Response of Acacia tortilis to utilization by elephants in a semi-arid African savanna

The impact of elephant feeding on individual growth and population trends of Acacia tortilis and Acacia nilotica was studied in the semi-arid Venetia-Limpopo Nature Reserve between 1996 and 2000, comparing Acacia Woodland and Riparian Woodland. Monitoring of permanent transects revealed that elephants reduced Acacia tortilis density of Acacia Woodland from 173 to 68 stems/ha between 1996 and 2000, but population size structure remained unchanged because height selection varied. Annual mortality of Acacia tortilis ranged from 9-37%, lagging utilization by elephants by 2-3 years, and decreased once tree density had declined. Elephants ceased to use Acacia Woodland once density had declined to that corresponding with the less used A. tortilis in Riparian Woodland. Elephants utilized A. tortilis by removing canopy branches, pollarding, uprooting or debarking stems, which depended on tree size and previous use. Pollarded or uprooted stems resprouted poorly and usually died. Survival following canopy removal depended on the severity of defoliation. Debarking was not observed to kill A. tortilis because debarked stems were subsequently uprooted or pollarded. By contrast, debarking killed A. nilotica trees or substantially reduced their size. Prediction of population trends further requires consideration of rainfall, other herbivores, and fire. The vulnerability of Acacia populations to utilization by elephants is increased because the food staple, Colophospermum mopane, is tolerant of severe utilization by elephants.