Leopards are captured using a combination of free darting and soft-release foot snares. Adult leopards are fitted with a 300-400 g GPS collar (Vectronic-Aerospace) or a 150 g VHF collar (Sirtrack). Subadult leopards (>12 months old) are only fitted with VHF collars and a drop-off device is included. We deploy drop-offs to accommodate for growth and the unlikely event that a collar fails or an individual disperses out of range and cannot be recaptured. They are programmed according to the animal’s age (we typically set them to release after 12 months). We attempt to locate every collared leopard at least once every 3 days and record their location to the nearest 50 m using a hand-held GPS receiver. If a close approach is not possible, the animal’s position is determined by radio-triangulation and plotted on a 100 × 100 m grid map. When leopards are observed, we record a wide variety of ecological data, including information on hunting attempts, kills, reproductive behavior, territorial activity, activity patterns, etc. We collect scat opportunistically for dietary analysis. GPS collars are programmed to take five fixes daily and we download data remotely to a handheld terminal every fortnight.
We conduct camera-trap surveys in Phinda, Mkhuze and the farms to the south of Phinda every 2 years to estimate density and gauge population trends over time. Each survey consists of 30 camera-trapping stations each of which has two camera-traps (a total of 60 camera-traps). The stations are positioned along roads, game trails and other areas likely to be frequented by leopards. Each survey covers 100-120 km2 and runs for 50 days. We monitor units every 3 days to replace film, exchange memory cards and change batteries. Abundance estimates are determined using capture-recapture models, as used elsewhere to generate rigorous estimates for other individually identifiable carnivore species. We have demonstrated that this is a reliable technique to estimate leopard population size. We calculate population density at each site using 2 methods: analysing capture histories in the software CAPTURE and dividing the resultant abundance by the size of the effectively sampled area that incorporates a buffer determined from photographic recaptures, and ii) using spatially explicit capture-recapture models in the software DENSITY. We also assess which factors (natural and anthropogenic) affect leopard abundance and distribution. We hope to extrapolate our estimates (and those from our leopard studies in the region e.g. Zululand Rhino Reserve, Hluhluwe-Imfolozi Park, etc.) using appropriate GIS data to estimate leopard population abundance throughout KwaZulu-Natal. We also provide photographs of non-target, priority species (e.g. black rhino, cheetah, wild dog) to EKZNW for management and monitoring purposes.
“Kezi” is one of the resident Phinda Reserve leopards tracked by the Munyawana staff. Currently, she has a dependent male cub, shown here posing for a camera trap. Click here to read more about Kezi and her cubs.
We conduct interviews with livestock owners and commercial game farmers to determine attitudes towards leopards (and other carnivore species) and assess avenues for conflict mitigation. Using semi-structured questionnaires and a participatory approach, we question landowners about the physical features of their properties, the presence of leopards on their land, previous problems experienced with leopards, husbandry techniques employed, the number of leopards removed from their properties, and their perceptions on the protocols for managing leopards in KwaZulu-Natal. This is part of an exhaustive process covering all socio-economic aspects relevant to leopard conservation in the province. We also respond to all livestock depredation incidents in our immediate study area.