場址效應為地震危害度評估重要影響因素之一。地表下30m內之平均剪力波速 (Vs30) 則是一個評估場址效應的重要參數，於在強地動預估式、地盤分類標準、液化潛能評估等，均有廣泛的應用。過去有許多探討區域性Vs30分布圖測繪之研究，主要方法係根據土壤物性與剪力波速轉換式求得區域鑽孔處之剪力波速與Vs30，再根據各鑽孔位置Vs30值，利用地質統計的方法對Vs30進行空間內插，但上述方法未能直接將地質及物性參數的空間變異性納入考慮。 本研究利用馬可夫隨機場 (Markov random field, MRF) 方法進行地質模型隨機場模擬，並統計各地層物性參數分布特性，以台北盆地測試區進行模擬，再利用前人土壤物性與剪力波速轉換式，計算剪力波速及Vs30，並量化Vs30之不確定性 (包含地質模型與物性參數) ，本研究亦探討有無考慮台北盆地基盤與礫石層空間分布對Vs30計算結果之影響。 研究結果顯示，未考慮台北盆地基盤的空間分布，在分析台北盆地北緣處Vs30時，相較於考慮台北盆地基盤的空間分布，低估了超過40%以上，顯示地質知識對地質模型建模及後續分析之重要性。本研究亦針對地質模型不確定性對Vs30不確定性之影響進行分析，結果顯示當地質模型不確定性越大的地方，其Vs30不確定性也越高，因此未來相關人員在計算Vs30時，地質模型與地質模型不確定性為需要列入考慮之重要因素。 ;Site effect is one of the important factors in assessing seismic hazards. The average shear wave velocity within the top 30 meters from the ground surface (Vs30) is a significant parameter used to evaluate site effect. It is widely applied in strong ground motion prediction equations, site classification standards, liquefaction potential assessment and etc. Previous studies have extensively explored the mapping of regional Vs30 distributions. The main approach involves determining the shear wave velocity at regional borehole locations based on the transformation functions of soil properties and shear wave velocity, and then interpolating Vs30 values through geological statistical methods. However, these methods do not directly account for the spatial variability of geology and soil physical properties. This study employs the Markov random field (MRF) method to simulate geological models as stochastic fields and statistically analyze the distribution characteristics of various lithological properties. The study area is the Taipei Basin, where simulations are conducted. Subsequently, utilizing established relationships between soil physical properties and shear wave velocity, this study calculates shear wave velocities and Vs30 values, quantifying the uncertainty associated with Vs30(including geological model and soil physical property uncertainties). This study also discusses the impact of considering the spatial distribution of the basement in the Taipei Basin on the calculation of Vs30. The results demonstrate that neglecting the spatial distribution of the basement in the Taipei Basin underestimates Vs30 by over 40% when analyzing the northern margin of the basin compared to considering the spatial distribution of the basement. This highlights the importance of incorporating geological knowledge in geological modeling and subsequent analyses. Furthermore, this study analyzes the influence of geological model uncertainty on the uncertainty of Vs30. The findings indicate that locations with greater geological model uncertainty correspond to higher uncertainty in Vs30. Therefore, future practitioners involved in Vs30 calculations should consider both the geological model and its associated uncertainty as crucial factors to be taken into account.
