Modeling the occupancy of sympatric carnivorans in a Mediterranean ecosystem

Abstract

Site occupancy provides a reasonable estimate of population status and trends, and it also provides an unbiased, cost-effective alternative method for large-scale, multispecies monitoring programs. In this study, we used camera-trapping data to determine carnivoran occupancy and associated environmental factors in Serra da Malcata Nature Reserve, Portugal. The study was intended as a precursor of further long-term multispecies monitoring programs. We estimated carnivoran species occupancy using a likelihood-based method, using the software PRESENCE. The major conclusions of the study were (1) fox occupancy tends to be independent of environmental factors; (2) stone marten occupancy is related with habitat variables, landscape structure, and preys; (3) common genet occupancy is related to broad leaf formations and preys; and (4) mongoose occupancy is higher in extensive areas of shrub habitats. Methodologically, we demonstrated the importance of modeling detection probabilities for species with low or variable detection rates. In the future, monitoring programs could benefit from incorporating estimates of detection probabilities into their design and analysis.