Abstracts

Anthropogenic activities, such as overexploitation (poaching, logging) and farming (livestock grazing), are the most serious threats to biodiversity and ecosystem services. The effects of these drivers may be synergistic and variable across different species. Many terrestrial large-bodied mammals experience range shrinkage and face extinction risks and population declines across the world. By these activities, humans either directly (prey poaching) or indirectly (logging and livestock grazing) affect the survival rates of large mammal species. Protected areas (PAs) have been the most effective tool to preserve native species. However, the effectiveness of PAs in relation to large mammal distribution or conservation in temperate forests has rarely been assessed, particularly at a large landscape scale. In this study, I assessed the effects of threats to seven native mammal species in the Hyrcanian forest of Iran, namely the Persian Leopard (Panthera pardus saxicolor), grey wolf (Canis lupus), brown bear (Ursus acrtos), bezoar goat (Capra aegagrus), Caspian red deer (Cervus elaphus maral), roe deer (Capreolus capreolus), and wild boar (Sus scrofa). In addition, I assessed whether there are direct effects of poaching on livestock depredation by large carnivores. I used a novel approach to survey mammal species occupancy over a large landscape (18 protected and non-protected areas) and walked 1204 km distributed randomly over 93 16-km2 cells. Field surveys resulted in 2876 animal signs of the above-mentioned species over three discrete surveys. I used single-season Bayesian occupancy modeling and estimated the occupancy and detection probability rates for each target species across the study sites. The results explicitly showed that grazing had negative and significant impact on the occupancy of the very patchily distributed Persian leopard ( = -1.65, Credibility Interval CI - 2.85 to -0.65), Caspian red deer ( = -1.36, CI -2.34 to -0.45) and roe deer ( = -1.61, CI -2.96 to -0.58) while logging negatively affected red deer ( = -0.82, CI -1.69 to -0.03). The intensities of grazing and logging were correlated (r = 0.59), followed by logging and poaching (r = 0.39), grazing and poaching (r = 0.37) (Chapter2). I estimated the population density of the main wild ungulate species hunted by poachers and Persian leopards using random encounter modelling from camera traps (2777 camera days), fecal standing crop sampling (38 km), direct line transect sampling (186 km) and double-observer point-counts (64 scans) (Chapters 3 and 4). The results suggest that, due to poaching, population densities of the Caspian red deer, bezoar goat and urial have decline by 66-89% compared to the 1970ies. However, wild boar abundance estimates have increased by 58% during the same period. Using wild prey encounter rates (1204 km) and interview data (n = 201), I estimated the effects of forest cover, IUCN category of reserves, distance to villages and livestock encounter rates on livestock depredation rates by Persian leopard and greyAdvisors/Committee Members: Waltert, Matthias (advisor), Balkenhol, Niko (referee), Mhlenberg, Michael (referee), Heymann, Eckhard W. (referee).