The ecology of large herbivores native to the coastal lowlands of the fynbos biome in the Western Cape, South Africa
The south-western Cape is a unique region of southern Africa with regards to generally low soil nutrient status, winter rainfall and unusually species-rich temperate vegetation. This region supported a diverse large herbivore (> 20 kg) assemblage at the time of permanent European settlement (1652). The lowlands to the west and east of the Kogelberg supported populations of African elephant, black rhino, hippopotamus, eland, Cape mountain and plain zebra, ostrich, red hartebeest, and grey rhebuck. The eastern lowlands also supported three additional ruminant grazer species - the African buffalo, bontebok, and blue antelope. The fate of these herbivores changed rapidly after European settlement. Today the few remaining species are restricted to a few reserves scattered across the lowlands. This is, however, changing with a rapid growth in the wildlife industry that is accompanied by the reintroduction of wild animals into endangered and fragmented lowland areas. These reintroductions, together with the realisation that we have limited knowledge of the functional role of native large herbivores in the fynbos ecosystem, provided the rationale for this study. Questions on large herbivore ecology were addressed at three different spatial scales. At the biome level, the reason for the absence of three ruminant grazers from the western lowlands was investigated. It was hypothesised that the absence of adequate high quality fodder in the form of C4-grass during the hot and dry summers made it impossible for buffalo, blue antelope, and bontebok to survive on the western lowlands. The results from carbon isotope analysis of late prehistoric, historic and contemporary large herbivore remains were consistent to this Summer Nutritional Stress Hypothesis. I found that eland, elephant, grey rhebuck, ostrich, and red hartebeest (all species that historically occurred in both coastal lowlands) can survive with very little (< 15%) C4 grass in their diet. In contrast, bontebok utilized at least 43% C4 grass biomass in what was considered their natural habitats. At a regional level, I tested the hypothesis that the large herbivores avoid nutrient-poor sandstone, sand, and limestone fynbos shrublands in favour of the more nutrient-rich shale renosterveld habitats. Support for this Renosterveld Preference Hypothesis was found by means of dung count surveys, which showed that both eland and bontebok readily utilize renosterveld, but avoid sandstone and limestone fynbos. In the latter they only utilize grassy microhabitats such as karstic sinkhole depressions. The same hypothesis was addressed in a novel way by using strontium isotope analysis and concluded that the technique needs more refinement for it to produce reliable results. At a landscape level, interactions between fire and grazing by native large herbivores in relation to renosterveld vegetation dynamics were addressed. I conclude that the disappearance of the native herbivores probably had little bearing on the putative structural changes in renosterveld (grassland-shrubland dynamics). Support was found for the notion that a high fire frequency followed by intense grazing by livestock could have converted original renosterveld grasslands to unpalatable shrublands. Herbivory by native grazers/browsers, or the release from it, cannot by itself bring about the vegetation-state (structural) changes in renosterveld patches which had already been altered to herbivoretolerant plant communities. However, in combination with fire, the presence or absence of large herbivores can change the trajectory of the system among the alternative structural states.