A Methodological Comparison of Three Models for Gully Erosion Susceptibility Mapping in the Rural Municipality of El Faid (Morocco)

Erosion is the main threat to sustainable water and soil management in Morocco. Located Souss-Massa watershed, rural municipality of El Faid remains an area where gully erosion a major factor involved degradation flooding. The aim this study predict spatial distribution at scale evaluate predictive capacity three prediction methods (frequency ratio (FR), logistic regression (LR), random forest (RF)) for characterization gullying vulnerability. Twelve predisposing factors underlying formation were considered mapped (elevation, slope, aspect, plane curvature, slope length (SL), stream power index (SPI), composite topographic (CTI), land use, wetness (TWI), normalized difference vegetation (NDVI), lithology, cover (C factor). Furthermore, 894 gullies digitized using high-resolution imagery. Seventy-five percent randomly selected used as training dataset, whereas remaining 25% validation purposes. accuracy was evaluated under curve (AUC). Results showed that most contributed prevalence (slope, CTI, LS). fitted models revealed RF model had better quality, with best AUC (91.49%). produced maps represent valuable tool management, conservation, protecting human lives against natural hazards (floods).