深層含水地層之壓縮行為

Abstract

超抽地下水引起之地層下陷以往多著重於地層中壓縮性較大之阻水層，然而根據現場觀測發現，部份含水層(即砂土層)存在不可忽視之壓縮性，對地層下陷有顯著影響。由於傳統鑽探不易取得不擾動砂性土樣，以擾動試體進行室內試驗所得之壓縮行為亦無法代表地層真實特性。因此本研究以濁水溪沖積扇為例，將現地擾動土樣利用霣降壓密製作室內試驗試體，並於試驗過程中以剪力波元件量測剪力波速，當波速與相對應之地層波速相同時，即視試體已回復現地狀態，再以此試體模擬地下水位升降造成有效應力改變之反覆荷重加載，探討因抽水導致深層含水層之壓縮行為。室內試驗結果也嘗試與現場監測資料比對，驗證室內試驗代表性。 研究結果顯示現地土壤狀況受到沉積方式(顆粒排列狀態)、應力歷史與其它環境因素影響，當試體波速與現地相同時，試體應力狀態略低於理論推估值，因此以波速比對將試體回復現地真實狀態為一可行的方法，但未來試體製作應考慮現地土壤形成過程。此外，本研究亦完成一系列模擬地下水位變化之室內壓縮試驗，結果顯示濁水溪沖積扇之深層含水地層會因週期性反覆荷重產生塑性變形，且彈性行為不明顯，此結果驗證含水層會因地下水位變化產生不可忽視之壓縮量，評估地層下陷時應考慮其影響性。同時研究中也發現當含水層內夾多層薄細料時，將會影響整體排水特性造成土層壓縮與時間有關之壓密沉陷，此與現場監測成果一致，可見室內試驗能良好反應現地深層含水層實際壓縮行為。

The excessive land subsidence induced by over-pumping of groundwater is usually explained by the consolidation of aquitards. However, some evidence from field measured data has shown that, in some cases, the compression of some aquifer (e.g., sandy layer) may be quite significant. Generally, it is difficult to obtain undisturbed sandy samples in field. Because the results of laboratory experiment using disturbed specimens were unable to express the real characteristic of in-situ layer, this study took the disturbed soils in Choshui River alluvial fan to reconstitute by pluviation and consolidation and try to examine the compressibility of deep aquifer induced by groundwater withdraw. During consolidation, shear wave velocity were measured by bender elements embedded in the upper and bottom plates of the consolidation device. The state of the reconstituted material was examined by its shear wave velocity in order to match the same wave velocity as obtained from in-situ P-S logger tests. After material reconstitution, the specimens were subjected to loading/unloading cycles to simulate that effective stress changes due to the groundwater-level fluctuation. The results of laboratory test were also compared with field data to verify the validity of the laboratory model test. Experimental results show that the stress state of in-situ layer was influenced by the soil deposition (e.g., the arrangement of soil particle), stress history and others. These factors cause the stress condition of samples were not consistent with the state in field when shear wave velocity were the same in laboratory and in field. Because the stress state were not consistent, the way that reconstituted specimens were returned to original condition by comparing shear wave velocity is not better. However, it is still acceptable. In the future, the factors should be overcome while preparing the remold specimens for laboratory experiments. A series of laboratory consolidation test for modeling the groundwater-level fluctuation were conducted in this study. The results indicated the deep aquifer of Choshui River alluvial fan was compressed by loading/unloading cycles, and the rebound of sandy layer was not clear as the stress was unloading. These findings show the compression of aquifer due to groundwater-level fluctuation is plastic. Hence, the compressibility of deep aquifer should be considered when estimating the land subsidence. This study also finds the compression of aquifer with thin clayey layers is time-dependent. Similar phenomenon was observed in field. Thus, the laboratory test may explain the land subsidence phenomenon contributed by the compression of deep aquifer.