本研究強調了平均剪力波速（VS30，指深度至30公尺的時間平均剪力波速）在理解地震事件中場址放大效應方面的重要角色，並將其應用於印尼西蘇拉威西馬穆朱市的都市尺度機率式地震危害評估（PSHA）。VS30 是透過現場量測，包括多波道表面波分析法（MASW）與單站頻譜比法（HVSR）所獲得。MASW數據用於獲取相速度的頻散曲線，而HVSR則是利用微地動資料計算。之後再根據頻散曲線和/或HVSR進行速度剖面的逆推。針對小於30公尺的剪力波速度剖面，採用外推法估算其VS30。 進行機率式地震危害評估（PSHA）需對地震活動有全面理解。傳統PSHA假設地震事件彼此獨立，因此，本研究整合國際地震中心（ISC）1976年至2009年與印尼氣象、氣候暨地球物理局（BMKG）2009年至2023年的地震目錄，並透過去群集程序剔除前震與餘震。在考慮規模完備性之後，僅納入規模Mw ≥ 4.6的地震事件進行分析。 為了更準確地表達淺層地殼地震活動，本研究提出區域震源模型與平滑模型的組合，以降低認知性不確定性。在評估斷層破裂機率時，並考量自上次破裂以來的時間因素，本研究採用了布朗運動通過時間（BPT）模型。地動預估方程式（GMPE）則選用NGA-West2模型，並將其估算結果與2021年發生於研究區域的Mw 6.2馬穆朱地震的RotD50觀測數據進行比較。接著，依據歐氏距離排名（EDR）指標對GMPE進行權重排序。 最後，綜合由地形坡度、地貌分類以及現地量測所獲得的VS30數據，建立都市尺度的地震危害地圖，提供工程師與決策者用於加強地震災害緩減策略的參考依據。;This study highlights the critical role of time-average shear wave velocity to 30 meters depth (VS30) in understanding site amplification during seismic events, applying it to urban-scale probabilistic seismic hazard assessment (PSHA) for Mamuju, West Sulawesi, Indonesia. VS30 was determined through field measurements using multichannel analysis of surface waves (MASW) and horizontal-to-vertical spectral ratio (HVSR). Dispersion curves of phase velocity have been derived from MASW data, while HVSR curves have been computed using microtremor data. Inversion for VS profiles was performed using dispersion curves and/or HVSR. Extrapolation of VS30 was applied for several VS profiles less than 30 meters. Conducting a PSHA requires a thorough understanding of seismic activity. A traditional PSHA rests on the assumption that earthquakes are independent of each other; therefore, combination of seismic catalogs from International Seismological Centre (1976-2009) and Indonesian Agency for Meteorological, Climatological and Geophysics (2009-2023) were initially declustered to remove foreshocks and aftershocks. Considering the magnitude of completeness, only events with Mw ≥ 4.6 were included in the analysis. To represent shallow crustal seismic activity, this study proposed a combination of an area source model and a smoothing model to reduce epistemic uncertainty. To accurately estimate fault rupture probability while accounting for the time elapsed since the last rupture, this study employed the Brownian passage time (BPT) model. Ground motion prediction equations (GMPEs) from NGA-West2 were examined by comparing the estimated ground shaking with the RotD50 observations of the 2021 Mw 6.2 Mamuju Earthquake that occurred in the study region. GMPEs were then weighted based on the Euclidean distance-based ranking (EDR) index. VS30 maps derived from topographic slope, geomorphology, and field measurements were integrated to develop a city-scale hazard map, providing outcomes that engineers and stakeholders can use to improve seismic hazard mitigation strategies.
