淺層 S 波速度構造探討與近場強地動模擬

Estimation of Shallow S-wave Velocity Structure and Near-field Strong Ground Motion Simulation

吳澄峰(Cheng-Feng Wu)

指導教授 ： 黃蕙珠

國立中正大學/理學院/地球與環境科學系地震研究所/博士(2012年)

摘要

本研究旨在利用微地動陣列量測的方式及接收函數的估算，獲得不同地區的淺層S波速度構造及其岩盤之分佈，並藉由井測及地球物理資料的比對，確認所得速度構造之可信度，最後將速度構造結果應用於地震動的模擬。本文之研究成果，可綜合如下：
(1)根據TCDP井測結果，於深度597～1705 m範圍內之平均S波速度為1.860 km/s，而本研究利用微地動陣列量測所得結果於深度585～1710 m之平均S波速度為1.899 km/s，二者結果相近。此外，根據反射震測結果，錦水頁岩層、車籠埔斷層面及三義斷層之深度分別介於900～1200 m、1100 m及1800 m；而本研究之結果，其深度分別介於855～1395 m、1125 m及1755～1800 m。
(2)在ESG2006之盲測試中，有關理論微地動波形的分析上，本研究所得結果與主辦單位所假設之地層S波速度構造相當接近。而在實際觀測資料之分析，本研究逆推結果與主辦單位所提供之鑽井資料(300 m)，在淺地層200 m內結果相近。由實例分析可知本研究所得結果準確性很高。
(3)根據埔里地區16個微地動陣列測點之分析結果，若假設此地區第三紀基盤之S波速度為2000 m/s，則第四紀沈積層的厚度約為300～870 m，此結果與反射震測結果(約為300～600 m)相似，且二者之基盤深度變化趨勢亦相近，皆由盆地邊緣向盆地中心逐漸增加。
(4)根據嘉義地區46個微地動陣列測點之分析結果，若此地區之岩盤S波速度為1500 m/s，則嘉義地區之沖積層厚度約560～1400 m，且由東向西漸增。所得之三維地下速度構造顯示，當VS小於1500 m/s時，地層之厚度由西向東遞減，當VS大於1500 m/s時，地層之厚度則由西向東遞增，此結果可能與嘉義地區之沉積環境有關。另外，本研究亦與地電阻資料做比對，於深度40 m及250 m，此地區之西邊存在較低電阻率，且S波速度明顯較低，而東邊則具較高之電阻率，且S波速度也相對較高。
(5)本研究利用接收函數法分析花蓮地區16個強震站資料，進而逆推淺層速度構造，並與井測資料做比對，於大漢技術學院之淺地層200 m內，接收函數法所得速度構造與修正後之井測資料及微地動陣列量測之比對結果良好；於榮民大理石廠測站，接收函數法與微地動陣列量測所得結果相近，但與井測資料仍有若干差異。於淺層速度的側向變化：當深度為200 m及600 m，位在米崙礫石層上方之測站速度較高。若假設此地區岩盤之S波速度為1000 m/s，則該區域之沖積層厚度約為130～660 m，且沖積層的厚度由東向西遞增。
(6)利用接收函數法分析台北盆地及嘉義地區之岩盤深度變化，探討強震站下方因速度不連續所產生之轉型波Ps與P波之時間差(Ps-P)，可推求岩盤的深度分佈。台北盆地之Ps-P時間差為0.295～0.925 sec，且由東南向西北漸增，變化趨勢亦與當地岩盤形貌相符。另外，嘉義地區之Ps-P時間差為1.010～1.690 sec，且由東向西遞增，此結果與嘉義地區之微地動陣列量測結果相吻合，即沖積層厚度(或岩盤深度)由東向西漸增。
(7)本研究利用微地動陣列所得之淺層S波速度構造，並結合此地區相關之地殼速度構造、震波衰減特性及地震震源的研究成果，利用隨機格林函數法進行1999年10月22日嘉義地震之寬頻地震動模擬。根據8個強震站之模擬結果顯示，在震央東南方之測點，其加速度、速度與位移之振幅皆有明顯低估的情形，而在震央西北方之測點則振幅高估，此問題可能與震源破裂之方向性(directivity)有關，另外，本研究僅採用一維速度構造，並假設震源為均勻破裂，而這些皆會造成模擬結果的差異。

關鍵字

微地動陣列量測 ； 接收函數 ； S波速度構造 ； 地震動模擬

並列摘要

The objectives of this study are to estimate the shallow S-wave velocity structures and to understand the bedrock distributions at different areas using the microtremor array measurements and the receiver-function methods. In order to examine the validity of the results, we compare the results with those from the well logging date and the geophysical data. Finally, we do the ground motion simulations based on the inverted S-wave velocity structures using the microtremor array measurements. The above results can be summarized as follows:
(1) At the TCDP drilling sites, the inverted S-wave velocity gradually increases from 1.52 km/sec to 2.22 km/sec at depths between 585 m and 1710 m using the microtremor array measurements. This result is similar to those from the velocity logs, which range from 1.4 km/sec at a depth of 597 m to 2.98 km/sec at a depth of 1705 m. According to the results of the seismic reflection method, the depths of the Chinshui Shale, the Chelungpu fault and the Sanyi fault are 900-1200 m, 1100 m and 1800 m, respectively; while the results of this study, its depths are 855-1395 m, 1125 m and 1755-1800 m, respectively.
(2) For the blind test of ESG2006, the inverted S-wave velocity structure in the study is very close to the theoretical data of ESG. For the real data analysis, the inverted S-wave velocity structures can match well with the well logging data at the depths 0-200 m. It means that the results from this study are reliable.
(3) Based on the inverted results from the microtremor array measurements at 16 sites, the depths of the Quaternary sediments are between 300 m and 870 m in the Puli area if the S-wave velocity of the bedrock is assumed to be 2000 m/sec. We also compare our results with those from the seismic exploration method. Both patterns are similar. The depths of the bedrock gradually increase from the basin’s edge to its center.
(4) In order to explore the shallow S-wave velocity structures of the Chiayi area, we conduct microtremor array measurements at 46 sites. If the S-wave velocity of the bedrock is assumed to be 1500 m/sec, the depths of the alluvium are between 560 and 1400 m gradually increasing from east to west. The thickness of the sediments increases from east to west while the S-wave velocities are in the range of 270 and 1500 m/sec; whereas, the thickness of the sediments decreases from east to west while the S-wave velocities are between 1500 and 2370 m/sec. The results are in good agreement with the geological and geophysical information.
(5) In order to investigate the shallow S-wave velocity structures in the Hualien area, we analyze the strong-motion data recorded by 16 stations of the SMART2 array using the receiver-function method. According to the inverted results, the shallow velocity structures (0-200 m) are similar to those from the well-logging data and the microtremor array measurements at the Dahan site. The thickness of the sediments at the Hualien area is about 130-660 m if the S-wave velocity of the bedrock is assumed to be 1000 m/sec The S-wave velocity decreases from east to west while the alluvial thickness increases from east to west.
(6) In order to detect the distribution of the seismic bedrock, we estimate the receiver functions by adopting the iterative deconvolution method. In the Taipei basin, the stacking receiver functions clearly reveal that the peak signals of the Ps-P time appear at about 0.295-0.925 sec. Both the Ps-P time and the alluvial thickness gradually increase from the southeast to the northwest. The results are consistent with those from the seismic exploration method and the well logging data. In the Chiayi area, the ranges of the Ps-P time are between 1.01 and 1.69 sec while the alluvial thickness increases from east to west. These results are in good agreement with those from the microtremor array measurements at the Chiayi area.
(7) We integrate the velocity structure between the crustal velocity structure and the near-surface velocity structure from the microtremor array measurements. In order to simulate the ground motions of the Chiayi earthquake (ML=6.4) in 1999, we use the stochastic Green’s function method based on the velocity structure, the Q values and its source rupture model. According to the simulation results of eight strong motion stations, the waveform amplitudes are obviously underestimated while the stations are located at the northwest of the epicenter. On the contrary, the waveform amplitudes are overestimated while the stations are located at the southeast of the epicenter. This phenomenon can be attributed to the rupture directivity. Moreover, the 1D velocity structure and uniform rupture model used also lead to some discrepancies between the observed and the synthetic data.

並列關鍵字

microtremor array measurement ； receiver function ； S-wave velocity structure ； ground motion simulation

參考文獻

Aki, K., and P. G. Richards (1980). Quantitative Seismology, theory and methods, W. H. Freeman, 932.

Ammon, C. J., G. E. Randall, and G. Zandt (1990). On the nonuniqueness of receiver function inversions, J. Geophys. Res., 95, 15303-15318.

Baumont, D., F. Courboulex, O. Scotti, and G. Stavrakakis (2002). Slip distribution of the MW 5.9, 1999 Athens earthquake inverted from regional seismiological data, Geophys. Res. Lett., 29(15), 1638.

Bonnefoy-Claudet, S., F. Cotton, and P. Y. Bard (2006). The nature of noise wave-field and its applications for site effects studies-A literature review, Earth Sci. Rev., 79, 205-227.

Boore, D. M. (2003). Simulation of ground motion using the stochastic method, Pure Appl. Geophys., 160, 635-676.