An overview of soil erosion modelling

Soil erosion is one of the most important environmental issues in natural and synthetic territories. It can lead to loss of fertility, slope instability, soil truncation; etc. which causes irreversible effects on the poorly renewable soil resource. Therefore, understanding the key parameters and factors to model soil erosion will enable the conservation of soil system goods, services and resources, and will avoid the damage outside of fields caused by transported and accumulated sediments and water. In view of this, a review was carried out on previous studies to examine the concept of soil erosion and review various soil models widely used in literature. It was found that several models are used for soil assessment and prediction and these models are classified into physical (e.g. WEPP), conceptual (SEDNET) and empirical (USLE). The Universal Soil Loss Equation (USLE) and its modifications were found to be the most commonly used soil erosion models due to its simplicity, ease of use and the ability to integrate the various ecosystem parameters successfully. Furthermore, it was found that one of the major limitations associated with the use of models is lack of data for validation especially in large areas where obtaining ground data is not feasible. Although some researchers have suggested the use of correlation between modeled erosion results with factors such as land cover and management factor and soil erodibility factor as well as results of land use change analyses as alternatives for validation purpose. Others correlate the predicted soil erosion results with that of sediment yield. Some validated their soil erosion data with results of land use change analyses, slope length and slope steepness factor, land cover and management factor and soil erodibility factor. However, there is still ambiguity in the knowledge of our understanding as to which soil erosion prediction model to use. Keywords: soil erosion, model, validation, USLE, RUSL