Application of the USLE in a Savannah Environment: Comparative Experiences from East and West Africa

Abstract

The use of empirical models for predicting erosion hazard in Africa is widespread. One of the most commonly used models is the Universal Soil Loss Equation (USLE). This paper explores the applicability of the USLE in the East and West African Savannah environments, using both primary and secondary data. The results show that the USLE can be applied to Savannah conditions if some level of instrumentation is provided to enable model calibration and validation. The rainfall erosivity factor (R) is one variable which is calculated differently in both the East and the West African Savannah, reflecting differences in the storm characteristics between the two regions. The soil erodibility factor (K) is applied to cultivated soils with relatively high organic matter but does not adequately predict soil loss if the soils have large quantities of stone cover or if they exhibit surface sealing properties. The topographic factors (LS) are transferable, but in some parts of the Savannah, very steep slopes are cultivated, limiting the accuracy of the LS factor. The crop and management factor (C) is applicable for single stand crops, but more studies are required to develop C factors for multiple cropping and rangelands as these are more prevalent in the Savannah. The conservation practices factors (P) as described in the USLE routines do not adequately cater for soil conservation structures found in Africa, such as trash lines and stone lines. Thus, local indices need to be developed. In general, the USLE tends to underestimate soil erosion in East Africa while overestimating it in the West African Savannah. User-friendly Geographical Information Systems (GIS) applications that depict the uncertainty of model estimates should be explored in the future.

Interdisciplinary on-site evaluation of stone bunds to control soil erosion on cropland in Northern Ethiopia

“…the support practice factor indicating reduced soil erosion potential due to farming practices and conservation measures) for stone bunds was estimated at 0.32 (Desta et al, 2005). Mati and Veihe (2001), while stating that ''experimentally derived P factor values in East and West Africa are few, partly due to the difficulties associated with setting up erosion plots that contain soil conservation measures'', present a P value of 0.18 for fanya juu terraces, stone lines and contour bunds. In Ecuador, stone bunds and grass strips established on steep runoff plots with loosened volcanic soils nearly eliminated runoff and erosion (De Noni et al, 2000) which is equivalent to a P-value of 0.…”

Mapping soil erosion risk in Rondônia, Brazilian Amazonia: using RUSLE, remote sensing and GIS

“…It has been extensively used to estimate soil erosion loss, to assess soil erosion risk, and to guide development and conservation plans in order to control erosion under different land-cover conditions, such as croplands, rangelands, and disturbed forest lands (Millward and Mersey, 1999;Boggs et al, 2001;Mati and Veihe, 2001;Angima et al, 2003). The RUSLE is expressed as:…”

Dynamic Assessment of Soil Erosion Risk Using Landsat TM and HJ Satellite Data in Danjiangkou Reservoir Area, China

“…Concerning raindrop impact, the kinetic energy with which raindrops impact the soil surface and provoke detachment of soil particles depends on their velocity and size [61]. However, the RUSLE is a model for predicting erosion which estimates long-term average annual soil erosion amounts [62][63][64]. It is applied to slope land, needs more accurate input data and has little consideration for the seasonal variation of vegetation.…”