The theory of predation risk effects predicts behavioral responses in prey when risk of predation is not homogenous in space and time. Prey species are often faced with a tradeoff between food and safety in situations where food availability and predation risk peak in the same habitat type. Determining the optimal strategy becomes more complex if predators with different hunting mode create contrasting landscapes of risk, but this has rarely been documented in vertebrates. Roe deer in southeastern Norway face predation risk from lynx, as well as hunting by humans. These two predators differ greatly in their hunting methods. The predation risk from lynx, an efficient stalk-and-ambush predator is expected to be higher in areas with dense understory vegetation, while predation risk from human hunters is expected to be higher where visual sight lines are longer. Based on field observations and airborne LiDAR data from 71 lynx predation sites, 53 human hunting sites, 132 locations from 15 GPS-marked roe deer, and 36 roe deer pellet locations from a regional survey, we investigated how predation risk was related to terrain attributes and vegetation classes/structure. As predicted, we found that increasing cover resulted in a contrasting lower predation risk from humans and higher predation risk from lynx. Greater terrain ruggedness increased the predation risk from both predators. Hence, multiple predators may create areas of contrasting risk as well as double risk in the same landscape. Our study highlights the complexity of predator–prey relationship in a multiple predator setting.
In this study of risk effects in a multi-predator context, LiDAR data were used to quantify cover in the habitat and relate it to vulnerability to predation in a boreal forest. We found that lynx and human hunters superimpose generally contrasting landscapes of fear on a common prey species, but also identified double-risk zones. Since the benefit of anti-predator responses depends on the combined risk from all predators, it is necessary to consider complete predator assemblages to understand the potential for and occurrence of risk effects across study systems.