吳秉昇王冠棋Wu, Bing-ShengWang, Kuan-Chi陳宇軒Chen, Yu-Hsuan2022-06-082026-07-292022-06-082021https://etds.lib.ntnu.edu.tw/thesis/detail/0cbb20e7aa81dba753947c3a90ed6c74/http://rportal.lib.ntnu.edu.tw/handle/20.500.12235/117813氣候變遷影響引發的災害現象，成為環境壓力的一部分，使得水資源相關議題逐漸 被重視，地層下陷在水資源管理相關議題所造成的災害現象中是許多國家必需面對的課 題。就台灣而言雲林土庫鎮、元長鄉與褒忠鄉交界處為下陷速度最快、累積下陷最劇烈的 區域，該地同時也為台灣高鐵必經處以及台灣農業重要的發展區域。為增進地層下陷造成 地區差異與擬定有效的調適策略，本研究以災害脆弱度 (vulnerability) 的概念進行切入， 將區域、地方性 (locality) 以及小尺度的層面納入環境政策制訂進行詳細的脆弱度評估， 並提出兩個研究問題，一為如何運用脆弱度指標分析地層下陷及對在地危害的影響。再 者.如何將上述影響透過地理資訊 (GIS) 取徑建置雲林地層下陷潛在風險的空間模型。故 本研究將雲林地層下陷所面對的氣候變遷之影響，以損失 (loss) 及衡量損失之脆弱度為主 軸進行探討。透過雲林地層下陷的個案以理解全球氣候變遷下區域與地方所面對的災害現 象，利用特徵選取方法篩選出關聯性高的脆弱度因子，同時計算權重，得到脆弱度指標。 再藉由風險的概念，將脆弱度指標與在地進行連結，經由地理資訊系統以視覺化方式取徑 建置雲林地層下陷災害的模型，以空間關聯作為探討各項因子在地區性的影響。研究結果 表明，特徵選取方法可以有效改善過往脆弱度選取的困難，並保留資料的完整性。再者， 風險指標與地層下陷的嚴重程度雷同，高風險聚集在土庫鎮、元長鄉與褒忠鄉為主.作為 提供理解當地災害調適的考量。Land subsidence hazard is an urgent environmental issue that many countries are facing in the context of global climate change. In Taiwan, the land subsidence hazard is mainly caused by the overuse of groundwater in places such as Changhua, Yunlin, Chiayi and Pingtung Counties. In particular, Tuku, Yuanchang, and Baozhong Townships in Yunlin County have been facing the most rapidly land submerging. Meanwhile, Yunlin is an important agricultural zone in Taiwan, and the place where Taiwan high speed rail infrastructure was constructed to connect northern and southern Taiwan. This study demonstrates that deepening the understanding of environmental issues of land subsidence in Taiwan can help to develop environmental policies for Taiwan government to address challenges of global climate change. As a result, this study conducts research with a focus on vulnerability and risk of land subsidence in the context of global climate change. The contributions of this research are listed below. First of all, despite given studies on global climate change have attracted much attention in academia, related studies on environmental vulnerability or risk have paid less attention on the influences of climate change at a small scale; furthermore, only a few studies have conducted analysis to systematically map out the “loss” caused by global climate change. Therefore, with an aim to fill the gap of the scholarship of global climate change, this study addresses the phenomenon of land subsidence, vulnerability, risk, and their spatial relationship by using GIS related technologies. The contribution of this thesis is to develop a spatial model to address the land subsidence hazard in the context of global climate change based on the village-level data of Taiwan. The approach of this study is first of all, organizing and filtering the indexes of environmental vulnerability by using the Boruta analysis; following that, this study visualizes land subsidence by modeling the spatial relationships of these indexes in order to explain the risk and the loss in this regard. Finally, this study offers advices for the policy making based on findings. The model this study develops shows that the villages of potential risk (and loss) are places affected by serious land subsidence hazard.Boruta地層下陷風險空間模型脆弱度Borutaland subsidencelossriskspatial modelingvulnerabilityYunlin學術論文