Quantifying perceived risk in a small mesocarnivore, the bat-eared fox

Abstract

The perceived risk of predation can induce anti-predator responses such as the spatial and temporal avoidance of predators. However, such responses come with a level of cost that can potentially have implications for fitness – described as ‘non-lethal effects’. While the non-lethal effects of predators on herbivore prey are well investigated, the non-lethal impacts of predators on mesopredators/mesocarnivores are less understood. Importantly, there is reason to expect mesopredators’ anti-predator responses to be greater than those of herbivores, considering that apex predators represent both predation risk and competition. In this thesis, the effects of temporal, spatial, social and anthropogenic factors on the perceived risk of a small mesopredator, the bat-eared fox (Otocyon megalotis), were explored using both experimental and observational approaches. The anti-predator behaviours of this species are virtually undescribed and as large predators, e.g. lions (Panthera leo) and spotted hyaenas (Crocuta crocuta), were historically extirpated from the area, it was unclear if anti-predator responses would have disappeared, or still remain. Using giving-up-density (GUD) experiments, I demonstrated that bat-eared foxes experience greater perceived risk in dark conditions and lower perceived risk in the presence of humans. Vigilance, however, did not appear to vary with these same factors, suggesting that GUDs are capable of detecting more subtle differences in perceived risk. Furthermore, by evaluating how bateared foxes use high-cost vigilance (which interrupts other activities) and low-cost vigilance (which occurs simultaneously with other activities), I demonstrated that fox vigilance behaviour is dynamic. Vigilance was generally focused towards that of low-cost, with the occasional use of high-cost vigilance under certain conditions. High-cost vigilance increased with vegetation height, in the presence of adult conspecifics, and in winter. These effects were most likely due to impeded lines of sight, higher levels of competition, and increased social interactions respectively. My results suggest that in areas of low predation risk, mesopredators retain responses to certain cues of risk, but adapt behaviours to reduce the associated costs, allowing more time to be allocated to other activities. Finally, I determine that personality and plasticity was evident in this population of bat-eared foxes, varying across lunar illumination, wind speed, and temperature. Interestingly, these patterns were only distinct when vigilance was classified as high- and low-cost, and patterns were masked when vigilance types were combined. Individual foxes demonstrated distinct strategies when engaged in high-cost vigilance, where duration of vigilance did not fluctuate among individuals but rate varied significantly. Comparatively, individuals consistently differed in both bout duration and frequency of low-cost vigilance. I propose that the area’s low predation pressure is unlikely to constrain individual variation in behaviours. Thus, individual differences in high-cost vigilance may also be adaptive – in contrast to the ecological hypothesis of Favreau et al. (2014), whereby individuals that experience similar ecological conditions behave in a similar manner. Until this study, personality and plasticity in different types of vigilance behaviours has never been demonstrated in mesopredators. Ultimately, my research highlights that when predation pressure is extremely low, it is premature to assume that anti-predator behaviours have been lost. Anti-predator behaviours may still persist, and vary with spatiotemporal changes, in the presence of conspecifics, and amongst individuals. Future research on mesopredator responses to perceived risk should consider investigating different types of vigilance behaviour, as well as the inclusion of individual differences. Combining vigilance types may mask biologically salient differences in personality and plasticity, and distinct behavioural patterns may be undetectable without the consideration of individual variation. Importantly, these differences may be crucial in revealing information on the ecological constraints placed on populations. Key terms: Giving up densities, Habituated foxes, High-cost vigilance, Individual variation, Low-cost vigilance, Observer effects, Otocyon megalotis, Spatiotemporal effects.