Using non-invasively collected genetic data to estimate density and population size of tigers in the Bangladesh Sundarbans

Population density is a key parameter to monitor endangered carnivores in the wild. The photographic capture-recapture method has been widely used for decades to monitor tigers, Panthera tigris, however the application of this method in the Sundarbans tiger landscape is challenging due to logistical difficulties. Therefore, we carried out molecular analyses of DNA contained in non-invasively collected genetic samples to assess the tiger population in the Bangladesh Sundarbans within a spatially explicit capture-recapture (SECR) framework. By surveying four representative sample areas totalling 1,994 km2 of the Bangladesh Sundarbans, we collected 440 suspected tiger scat and hair samples. Genetic screening of these samples provided 233 authenticated tiger samples, which we attempted to amplify at 10 highly polymorphic microsatellite loci. Of these, 105 samples were successfully amplified, representing 45 unique genotype profiles. The capture-recapture analyses of these unique genotypes within the SECR model provided a density estimate of 2.85 ± SE 0.44 tigers/100 km2 (95% CI: 1.99-3.71 tigers/100 km2) for the area sampled, and an estimate of 121 tigers (95% CI: 84-158 tigers) for the total area of the Bangladesh Sundarbans. We demonstrate that this non-invasive genetic surveillance can be an additional approach for monitoring tiger populations in a landscape where camera-trapping is challenging.