Simple Spatially-distributed Model for Inundation Simulations

台灣地區由於水文及地文條件特殊，當暴雨事件發生時，常在短時間內導致淹水災情。傳統的淹水模式，雖可獲得良好的模擬結果，但須花費較長的演算時間，無法在緊急應變前置時間內完成淹水模擬。研究主要目的是建立一套適用於台灣地區之淹水計算模式，銜接中央氣象局定量降水預報系統(QPESUMS system)，利用雷達回波資訊推估之雨量整合地面雨量站觀測之雨量，進行即時演算淹水模擬，並以納莉颱風事件評估不同網格尺度及演算時距對模式運算時間及模擬精度之影響。究區域為基隆河流域，採用海棠颱風及柯羅莎颱風兩場颱洪事件進行淹水模擬，成功銜接QPESUMS雷達雨量資訊，及在短時間內完成淹水演算工作。未來若能加入1-3小時預報降雨量，在短時間內完成淹水模擬，模擬結果可提供做為1-3小時之淹水預報，有助於決策單位進行災情分析研判及後續相關防救災工作，提昇應變作業之效能。Although traditional inundation models have been applied with good accuracy in Taiwan, they usually require a long computing time for simula-tions. However, the meteorological and geographical conditions in Taiwan frequently cause inundation within a short time period when storm occurs. The lead-time for emergency response in too short to indicate the areas with high flood risks for evacuation by using the traditional inundation models.he study aims to establish an inundation model for Taiwan and inte-grate the QPESUMS system which constructed and developed by the Cen-tral Weather Bureau. The radar precipitation by the QPESUMS system, as well as the rain-gauge records, are considered in the inundation model for real-time simulations. The precipitation data of typhoon NARI were simu-lated and evaluated different scale of grid size that the accuracy and effi-ciency of model would be suggested for practical applications.he Keelung River basin is adopted as the study areas of the inunda-tion model. By use of QPESUMS radar precipitation for the typhoon HAITANG and KROSA, the inundation simulations can be calculated in a short time. The model will be executed in the future, to simulate the flood scenarios induced by the occurring and the forecasted rainfalls. The inun-dation will be predicted in 1-3 hours ahead to help the emergency managers taking proper strategies for disaster mitigations.目 錄 誌 I 要 IIbstract III 錄 IV目錄 VII目錄 X一章 緒 論 1 1-1 研究目的 1 1-2 文獻回顧 2二章 淹水模式與雷達估計降雨 6 2-1 淹水計算模式 6 2-1-1 網格劃分及其基本假設 6 2-1-2 基本方程式 6 2-1-3 差分式之建立 8 2-1-4 求解方法 8-1-5 初始及邊界條件 10 2-2 雷達-雨量站整合降雨估計 11 2-2-1 雷達觀測原理與應用 11 2-2-2 QPESUMS系統 12 2-2-3 雷達-雨量站整合降雨技術 13三章 研究區域概述 15 3-1 流域概況 15 3-2 模擬區域 16 3-3 歷年災害 16四章 資料整合與輸入 18 4-1 地形資料 18 4-1-1 數值地形 18 4-1-2 地理資料處理 18 4-2 土地利用 19 4-3 雨量資料 19 4-3-1 地面雨量站資料 19 4-3-2 QPESUMS雷達雨量前置處理 20 4-3-3 QPESUMS雷達雨量與淹水模式之銜接 21五章 模式驗證 23 5-1 納莉颱風事件 23 5-2 網格尺度與演算時距之選定 24 5-3 驗證之模擬結果 25六章 模式應用 28 6-1 海棠颱風事件 28 6-2 科羅莎颱風事件 30七章 結論與建議 32 7-1 結論 32 7-2 建議 33考文獻 34錄 8