This study considers the dynamical interaction of two predatory carnivores (Lions (Panthera leo) and Spotted Hyaenas (Crocuta crocuta)) and three of their common prey (Buffalo (Syncerus caffer), Warthog (Phacochoerus africanus) and Kudu (Tragelaphus strepsiceros)). The dependence on spatial structure of species’ interaction stimulated the author to formulate reaction-diffusion models to explain the dynamics of predator-prey relationships in ecology. These models were used to predict and explain the effect of threshold populations, predator additional food and prey refuge on the general species’ dynamics. Vital parameters that model additional food to predators, prey refuge and population thresholds were given due attention in the analyses. The stability of a predator-prey model for an ecosystem faced with a prey out-flux which is analogous to and modelled as an Allee effect was investigated. The results highlight the bounds for the conversion efficiency of prey biomass to predator biomass (fertility gain) for which stability of the three species ecosystem model can be attained. Global stability analysis results showed that the prey (warthog) population density should exceed the sum of its carrying capacity and threshold value minus its equilibrium value i.e., W >(Kw + $) −W . This result shows that the warthog’s equilibrium population density is bounded above by population thresholds, i.e., W < (Kw+$). Besides showing the occurrence under parameter space of the so-called paradox of enrichment, early indicators of chaos can also be deduced. In addition, numerical results revealed stable oscillatory behaviour and stable spirals of the species as predator fertility rate, mortality rate and prey threshold were varied. The stabilising effect of prey refuge due to variations in predator fertility and proportion of prey in the refuge was studied. Formulation and analysis of a robust mathematical model for two predators having an overlapping dietary niche were also done. The Beddington-DeAngelis functional and numerical responses which are relevant in addressing the Principle of Competitive Exclusion as species interact were incorporated in the model. The stabilizing effect of additional food in relation to the relative diffusivity D, and wave number k, was investigated. Stability, dissipativity, permanence, persistence and periodicity of the model were studied using the routine and limit cycle perturbation methods. The periodic solutions (b 1 and b 3), which influence the dispersal rate (') of the interacting species, have been shown to be controlled by the wave number. For stability, and in order to overcome predator natural mortality, the nutritional value of predator additional food has been shown to be of high quality that can enhance predator fertility gain. The threshold relationships between various ecosystem parameters and the carrying capacity of the game park for the prey species were also deduced to ensure ecosystem persistence. Besides revealing irregular periodic travelling wave behaviour due to predator interference, numerical results also show oscillatory temporal dynamics resulting from additional food supplements combined with high predation rates.

Predator reintroductions are becoming increasingly more common for multiple reasons, including assisting with the conservation of a predator species, the restoration of ecosystem functions and the economic benefits of their reintroduction for ecotourism ventures. There remains however, little knowledge on prey species responses to these predator reintroductions. As such, the reintroduction of cheetah (Acinonyx jubatus) into Samara Private Game Reserve provided an opportunity to investigate prey responses to predator reintroduction across a range of spatial and temporal scales. More specifically, the aim of this study was to investigate the effect of cheetah predation risk on habitat use and behavioural responses of the resident ungulate community. Samara is divided into “predator present” and “predator absent” sections, providing the opportunity to conduct a comparative study investigating the effect of cheetah on prey responses. It was hypothesized that different ungulate species would respond differently to the presence of cheetah, depending on differences in perceived vulnerability to cheetah predation. To address this, shifts in habitat use, and behavioural responses of the ungulate community reflected at landscape and patch scale were investigated. Overall, ungulate species have not shifted habitat use since cheetah reintroduction, this was possibly related to life history strategy constraints and the need for individuals to obtain suitable forage and therefore remain in specific habitats. While no shift in habitat use was observed for the majority of ungulate species, a shift in behaviour was observed for kudu (Tragelaphus strepsiceros) at a landscape scale, where kudu dedicated more time to vigilance and less time to foraging within the predator section. In addition, individuals within smaller kudu groups were observed to be more vigilant than individuals within larger groups of kudu within the predator section. Although the other three ungulate species monitored did not increase time spent vigilant within the predator section, they still maintained relatively high levels of vigilance, potentially as a means of social monitoring. At a patch level, ungulate species responded strongly to a predator cue as a proxy for cheetah proximity, and increased time spent vigilant with a trade-off of lower foraging effort. This vigilant response was strongest for kudu. Differences in perceived predation risk were reflected within eland (Tragelaphus oryx) and kudu species demographic classes, with juvenile eland and kudu, adult female kudu and subadult female kudu spending more time vigilant within manipulated patches than respective males of each species. Kudu were also observed adopting fine-scale behavioural responses to minimize predation risk within patches. The asymmetrical prey species response to perceived predation risk from cheetah supports the hypothesis that different species respond differently to the presence of a predator. Furthermore, this study illustrated the importance of measuring prey responses to predation risk across multiple scales and highlighted the need to replicate this study for a number of different sites where predators have been reintroduced, to better understand the range of factors influencing these predator-prey interactions.

Recent research has focused on the non-lethal effects of predator intimidation and fear, dubbed non-consumptive effects (NCEs), in which prey actively change their behavior and habitat use in response to predator chemical cues. Although NCEs can have large impacts on community structure, many studies have ignored differences in predator population structure and properties of the natural environment that may modify the magnitude and importance of NCEs. Here, I investigated the roles of predator size and density (i.e. biomass), as well as habitat characteristics, on predator risk assessment and the magnitude of consumptive and NCEs using blue crabs, mud crabs, and oyster prey as a model system. Predation experiments between blue crabs and mud crabs demonstrated that blue crabs consume mud crabs; however, the consumptive effects were dependent upon blue crab body size and habitat type. When mud crabs were exposed to chemical cues from differing biomasses of blue crabs in laboratory mesocosms, mud crab activity and predation on oysters was decreased in response to high biomass treatments (i.e. large and multiple small blue crabs), but not to low biomass predators (i.e single small blue crab), suggesting that risk associated with predator size is perceptible via chemical cues and is based on predator biomass. Further experiments showed that the perception of risk and the magnitude of the NCEs were affected by the sensory cues available and the diet of the blue crab predator. The NCE based on blue crab biomass was also demonstrated in the field where water flow can disperse cues necessary for propagating NCEs. Properties of water flow were measured within the experimental design and during the experiment and confirmed cage environments were representative of natural conditions and that patterns in NCEs were not associated with flow characteristics. These results affect species conservation and commercial fisheries management and demonstrate that we cannot successfully predict NCEs without considering predator size structure and the contexts under which we determine predator risk.