Assessment of flood hazard mapping in urban areas using entropy weighting method: a case study in Hamadan city, Iran

Abstract

Flood is one of the major natural disasters which cause enormous casualties and damages particularly in urban areas. In urban areas, studies on flood hazards have been accompanied by tensions for various reasons, including complexity of urban levels, different spatial modeling indices, lack of accurate hydrological data, and precise modeling of land surface simulations. This paper used a Constrained Delaunay Triangular Irregular Network to model fine urban surfaces (based on the detailed ground sampling data), and subsequently discusses five indicators regarding the dangers of flood, namely (1) elevation, (2) slope, (3) distance to discharge channels, (4) index of development and persistence of the drainage network (IDPR), and (5) infiltration rate. In the next step for flood hazard mapping, the combination of geographical information systems and the entropy weight method as the multi-criteria decision analysis was used to combine the indicators. The proposed methodology was used for Hamadan city that is located in the central part of Hamadan Province in Iran where several floods occur annually. The flood hazard mapping indicates that approximately 15.83% of the total study area is classified as very highly hazardous, 31.72% as hazardous, 20.11% as moderate, 16.02% as minor, and 16.32% as the least hazardous. Finally, superimposition and receiver operating characteristic (ROC) curve methods were used to verify the accuracy of the obtained flood hazard map. In terms of superimposition and ROC curve, the accuracy of the model was approximately 70% and 73%, respectively.