圖書典藏及影音數位平台

本計畫針對國內河堤沖刷破壞潛勢較高之大甲溪與大安溪，選擇數處易致災河段進行沖刷監測，研發以沖刷資訊建立之河堤沖刷破壞預警方法，同時，研擬河堤風險評估架構，結合現場監測、預警機制與互動淹水模擬等技術，藉由物聯網與雲端資訊系統判別溢潰堤潛勢並評估淹水範圍，以作為相關單位進行河防安全改善以及防災應變時之工具與參考資訊。

Typhoon often has a significant impact on the safety of dike in Taiwan, because of the steep terrain characteristics and high-intensity rainfall. A dike failure early warning framework along with the flood simulation technology is essential which could effectively reduce the damage caused by dike failure. We have installed the dike monitoring system in two specified reaches of Dajia and Puzi Rivers to obtain the real-time pore water pressure, scouring depth, and water stage. Besides, the dike failure threshold curves for the two reaches were developed which forecast the dike failure probability during a storm event. However, such approach is difficult to widely applied in the field due to the high cost of the monitoring system. Since most of the dike failures in Taiwan are caused by the severe bed scouring near the dike foundation, the purpose of this study is to develop a dike failure threshold curve which requires the scouring depth only. The project installed the scouring depth and water stage real-time monitoring system in 6 frequently damaged dikes in the Daan and the Dajia Rivers. The dike failure threshold curves for the 6 study reaches have been derived which forecast the dike failure probabilities for various lead-time. In addition, a real-time flood risk assessment framework has been developed which incorporating the monitoring system, the early warning threshold curve, and the flood simulation.