TCDP井下地震儀—微地震之觀測與震源特性分析

TCDP Borehole Seismometers Array – Microearthquake Observations and Seismic Source Characteristics Investigation

林彥宇(Yen-Yu Lin)

指導教授 ： 馬國鳳

國立中央大學/地球科學學院/地球科學系/博士(2014年)

摘要

台灣車籠埔鑽探計畫（Taiwan Chelungpu Drilling Project）之井下地震儀（簡稱TCDPBHS）安裝於1999年集集地震最大同震變形（12 m）之車籠埔斷層北段。井下地震儀所記錄之高品質地震波形，使我們能夠對於研究區域產生之微地震、群組地震的活動情形，及其震源特性進行深入的分析與探討。震源尺度關係一直為地震學家所關心的議題－小地震與大地震的行為是否存在相通性？由於微地震破裂行為暗示地震的發震特性，了解微地震破裂物理，可幫助我們進而探討較大地震的破裂行為。因微地震波形紀錄的訊號雜訊比較低，較難為地表地震儀所記錄，因此需仰賴高品質的井下地震儀，如TCDPBHS。
本研究分析2006年11月至2007年TCDPBHS之紀錄，共270筆微地震（M < 2.0）之分佈發現，集集地震車籠埔斷層北段主要破裂帶上，幾乎沒有任何地震發生。此觀測顯示該區域在集集地震發生時，已將斷層上累積的剪應力完全釋放，故在間震期呈現被鎖定的狀態。區域性微地震（Δtsp < 2 s）分佈在10~15 km深的水平地震帶上，此地震帶可能為臺灣西部褶皺逆衝帶的滑脫面，且位於三義埔里地震帶的北部。其孕震機制可能為滑脫面之形變結果，亦或是為液體富集形成區域弱面，而產生許多微地震。本研究利用交互相關法，搜尋2006年11月至2009年9月共35個月的地震紀錄，共發現287個群組地震。這些群組地震之發震位置與規模大小，均與背景地震活動類似，且其類型多為突發型群組地震－即在短時間內發生數筆地震之後歸於平靜。群組地震之發震間距不規則，顯示本研究區域之孕震機制，與擁有規則性發震間距重複地震的美國加州潛移帶（creeping zone）有很大的差異。群組地震之發震間距有兩個峰值，一為分鐘至小時的尺度（10^-1~10^0 hr），另一則為年的尺度（10^3.5 hr）。此外，本研究也觀測到同一群組地震中，地震規模不同，但P波波形相似的特殊的行為，顯示這些群組地震似乎不遵守地震自我相似的尺度關係。
為了探討微地震的震源尺度關係，本研究利用ω^2震源模型及Q對頻率無關的模型假設，對地震觀測頻譜進行擬合分析並求取震源參數。對242筆規模Mw 0.0~2.0之微地震進行尺度分析發現，應力降與地震矩有正相關的情形，視應力與地震矩也有正相關。為排除路徑效應的干擾，本文利用群組地震其路徑效應相似的特性，使用經驗格林函數法與理論路徑衰減模型修正法，求得地震之震源時間函數Tw，結果發現，無論使用何種方法，Tw均不遵守地震自我相似性的Tw ∝ M0^3。經驗格林函數法及理論路徑衰減模型修正法結果均顯示，無論地震規模大小，Tw為常數。本研究結果為微地震在群組地震中不遵守自我相似性的證據。
TCDPBHS紀錄中發現地震群組中，發震間距為分鐘至小時尺度的地震群組，似乎很難使用應力重新累積的理論解釋，此暗示著特殊的孕震構造。因此，本研究嘗試使用Rate-and-state friction law，模擬短時間重複發震之群組地震行為的極端例子，發震間距為4 s的群組地震。由二維模擬潛移的斷層面上有三個速度弱化（velocity weakening）區域的結果發現，如果增加三個速度弱化的區域之間的破裂障礙強度，並且使用Slip law，破裂障礙間距D約為2 m，可以達到符合短暫重複發震群組地震的觀測，此結果暗示在本研究區域的構造可能並非單一斷層。由三維模擬結果可以證實，若本研究區域液體富集且孔隙液壓變化大（超越50%正向應力），則可模擬出短暫重複發震群組地震行為。

關鍵字

微地震 ； 井下地震儀 ； 臺灣車籠埔鑽探計畫 ； 微地震尺度分析 ； 群組地震

並列摘要

Microearthquakes with magnitude down to 0.3 were detected by the Taiwan Chelungpu fault Drilling Project Borehole Seismometers (TCDPBHS). Despite the large co-seismic slip of 12 m at the drill site during the 1999 Chi-Chi earthquake, our studies show little seismicity near the TCDPBHS drill site 6 years after the Chi-Chi main shock. The microearthquakes clustered at a depth of 9-12 km, where the Chelungpu thrust fault turns from a 30-degree dipping into the horizontal decollement of the Taiwan fold-and-thrust tectonic structure. This observation suggests that the thrust belt above the decollement is locked during this interseismic period. A cross-correlation (CC) was made to the identified microeartqhuakes, 287 clusters were discovered for CC > 0.8. These clusters are mostly burst-type, as which occur in a short time period. The examination of the interseismic time interval within the clusters reveals two significant peaks in time intervals, as if 10^-1~10^0 hr (minutes to hour) and 10^3.5 hrs (year). For the similarity in waveforms, we observed unique earthquake clusters, which have near constant P- and S-wave durations regardless the magnitudes of events within the clusters.
Further studies on source scaling from the investigation of source parameters of 242 microearthquakes, we used SH-wave spectra by fitting ω^2-shaped Brune source spectra with a frequency-independent Q model. We find that the static stress drop increases significantly with increasing seismic moment, and also the similar positive feature for the apparent stress scaling with seismic moment. To avoid the contamination from attenuation, we further analysis the data for events within clusters to remove the path effect of the events in the clusters. The Empirical Green’s function (EGF) and Futterman Q correction methods are utilized. The derived source time function, Tw, from the both methods showed similar feature as the Tw does not follow the earthquake self-similarity scaling of Tw ∝ M0^3. The results obtained a nearly constant Tw with moment. Our observations provide a direct evidence of an earthquake non-self-similarity behavior for events ranging from Mw 0.0 - 2.0 within the cluster.
To explore the potential mechanisms for rapid event recurrence (sec-hour), we developed a model of repeating earthquakes based on rate-and-state friction. In the model, several small patches governed by steady-state velocity-weakening friction are located in a close proximity to each other and surrounded by a larger velocity-strengthening region with a background loading slip rate. Our modeling results indicate that the rapid triggering does not occur in the long-term response with typical lab parameters. However, a model with stronger barriers and slip law of state variable evolution matches the observations, suggesting high heterogeneity of the fault zone. In another way, we develop a model with pore pressure variation indicating fluid flows. The modeling results suggest high pore pressure change ( >50% normal stress) with artificial time drifting may also trigger the rapid triggered seismic clusters.

並列關鍵字

Microearthquake ； Borehole seismometer ； TCDP ； Seismic source scaling ； Earthquake cluster

參考文獻

Abercrombie, R. E., 1995. Earthquake source scaling relationships from -1 to 5 ML using seismograms recorded at 2.5-km depth, J. Geophys. Res., 100(12), 24015-24036.

Abercrombie, R. E. & Rice, J. R., 2005. Can observations of earthquake scaling constrain slip weakening?, Geophys. J. Int., 162, 406-424.

Ake, J., Mahrer, K., O’Connel, D. & Block, L., 2005. Deep-Injection and Closely Monitored Induced Seismicity at Paradox Valley, Colorado, Bull. Seism. Soc. Am., 95(2), 664-683.

Aki, K., 1967. Scaling law of seismic spectrum, J. Geophys. Res., 72, 1217-1231.

Ampuero, J. & Rubin, A. M., 2008. Earthquake nucleation on rate-and-state faults : aging and slip laws, J. Geophys. Res., 113, B01302, doi:10.1029/2007JB005082.