The impact of elephants on the marula tree Sclerocarya birrea

This study determined the abundance, density and population structure of the marula tree, Sclerocarya birrea, in three game reserves in South Africa, and assessed patters and amounts of new and cumulative impact of elephants. Elephant feeding was very patchy so several attributes of individual trees, sampled transects and communities that might influence elephant herbivory were investigated. The incidence and type of elephant impact (bark, branch or stem breakage) were significantly related to tree diameter, but not to fruiting nor proximity to roads. At the transect level, elephant impact was influenced by density of marula trees, but was not influenced by proximity to roads, or proportion of marula trees bearing fruits in the vicinity. At the community level, elephant impact was higher on reserves with higher total marula densities. Fourfold differences in elephant densities (0.08-0.30 elephants km(-2)) did not explain marula consumption: the percentage of trees with branch damage was similar across reserves and bark damage was inversely proportional to elephant density. Variation across reserves may reflect local and landscape-level marula tree abundance, differences in alternative food plants and individual feeding habits. The recorded levels of impact appeared to be sustain able because mortality rates were low, affected trees often recovered, and small trees were not preferentially preyed upon.