都市雨水下水道淹水圖資建置與分析應用-以永和排水分區為例

Abstract

全球暖化與氣候變遷所帶來的極端氣候與降雨現象，短延時強降雨暴雨強度越來越大，嚴峻考驗排水設施的排水能力，使得都會區淹水災害事件頻傳。台灣位處副熱帶季風區，且於環太平洋地震帶上，因此颱風、洪水、地震及坡地災害等經常發生，又因台灣地理環境特殊，地形陡峭，山勢高聳，加上降水的時間及空間分布呈現不均勻的現象，造成台灣最多生命財產損失的災害即是洪水災害。水利署在2018年完成了第三代淹水潛勢圖資的公開作業，這些資料被用來規劃淹水防災措施。然而，淹水潛勢圖的繪製所使用的設計雨型是以Horner雨型為基礎，並在空間上採用均一分布。這種方法與實際降雨的時空間分布情況存在差異，導致淹水潛勢圖可能未能完全準確地反映真實的淹水風險。為增強水利防災人員對各地區淹水情況的決策判斷支援能力，本研究利用真實雨量數據製作暴雨時空間分布序列資料，生成大量符合實際降雨特性的不同模擬雨型。透過這些模擬雨型，結合SWMM淹水模式工具完成事件的淹水模擬。進一步分析這些模擬事件中積淹水的發生機率，製作淹水機率圖資提供防救災工作的參考，以不同的角度評估淹水風險，強化對淹水事件的預測和應對能力。考量到降雨紀錄及下水道設施圖資完整性，本研究選擇新北市永和地區進行深入分析，台灣降雨類型多變，降雨機率也極高，降雨資料相對龐大，因此本研究僅針對災害損失較嚴重的颱風類型降雨數據，搜集永和區雨量觀測站資料，篩選颱風事件雨量紀錄，並從中挑選降雨及延時足夠的降雨事件，以進行對時雨量的時間與空間特性分析。本研究期許能以現有技術或工具，建立氣候變遷情境下災害衝擊的評估，且透過評估成果的判讀，連結其與實際之應用，以此做為後續風險評估的重要參考依據。The increasing frequency and intensity of extreme weather events and rainfall patterns resulting from global warming and climate change pose significant challenges to urban drainage systems. Taiwan, situated in the subtropical monsoon region and along the Pacific Ring of Fire, frequently experiences typhoons, floods, earthquakes, and slope disasters. The unique geographic characteristics of Taiwan, including steep terrain and uneven distribution of rainfall in time and space, contribute to flooding disasters being the most common and costly in terms of loss of life and property.In 2018, the Water Resources Agency completed the public release of the third-generation flood susceptibility maps, which are utilized for planning flood disaster prevention measures. However, these maps are created using the Horner rainfall model with uniform spatial distribution, which may not accurately represent the actual spatiotemporal distribution of rainfall. Consequently, there may be discrepancies between the flood susceptibility maps and the actual flooding situations. To enhance the decision-making support capability of water resources disaster prevention personnel regarding flood situations in different regions, this study employs real rainfall data to develop a spatiotemporal distribution probability model for heavy rainfall events. Numerous simulated rainfall patterns that match actual rainfall characteristics are generated, and flood simulations for various scenarios are conducted using the SWMM flood simulation model. The probability of accumulated flooding occurrence in these simulated events is further analyzed to produce flood probability maps. These data will provide valuable references for disaster prevention and relief efforts, enabling a comprehensive assessment of flood risks and strengthening the prediction and response capabilities for flood events. Considering the completeness of rainfall records and sewer infrastructure data, this study focuses on in-depth analysis of the Yonghe District in New Taipei City. Given the variability of rainfall types and high rainfall probability in Taiwan, only rainfall data from severe typhoon events with significant disaster losses are selected for analysis. The temporal and spatial characteristics of rainfall intensity are analyzed accordingly. This study aims to establish an assessment of disaster impacts under climate change scenarios using existing technologies or tools. Through the interpretation of assessment results and their connection with practical applications, this assessment will serve as a crucial reference for subsequent risk assessments.