芮氏規模是根據地震波紀錄之振幅來計算，各測站場址條件的差異會對振幅造成不等的放大效應而影響芮氏規模的計算。為了準確表示地震事件的芮氏規模，利用了從多個測站獲得的平均規模值。dML代表地震事件的規模與各個測站所記錄到的規模之間的差異。目前，台灣中央氣象署在決定芮氏規模時並未應用dML修正。我們觀察到dML值的範圍從-0.71到0.92不等。當應用dML修正時，事件規模的標準差可以從0.36減少至0.22，顯示出規模更一致且更加可靠。對於僅有震央附近測站記錄到的較小規模地震，或是區域型的觀測網，可以更顯出dML修正應用的重要性。在這種情況下，所有測站可能都處於類似的地質條件（例如，都位於硬岩或軟土上），這種同質性會導致規模估計的系統性偏差。因此，使用dML修正有助於將這些測站調整到統一的地震場址條件來減少這些偏差。本研究提議利用已知的場址參數來計算dML。在這些參數中，VS30在本研究中與dML顯示出最強的相關性。此外，放置在井下的地震儀大大減少了來自地表噪訊的干擾，進而獲得高品質的地動訊號。由於地表與井下地震儀之間的地震波速度差異，導致計算出的芮氏規模結果（DeltaML）有所不同。使用井下記錄來決定芮氏規模時，由於其振幅較地表記錄小，會導致低估。本研究的目的是探討dML與場地參數之間的相關性，並評估dML修正對芮氏規模的影響。此外，我們還旨在評估由於場址放大效應導致的地表和井下記錄的芮氏規模計算差異（DeltaML）。;The local magnitude (ML) of an earthquake, calculated from the amplitude of seismic waves, can vary depending on site conditions at different recording stations. These site conditions cause different amplification effects on the recorded seismic waves, impacting the calculated ML values. To represent the ML of an earthquake event, the average magnitude value obtained from many stations is utilized. The term “dML” signifies the difference between the event magnitude and individual station magnitudes. The use of dML corrections is not currently applied by the Central Weather Administration (CWA) of Taiwan in their magnitude determinations. The observed dML values ranged widely from -0.71 to 0.92 magnitude units. When dML corrections are applied, the standard deviation of the event magnitude decreases from 0.36 to 0.22 magnitude units, indicating a more consistent and reliable magnitude determination. The application of dML corrections is especially critical for earthquakes of small magnitudes. These events are typically recorded only by stations near the epicenter, which might all be situated on similar ground conditions (e.g., all on hard rock or all on soft soil). This homogeneity can lead to systematic biases in magnitude estimates. Therefore, using dML corrections helps mitigate these biases by adjusting those stations to a united seismic site condition. This study proposes utilizing known site parameters to calculate dML. Among these parameters, VS30 showed the strongest correlation with dML in this study. Moreover, seismic instruments placed downhole substantially minimize interference from surface noise, enabling the acquisition of high-quality ground motion signals. The difference in seismic wave velocities between the layers at the surface and downhole seismometers leads to divergence in the calculated local magnitude results (DeltaML). When employing downhole recordings for local magnitude determination, an underestimation occurs due to their smaller amplitudes compared to surface recordings. The goal of this study is to explore the correlation between dML and site parameters and to evaluate the impact of dML corrections on local magnitude determinations. Additionally, we aim to evaluate differences (DeltaML) in local magnitude calculations resulting from ground and downhole recordings due to site amplification effects.
