Pastoralism and Kalahari Rangeland Soils

This chapter presents field-based evidence of the changes in soil properties resulting from grazing-related disturbance and shrub encroachment in the south-west Kalahari. The objectives were: (i) to determine the impact of different grazing intensities on soil CO2 efflux, soil organic carbon (SOC) and chlorophyll a (a proxy for photosynthetic biomass); and (ii) to quantify differences in microclimate, SOC and soil CO2 efflux between open-grass sites, and the shrub/tree canopy. Experiments were conducted in an open fine-leaf savannah on Kalahari Sand soils in south-west Botswana. Vegetation cover is a mix of perennial (Eragrostis sp.; Poaceae) and annual (Schmidtiaspp.; Poaceae) grasses, woody shrubs (Grewia flava; Tiliaceae and Acacia mellifera; Leguminosae) and trees (predominantly A. erioloba; Leguminosae). Pastoral farming on unfenced communal areas is the principal livelihood activity across most of the Kalahari. The grazing land available to farmers without tenure is, however, shrinking as more is sold to private owners. This increases livestock stocking densities and communal grazing lands in Botswana typically exceed recommended carrying capacities. This leads to detrimental changes in soil properties, at least in the short term, and longer-term changes in the balance between shrubs and grasses. The research has shown that intensive grazing damages the biologically crusted sand surface resulting in a reduction in the amount of C fixed by autotrophic organisms. Light disturbance of the soil, however, does not appear to have an adverse effect on the ability of the organisms to photosynthesize and add C to the soil. Thus, managed grazing, where soils are only lightly disturbed, could help maintain a positive C balance in African drylands. Consequently, it is vital that land management decisions take a holistic view that includes consideration of how land use changes may affect the behaviour of human populations that rely directly on the managed land for their livelihoods.