利用區域深震之接收函數研究地殼速度構造

Abstract

一般傳統的接收函數法（Receiver Function），乃是利用近乎垂直入射的遠震P波資料來研究測站下之地殼與上地幔速度構造。若在選取適當的入射角度範圍下，區域深震亦可依此原理用來計算接收函數，且由於區域地震的傳播距離較短，能量沿著傳播路徑之衰減與擴散較少，規模較小之地震亦可使用。因此，當遠震資料有數量不足或方位上分布不均時，區域深震能提供額外的資料作接收函數分析，以提高分析結果之可信度。 為了證實以上論點，本研究以中國東北的牡丹江測站資料，比較遠震以及區域深震兩種資料所作之接收函數結果。為了避免其他因素影響結果，兩組地震資料所選取之方位角相似，且為符合入射角之條件以及有足夠長之P波資料作接收函數計算，對於區域深震資料，我們選取震央距2至5度、深度大於400公里之隱沒帶深震資料作分析。比較結果顯示，區域深震與遠震兩組資料所得之接收函數結果相當一致，且推算出的地殼厚度約為36至40公里，亦與前人之研究結果相符，顯示區域深震接收函數之可行性。 此外，在區域深震與遠震接收函數結果中，兩者的PpPms波到時約有0.4秒的差異，此差異對逆推構造之結果影響極小，而造成此時間差之原因，本研究提出五種可能因素，但經過仔細計算與確認後皆無法完全解釋。雖然對於造成此時間差的原因，目前仍不清楚，但並不影響牡丹江站之速度逆推結果。

For the conventional receiver function (RF) analysis, waveforms of steeply impinging P-waves from tele-earthquakes are used to investigate the crust and upper mantle structures beneath seismic stations. In principle, regional deep events can also be used to calculate RFs as long as the ray parameters are in appropriate range similar to those from teleseismic events. Due to shorter epicentral distances, the regional body waves experience less geometrical spreading and attenuation along the paths of propagation. Therefore, regional events potentially could provide additional data for the RF analysis when available teleseismic events are limited. Based on our extensive tests, the results show that the RFs estimated using regional deep earthquakes are comparable to those from teleseismic events. We perform side-by-side comparisons of RFs calculated from different sets of data at a narrow azimuth range to avoid other complications. For the regional case, we select data from events deeper than 400 km with epicentral distances between 2° and 5° so that the incident angles are small and the P wave trains are sufficiently long for deconvolution. Using station MDJ and deep events in Japan Sea as examples, we successfully estimate the crustal thickness under northeast China to be 36~40 km, consistent with previous studies for this area. Moreover, we discover that there is about 0.4 s of time difference in PpPms-P between the regional and teleseismic RFs. We examine five possible scenarios and none of them are able to provide satisfying explanation to the observation. However, such differences in PpPms-P do not significantly influence the inversed seismic velocity model. Here the regional RF analysis is considered to efficiently determine seismic structure, as well as the results compiled from the teleseismic RF.