大地震破裂方向性研究 研 究 生：張若磐 指導教授：王乾盈 博士 摘 要 大地震的破裂方向性研究提供了對地震基本斷層參數的了解， 這 些參數間的關係(稱為震源尺度律)可描述大地震破裂的一些基本物理特 性。本研究利用表面波走時差的方法來決定大地震的破裂方向性，即利用主震 及其附近參考地震的表面波走時差來評估各測站的震源歷時，並進行破裂方向性 分析，但是，若大地震附近並無合適的參考地震時，此法並不適用，因此，本研 究進一步提出一個新方法來解決參考地震的問題，就是利用全球表面波相速分 布圖來取代參考地震，在相速圖上計算從地震至測站的走時可視為參考地震的表 面波走時，也就是合成格林函數的觀念。藉此可以快速及有效的決定大地震的破 裂方向性，克服無真實參考地震的缺點。 本研究共分析由1999 至2008 年震矩規模大於7.6 的地震，共計8 個，包括 6 個逆斷層及2 個走向平移斷層，採用震央距離30°-90°的表面波資料，逆斷層型 態的地震利用雷利波分析，而走向平移斷層型態的地震則利用洛夫波分析。除 2004 年蘇門達臘地震採用週期150 秒雷利波分析外，其餘皆利用週期100 秒的雷 利波或洛夫波。結果顯示本研究所提的分析方法確實可行，對一些無合適參考地 震的大地震(如2001 年崑崙地震、2008 年四川地震)都能得到合理的破裂方向性 分析。其次，整體而言，震源歷時及破裂長度隨地震矩(震矩規模)增加而增加， 其中以2004 年蘇門達臘地震(Mw 9.0-9.3)具有最長的震源歷時及破裂長度；以同 規模地震而言，走向平移斷層較逆斷層有較長的破裂長度及震源歷時。再則，破 裂方位分析能有效判斷走向平移斷層型態地震的斷層面，然對逆斷層型態的地震 則無法全部判讀。因整個震源歷時包括破裂時間和震源上揚時間，故由整個震源 歷時易低估破裂速度，透過表面波頻譜節點週期的分析可評估震源上揚時間，因 此可得到較合理的破裂速度。本研究所提方法的優點在於不需要主震附近的參考 地震，因此，當大地震發生後，可快速有效的決定其破裂方向性和斷層參數，而 且也適合用來重新評估一些歷史地震的斷層參數，如早期的1960 年智利地震和 1964 年阿拉斯加地震等。 Rupture Directivity Analysis for Large Earthquakes Postgraduate：Jo-Pan Chang Adviser：Dr. Chien-Ying Wang Abstract Rupture directivity analysis for large earthquakes can provide some basic fault parameters. Relationships between these fault parameters, the so-called scaling law, lead to understand the physical properties of earthquakes. Many methods and seismic data can investigate the rupture directivity of an earthquake. In this study, we use the differences of surface-wave travel time between the mainshock and reference earthquake to determine the fault parameters for large earthquakes. Above mentioned, constraint on such work is that there must be reference earthquakes in the vicinity of the mainshock. Here, we propose a new method to analyze the rupture directivity of an earthquake without using reference earthquakes. That is, we calculate the surface-wave travel time using the global surface-wave phase-velocity maps to be the travel times from reference earthquakes. The concept is similar to the Green’s function. In this study, 8 large earthquakes occurring from 1999 to 2008 with Mw 7.6-9.0(9.3) are analyzed using surface waves with epicentral distances between 30° and 90°. We use the Rayleigh waves to perform the rupture directivity analysis for trust-type earthquakes; on the contrary, using the Love waves to analyze the rupture directivity for strike-slip-type earthquakes. Results show that the proposed method is valid for the rupture directivity analysis of a large earthquake. On the whole, the source duration and rupture length increase with seismic moment (or Mw). Among these analyzed earthquakes, the 2004 Sumatra-Andaman earthquake (Mw 9.0-9.3) has the longest source duration and rupture length. Besides, earthquakes with strike-slip-type mechanism have relatively larger source duration and rupture length than those with thrust-type mechanism. The optimal rupture azimuth can efficiently provide the judgment on which is the fault plane from the beach ball for the strike-slip earthquakes. However, it is ambiguous to judge the fault plane for the trust-type earthquakes. The rupture velocity is easily underestimated from the whole source duration (including the rupture time and rise time). By investigating the period of spectral node, we can determine the rise time, and then derive the reasonable rupture velocity. The advantage of our proposed method is that it is not in want of the reference earthquakes and can quickly and efficiently determine the fault parameters of large earthquakes from rupture directivity analysis. Also, the proposed method is appropriate to re-examine some historical earthquakes, such as the 1960 Chile earthquake and 1964 Alaska earthquake.
