Comparison of methods for estimating Amur tiger abundance

Abstract

The Amur tiger Panthera tigris altaica is endangered throughout its range. Estimating abundance to monitor population trends is needed to judge success of conservation efforts. However, low densities and elusive behavior make precise estimates difficult to achieve. Managers must employ sampling methods that maximize precision while remaining feasible under constraints inherent to areas tigers inhabit. We applied camera trap, hair snare, DNA collection, scent-matching dog, morphometric track identification and track survey methods and developed a rubric to evaluate their efficacy in estimating tiger abundance in Russia. We conducted sampling over the course of one year in Sikhote-Alin Biosphere Zapovednik (SABZ). We ranked candidate methods using logistical, statistical and cost criteria to evaluate their efficacy for monitoring tiger numbers. Due to logistical and technical constraints, we relied on data from a parallel study to derive three statistical criteria for the DNA collection method. Camera traps, DNA collection, and track surveys were the only methods with sufficient sample sizes to estimate abundance for each sampling session during which the method was tested. DNA collection and camera traps were promising statistically, though poor-performing for logistics and cost. Technical improvements in camera trap design since the time of this study may improve the method's statistical and cost performance compared to our evaluation. Conversely, track surveys were most efficient logistically and financially, but poor statistically. Method choice for any given monitoring situation should consider local conditions and the specific objectives of the monitoring program in light of all three criteria -- statistical, logistical and cost. Failure in any of these areas can lead to failure of a monitoring program.