Assessing the relative importance of landscape and husbandry factors in determining large carnivore depredation risk in Tanzania’s Ruaha landscape

• Published in: Biological conservation 2014-12, Vol.180, p.241-248
• Main Authors: Abade, L., Macdonald, D.W., Dickman, A.J.
• Format: Article
• Language: English
• Subjects: Animal, plant and microbial ecology; Applied ecology; Biological and medical sciences; Conservation, protection and management of environment and wildlife; Crocuta crocuta; ENFA; Fundamental and applied biological sciences. Psychology; General aspects; General aspects. Techniques; Grazing; Human–carnivore conflict; Husbandry; Land; Landscapes; Livestock; Mathematical models; Maxent; Methods and techniques (sampling, tagging, trapping, modelling...); Modelling; Panthera leo; Panthera pardus; Parks, reserves, wildlife conservation. Endangered species: population survey and restocking; Raw materials; Risk; SVM
• ISSN: 0006-3207; 1873-2917
• DOI: 10.1016/j.biocon.2014.10.005

•We generated a predictive risk map of large carnivore predation on livestock.•Traditional husbandry was insufficient to outweigh innate risks of the landscape.•Predation risk increased closer to rivers, in low altitude and vegetation cover.•41% of the study area was mapped as high-risk for livestock grazing. Tanzania’s Ruaha landscape, centred around Ruaha National Park (RNP), is an international priority area for large carnivores, supporting >10% of the world’s lions (Panthera leo) and globally significant populations of leopards (Panthera pardus), spotted hyaenas (Crocuta crocuta) and other carnivores. However, Ruaha’s carnivore populations cause intense conflict with local people, mainly due to livestock depredation, and are exposed to alarming rates of retaliatory killing, especially on village land adjacent to RNP. Depredation risk is likely to be influenced by both habitat features and livestock husbandry, but the specific environmental risk factors, and relative importance of habitat and husbandry, have never been assessed in this landscape. Here, we assessed which ecogeographic variables (EGVs) were associated with depredation risk for grazing livestock on village land, and generated a predictive map of large carnivore predation risk, based on species distribution modelling algorithms (SDMs). Secondly, we investigated the relative influence of husbandry and EGVs on depredation risk of enclosed stock, based on a generalized linear model. Grazing livestock predation risk was higher closer to rivers, and in areas of lower elevation and low percentage of tree cover, with 41% of the area mapped as high-risk. For enclosed stock, predation risk was mostly influenced by low percentage of tree cover and increased rainfall, with no discernible influence of current husbandry, which suggests that traditional husbandry was insufficient to outweigh the innate predation risks associated with high-risk landscape areas. Adopting new husbandry methods, such as specialised guarding dogs and fortified livestock enclosures, could be valuable for reducing depredation and carnivore killing in the Ruaha landscape.