1999年的集集大地震撼動全台,驚醒的許多夢中人,不但造成台灣人民傷亡、房屋倒塌、更引發許多邊坡產生滑動破壞,也造成濁水溪沖積扇地區許多起液化事件發生。 在國內外許多地震事件中,經常發現地下水位在地震發生前後有異常變化之現象。1999年集集地震發生時,由經濟部水資源局在鄰近於車籠埔斷層西側的濁水溪沖積扇建立之地下水位監測井網,恰好紀錄了濁水溪扇洲於地震期間之水位變化。 為能了解含水層受地震力作用下其孔隙水壓變化的關係,本研究利用有限差分軟體FLAC程式建構孔隙水壓力動態反應模式的概念模型,此概念模型利用到地下水觀測井之地層剖面資料,並進行合理的土層簡化以利分析,接著利用中央氣象局在濁水溪沖積扇區佈設的強震測站,並經由內插得到地下水觀測井口的地表加速度歷時紀錄,接著使用SHAKE程式當中的反捲積(deconvolution)計算功能推得地下水觀測井基盤位置的地震加速度歷時資料,最後利用FLAC程式建立地下水觀測井的地層剖面,並分別輸入南北向以及東西向的地震加速度歷時紀錄,藉由數值軟體得到分析結果以了解地震與地下水位變化之相關機制。 進一步根據濁水溪沖積扇的地下水觀測井網於集集地震發生時所紀錄到的實際地下水位變化量等資料,與數值分析結果進行比較及討論,經分析後發現以東西向震波作為數值模型的輸入依據所得的結果較南北向震波優異,並其結果與實際地下水位變化相符合,也與整體趨勢線一致,所以模擬成效相當不錯。
The Chi-Chi earthquake truck Taiwan in 1999, not only caused casualties, house collapses, slope failures, but also caused many liquefication in the Choshui alluvial fan. The abnormal change of groundwater level has often accompanied earthquake incidents around the world. During the Chi-Chi earthquake, a network of groundwater monitoring wells located in Choshui River alluvial fan near the west side of Chelungpu fault recorded the abrupt changes of groundwater level which clearly showed the effect of this major earthquake. In order to understand the relation of the pore pressure change of the aquifer subjected to qarthquake, this study used finite-difference software, FLAC, to construct the conceptual model of the pore pressure dynamic response pattern. The ground stratigraphic section data of the groundwater monitoring wells and reasonable simplify cation of soil layer to were used in analysis. The ground acceleration records of the Central Weather Bureau in the Choshui River alluvial fan, and records at the groundwater monitoring wells were determined by way of the interpolation. The deconvolution analysis were performed SHAKE to get the earthquake acceleration time histories for the base input of the groundwater monitoring wells. Comparisons of the analysis results with the groundwater level record obtained during the Chi-Chi indicated that the result of east-west seismic wave compares the north-south seismic wave as the numerical model input basis obtained outstanding, and its result and the actual groundwater level change tallies, also with overall tendency consistent, therefore the simulation result is quite good.