王鹏^1,2,,
王宝善^1,3
1. 中国地震局地球物理研究所, 北京 100081
2. 山东省地震局, 济南 250102
3. 中国科学技术大学地球和空间科学学院, 合肥 230026

基金项目: 国家重点研发计划（2016YFE0109300）、中国地震局地震科技星火计划项目（XH18026Y）、山东省自然科学基金（ZR2017QD014）、山东省重点研发计划（2016GSF120011）和山东省地震局科研基金项目（JJ1704Y）联合资助


详细信息

作者简介: 王鹏, 男, 1983年生, 副研, 中国地震局地球物理研究所博士研究生, 主要从事微震识别和震源参数反演方面的研究.E-mail:wangpengeq@163.com

中图分类号: P315

收稿日期:2019-04-19
修回日期:2019-07-30
上线日期:2020-05-05

Stress drop estimation of Chang Island earthquake swarm by generalized superposition inversion method

WANG Peng^1,2,,
WANG BaoShan^1,3
1. Institute of Geophysics, China Earthquake Administration, Beijing 100081, China
2. Shandong Earthquake Agency, Jinan 250102, China
3. School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China


MSC: P315

--> Received Date: 19 April 2019
Revised Date: 30 July 2019
Available Online: 05 May 2020


摘要
摘要:地震应力降是表征震源特性的一个重要参数，通常由震源谱计算得到.但如何从大量地震的观测谱中校正路径和台站效应，准确地分离出震源谱并计算应力降，仍是一个难题.本文采用广义叠加反演方法通过分阶段叠加从观测谱中迭代分离出震源项、台站项和路径项，并利用分震级的叠加震源项与理论谱的拟合得到经验格林函数，再利用由经验格林函数校正后的真实震源谱估算出应力降.该方法不需要知道仪器响应、场地响应和传播路径上的衰减系数，并且不用事先假设震源具有自相似的恒应力降特征，计算效率很高，适用于反演大量地震的震源参数.我们将广义叠加反演方法应用于长岛震群，利用S波谱估算了1431个M_L≥1.0地震的应力降.结果表明，广义叠加反演方法获得的震源参数和用传统方法得到的结果一致；长岛震群地震的应力降较小（中值是0.3 MPa），变化范围很大（0.009~3.04 MPa之间），属于低应力降事件集；同时应力降中值随地震矩增大而增大，偏离了自相似理论.应力降空间变化不均匀，高应力降事件主要集中在震群的端部（NW端），到最近断层距离的依赖性不强，在9、10 km处的应力降中值略高于其他深度.根据应力降的变化特征和区域构造条件，我们推测地下流体在长岛震群的活动中起到一定的作用.
关键词: 广义叠加反演/
震源谱分离/
长岛震群/
应力降/
震源参数/
拐角频率

Abstract:Seismic stress drop is an important parameter to characterize the seismic source, which is usually calculated from the source spectrum. However, it is still challenging to calibrate the path and station effects from the observed spectra of a large number of earthquakes, then accurately separate the source spectrum and calculate the stress drop. In this paper the generalized superposition inversion method is used to iteratively isolate the source term, station term and path term from the observed spectrum by progressive superposition. Then the empirical Green's function is obtained by fitting the theoretical spectrum to the average source terms. The stress drops of each event are then estimated from individual source spectrum corrected by the empirical Green's function. This method does not depend on the knowledge of the attenuation coefficients of instrument response, site response, or propagation path, and is not banded to any preferred stress drop model. It has high computational efficiency and is suitable for inversion of source parameters of a large number of earthquakes. We apply the generalized superposition inversion method to the Chang Island swarm and estimate the stress drop of 1431 earthquakes with M_L≥1.0. The results show that the source parameters obtained by the generalized superposition inversion method are consistent with those obtained by the traditional method; the stress drop of the Chang Island earthquake swarm is small (the median is 0.3 MPa), and the variation range is large (between 0.009 and 3.04 MPa), which belongs to the set of low stress drop events; at the same time, the median stress drop increases with seismic moment and deviates from the self-similarity theory. Spatial variation of stress drop is inhomogeneous. High stress drop events are mainly concentrated at the Northwest of the swarm (NW), and the dependence on the distance from the nearest fault is not clear. The median stress drop at 9 km and 10 km is slightly higher than that at other depths. According to the variation characteristics of stress drop and regional tectonic conditions, we infer that the underground fluid plays a role in the activity of Chang Island earthquake swarm.
Key words:Generalized superposition inversion/
Source spectrum separation/
Chang Island earthquake swarm/
Stress drop/
Source parameters/
Corner frequency



PDF全文下载地址:

http://www.geophy.cn/data/article/export-pdf?id=dqwlxb_15459