COS 76-7
Death, destruction and avoidance of adult trees by elephants in savanna woodland
In plant communities characterized by high productivity and high species turnover, such as many C4 savannahs, identifying the relative influence of biotic and abiotic controls is challenging because of the number and interactivity of different controlling agents. African savanna elephants (Loxodonta africana africana) have large, well-documented impacts on tree survival, growth and reproduction in areas where they occur in medium to high densities. However, savanna woodland communities are also strongly regulated by fire and rainfall, and these factors may modify the impact of elephants. In this study we followed the fates of over 650 individually-marked trees across a large rainfall gradient in the Serengeti Ecosystem in Tanzania from 2009 to 2013. Our goals were to: (1) characterize the magnitude of impact of elephants on adult trees, and (2) to determine whether damage and preference for certain trees is mediated by rainfall and experimental burning.
Results/Conclusions
Elephant browsing was highly selective towards certain species, after accounting for their availability in the landscape: A. robusta – the most dominant species in the ecosystem – was avoided while Acacia senegal and A. tortilis were highly preferred. Mortality of adult trees (>2meters in height) by elephants averaged 5.9% per year from 2009-2013 across all tree species, and a majority of the trees showed some form of damage from elephants. Trees in experimentally burned plots were damaged 10% less than those in unburned plots; this preference for unburned areas appeared to drive higher tree survival in these areas following damage by elephant. Despite clear trade-offs in tree growth rates and rainfall, trees in the driest (~500mm rainfall per year) and wettest (~1000mm per year) areas received less damage and died less frequently than those in areas with intermediate (~700mm per year) rainfall, a pattern consistent with elephant distribution. Given the fluctuations in elephant densities in many areas due to poaching, habitat restriction and (in the case of the Serengeti) population recovery, we believe it is important to understand the interactive impacts of elephants on woody cover as a basis for predicting changes in savannah structure and composition.