生長正斷層錯動引致覆土層剪切帶發展之研究

Abstract

活動斷層錯動引致之地震，因地盤錯動所產生永久或塑性變形，會引致結構物及重要的維生管線嚴重變形而損壞。臺北盆地西緣的山腳斷層屬於高活動度之正斷層，山腳斷層之錯動將造成覆蓋於臺北盆地的第四紀沉積物變形，進而造成影響區域內之結構物或交通建設及維生管線（油、氣、輸水管...等）破壞。 由鑽孔及定年資料推斷山腳斷層有生長斷層 (growth fault) 現象，此種斷層錯動所造成上覆蓋沉積層之變形行為，與數值模擬單一覆土層錯動影響之異同為本研究探討重點。本研究建置砂箱物理模型以無凝聚性砂土進行模擬山腳斷層在有生長斷層情形下，剪切帶發展範圍及地表差異變形之影響，並針對生長正斷層分次錯動的影響進行試驗。實驗結果顯示正斷層如含有生長斷層，當基盤錯動時，剪切帶會沿原覆土層之剪切帶弱面向上發展。且此剪切帶會比單一覆土層，更為快速發展至地表，錯移率（基盤錯移量/下盤覆土層厚度）僅需單一覆土層之約1/3（或5/16）。本研究以分離元素法二維分析程式(PFC2D)進行砂箱試驗成果之模擬，模擬生長正斷層錯動對上覆土層之剪切帶發展速度及影響範圍。模擬結果顯示與砂箱試驗成果甚為接近。 依據上述試驗結果，本研究建立五股剖面的幾何模擬模型，以含生長層之上盤厚度H係以原來上盤厚度之5/16計算為新的上盤厚度，推算含生長斷層平均錯動量為2.5m之錯移率與斷層尖端發展位置之關係，推論得出當臺北盆地第四紀沉積物厚度低於510m時，含生長層之斷層錯動時上覆土層之剪切帶，均會影響至地表，而當覆土層厚度接近680m時，剪切帶之影響深度為地表下約20餘公尺，約為656m。可推論：含生長斷層之平均錯動量越大，其斷層尖端發展位置會越接近地表，亦即剪切帶的發展會越接近地表。 且經PFC2D進行第四季沉積物厚達679m的五股剖面模擬，亦顯示生長正斷層錯動2.5m後，剪應變亦會發生在近地表處。因此針對山腳斷層之週邊之重大工程設計分析，應需考慮生長斷層之影響，進行鑽孔調查是否有生長斷層之證據，生長斷層之剪切帶發展代表最近一次斷層錯動對上覆土層之影響，因此剪切帶之可能範圍內，須特別加以注意。

A fault slip can cause the deformation of shallow soil layers and destroy infrastructures. The Shanchiao Fault on the west side of the Taipei Basin is one such fault. The activities of the Shanchiao Fault have caused the quaternary sediment beneath the Taipei Basin to become deformed, damaging structures, traffic construction, and utility lines in the area. Data on geological drilling and dating have been used to determine that a growth fault exists in the Shanchiao Fault. In an experiment, a sandbox model was built using noncohesive sandy soil to simulate the existence of a growth fault in the Shanchiao Fault and forecast the effect of the growth fault on shear-band development and ground differential deformation. The experimental results indicated that when a normal fault contains a growth fault at the offset of the base rock, the shear band develops upward beside the weak side of the shear band of the original-topped soil layer, and surfaces considerably faster than that of the single-topped layer. The offset ratio required is approximately one-third that of the single-cover soil layer. In this study, a numerical simulation of the sandbox experiment was conducted using a discrete element method program, PFC2D, to simulate the upper-covering sand layer shear-band development pace and the scope of a growth normal fault slip. The simulation results indicated an outcome similar to that of the sandbox experiment. According to the above test results, the Wuku profile geometric simulation model established in this study, the new hanging wall thickness H is 5/16 of the original thickness in every sedimentary strata of the growth normal fault. The simulation results show that the more average offset will lead the shear-band propagation reached close to the ground surface. The PCF2D program was used to create a model for simulating SCF-2 and WK-1E profiles and the shear-band propagation reached the particle surface in the final 2.5-m slip of this growth normal fault numerical model. The simulation results can be applied to the design of construction projects near fault zones.