Baseline for monitoring and habitat use of medium to large non-volant mammals in Gran Sabana, Venezuela

The Gran Sabana is a region of great biogeographical and conservation value that has been recently threatened due to increasing overexploitation of natural resources and illegal mining. Systematic survey methods are required in order to study species responses to landscape transformation. The main objectives of this study were: 1) to test the relationship between habitat types and mammal species presence in the Gran Sabana and Canaima National Park (NP), and 2) establish baseline methodology that can set guidelines for future, considering sampling limitations, conservation opportunities and increasing threats to biodiversity in this region. We implemented a stratified sampling design using camera traps for monitoring medium and large mammals in two previously under-sampled regions in the Gran Sabana, south-east Venezuela. We analysed time-series of remotely sensed vegetation indices to classify habitat types and summarized relative abundance of mammals, naïve occupancy, and habitat fidelity for each type. With a sampling effort of 5,523 camera*days, 29 species of mammals were recorded, including endangered Priodontes maximus, and vulnerable: Speothos venaticus, Tapirus terrestris, and Panthera onca. Cuniculus paca and Dasyprocta leporina were the most frequently registered, while Hydrochoeris hydrochaeris, Leopardus wiedii, and Leopardus tigrinus had few records. Most of the species were associated with forest (13), fewer with transitional shrub (7) and savanna (5) habitats, but only some of these were statistically significant. Cerdocyon thous show significant association with savanna (P < 0.01), whereas Leopardus pardalis (P < 0.05) and Cuniculus paca (P < 0.05) with shrub-intermediate habitat. As many as seven species: Dasypus kappleri, Dasyprocta leporine, Mazama americana, M. gouazoubira, Nasua nasua, Priodontes maximus, and Tapirus terrestris have statistically significant association to forest habitat (P < 0.05). We present the first record Myoprocta pratti f