Long-term interactive effects of elephants, fire, and rainfall determine woody plant biomass and carbon storage in African savanna
Disturbance and resource availability are hypothesized to be key determinants of the distribution and functioning of tropical savannas, but their interactive effects are both complex and poorly known. In African savannas, fire and herbivory, especially by elephants, are key disturbance agents, where frequent fires and intense browsing reduce tree biomass and potentially carbon storage. However, the effect of these disturbances likely depends upon resource availability and the ability of vegetation to recover following disturbance.
We examined the effects of fire and elephants on woody biomass in Kruger National Park, South Africa, where experimental plots have been maintained under different fire regimes for ~60 years and span a broad rainfall gradient. A key feature of our study design is that this period encompassed a large increase in elephant abundance caused by the mid-1990s suspension of elephant culling.
Before the cessation of culling, woody biomass diverged across fire treatments, with biomass being inversely related to fire frequency. Following the increase in the elephant population, however, the effects of fire were muted to the extent that plots experiencing fire vs. protected from fire converged on similar levels of woody biomass. Furthermore, the effect of fire alone changed across the rainfall gradient, with tree biomass being more sensitive to fire in dry sites but accumulating in the wettest sites even at annual or triennial burn intervals. Contrastingly, increased elephant populations resulted in decreased woody biomass, regardless of rainfall across sites. These results suggest that while elephants and fire can affect tree biomass in savannas, variation in environmental conditions may fundamentally alter both their individual and interactive effects. Moreover, the substantial and rapid changes in woody biomass carbon stocks following the rise in elephant populations suggest that elephants are likely to play a dominant role in the carbon cycle of African savannas.