利用合成孔徑雷達差分干涉技術探討莫拉克風災大規模潛勢區位之地表變形：以南投縣仁愛鄉為例

Investigation of Typhoon Morakot Triggered Deep-Seated Landslides by Using Differential Interferometric Synthetic Aperture Radar (D-InSAR): An Example of Renai Township, Nantou County

陳柔妃(Rou-Fei Chen) ； 吳若穎(Ruo-Ying Wu)

《中華防災學刊》 9卷2期 (2017/08) Pp. 125-131

摘要

近幾年來大規模崩塌所衍生災害為坡地防災的重點，2009年莫拉克颱風期間，由潛在大規模崩塌轉換成劇變式山崩，在小林村造成450人傷亡的嚴重災情。如何找出台灣全島大規模崩塌潛勢區位、評估其災前災後的影響，快速地提供地表變形資訊，成為後續大規模崩塌災害防治重要的參考。過去幾年間，經濟部中央地質調查所針對莫拉克颱風受災集水區，藉由產製高精度空載光達(Airborne LiDAR)數值地形，配合同步高解析度航照影像，輔以地質資料及野外勘查，判釋出坐落於臺灣南、中、東部區共56處，具保全對象(包括聚落、水庫、道路及橋梁等)之大規模崩塌潛勢區位。然而大規模崩塌潛勢區位之變形作用極為複雜，藉此行政院農業委員會水土保持局，為持續掌握大規模崩塌潛勢區位之地表變形，首度使用日本衛星雷達影像(ALOS/PALSAR)，進行大規模崩塌潛勢區位活動性之可行性評估，突顯此項技術有助於大規模崩塌災害後續防治之工作。本研究利用莫拉克風災前後日本衛星雷達影像(ALOS/PALSAR)產置崩塌區位內之地表變形量，藉此評估單一崩塌潛勢區內活動性較高之區域，將有助於後續之減災工作。

關鍵字

合成孔徑雷達差分干涉技術 ； 大規模崩塌 ； 日本衛星雷達影像

並列摘要

In recent years, disasters induced by deep-seated landslides have been the focus of slope-land disaster prevention works. During Typhoon Morakot in 2009, the deep-seated landslide in Hsiaolin Village transformed into a catastrophic landslide and killed over 450 people. How to find deep-seated landslides on the main island of Taiwan, assess their pre-disaster and post-disaster impacts, and swiftly provide surface deformation data are rather significant as they are important references for disaster prevention and mitigation works. Over the last few years, Central Geological Survey (CGA), Ministry of Economic Affairs (MOEA) targeted Typhoon Morakot affected areas to produce high-precision airborne LiDAR digital terrains and, together with high-resolution aerial imagery, geographical data and field survey, successfully identified 56 significant deep-seated landslides with secured objects (including villages, reservoirs, roads and bridges) in the South, Center and East of Taiwan. Nevertheless, the activity and behavior of deep-seated landslides are extremely complicated. To continuously monitor surface deformation of deep-seated landslides, Soil and Water Conservation Bureau (SWCB) of Council of Agriculture (COA) has therefore, for the first time, assessed the feasibility of evaluating the activity of deep-seated landslides using Japanese ALOS/PALSAR imagery, showing the importance of this technology in preventing and mitigating deep-seated landslide disasters. To facilitate the following disaster mitigation works, we have produced the surface deformation rate with pre- and post-Typhoon Morakot ALOS/PALSAR imagery to identify relatively active areas in a potential landslide site in this study.

並列關鍵字

D-InSAR ； deep-seated landslide ； ALOS/PALSAR