Landslide susceptibility mapping for western coastal districts of India using geospatial techniques and eXplainable Artificial Intelligence

Landslide susceptibility mapping (LSM) assists in identifying probable zones for future landslide occurrences within a given location by considering various landslide-triggering factors. Most significantly, this mapping contributes to regional planning and the landslide mitigation procedure and raises public awareness and education on landslides. In the current study, LSM was conducted for western coastal districts of India using fourteen landslide triggering factors. For locating landslide-susceptible areas and to identify the best preforming model, a comparison between frequency ratio (FR), logistic regression (LR), machine learning (ML) and artificial intelligence models was performed. ML models used in this study were random forest (RF), support vector machine (SVM), extreme gradient boosting (XGB) and deep neural network (DNN). Most of the area was covered by very low class, i.e., 60.12% followed by low (13.50%), moderate (10.54%), high (8.04%) and very high (7.79%) classes, respectively. From the variable importance plots, it was found that factors such as slope, TRI, LS-factor, distance to road and rainfall were the most significant landslide-triggering factors. The area under the ROC curve (AUC) was utilised to validate the models. The results of the AUC revealed that the RF model showed an excellent accuracy rate of 0.993, followed by XGB (0.992), SVM (0.955), DNN (0.949), LR (0.919), and FR (0.906) model. The ranking based on multiple model evaluation parameters using validation dataset revealed DNN as the best-performing model. It was concluded that the performance of ML models was excellent compared to the FR model. The results of this study could help to identify landslide-vulnerable areas and adopt suitable preventive measures for mitigating the likely occurrence of future landslide events.