利用EEMD及超导重力数据探测超低频地球自由振荡模态分裂信号:探测实例

Observation of Ultralow-Frequency Earth Free Oscillation Mode Multi-Splitting Based on EEMD and Global Superconducting Gravimeter Data:A Case Study

利用2004年苏门答腊9.0级地震后GGP台网中15台超导重力仪的18组观测数据,采用EEMD分解技术及叠积技术,检测了频率小于1mHz的低频地球自由振荡的简正模(3S1、0S4、0S5)及其频谱分裂现象。通过对原始分钟间隔的超导重力数据进行预处理得到重力残差,然后运用EEMD方法将其分解到处于不同频段上的IMF分量,提高了探测低频自由振荡信号的信噪比,使一些低频自由振荡信号的谱线分裂更清晰地显现出来。结果表明,经过EEMD分解过的重力残差能更有效地检测低频自由振荡信号,得到信噪比更高的低阶球型振荡分裂谱线。该研究进一步证实了EEMD数据处理方法的有效性以及超导重力仪在低阶地球自由振荡检测中的优越性。

In this study,we select 15 sets of 18 of superconducting gravimeter（SG）records from GGP（global geodynamics project）stations after the 2004 Sumatra Mw 9.0earthquake and use EEMD（ensemble empiricalmode decomposition）and singlet stacking technology to detect low-frequency earth＇s free oscillations mode（3S1,0S4,0S5）and their spectral splitting with frequencies less than 1mHz.After removing the tidal and local atmospheric pressure effects from the original minute-interval SG records,we obtain a residual gravity data set.Then,EEMD is applied to this SG time-series to obtain different IMF（intrinsic mode function）on different frequencies.This will significantly reduce the possibility of mode mixing and end effect,and it could improve some low-frequency seismic signals＇ SNR（signal-to-noise ratio）.Therefore,EEMD could enable some splitting spectral of low-frequency free oscillation signals to be observed more clearly.Through comparisons of the normalized amplitude of the results obtained without using EEMD and after using EEMD of the residual gravity records,the experimental results show that when EEMD is applied to residual gravity records,it can be more effective for observation of the Earth＇s low-frequency signals and will obtain higher resolution of low-order spherical oscillations＇ singlets.This study demonstrates that EEMD is effective in data-processing and that the superconducting gravimeter is superior in detecting the low-order earth＇s free oscillations.