The role of topography in the emergence of African savannas

Abstract

The coexistence of trees and grasses in savannas are not well understood even though savannas occupy a wide area of West Africa. In this study, a hypothesis is proposed to investigate the question of how trees and grasses coexist in a region. The hypothesis suggests that the variation in elevation leads to the variation in soil moisture, which in turn can explain the coexistence of trees and grasses in savannas. To test this hypothesis, experimental simulations are performed using biospheric model, IBIS, and distributed hydrologic model, SHE. We, first, estimate the amount of rainfall required for trees and grasses under a certain atmospheric condition. Here, the variation of rainfall is prescribed to force a similar variation of soil moisture. A 30% decrease in rainfall is sufficient to simulate grasses at 9°N. A 100% increase in rainfall is sufficient to simulate trees at II°N. However, even with a five fold increase in rainfall, the model fails to simulate trees at 13°N. To study the influences of topography explicitly, a distributed hydrologic modeling is performed using SHE. The results suggest that the variation of the depth to water table induced by the varying elevation is highly correlated with the variation of soil moisture. Consequently, an asynchronous coupling of SHE and IBIS is designed to investigate the stated hypothesis. The coupling is performed by modifying IBIS to include the groundwater table as a boundary variable. The modified IBIS simulates both trees and grasses according to a different water table boundary condition in natural savannas of 11°N. The shallow water table of valleys allows the growth of trees, and the deep water table of hills allows the growth of grasses. The simulations in this study suggest that the variability of soil moisture resulting from the topographic variation can be a determinant of savanna ecosystems. Moreover, grasslands in 13°N cannot be changed into forests only by adjusting soil moisture. It suggests that the role of soil moisture can be significant to dictate the vegetation type only in a certain window characteristic of savanna climate.