Flood occurs as a result of high intensity rainfall, long-term rainfalls and snowmelt which flows out of the main river channel onto the floodplain areas and damages to buildings, roads, and facilities and causing life losses. This study aims to implement Extreme gradient boosting method for the first time in flood susceptibility modelling and compare its performance with three advanced benchmark models including Frequency Ratio (FR), Random Forest (RF), and Generalized Additive Model (GAM). The input factors include altitude, slope, plan curvature, profile curvature, stream power index (SPI), topographic wetness index (TWI), distance from rivers, normalized difference vegetation index (NDVI), rainfall, land use, and lithology. For running the models, 243 flood locations were detected by field surveys and national reports. The same number of locations were randomly created in the study regions and considered as nonflood locations. Both flood and non-flood locations were fed into the models and output flood susceptibility maps were produced. To evaluate the efficacy of the algorithms, receiver operating characteristics (ROC) curve were implemented. The results of the current research showed that the RF model and EGB had the best performances with the area under ROC curve (AUC) of 0.985, and 0.980, followed by the GAM and FR model with AUC values of 0.97, and 0.953, respectively.
2The results of factor importance by the RF model showed that distance from rivers had an important influence on flood susceptibility mapping (FSM), followed by profile curvature, slope, TWI, and altitude. Considering the high performances of the RF and EGB models in flood susceptibility modelling, application of these models is recommended for such studies.