Assessing the Drivers of Distribution for a Cryptic Species Over a Large and Rugged Landscape: Occupancy Modeling of...

Hunting and wildlife trade in Lao PDR (also known as Laos) has left species of larger wild fauna existing at low densities, almost exclusively in remote areas of rugged forest, and wary of human presence. This makes surveying of larger fauna in the country difficult as observation of individuals is remarkably rare. Due to the critical nature of faunal conservation in Lao PDR (many globally threatened species facing the immediate risk of extirpation) the difficulty of conducting faunal surveys does not supersede the need to do so. Occupancy estimation and modeling, using repeated sampling of presence/non-detection of a site, offers a potential solution to this challenge. We conducted presence/non-detection sampling of 80 sites spread across the potentially suitable habitat for the Critically Endangered northern white-cheeked gibbon in Nam Et-Phou Louey National Park in the northeast of Lao PDR. We used maximum likelihood models to test the significance of several natural and anthropogenic covariates on the population's detection and occupancy probability to assess the drivers of the population's distribution. Detection was found to be a function of cloud cover, the proportion of bamboo forest, and topographic roughness of the site. Occupancy was found to be a function of human usage of the site, distance from the nearest road, and the uninterruptedness of forest cover. These results suggest that the viability of this species in Nam Et-Phou Louey National Park relies on preventing the expansion of all forms of roads into the park, maintaining continuous forest cover, and mitigating human presence in the habitat.