| 摘要: | 地層下陷是台灣目前急需解決的環境問題之一，除沿海低漥地區因地層下陷所產生之淹水問題外，台灣高鐵路線上之下陷嚴重區，將造成交通安全層面上的疑慮。目前高鐵路線上所經過之下陷區屬濁水溪沖積扇最為嚴重，其中以虎尾、土庫、元長及褒忠等鄉鎮為主要下陷區。地層下陷的觀測方法有許多種，其中精準度最高且分布最廣的觀測資料為水準測量。但此方法之缺點是監測時間為一年一次，監測頻率過長，而地下水水位為影響地層下陷影響的主因，地下水會隨著旱季或雨季而有水位的高低起伏，因此一年一次的監測頻率是明顯不足的。為解決各種觀測資料在時間、空間上精度與頻率差異，所產生資料無法直接互補之問題，本研究專注於濁水溪沖積扇地下水區，使用kriging插值，假定採樣點之間的距離或方向可以用於反映插值結果的空間相關性，利用kriging插值出水準測量資料與GPS固定站資料相同時間內的地表高程變化圖，並計算兩者間的誤差，再以此誤差對各個月份的GPS地表沉陷進行融合，結合水準測量結果的精準度以及GPS固定站的高監測頻率，得到不同月份的地表沉陷分布，用來分析地下水位與地層下陷的關係，發現地表高程變化會根據不同的地質條件及雨季與旱季的地下水水位變化，而有抬升及下陷的發生。此外，本研究藉由磁環式監測井與水準樁的觀測資料進行整合，評估地表下300m之壓縮量，並與地表觀測下陷分布比較，探討區域較深層的壓縮狀況，結果表明深層壓縮的熱區與嚴重地層下陷熱區相吻合，推斷深層抽水造成的壓縮對於雲林土庫地區的地層下陷有明顯的占比。為了探討地下水水位與地層下陷間的關聯，以DTW方法分析雲林高鐵沿線3km封井防治對策執行前後地下水位與地層下陷之關係，發現在封井後的淺層地下水位與地層下陷有較高的相似性，說明減抽前主要為深層抽水影響地層下陷。;Land subsidence is one of the environmental problems that Taiwan needs to solve urgently. In addition to the flooding problem caused by the land subsidence in the coastal low-lying areas, the serious subsidence of the Taiwan high-speed railway line will cause doubts on the level of traffic safety. At present, the Chuoshui alluvial fan is the most serious subsidence area passing through the high-speed railway line. Among them, Huwei, Tuku, Yuanchang and Baozhong towns are the main subsidence areas. There are many kinds of observation methods for land subsidence, among which the most accurate and most widely distributed observation data is leveling. However, the disadvantage of this method is that the monitoring time is once a year and the monitoring frequency is too long. The groundwater level is the main cause of the impact of land subsidence. The groundwater level will fluctuate with the dry season or the rainy season, so the monitoring frequency is once a year. It is obviously insufficient. In order to solve the problem of the difference in accuracy and frequency of various observational data in time and space, the generated data cannot be directly complementary. This research focuses on the groundwater area of the Chuoshui alluvial fan and using kriging interpolation, assuming that the distance or direction between the sampling points can be used to reflect the spatial correlation of the interpolation results, using kriging to interpolate the leveling data and the GPS fixed station data within the same time The surface elevation change map of, and calculate the error between the two, and then use the error to integrate the GPS surface subsidence of each month, and combine the accuracy of the leveling results and the high monitoring frequency of the GPS fixed station to obtain the surface subsidence of different months The distribution is used to analyze the relationship between groundwater level and land subsidence. It is found that changes in surface elevation will cause uplift and subsidence according to different geological conditions and changes in groundwater level in rainy and dry seasons. In addition, this study integrates the observation data of magnetic ring monitoring wells and leveling piles to evaluate the compression of 300m below the surface, and compares it with the distribution of observations on the surface to explore the compression status of the deeper layers of the region. The results show that the heat of deep compression The area coincides with the severe land subsidence hot zone, and it is inferred that the compression caused by deep pumping has a significant proportion of the land subsidence in the Yunlin Tuku area. In order to explore the relationship between groundwater level and land subsidence, the DTW method was used to analyze the relationship between groundwater level and land subsidence before and after the implementation of the prevention and control measures for the 3km well closure along the Yunlin High-speed Railway. The similarity indicates that before the reduction of pumping, it is mainly due to deep pumping that affected the land subsidence. |
