Thomas, Andrew D. and Dougill, Andrew J. (2007) Spatial and temporal distribution of cyanobacterial soil crusts in the Kalahari: implications for soil surface properties. Geomorphology, 85 (1-2). pp. 17-29. ISSN 0169-555X
File not available for download.Abstract
Localised patterns of erosion and deposition in vegetated semi-arid rangelands have been shown to influence ecological change and biogeochemical cycles. In the flat, vegetated Kalahari rangelands of Southern Africa the factors regulating erodibility of the fine sand soils and the erosivity of wind regimes require further investigation. This paper reports on the spatial and temporal patterns of cyanobacterial soil crust cover from ten sites at five sampling locations in the semi-arid Kalahari and discusses the likely impact on factors regulating surface erodibility and erosivity. Cyanobacterial soil crust cover on Kalahari Sand varied between 11% and 95% of the ground surface and was higher than previously reported. Cover was inversely related to grazing with the lowest crust cover found close to boreholes and the highest in the Game Reserve and Wildlife Management Zone. In grazed areas, crusts form under the protective canopies of the thorny shrub Acacia mellifera. Fenced plot data showed that crusts recover quickly from disturbance, with a near complete surface crust cover forming within 15 months of disturbance. Crust development is restricted by burial by wind blown sediment and by raindrop impact. Crusts had significantly greater organic matter and total nitrogen compared to unconsolidated surfaces. Crusts also significantly increased the compressive strength of the surface (and thus decreased erodibility) and changed the surface roughness. Establishing exactly how these changes affect aeolian erosion requires further process-based studies. The proportion of shear velocity acting on the surface in this complex mixed bush–grass–crust environment will be the key to understanding how crusts affect erodibility.
Impact and Reach
Statistics
Additional statistics for this dataset are available via IRStats2.