| 摘要: | 福建沿海地區接近歐亞板塊與菲律賓海板塊的邊界，其鄰近的地體構造包含兩個隱沒系統－馬尼拉隱沒系統及琉球隱沒系統和一個碰撞系統－臺灣造山帶。在研究地體構造時，震波非均向性的研究有助於了解岩石圈變形和地函流場機制。以往透過地震資料研究臺灣本島的非均向性性質時，所得結果可知其非均向性方向大致與臺灣造山帶走向平行，呈現近南北向的非均向性方向，此種排列形式與岩石圈變形有關。然而，就福建沿海地區而言，其地理位置與臺灣僅以臺灣海峽分隔，且為中國大陸東南部主要的強震活動區，但關於福建沿海區域性的非均向性特性和地函流場機制，其相關研究相對於臺灣地區較為少。
為了解中國大陸東南方－福建沿海地區複雜的地體動力機制，探討地震非均向性是相當重要的，相較於過去較大尺度的研究，以往這些研究的目的通常為得到大範圍的非均向性變化，因此只用零星的測站所得資料去描述，本研究期望以密集的區域性測站所能得到的高解析資料為目標，得到小範圍鄰近測站的非均向性變化，並以此彌補全球尺度的非均向性結果的不足。因此，本研究針對福建沿海地區的17個南北向排列的寬頻地震測站（broadband seismic stations）進行地震資料的處理，其資料來源時間段約為2008年8月至2010年6月，對此資料進行：（1）橫向分量最小能量法（Transverse Minimization Method, SC）之分析，及（2）分離強度法（Splitting Intensity Method, SI），藉此兩種方法觀察並探討福建沿海地區較區域性也較精細的地函流流場及其非均向性變化。
兩種方法的觀測結果顯示：（1）由遠震剪力波經SC法之分析，在17個測站資料中得到共27組分離參數（splitting parameters），平均快慢波之間的時間差（delay time, δt）約為2.4秒，最大可達3.8秒，其分離時間差與臺灣地區所得之結果相比明顯較高，表示其非均向性厚度也大於臺灣地區。而各測站所得的快波極化方向（fast polarization, ϕf）卻因不同地震事件相對於測站的震源後方位角（back-azimuth, ϕb）不同而產生變化，因此觀察到多方向性的快波極化方向解，部分測站以平行海岸線的形式存在，這樣的結果也表示中國大陸東南沿海下方存在複雜的地函流場機制。（2）藉由測量剪力波分離強度，在17個測站測得共193組強度參數，分別對每個測站的資料點做曲線擬合後，平均快慢波之間時間差約為1.04秒，快波極化方向於北段（北緯24.5度以北）呈現平行於海岸線方向，南段（北緯24.5度以南）呈現垂直海岸線方向。將本研究所得結果與全球非均向性速度構造交互比對，推測此變化應是由地函流流場與隱沒板塊之間的相互影響所造成的。;The tectonics of southeastern China and its surrounding regions involve two active subduction systems (the Manila and Ryukyu subductions) and one collision system (the Taiwan orogen). In this study, we use teleseismic SKS/SKKS splitting as a tool for investigating seismic anisotropy. The splitting parameters, fast-polarization azimuth (ϕf) and delay time (δt), can provide key information for understanding the geodynamic process in this region. Measurements of δt and ϕf can be used as indicators for the information of the product of deformation magnitude and strain direction. In previous studies, only few seismic stations in this region were available for the analysis. Therefore, in this study we selected 17 seismic stations along the Fujian coastline to obtain more shear wave measurements. Two methods (transverse minimization and splitting intensity method) are applied to obtain the splitting parameters for testing the reliability of the results. Transverse minimization method [Silver and Chan, 1991] utilizes a grid search approach to determine the pair of splitting parameters (ϕf, δt) by miniming the energy on the transverse component. Splitting intensity method, also named Multichannel method [Vinnik et al., 1989; Chavrot, 2000], is defined as the amplitude of the transverse component relative to the time derivative of the radial component. According to the sinusoidal function S=δt sin2(ϕ_b-ϕ_f ), splitting intensity (S) depends on the angle of back-azimuth (ϕb) and ϕf and on the δt between the two shear waves.
From transverse minimization method (SC), the results indicate that the average delay time of teleseismic shear waves observed from 17 stations is 2.4 s and the maximum is 3.8 s, suggesting that seismic anisotropy at least down to 300 km depth. The average split time delay (2.4 s) beneath Fujian coastline is greater than the average result observed from Taiwan and also indicates stronger seismic anisotropy. The disorder distribution of fast directions observed from 17 stations may be caused by different back-azimuths with different teleseismic events, and reveal the fact of complex geodynamic mechanisms beneath southeastern China. From splitting intensity method (SI), each station can observe a pair of splitting parameter (ϕf, δt) by sinusoidal curve fitting. The average split time delay is 1.04 s, which is lower than the result (2.4 s) observed from transverse minimization method. The fast directions can be discussed in two parts. As a result, along the Fujian coastline, compared with the global tomography, it can be explained that the variations of fast-polarizations (ϕf) could relate to the EW mantle flow created by NS collision between the India and Eurasian Plates, influenced by the Taiwan orogen (the collision between Eurasian and Philippine sea plates) and two subduction systems (the Manila and Ryukyu subductions). |
