2012-01-012024-05-15https://scholars.lib.ntu.edu.tw/handle/123456789/663643摘要：了解位於地表下的斷層尖端擴展受上覆土層對斷層跡修飾的效應，以及探討斷層作用時，隨斷層尖端擴展主要變形區內土體內應力、應變的空間分布及其對結構物之影響，學理上對斷層帶退縮距離的決定相當重要。另外加強對斷層錯動-土壤-結構互制的了解，針對直接承受斷層錯動衝擊的結構物，如何強化結構設計，盡量減少衝擊的程度也是地震減災可行的方法。 目前對於活動斷層近地表變形特性的研究在研究方法大致上可以分為現地調查、物理模型試驗以及數值分析等三大類，本計畫之研究為突破單一方法學提出結果之侷限性，因此併用現地調查、物理模型實驗(離心機砂箱試驗、1g砂箱實驗)及數值分析之方法，分別進行基本行為觀察及定量分析。 第一年(100年)研究成果顯示，現地調查方面：以橫跨集集地表破裂跡的的竹山槽溝及相鄰地區的調查結果為例，近斷層之變形特徵(包括地表的破裂變形帶範圍及變形帶與槽溝剖面中的變形特徵)，基本上是無法由航照或衛星影像判釋得知，雖集集地震距今已10年以上，但部份現地地表仍保留震後的重要地形變化的特徵(如斷層的大致崖高、錯動方向與錯距)。透過槽溝剖面輔以鄰近補充鑽井岩心產狀及試驗報告顯示，上覆變形土層主要為砂礫層、卵礫石層、偶夾黏土層及砂層，對比過去工程鑽井文獻，因研究目的的差異造成對槽溝及地質連續取樣鑽井對岩層描述所用沉積學的描述與一般工程鑽井針對土壤工程分類與力學特性對土壤岩層描述與分類不盡相同，本研究初步嚐試採用土壤力學概念，進行槽溝變形數值模擬模型之建立。 砂箱實驗方面: 鑑於過去對影響斷層跡出現在地表位置的因素之研究，優先探討上覆土層厚度及相對密度這兩個變因，對斷層引致的變形範圍及斷層跡出露地表位置的影響程度。第一年(100年)初步離心機實驗成果顯示，傾角60o的斷層角錯動時，在1g及10g的試驗條件下，地表高程劇烈變化影響範圍約等於覆土層厚度；而相對密度的大小，並不會顯著影響表土層劇烈變形的範圍。 未來整合試驗成果，建立所得之正規化地表頂升量與正規化距斷層尖端出露距離的關係。可以提供不同斷層垂直錯動量與距斷層尖端出露不同距離之地表垂直頂升量的預測。 數值分析方面:第一年研究已分別嘗試分離元素法 (PFC2D程式)、連續體模式之有限差分法(FLAC程式)及邊界元素法進行砂箱成果或現地變形剖面之模擬。研究成果顯示，利用邊界元素法可推求逆斷層觸發地表變形帶中應變與離斷層跡相對位置的關係；有限差分法(FLAC2D)可直接求取逆斷層滑移歷程下，上覆土層應力及應變場分布情形；分離元素法(PFC2D)可觀察不同土層材料及組成(純砂、純黏土、互層)，在不同逆斷層滑移量下，地表變形特性。邊界元素法的優點在計算上效能最高，唯因原理與假設條件限制，無法進行土壤結構互制模擬；連續體模式之有限差分法的優點在已有分析經驗最為豐富，且可以進行土壤－結構互制之分析；而分離元素法的優點在於它最能模擬接近地表土層之粘土或砂土不同的變形行為，對土層中變形剪切帶發育行為較為相符，而且具有模擬斷層多次歷史活動所產生地表附近生長斷層與生長地層特性之潛力。 本年度選擇正斷層(山腳斷層)為主要研究對象，探討斷層兩側岩層變形行為(包含剪切帶發育、遷移及影響參數等)、斷層錯移在近地表可能影響範圍；進行離心機砂箱實驗，探討斷層錯動時，上覆岩土層變形行為及對結構物之影響。並以砂箱實驗配置與實驗所得之材料參數進行數值分析，以測定邊界條件及輸入參數之合理性。並嚐試用不同數值分析方法，建立相同模型，進行不同分析法間相互比較，視分析模擬成效，選用最合宜的模式。再將已驗證之數值分析工具，以全尺度數值分析探討斷層作用時上覆土層變形行為，及其對人工結構物之影響，並提出較合理的安全性探討指標，作為判斷建築物或設施安全性的參考。 <br> Abstract: The determination of theoretical setback widths used to avoid the earthquake hazard due to distortion of ground induced by faulting is deemed as essential as the understanding on how propagation of fault-tip affects the width of deformation zone on the ground of the soil/rock mass, the spatial distribution of strain and stress within the zone, and the foundation response. In addition, modification of the design methods of the foundations by studying the interaction among the fault rupture, geomaterial and foundation is a useful means to mitigate an earthquake hazard associated with soil structure interaction inducing by faulting Currently the approaches to investigate the near ground deformation due to reactivation of active faults can be categorized as (1) field investigation (2) laboratory testing and (3) numerical modeling. In order to break through the limitation of individual approach, field investigation, numerical modeling along with laboratory testing, including sand box experiments under both 1-g and centrifuge conditions (hereinafter referred to as centrifuge testing), will be conducted for characterizing and quantitating deformation induced by faulting in this study. The result of the first-year (2011) study shows that in aspect of field investigation, taking the Chushan trench which straddled the Chi-Chi ground rupture and its neighborhood as an example, the near-fault deformation features (including the widths of zone of fractures and shear zone on the ground and features on the walls of the trench) cannot be possibly recognized on aerial photographs or satellite images; however, some important topographic features (such as rough height of the fault escarpment, the direction of fault slip and the fault separation) formed during the earthquake remains even after more than ten years since the Chi-Chi earthquake. Through a compilation of sedimentary columns of the trench walls, the drilling cores from the nearby sites and the previous report on the soil tests, the deformed cover are composed by layers of sandy gravel and pebble-sized gravel, and alternative clayey and sandy layers. The description and classification of soils are different owning to the purposes of studies. Therefore, the description and classification for the sedimentary columns of the trench walls and geologically wire-line coring samples are quite different from the ones for the engineering drilling which emphasizes geotechnical classification and the mechanical properties of soils. This study has preliminarily tried building numerical models based on the concept of soil mechanics for simulating the features on the walls of the Chushan trench. In aspect of centrifuge testing, the effect of two factors, which are the thickness of the soil cover and its density is firstly explored on the extent of fault-induced deformation and the location of fault trace on the ground in the light of previous studies. The preliminary result of the first-year (2011) centrifuge testing shows that under both centrifuge 1- and 10-g conditions for faulting with a fault angle of 60° the extent of severe change of the soil-top elevation is approximately equal to the thickness of overburden soils; however, the variation on the relative density of soils doesn`t affect much on the extent of severe distortion of the ground. In the future an integrated testing result of relation between normalized ground surface uplift and normalized distance in reference to the location of fault trace on the ground surface will be built to provide a prediction on the vertical component of fault separation and the uplift in any position on the ground surface. In aspect of numerical modeling, a discrete element method (PFC2D software), a finite differential method (FLAC software) and a boundary element method have been tried simulating the deformation in sand boxes and on the walls of trenches during the first year. Using the boundary element method, the relation of strain within a deformation zone induced by thrust faulting and its location in reference to the fault trace can be derived. Using the finite differential method (FLAC software), the distribution of stress within the overburden soil can be derived in any stage of thrusting faulting. Using the discrete element method (PFC2D software), the evolution of deformation within soil cover consists of multiple layers with a different mechanical property (pure sandy soil, pure clayey soil or layered soil) is observable with the increase of fault slip. The merit of boundary element method is its highly computing efficiency; however, its drawback is incapable of analyzing the interaction between soil cover and engineering structures because of its theoretical basis and assumptions. The merit of finite differential method is its abundance of analyses in previous studies and capable of analyzing soil-structure interaction. The merit of discrete element method is that its simulations closely resemble the deformation behaviors of clayey soil or sandy soil of the ground-surface cover and development of shear zone within, in addition to its potential to simulate growth faults and growth strata for multiple earthquake events. The target of this year is pined on the type of normal faults (Shanchiao fault) to investigate deformation behavior of the soil/rock mass straddled the fault trace (including initiation and development of the shear zone, and its affecting factors) and affected extent due to its reactivation. The centrifuge testing will be performed to gain insights on the deformation behavior of the soil/rock mass associated with faulting as well as the response of the foundation. Boundary conditions and parameter values for numerical models will be verified based on the configurations of the centrifuge testing and its resultant parameter values. Thus, full-scale numerical modeling can be carried out for studying the deformation behavior of the soil/rock mass and its effect on engineering structures. In consequence, reasonable safety indices can be suggested as safety assessment of buildings and infrastructure. Taking these different approaches, we anticipate to enhance our understanding on how propagation of fault-tip affects the width of deformation zone on the ground of the soil/rock mass and the spatial distribution of strain and stress within the zone. These analyses shall help us determine theoretical setback widths used to avoid the earthquake hazard due to distortion of ground induced by faulting. In addition, modification of the design methods of the foundations by studying the interaction among the fault rupture, geomaterial and foundation is another useful means to mitigate an earthquake hazard associated with soil-structure interaction inducing by faulting.活動斷層近地表變形離心機砂箱實驗土壤結構互制退縮距離減災工法active faultsnear-ground deformationcentrifuge testingsoil -structure interactionsetback widthhazard mitigation strategy