基于背景噪声经验格林函数的地震准确定位精度分析——以2008年甘肃武都地震为例  

Accuracy analysis of locating earthquakes with ambient noise Empirical Green′s Functions—A case study of the 2008 Wudu earthquake(Gansu,China)

作　　者：王烁帆 倪四道 王伟涛[3] 曾祥方 谢军 韩宏博 WANG ShuoFan;NI SiDao;WANG WeiTao;ZENG XiangFang;XIE Jun;HAN HongBo(State Key Laboratory of Geodesy and Earth′s Dynamics,Innovation Academy for Precision Measurement Science and Technology,Chinese Academy of Sciences,Wuhan 430077,China;College of Earth and Planetary Sciences,University of Chinese Academy of Sciences,Beijing 100049,China;Institute of Geophysics,China Earthquake Administration,Beijing 100081,China;School of Earth and Space Sciences,University of Science and Technology of China,Hefei 230026,China)

机构地区：[1]中国科学院精密测量科学与技术创新研究院,大地测量与地球动力学国家重点实验室,武汉430077 [2]中国科学院大学地球与行星科学学院,北京100049 [3]中国地震局地球物理研究所,北京100081 [4]中国科学技术大学地球与空间科学学院,合肥230026

出　　处：《地球物理学报》2022年第8期2904-2916,共13页Chinese Journal of Geophysics

基　　金：国家重点研发计划课题(2018YFC1504202);国家自然科学基金(41874068,42030311)资助.

摘　　要：准确的震源位置能够为抗震减灾工作和地球内部结构研究提供关键信息.在震中附近观测台网密集且方位角覆盖良好的情况下,通过拟合多个台站P波、S波的观测到时,能够得到准确的震源位置.而在台网稀疏的地区,由于缺少可靠的三维速度结构模型,往往造成较大的地震定位误差.近年来发展了基于背景噪声经验格林函数(EGFs)对地震波形进行校正的重定位方法,能够有效地压制路径上复杂速度结构体的影响,提高了地震定位精度,为稀疏台网情形下地震准确定位研究提供了一个新思路.本文选取由InSAR观测到准确位置的2008年甘肃武都M_(S)5.5地震作为测试案例,对稀疏台网下基于背景噪声格林函数地震准确震中测定方法进行了定量评估.利用震中附近多个流动台分别作为参考台,提取其与固定台站之间的背景噪声格林函数(EGFs),然后使用噪声Rayleigh面波格林函数对地震波形进行校正,重定位武都地震震中,并与真实震中位置进行对比.结果表明:使用距离震中30 km以内的参考台,利用10~30 s频段的面波走时信息,噪声叠加时长为一个月,定位精度在5 km以内;当噪声叠加时间一年以上,重定位精度优于1 km.本文进一步针对缺少近台以及台网更加稀疏的情况进行了测试,发现使用10个左右固定台进行重定位,基于较高质量的噪声EGFs频散数据,定位结果精度可达3 km,从而给出了该方法的高精度定位所需的固定台网及参考台站的观测指标体系.Accurate earthquake locations play important roles in seismic hazard mitigation and studies of Earth′s structure.In regions with dense seismic network,earthquake location can be accurately obtained from the observations of seismic waves(P,S)travel-time.With sparse network,the lack of a reliable velocity structure model usually results in the location of poor precision.In recent years,an earthquake location method based on ambient noise Empirical Green′s Functions(EGFs)has been developed,providing a new approach for the accurate location of earthquakes in the case of a sparse seismic network.This method uses the noise EGFs surface wave group travel-time to calibrate the travel-time of seismic phases computed with 1D velocity models.It can effectively suppress the location error due to anomalous velocity structure along the propagation path,thereby improving the earthquake location accuracy.This study selects the 2008 Wudu M_(S)5.5 earthquake(Gansu,China)with the real location observed by InSAR as the standard,to evaluate the accurate epicenter determination method based on the ambient noise empirical Green′s function under the sparse network quantitatively.We use several portable stations near the epicenter as reference stations to extract the ambient noise empirical Green′s function.The seismic waveforms are then corrected using the ambient noise surface wave empirical Green′s function to relocate the Wudu earthquake and compared with the real epicenter.The results show that using a reference station within 30 km from the epicenter and the 10~30 s band Rayleigh wave,and the noise stacking is one month,the positioning accuracy is within 5 km.When the noise cross-correlation stacking time is more than one year,the relocation accuracy is better than 1 km.We further conduct tests for a sparser seismic network.It shows that using about 10 remote stations with reliable noise EGFs dispersion for relocation,the relocation results with deviations within 3 km can also be obtained,giving the observation system of

关 键 词：背景噪声格林函数 地震定位 稀疏台网 2008年甘肃武都地震 

分 类 号：P315[天文地球—地震学]