台灣東部縱谷北端區域的地震構造特性

Abstract

台灣位於歐亞板塊與菲律賓海板塊的碰撞帶上，位於台灣東部，介於中央山脈與海岸山脈間的縱谷縫合帶正是兩個板塊縫合帶的位置，而在台灣東部外海呈東西向的琉球隱沒系統，從日本西南外海一路向西南延伸交會至縱谷縫合帶北端，因此在縱谷北端區域有頻繁的地震活動與複雜的地質構造。雖然此區域先前已經有許多關於大地構造的研究，但詳細的地震構造研究仍然很少。近年來陸續有重新定位後的地震資料被發表出來，本研究試圖利用這些地震資料進一步探討此區域地底下構造的分布，並與地表的一些構造特徵去進行比對以及探討，建構出此區域的構造模型。 由1990年至2010年間的地震震源分布與震源機制解結果發現，在縱谷北端區域的底下十公里處有一群地震的密集帶，此地震群一直往北北東方向延伸至北方的立霧溪沖積扇附近，而在美崙台地西方山區存在另一群地震群，此地震群明顯呈現向東傾斜的線性分布，此線性分布從深部十公里往西連接至地表，其中具有代表性的震源機制解顯示以逆衝機制為主，故本研究推斷此地震群可能代表一條向東傾的逆衝構造存在於中央山脈山區附近，此構造與1951 年的花蓮地震其發震的米崙斷層為兩個不一樣的斷層。利用同樣的分析方法，本研究在立霧溪沖積扇附近也發現另一群向南傾斜的地震序列，本研究認為這群地震序列代表著，在立霧溪沖積扇附近有另一條東西走向並向南傾的逆衝構造，而且與深十公里的地震群連結成一個鏟形的構造系統。根據以上的地震資料分析結果，本研究針對構造可能出露的位置進行地表的調查，由調查結果發現，位於立霧溪口的沖積扇扇面呈現南北不對稱的分佈，另外，也在美崙台地西方山區發現不尋常的陡崖地形，諸此現象，本研究推測這些現象可能跟本研究發現的構造有所關聯。再比對此區域的地質圖後，本研究認為這兩條構造可能依循岩性的交界相連形成一個弧形的逆衝構造系統，同樣地，美崙台地周圍可能也有同樣的弧形斷層系統的活動，驅使美崙台地的抬升。而位於縱谷北端區域的弧形構造系統，可能代表的是縱谷沉積物在出了縱谷後並沒有跟著海岸山脈以及菲律賓海板塊隱沒至歐亞板塊之下，反而是向北以及向西逆衝上來，而美崙台地的形成代表著另一個相同機制的小型系統發育。

The island of Taiwan is located at the collisional boundary between the Philippine Sea and the Eurasian plates. In eastern Taiwan, the Longitudinal Valley between the Central Range and the Coastal Range is generally considered as the suture zone. The Ryukyu subduction system, on the other hand, extends southwestward from offshore eastern Taiwan, and intersects with the northernmost Longitudinal Valley suture. As a result, this area is characterized by frequent earthquakes and complex geological structures. Although there have been several tectonic investigations in this area, detailed knowledge of seismotectonic characteristics of this complicated region is still very limited. Recently, we have obtained relocated earthquake data by combining records from seismic stations of the Japan Meteorological Agency (JMA) and data from Taiwan Central Weather Bureau Seismic Network (CWBSN) and Taiwan Strong Motion Instrumentation Program (TSMIP). Therefore, we attempted to analyze this area in detail using these high resolution seismic data, together with tectonic geomorphic investigations. The distribution of seismicity shows that many earthquakes occurred at about 10km deep in this area, and a linear seismicity cluster extends to the surface west of the Hualien City. We infer that this seismicity cluster represents a structure that crops out within the eastern flank of the Central Range. This structure would be different from the Milun fault, which is a well-known structure in the Hualien area and ruptured during an M7.3 earthquake in October 1951. Another major feature in this area is a listric cluster of seismicity that appears to extend to the surface near the Liwu River mouth, north of the Hualien City. This cluster likely represents an E-W striking structure that dips to the south. Based on the focal mechanisms of several recent moderate earthquakes that occurred within these seismic clusters, we suggest that these two structures are reverse faults. Our field survey revealed that the fan terraces at the Liwu River mouth appears to be asymmetrical, with its southern part higher than its northern part. This phenomenon may have been produced by the activity of the south dipping reverse fault near the Liwu River mouth. Summarizing above, we found two previously unidentified reverse faults in the northernmost part of the Longitudinal Valley suture: the E-W striking structure near the Liwu river mouth and the N-S striking structure within the eastern flank of the Central Range. These two structures may be connected. The Milun fault and an unknown E-W striking reverse fault north of Milun tableland may be a branching system similar to this system. In the northernmost part of the Longitudinal Valley suture, we propose that as the Coastal Range subducts northward beneath the Eurasian plate with the Philippine Sea plate, the shallow sediments of the Longitudinal Valley, being a buoyant block, do not subduct, but overthrust northward and westward instead.