There is a need for a refined understanding of large carnivore prey preference and carrying capacity (K). To date, K estimates for large carnivores have been developed from predictions of carnivore diet at a prey and predator species-level. These predictions therefore assume that all social classes within a carnivore species display similar prey preferences and that all demographic classes within a prey species are equally preferred or avoided. The objective of this study was to investigate the importance of including prey demographics and carnivore social class in carnivore diet descriptions and thereby K estimates, using cheetah Acinonyx jubatus as a study species. It was predicted that prey sex, prey age and cheetah social class influence cheetah prey preferences, when they influence the risk and ease of prey capture, and that their inclusion in a K model would improve its predictive strength. Based on an analysis of 1290 kills from South Africa, male coalition cheetah were found to prefer a broader weight range of prey than solitary cheetah. Prey demographics further influenced cheetah prey preference, when it corresponded to differences in prey size and the presence of horns. The current species-level K regression model for cheetah is based on preferred prey and thus omits highly abundant antelope that often comprise the majority of the diet, an artefact of the way in which preferences are calculated. A refinement of the species-level K regression model, to account for prey demographic- and cheetah social class-level differences in diet and the biomass of accessible prey (defined in this study as all non-avoided prey) instead of just preferred prey, doubled the predictive strength of the K model. Because group-hunting enabled predation on a broader weight range of prey, cheetah K was influenced by the ratio of male coalition cheetah to solitary cheetah in the population. The refined K regression model is derived from ecosystems supporting an intact carnivore guild. A mechanistic approach to estimating K, based on Caughley‟s (1977) maximum sustainable yield model, therefore better predicted cheetah K in systems devoid of lion Panthera leo and African wild dog Lycaon pictus, which were found to suppress cheetah density. This study improves our understanding of the relationships between prey demographics, cheetah social classes and intra-guild competition in determining cheetah prey preferences and K. This study therefore paves the way for similar work on other large carnivores.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:nmmu/vital:10754 |
| Date | January 2012 |
| Creators | Clements, Hayley Susan |
| Publisher | Nelson Mandela Metropolitan University, Faculty of Science |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis, Masters, MSc |
| Format | v, 129 leaves, pdf |
| Rights | Nelson Mandela Metropolitan University |
Page generated in 0.0023 seconds