摘要
本文利用基于图像处理器加速的模板匹配定位法(Graphics Processing Unit-based Match&Locate, GPU-M&L)和双差定位法(HypoDD),对上海及邻区13个台站记录的2011年至2020年共10年的连续地震数据资料进行分析.首先从中国地震台网中心提供的146个地震事件目录中挑选了136个地震事件作为模板事件,使用模板匹配定位技术对上海及邻区10年的连续资料进行遗漏地震事件的扫描和检测,共识别出824个地震事件,约为台网中心提供地震目录事件数量的5.5倍.然后对识别出的地震事件通过深度去噪方法(DeepDenoiser)将信号与噪声分离,并对去噪后地震波形的频率和振幅特性分析来进一步确认识别出的地震事件.同时利用基于机器学习的震相拾取技术(PhaseNet),对去噪后的333个地震事件进行了震相拾取.检测后的地震目录完备震级由台网目录的Mc1.0降为Mc0.8.最后利用双差定位法对479个地震事件进行精定位,精定位的结果显示,上海地区整体地震活动性较弱,地震的空间分布相对较为分散,定位后的地震事件部分集中于安角断凹,部分事件沿北东向的枫泾—川沙隐伏断裂带和沿北西向的南通—上海断裂分布.我们的结果,为研究上海及邻区地震活动性、地震发生灾害程度和风险性评价等,奠定了重要的数据基础.
In this paper, we utilize the graphics processing unit-based match & locate(GPU-M&L) and double-difference location algorithm(HypoDD) methods to study the seismic activity in Shanghai and its adjacent areas based on continuous seismic data between 2011 and 2020 recorded by 13 seismic stations. Firstly, 136 events are selected as templates from 146 seismic events listed in the routine catalog provided by China Earthquake Network Center. By scanning and detecting missing seismic events through the continuous data with the GPU-M&L technique, 824 events are identified which are ~5.5 times more events than those recorded in the routine catalog. Secondly, seismic signal denoising and decomposition using deep neural network(DeepDenoiser) is applied to separate signal and noise for the newly detected seismic events. After DeepDenoiser, by checking the frequency and amplitude characteristics of seismic waveform, we further check the newly detected seismic events. The phase-picking technology based on machine learning(PhaseNet) is used to pick up the seismic phases of 333 events after DeepDenoiser. We reduce the magnitude of completeness Mcfrom 1.0 of the routine catalog to 0.8. Finally, the HypoDD is applied to relocate the 479 detected events. The results show that the seismicity of Shanghai area is weak, and the spatial distribution of earthquakes in Shanghai area is generally sporadic. Some events are distributed around the Anjiao fault depression, and some events are distributed along the NE-striking Fengjing-Chuansha concealed fault and the NW-striking Nantong-Shanghai fault. Our results provide an important data base for a further study of seismic activity, seismic hazard and risk assessment in Shanghai and its adjacent areas.
作者
张雅楠
李红谊
张盛中
李炎臻
黄雅芬
钟卫星
ZHANG YaNan;LI HongYi;ZHANG ShengZhong;LI YanZhen;HUANG YaFen;ZHONG WeiXing(School of Geophysics and Information Technology,China University of Geosciences(Beijing),Beijing 100083,China;Shanghai Sheshan National Geophysical Field Observation and Research Station,Shanghai 201602,China;Information Network Center,China University of Geosciences(Beijing),Beijing 100083,China;Sheshan Earthquake Monitoring Center,Shanghai Earthquake Agency,Shanghai 201602,China)
出处
《地球物理学报》
SCIE
EI
CAS
CSCD
北大核心
2023年第3期1113-1124,共12页
Chinese Journal of Geophysics
基金
国家自然科学基金项目(U1939203)
上海佘山地球物理国家野外科学观测研究站开放基金(2020K02)
北京市自然科学基金项目(8212041)共同资助。
关键词
上海地区
微震检测
模板匹配定位技术
深度去噪
双差定位
Shanghai and its adjacent areas
Microseismic detection
Template match and locate
DeepDenoiser
Double-difference location