Study On SDSS-based Decision Support For Reducing Urban Flood At The Tidal Estuary

Large cities located at tidal estuaries such as Shanghai, Guangzhou and Tianjin are playing an important role in economy and society development in world. Because of the special geographical locations, climate and topography, those cities are threatened by flood, although meanwhile they benefit from abundant water resources and convenient transportation. Once flood happens, it will result in a huge loss in the national economy. Compared with those in large watersheds, inland area and other cities, the floods at urbanized tidal estuaries are very complicated and unique. However, very few researches have been done on flood-reduction for cities located at tidal estuaries in the past decades. Therefore, it is necessary and urgent to explore related methods and technologies, particularly the application of modern information technology.Based on the above discussion, the cities located at tidal estuaries are chosen as the study area, and Shanghai is taken as example in this thesis. The theories, methods and key technologies used in decision-making are discussed for the purpose of reducing flood disaster. A SDSS featured by real-time, dynamic, forecasting, early warning and layered-running is also established, by connecting a few key links such as data collecting, flood process forecasting and information sharing on the network, etc. This research not only gives an example of technology and application for other researches, but also provides scientific basis for reducing economic and social loss in shanghai city.First of all, this thesis analyzes the structure, characteristics, main disaster-causing factors and space-time distribution of flood disaster from the view of large complex system of flood disaster, based on plenty of historical data. A SDSS framework for reducing flood is then brought out on the basis of SDSS theory and technology. Next, the flood process is simulated by integrating GIS with hydrological model and database. Astronomical tide, typhoon and rainstorm are considered in this SDSS as well. Next, it is discussed how the information can be shared by city and county flood-control headquarters based on WebGIS, which extends SDSS applications of C/S mode. The SDSS is established, applied into the real problems and then verified, which provides a scientific basis for flood-reduction in Shanghai. Some solutions are also suggested in terms of the pitfalls of this system.This thesis has three main contributions:1. In terms of cities located at tidal estuaries, several key links including data collecting, flood process simulating and information sharing on the network are integrated based on SDSS, and a theoretical framework and constructing methodfeatured by integrating C/S and B/S modes, real-time, dynamic, forecasting and layered-running are brought out.The flood in the study area is a result from the interaction of disaster-breeding environment, disaster-causing factors and disaster-affected body. It is caused by astronomical tide, typhoon, rainstorm, flood upstream, sea level rise, ground subsidence and human activities. The current systems have quite a few pitfalls such as delayed information-sharing, inaccurate forecasting and warning, slow response to emergency, etc. This system is able to quickly gather data of weather, water regime, rain regime, hydraulic engineering, etc. It also can effectively simulate and forecast flood process by integrating GIS with hydrological models, and use WebGIS technology to publish information on Internet/Intranet.2. Using integrated hydrological models with GIS and database to simulate and forecast flood process dynamicly should also be considered an important contribution. The inherent mechanism of flood is also explored from the view of method.In terms of uban waterlogging caused by rainstorm and storm surge caused by typhoon, a system is established considering astronomical tide, typhoon, rainstorm, etc., based on the designed SDSS framework, the integrated GIS, hydrological model, database and middle-ware. This system can well simulate the process and space-time distri...