The flood hazard risk zonation is of great importance to the control and prevention of flood disaster. Taking the catchment area along Han River in Hubei Province as an example， based on the flood hazard risk evaluation index system， ArcGIS was used to normalize each evaluation index. The grid of the research area connected with the historical flood disaster point was accordingly established in order to obtain the statistics of each evaluation index and the corresponding disaster occurrences in each grid. Afterwards， according to the binary Logistic regression principle， SPSS was used for binary logistic regression analysis， so as to obtain the correlation between each evaluation index and the occurrence of flood disaster. On the basis of the above， the risk probability calculation method of Logistic regression model was applied to draw the flood hazard zoning map of the study area in ArcGIS. The results showed that flood hazard risk in the study area coexisted with the two aggregation modes of high and low value. And flood hazard risk was divided into five levels： The highest， higher， moderate， lower and the lowest regions. The area proportion of each risk region was 8.3%， 12.5%， 20.6%， 19.2% and 39.4%， respectively.

Keywords： Flood hazard ; Susceptibility mapping ; Han River ; GIS ; Logistic regression model