Detection of Susceptible Areas to Flooding and its Most Important Contributing Factors Using the Maximum Entropy Model in the Tashan Watershed, Khuzestan

One of the most important measures for flood management is preparation of a flood susceptibility map. The purpose of this study was to identify susceptible areas to flooding using the maximum entropy model in the Tashan Watershed, Khuzestan Province. A flood inventory map was prepared for the statistical analysis. Of the 169 flooding occurrences, 70% were used for the model calibration and 30% were used for validation. Ten flooding factors, namely: altitude, slope aspect, distance from river, drainage density, slope angle, land use, distance from road, topographic wetness index, plan curvature and lithology were used. The effect and contribution of each environmental parameter was calculated using the response curves and the Jackknife method. Finally, a flooding susceptibility map was prepared in four classes. The receiver operating characteristic curve was used to evaluate the accuracy of modeling. The results indicated that the maximum entropy model had a very good accuracy (AUC=0.885) in identifying the prone areas to flooding, and the land use and distance from road with 50.5% and 20.6% contribution, respectively, had the most impact on the occurrence of flood inundation. Also, about 25% of the watershed area was highly sensitive to flooding. Regarding to very good prediction accuracy of maximum entropy model in detecting susceptible areas to flooding, it is recommended to apply this model for the preparation of the flooding map in other watersheds, especially areas lucking hydrometric stations.