Forest habitats in a mixed urban-agriculture mosaic landscape: patterns of mammal occupancy

Abstract

Conservation planning for biodiversity within anthropogenic landscapes is crucial given the rate of habitat conversion and human population growth. Investigating anthropogenic impacts on the persistence of biodiversity is key to management decision-making. We investigated the influence of protected areas (PAs), agriculture and urbanisation on the occupancy of mammal communities in an anthropogenic matrix containing indigenous forest fragments of the Coastal Belt of southern KwaZulu-Natal, South Africa. We integrated camera-trap mammal data, land-use and human population density within occupancy models, and compared occupancy of individual species across the land-use mosaic. We modelled occupancy of seven mammal species with sufficient naive occupancy (> 0.20, range 0.25–0.87). The occupancy of Philantomba monticola was positively influenced by human population size and was higher within urban areas compared with PAs. Although human population size positively affected Hystrix africaeaustralis occupancy, it along with Atilax paludinosus had a lower occupancy within urban areas. Tragelaphus scriptus and Potamochoerus larvatus overall had higher and Sylvicapra grimmia had lower occupancies within PAs. Species were variable in their response to the anthropogenic changes in the landscape. For example, occupancy of P. monticola was low within PAs but high in areas where change in land ownership and loss of habitat are threats. For other species, it appeared that the density of infrastructure of the urban landscape, rather than human population density, affected them negatively. However, seasonal differences within different management regimes also provided context-specific influences on occupancy and detectability. We emphasize the importance of natural vegetation patches within anthropogenic landscapes for maintaining native fauna, whilst stressing the value of temporally replicated, multi-species, regional-scale studies when making conservation decisions.