Flood Hazard Mapping in Ilam City using Evidence Belief Function Model

Floods are one of the most dangerous and catastrophic hydrological hazards in the world. It is a condition in which the flow of the river and the water level suddenly increase and causing huge losses of life and property. The purpose of this study is flood hazard mapping in Ilam city using evidence belief function model. Factors affecting flood occurrence considered in this study are slope, aspect, elevation, distance to river, river density, topographic wetness index (TWI), stream power index (SPI), curvature, lithology, soil science, land use and rainfall. After collecting the required information and layers, 126 flood points were identified in the study area. Flood locations were randomly divided into two groups of 70% (89 points) and 30% (37 points) for modeling and validation, respectively. Using the evidence belief function (EBF) model, a flood hazard mapping of Ilam city was prepared. The prepared flood susceptibility map was classified into 5 classes: very low, low, medium, high and very high. To evaluation the performance of the model, the area under the AUC diagram obtained from the ROC curve was used. According to the evaluation criterion used in this study (ROC) and according to the validation data, the EBF model (0.914) had a good performance in flood potential detection in the study area. Finally, according to the final flood hazard map, the eastern and northeastern regions of Ilam are in low risk. Also, the results of the study showed that flooding in Ilam city is due to environmental factors, including high slope and land use change. Based on the flood hazard map, appropriate management measures can be taken in the future to reduce flood damage and casualties.