Monitoring carnivore populations at the landscape scale: occupancy modelling of tigers from sign surveys

摘要: Summary 1. Assessing spatial distributions of threatened large carnivores at landscape scales poses formidable challenges because their rarity and elusiveness. As a consequence logistical constraints, investigators typically rely on sign surveys. Most survey methods, however, do not explicitly address the central problem imperfect detections animal signs in field, leading to underestimates true habitat occupancy distribution. 2. We assessed for tiger Panthera tigris metapopulation across c. 38 000-km2 India, employing spatially replicated detections. Ecological predictions about presence were confronted with detection data generated from sampling 205 sites, each 188 km2. 3. A recent model that considers Markovian dependency among replicates performed better than standard (ΔAIC = 184·9). A formulation this fitted best showed density ungulate prey levels human disturbance key determinants local presence. Model averaging resulted replicate-level probability  = 0·17 (0·17) estimate  = 0·665 (0·0857) or 14 076 (1814) km2 potential 21 167 km2. In contrast, traditional presence-versus-absence approach underestimated by 47%. Maps probabilities site clearly identified source populations higher densities matched observed variations, suggesting utility population assessments scales. 4. Synthesis applications. Landscape-scale surveys can efficiently assess carnivore elucidate factors governing presence, provided ecological observation processes are both modelled. Occupancy using be used reliably identify sources help monitor metapopulations. Our results reinforce earlier findings depletion drivers extinctions tigers persist even human-dominated landscapes through effective protection populations. facilitates efficient targeting conservation interventions and, more generally, provides basis reliable integration monitoring between scales.