Mass Movement Hazard Assessment at a Medium Scale Using Weight of Evidence Model and Neo-predictive Variables Creation

Abstract

In this paper, mass Movements hazard (MMH) was evaluated by using weights of evidence model (WOE). The Bab region (Central Rif, Morocco) is known as one of the most areas prone to mass movements (MM). High resolution remote sensing data interpretation and a systematic field survey were performed in order to obtain an inventory of MM. A total of 247 MM of the following types: landslides (45%), rock-falls (48%) and debris flows (7%) were identified, which covered an area of 20 \({\text {km}}^{2}\) and accounting for 3% of the study area (630 \({\text {km}}^{2}\)). Then, nine conditioning parameters characterizing topographical, geological, and environmental conditions were mapped, hierarchized and integrated as the main parameters, controlling the occurrence of MM. To eliminate the causal dependency a strategy based on the establishment of three neo-predictive variables (NPV) with a geomorphological significance from the parameters responsible for the conditional independency violation was implemented. Several simulations with several combinations of parameters were performed in order to examine the influence of each variable on the predictive power of MMH. The prediction capability of each model is determined by the area under the curve value and the best combination model was selected to generate the MMH map.