Department of Biology, Norwegian University of Science and Technology, 7491 Trondheim, NORWAY.
This thesis examines the interaction between woody plants, elephants and other browsers in a semi-arid savanna in northern Botswana, southern Africa. Particularly I studied how woody plants respond to herbivory, how browsers respond to previously browsed trees, and how the browsers share food resources.
Browsing pressure by twig biting ungulates showed a unimodal relationship with shoot vigour (first PCA axis scores generated from four shoot variables) of 14 tree species, which may be caused by low quantity and quality of food on trees with low vigour, and too large shoot diameters on the most vigorously growing tree species. Browsing pressure by elephants showed no relation with shoot vigour of plant species. Elephants, giraffe, impala and kudu largely used different food resources in terms of browse species and height levels selected. There was little resource-use overlap (Schoener’s index) between the herbivore species. Elephants (ca. 3000 kg) predominantly browsed other woody species than those browsed by giraffe (ca. 1000 kg), impala (ca. 50 kg) and kudu (ca. 200 kg). Differences in body size could not explain this difference in food choice, as giraffe, impala and kudu browse the same tree species independent of considerable difference in body size. The differences in food selectivity may instead be explained by difference in digestive systems. Elephants are hindgut fermenters whereas the others are foregut fermenters (ruminants).
Trees that were strongly affected by elephant browsing had more shoots at low height levels than individuals without elephant impact. Impala and kudu preferred to browse from trees previously browsed by elephant rather than from trees without any elephant impact. This indicates that elephants facilitate the foraging by these two species. Elephants themselves are known to rebrowse previously impacted trees, and were found to recognize and preferentially browse trees that had been subject to simulated browsing three years earlier and since then protected from large herbivores.
The study did not find evidence that elephants compete with and deplete food for other browsers. Results in this thesis do not justify manipulation of elephant population and distribution as a means to increase populations of the other animal species investigated in this thesis. Instead it seems that elephants may act to facilitate browsing by other investigated animal species.