Abstract: Flood risk can, in general terms, be defined as probability time consequence. It consists of flood hazard analysis, vulnerability analysis and damage evaluation. A variety of methods have been developed and applied. Among them, Quantitative Risk Analysis (QRA) is a method of quantifying risk through systematic examination of the factors contributing to the flood hazard and affecting the severity of flood consequence, their interaction and relative contribution to the occurrence of the flood. The QRA technique is well established in many fields such as chemical engineering and hazardous materials processing. The application of QRA to flood risk is relatively new and still under development. While the basic risk assessment concepts and tools can be used, the methodologies need to be adapted. Category-based model for flood risk analysis is used to assign a value to each driven-factor such as triggering factor of rainfall, dam break, ground surface conditions of topography, land cover, and others. The keys to the model are to synthesis the spatial-referenced data and create the risk zone. The diffusion of Geographical Information Systems (GIS) technology opens up a range of new possibilities for hazard mitigation and disaster management. Microzonation is greatly facilitated by the kind of automation that GIS offers, especially as it involves comparison, indices and overlays in much the same way that GIS does. In this article, ArcInfo GIS has been chosen to quantitatively represent the influencing factors, spatialize the data into the uniform grid system, and transfer all the data item into the effect degrees on the probability of flooding. At last, with the support of Arc/Info GRID model, a categorical model for flood risk zonation has been put forward. The approach has been applied to the Liaohe river basin, the north-eastern of China, flood disaster risk zonation. The results show that the flood risk of the lower reaches of the Liaohe river is more serious than other places, which accord with the fact. The case study showes that the GIS-based category model is effective in flood risk zonation.

Key words: Geographical Information System, Flood hazard, Risk zonation