利用1993年至2009年台灣強震網(TSMIP)波形資料逆推點源地震矩

Abstract

震源參數推導為評估地震潛能的重要依據，它除了可顯示地震斷層的型態外，更可反映區域的應力(應變)狀態。利用寬頻地震網資料逆推點源地震矩(位移場波形逆推)為目前最常計算震源參數方法之一，然而，若能將波形反演的技術應用在其他波形上(如：加速度強震網)，對孕震帶的了解亦是有益的。本研究利用1993年至2009年台灣強震網資料(TSMIP)逆推點源地震矩(速度場波形逆推)，並分析台灣地震震源機制序列。我們測試了不同的速度模型和濾波頻寬，成功地反演107個規模(Mw)大於4.8的地震，同時將其結果與區域地震網(BATS)及全球地震網(GCMT)作比較，具有可信度。但是對於規模(Mw)為5.0或更小的地震，我們所計算的規模仍有高估0.5至1.0誤差值。此外，除利用速度波形逆推點源地震矩外，我們也測試了加速度波形逆推，其結果與速度波形計算相同。此研究中我們特別討論臺灣梅山區域孕震帶特性，發現其具有複雜右移走向滑移斷層系統。期望藉由此研究所討論的方法亦可應用在其他強震網資料，增加區域地震震源機制目錄，協助我們更瞭解某區域地震斷層型態及應力隨時間變化的特性。

Deriving earthquake source parameters is important for the purpose of seismic hazard mitigation. Moment tensors are among the most important earthquake source parameters for seismologists, and are now routinely derived using modern broadband seismic networks around the world. However, it is advantageous to apply similar waveform inversion techniques to other available waveforms, including accelograms recorded by strong motion stations. We inverted the source mechanisms of the Taiwan earthquake sequence between 1993 and 2009 by applying the regional moment tensor inversion method using data from the Taiwan Strong Motion Instrumentation Program (TSMIP) accelerograph stations. Through testing different velocity models and filter passbands, we were able to successfully derive moment tensor solutions for 107 MW>= 4.8 earthquakes. The solutions agree well with other available high quality moment tensor catalogs derived with local and global networks. However, for MW= 5.0 or smaller events, we consistently over estimate the moment magnitudes by 0.5 to 1.0. We have tested accelograms, and velocity waveforms integrated from accelograms, for the inversions, and found the results are similar. We use part of the catalogs to study seismogenic structures of Meishan, Taiwan region, and found that it is dominated by complex right-lateral strike-slip fault system. The procedure developed from this study may be applied to other strong motion datasets to compliment or fill the gaps in catalogs from regional broadband networks and teleseismic networks.