Carnivore Detection at the Domestic/Wildlife Interface within Mpumalanga Province, South Africa.

South African protected areas account for 8% of the total landmass according to World Bank indicators. Effective conservation of biodiversity in protected areas requires the development of specific reserve management objectives addressing species and disease management. The primary objective of the current study was to identify predictors of carnivore detection in an effort to inform carnivore species management plans on Andover and Manyeleti nature reserves in South Africa. A limited number of camera traps were placed randomly using a grid system. Species detection data were analysed using mixed-effects logistic regression and Spearman's correlation coefficients. Deterministic inverse distance weighted distribution maps were used to describe the spatial distribution of carnivore species. Camera traps identified similar species as traditional call-up surveys during the study and would be useful as an adjunct census method. Carnivore detection was associated with several variables, including the presence of specific prey species. The measured intra-and interspecies interactions suggested the risk of disease transmission among species, and vaccination for prevalent diseases should be considered to manage this risk.