The use of erosion proxies for the spatial assessment of erosion in a watershed and modelling the risk in a GIS

This paper presents a new approach to soil erosion risk assessment and modelling. The approach recognizes that erosion risk is linked to biophysical landscpe features, societal activities and spatia-temporal attributes of the landscape processes. The assessment involves making measurements of soil erosion in the field and linking the erosion features to some selected landscape elements that act as either drivers or disruptors of erosion. These drivers or barriers of erosion are referred to in this paper as erosion proxies. Examples of the erosion proxies in a watershed include drainage ditches, field boundaries, footpaths, animal tracks and other man made elongate features that cause water concentration and flow in the watershed. These are referred to in this paper as erosion drivers. Barrier proxies or distruptors of erosion inclUde: hedges, closed fences, grassed field boundaries, trashed field boundaries, bunded field boundaries, barrier ditches, constructed dykes or built earthdams. Erosion risk is on the other hand defined as the potential for the occurrence of soil erosion due to the presence or absence of the proxies. The watershed could be at high or low risk depending on the intensity of occurence of the erosion proxies. The approach views the watershed more from the principles of energy and matter flows in them rather than by assessing the individual factors in a deterministic erosion model such as Wischmeir's (1978) Universal Soil Loss Equation (USLE). The net risk of erosion in the watershed, catchment, or sub-catchment is modelled by summing up the total length of the erosion drivers and the total length of the disruptors in a GIS database. The ratio retween the sum of the drivers, and the total sum of the drivers and the disruptors provides an indication of whether the catchment is at high or low risk. The method uses the spatial characteristics of the erosion proxies to extract them from the broader landscape mosaic through visual interpretation of aerial photographs or satellite images. The results of the interpretations are afterwards digitized into a geographic information system (GIS) database and processed to paper maps. The produced maps are then carried to the field to characterize and link the proxies to the presence or absence of soil erosion features. In the field, the occurrence of different soil erosion features on each erosion proxy is measured and recorded. The influence of the individual erosion proxy is obtained by using the analysis of variance (ANOVA) and the F test. The statistical analysis forms the basis for ranking the erosion proxies into different risk categories and for selecting the best mitigation options. The method is tested on an agricultural test area at the National Agricultural Research Laboratcries (NARL) in Nairobi. The results indicate that grassed field boundaries (an example of an erosion proxy) form good management practice for conserving the NARL sub-catchment. Other proxies for use against the risk of erosion include constructed earth and stone bunds along the field boundaries, trash and stover cover along the field boundaries or any other method that disrupts water flow in channels on the catchment or sub-catchment. Due to this method of assessment, field plots and large portions of the test area at the National Agricultural Research Laboratcries compound is now well conserved and suffers a minimum risk of soil erosion by water.