地球物理学报  2015, Vol. 58 Issue (1): 125-133   PDF    
帕米尔—兴都库什深俯冲残留体对410 km间断面起伏形态的影响
眭怡1,2, 周元泽1,2, 王晓冉1,2    
1. 中国科学院计算地球动力学重点实验室, 北京 100049;
2. 中国科学院大学地球科学学院, 北京 100049
摘要:受俯冲残留体影响的410 km间断面起伏形态的研究对于确定地球内部物质构成及地球动力学过程具有重要作用.帕米尔—兴都库什俯冲区域拥有全球少有的中、深源地震,为研究410 km间断面起伏提供了良好的资源.利用日本Hi-net地震台网和美国TA台阵记录的帕米尔—兴都库什俯冲区域的6个震源深度为154.0~220.9 km、震级为Mb5.6~6.4的中、深源地震的短周期/宽频带波形资料,经过4次根倾斜叠加处理,获得了36组Hi-net子台网和TA记录资料的倾斜叠加灰度图,从中提取了与410 km间断面相关的次生转换震相SdP,发现受俯冲残留体影响下的410 km间断面的深度位于372~398 km.较之持续俯冲的西太平洋地区海洋岩石圈,研究区域俯冲滞留体对于410 km间断面的相变线的影响要小得多.
关键词帕米尔—兴都库什     410 km间断面     N次根倾斜叠加    
The effects of stagnant slabs on the topography of 410 km discontinuity beneath the Pamir-Hindu Kush
SUI Yi1,2, ZHOU Yuan-Ze1,2, WANG Xiao-Ran1,2    
1. Key Laboratory of Computational Geodynamics, Chinese Academy of Sciences, Beijing 100049, China;
2. College of Earth Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
Abstract: The effects of stagnant slabs on the topography of the 410 km discontinuity is helpful in determining the composition of Earth's interior and understanding the related geodynamical process. The different stages of subduction can introduce different temperature and material anomalies on mantle discontinuities. Comparing with the subducting slabs around the western Pacific regions, the subducted oceanic lithosphere beneath the Pamir-Hindu Kush should have less effects on the phase transition of mantle discontinuities. Pamir-Hindu Kush is a rare subduction region with a lot of mid-deep earthquakes. Broadband/short period waveform data of 6 earthquakes in the subducted slabs beneath Pamir-Hindu Kush, whose focal depths are between 154.0~220.9 km and magnitudes are between Mb5.6~6.4, recorded by Hi-net seismic network, Japan, and one event among them by the TA seismic array, USA, were processed with the 4-th root slant stack method. 36 vespegrams deduced from the observed waveform data from Hi-net sub-networks and the TA array were used to obtain the conversion phase SdP, then the conversion depths around 410km were determined based on IASP91.
For the Ev.4, the conversion points picked from the vespegrams of Hi-net at the epicentral distances of 40°~50° and the TA at about 95° respectively show almost the same conversion depths at 390 km and which means that the conversion depths from different sub-networks of Hi-net and TA array are consistent. Based on the distribution of 36 conversion points around the depth of 410 km related to picked SdP phases, the 410 km discontinuity is uplifted to the depths between 372~398 km and has relatively complicated structure. Assuming the Clapeyron slope as -2.9 MPa/K (Bina and Helffrich, 1994), the subducted slabs cause a low temperature anomaly about 250 to 360 K. Combining with geological and geochemical results, the continent-continent collision of Pamir-Hindu Kush happened at 102~85 Ma (Ali et al., 2002), so the stagnant materials of subducted slabs have been warmed up since then. The stagnant materials including the garnet and pyroxene, and some volatiles make the 410 km discontinuity structure complicated. Because of long-period warming-up process of the subducted slabs beneath Pamir-Hindu Kush, the stagnant materials have less impacts on the 410 km discontinuity than those from subducting slabs around the western Pacific region where the discontinuity is up-lifted by 60~70 km (e.g., Collier and Helffrich,1997; Collier et al., 2001). 36 vespegrams deduced from the observed waveform data of Hi-net sub-networks and the TA array of 6 mid-deep events beneath Pamir-Hindu Kush were used to retrieve the conversion phases SdP caused by the 410 km discontinuity. The conversion depths around 410 km were determined based on IASP91. From the distribution of conversion points, the 410 km discontinuity is up-lifted to the depths around 372~398 km caused by the long-period warming-up stagnant materials from the subducted oceanic lithosphere. Comparing with the western Pacific subducting slabs, the subducted stagnancy has less impacts on the topography of the 410 discontinuity.
Key words: Parmir-Hindu Kush     410 km discontinuity     N-th root slant stack    
1 引言

地震波可以直接带来地球内部结构的信息,人们利用强体波震相、面波乃至地球自由振荡等的信息,获得了一系列可靠的一维径向分层地球模型,如PREM(Dziewonski and Anderson,1981),IASP91(Kennett and Engdahl,1991),AK135(Kennett et al.,1995).410 km间断面作为地幔转换区的上界面出现在这三个速度模型中,其精细结构对于认识地球内部物质组成及地球内部运移的地球动力学过程具有重要意义(Deuss,2009).

人们利用长周期PP和SS前驱波(Flanagan and Shearer,1998a1998b; Deuss,2009)、ScS多次反射波(Revenaugh and Jordan,1991)等,获得了410间断面的全球性存在及其大尺度上的起伏形态.结合高温高压矿物物理实验和理论分析,410 km间断面被认为是α相橄榄石到β相橄榄石(也称瓦兹利石)的放热相变界面(如,Bina and Helffrich,1994; Collier et al.,2001; 金振民等,2013).在热地幔柱地区410 km间断面相变线表现为下降;而在冷的俯冲板片及其残留物存在的情况下,410 km间断面 相变线表现为上升(Bina and Helffrich,1994; Song et al.,2004; Obayashi et al. 2006; Deuss,2009).

随着世界上受地震灾害影响比较严重且经济发达或相对发达的国家对于测震工作的重视,美国、日本、欧洲部分地区和中国均开展了高密度、高质量固定或流动地震台网布设工作,以加强地震活动监测.近些年来,大量高密度台网的高质量观测波形资料不断积累,为地球内部结构研究提供了观测数据支撑.基于接收函数方法提取到相对低频的PdS转换 震相(d为转换深度),人们对于区域尺度上的410 km 间断面的存在形态有了更多的认识(如Zhang et al.,2010; Tian et al.,20102011),但相比于受冷或热物质影响的局部大起伏而言,这些结果显示的起伏则要小得多(如Li et al.,2000).进一步利用相对短周期波形资料来研究受冷或热的局部温度异常影响下的410 km间断面小尺度上的结构形态就显得很重要.

基于台网/台阵资料叠加方法提取离源弱次生震相是研究地幔间断面小尺度起伏形态的主要方法之一(Rost and Thomas,2002).弱次生转换震相SdP是以S波离源到达深度为d的间断面之后转换为P波,并为远处地震台站所记录.一般而言,考虑到减少横向不均匀的影响,常用30~90°震中距范围内的地震记录来提取SdP(臧绍先和周元泽,2002; Rost and Thomas,2002).中、深源地震的波形记录中的SdP因震源深度与附近间断面距离较小,因而相应的地震射线具有较小的Fresnel区,这有助于从观测波形的叠加图像中识别到该震相,从而能够确认地幔间断面的小尺度上的存在及起伏形态.

西太平洋俯冲区拥有全球最丰富的中、深源地震资源,部分地区密集地震台网可以在合适的震中范围内记录这些地震.很多研究者基于这些资料,利用倾斜叠加方法来提取转换震相SdP,很好地确认了西太平洋地区(主要是伊豆—小笠原和汤加—斐济地区)冷俯冲物质影响下的410 km间断面的相 变线深度的变化及复杂结构(如Collier et al.,2001). 陆陆碰撞区下方的俯冲残留物质较之持续俯冲过程中的海洋岩石圈俯冲带来滞留物质要少(Fukao and Obayashi,2013),温度异常也会弱,因而,410 km间断面相变深度所受到的影响也会小.

帕米尔—兴都库什地区位于印度板块与欧亚板块碰撞的西构造结是世界上少有的陆陆碰撞中、深源地震区(如宁杰远和臧绍先,1990孙文斌等,2009Negredo et al.,2007; Sippl et al.,2013).基于震源机制和高精度地震重定位结果(Pegler and Das,1998; Sippl et al.,2013; Bai and Zhang,2015)表明,兴都库什地区板片向北俯冲是重力引起的,而帕米尔地区板块向南俯冲则是由上地幔流所决定的,这样的双向俯冲(如图 1所示)构成了喜马拉雅西构造结区域复杂的高原地貌(张浪平等,2014).由于帕米尔—兴都库什地区海洋板块俯冲进入了上地幔深部,直至地幔转换区处(Negredo et al.,2007),而大陆岩石圈俯冲则紧随其后,因而在本区域具有丰富的中-深源地震最深可以到达380 km左右(Pegler and Das,1998).选择本区域研究410 km间断面受俯冲滞留物质影响下的起伏形态,可以很好地与海洋岩石圈俯冲板块对该间断面的影响进行对比分析.

图 1 地震分布图
以沙滩球表示的矩张量解来自http://www.globalcmt.org,没有矩张量解的地震事件以大圆点表示,等值线给出的是Wadati-Benioff带在地幔中存在的深度分布(Gudmundsson and Sambridge,1998),等值线上的白色数字为深度,单位为km.
Fig. 1 The distribution of the events used
The focal mechanisms labeled with beach balls are the CMT solutions from http://www.globalcmt.org and the solid circle for another one without CMT solution. The contours are for the extension of the Wadati-Benioff slab(Gudmundsson and Sambridge,1998). The white numbers are the depths of the contous with units in km.

本文基于高密度的日本Hi-net地震台网和Earth Scope计划组成部分的USArray流动台阵(TA: Transportable Array)所记录的地震波形数据,利用N次根倾斜叠加方法,提取P尾波中的SdP次生转换震相,分析了帕米尔—兴都库什地区 俯冲板块物质影响下的410 km间断面的存在及形态. 2 数据处理

由于震源过于复杂或者来自地表反射的强pP和sP震相及多次波震相会对弱次生SdP震相的拾取带来很大的干扰,因而我们在收集地震波形资料时需要限制地震事件的震级,以保证所使用的波形资料具有相对简单的震源时间函数,从而有助于提取次生震相.本文中选用的地震事件震级范围为Mb5.6~6.4,震源时间函数时长不超过4 s,且具有足够高的信噪比.虽然本区域有较多的中、深源地震,但受限于台网记录时间以及波形资料选择的要 求,我们收集了2004年以来的6个地震的Hi-net台网(图 2a)短周期波形资料;同时还收集了其中一个地震的TA台阵(图 2b)宽频带波形资料.地震分布如图 1所示,相关地震参数见表 1.本文中所用到的地震波形资料来自于日本防灾科学技术研究所(http://www.hinet.bosai.go.jp)和美国地震学联合会(IRIS,http://www.iris.edu).

图 2 地震台网/台阵的台站分布图
(a)Hi-net台网的分布,其中H1—H9为事件Ev.5可用资料对应的各子台网台站分布图,圆点为所有的Hi-net台站分布;(b)TA的台站(三角形)分布图.
Fig. 2 Station distributions of the Hi-net seismic network and the TA
(a)The station distribution of the Hi-net. The sub-network distributions(H1—H9)of Hi-net which provided the waveform data of the Ev.5 are labeled with different color legends,and the small dots are for all others.(b)The station distribution of TA labeled with triangles.

表 1 本文使用的地震事件相关参数 Table 1 Paramets of the events used

震源深度的确定直接影响到转换点深度反演的精度,因而准确的震源深度是非常重要的.考虑到本文所使用的局部地区地震台网来进行震源深度确定并不能获得很好的深度精度,因而本文选用了目前震源深度确定最好的EHB数据库(http://www.isc.ac.uk/EHB/index.html)给出的2008年前(含2008年)的地震位置及深度参数(Engdahl et al.,1998),而2009年起的地震参数则来自于国际地震中心(http://www.isc.ac.uk)提供的地震位置及基于全球波形资料的pP震相确定的震源深度.

文中我们采用0.2~1.0 Hz的滤波频段对地震波形进行滤波.这一滤波频段可以很好地拾取出与 地幔间断面相关的SdP转换震相(Castle and Creager,2000),同时压制噪声.在处理中,我们进一步人工删除受脉冲噪声等影响的非正常波形记录.

地幔间断面比较弱,来自这样速度界面的转换震相SdP相对也是很弱的,一般地是无法直接观察到的.图 3所示为Ev.5被Hi-net子台网H6(如图 2a)所记录的波形资料图.各记录以直达P震相的最强峰值进行归一化处理,并将其到达时刻归零.图中我们以虚线标识了直达P震相和后面倾斜叠加结果中识读到的S376P次生转换震相.该次生震相直接从波形图上则很难观测到,甚至应该较强的pP震相也很难被清晰地识读出.

图 3 Hi-net子台网H6记录到的Ev.5地震垂向波形
0 s时刻虚线标注了归一化的直达P震相,标有S376P的虚线为对应着来自376 km速度界面的次生转换震相.
Fig. 3 Waveform data of Ev.5 recorded by the sub-network H6 of the Hi-net
The dashed line with “P” at 0s is for the peaks of the direct P phases and the dashed line with “S376P” is for the arrival times of the conversion phases from the 410 km discontinuities.

为了有效地获取SdP次生转换震相,我们使用了N次根倾斜叠加方法来增强次生震相并压制噪 声(Kanasewich et al.,1973;McFadden et al.,1986臧绍先和周元泽,2002; Rost and Thomas,2002).由于目标震相SdP位于强直达震相P之后,因而N次根叠加处理中,我们选用直达P震相作为参考震相.本文所用的地震震源时间函数比较简单,叠加处理过程中,我们将直达P震相的峰值走时归为0,并基于峰值对相应的波形资料进行幅度归一化处理(如图 3),因而在叠加图灰度图上直达P震相的走时差和慢度差均为0.

在具体叠加处理中,TA台阵震中范围相对较小,且可用台站资料比较少,我们对所有资料直接进行4次根叠加处理.日本Hi-net台网具有很高的台站密度,且质量较高.根据资料的实际质量情况,我们将Hi-net台网的信噪比超过3的波形资料按照震中距和方位角范围进行分组,形成一个地震为多个Hi-net子台网(如图 2a)所记录的波形资料组.进一步地,我们分别对各组资料进行4次根倾斜叠加处理,获得各地震-子台网的走时差-慢度差域的相对幅度图,从而可以对相同地震不同台网/子台网和不同地震相同子台网的叠加结果进行对比,进而分析速度界面的起伏,并验证结果的可靠性.为更好地显示弱次生震相的存在,进一步求取叠加图的包络线并求取20倍的以10为底的对数值(dB),据此绘制出走时差-慢度差域的灰度图(Vespegram).参考理论走时差和慢度差,可以从灰度图中读取出慢度差理论值±0.1 s/(°)范围内的次生震相SdP;在此基础上基于IASP91模型反演出相应的转换点深度及位置.详细的资料分析处理过程可以参看臧绍先和周元泽(2002). 3 结果与讨论

我们以Ev.5为例分析台网资料的4次根倾斜 叠加结果.图 4是地震Ev.5为Hi-net各子台网记录波形的4次根倾斜叠加结果,H6子台网记录的波形资料见图 3.各叠加图上方的箭头对应的数字标注出了可能的次生转换震相对应的转换点深度.由于震源深度的限制,较强的直达震相P和pP之间的次生转换震相均与360~381 km深度上速度跃变面有关系.

图 4 Hi-net所记录地震Ev.5的波形资料各子台网4次根倾斜叠加灰度图灰度图横坐标为相对于直达P的走时差Δt,纵坐标为相对于直达P的慢度差Δp.箭头所对应的数字为转换震相对应的转换深度,单位为 km.Fig. 4 The vespegrams of the Ev.5 derived from the 4-th root slant stacking with the data by Hi-net sub-networks. H1—H9 were shown at Fig. 2b The numbers above arrows are the conversion depths(in km)related to SdP phases.

类似地,我们将各地震为Hi-net不同子台网和TA台阵记录到的波形资料进行4次根倾斜叠加处理,共获得36组能够清晰地观察到与410 km间断面相关SdP次生震相的4次根倾斜叠加灰度图.相关的转换点深度见表 2.不是每个子台网都能识读到具有很好的强度和符合走时差-慢度差关系的次生转换震相,因而表中存在着部分子台网转换点深度的缺失.

表 2 叠加图给出的转换点深度列表 Table 2 Conversion depths from the vespegrams

本文中只有一个地震(Ev.4)共同为TA台阵和Hi-net台网所记录,并给出了很好的倾斜叠加结果.TA台阵在95°左右的震中距记录到该地震,而Hi-net则在40°~50°震中距上记录到该地震.从转换点平均深度上来看,Hi-net台网各子台网给出的转换点的平均深度则位于393 km上,而TA给出的转换点深度是390 km,这说明各台网/子台网叠加结果具有很好的一致性.

相较于其他地震,Ev.5给出的转换点深度要浅得多.一般地,转换点深度误差来源于震源深度误差(±5 km)、相对走时识读误差(±5 km)以及俯冲板块速度异常引起的误差(±1 km),总体而言,转换点深度误差在±11 km(Collier and Helffrich,1997).各子台网相应的倾斜叠加结果图上的次生震相对应的转换点深度的差异与局部区域的速度跃变和走时差读取误差有一定关系(Zang et al.,2006).因此,该地震给出的转换点深度有些许存疑.

我们将各转换点投影到地球平面上,可以看到410 km间断面存在的深度位于372~398 km附近(图 5),即在俯冲板块及其残留物质的影响下抬升约20~50 km.一般地,410 km间断面附近的克拉伯龙(Clapeyron)斜率为-2.9 MPa/K(Bina and Helffrich,1994),那么每抬升8 km则约有100 K 的低温异常(Flanagan and Shearer,1998b),因而相对于周围地幔而言本文研究区域下方俯冲物质导致了250~630 K的低温异常.

图 5 410 km间断面相关的SdP转换点深度分布
三角形为转换点的位置,其附近的数字为转换点平均深度(表 2).其他说明同图 1.
Fig. 5 The SdP conversion points related to the 410 km discontinuity
Triangles are for the conversion points locations and the related numbers are for the average depths of the conversion points(Table 2). All others are same as Fig. 1.

基于地质学与地球化学的研究成果,本文研究 区域陆陆碰撞的时间大致在102~85 Ma(Ali et al.,2002),因而海洋岩石圈俯冲残留应该在此之前就已经进入地球深部了.同时,从地震层析成像结果来看(Koulakov and Sobolev,2006Villaseñor et al.,2003转引自Negredo et al.,2007),帕米尔—兴都库什俯冲板块在地球内部250 km深度处存在着减薄形态,表明俯冲板块处于拆离过程中,最深部可能到达地幔转换区中.俯冲板块的残留体会给地球 深部带来不同的物质,如石榴子石、辉石,甚至水等挥发分物质(Weidner and Wang,2000),这会导致在α相橄榄石到β相橄榄石相变之外,还有石榴子石和辉石等的相变;同时水的存在也会导致部分熔融的发生,进而影响到410 km间断面相变线的起伏形态.

因而,相比于410 km间断面抬升达60~70 km(如Collier and Helffrich,1997; Collier et al.,2001)的西太平洋俯冲区域而言,本研究区域下方的俯冲残留体带来的低温异常要小,相应的抬升也就要小不少.当然俯冲残留物质本身也会导致该区域物质构成的调整,而这需要进一步更精细的地震学工作,并在此基础上,结合高温高压矿物学成果进行分析. 4 结论

本文利用日本Hi-net地震台网和美国TA台阵记录的帕米尔—兴都库什俯冲区域的6个震源深度为154.0~220.9 km、震级为Mb5.6~6.4的中、深源地震的短周期/宽频带波形资料,经过4次根倾斜叠加处理,获得了36组Hi-net子台网和TA记录资料的倾斜叠加灰度图,从中提取了与410 km间断面相关的次生转换震相SdP,发现受俯冲残留 体影响下的410 km间断面的深度位于372~398 km.较之持续俯冲的西太平洋地区海洋岩石圈对于410 km 间断面的相变线的影响要小得多.

致谢 感谢两位评阅人宝贵意见.
参考文献
[1] Ali A, Nakamura E, Yamamoto H. 2002. Sm-Nd mineral ages of pegmatite veins and their host rocks from Swat area, Chilas complex, northern Pakistan. J. Asian Earth Sci., 21(3): 331-339.
[2] Bai L, Zhang T Z. 2015. Complex deformation pattern of the Pamir-Hindu Kush region inferred from multi-scale double-difference earthquake relocations. Tectonophysics, 638: 177-184, doi: 10.1016/j.tecto.2014.11.006.
[3] Bina C R, Helffrich G. 1994. Phase transition Clapeyron slopes and transition zone seismic discontinuity topography. J. Geophys. Res., 99(B8): 15853-15860.
[4] Castle J C, Creager K C. 2000. Local sharpness and shear wave speed jump across the 660 km discontinuity. J. Geophys. Res., 105(B3): 6191-6200.
[5] Collier J D, Helffrich G R. 1997. Topography of the "410" and "660" km seismic discontinuities in the Izu-Bonin Subduction Zone. Geophys. Res. Lett., 24(12): 1535-1538.
[6] Collier J D, Helffrich G R, Wood B J. 2001. Seismic discontinuities and subduction zones. Phys. Earth Planet. Inter., 127(1-4): 35-49.
[7] Deuss A. 2009. Global observations of mantle discontinuities using SS and PP precursors. Surv. Geophy., 30(4-5): 301-326.
[8] Dziewonski A M, Anderson D L. 1981. Preliminary reference Earth model. Phys. Earth Planet. Inter., 25(4): 297-356.
[9] Engdahl E R, van der Hilst R, Buland R. 1998. Global teleseismic earthquake relocation with improved travel times and procedures for depth determination. Bull. Seism. Soc. Am., 88(3): 722-743.
[10] Flanagan M P, Shearer P M. 1998a. Global mapping of topography on transition zone velocity discontinuities by stacking SS precursors. J. Geophys. Res., 103(B2): 2673-2692.
[11] Flanagan M P, Shearer P M. 1998b. Topography on the 410-km seismic velocity discontinuity near subduction zones from stacking of sS, sP, and pP precursors. J. Geophys. Res., 103(B9): 21165-21182.
[12] Fukao Y, Obayashi M. 2013. Subducted slabs stagnant above, penetrating through, and trapped below the 660 km discontinuity. J. Geophys. Res., 118(11): 5920-5938.
[13] Gudmundsson Ó, Sambridge M. 1998. A regionalized upper mantle (RUM) seismic model. J. Geophys. Res., 103(B4): 7121-7136.
[14] Jin Z M, Zhang Z J, Wu Y, et al. 2013. Mantle transition zone and research methods. // Ding Z L ed. Research Methods in Solid Earth (in Chinese). Beijing: Science Publishing Company: 402-434.
[15] Kanasewich E R, Hemmings C D, Alpaslan T. 1973. Nth-root stack nonlinear multichannel filter. Geophysics, 38(2): 327-338.
[16] Kennett B L N, Engdahl E R. 1991. Traveltimes for global earthquake location and phase identification. Geophys. J. Int., 105(2): 429-465.
[17] Kennett B L N, Engdahl E R, Buland R. 1995. Constraints on seismic velocities in the Earth from traveltimes. Geophys. J. Int., 122(1): 108-124.
[18] Koulakov I, Sobolev S V. 2006. A tomographic image of Indian lithosphere break-off beneath the Pamir-Hindu Kush region. Geophys. J. Int., 164(2): 425-440.
[19] Li X, Sobolev S V, Kind R, et al. 2000. A detailed receiver function image of the upper mantle discontinuities in the Japan subduction zone. Earth Planet. Sci. Lett., 183(3-4): 527-541.
[20] McFadden P L, Drummond B J, Karvis S. 1986. The N-th root stack: theory, application and examples. Geophysics, 51(10): 1879-1892.
[21] Negredo A M, Replumaz A, Villaseñor A, et al. 2007. Modeling the evolution of continental subduction processes in the Pamir-Hindu Kush region. Earth Planet. Sci. Lett., 259(1-2): 212-225.
[22] Ning J Y, Zang S X. 1990. The distribution of Earthquakes and Stress State in Pamir-Hindu Kush Regions. Chinese J. Geophys. (in Chinese), 33(6): 657-669.
[23] Obayashi M, Sugioka H, Yoshimitsu J, et al. 2006. High temperature anomalies oceanward of subducting slabs at the 410 km discontinuity. Earth Planet. Sci. Lett., 243(1-2): 149-158.
[24] Pegler G, Das S. 1998. An enhanced image of the Pamir-Hindu Kush seismic zone from relocated earthquake hypocentres. Geophys. J. Int., 134(2): 573-595.
[25] Revenaugh J, Jordan T H. 1991. Mantle layering from ScS reverberations: 1. Waveform inversion of zeroth-order reverberations. J. Geophys. Res., 96(B12): 19749-19762.
[26] Rost S, Thomas C. 2002. Array seismology: methods and applications. Rev. Geophys., 40(3): 2-1—2-27, doi: 10.1029/2000RG000100.
[27] Sippl C, Schurr B, Yuan X, et al. 2013. Geometry of the Pamir-Hindu Kush intermediate-depth earthquake zone from local seismic data. J. Geophys. Res., 118(4): 1438-1457.
[28] Song T R A, Helmberger D V, Grand S P. 2004. Low-velocity zone atop the 410-km seismic discontinuity in the northwestern United States. Nature, 427(6974): 530-533.
[29] Sun W B, He Y S, Chang Z, et al. 2009. The plate subduction and stress state in the Pamir-Hindu Kush region. Seism. Geol. (in Chinese), 31(2): 207-217.
[30] Tian X B, Zhao D P, Zhang H S, et al. 2010. Mantle transition zone topography and structure beneath the central Tien Shan orogenic belt. J. Geophys. Res., 115(B10): B10308, doi: 10.1029/2008JB006229.
[31] Tian X B, Teng J W, Zhang H S, et al. 2011. Structure of crust and upper mantle beneath the Ordos Block and the Yinshan Mountains revealed by receiver function analysis. Phys. Earth Planet. Inter., 184(3-4): 186-193.
[32] Villaseñor A, Spakman W, Engdahl E R. 2003. Influence of regional travel times in global tomographic models. EGS-AGU-EUG Joint Assembly, Abstracts from the meeting held in Nice, France, 6-11 April 2003, abstract 8614.
[33] Weidner D J, Wang Y B. 2000. Phase transformations: implications for mantle structure. // Karaoto S I, Forte A, Liebermann R, et al., Earth's Deep Interior: Mineral Physics and Tomography from the Atomic to the Global Scale, Vol. Geophysical Monograph Series 117. American Geophysical Union: 215-235.
[34] Zang S X, Zhou Y Z. 2002. N-th root slant stack and its application in study of mantle discontinuities. Chinese J. Geophys. (in Chinese), 45(3): 407-415, doi: 10.3321/j.issn:0001-5733.2002.03.012.
[35] Zang S X, Zhou Y Z, Ning J Y, et al. 2006. Multiple discontinuities near 660 km beneath Tonga area. Geophys. Res. Lett., 33(20): L20312, doi: 10.1029/2006GL027262.
[36] Zhang L P, Shao Z G, Li Z H. 2014. Dynamic action of mutual subduction between the Indian and the Eurasia plates in Hindu Kush-Pamir region. Chinese J. Geophys. (in Chinese), 57(2): 459-471, doi: 10.6038/cjg20140212.
[37] Zhang Z J, Yuan X H, Chen Y, et al. 2010. Seismic signature of the collision between the east Tibetan escape flow and the Sichuan Basin. Earth Planet. Sci. Lett., 292(3-4): 254-264.
[38] 金振民, 张忠杰, 吴耀等. 2013. 地幔转换带和研究方法. //丁仲礼主编. 固体地球科学研究方法. 北京: 科学出版社: 402-434.
[39] 宁杰远, 臧绍先. 1990. 帕米尔—兴都库什地区地震空间分布特征及应力场特征. 地球物理学报, 33(6): 657-669.
[40] 孙文斌, 和跃时, 常征等. 2009. 帕米尔—兴都库什地区板块俯冲及其应力状态. 地震地质, 31(2): 207-217.
[41] 臧绍先, 周元泽. 2002. N次根倾斜叠加在地幔间断面研究中的应用. 地球物理学报, 45(3): 407-415, doi: 10.3321/j.issn:0001-5733.2002.03.012.
[42] 张浪平, 邵志刚, 李志海. 2014. 印度板块与欧亚板块在兴都库什—帕米尔地区相互俯冲的动力作用分析. 地球物理学报, 57(2): 459-471, doi: 10.6038/cjg20140212.