Factors Affecting Mammalian Occupancy and Species Richness in Annapurna Conservation Area, Nepal

Abstract

ABSTRACT Species richness is an important metric used for undertaking conservation management decisions. However, species richness estimates are influenced by species detection probabilities, with potential to entirely overlook species during surveys. Occupancy models which account for imperfect detection provide unbiased estimates, ensuring accurate estimates of richness. We carried out a camera trap survey in the mountains of north‐central Nepal during 2017 and documented a total of 21 mammal species. Here, we used multi‐species occupancy models within a Bayesian hierarchical framework to reassess our initial species richness estimate and to understand the influence of environmental covariates on occupancy and species richness of mammals in the area. Our model estimated the mean species richness was ~26 species (95% CRI: 21–36 species), suggesting we might have missed ~5 species during the survey. The mean probability of occupancy and detection of mammal species were estimated to be 0.2895%CRI:0.08–0.46 and 0.02 (95% CRI:0.01–0.03) respectively. Mammalian species richness of the area had an anticipated positive relationship with tree canopy cover β=1.908,95%CI=0.989–2.827,p=1.95e−04 though its positive relationship with anthropogenic disturbance was surprising β=1.339,95%CI=0.334–2.344,p=0.012. Mammalian species richness had a quadratic relationship with elevation as is expected. This research contributes to baseline information of the mammal community ecology in north‐central Nepal and supports the need for future multi‐season surveys to understand the influence of temporal factors on mammalian community and species richness in the area.

Keywords

• Bayesian hierarchical model
• detection probability
• imperfect detection
• mammal community
• multi‐species occupancy models
• species richness