台北盆地的場址效應放大效應－譜比法應用於強震資料與理論分析的探討

Site Amplification in Taipei Basin – Spectral Ratio Analysis for Strong Motion Data and Theoretical Simulations

許嘉峻(Chia-Chun Hsu)

指導教授 ： 陳浩維

國立中央大學/地球科學學院/地球物理研究所/碩士(2007年)

摘要

台北盆地座落於台灣北部，為人口及建物相當密集之區域，世界第一大樓－台北101亦座落於本研究地區。當地震波傳遞經過較為鬆軟的土層時，往往會在此地層上產生較大的振幅及地震災害。為探討此盆地之放大效應，於本研究使用西元1999年9月21日發生的集集地震和2002年3月31日發生於花蓮外海的地震及2004年10月23日發生於四獸山地震之強震地震資料，進行譜比法分析其盆地放大效應之影響。由強震資料分析結果得知：台北盆地地區的主要共振頻率介於0.5~2Hz間，而外部的山區則有較高的共振頻率。使用S波紀錄時窗分析實際資料所得譜比值將大於使用雜訊時窗所計算的譜比值。由S波紀錄時窗所得的頻譜圖可以解析出較多的放大頻率。使用S波紀錄時窗計算出來的放大效應趨勢方向性約略與地震波傳遞的方向相互垂直，此應為選取之時窗內主要的能量以SV波為主；使用雜訊波紀錄時窗所計算出來的放大效應趨勢方向性約略與震波傳遞的方向相互平行，此意含所選取的雜訊波時窗當中其波相主要以雷利波為主。地震郭玫較大的921集集地震與331花蓮外海地震的分析結果，無明顯的地形放大效應。此應為盆地的放大效應大過於地形效應所造成之現象。相對的，規模較小的四獸山地震，由雜訊分析則呈現較明顯的地形放大效應。資料分析結果顯示，不同的地震在不同的方向上會出現不同的路徑效應，若再加上淺部場址效應，則使得台北盆地最終的盆地放大效應於共振頻率與放大倍率在空間分布上顯得相對更加複雜。
由理論模擬與譜比值分析可知：考慮不同的入射角對應的折射或反射係數，可知當造成全反射時，會有震幅與能量變大的現象。透過簡單模型的理論計算，可得出共振頻率的高低與S波速度的大小成正比且與4倍岩層厚度成反比。入射角度、S波速度、密度與Q值為影響放大倍率的主因。改變速度構造間之阻抗對比，分析結果亦指出上述之關係式不因阻抗對比改變而有明顯變動。二維SH波的波傳模擬的結果指出：淺部的速度構造對於台北盆地的振幅放大與震動歷程有相當程度上之影響。盆地邊緣的共振累積能量分布亦較盆地中心為大。盆地較深的地區其能量分布亦較盆地較淺區域來的大。將來可藉由不同的震源陣列，產生不同的平面波入射角度，進一步可探討與分析不同角度入射盆地效應的影響。由理論模擬得知，未來需建立一個較完善之淺部速度構造資料庫，對台北盆地之放大效應進行有系統的模擬與分析，以便有更符合實際及更佳的成果及解釋。

關鍵字

盆地放大效應 ； 譜比法 ； 場址效應

並列摘要

Located in the Northern Taiwan, Taipei is the city where dense population and many high rise buildings like “Taipei 101” may cause concerns on earthquake hazards. Large amplification effects may cause serious damages when seismic waves impinging upon soft soil layer sitting on the top of Taipei basin. Three typical earthquake data sets including Chi-Chi earthquake (Sept. 21., 1999); Hualien earthquake (March 31, 2002) and Shisoshan earthquake (Oct. 23, 2004) were used for site amplification effects studies. Based upon spectral ratio analysis technique for S-wave and noise wave phases, the dominant resonant frequencies ranging 0.5 to 2.0 Hz was defined in the Taipei basin. Higher resonance frequency can be obtained in area with obvious topography variations. The estimated amplification values from spectral analysis base on HVSR methods appears to have higher response than HVNR method. Base on HVNR study, all three earthquakes data show strong amplifications and higher resonant frequency from topography around Taipei basin. More resolvable resonance frequency can be obtained from know S-wave than noise phases. Directional spatial distribution of amplification is mainly correlated to the types of wave phases used in HV method. For Sv-wave, direction of propagation is roughly perpendicular to the trend corresponds to high amplification area. While base on noise data, mainly corresponds to Raleigh waves, the direction of propagation is parallel to the trend with high amplification values. The large magnitude including 921 and 331 earthquakes will have less site effects resulting from topography than basin amplification effects. For smaller Shisoshan earthquake, a topography effect becomes more apparent. Combination of path and site effects produces even more complicated resonant frequency and amplification patterns.
Base on the calculations of reflection and transmission coefficients varies with incidence angle, drastic increases in amplitude as conditions meets the total reflection. From a simple model, we have found positive relationship between fundamental resonant frequency and S-wave speed but inverse proportional to the four times of layer thickness. Incidence angle, S-wave velocity, density and quality factor are the main parameters affecting amplification and Horizontal to vertical signal spectral ratio. The changes in impedance contrast may not strongly affects the resonant frequencies and conclusions obtained from high contrast model. Two dimensional SH wave simulation of basin structure shows that: the shallow S-waves velocity specification is critical for the quantitative analysis of basin amplification effects. With increasing details in ultra-shallow S- wave velocity variations, the estimated cumulative kinetic energy from synthetic data increases accordingly which also indicates that thicker layer will preserve more kinetic energy obtained from recorded strong motion data. Basin edges also show larger kinetic energy than basin center which explain well with the observed phenomena. For further detailed simulation, a synthesis of plane waves impinging upon formation boundary with different incidence angles for basin amplification effects studies should be actively pursuit. From our preliminary simulation, results indicate that it is important to systematically measure and set up a near-surface S-wave velocity database in Taipei metropolitan area for future more practical usages in strong motion analysis and interpretations.

並列關鍵字

Spectral ratio method ； basin amplification effects ； site effects

參考文獻

黃慈銘，李錫堤，廖啟雯，2002，台北盆地松山層土壤性質及剪力波速之空間分析，地工技術，90，41-54。

廖啟雯、李錫堤，1998，地下地質分散式資料庫建置與應用－以台北盆地為例，中華地理資訊學會學術研討會論文摘要集，第16頁。

Gardner, G. H. F., Gardner, L. W. and Gregory, A. R., 1974, Formation velocity and density－the diagnostic basics for stratigraphic traps, Geophysics, Vol. 39, No. 6, 770-780.

Lee, S. J., et al., 2007, Simulation of Strong Ground Motion and 3D Amplification Effect in the Taipei Basin by using a Composite Grid Finite-Difference Method.（submitted）

Parolai, S., et al., 2001, Comparison of different site response estimation techniques using aftershocks of the 1999 Izmit earthquake, BSSA, Vol. 94, 1069-1108.