| Summary: | 碩士 === 國立臺北科技大學 === 土木工程系土木與防災碩士班 === 105 === At present, the most common flood model in Taiwan is using full two-dimensional way to run flow simulation and overland simulation. But using this way to run flood simulation will cost too much time that it doesn’t meet the needs of the requirement of disaster prevention. In order to reduce the flood model computing time, there comes the concept of one-dimensional coupled two-dimensional flood model. Which is using one-dimension flow model to simulate the water level and discharge of the channels. If there’s overflowing occurrence then the model will use the overland model to simulate the water movement on the ground. This coupled method can effectively reduce the flood model simulation time and simultaneously it’s meet the needs of the requirement of disaster prevention. At present, Water Resources Planning Institute, Water Resources Agency, Ministry of Economic Affairs (known as WRAP) has built a flood model based on this concept and within the micro-catchments that sized are around 2 square kilometers. But will the resolution of the catchments scale become a major impact to the flood simulation results. It’s a subject that has not been discussed in recent years.
Therefore, this study will explore how much impact on the flood simulation results when using different catchments scale in the one-dimension flood model. So in this study, I’m using the module from WRAP and select Taichung City and Tainan Annan District as research area. Then use the concept of the Horton-Strahler Number for channels classification, and then combining the catchments and editing the flood model based on this classification. Taichung City can result in 7 versions of flood model. Tainan Annan District can result in 5 versions of flood model. After that, select numbers of known flooded typhoon events and using the QPESUMS（Quantitative Precipitation Estimation and Segregation Using Multiple Sensor）data from Central Weather Bureau to run the flood simulation and analysis. At Taichung study area using the 2008 typhoon Kalmaegi shows that results of the flood area at scale 2 is 64％ less than scale 1’s flood area, maximum flood depth drop 7 cm and the maximum flood occurred time is 200 minutes delay. Tainan study area using the 2016 typhoon Megi shows that results of the flood area at scale 2 is 15％ less than scale 1’s flood area, maximum flood depth increase 1 cm and the maximum flood occurred time is 10 minutes delay. results of the flood area at scale 3 is 45％ less than scale 1’s flood area, maximum flood depth drop 2 cm and the maximum flood occurred time is 20 minutes delay. This study can conclude that it is necessary to use the highest resolution of the catchments while building two-dimension flood model, otherwise the flood simulation results from the flood model will not be accurate as the high resolution flood model.
|
|---|
