Statistical methods for camera trap surveys of snow leopard populations

Abstract

The population size of a species is critical to its conservation policy. In the case of the snow leopard (Panthera uncia), it is the difference between being classified as a vulnerable versus an endangered species in the IUCN Red list. Camera traps have been shown to be an effective tool to monitor the snow leopard and spatial capture-recapture (SCR) a robust statistical framework to infer its ecology. We develop novel extensions to SCR to investigate snow leopard ecology from camera trap surveys.

SCR assumes that individuals detected on camera traps are perfectly identified. Misidentifications however are common. The most common error of not being able to recognise a previously detected individual results in single-encounter capture histories that can bias abundance estimates by at least 25% at the recorded identification error rates. We develop the SCR-R model that conditions on at least R detections of an individual, of which SCR is a special case of R = 1. We show estimates to be nominally unbiased when there are misidentification errors using the SCR-2 method. We apply our method to multiple surveys of snow leopards in Mongolia.

Integrating data from multiple surveys can also improve inference. We are interested in integrating telemetry data in camera trap surveys within SCR. Current methods to do this in maximum likelihood estimation methods are inadequate. We develop novel models to bridge gaps in existing models by ensuring data from both surveys arise from a common space use distribution and by jointly modelling the data of an individual from both datasets rather than treating them as independent. We show that telemetry integrated SCR models can increase the precision of parameter estimates. We also investigate inference on parameters beyond density from these models. We apply these models to a multi-year survey of a snow leopard population in Tost, Mongolia to investigate population density, space use heterogeneity and philopatry.

Date of Award	3 Jul 2025
Original language	English
Awarding Institution	University of St Andrews
Supervisor	David Louis Borchers (Supervisor), Hannah Worthington (Supervisor) & Khoustubh Sharma (Supervisor)