Landslide susceptibility mapping by remote sensing and GIS using frequency ratio, value of information and logistic regression methods in the N'fis watershed, High Atlas of Marrakech (Morocco)

The primary aim of this study was to assess landslide susceptibility in the N'fis watershed by leveraging Geographic Information System (GIS) and remote sensing data. Three statistical models, namely frequency ratio (FR), information value (IV) and logistic regression (LR), were applied and compared. The process began with mapping an inventory of 156 landslides where 70% of the data served to train the models and the remaining 30% was used for validation. The second step involved thematic mapping of 14 causative factors. Subsequently, landslide susceptibility was evaluated using the three methods and the resulting maps were classified into five risk levels. The accuracy of the maps produced by each model was then assessed using the area under curve (AUC) method. The findings revealed that the LR model exhibited the highest accuracy (89.71%) followed closely by the FR model (85.31%) while the IV model demonstrated the least accuracy (72.58%).