A death trap in the landscape of fear

A crucial element in the the landscape of fear concept is that prey animals are aware of varying levels of predation risk at a spatial scale. This often leads to a negative spatial relationship between prey and predator in which prey avoid the most risky sites in the landscape. In this paper, we argue that our understanding of large carnivore-ungulate interactions is biased by studies from highly heterogeneous landscapes (e.g. the Yellowstone National Park). Due to a high availability of refuges and foraging sites in such landscapes, prey are able to reduce predation risk by showing habitat shifts. Besides the spatial heterogeneity at the landscape scale, the ungulate response to predation risk can be affected by the hunting mode (stalking vs. cursorial) of the predator. We propose that prey cannot easily avoid predation risk by moving to less risky habitats in more homogenous landscapes with concentrated food resources, especially where the large carnivores' assemblage includes both stalking and cursorial species. No distinct refuges for prey may occur in such landscapes due to equally high accessibility to predators in all habitats, while concentrated resources make prey distribution more predictable. We discuss a model of a densely forested landscape based on a case study of the Bialowieza Primeval Forest, Poland. Within this landscape, ungulates focus their foraging activity on small food-rich forest gaps, which turn out to be death traps as the gaps are primarily targeted by predators (stalking lynx and cursorial wolf) while hunting. No alternative of moving to low predation risk areas exist for prey due to risk from wolves in surrounding closed-canopy forest. As a result, the prey is exposed to constant high predation pressure in contrast to heterogeneous landscapes with less concentrated resources and more refuge areas. Future research should focus on explaining how ungulates are coping with predation risk in these landscapes that offer little choice of escaping predation by considering behavioural and physiological (e.g. metabolic, hormonal) responses.