A comparative assessment of decision trees algorithms for flash flood susceptibility modeling at Haraz watershed, northern Iran.

An ensemble prediction of flood susceptibility using multivariate discriminant analysis, classification and regression trees, and support vector machines

Environmental hazards: Assessing risk and reducing disaster

A new artificial intelligence (AI) model, called Bagging-LMT - a combination of bagging ensemble and Logistic Model Tree (LMT) - is introduced for mapping flood susceptibility. A spatial database was generated for the Haraz watershed, northern Iran, that included a flood inventory map and eleven flood conditioning factors based on the Information Gain Ratio (IGR). The model was evaluated using precision, sensitivity, specificity, accuracy, Root Mean Square Error, Mean Absolute Error, Kappa and area under the receiver operating characteristic curve criteria. The model was also compared with four state-of-the-art benchmark soft computing models, including LMT, logistic regression, Bayesian logistic regression, and random forest. Results revealed that the proposed model outperformed all these models and indicate that the proposed model can be used for sustainable management of flood-prone areas.

A novel hybrid artificial intelligence approach for flood susceptibility assessment

A comparative assessment of flood susceptibility modeling using Multi-Criteria Decision-Making Analysis and Machine Learning Methods

Flood is one of the most prevalent natural disasters that frequently occur in the northern part of Iran reported in hot spots of flood occurrences The main aim of the current study was to prepare flood susceptibility maps using four models, namely frequency ratio (FR), weights-of-evidence (WofE), analytical hierarchy process (AHP), and ensemble of frequency ratio with AHP (FR-AHP), and to compare them at Haraz Watershed in Mazandaran Province, Iran A total of 211 flood locations were prepared in GIS environment, of which 151 locations were randomly selected for modeling and the remaining 60 locations were used for validation aims In the next step, 10 flood-conditioning factors were prepared including slope angle, plan curvature, elevation, topographic wetness index, stream power index, rainfall, distance from river, geology, landuse, and normalized difference vegetation index The receiver operating characteristic curve and the area under the curve (AUC) were created for different flood susceptibility maps Validation of results showed that AUC values for success rate in training data set, for FR, WofE, AHP, and FR-AHP, were 9707, 9896, 9591, and 8619 % with prediction rates of 09657 (9657 %), 09596 (9596 %), 09492 (9492 %), and 08469 (8469 %), respectively Moreover, the results showed that the frequency ratio model had the highest AUC in comparison with other models Generally, the four models show a reasonable accuracy in flood-susceptible areas The results of this study can be useful for managers, researchers, and planners to manage the susceptible areas to flood and reduce damages

A GIS-based flood susceptibility assessment and its mapping in Iran: a comparison between frequency ratio and weights-of-evidence bivariate statistical models with multi-criteria decision-making technique

Edris Maroufinia

Papers

A GIS-based flood susceptibility assessment and its mapping in Iran: a comparison between frequency ratio and weights-of-evidence bivariate statistical models with multi-criteria decision-making technique