以衛星大地測量分析臺灣西南部地殼形變;Crustal Deformation in Southwestern Taiwan from Geodetic Observations

NCU Institutional Repository > 地球科學學院 > 地球物理研究所 > 博碩士論文 > Item 987654321/97706

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://ir.lib.ncu.edu.tw/handle/987654321/97706

题名:	以衛星大地測量分析臺灣西南部地殼形變;Crustal Deformation in Southwestern Taiwan from Geodetic Observations
作者:	林耕霈;Lin, Geng-Pei
贡献者:	地球科學學系
关键词:	地殼形變;斷層潛移;雷達差分干涉;泥灌入體;Crustal Deformation;Fault Creep;InSAR;Mud Diapir
日期:	2025-08-28
上传时间:	2025-10-17 11:48:41 (UTC+8)
出版者:	國立中央大學
摘要:	臺灣的西南部地區是弧陸碰撞下演育較晚的構造區域，區域裡有著密集的斷層和褶皺構造，在1940年代也曾經發生大型災害地震，但近年僅山區發生中尺度地震，例如2010年甲仙地震，平原地區則少有大型地震，可謂是臺灣陸域的一地震空區。由於臺灣西南部相對開發密度高的中北部都會區，仍有較多的土地可作利用，目前也有新興的科技和工業園區正在運作或是即將開發，因此精確了解西南部的已知斷層活動情形和潛在的地質災害機制，成為一個重要議題。本研究利用ESA的Sentinel-1衛星自2017至2024年以來累積的SAR影像，利用經過改進的PSInSAR技術，獲取高空間解析度（20公尺）的震間期地殼形變場，利用高解析度變形場，提出一個利用衛星升軌和降軌資料，分離出三維速度場的新流程，過程中可以優先保留垂直向準確度，再分離出準確度相對較低的水平向分量，最後將與GNSS三分量資料融合，使得融合過程得以保持原始InSAR資料空間解析度精度，又無須先將GNSS速度場投影至衛星視線方向（LOS），得以避免引入GNSS較大的垂直向誤差。第三章成果顯示，許多斷層可能有淺部潛移，或斷層鄰近的近地表褶皺作用。根據判釋沿著已知斷層跡的變形邊界，本研究提出後甲里斷層南北的延伸，尤其向北延伸至新化斷層，並可能連結至口宵里斷層。鳳山構造也有明顯的向南北延伸變形，北至右昌構造，南可延伸至屏東平原海濱，總長度達到目前臺灣地震模型版本的兩倍。本研究也辨識了兩條先前未見登載的變形邊界，是未來調查疑似斷層的重要標的，一條是小崗山斷層西南方相似走向的構造，另一條則是位於屏東平原中央南北走向的構造。利用高解析度變形場，本研究進一步獲得200公尺解析度的水平應變率場，這是震間期地殼應變率場少有的超高解析度。展示了西南部的應變率集中在兩組東北西南向的構造線上，許多斷層因為潛移或是近地表的褶皺作用，出現了密集於斷層線上的高應變率，西南部最高的應變率出現在車瓜林斷層與右昌構造一帶（2.4 microstrain/yr）。相較於前人的成果，本研究的應變率場清晰的提供了分辨斷層潛移訊號和背景應變率的依據，這使得後續分析測地震矩累積率時，比起僅使用GNSS資料的求得的5.58 Mw/yr，獲得更加精確的6.24 Mw/yr。最後本研究第五章為2016美濃地震的同震地表抬升，以泥貫入體構造的角度，從測地學的觀點以類岩床模型模擬，論證了同震泥貫入體作用的可能性，為地震和地質構造的互動拓展了新的視野。 ;Southwestern Taiwan is a later developed member of Taiwan orogenic belt, a region characterized by dense faults and folds, and it experienced several large destructive earthquakes in the 1940s. However, in recent decades, only moderate-magnitude earthquakes have occurred in mountainous areas, for example, the 2010 Mw 6.2 Jiashian earthquake, while the plain region can be considered as a seismic gap within the onshore Taiwan. Compared to the highly urbanized regions in central and northern Taiwan, the southwest still possesses considerable land available for use. Emerging industrial and science parks are either already operating or planned for future development. Therefore, accurately understanding the activity of known faults and assessing potential geological hazards in southwestern Taiwan has become a critical issue. I utilized Synthetic Aperture Radar (SAR) images acquired by ESA’s Sentinel-1 satellites from 2017 to 2024. An improved PSInSAR approach was applied to derive high-resolution (20-meter) interseismic deformation fields. Based on this refined deformation data, I proposed a novel workflow to decompose three-dimensional velocity fields by integrating ascending and descending track InSAR datasets. The procedure prioritizes the accuracy of the vertical component, which is then used to infer the relatively noisy horizontal components. Finally, this study integrated the results with 3D GNSS velocity data, enabling the retention of the original InSAR spatial resolution. And did not project GNSS data into the satellite’s line-of-sight direction, thus avoiding the large vertical uncertainties associated with GNSS. The results revealed that many faults in the region may exhibit shallow aseismic creep or are associated with near-surface folding. Based on deformation boundaries interpreted along known fault traces, I suggest possible extensions of the Houchiali Fault, especially northward towards the Hsinhua Fault and potentially to the Koshawli Fault. The Fengshan structure also shows significant north–south extensions, from the Youchang structure in the north to the coastal region of the Pingtung Plain in the south, with a total length nearly twice than current Taiwan Earthquake Model (TEM) version. Additionally, this work identified two previously undocumented deformation boundaries that merit further investigation as potential fault candidates: one at the southwest of the Hsiaokangshan Fault with similar strike, and another north–south striking structure in the central Pingtung Plain. Using the high-resolution deformation field, this study further derived a 200-meter resolution horizontal strain rate field, it was the highest resolution interseismic strain fields ever reported. Chapter 4 shows that strain is concentrated along two sets of NE–SW trending structures, where several faults show high strain rates concentrated near their traces, potentially due to aseismic creep or near-surface folding. The highest strain rate (2.4 microstrain/year) is observed near the Chekualin Fault and the Youchang structure. Compared to previous studies, our strain rate field provides a clearer distinction between localized fault creep and background strain, allowing a more accurate estimation of geodetic moment accumulation. While previous GNSS-only studies estimated an accumulation rate of ~5.58 Mw/yr, our integrated approach yields a precise value of 6.24 Mw/yr. Finally, Chapter 5 examines the 2016 Meinong earthquake’s co-seismic uplift from an alternative perspective, proposing a sill-like model to simulate the additional deformation from coseismic triggered mud diapirism. This approach expands the conceptual framework of how earthquake and geological processes may interact in southwestern Taiwan.
显示于类别:	[地球物理研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	25	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....