本研究利用接收函數推求橫跨台灣中部莫荷面深度，地震資料係利用TAIGER 計畫於2006 年至2008 年間設置於台灣中部之12 個寬頻地震站(TGC01 至TGC12)，震央距介於30 至90 度且規模大於6.0 以上的地震事件。經過資料篩選後，挑選TGC06 至TGC12 共7 個寬頻地震站，另外加入氣象局地震觀測網(CWBSN)魚池站(TYC)與中研院寬頻地震網(BATS)雙龍站(SSLB)，使測線更為完整。資料分析使用Levin and Park(2000)所提出多重窗制相關估計法(MultipleTaper Correlation estimate, 簡稱MTC)，求得徑向接收函數；接著利用Zhu and Kanamori (2000) 所提出H-κstacking 方法，求取莫荷面深度與κ值(即Vp/Vs)。根據研究結果顯示，除了TGC09 與TGC12 外，莫荷面深度皆介於30-35 公里間，在TGC09 並未得到一個較為合理的莫荷面深度值，TGC12的莫荷面則落在17-20 公里間，顯示花東縱谷以東已進入菲律賓海板塊範圍。此外，比較前人研究速度模型與密度模型、地震事件後方位角分群計算和共同地震事件計算的 結果，本研究可歸納以下結論：1.莫荷面深度分佈與Yen and Yeh (1998)利用重力觀測推求之密度模型結果最為相似；2.依後方位角(bazk azimuth)分群作計算，所得到的深度與未分群之結果並無顯著差異；3.各測站計算使用的地震事件相同與否，對於深度結果 影響並不顯著，而與資料品質好壞及地震事件數量多寡較為相關；4.在TGC09 所得到的深度為20 公里，與一般大陸性地殼的莫荷面(30 至40 公里)深度差異甚大，推論TGC09 的波向不明顯，可能與中央山脈下方有ㄧ高速帶(Lin, 2007)的存在有關。 We use data from TAIGER project to deduce the Moho depths cross Central Taiwan byreceiver function. The selected data is mainly from seven of the twelve broadband seismograph stations, TGC06 to TGC12, of the TGC-line of TAIGER from 2006 to 2008, and we also add two other stations for the data integrity, the SSLB station of Broadband Array in Taiwan for Seismology (BATS) and the TYC station of Central Weather Bureau Seismograph Network (CWBSN). From these nine stations, we pick up the earthquake events whose epicenters and magnitudes are between 30 to 90 degree and greater than 6.0, respectively. The Multiple-Taper Correlation estimation is used to calculate the radial receiver function as the first step for data analysis; then, we compute the possible Moho depth and κ value (Vp/Vs) by H-κ stacking. The Moho depths beneath these stations are approximated 30 to 35 km except TGC09 and TGC12. We don’t get a sensible depth in TGC09. The Moho depth beneath TGC12 around 17 to 20 km indicates that TGC12, to the east of the Longitudinal Valley, is located above the Philippine Sea Plate. According to further analysis and discussion with former researches, we can have the following conclusions: (1) Comparing the result with former two velocity models and one density model, our result is mostly consisted with the density model proposed by Yen and Yeh (1998). (2) We re-group the events by back azimuth into four quadrants and get the similar Moho depths as previous ones. (3) We use the same events for different stations to identify if event selection could be an important factor for H-κ stacking or not. The result shows that one of the most important factors is not using the same events for different stations but the data quality and quantity. (4) The extremely shallow Moho depth beneath TGC09 about 20 km, different from average depth beneath continental crust, is mostly related with a high velocity zone beneath Central Taiwan (Lin, 2000).
