Historical exploration of flash flood events and producing flash-flood susceptibility maps are crucial steps for decision makers in disaster management. In this article, classification and regression tree (CART) methodology and its ensemble models of random forest (RF), boosted regression trees (BRT) and extreme gradient boosting (XGBoost) were implemented to create a flash-flood susceptibility map of the Bâsca Chiojdului River Basin, one of the areas in Romania that is constantly exposed to flash floods. The torrential areas including 962 flash flood events were delineated from orthophotomaps and field observations. Furthermore, a set of conditioning forces to explain the flash floods was constructed which included aspect, land use and land cover (LULC), hydrological soil groups lithology, slope, topographic wetness index (TWI), topographic position index (TPI), profile curvature, convergence index and stream power index (SPI). All models indicated the slope as the most important factor triggering the flash flood occurrence. The highest area under the curve (AUC) was achieved by the RF model (AUC = 0.956), followed by the BRT model (AUC = 0.899), XGBoost model (AUC = 0.892) and CART model (AUC = 0.868), respectively. The results showed that the central part of the Bâsca Chiojdului river basin, which covers approximately 30% of the study area, is more susceptible to flash flooding.

Concerning the significant increase in the negative effects of flash-floods worldwide, the main goal of this research is to evaluate the power of the Analytical Hierarchy Process (AHP), fi (kNN), K-Star (KS) algorithms and their ensembles in flash-flood susceptibility mapping. To train the two stand-alone models and their ensembles, for the first stage, the areas affected in the past by torrential phenomena are identified using remote sensing techniques. Approximately 70% of these areas are used as a training data set along with 10 flash-flood predictors. It should be remarked that the remote sensing techniques play a crucial role in obtaining eight out of 10 flash-flood conditioning factors. The predictive capability of predictors is evaluated through the Information Gain Ratio (IGR) method. As expected, the slope angle results in the factor with the highest predictive capability. The application of the AHP model implies the construction of ten pair-wise comparison matrices for calculating the normalized weights of each flash-flood predictor. The computed weights are used as input data in kNN–AHP and KS–AHP ensemble models for calculating the Flash-Flood Potential Index (FFPI). The FFPI also is determined through kNN and KS stand-alone models. The performance of the models is evaluated using statistical metrics (i.e., sensitivity, specificity and accuracy) while the validation of the results is done by constructing the Receiver Operating Characteristics (ROC) Curve and Area Under Curve (AUC) values and by calculating the density of torrential pixels within FFPI classes. Overall, the best performance is obtained by the kNN–AHP ensemble model.