Estimating the density of a globally important tiger (Panthera tigris) population: Using simulations to evaluate survey design in Eastern Thailand. (January 2020)

• Record Type: Journal Article
• Authors: Ash, Eric; Hallam, Chris; Chanteap, Prawatsart; Kaszta, Żaneta; Macdonald, David W.; Rojanachinda, Wiroj; Redford, Tim; Harihar, Abishek
• Is Part Of: Biological conservation. Volume 241(2020)
• Publication Date: 2020-01
• Subjects: Panthera tigris -- Spatially-explicit capture-recapture -- Simulations -- Dong Phayayen-Khao Yai Forest Complex -- Population density -- Thailand; Conservation of natural resources -- Periodicals; Nature conservation -- Periodicals; Ecology -- Periodicals; Environment -- Periodicals; Environmental Pollution -- Periodicals; Electronic journals; 333.9516
• DOI: 10.1016/j.biocon.2019.108349
• Languages: English
• ISSNs: 0006-3207

Abstract

Abstract: Spatially explicit capture-recapture analysis is widely utilized for estimating densities of tigers ( Panthera tigris ). However, developing a robust study design capable of meeting assumptions and achieving study objectives may be difficult, particularly for low-density populations. Study design decisions for such fieldwork can be aided by simulations. Our goal was to (1) use simulations to investigate and evaluate study design and (2) generate a reliable estimate of density for a population of tigers in Thailand's Dong Phayayen-Khao Yai forest complex. Scenarios were parameterized with a range of potential density estimates ( D ̂ ) and detection function parameters ( g 0 and σ ). We designed a field-based trap configuration identified and compared it with simulated performance of a regular trapping array, over 45-day and 60-day sampling occasions. We compared simulation results (i.e. number of individuals [ n ], detections [ ndet ], relative standard error [RSE] and relative bias [RB]) and identified that the non-regular trapping array deployed for 60 sampling days would generate reliable density estimates. Our survey produced a density estimate of 0.63 ± SE0.22; (0.32–1.21) tigers per 100 km 2, from a model incorporating variation in sex for g 0 and σ, and a population estimate of 20 (14–33). Simulations closely reflected actual results under the null model. Our survey design performed reasonably well, generating a sufficient number of detections and individuals to estimate density of a globally important tiger population. Our results suggest simulations and use of non-regular trap arrays may be beneficial for areas with low species density in which generating sufficient detections is particularly challenging. Highlights: Tiger density in Dong Phayayen-Khao Yai (DPKY) was 0.63 (0.32–1.21) tigers/100 km 2 . The estimated population for the study area was 20 (14–33) individuals. We used simulations to validate a non-regular survey design prior to deployment. Simulations may be beneficial for designing density surveys for low-density species. Breeding in DPKY reinforces the global importance of this tiger population.