龙门山断裂带卫星重力场特征及其发震机制

引用本文

汤井田, 杨磊, 任政勇, 胡双贵, 徐志敏. 龙门山断裂带卫星重力场特征及其发震机制[J]. 地震地质, 2019,41(5): 1136-1154.
TANG Jing-tian, YANG Lei, REN Zheng-yong, HU Shuang-gui, XU Zhi-min. CHARACTERISTICS OF SATELLITE GRAVITY FIELD AND SEISMOGENIC MECHANISM IN THE LONGMENSHAN FAULT ZONE[J]. SEISMOLOGY AND GEOLOGY, 2019,41(5): 1136-1154. 复制到剪切板

DOI:10.3969/j.issn.0253-4967.2019.05.005
Permissions

《地震地质》编辑部所有

龙门山断裂带卫星重力场特征及其发震机制

汤井田^1,^2,³, 杨磊¹, 任政勇^1,^2,^3,^*, 胡双贵^1,^2,³, 徐志敏⁴

1)中南大学, 地球科学与信息物理学院, 长沙 410083

2)中南大学, 有色金属成矿预测与地质环境监测教育部重点实验室, 长沙 410083

3)中南大学, 自然资源部覆盖区深部资源勘查工程技术创新中心, 长沙 410083

4)承德石油高等专科学校, 河北省仪器仪表工程技术研究中心, 承德 067000

*通讯作者: 任政勇, 男, 教授, 从事重磁电正反演及信号处理反演成像研究, E-mail: renzhengyong@csu.edu.cn。

〔作者简介〕汤井田, 男, 1965年生, 1992年于中南工业大学获地球探测与信息技术专业博士学位, 教授, 博士生导师, 主要从事电磁场理论和应用、地球物理信号处理及反演成像等研究, E-mail: jttang@csu.edu.cn。

收稿日期: 2019-01-21

基金项目: 国家重点研发计划 “深地资源勘查开采”重点专项(2018YFC0603202)、国家自然科学基金(41574120)和湖南省研究生科研创新项目(CX20190084)共同资助

摘要

为了整体了解青藏高原东南缘龙门山断裂带的卫星重力场特征及其发震机制, 文中基于重力卫星数据展现出研究区域的地形信息, 并通过解算反映莫霍面起伏的区域重力异常, 反演出研究区域的莫霍面深度, 同时运用GPS速度场数据探测了研究区域地壳的形变速率及方向; 最后, 结合龙门山断裂带的构造背景及前人已有的深部探测结果, 分析其发震前后重力时变场的动态特征并探讨了发震机理。研究结果表明: 青藏高原东缘深部物质逐年向E流展, 在龙门山断裂带处强势受阻, 持续的挤压、碰撞过程形成了深部莫霍面的 “陡降带”, 长期的应力作用利于逆冲推覆和走滑构造的形成; 龙门山地震带震前处于明显的重力变化梯度带, 深部物质运移的流速差异大, 流体压力增强, 有助于地壳流体入侵和软流圈物质底侵上涌; 物质和能量的持续积累最终将导致深部应力严重失衡、深部薄弱构造剪切破裂, 从而发生逆冲、推覆构造运动的大地震。

关键词: 龙门山断裂带; 卫星重力场; 莫霍面; GPS速度场; 重力时变场; 地震

中图分类号:P315.72⁺ 文献标志码:A 文章编号:0253-4967(2019)05-1136-19

CHARACTERISTICS OF SATELLITE GRAVITY FIELD AND SEISMOGENIC MECHANISM IN THE LONGMENSHAN FAULT ZONE

TANG Jing-tian^1,^2,³, YANG Lei¹, REN Zheng-yong^1,^2,³, HU Shuang-gui^1,^2,³, XU Zhi-min⁴

1)School of Geosciences and Info-Physics, Central South University, Changsha 410083, China

2)Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring of Ministry of Education, Central South University, Changsha 410083, China

3)Technical Innovation Center of Coverage Area Deep Resources Exploration, Ministry of Natural Resources, Central South University, Changsha 410083, China

4)Chengde Petroleum College, Hebei Instrument & Meter Engineering Technology Research Center, Chengde 067000, China;

Abstract

Longmenshan fault zone is a famous orogenic belt and seismic zone in the southeastern Tibetan plateau of China. The Wenchuan M_S8.0 earthquake on May 12, 2008 and the Ya’an M_S7.0 earthquake on April 20, 2013 occurred in the central-southern part of Longmenshan fault zone. Because of its complex geological structures, frequent earthquakes and special geographical locations, it has attracted the attention of many scholars around the world. Satellite gravity field has advantages in studying gravity field and gravity anomaly changes before and after earthquake. It covers wide range, can be updated regularly, without difficulty in terms of geographical restrictions, and is not affected by environmental factors such as weather, terrain and traffic. Therefore, the use of high-precision Earth satellite gravity field data inversion and interpretation of seismic phenomena has become a hot topic in earth science research. In order to understand satellite gravity field characteristics of the Longmenshan earthquake zone in the southeastern Tibetan plateau and its seismogenic mechanism of earthquake disasters, the satellite gravity data was used to present the terrain information of the study area. Then, by solving the regional gravity anomaly of the Moho surface, the crustal thickness of the study area was inverted, and the GPS velocity field data was used to detect the crustal deformation rate and direction of the study area. Combining the tectonic setting of the Longmenshan fault zone and the existing deep seismic sounding results of the previous researchers, the dynamic characteristics of the gravity time-varying field after the earthquake in the Longmenshan earthquake zone was analyzed and the mechanism of the earthquake was explored. The results show that the eastward flow of deep materials in the eastern Tibetan plateau is strongly blocked at the Longmenshan fault zone. The continuous collision and extrusion process result in a “deep drop zone” in the Moho surface, and the long-term stress effect is conducive to the formation of thrust-nappe and strike-slip structures. The Longmenshan earthquake zone was in the large-scale gradient zone of gravity change before the earthquake, the deep plastic fluid material transport velocity differed greatly, the fluid pressure was enhanced, and the rock mechanical strength in the seismic source region was weakened, which contributed to the intrusion of crustal fluid and the upwelling of the asthenosphere. As a result, the continuous accumulation of material and energy eventually led to continuous stress imbalance in the deep part and shear rupture of the deep weak structure, causing the occurrence of the thrust-nappe and strike-slip earthquake.

Keyword: Longmenshan fault zone; satellite gravity field; Moho surface; GPS velocity field; gravity time-varying field; earthquake

Show Figures

0 引言

地球重力场模型已逐渐成为人类感知地球内部物质变化的一种重要手段。自21世纪以来, 随着重力卫星技术的快速发展, 国际上先后发射了CHAMP、 GRACE和GOCE 3类地球重力场观测卫星, 极大促进了大地测量学、海洋科学以及固体地球物理学等学科领域的发展。先进的卫星重力测量技术为我们刻画出高精度的全球重力场模型, 为大尺度区域重力异常特征及重力反演研究提供了极大帮助。此外, 卫星重力时变场研究也如雨后春笋般逐渐兴起。 GRACE(Gravity Recovery and Climate Experiment)双星系统提供的月重力场模型为连续监测全球地表物质迁移和再分布情况提供了直观的观测手段, 目前已经逐渐适用于全球区域储水量变化、冰川融化以及地震监测等研究(郑秋月等, 2015; 严畅达等, 2018)。

在基于高精度地球重力场模型解算重力异常的研究中, 郭良辉等(2012)利用优化滤波方法对中国大陆布格重力异常进行了处理和分析; 陈石等(2015)利用EGM2008自由空气异常模型得到了蒙古国周边区域的重力异常分布特征; 郭东美等(2015)基于EIGEN-6C2模型反演分析了青藏高原的地壳特征; Xu等(2017)对青藏高原布格重力异常进行小波多尺度分析, 并进行了莫霍面深度反演研究。在与GRACE卫星重力时变场相关的地震研究中, 段虎荣等(2009)利用GRACE卫星数据分析了汶川地震发震前后的重力场变化; 陈国雄等(2012)以2008年汶川地震为例, 利用GRACE卫星的重力时变场信息, 给出了多尺度变化场与汶川地震的关系; 姜永涛等(2014)基于去相关滤波方法, 结合3次实际地震对青藏高原东缘卫星重力时变场进行了特征分析; 严畅达等(2018)总结了目前GRACE卫星重力数据在地震研究中的应用, 对新一代GRACE-Follow on卫星提高地震重力信号观测精度的发展提出了展望。

对于青藏高原东南缘龙门山断裂带这一热点区域, 目前已有许多地球物理深部探测结果。 Huang等(2002)、楼海等(2008)和郑勇等(2013)利用获得的龙门山断裂带附近区域的波速结构及地壳厚度结果, 揭示了巴颜喀拉块体与四川盆地相互作用及地壳形变的机理; Zhao等(2012)和詹艳等(2013)分析了龙门山断裂带中段及西南段的大地电磁剖面的地壳电性结构, 结果表明其地下高阻体阻碍了青藏高原深部物质向E运移。上述成果多集中于结合详细的区域构造资料, 依据特定地质构造方向的二维剖面测量数据进行相关研究, 而对于大尺度区域的地质构造进行研究则较为困难。与地面测量手段相比, 卫星重力测量技术具有资料定期更新, 研究范围广, 无困难地域限制, 不受天气、地形和交通等环境因素影响的优点, 具有明显优势。目前, 重力卫星测量技术在青藏高原东南缘的应用亦取得了一定成果(陈国雄等, 2012; 姜永涛等, 2014; 毕奔腾等, 2016; Xu et al., 2017), 然而结合高精度卫星重力场的区域地质构造解释及重力时变场的深部物质迁移变化的研究较少。事实上, 多种卫星重力场资料的联合分析能更好地反映研究区域构造特征及其孕震机理。

鉴于此, 本文选取青藏高原东南缘为研究区域, 以地震频发的龙门山断裂带为主要研究目标。基于高精度地球重力场WGM2012(World Gravity Map 2012模型数据展现研究区域的基本区域地质构造概况, 通过解算反映莫霍面起伏的区域重力异常, 利用改进的Parker-oldenburg界面反演方法进行了研究区域莫霍面深度反演研究, 并结合GPS(Global Positioning System)速度场数据分析, 给出龙门山断裂带的孕震构造及深部物质运移特征; 同时以此为基础, 分析了龙门山地震带发震前后GRACE卫星重力时变场的动态变化特征, 揭示了发震前后深部物质流动及能量转换情况, 初步探究其发震机理, 期望利用全球卫星重力测量技术, 对龙门山断裂带的地震监测或短临预报工作提供有价值的参考。

1 青藏高原东南缘区域的构造特征

1.1 区域地形信息

青藏高原东南缘的龙门山断裂带是中国地震和构造运动十分活跃的地带之一, 2008年5月12日的汶川M_S8.0地震和2013年4月20日的雅安市芦山县M_S7.0地震分别发生在龙门山断裂带的中部和南部。龙门山断裂带在仅仅5a内就发生了2次大地震, 多发的地震、复杂的地质构造以及特殊的地理位置使得龙门山断裂带成为研究热点。

WGM2012地球重力场模型是基于全球超高阶重力场模型EGM2008(Earth Gravitational Model 2008)和全球地形高程参考模型ETOPO1(ETOPO1 Global Topography)计算的新型高精度重力场模型, 其重力场模型阶数达到2 190阶, 具有2'× 2'高精度网格数据。该模型参考实际地球模型, 考虑了地表不同组成成分的质量贡献, 根据大地测量学及地球物理学相关原理严格进行相应的重力数据处理, 为我们了解青藏高原东南缘区域(24° ~36° N, 96° ~108° E)的地形及重力异常空间分布特征提供了帮助。

从青藏高原东缘地形图(图1a)中可以发现, 龙门山断裂带整体呈NE走向, 断裂带长约500km, 宽约70km, 由数条大断裂带构成(张培震, 2008)。龙门山断裂带所处的地形是一个较为明显的过渡带, 其区域梯度变化大, 地形坡度陡峭, 西北为青藏高原的羌塘地块及巴颜喀拉地块, 西南侧为川滇地块, 东临四川盆地。龙门山断裂带海拔约1 500m, 其地形高度差最高可达4 500m, 从而有效形成了地形剧烈变化的梯度带。自由空气重力异常是反映地形特征的重力场信息, 地形与自由空气重力异常值呈正相关, 自由空气重力异常值越大则该处地形越高。图1b显示, 龙门山断裂带以西的巴颜喀拉和羌塘块体为较大的正异常, 以东的四川盆地区域则是较小的负异常, 而龙门山断裂带处于一个明显的正、负异常过渡带上, 是巴颜喀拉地块与四川盆地之间的明显界限。

Figure Option
View Download New Window

图 1 青藏高原东南缘地形信息
a 青藏高原东南缘高程; b 青藏高原东南缘自由空气重力异常。黑色线为主要区域构造线(邓起东等, 2002)。 QTB 羌塘地块; CDB 川滇地块; SCB 四川盆地; BHB 巴颜喀拉地块; LMSF 龙门山断裂带。红点为汶川地震震中, 黑点为雅安地震震中Fig. 1 Topographical information of southeastern Tibetan plateau.

1.2 区域莫霍面深度反演

基于正常地球重力场参考模型, 在自由空气重力异常场信息的基础上, 利用Fullea等(2008)提出的大尺度范围地形改正算法, 经中间层校正及基于1'× 1'的ETOPO1高程数据的地形校正来获得布格重力异常数据, 从而将重力异常解算到大地水准面, 其中地壳密度取平均值2.67g/cm³, 地形改正半径取为167km。此外, 布格重力异常是由地表浅部至地壳深部不同地质体所产生的重力异常的叠加场, 消除沉积层的干扰可提高莫霍界面深度的反演精度。近年来, 小波多尺度分析成为位场分离研究中的有用工具(姜磊等, 2014; Xuan et al., 2016), 本文采用小波多尺度分析方法进行沉积层校正, 选用双正交bior 4.4小波基函数进行多尺度分析, 以分离研究区域的布格重力异常, 然后采用径向对数功率谱方法对小波近似信息进行平均场源约束计算, 从而得到反映莫霍面起伏的区域场重力异常。

径向对数功率值与径向频率基本呈线性变化, 通过绘制径向功率谱和圆波数的谱图, 可计算出最低波数段内的曲线斜率并近似推断地下异常场源的深度(Xu et al., 2017):

$\overset{—}{h} \approx - \frac{Δln P}{2 Δ r}$ (1)

式中, Δ lnP和Δ r分别为径向功率谱和圆波数的变化率, $\overset{—}{h}$ 为场源平均深度。

	Figure Option View Download New Window
	图 3 青藏高原东南缘莫霍面深度反演图 a 莫霍面深度正演与观测值之差的数据点统计; b 反演莫霍面的深度Fig. 3 Depth map of Moho surface in southeastern Tibetan plateau.

从图2a中可以发现布格重力异常在四川盆地区域、川滇地块南部的负异常幅值在0~150mGal变化, 而青藏高原的巴颜喀拉、羌塘块体及川滇地块的西北部则处于幅值为400~600mGal的较高负异常区域。布格重力异常与地壳厚度有着密切的联系, 一般而言, 布格重力异常值越小, 区域地壳越厚, 这在一定程度上反映出龙门山断裂带处于地壳厚度变化显著的过渡带上。

	Figure Option View Download New Window
	图 2 青藏高原东南缘反映莫霍面的区域重力异常计算 a 青藏高原东南缘布格重力异常; b 五阶小波近似反映的莫霍面区域异常; c 区域异常径向对数功率谱分析Fig. 2 Regional gravity anomaly calculation of Moho surface in southeastern Tibetan plateau.

之后利用小波多尺度分析来分离研究区域的布格重力异常。图2b为五阶小波的近似信息, 经径向对数功率谱分析(图2c)计算得到的场源平均深度约47.36km, 与全球地壳模型CRUST1.0(Laske et al., 2013)给出的研究区域莫霍面的平均深度相近, 可见五阶小波的近似信息基本反映了研究区域莫霍面起伏的区域重力场。

为了加深对区域构造环境的认识, 下一步利用反映莫霍面起伏的区域重力异常数据进行青藏高原东南缘的莫霍面深度反演。在频率域中莫霍面界面的反演中常使用Parker-oldenburg位场迭代反演方法:

$\begin{array}{l} F [Δ g (x)] = - 2 πG ρ e^{- k z_{0}} \overset{\infty}{\sum_{n = 1}} \frac{{(- k)}^{n - 1}}{n!} F [h^{n} (x)] (2) \\ F [h (x)] = - \frac{F [Δg (x)] e^{- k z_{0}}}{2 πG ρ} + \overset{\infty}{\sum_{n = 2}} F [h^{n} (x)] \cdot \frac{{(- k)}^{n - 1}}{n!} \end{array}$ (3)

$\{\begin{array}{l} filter (k) = \frac{1}{2} [1 + \cos (\frac{k - 2 π a}{2 (b - a)})], a < k < b \\ filter (k) = 0, k > b \\ filter (k) = 1, k < a \end{array}$ (4)

式中, x为位置矢量, Δ g $(x)$ 为区域重力异常, h(x)为参考水平面向下深度, F $[Δ g (x)]$ 和F $[h (x)]$ 为其对应参数的傅里叶变换值; G为万有引力常数, ρ 为反演界面的密度差, z₀为界面平均深度, k为径向圆频率值, filter为一低通滤波器, a和b均为滤波参数。

由于重力异常的高频成分会导致反演迭代结果不稳定, 为保证迭代收敛, 采用改进的Parker-oldenburg反演算法, 利用公式(4)对重力异常傅里叶谱结果进行低通滤波处理, 并对边缘数据进行边界锥形余弦滤波处理(Gó mez-Ortiz et al., 2005)。通过公式(2)和(3)可以发现参考平均深度z₀及壳幔密度差ρ 的选取十分重要, 于是在研究区域采用精准度较高的人工地震探测结果(徐志萍等, 2018), 选用平均深度46km、壳幔密度差0.4g/cm³作为界面反演的初始参考值。经过6次迭代反演计算, 连续2次迭代的莫霍面深度差值的标准差< 0.05, 故结束迭代计算, 得到的莫霍面深度反演结果如图 3所示。对比莫霍界面深度的正演值与观测值之差(图3a)可发现, 大部分数据点分布在-10~10mGal, 其平均值为1.022mGal, 标准差为1.653mGal, 基本符合正态分布特征, 说明界面反演结果可靠性较好。此结果与前人利用重力数据资料反演得到的莫霍面深度基本一致(张恩会等, 2015; Xuan et al., 2016; 徐志萍等, 2018)。

根据莫霍面深度反演结果(图3b), 龙门山断裂带两侧的地壳厚度自西向东逐渐减小。青藏高原东缘的羌塘块体及巴颜喀拉块体的莫霍面深度约达65km, 川滇地块北部区域的莫霍面深度约55km, 四川盆地则约深40km。而龙门山断裂带附近区域的莫霍面深度变化剧烈, 两侧地壳厚度差约达25km, 巨大的地壳差异带使其成为构造活跃区域及应力作用的耦合地段(滕吉文等, 2008)。自NW的巴颜喀拉块体至SE的四川盆地区域, 龙门山断裂带成为该研究区域莫霍面深度变化最为剧烈的地区, 其极有可能是在印度洋板块与欧亚板块长期的地质构造运动中, 深部壳幔物质受到四川盆地强硬地壳的阻挡而形成的。

1.3 区域GPS速度场

此外, 现代大地测量技术能使我们能以更高的精度和时空分辨率认识区域地壳运动以及形变特征, 为探索断层活动和地震特性提供强有力的支持(伍吉仓等, 2018)。 GPS作为一种研究地壳形变的重要技术, 以其定位精度高、全天候作业等优点广泛应用于地壳形变的监测和研究。在图3b所示的莫霍面深度反演结果的基础上, 基于1998— 2014年GPS流动站相对于稳定欧亚大陆速度场的数据资料(Zhao et al., 2015)绘制出GPS速度场, 如图 4所示。

	Figure Option View Download New Window
	图 4 GPS流动站1998— 2014年相对于稳定的欧亚大陆速度场Fig. 4 Horizontal GPS velocities with respect to stable Eurasia from 1998 to 2014.

龙门山断裂带的形成可追溯到新生代时期, 印度洋板块与欧亚板块的一系列碰撞、挤压作用激发了深部壳幔物质的侧向移动, 造成了喜马拉雅造山带构造向NE的侧向挤压, 使得龙门山断裂带发生大规模的构造变形、抬升(楼海等, 2008; 詹艳等, 2013)。从GPS速度场(图4)中可以进一步发现, 在两大板块一系列构造的长期碰撞、挤压作用下, 地壳向青藏高原东缘变形移位, 亦是青藏高原深部物质向E流展的表现; 地表水平形变速率自西向东逐渐减小, 青藏高原整体大构造水平运动速率约20mm/a, 而整个龙门山断裂带的滑动速率约3mm/a, 相对于整个青藏高原大构造环境活动而言, 龙门山断裂带的地表水平变形运动并不明显。此外, 青藏高原深部存在丰富的物质热流活动, 深部物质流至龙门山断裂带附近时, 受到以龙门山为NW边界的四川盆地岩石圈的阻挡, 在形成逆冲推覆构造的同时, 脆弱态的下地壳相互挤压碰撞, 在接触部位形成了破碎带(毕奔腾等, 2016)。部分物质流被稳定的四川盆地刚性块体所阻挡, 受迫向上移动并不断侵入到破碎岩石中(毕奔腾等, 2016; 王志等, 2017), 这有助于该区域地壳的抬升, 也促进了物质流的大量累积。此外, GPS监测数据还显示部分物质受到四川盆地岩石圈的强烈阻挡, 转而向川滇地块SE向流动。

地震发生的主要原因是板块之间相互碰撞、挤压, 从而造成板块边缘及内部发生明显的破裂及位错现象(滕吉文等, 2008; 祝意青等, 2010)。通过分析WGM2012高精度地球重力场信息可以清晰地发现, 龙门山断裂带不仅位于地形差异较大的陡变带, 亦是地壳厚度变化剧烈的过渡带。在大背景下, 青藏高原的下地壳和上地幔主要呈低密度、低速度、高电导率和高泊松比异常的特征, 并存在部分熔融和流体作用(Wang et al., 2010; 王志等, 2017; 雷建设等, 2018; 徐志萍等, 2019)。深部物质东流的进程受到了以龙门山为NW边界的四川盆地刚性块体的阻隔, 使得巴颜喀拉地块和四川盆地之间形成了高强度的挤压状态, 亦促进了应力的不断积累(滕吉文等, 2008; 王志等, 2017)。龙门山断裂带两侧地壳厚度的显著差异反映了地壳内部受到了EW向强烈挤压的趋势, 深部物质流的软弱以及长期向E的挤压应力作用有利于逆冲推覆和走滑构造的形成, 造就了龙门山断裂带特殊的地质构造背景及地震的诱发环境。

2 青藏高原东南缘重力时变场特征

2.1 重力时变场的计算

在巨大地震孕育及发生的过程中, 震源区域会发生较为明显的物质运移及质量分布变化, 重力场值也会产生明显变化。利用GRACE卫星的重力时变场信息, 可以较为精确地监测地震前后物质及能量的变化情况。

由GRACE卫星重力场模型得到的大地椭球面上的重力时变场变化可写为

$Δ g (r, θ, λ) = \frac{GM}{r^{2}} \overset{\infty}{\sum_{n = 2}} (n - 1) {(\frac{R}{r})}^{n} \cdot \overset{n}{\sum_{m = 0}} [(Δ {\overset{—}{C}}_{nm} \cos mλ + Δ {\overset{—}{S}}_{nm} \sin mλ)] \cdot {\overset{—}{P}}_{nm} (\cos θ)$ (5)

式中, θ 、 λ 分别为地心余纬和经度, r为观测点到地球质心之间的距离, Δ g(r, θ , λ )为观测点的重力时变差值, GM为地心引力常数, R为地球的平均半径, n、 m分别为模型位系数的阶和次, ${\overset{—}{P}}_{nm}$ (cosθ )为完全正则化勒让德函数, Δ ${\overset{—}{C}}_{nm}$ 、Δ ${\overset{—}{S}}_{nm}$ 为所选重力场模型的相应阶次的球谐系数差值。

GRACE卫星双星系统通过测量星间距求解月重力场模型, 但由于测量方式及轨道倾角等因素, 使得月重力场模型球谐系数具有很强的相关性, 球谐误差随着阶次的增大而逐渐增加, 研究区重力时变场的解算结果会在SN向产生较为明显的SN向条带状波纹, 因此需要设计平滑核心函数对其位系数进行限制(Swenson et al., 2006)。本文采用德国地学研究中心GFZ(German Research Center for Geosciences)发布的Level-2 RL-05地球重力模型, 该数据为已处理的规格化重力场球谐系数, 最大阶数为90阶, 该模型在计算中已经扣除了大气、海洋极潮、固体潮及海潮等的影响(Christoph et al., 2012), 其半波长空间分辨率可达350km。此外, 为了尽可能地压制周期性区域水文特征引起的重力时变场干扰, 采用每年冬季1— 3月的平均重力场值作为其年度均值。利用德国地学研究中心的Kusche等(2009)提出的去相关平滑滤波方法, 经测验选用优于高斯类滤波结果的DDK5滤波方法(姜永涛等, 2016), 选取青藏高原东缘震源区域范围(24° ~36° N, 96° ~108° E), 以2005— 2009年、 2009— 2013年2类尺度为4a的重力年平均重力场进行年度差分及年度累积计算, 基于此比较重力时变场的动态变化。

2.2 重力时变场的动态变化与发震机制分析

图 5、图6、图 7和图 8分别为2008年汶川地震与2013年雅安地震前后重力时变场的变化特征。从图 5和图 6中可以发现, 汶川地震(红点处)发生前, 2005— 2007年龙门山断裂带附近的重力场值剧烈增大(幅值达约3μ Gal), 并于2008年呈缓慢增大的态势, 正重力变化场的梯度向汶川附近区域逐年增强。雅安位于龙门山断裂带的南侧, 重力场变化则较为平缓。重力时变场变化是空间重力场变化的表现, 可以通过研究区域深部的壳幔物质迁移来定性分析卫星重力场的变化。按照板块运动的地幔对流说(傅容珊等, 1998)相关理论, 将深部壳幔物质看作软流性物质, 若地壳及上地幔物质流动速度增大、壳幔物质增多, 将产生较为明显的正重力场变化; 而地壳及上地幔物质流动速度减小、壳幔物质减少, 则将产生负重力场变化。从2005— 2007年的重力变化场中可以发现, 汶川地区重力场值上升明显, 正重力场变化梯度在2007年达到最大, 龙门山断裂带SW侧较大范围内持续正重力场值的变化, 地下物质会聚流入持续增加, 能量大量累积。与2007年相比, 在2008年则产生了较大的正负重力场变化梯度带, 反映该时间段内地下物质发生较大的差异运动, 处于物质膨胀和收缩的过渡部位(申重阳等, 2009; 陈国雄等, 2012), 潜在震源的相对薄弱部位极易产生剪应力, 使得构造破裂。其次, 比较2008年与2009年汶川地震前后重力场变化可发现, 龙门山断裂带东、西两侧出现了相反的重力场变化, 这有可能是对震后壳幔物质重新调整过程的反映。

	Figure Option View Download New Window
	图 5 青藏高原东南缘年度差分时变重力场(2005— 2009年)Fig. 5 Annual differential time-varying gravity field in southeastern Tibetan plateau(2005— 2009).

	Figure Option View Download New Window
	图 6 青藏高原东南缘年度累积时变重力场(2005— 2009年)Fig. 6 Annual cumulative time-varying gravity field in southeastern Tibetan plateau(2005— 2009).

	Figure Option View Download New Window
	图 7 青藏高原东南缘年度差分时变重力场(2009— 2013年)Fig. 7 Annual differential time-varying gravity field in southeastern Tibetan plateau(2009— 2013).

	Figure Option View Download New Window
	图 8 青藏高原东南缘年度累积时变重力场(2009— 2013年)Fig. 8 Annual cumulative time-varying gravity field in southeastern Tibetan plateau(2009— 2013).

从图 7和图 8中亦可以看出, 雅安地震(黑点处)发生前, 2009— 2013年龙门山断裂带南侧附近的重力场值先较大增长, 后轻微下降, 最后又快速上升, 其累积重力场值保持上升态(约2μ Gal), 说明其深部物质会聚运动在该时间段内总体是增强的。 2009— 2012年的重力场变化则侧面反映出深部物质流先快速积累, 后出现深部物质回流的现象, 这可能是汶川大地震余震的干扰以及大地震对局部重力场的影响(姜磊等, 2014)。该区域重力场值在2012— 2013年又开始快速上升, 深部能量达到临界点时造成深部断层破裂, 从而引发地震。而汶川地震震中区(红点处)的重力值则是逐年减小, 2013年才有增大趋势, 这可能是汶川地震后深部物质部分回流, 震后深部物质重新调整恢复的体现。此外, 还发现该区域震前基本处于正负重力变化梯度带附近, 剪切走滑构造断层在持续失稳的状态下亦容易破碎断裂。

2次地震的显著共同点是震前物质流动速度差异大, 累积重力场动态变化基本处于上升态, 震前震源区基本处在较大的正重力梯度带或正负重力梯度带的中心位置。该区域极易产生物质压缩、膨胀现象, 且又位于震源附近相对薄弱的区域, 在震源附近应力持续失去平衡的影响下极易产生剪应力, 从而使深部区域构造破裂。此外, 结合青藏高原东缘区域的构造背景及前人已有的深部探测结果, 可得到青藏高原东南缘深部的动态影像(图9)。由于印度洋板块向欧亚板块俯冲, 青藏高原深部物质逐年向E流展。因受到四川盆地刚性块体的强势阻挡, 龙门山断裂带下地壳部分流体受迫上移并侵入到孕震层的破碎岩石中, 使流体应力增强, 减弱了岩石的力学强度和断层面处的有效正应力, 改变了局部岩石的剪切应力(王志等, 2017; 雷建设等, 2018), 影响了震区断层活动及地震的触发。此外, 深部物质的东流及SE向的运移迹象, 明显具有向汶川及雅安会聚的趋势, 以致在龙门山断裂带中段及SW段形成了一定的物质和能量的累积。震前深部流体运移流速差异大、运动活跃, 导致震前龙门山断裂带深部的物质流大幅度增加。在青藏高原深部物质东流的动力学影响下, 龙门山断裂带深部形成的高温高压造成壳幔解耦, 壳幔岩石圈物质部分熔融(Wang et al., 2010; 毕奔腾等, 2016; 朱介寿等; 2017)。毕奔腾等(2016)汇总了一系列前人关于青藏高原东缘的地球物理和地球化学深部探测结果, 一系列证据说明了该区域岩石圈底部软流圈物质存在底侵上涌运动, 并有NE向的上涌位移趋势。重力时变场的动态变化说明震前龙门山震区深部物质流积累, 震前物质流动速度差异表明其挤压流动能力增强。面对四川盆地刚性块体的强势阻挡, 高强度的挤压力迫使软流圈物质底侵上涌, 强烈的上拱力亦促进地壳的抬升。物质和能量的持续累积, 影响了龙门山断裂带的隆升和其地震的诱发。

	Figure Option View Download New Window
	图 9 青藏高原东南缘地形及莫霍面深度影像 a 青藏高原东南缘地形及莫霍面深度的三维影像, 红色虚线为断面剖线; b SE向二维剖面深部物质流动示意图, 蓝色箭头为来自GPS的测量结果, 红色箭头引自文献(毕奔腾等, 2016)Fig. 9 The topography and Moho depth image of southeastern Tibetan plateau.

3 讨论与结论

本文利用高精度卫星重力场数据, 对青藏高原东南缘龙门山断裂带的区域构造及重力时变动态场特征进行了细致研究, 并结合前人已有的地球物理、地球化学等深部探测结果, 得到以下几点认识:

(1)印度洋板块向欧亚板块的俯冲等因素激发了深部壳幔物质的流动, 青藏高原深部物质运移逐年向E流展, 受到以龙门山为NW边界的四川盆地刚性块体的强势阻挡, 部分深部物质转而产生了向SE运移的迹象, 造成对龙门山断裂带的碰撞、挤压, 导致其成为地形差异较大的陡变带和深部地壳厚度变化强烈的过渡地段, 长期的应力作用亦促进了其逆冲推覆和走滑构造的形成, 造就了其特殊的地震孕育环境。

(2)通过研究汶川及雅安2次大地震发震前后重力时变场特征, 发现震前物质流动速度差异大, 而年度累积重力场变化处于上升态, 震前震源区基本处在明显的正负重力梯度带的中心位置, 说明其属于震源附近相对薄弱的区域, 容易产生剪应力而使深部构造破裂, 满足剪切走滑发震断层的形成条件。震后龙门山断裂带两侧的重力变化趋势相反, 这可能是对震后壳幔物质重新调整恢复的反映。重力场的动态变化也表明, 震前一段时间内, 重力场值的持续上升和较大的正负重力梯度带变化很可能是深部能量释放的前兆。

(3)结合重力时变动态场特征、研究区的构造背景及前人已有的深部探测研究结果, 初步探讨了龙门山断裂带的发震机理。青藏高原深部物质东流受到四川盆地刚性块体的强势阻挡, 部分下地壳物质流受迫上移, 侵入到孕震层的破碎岩石中, 降低了岩石力学强度和断层面的应力作用, 影响了地震的触发, 却侧面促进了地壳的抬升。震前, 深部物质流体向汶川及雅安流动会聚的趋势明显增强, 震区深部物质流体及其应力累积不断增加并几乎呈现圈闭状态。随着流体压力的增强, 龙门山断裂带深部的软流圈物质被迫底侵上涌, 亦有助于地壳的抬升。这在一定程度上阐释了龙门山断裂带处于地壳厚度变化强烈地段的成因, 亦是地震触发的重要诱因。在这种长期、复杂的深部运动的趋势下, 震区形成了物质和能量的过度累积, 进入临震闭锁状态。应力的持续失衡将造成深部断层走滑破裂, 能量爆发释放, 最终发生逆冲推覆构造运动的大地震。

本文的研究结果也表明, 利用全球重力卫星测量技术研究区域静态重力场及时变重力场动态变化特征, 有益于探索震区的孕震构造环境、发震前后的深部物质流动及能量变化情况。不断发展的高精度重力卫星测量和其它地球物理勘探技术能够为未来地震运动的中长期预测提供重要的震例经验, 为后续发震机理的研究提供有价值的参考。

致谢 BGI官方网站及德国地学研究中心(GFZ)为本研究提供了原始卫星重力数据; 文中部分图件采用GMT软件进行绘制; 审稿专家为本文提供了建设性的修改意见。在此一并表示感谢!

The authors have declared that no competing interests exist.

参考文献

View Option

[1]	毕奔腾, 胡祥云, 李丽清, 等. 2016. 青藏高原东北部多尺度重力场及其地球动力学意义[J]. 地球物理学报, 59(2): 543—555. [本文引用:5]
[2]	BI Ben-teng, HU Xiang-yun, LI Li-qing, et al. 2016. Multi-scale analysis to the gravity field of the northeastern Tibetan plateau and its geodynamic implications[J]. Chinese Journal of Geophysics, 59(2): 543—555(in Chinese). [本文引用:1]
[3]	陈石, 王谦身. 2015. 蒙古及周边地区重力异常和地壳不均匀体分布[J]. 地球物理学报, 58(1): 79—91. [本文引用:1]
[4]	CHEN Shi, WANG Qian-shen. 2015. Gravity anomalies and the distributions of inhomogeneous masses in the crust of Mongolia and its surrounding regions[J]. Chinese Journal of Geophysics, 58(1): 79—91(in Chinese). [本文引用:1]
[5]	陈国雄, 孙劲松, 刘天佑. 2012. GRACE卫星时变重力场的小波多尺度分解: 以2008年汶川 M_S8. 0大地震为例[J]. 武汉大学学报(信息科学版), 37(6): 679—682. [本文引用:3]
[6]	CHEN Guo-xiong, SUN Jin-song, LIU Tian-you. 2012. Wavelet multi-scale decomposition of time variable gravity field detected by GRACE satellite: A case from Wenchuan M_S8. 0 earthquake, 2008[J]. Geomatics and Information Science of Wuhan University, 37(6): 679—682(in Chinese). [本文引用:1]
[7]	邓起东, 张培震, 冉勇康, 等. 2002. 中国活动构造基本特征[J]. 中国科学(D辑), 32(12): 1020—1030. [本文引用:1]
[8]	DENG Qi-dong, ZHANG Pei-zhen, RAN Yong-kang, et al. 2002. Basic characteristics of active tectonics of China[J]. Science in China(Ser D), 32(12): 1020—1030(in Chinese). [本文引用:1]
[9]	段虎荣, 张永志, 刘锋, 等. 2009. 利用GRACE卫星数据研究汶川地震前后重力场的变化[J]. 地震研究, 32(3): 295—298. [本文引用:1]
[10]	DUAN Hu-rong, ZHANG Yong-zhi, LIU Feng, et al. 2009. Study of the gravity variance in Chinese mainland before and after Wenchuan earthquake with GRACE gravity models[J]. Journal of Seismological Research, 32(3): 295—298(in Chinese). [本文引用:1]
[11]	傅容珊, 黄建华, 徐耀民, 等. 1998. 青藏高原—天山地区岩石层构造运动的地幔动力学机制[J]. 地球物理学报, 41(5): 658—668. [本文引用:1]
[12]	FU Rong-shan, HUNAG Jian-hua, XU Yao-min, et al. 1998. Study of the mantle dynamics of the lithosphere movements in the region from Qinghai-Xizang plateau to Tianshan Mountain[J]. Chinese Journal of Geophysics, 41(5): 658—668(in Chinese). [本文引用:1]
[13]	郭东美, 鲍李峰, 许厚泽. 2015. 基于EIGEN-6C2重力场模型反演青藏高原地壳结构[J]. 地球科学, 40(10): 1643—1652. [本文引用:1]
[14]	GUO Dong-mei, BAO Li-feng, XU Hou-ze. 2015. Tectonic characteristics of the Tibetan plateau based on EIGEN-6C2 gravity field model[J]. Earth Science, 40(10): 1643—1652(in Chinese). [本文引用:1]
[15]	郭良辉, 孟小红, 石磊, 等. 2012. 优化滤波方法及其在中国大陆布格重力异常数据处理中的应用[J]. 地球物理学报, 55(12): 4078—4088. [本文引用:1]
[16]	GUO Liang-hui, MENG Xiao-hong, SHI Lei, et al. 2012. Preferential filtering method and its application to Bouguer gravity anomaly of Chinese continent[J]. Chinese Journal of Geophysics, 55(12): 4078—4088(in Chinese). [本文引用:1]
[17]	姜磊, 李德庆, 徐志萍, 等. 2014. 芦山 M_S7. 0地震震前GRACE卫星时变重力场特征研究[J]. 地震学报, 36(1): 84—94. [本文引用:2]
[18]	JIANG Lei, LI De-qing, XU Zhi-ping, et al. 2014. Characteristics of gravity field before the Lushan M_S7. 0 earthquake based on time variable gravity data from GRACE satellite[J]. Acta Seismologica Sinica, 36(1): 84—94(in Chinese). [本文引用:1]
[19]	姜永涛, 王丽美, 高春春, 等. 2016. GRACE揭示的现阶段青藏高原及邻域重力变化趋势[J]. 地震研究, 39(4): 574—578. [本文引用:1]
[20]	JIANG Yong-tao, WANG Li-mei, GAO Chun-chun, et al. 2016. Long-term gravity change of Tibet plateau and its surrounding area derived from GRACE[J]. Journal of Seismological Research, 39(4): 574—578(in Chinese). [本文引用:1]
[21]	姜永涛, 张永志, 王帅, 等. 2014. 基于卫星重力的青藏高原东缘多尺度时变重力场研究[J]. 地震工程学报, 36(2): 353—359. [本文引用:2]
[22]	JIANG Yong-tao, ZHANG Yong-zhi, WANG Shuai, et al. 2014. Study on multi-scale temporal gravity field in the eastern margin of Tibetan plateau based on satellite gravity[J]. China Earthquake Engineering Journal, 36(2): 353—359(in Chinese). [本文引用:1]
[23]	雷建设, 赵大鹏, 徐锡伟, 等. 2018. 龙门山断裂带深部结构与2008年汶川地震发震机理[J]. 科学通报, 63(19): 1906—1916. [本文引用:2]
[24]	LEI Jian-she, ZHAO Da-peng, XU Xi-wei, et al. 2018. Deep structure of the Longmenshan fault zone and mechanism of the 2008 Wenchuan earthquake[J]. Chinese Science Bulletin, 63(19): 1906—1916(in Chinese). [本文引用:1]
[25]	楼海, 王椿镛, 吕智勇, 等. 2008. 2008年汶川 M_S8. 0地震的深部构造环境: 远震P波接收函数和布格重力异常的联合解释[J]. 中国科学(D辑), 38(10): 1207—1220. [本文引用:1]
[26]	LOU Hai, WANG Chun-yong, LÜ Zhi-yong, et al. 2008. The deep tectonic environment of the Wenchuan M_S8. 0 earthquake, 2008: The joint explanation with teleseismic P-wave reception function and Bouguer gravity anomaly[J]. Science in China(Ser D), 38(10): 1207—1220(in Chinese). [本文引用:1]
[27]	申重阳, 李辉, 孙少安, 等. 2009. 重力场动态变化与汶川 M_S8. 0地震孕育过程[J]. 地球物理学报, 52(10): 2547—2557. [本文引用:1]
[28]	SHEN Chong-yang, LI Hui, SUN Shao-an, et al. 2009. Dynamic variations of gravity and the preparation process of the Wenchuan M_S8. 0 earthquake[J]. Chinese Journal of Geophysics, 52(10): 2547—2557(in Chinese). [本文引用:1]
[29]	滕吉文, 白登海, 杨辉, 等. 2008. 2008汶川 M_S8. 0地震发生的深层过程和动力学响应[J]. 地球物理学报, 51(5): 1385—1402. [本文引用:3]
[30]	TENG Ji-wen, BAI Deng-hai, YANG Hui, et al. 2008. Deep processes and dynamic responses associated with the Wenchuan M_S8. 0 earthquake of 2008[J]. Chinese Journal of Geophysics, 51(5): 1385—1402(in Chinese). [本文引用:1]
[31]	王志, 王绪本, 黄润秋, 等. 2017. 龙门山断裂带多参数深部结构成像与地震成因研究[J]. 地球物理学报, 60(6). 2068—2079. [本文引用:4]
[32]	WANG Zhi, WANG Xu-ben, HUANG Run-qiu, et al. 2017. Deep structure imaging of multi-geophysical parameters and seismogenesis in the Longmenshan fault zone[J]. Chinese Journal of Geophysics, 60(6): 2068—2079(in Chinese). [本文引用:1]
[33]	伍吉仓, 王岩, 吴伟伟. 2018. 青藏高原东南缘现今地壳运动速度场分析[J]. 大地测量与地球动力学, 38(2). 116—124. [本文引用:1]
[34]	WU Ji-cang, WANG Yan, WU Wei-wei. 2018. Analysis of current crustal movement velocity field of southeastern Tibetan plateau[J]. Journal of Geodesy and Geodymamics, 38(2): 116—124(in Chinese). [本文引用:1]
[35]	徐志萍, 王夫运, 姜磊, 等. 2018. 川滇地区莫霍面深度和地壳厚度[J]. 地震地质, 40(6): 1318—1331. doi: DOI:10.3969/j.issn.0253-4967.2018.06.009. [本文引用:2]
[36]	XU Zhi-ping, WANG Fu-yun, JIANG Lei, et al. 2018. The depth of Moho interface and crustal thickness in Sichuan-Yunnan region[J]. Seismology and Geology, 40(6): 1318—1331(in Chinese). [本文引用:1]
[37]	徐志萍, 王夫运, 姜磊, 等. 2019. 龙门山中南段地壳上地幔三维密度结构[J]. 地震地质, 41(1): 84—98. doi: DOI:10.3969/j.issn.0253-4967.2019.01.006. [本文引用:1]
[38]	XU Zhi-ping, WANG Fu-yun, JIANG Lei, et al. 2019. The three dimensional density structure of crust and upper mantle in the central-southern part of Longmenshan[J]. Seismology and Geology, 41(1): 84—98(in Chinese). [本文引用:1]
[39]	严畅达, 徐亚. 2018. GRACE卫星重力在地震研究中的应用进展[J]. 地球物理学进展, 33(3): 1005—1012. [本文引用:2]
[40]	YAN Chang-da, XU Ya. 2018. Progress of application of GRACE satellite gravity in research of earthquakes[J]. Progress in Geophysics, 33(3): 1005—1012(in Chinese). [本文引用:1]
[41]	詹艳, 赵国泽, Unsworth M, 等. 2013. 龙门山断裂带西南段4·20芦山7. 0级地震区的深部结构和孕震环境[J]. 科学通报, 58(20): 1917—1924. [本文引用:1]
[42]	ZHAN Yan, ZHAO Guo-ze, Unsworth M, et al. 2013. Deep structure beneath the southwestern section of the Longmenshan fault zone and seismogenic context of the 4·20 Lushan M_S7. 0 earthquake[J]. China Science Bulletin, 58(20): 1917—1924(in Chinese). [本文引用:1]
[43]	张恩会, 石磊, 李永华, 等. 2015. 基于抛物线密度模型的频率域三维界面反演及其在川滇地区的应用[J]. 地球物理学报, 58(2): 556—565. [本文引用:1]
[44]	ZHANG En-hui, SHI Lei, LI Yong-hua, et al. 2015. 3-D interface inversion of gravity data in the frequency domain using a parabolic density-depth function and the application in Sichuan-Yunnan region[J]. Chinese Journal of Geophysics, 58(2): 556—565(in Chinese). [本文引用:1]
[45]	张培震. 2008. 青藏高原东缘川西地区的现今构造变形、应变分配与深部动力过程[J]. 中国科学(D辑), 38(9): 1041—1056. [本文引用:1]
[46]	ZHANG Pei-zhen. 2008. The tectonic deformation, strain distribution and deep dynamic processes in the eastern margin of the Qinghai-Tibet plateau[J]. Science in China(Ser D), 38(9): 1041—1056(in Chinese). [本文引用:1]
[47]	郑秋月, 陈石. 2015. 应用GRACE卫星重力数据计算陆地水变化的相关进展评述[J]. 地球物理学进展, 30(6): 2603—2615. [本文引用:1]
[48]	ZHENG Qiu-yue, CHEN Shi. 2015. Review on the recent developments of terrestrial water storage variations using GRACE satellite-based datum[J]. Progress in Geophysics, 30(6): 2603—2615(in Chinese). [本文引用:1]
[49]	郑勇, 葛粲, 谢祖军, 等. 2013. 芦山与汶川地震震区地壳上地幔结构及深部孕震环境[J]. 中国科学(D辑), 43(6): 1027—1037. [本文引用:1]
[50]	ZHENG Yong, GE Can, XIE Zu-jun, et al. 2013. Crustal and upper mantle structure and the deep seismogenic environment in the source regions of the Lushan earthquake and the Wenchuan earthquake[J]. Science in China(Ser D), 43(6): 1027—1037(in Chinese). [本文引用:1]
[51]	祝意青, 梁伟锋, 徐云马, 等. 2010. 汶川 M_S8. 0地震前后的重力场动态变化[J]. 地震学报, 32(6): 633—640. [本文引用:1]
[52]	ZHU Yi-qing, LIANG Wei-feng, XU Yun-ma, et al. 2010. Dynamic variation of gravity field before and after Wenchuan M_S8. 0 earthquake[J]. Acta Seismologica Sinica, 32(6): 633—640(in Chinese). [本文引用:1]
[53]	朱介寿, 王绪本, 杨宜海, 等. 2017. 青藏高原东缘的地壳流及动力过程[J]. 地球物理学报, 60(6): 2038—2057. [本文引用:1]
[54]	ZHU Jie-shou, WANG Xu-ben, YANG Yi-hai, et al. 2017. The crustal flow beneath the eastern margin of the Tibetan plateau and its process of dynamics[J]. Chinese Journal of Geophysics, 60(6): 2038—2057(in Chinese). [本文引用:1]
[55]	Christoph D, Frank F, Christian G, et al. 2012. GFZ GRACE level-2 processing document for level-2 product: Release 0005 [R]. Potsdam: GFZ: 20. [本文引用:1]
[56]	Fullea J, Fern’ndez M, Zeyen H. 2008. FA2BOUG: A FORTRAN 90 code to compute Bouguer gravity anomalies from gridded free-air anomalies: Application to the Atlantic-Mediterranean transition zone[J]. Computers & Geosciences, 34(12): 1665—1681. [本文引用:1]
[57]	Gómez-Ortiz D, Agarwal B N P. 2005. 3DINVER. M: A MATLAB program to invert the gravity anomaly over a 3-D horizontal density interface by Parker-Oldenburg’s algorithm[J]. Computers & geosciences, 31(4): 513—520. [本文引用:1]
[58]	Huang J L, Zhao D P, Zheng S H. 2002. Lithospheric structure and its relationship to seismic and volcanic activity in southwest China[J]. Journal of Geophysical Research: Solid Earth, 107(B10): 13—14. [本文引用:1]
[59]	Kusche J, Schmidt R, Petrovic S, et al. 2009. Decorrelated GRACE time-variable gravity solutions by GFZ, and their validation using a hydrological model[J]. Journal of Geodesy, 83(10): 903—913. [本文引用:1]
[60]	Laske G, Master G, Ma Z T, et al. 2013. Update on CRUST1. 0: A 1-degree global model of earth’s crust[C]. Geophysics Research Abstracts, vol 15, EGU2013: 2658. [本文引用:1]
[61]	Swenson S, Wahr J. 2006. Post-processing removal of correlated errors in GRACE data[J]. Geophysical Research Letters, 33(8): L08402. [本文引用:1]
[62]	Wang Z, Zhao D, Wang J. 2010. Deep structure and seismogenesis of the north-south seismic zone in southwest China[J]. Journal of Geophysical Research: Solid Earth, 115(B12): B12334. [本文引用:2]
[63]	Xu C, Liu Z W, Luo Z C, et al. 2017. Moho topography of the Tibetan plateau using multi-scale gravity analysis and its tectonic implications[J]. Journal of Asian Earth Sciences, 138: 378—386. [本文引用:3]
[64]	Xuan S B, Shen C Y, Li H, et al. 2016. Structural interpretation of the Chuan-Dian block and surrounding regions using discrete wavelet transform[J]. International Journal of Earth Sciences, 105(5): 1591—1602. [本文引用:2]
[65]	Zhao B, Huang Y, Zhang C H, et al. 2015. Crustal deformation on the Chinese mainland during 1998—2014 based on GPS data[J]. Geodesy and Geodynamics, 6(1): 7—15. [本文引用:1]
[66]	Zhao G Z, Unsworth M, Zhan Y, et al. 2012. Crustal structure and rheology of the Longmenshan and Wenchuan M_W7. 9 earthquake epicentral area from magnetotelluric data[J]. Geology, 40(12): 1139—1142. [本文引用:]

2016

0.0

... 毕奔腾等, 2016 ...

... 此外, 青藏高原深部存在丰富的物质热流活动, 深部物质流至龙门山断裂带附近时, 受到以龙门山为NW边界的四川盆地岩石圈的阻挡, 在形成逆冲推覆构造的同时, 脆弱态的下地壳相互挤压碰撞, 在接触部位形成了破碎带(毕奔腾等, 2016) ...

... 部分物质流被稳定的四川盆地刚性块体所阻挡, 受迫向上移动并不断侵入到破碎岩石中(毕奔腾等, 2016 ...

... 毕奔腾等, 2016 ...

... 毕奔腾等(2016)汇总了一系列前人关于青藏高原东缘的地球物理和地球化学深部探测结果, 一系列证据说明了该区域岩石圈底部软流圈物质存在底侵上涌运动, 并有NE向的上涌位移趋势 ...

2016

0.0

2015

0.0

陈石, 王谦身. 2015. 蒙古及周边地区重力异常和地壳不均匀体分布[J]. 地球物理学报, 58(1): 79—91.

The Mongolia arc structural belt has a complicated formation and development history. It is an important boundary that affect the internal tectonic deformation and evolution of China mainland. From the gravity anomalies to study the crustal deformation patterns, to further reveal the continental crust and lithospheric mantle deformation characteristics in the northern part of China, has an important geological and geophysical significance.#br#In order to study the density structures of sources at different depths, we need to separate the gravity anomalies and understand tectonic deformation patterns from the field sources of different depths. First, on the basis of the EGM2008 model, we have the free air gravity anomalies, using the gravity correction approach, calculated Bouguer and Airy gravity anomalies in the Mongolia and its surrounding areas. Then, introduce the Crust 1.0 crustal model, and establish the reference density model from the surface to the 55 km depth in Mongolia and its surrounding areas. Finally, we extract 6 profiles and determine the hierarchical gravity anomalies in each profile, and make the quantitative analysis on gravity anomaly features of the crust-lithospheric mantle density field source bodies of in Mongolia and its surrounding areas.#br#On the basis of above data and methods, our results include: (1) The gravity anomaly models in Mongolia and its surrounding areas indicate the obvious differences of tectonic deformation between the Hangai dome and Altay Mountains in the western region. There is no area with apparent high-value isostatic gravity anomalies in the Hangai dome. But the high-value regions are concentrated in the southwest of the Altay Mountains. We infer that this feature is related to the new tectonic movement.(2)Based on the reference crustal density model by Crust 1.0, we have investigated the distributions of lateral inhomogeneous bodies in the crust and upper mantle, and find the lateral density structure is important to forward modeling of the gravityanomalies caused by the Moho relief. (3)We have the comprehensive profiles with density structure from the Crust 1.0 model. The results indicate that the density feature is consistent with deep crustal deformation and structures of large active faults.#br#From the analysis of the hierarchical gravity anomalies and the reference density models, we suggest a dynamic model that accounts for the deep crustal movement from west to east in Mongolia, in which the force source is not from the squeeze process between the Siberian plate and Eurasian plate, rather related to the northeastward movement of the block in the southwestern Altay Mountains. The formation of strike-slip faults in both north and south of the Hangai Dome is also related to this movement. The deformed interface determined by the Crust 1.0 model is consistent with the distribution of surface ruptures. There is the correlation between the topography and deep deformation, which consists with the isostatic hypothesis. Therefore, it is a feasible approach to improve gravity inversion using Bouguer gravity anomalies based on the lateral inhomogeneous density model in regions with complex structures. Our results also indicate that the consistency between deep and shallow mass distribution and deformation in the arc-like tectonic deformation zone of Mongolia.

1. Institute of Geophysics, China Earthquake Administration, Beijing 100081, China; 2. Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China

基于全球EGM2008自由空气重力异常模型,本文计算了蒙古及周边地区的布格重力异常和Airy-Heiskanen均衡重力异常.在此基础上,本文采用Crust 1.0地壳模型为参考,通过重力正演方法,对蒙古及周边地区不同深度地壳密度结构模型的重力异常进行了计算,并对得到的正演布格重力异常与实际重力异常进行了对比和分析.研究结果表明:蒙古西部杭爱山地区与阿尔泰山地区的构造变形差异性明显,现今均衡重力异常中杭爱山周边没有明显的均衡异常高值区,而阿尔泰山地区西南方向存在均衡重力异常高值分布,分析与新构造运动密切相关;Crust 1.0模型给出的壳幔横向密度不均匀体分布对于计算Moho面起伏引起的重力异常作用明显;Crust 1.0给出的地壳内界面变形可以反映深大活动断裂的深部构造变形.研究结果对于认识蒙古东西部构造特征差异,以及现今西部活动断裂的地球物理场特征具有参考意义,也可以为进一步应用Crust 1.0模型为参考开展三维密度结构反演提供一定帮助.

... 陈石等(2015)利用EGM2008自由空气异常模型得到了蒙古国周边区域的重力异常分布特征 ...

2015

0.0

2012

0.0

陈国雄, 孙劲松, 刘天佑. 2012. GRACE卫星时变重力场的小波多尺度分解: 以2008年汶川 M_S8. 0大地震为例[J]. 武汉大学学报(信息科学版), 37(6): 679—682.

The time variable gravity in Wenchuan and its surrounding areas was calculated by using the GRACE data published by UTCSR.Then the multi-scale wavelet analysis was used to decompose the satellite gravity field,therefore,details and approximations reflecting different depth of geological bodies were obtained.By studying changes in the gravity field of every detail and approximation component,combining with the features of regional tectonic movement,a preliminary explanation and discussion on the features of time variable gravity as well as the dynamic mechanism of Wenchuan earthquake was presented.The results show that the change in gravity which caused by the tectonic deformation and deep mass transfer in Wenchuan area is revealed effectively by the fourth-order detail and approximation.Moreover,the features of changes in gravity before earthquake meet the conditions of earthquake preparation and occurrence.

1 Institute of Geophysics and Geomatics,China University of Geoscience,388 Lumo Road,Wuhan 430074,Cina

根据UTCSR发布的GRACE卫星资料,计算了汶川及其周边地区的卫星时变重力场,并利用小波多尺度分析方法对卫星重力场进行分解,得到了反映不同深度的重力场细节和逼近。通过研究各个细节和逼近成分的重力场变化,结合区域构造运动的特点,对GRACE观测的时变重力场及汶川地震的动力机制进行了初步解释和讨论。结果表明,4阶细节和4阶逼近较好地揭示了汶川地区构造形变和深部物质流动引起的重力变化,而且这种震前重力变化特征满足地震孕育发生的条件。

... 陈国雄等(2012)以2008年汶川地震为例, 利用GRACE卫星的重力时变场信息, 给出了多尺度变化场与汶川地震的关系 ...

... 目前, 重力卫星测量技术在青藏高原东南缘的应用亦取得了一定成果(陈国雄等, 2012 ...

... 陈国雄等, 2012), 潜在震源的相对薄弱部位极易产生剪应力, 使得构造破裂 ...

2012

0.0

2002

0.0

2002

0.0

2009

0.0

... 在与GRACE卫星重力时变场相关的地震研究中, 段虎荣等(2009)利用GRACE卫星数据分析了汶川地震发震前后的重力场变化 ...

2009

0.0

1998

0.0

傅容珊, 黄建华, 徐耀民, 等. 1998. 青藏高原—天山地区岩石层构造运动的地幔动力学机制[J]. 地球物理学报, 41(5): 658—668.

The structures and neotectonic movements of the lithosphere in the region from Qnghai-Xizang Plateau to Tianshan Mountain System have been inveshgared in this paper. Two scale mantle convection models are employed to explain the lithospheric dynamics of this region. One is a large scale mantle convection pattem, which is calculated by using the global geoid anomalies, plare absolute motions and mantle seismic tomography data set,and anather is a small scale convection model in the upper mantle of this region constrained on the data of the regional isostatic gravity anomalies. Comparing these models with the geophysical and geodetic observations shows that the large scale mande convechon may drive the lithosphere motions of the China continent from North-South direction in the west part to the North-East or South-East direction in the east part. Meantime, the small scale convection system in the upper mantle of this region may control the recent uplift movements of the Qnghai-Xizang Plateau and the Tianshan Mountain System. Furthermore, a three step uplift model of the Qinghai-Xizang Plateau: "uplift after break off-uplift with compression-uplift with small scale convechon" is to be proposed and the deep mantle dynamical background of the lithosphere movements in this region are to be discussed as well.

University of Science and Technology of China, Hefei, 230026, China

利用全球重力大地水准面异常、板块绝对运动及全球地震层析成像数据，计算了青藏高原—天山地区岩石层下部地幔大尺度对流格局以及此种尺度对流驱动下岩石层内应力场分布；同时，利用区域均衡重力异常数据反演青藏高原中、北部到天山地区上地幔小尺度对流模型．结果表明，大尺度的地幔物质运移过程可能驱动着中国大陆岩石层整体从西部以南北方向为主的运动转向东部地区以北东和南东方向的运动；而该区域上地幔小尺度上升流动支持了现代青藏高原和天山地区的抬升运动．提出和讨论了青藏高原隆升的“断离隆升—挤压隆升—对流隆升”三阶段模式，并探讨了大陆岩石层构造运动的地幔深部动力学背景．

... 按照板块运动的地幔对流说(傅容珊等, 1998)相关理论, 将深部壳幔物质看作软流性物质, 若地壳及上地幔物质流动速度增大、壳幔物质增多, 将产生较为明显的正重力场变化 ...

1998

0.0

2015

0.0

... 郭东美等(2015)基于EIGEN-6C2模型反演分析了青藏高原的地壳特征 ...

2015

0.0

2012

0.0

郭良辉, 孟小红, 石磊, 等. 2012. 优化滤波方法及其在中国大陆布格重力异常数据处理中的应用[J]. 地球物理学报, 55(12): 4078—4088.

We presented the preferential filtering method for denoising and anomaly separation of gravity data based on Green's equivalent-layer concept and Wiener filter. Compared to the conventional upward continuation and the preferential continuation, the preferential filtering method has the advantage of no requirement of continuation height. The data test on the synthetic gravity data showed that the preferential filtering method produced better separation of gravity anomaly than both the band-pass filtering and the conventional upward continuation. We then used the preferential filtering method to suppress the noise in the Bouguer gravity anomaly data of Chinese continent and separate the regional gravity anomaly. Then with constraints of the scattered Moho depth information deduced from the deep seismic detection, the regional gravity anomaly was inversed to obtain the Moho depth distribution of the whole Chinese continent. The preferential filtering method provides technique support for the deep detection and tectonic study of Chinese continent.

1. Key Laboratory of Geo-detection (China University of Geosciences, Beijing), Ministry of Education, Beijing 100083, China; 2. School of Geophysics and Information Technology, China University of Geosciences, Beijing 100083, China; 3. Institute of Geophysics, China Earthquake Administration, Beijing 100081, China

在优选延拓法的理论基础上,研究提出基于格林等效层概念和维纳滤波器的优化滤波法,用于对重力异常数据进行去噪和分离.与传统向上延拓法和优选延拓法相比,优化滤波法分离异常与延拓高度无关,不需要已知延拓高度,具有一定的优势.理论重力模型数据的去噪和异常分离试验表明优化滤波法有效,异常分离效果优于传统向上延拓法和带通滤波法.利用优化滤波法对中国大陆重力异常数据去噪和异常分离,得到有效的布格重力异常和区域重力异常.以中国大陆深地震探测推断的莫霍面深度信息为约束,对区域重力异常数据进行密度界面约束反演,得到中国大陆莫霍面深度分布.本文方法为中国大陆深部探测和区域构造研究提供一定的技术支撑.

... 在基于高精度地球重力场模型解算重力异常的研究中, 郭良辉等(2012)利用优化滤波方法对中国大陆布格重力异常进行了处理和分析 ...

2012

0.0

2014

0.0

... 近年来, 小波多尺度分析成为位场分离研究中的有用工具(姜磊等, 2014 ...

... 2012年的重力场变化则侧面反映出深部物质流先快速积累, 后出现深部物质回流的现象, 这可能是汶川大地震余震的干扰以及大地震对局部重力场的影响(姜磊等, 2014) ...

2014

0.0

2016

0.0

... 利用德国地学研究中心的Kusche等(2009)提出的去相关平滑滤波方法, 经测验选用优于高斯类滤波结果的DDK5滤波方法(姜永涛等, 2016), 选取青藏高原东缘震源区域范围(24#cod#x000b0 ...

2016

0.0

2014

0.0

姜永涛, 张永志, 王帅, 等. 2014. 基于卫星重力的青藏高原东缘多尺度时变重力场研究[J]. 地震工程学报, 36(2): 353—359.

In this paper，we calculate annual satellite gravity changes in the eastern margin of the Tibetan Plateau from 2003 to 2012 by using the Gravity Recovery and Climate Experiment（GRACE） time-variable gravity field models with decorrelated filtering.By contraposing the three strong earthquakes that occurred recently in this region，including the Wenchuan MS8.0，Yushu MS7.1，and Ya’an MS7.0 earthquakes，we analyze the changes in satellite gravity field features related to the strong earthquakes with the annual gravity changes before and after each earthquake.For analyzing the relationship between these earthquakes and the gravity change mode in detail，we also caculate the monthly gravity change for the Wenchuan MS8.0 and Yushu MS7.1 earthquakes.Moreover, we use the recently relesed weekly GRACE gravity models to caculate the point-wise gravity change near Longmenshan fault for the Wenchuan MS8.0 earthquake.The regional annual differentiated dynamic gravity changes image indicates that all three earthquakes occurred in a period of small gravity change，which means a small migration of crust-mantle material occurred according to the mantle convection theory.The same feature is indicated in the regional annual cumulative dynamic gravity changes image.These gravity changes likely indicate a medium-short term earthquake precursor.In the monthly differentiated dynamic gravity changes image，the gravity change mode shows significant changes in both the Wenchuan MS8.0 earthquake and the Yushu MS7.1 earthquake，which indicates that the gravity gradient change direction altered from perpendicular to the plane of a fault to parallel.These change in the gravity change mode may confirm the theory of post-earthquake potential field restoration.The nine points of the weekly gravity change sequences in Longmenshan fault area that occurred during 2008 indicate that a nearly nine-week contrary gravity change trend occurred between the eastern and western areas of the fault after the occurrence of the Wenchuan MS8.0 earthquake.These gravity mode changes may be the reflection of an adjustment procedure of the crust-mantle material after the earthquake.Gravity measurement is an important subject in earthquake prediction research.Satellite temporal gravity analysis greatly enriches and complements the deficiency of the ground gravity network in time and space scales and can be quickly used for the study of crustal movement and earthquake prediction.

利用经过去相关滤波处理的GRACE时变重力场模型获得了青藏高原东缘2003—2012年的卫星年重力变化图像，并针对该区域近年发生的三次特大地震，结合震前及震后月重力场变化图像，分析与强震有关的卫星重力场变化特征。从区域年重力变化图像可以看出，三次大震均发生在年重力变化较低的时段内，震前小幅值变化可能是地震发生的中短期前兆；从汶川地震和玉树地震发生前后的月重力场变化图像可以发现，发震前后断层附近的重力变化模式发生变化，这可能印证了震后位场变化恢复理论；从汶川地震前后的龙门山断层附近点上的周重力变化趋势可以明显发现，汶川地震发生（第20周）后近9周的时间，断层东西侧呈现了相反的重力变化特征，这可能是对震后壳幔物质调整过程的反映。

... 姜永涛等(2014)基于去相关滤波方法, 结合3次实际地震对青藏高原东缘卫星重力时变场进行了特征分析 ...

... 姜永涛等, 2014 ...

2014

0.0

2018

0.0

... 雷建设等, 2018 ...

... 雷建设等, 2018), 影响了震区断层活动及地震的触发 ...

2018

0.0

2008

0.0

... 龙门山断裂带的形成可追溯到新生代时期, 印度洋板块与欧亚板块的一系列碰撞、挤压作用激发了深部壳幔物质的侧向移动, 造成了喜马拉雅造山带构造向NE的侧向挤压, 使得龙门山断裂带发生大规模的构造变形、抬升(楼海等, 2008 ...

2008

0.0

2009

0.0

申重阳, 李辉, 孙少安, 等. 2009. 重力场动态变化与汶川 M_S8. 0地震孕育过程[J]. 地球物理学报, 52(10): 2547—2557.

Based on the gravity observations from 1998 to 2007 of China mainland which repeated every 2~3 years, and considering the GPS and leveling measurements and the regional environment of tectonics and geodynamics, we analyze the characteristics of regional dynamic gravity field evolution and the preparation mechanics of the Wenchuan M s 8.0 earthquake in 2008. The result shows that variations of the regional gravity field reflect the dynamic effect of eastward flow of the Tibetan plateau and the preparation information of mid-and-long term, that is 2 to 10 years, of the Wenchuan earthquake. The obvious gravity indicators related to the Wenchuan earthquake preparation are the positive gravity changes in southwest of epicenter which lasted many years, which is in favor of the continuous accumulation of seismic energy, and the relatively large-scale gradient zone of gravity changes, which is in favor of occurrence of seismic rupture. The gravity change process associated with the earthquake preparation is increasing, accelerated increasing and decelerated increasing followed by the occurrence of the earthquake. The total gravity change accumulated over 8 years may come up to 200×10 -8 m·s -2 . The preparation and occurrence of the Kunlunshan M s 8.1 earthquake in 2001 and the adjustment after this earthquake significantly affected the regional dynamic gravity changes and the preparation of the Wenchuan earthquake. The gravity change in the Songpan-Garzê block is negative, which may reflects the local uplift of deep crust-mantle and the crust expansion that resulted from the relatively high temperature inside, which might promote the thrust or nappe movement and the occurrence of the large earthquake.

1 Institute of Seismology, China Earthquake Administration, Wuhan 430071,China 2 Crustal Movement Laboratory, Wuhan 430071,China

基于中国大陆1998～2007年（复测周期2～3年）流动重力观测数据，结合GPS、水准观测成果和区域地质构造动力环境，分析研究了汶川8.0级地震区域重力场动态变化演化特征和孕震机理.结果表明：区域重力场动态演化大体反映了青藏高原物质东流的动态效应和汶川大震孕育的中长期（2～10年）信息；汶川大震孕育的显著重力标志为震中西南持续多年的正重力变化（上升）和出现较大规模的重力变化梯级带，前者有利于地震能量的不断积累，后者有利于地震剪切破裂的发生；与地震孕育相关重力场变化总体呈增大—加速增大—减速增大—发震的过程；8年累积重力变化幅差最大约200×10 -8 m·s -2 ；2001年昆仑山口8.1级地震孕育发生和震后恢复调整，对区域重力场动态变化和汶川大震的孕育发展具有重要影响；松潘—甘孜块体一般呈现负重力变化，可能反映深部壳幔局部上隆、壳内温度较高而膨胀，有利于逆冲或推覆体运动的形成和大震的发生.

... 与2007年相比, 在2008年则产生了较大的正负重力场变化梯度带, 反映该时间段内地下物质发生较大的差异运动, 处于物质膨胀和收缩的过渡部位(申重阳等, 2009 ...

2009

0.0

2008

0.0

滕吉文, 白登海, 杨辉, 等. 2008. 2008汶川 M_S8. 0地震发生的深层过程和动力学响应[J]. 地球物理学报, 51(5): 1385—1402.

The Wenchuan M s 8.0 earthquake of 2008 occurred on the Longmenshan tectonic zone which seems not very active from available data. And it is reported that no noticeable precursors have been observed before the event. It might imply that the crustal deformation at depth associated with this gigantic shock had no expression on the surface or shallow subsurface which was detectable. In this article, we attempt to analyze the processes of brewing and generation of this event at depth that can account for the reason of this devastating earthquake. Tectonically, it is obvious that the India-Eurasia plate collision serves as the most important dynamic setting of major earthquakes in southwestern China. The eastern end of the Himalayan chain is moving rapidly in NNE direction into the Tibetan plateau, forcing deep material to flow toward east. Due to the obstruction of the NE trending Longmenshan tectonic zone in the northwestern margin of the Sichuan basin, the partial flow turns to southeast. Longmenshan is a conspicuous boundary of topography and crust structure, across which the elevation difference is as large as 3500±500 m, with crustal thickness 60±5km in its west and 40±2 km in its east, respectively. Thus it must experiences intense deformation and stress accumulation, though very slow without appreciable manifestation. Based on available data, it is inferred that material of lower crust and covering strata of upper mantle is flowing eastward along two deep slide surfaces, of which one is the layer of low-velocity and low-resistivity (20~25 km depth), and the other is the top of mantle asthenosphere. Hindered by rigid mattter beneath the Sichuan basin, the flow of lower crust and covering strata of upper mantle material turns overthrusting at high angles at the Longmenshan. On the other hand, three thrust faults visible west of the Longmenshan extend downward to depth 20 km, where they converge into one active zone. Thus the Longmenshan tectonic zone is a place where occurs the coupling between slippage on faults of shallow subsurface and material flow of lower crust and covering strata of upper mantle. Under the intense exchange of matter and energy, the source medium ruptures suddenly at a large scale to release a huge amount of elastic strain, resulting in the Wenchuan M s 8.0 earthquake.

Institute of Geology and Geophysics, Chinese Academy of Sciences,Beijing 100029,China

汶川 M s 8.0强烈地震发生在一条现今并不活动的龙门山构造带上，造成了以汶川、映秀为中心及其周边地域的严重破坏和人员的重大伤亡.然而强烈震发生前却未见有可能的确切征兆或浅表层异常活动，即浅层过程与地震发生的深层过程并不匹配.为此对这次强烈地震“孕育”、发生和发展的深层过程进行了分析和探讨,初步研究表明：①在印度洋板块与欧亚板块陆—陆碰撞、挤压作用下，喜马拉雅造山带东构造结向NNE方向顶挤、楔入青藏高原东北缘，迫使高原深部物质向东流展，在受到以龙门山为西北边界的四川盆地阻隔下，一部分物质则转而向东南侧向运移；②龙门山地带在地形上差达3500±500 m左右，而地壳厚度在龙门山西北部为60±5 km左右，四川盆地为40±2 km左右，而龙门山地带与其东、西两侧相比则为地壳厚度变化幅度达15~20 km的突变地域，即为应力作用的耦合地带；③中、下地壳和地幔盖层物质以地壳低速层、低阻层（深20~25 km）为第一滑移面，以上地幔软流层顶面为第二滑移面，且在四川盆地深部“刚性”物质阻隔下，深部壳、幔物质以高角度在龙门山构造带和四川盆地的耦合地带向上运移（或称逆冲），且在龙门山地表三条断裂构成的断裂系向下延伸到20 km左右深处汇聚，二者强烈碰撞、挤压、震源介质破裂；在物质与能量的强烈交换下，应力得到释放，故形成了这次 M s .0强烈地震.为此从深部初步揭示了这次强烈地震“孕育”、发生和发展的深层动力过程.

... 而龙门山断裂带附近区域的莫霍面深度变化剧烈, 两侧地壳厚度差约达25km, 巨大的地壳差异带使其成为构造活跃区域及应力作用的耦合地段(滕吉文等, 2008) ...

... 地震发生的主要原因是板块之间相互碰撞、挤压, 从而造成板块边缘及内部发生明显的破裂及位错现象(滕吉文等, 2008 ...

... 深部物质东流的进程受到了以龙门山为NW边界的四川盆地刚性块体的阻隔, 使得巴颜喀拉地块和四川盆地之间形成了高强度的挤压状态, 亦促进了应力的不断积累(滕吉文等, 2008 ...

2008

0.0

2017

0.0

... 王志等, 2017), 这有助于该区域地壳的抬升, 也促进了物质流的大量累积 ...

... 王志等, 2017 ...

... 王志等, 2017) ...

... 因受到四川盆地刚性块体的强势阻挡, 龙门山断裂带下地壳部分流体受迫上移并侵入到孕震层的破碎岩石中, 使流体应力增强, 减弱了岩石的力学强度和断层面处的有效正应力, 改变了局部岩石的剪切应力(王志等, 2017 ...

2017

0.0

2018

0.0

... 3 区域GPS速度场此外, 现代大地测量技术能使我们能以更高的精度和时空分辨率认识区域地壳运动以及形变特征, 为探索断层活动和地震特性提供强有力的支持(伍吉仓等, 2018) ...

2018

0.0

2018

0.0

... 的选取十分重要, 于是在研究区域采用精准度较高的人工地震探测结果(徐志萍等, 2018), 选用平均深度46km、壳幔密度差0 ...

... 徐志萍等, 2018) ...

2018

0.0

2019

0.0

... 徐志萍等, 2019) ...

2019

0.0

2018

0.0

... 严畅达等, 2018) ...

... 严畅达等(2018)总结了目前GRACE卫星重力数据在地震研究中的应用, 对新一代GRACE-Follow on卫星提高地震重力信号观测精度的发展提出了展望 ...

2018

0.0

2013

0.0

詹艳, 赵国泽, Unsworth

, 等. 2013. 龙门山断裂带西南段4·20芦山7. 0级地震区的深部结构和孕震环境[J]. 科学通报, 58(20): 1917—1924.

2008年汶川 M w7.9级地震后, 于2009~2011年间, 穿过汶川地震所在的龙门山断裂带中段和2013年芦山7.0级地震所在的龙门山断裂带西南段, 沿两条剖面开展了大地电磁观测, 使用相位张量和二维反演等多种先进的分析解释方法, 获得了两条剖面的可信度高的二维电性结构. 通过对龙门山西南段和芦山地震区深部结构的深入研究发现, 它们与龙门山断裂带中段及汶川地震区深部结构既有相同之处, 也有明显差别. 西南段西侧松潘-甘孜地块上地壳高阻层下约10 km深度出现地壳高导层(HCL), 比中段的深度小. 西南段自地表到上地幔顶部也存在龙门山高阻体(HRB), 但是龙门山高阻体规模比中段的小, 其内既没有与西侧地壳高导层相通的低阻层, 也没有像汶川地震区那样在龙门山高阻体内部震源区附近出现局部的低阻区. 芦山地震发生在龙门山高阻体东侧边界附近, 在双石-大川断裂附近, 但是汶川地震发生在龙门山高阻体内部, 这些构成了芦山地震和汶川地震孕震环境的差异. 对青藏高原东部地壳高导层与地壳变形的关系, 以及龙门山断裂带不同位置潜在的地震危险性进行了分析.

... 詹艳等, 2013) ...

2013

0.0

2015

0.0

张恩会, 石磊, 李永华, 等. 2015. 基于抛物线密度模型的频率域三维界面反演及其在川滇地区的应用[J]. 地球物理学报, 58(2): 556—565.

As an important way to reveal the internal structure of the earth, inversion of the density interface is a main subject of gravity research for a long time. Inversion of field data in the frequency domain provides a quick and efficient data analysis method. Combining the advantages of the frequency domain and parabolic density function, this paper presents a processing method for gravity anomaly data of a three-dimensional underground interface using parabolic density-depth function in the frequency domain. The method we used to make forward modeling of the gravity anomaly is based on the algorithm proposed by Parker and Oldenburg which uses constant density contrast. We substituted the constant density function in gravity forward formula for the parabolic density function. Then the Fast Fourier Transform (FFT) was applied to both sides of the equation. We derived the gravity anomaly formula of 3D density interface with parabolic density contrast after calculating the involved integral. To invert the density interface depth, an iterative algorithm was preformed to adjust the interface depth step by step until the difference between the theoretical anomalies and observations reduced to the threshold value. In addition, in order to avoid the divergence of the algorithm which is difficult to avoid for the calculation in the frequency domain, we also added a low-pass filter to the codes. We applied our method to a theoretical model which has 91×71 grid data with 10 km interval in x -axis and y -axis, and a complex 3D interface with parabolic density contrast variation in depth. In the synthetic test, we compared the method proposed in this paper with the constant, exponential and binomial functions. Besides the points with acute topography fluctuation, the results show that the inversion using parabolic function provides good and stable estimates of the interface depth of the theoretical model with RMS 0.02 km, and that the constant density function inversion gives the worst estimate of the theoretical model. According to the comparison, the choice of density parameters is a key factor in interface inversion. For the density of real situation, which varies more rapidly at smaller depth and less rapidly at larger depth, the method provides better approximations to density interface depth. In addition, we applied our method to invert the satellite regional gravity anomaly data with 10 km interval in x -axis and y -axis for the Sichuan-Yunnan region, which has complex geological structures with a large number of major faults, and negative gravity anomalies varying from -540~-90 mGal. We applied the method to invert the Moho depth below this region using reference level 35.5 km and density contrast -0.63 g·cm -3 at the earth surface with attenuation coefficient 0.0018. We compared the result with receiver function from Li et al. (2014) and other geological information. The differences between results of receiver function and our method at most seismic stations are less than 6 km. Therefore the Moho discontinuation derived from this method shows good correlation with the results of receiver function, other prior geophysical and geological information. Density contrast interface inversion is one of the primary subjects in gravity field research for understanding the earth's interior structure. This work combines the merits of the parabolic density model and the algorithm in the frequency domain, and applies the parabolic density function to the Parker-Oldenburg forward modeling and inversion algorithm. The tests on synthetic data and real data prove the new approach is rapid and effective. By inverting gravity anomaly data using this new approach, we obtained the Moho depth distribution in the Sichuan-Yunnan area, which shows that the Moho interface is shallow in the southeast but deep in the northwest. The Longmen Shan fault zone is a transition zone of the Moho depth ranging from 42 km to 58 km.

1. Institute of Geophysics, China Earthquake Administration, Beijing 100081, China; 2. Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China; 3. CNPC Xibu Drilling Engineering Sulige Gas Field Development, Ordos Uxin Qi 017300, China

密度界面反演作为了解地球内部结构的一种重要方法,长期以来都是重力学研究的主要内容.本文结合抛物线密度模型及频率域算法的优点,将抛物线密度函数应用于Parker-Oldenburg算法,经过理论推导得到了抛物线密度模型的频率域公式,从而建立了基于抛物线密度模型的三维密度界面重力异常正反演的算法和流程.理论模型数据试验表明本方法快速、有效,适用于大多数浅部比深部增加更快的实际地壳密度.研究中还利用该方法对川滇地区重力异常进行了反演,获得了该区的莫霍面深度分布,并与接收函数研究结果进行对比分析,进一步验证了本文方法的正确性和有效性.

... 此结果与前人利用重力数据资料反演得到的莫霍面深度基本一致(张恩会等, 2015 ...

2015

0.0

2008

0.0

... 从青藏高原东缘地形图(图1a)中可以发现, 龙门山断裂带整体呈NE走向, 断裂带长约500km, 宽约70km, 由数条大断裂带构成(张培震, 2008) ...

2008

0.0

2015

0.0

... GRACE(Gravity Recovery and Climate Experiment)双星系统提供的月重力场模型为连续监测全球地表物质迁移和再分布情况提供了直观的观测手段, 目前已经逐渐适用于全球区域储水量变化、冰川融化以及地震监测等研究(郑秋月等, 2015 ...

2015

0.0

2013

0.0

2013

0.0

2010

0.0

... 祝意青等, 2010) ...

2010

0.0

2017

0.0

... 朱介寿等 ...

2017

0.0

2012

0.0

... 本文采用德国地学研究中心GFZ(German Research Center for Geosciences)发布的Level-2 RL-05地球重力模型, 该数据为已处理的规格化重力场球谐系数, 最大阶数为90阶, 该模型在计算中已经扣除了大气、海洋极潮、固体潮及海潮等的影响(Christoph et al ...

2008

0.0

2005

0.0

... 由于重力异常的高频成分会导致反演迭代结果不稳定, 为保证迭代收敛, 采用改进的Parker-oldenburg反演算法, 利用公式(4)对重力异常傅里叶谱结果进行低通滤波处理, 并对边缘数据进行边界锥形余弦滤波处理(G#cod#x000f3 ...

2002

0.0

2009

0.0

2013

0.0

... 0(Laske et al ...

2006

0.0

... GRACE卫星双星系统通过测量星间距求解月重力场模型, 但由于测量方式及轨道倾角等因素, 使得月重力场模型球谐系数具有很强的相关性, 球谐误差随着阶次的增大而逐渐增加, 研究区重力时变场的解算结果会在SN向产生较为明显的SN向条带状波纹, 因此需要设计平滑核心函数对其位系数进行限制(Swenson et al ...

2010

0.0

... 在大背景下, 青藏高原的下地壳和上地幔主要呈低密度、低速度、高电导率和高泊松比异常的特征, 并存在部分熔融和流体作用(Wang et al ...

... 在青藏高原深部物质东流的动力学影响下, 龙门山断裂带深部形成的高温高压造成壳幔解耦, 壳幔岩石圈物质部分熔融(Wang et al ...

2017

0.0

... Xu等(2017)对青藏高原布格重力异常进行小波多尺度分析, 并进行了莫霍面深度反演研究 ...

... Xu et al ...

... 径向对数功率值与径向频率基本呈线性变化, 通过绘制径向功率谱和圆波数的谱图, 可计算出最低波数段内的曲线斜率并近似推断地下异常场源的深度(Xu et al ...

2016

0.0

... Xuan et al ...

2015

0.0

... 2014年GPS流动站相对于稳定欧亚大陆速度场的数据资料(Zhao et al ...

2012