnivores in small protected areas are subject to a considerable dietary niche overlap. Wild dog and cheetah, particularly reproductive females with dependent offspring, displayed the greatest potential for dietary overlap and subsequent competition. Leopard and lion at the population species level exhibited greater degrees of foraging plasticity. Lion displayed a contrasting prey species preference to sympatric predators as they selected for prey items frequently avoided by cheetah, leopard and wild dog. The proposed wild dog reintroduction site is expected to sustain seven wild dog based on the availability of preferred prey biomass. However, the reintroduction of wild dog to the small protected area is expected to have negative lateral trophic influences on other species of conservation concern. This should be of vital importance to management of the protected area. As the influence of competition in food-web and population dynamics, particularly in resource poor environments may be profound, our research highlights the need to assess the influence of competitive forces in structuring and restoring large predators to portions of their historical range