Contribution of statistical models in flood susceptibility assessment in Essaouira Province-Morocco

Abstract

Floods are recurrent natural hazards in Morocco’s northern and central regions, causing considerable losses. This study developed a flood susceptibility map for the Essaouira provincial watersheds using three models: Analytic Hierarchy Process (AHP), Frequency Ratio (FR), and Weights-of-Evidence (WoE). Ten conditioning factors; slope, elevation, drainage density, rainfall, lithology, and topographic wetness index…, were integrated within a GIS framework. Model accuracy, evaluated using the area under the curve (AUC), exceeded 90% for all methods, confirming their reliability. Highly flood-prone zones were mainly identified near Essaouira city along the Ksob River and surrounding rural centers.