首页> 外文学位 >An efficient data-subspace two-dimensional magnetotelluric inversion and its application to high-resolution profile across the San Andreas Faults at Parkfield, California.
【24h】

An efficient data-subspace two-dimensional magnetotelluric inversion and its application to high-resolution profile across the San Andreas Faults at Parkfield, California.

机译:一种有效的数据子空间二维大地电磁反演及其在加利福尼亚帕克菲尔德市圣安德烈亚斯断层的高分辨率剖面中的应用。

获取原文
获取原文并翻译 | 示例

摘要

This dissertation is divided into two different aspects of the Magnetotelluric (MT) method: the development of numerical algorithms for interpreting and modeling of MT data, and the interpretation of high resolution MT data across the San Andreas Fault (SAF). In the first part, it involves two major developments: a three-dimensional MT forward modeling program on a parallel computer, and an efficient inversion scheme for two-dimensional MT data.;Modeling geologically realistic problems in 3-D requires such large grids that is not feasible on workstations. To allow for large problems, we developed an algorithm for 3-D forward modeling to run on a massively parallel distributed memory computer. The development involves a parallel implementation of existing serial algorithms, and the development of preconditioners to speed up the convergence of the iterative solver.;Generally, most reliable and efficient 2-D inversion methods require extensive computational time and memory. To overcome these difficulties, we present a new and much more efficient inversion algorithm, the reduced basis Occam's inversion (REBOCC). The algorithm is based on the fact that MT data are smooth and redundant. Therefore, the solution can be constructed from subsets of the data, without significant loss of detail, and still fit all data adequately. Numerical experiments on synthetic and real data sets show that REBOCC is stable and relatively faster than most other 2-D inversion algorithms.;Fluid which strongly influences electrical resistivities has been related to earthquake mechanisms at the SAF. Thus, MT which has an advantage over other methods for mapping the fluid distribution in the fault zone, was collected across the SAF at Parkfield, California, and were interpreted using REBOCC inversion, along with hypothesis and case studies. Three main features are required by the data: resistive Salinian granite to the west, a large volume of low resistivity to the east, and high conductivity (3 O - m) in the fault zone. These features are consistent with velocity models. Experiments with constrained inversions show that low resistivity (3 O - m, 250-m wide) can extend to 10-km depth. If interpreted to result from fluid, this would require a porosity around 8--18%, consistent with other geophysical observations.
机译:本文将大地电磁法分为两个不同的方面:开发用于解释和建模MT数据​​的数值算法,以及解释整个San Andreas断层(SAF)的高分辨率MT数据。在第一部分中,它涉及两个主要的发展:并行计算机上的三维MT正向建模程序和二维MT数据的有效反演方案;在3D中对地质现实问题进行建模需要如此大的网格,以至于在工作站上不可行。为了解决大问题,我们开发了一种用于3-D正向建模的算法,该算法可在大规模并行分布式存储计算机上运行。开发涉及并行实现现有串行算法,以及开发预处理器以加速迭代求解器的收敛。通常,最可靠,最有效的二维反演方法需要大量的计算时间和内存。为了克服这些困难,我们提出了一种新的,效率更高的反演算法,即简化的Occam反演(REBOCC)。该算法基于MT数据平滑且冗余的事实。因此,可以从数据子集构建解决方案,而不会造成细节损失,并且仍然可以完全适合所有数据。在合成和真实数据集上的数值实验表明,REBOCC是稳定的,并且相对于大多数其他二维反演算法而言相对较快。;对电阻率有很大影响的流体与SAF的地震机制有关。因此,在加利福尼亚州帕克菲尔德的SAF上收集了MT,它比其他在断层带中绘制流体分布图的方法更具优势,并使用REBOCC反演以及假设和案例研究对其进行了解释。数据需要具备三个主要特征:西部的电阻性萨利尼亚花岗岩,东部的大量低电阻率和断层带的高电导率(3 O-m)。这些特征与速度模型一致。约束反演的实验表明,低电阻率(3 O-m,250 m宽)可以延伸到10 km深度。如果将其解释为是由流体引起的,那么将需要大约8%-18%的孔隙度,这与其他地球物理观测结果是一致的。

著录项

  • 作者

    Siripunvaraporn, Weerachai.;

  • 作者单位

    Oregon State University.;

  • 授予单位 Oregon State University.;
  • 学科 Geology.;Geophysics.;Mathematics.
  • 学位 Ph.D.
  • 年度 1999
  • 页码 197 p.
  • 总页数 197
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号