Functional correlates for elephant spatial use in Southern African savannas and some implications for management

Abstract

The scientific evaluation of conservation management options for African elephants is shifting focus. Previous evaluation of methods to manage elephants within stable savannas is giving way to scientific development of management frameworks that reinstate ‘ecologically effective’ densities of elephants as interactive components of dynamic savanna ecosystems. Megaparks across the distributional range of elephants are proposed as a nonequilibrium management option to establish regionally stable elephant metapopulations while ameliorating local impacts. This approach relies on assumptions regarding elephant spatial use under non-equilibrium management conditions. It furthermore assumes that elephant populations fluctuate in response to intrinsic- (density-related) and extrinsic- (resource-related) factors that also enable emigration and immigration over broad spatial scales. For metapopulation dynamics in particular, it assumes spatial discontinuity in demographic responses of elephant sub-populations to landscape heterogeneity. I studied elephant spatial use across southern Africa’s west to east resource gradient, and within Kruger National Park (Kruger), to examine some of these assumptions. I used rainfall and vegetation productivity data, satellite collar locations and spatially explicit aerial survey data, to test whether elephant spatial use responded to the heterogeneous distribution of resources. I then tested whether elephant densities modified the influence of these factors on their distributions and movements. Thereafter, I explored whether the influence of density on movements provided a spatially driven mechanism for density dependent survival rates of subadult elephants. Elephant spatial use varied in relation to heterogeneity in the spatial and temporal distribution of food availability. Across six conservation areas spanning southern Africa, family group home ranges were explained by food availability during the wet season, and heterogeneity in the distribution of food availability during the dry season. Inter-seasonal spatial use was explained by inter-seasonal stability in food availability. In Kruger, elephant distributions were explained by the distribution of food availability within landscapes. Elephant densities modified these relationships. In Kruger, dry season distributions became more uniform within landscapes with increasing density, and across 13 conservation areas spanning southern Africa, dry season daily displacement distances of family groups increased with density. The survival of weaned calves decreased with increasing dry season daily displacement distances. This implies a spatially driven mechanism for density dependent population limitation. Collectively these findings supported assumptions that the response of elephant spatial use to intrinsic- and extrinsic- factors can drive fluctuations in population growth rates over broad spatial scales. The management actions of fencing and supplementary water negated the relationship between daily displacement distances and weaned calf survival. Practical application of effective conservation management, however, lags behind scientific understanding. Therefore, I interviewed elephant managers in South Africa to determine their reliance on science for decisions regarding elephants and maintenance of biodiversity. Managers did not base decisions on science, instead relying on their own experience, observations and “gut feelings”. Nevertheless, they valued science and identified several practical opportunities for scientists to secure better uptake of their work into management decisions. This dissertation thus contributes scientific support and identifies practical opportunities with which the science community can assist managers to move towards managing elephants as interactive components of dynamic savanna ecosystems.