Using an occupancy approach to identify poaching hotspots in protected areas in a seasonally dry tropical forest

Poaching activity has been described in the literature as harmful due to impacts on biodiversity, especially in protected areas. Although the main reason for this activity is subsistence, in many regions motivation goes beyond the limits of food necessity. We applied single-species, single season occupancy models to evaluate the spatial distribution of poachers and identify potential poaching hotspots in a mosaic of protected areas in the Caatinga domain, northeastern Brazil. We used camera-traps over a period of 200 days at 60 sites randomly selected. We used distances from human settlements, roads and the nearest water holes, frequency of game species and sampling effort as covariables that could influence poachers' occupancy and detectability and to identify potential poaching areas. Occupancy poachers were higher in sites with higher frequency of game species. Frequency of game species and distance from roads had a negative effect on the detectability of poachers. Spatial analysis indicated three critical poaching areas within and around the Boqueirão da Onça National Park, associated with roads and some isolated cattle and goat farms. In this study, we provided an assessment of poaching spatial patterns in relation to different landscape elements and biotic influences, indicating critical areas where enforcement efforts should be focused. Hotspots are clearly concentrated within and on the edge of National Park. The approach presented here to identify poaching hotspots is effective and economical, and therefore may be applied in other protected and non-protected areas throughout the world.