Lions, hyenas, and cheetahs: Spatiotemporal avoidance in a landscape of fear

Background/Question/Methods

Top predators can create a “landscape of fear” in which subordinate species alter their use of the landscape to minimize the risk of aggressive encounters. Fine-scale avoidance strategies may minimize risk without resulting in costly large-scale displacement that reduces access of the subordinate species to resources. We evaluated patterns of active spatial and temporal avoidance in a guild of large African predators in Serengeti National Park, Tanzania. Lions and hyenas reciprocally kill and steal food from each other; both species kill and steal food from cheetahs and are widely believed to suppress cheetah populations. We deployed 225 camera traps across 1,125km² to evaluate spatial and temporal patterns of habitat use and avoidance among lions, hyenas, and cheetahs and validated camera-trap captures against concurrent lion GPS collar data. We applied hurdle models (or “zero-altered Poisson” models) to photographic capture rates to measure long-term spatial attraction or avoidance, while controlling for habitat preferences and localized prey abundance. We then evaluated active temporal attraction or avoidance (as the probability of a species appearing at a site) in the hours immediately following camera-trap captures of predators and prey. 

Results/Conclusions

We frame our analyses according to the respective competitive abilities of three species. As the dominant competitor, lions are expected to gain access to their preferred landscape characteristics. Hyenas should be sensitive to both lions and habitat, whereas cheetahs should distribute themselves according to lions, hyenas, and habitat. However, despite high levels of interspecific aggression, we see no evidence that lions displace cheetahs or hyenas, nor any evidence hyenas displace cheetahs. Hurdle models indicate that despite dissimilar habitat preferences, hyenas appeared more often at sites with more lions (p<0.0001). After controlling for similar habitat preferences, cheetahs were similarly seen more often at sites frequented by lions (p=0.034). Poisson regression of the number of hyena and cheetah sightings vs. time-since-lion-sighting suggest that while hyenas actively follow lions (p<0.0001), cheetahs avoid lion-occupied areas for several hours (p=0.018). These patterns suggest that cheetahs are able to use patches of preferred habitat by avoiding lions on a moment-to-moment basis. Such fine-scale temporal avoidance is likely to be less costly than long-term avoidance of habitat hotspots, and may help explain why cheetahs are able to co-exist with lions despite high rates of lion-inflicted mortality.