Soil erosion hazard mapping using Analytic Hierarchy Process and logistic regression: a case study of Haffouz watershed, central Tunisia

The purpose of this study is to develop an overall methodology for assessing soil erosion hazard. Two models are implemented in Haffouz watershed (central Tunisia), which is generally prone to erosion processes. The first is a multi-parametric approach based on expert’s knowledge and Analytic Hierarchy Process (AHP), while the second is a statistical technique based on logistic regression (LR). Six parameters (slope gradient, annual precipitation, lithofacies, vegetation index, drainage density, and land use) are used for both methodologies. A thematic layer map is generated for each parameter using geographic information system (GIS) and remote sensing, while the weighting coefficients of each methodology are used for the compilation of AHP and LR final maps. Both models give similar results depicting that, in general, a moderate hazard condition of soil erosion was found in the study area. The proposed approaches also resulted in the identification of the highly eroded areas, requiring immediate action.