| 摘要: | 自 2017 年 1 月開始, 土耳其西海岸的比加半島, 恰納卡萊地區的圖茲拉地熱田附近發生了群發性地震。這是自 1970 年開始的連續地震監測以來該地區的第一次記錄到的群發性地震。由於該群發性地震的觸發機制尚有爭議，本研究旨在調查2017年艾瓦哲克 (恰納卡萊) 地震群體是否可能由圖茲拉地熱田的活動觸發。 在這方面，我們使用了 2011 年初至 2018 年底來自希臘統一地震台網 (HUSN) 和波加濟奇大學坎迪利天文台和地震研究所 (KOERI) 地震台網的地震數據。波形數據是從歐洲綜合數據檔案館 (EIDA) 獲得的。 使用非線性概率算法 NLLOC 獲得了 1602 個事件的絕對位置。使用雙差技術確定了 1558 個事件的精確相對位置。 重新定位的事件似乎描繪出一種類似於李斯特性斷層的構造，具有北西-東南向趨勢，深度在 5 km 至 〜20 km 之間。 由地震震源的分佈可知此斷層的再活動發生在群發性地震期間。 我們測量了 Vp/Vs 比率，估計了剪力波分裂參數 (時間差，快S波極化方向), 並計算了事件發生前、中、後的b值。 此外，我們還估計了地震期內注入持續時間 (t0) 的擴散率值 (D)。在群發性地震出現之前, 我們觀察到低b值 (〜0.79) 和高 Vp/Vs 比 (〜1.69), 並且時間差和快S波極化方向沒有明顯變化。地震開始後不久, Vp/Vs 比似乎穩定 (〜1.67), 時間差從 〜0.2 增加到0.44 s。 此時, D值高 (145 m2/s), t0最小。 在此期間, 快S波極化方向顯示出強烈的不穩定性。 到地震結束時, Vp/Vs比和時間差分別下降至〜1.53和0.11 s, 而b值較高 (1.34), D值減小至 35 m2/s (t0較)。 我們所觀察到的低b值說明在活斷層附近有一段高應力時期。再來是較高的Vp/Vs比, 且在群發性地震開始之後時間差的上升, 很有可能是由於充滿流體的岩石中的裂縫密度增加所致。 在群發性地震期間, 所觀察到的快S波極化方向的不穩定性可能與活斷層的高孔隙流體壓力有關；其中，流體的擴散可能具有關鍵作用。 這可能與研究區域附近的TGF地區的液體注入活動有關。;In the western coast of Turkey, an earthquake swarm occurred near the Tuzla Geothermal Field in Çanakkale region, Biga Peninsula, that started in January 2017. This was the first recorded swarm in the region since 1970, when the continuous seismic monitoring began. As the triggering mechanism of the swarm is still debatable, this study aims to investigate whether the 2017 Ayvacik (Çanakkale) earthquake swarm was potentially triggered by activities in the Tuzla Geothermal Field. In this respect, we used the earthquake data from both the Hellenic Unified Seismological Network (HUSN) and Bogazici University Kandilli Observatory and Earthquake Research Institute (KOERI) seismic network from early 2011 to the end of 2018. The waveform data was requested from the European Integrated Data Archive (EIDA). Absolute locations of 1602 events were obtained by using the nonlinear probabilistic algorithm NLLOC. Precise relative locations were determined for 1558 events using the double-difference technique. The relocated events appear to delineate a structure that resembles a listric fault with NW-SE trending and depths of 5 km to ~20 km. The distribution of the earthquake hypocenters suggests the reactivation of the fault during the swarm. We measured the Vp/Vs ratio, estimated shear wave splitting parameters (delay time , fast direction ), and calculated the b-value in the period before, during, and after the swarm. Further, we also estimated the diffusivity value (D) with the duration of injection (t0) in the swarm period. Before the swarm, we observed low b-value (~0.79) with high Vp/Vs ratio (~1.69) and no significant changes in delay time and fast direction. Soon after the swarm begun, Vp/Vs ratio seemed to be stable (~1.67) and delay time increased from ~0.2 to 0.44 s. At this time, the D value was high (145 m2/s) with the minimum t0. Fast direction showed strong instability during this period. By the end of the swarm, Vp/Vs ratio and delay time dropped down to ~1.53 and 0.11 s, respectively, while b-value was high (1.34) and D value was decreasing to 35 m2/s with longer t0. The observed low b-value before the swarm indicates a period of high stress near the activated fault. This was followed by high Vp/Vs ratio and an increase in delay time after the swarm begun, probably due to the increase of crack density within the fluid-filled rocks. The instability of the observed fast direction during the swarm can be associated with high pore-fluid pressure along the activated fault. We suggest that the diffusion of fluids possibly played a key role during the period of the swarm. This can potentially be associated with fluid injection activities at TGF site near the study area. |
