Is anthropogenic cougar mortality compensated by changes in natural mortality in Utah? Insight from long-term studies

Understanding the interplay between exploitation and natural mortality is essential to guiding sustainable conservation of wildlife. Exploitation of carnivores by humans has long been thought to result in compensatory reductions of natural mortality among survivors. If rates of human exploitation exceed natural mortality, however, such actions will 'add' to overall mortality and could imperil the sustainability of such actions. We applied competing risk analyses to ⩾16 years of data for heavily harvested and semi-protected cougar populations in Utah to test the additive and compensatory mortality hypotheses, while accounting for parameter uncertainty. We additionally tested for presence of the two primary mechanisms by which compensatory mortality can arise: density dependence and individual heterogeneity in mortality risks. Despite an opportunity for compensation in the heavily harvested population, we could not reject the additive mortality hypothesis when uncertainty in parameter estimates was accounted for. In the semi-protected population, however, we detected evidence for partial compensation of increased anthropogenic exploitation via reductions in natural mortality. As may be common in carnivore studies, we found that ignoring uncertainty in estimates of cause-specific mortality systematically led to biased conclusions regarding additive and compensatory mortality hypotheses. Efforts should be made to address and minimize this uncertainty in demographic studies of carnivores in order to avoid flawed management recommendations. To attain the necessary sample sizes for making sound inference, this may require that the spatial extent of management units be extended for territorial species with large home-range requirements.