| 摘要: | 濁水溪沖積平原（CRFP）是台灣中部農業和水產養殖活動的重要地區之一，地下水在該區是灌溉、家庭和工業用水的主要供應來源。幾十年來，CRFP 明顯增加的地下水抽取量導致了地層下陷。儘管相關單位已實施了一系列降低抽水量的政策，但地面仍然持續下陷，並可能危及人的生命和基本基礎設施的安全。為了能控制該區的沈陷速率，有必要對該區地面高程隨時間的變化進行監測。
本研究希望能藉由SBAS-DInSAR 技術處理 2018-2020 年間所取得的 90 多幅 Sentinel-1A SAR影像來觀察 CRFP近期的地層下陷型態。結果顯示地層下陷最嚴重的區域位於雲林縣中西部。沈陷中心為東勢鄉、元長鄉、四湖鄉、水林鄉、口湖鄉及崙背鄉、土庫鎮與虎尾鎮之交匯處。在這三年期間，雲林縣的最大累積沈陷量達到約–13 至–18 公分。此外根據與光學影像的比較結果，本研究注意到大多數沈陷的地方與農業區域有關。另一方面，彰化縣僅在年下陷速率為 2 至 3 公分/年時才出現中度沈陷。至2020年末，彰化縣的總累積沈陷量主要在–2 到 –5 公分之間變化。埔鹽鄉與溪湖鎮的最大沈陷量達到-7公分，較周圍來的嚴重。本研究另外還分析了累積位移、地下水位和降雨量之間的cross-correlation以評估這些物理量之間的相關性。分析結果顯示地面沈陷與地下水位波動之間有著中等的相關性，相關係數的範圍大約為 0.4 到 0.7。同時，降雨與位移的相關性相對較低，因此其對地表變形的影響並不顯著。
這項研究還提供了一個顯示高鐵沿線總累積位移的剖面圖。結果顯示了虎尾鎮、土庫鎮和元長鄉的鐵路段在研究期間內有著嚴重的沈陷。但即使在元長鄉與土庫鎮都觀測到了變形速度的減慢，虎尾地區下仍有較高的沈陷速率。因此應持續監測虎尾鎮、土庫鎮與元長鄉的沈陷情況，以便決策者提出對應下陷速率的控制政策，並最大限度地降低鐵路的潛在風險。
最後，本研究使用了具備多種強大功能的PCI Geomatica軟件來處理SAR圖像，並估計了 2018-2020 年間 CRFP的地層下陷時空發展情形。;The Choushui River fluvial plain (CRFP) is one of the critical regions for agricultural and aquacultural activities in Central Taiwan. In this region, groundwater is the main supply for irrigation, domestic and industrial usages. For decades, the amount of extracted groundwater in CRFP has significantly increased, resulting in land subsidence. Although the authorities had some policies to reduce the pumping rate, the ground has unexpectedly continued to subside, which might risk the safety of human lives and essential infrastructures. Therefore, monitoring the ground elevation changes in this area over time is necessary to control the subsiding rates.
This study aims to observe the recent land subsidence patterns in the CRFP by applying the SBAS-DInSAR technique to process over 90 Sentinel-1A SAR images acquired from 2018 – 2020. The results indicated that the regions that suffered most from land subsidence were located in the western central side of Yunlin County. The centers of subsidence were Dongshi, Yuanzhang, Sihu, Shuilin, Kouhu, and an intersection area of Lunbei, Tuku, and Huwei districts. After three years, the maximum cumulative displacement values in Yunlin County reached approximately –13 to –18 cm. In addition, it is also noticed that the most subsiding places correspond to the agricultural lands, based on the comparison with optical images. On the other hand, Changhua County only experienced moderate subsidence when the annual sinking rate varied from 2 to 3 cm/year. At the end of the study period, the total cumulative subsidence values in Changhua County mainly varied from –2 to –5 cm. Puyan and Xihu districts witnessed more serious subsidence than the surroundings, with the maximum displacement reaching –7 cm.
Furthermore, cross-correlation analyses between the cumulative displacements, groundwater levels, and rainfall were conducted to statistically assess the correlation between these quantities. The analyses indicated that the correlations between subsidence and groundwater level fluctuations were moderate, ranging from 0.4 to 0.7. Meanwhile, rainfall insignificantly impacted the surface deformation since its correlations with displacements were relatively low.
This study also provided a profile indicating the total cumulative displacements along the High-Speed Rail. It was shown that railroad segments in Huwei, Tuku, and Yuanzhang districts experienced severe subsidence during the research period. Especially, the sinking rate in Huwei district was relatively high even though Tuku and Yuanzhang have observed the decelerated deformation. Therefore, the subsidence in Huwei, Tuku, and Yuanzhang districts should be monitored continuously so that the decision-makers could introduce appropriate policies to control the sinking rates, minimizing potential risks to the railroad.
Finally, this study initially applied the PCI Geomatica software, which provides multiple powerful functions, to process SAR images more conveniently. The results of this study could be an estimation of the spatial and temporal development of land subsidence in the CRFP during 2018 – 2020. |
