Modelling land degradation in the Nepalese Himalaya

Land degradation is a crucial issue in mountainous areas and is manifested in a variety of processes. For its assessment, application of existing models is not straightforward. In addition, data availability might be a problem. In this paper, a procedure for land degradation assessment is described, which follows a four-step approach: (1) detection, inventory and mapping of land degradation features, (2) assessing the magnitude of soil loss, (3) study of causal factors, and (4) hazard assessment by applying decision trees. This approach is applied to a case study in the Middle Mountain region of Nepal. The study shows that individual mass movement features such as debris slides and slumps can be easily mapped by photo interpretation techniques. Application of soil loss estimation models helps get insight on the magnitude of soil losses. In the study area, soil losses are higher in rainfed crops on sloping terraces (highest soil loss is 32 tons/ha/year) and minimal under dense forest and in irrigated rice fields (less than 1 ton/ha/year). However, there is high frequency of slope failures in the form of slumps in the rice fields. Debris slides are more common on south-facing slopes under rainfed agriculture or in degraded forest. Field evidences and analysis of causal factors for land degradation helps in building decision trees, the use of which for modelling land degradation has the advantage that attributes can be ranked and tested according to their importance. In addition, decision trees are simple to construct, easy to implement and very flexible in adaptations. (C) 2003 Elsevier B.V. All rights reserved.