首页 | 本学科首页   官方微博 | 高级检索  
     检索      

球坐标系下多震相走时三参数同时反演成像
引用本文:黄国娇,白超英,钱卫.球坐标系下多震相走时三参数同时反演成像[J].地球物理学报,2015,58(10):3627-3638.
作者姓名:黄国娇  白超英  钱卫
作者单位:1. 河海大学地球科学与工程学院地质科学与工程系, 南京 211100; 2. 长安大学地质工程与测绘学院地球物理系, 西安 710054
基金项目:江苏省自然科学基金项目(BK20150799)、国家自然科学基金项目(41504038)、河海大学中央高校基本科研业务费项目(2014B13814)、教育部博士学科专项基金项目(20110205110010)资助.
摘    要:球坐标系下多震相走时三参数(速度、震源位置和反射界面)同时反演需要解决两个关键问题:(1)球坐标系下3D速度模型中多次透射、反射(折射)及转换波精确、快速的射线追踪;(2)同时反演时三种不同参数间的强耦合问题.为此,我们将直角坐标系下分区多步不规则最短路径算法推广至球坐标系中,进行区域或者全球尺度的多震相射线追踪.然后将其与适合多参数同时反演的子空间算法相结合,形成一种球坐标系下联合多震相走时三参数同时反演的方法技术.与双参数(速度和反射界面或速度和震源位置)同时反演的数值模拟对比分析显示:三参数与双参数的同时反演结果大体接近,并且它们对到时数据中可容许的随机噪声不太敏感.结果说明本文中的同时反演成像为一种提高成像分辨率,同时反演速度、震源位置和反射界面的有效方法.

关 键 词:球坐标系  分区多步不规则最短路径算法  多震相走时  三参数同时反演  子空间法  
收稿时间:2014-12-22

Simultaneous inversion of three model parameters with multiple phases of arrival times in spherical coordinates
HUANG Guo-Jiao,BAI Chao-Ying,QIAN Wei.Simultaneous inversion of three model parameters with multiple phases of arrival times in spherical coordinates[J].Chinese Journal of Geophysics,2015,58(10):3627-3638.
Authors:HUANG Guo-Jiao  BAI Chao-Ying  QIAN Wei
Institution:1. Department of Geology Science & Engineering, School of Earth Sciences and Engineering, Hohai University, Nanjing 211100, China; 2. Department of Geophysics, College of Geology Engineering and Geomatics, Chang'an University, Xi'an 710054, China
Abstract:To guarantee computational precision and improve the resolution of seismic traveltime tomography at a regional or global scale, we conduct a 3D simultaneous inversion of three different parameters (velocity, hypocenter location and reflector geometry) in spherical coordinates with multi-phase travel times. For this task, there are two key issues: (1) It needs an efficient and accurate arrival tracking algorithm for multiply transmitted, reflected (or refracted) and converted waves in a 3D variable velocity model with embedded velocity discontinuities (or subsurface interfaces), and (2) A subdimensional inversion solver is required which can easily search for different types of model parameters to balance the tradeoff between the different types of model parameter updated in the simultaneous inversion process. For these purposes, we first extend a popular grid/cell-based wavefront expanding ray tracing algorithm (the multistage irregular shortest-path ray tracing method, shorted as ISPM), which previously worked only in Cartesian coordinates at a local scale, to spherical coordinates appropriate to a regional or global scale. In 3D spherical coordinates a trapezoidal prism cell is used to divide the spherical earth model, except for the global and other irregular subsurface interfaces, where a trapezoidal cone is used. And in one previous paper we have discussed the computational accuracy of the multistage ISPM algorithm against the analytic solution of AK135 Travel Time Table for 49 kinds of global phases and concluded that the computational accuracy can be tuned to within the 0.1 s absolute time error when it works in the spherical coordinates. We then incorporate the subspace method to formulate a simultaneous inversion algorithm, in which the multiple classes of arrivals (including direct and reflected arrivals from different velocity discontinuities) can be used to simultaneously update the velocity fields, hypocenter location and reflector geometries. In order to illustrate the performance of inversion for three model class parameters, we have selected a regional model at a scale of 20°×20°×700 km. In the numerical experiment, different phases are used. For comparison, we design three different schemes. The first one is to simultaneously update both the velocity field and the source locations (referred as V+S inversion), the second is to simultaneously invert both the velocity field and the reflector positions (referred as V+R inversion), while the third scheme updates all three model parameter classes simultaneously (referred as V+R+S inversion). To account for picking uncertainty, random noise was added to the multiple sets of arrivals according to the expected picking errors. The inversion results show that the reconstructed velocity fields exhibit similar anomalous structures to the true field, but only partially being recovered. Interestingly, the recovered velocity fields are nearly the same regardless whether the two parameter class inversion or three parameter class inversion is attempted. The updated two subsurface interfaces are almost identical, no matter whether the V+R or the V+R+S simultaneous inversion algorithm is used. For the hypocenter location, the V+S inversion has a slightly better convergence to the real source locations than the V+R+S inversion, due to only two classes of the model parameters being updated in the simultaneous inversion process, but the difference is not so significant. The above results verify that the V+R+S simultaneous inversion scheme is feasible and applicable due to the almost equal capability with the constrained two model parameter class inversions (V+S or V+R). All three scheme algorithms are quite tolerant to modest errors in the travel time picks, which makes them robust enough for practical applications. The results show that the V+R+S simultaneous inversion method is an efficient and feasible scheme for updating the velocity field, reflector geometry and hypocenter location.
Keywords:Spherical coordinates  Multistage irregular shortest-path method  Multi-phase travel times  Simultaneous inversion of three model parameters  Subspace method
本文献已被 CNKI 等数据库收录!
点击此处可从《地球物理学报》浏览原始摘要信息
点击此处可从《地球物理学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号