完全剥离一阶模态nS1的三重分裂信号将有助于识别出其全部分裂谱线,进而更好地约束地球内部结构.理想情况下,球谐叠加法(SHS)与多台站实验技术(MSE)均可剥离一阶模态nS1的三重分裂信号,但部分学者持不同观点.本文基于对二者的理论分析进一步确认,在自耦合前提下,MSE方法可成功剥离nS1的三重分裂信号;而SHS方法在实际应用中无法成功剥离nS1的三重分裂信号,但可相对增强目标信号的振幅强度.此外,本文解释了MSE可剥离单线态信号的真实原因,并指出由于MSE未考虑全频段耦合影响,故仅适用于1mHz以下的低频模态.鉴于超导重力(SG)数据在1mHz以下比宽频地震数据具有更高的信噪比,因此,为验证本文结论并弄清MSE和SHS的实质,基于模拟数据及SG台站实测数据,本文利用MSE和SHS分别对一阶模态的合成信号及模态nS1进行了探测分析.实验结果表明,MSE可成功剥离一阶模态三重分裂信号,SHS则无法完全剥离,验证了我们的理论结论,表明前人部分结果需重新审视与评价.此外,本文基于三种不同方法并利用13个SG台站数据给出的nS1的分裂宽度比分别为1.008,1.000和1.001,远小于异常分裂判别临界值1.5,因此,3S1应是正常分裂模态.
Complete isolation of the triplet of the seismic normal mode nS1 may better help to tell the complete splitting lines and consequently better constrain the inner structure of the Earth. In theory, the Multi-Station Experiment (MSE) technique and the Spherical Harmonic Stacking (SHS) method can successfully strip the triplet of a mutiplet mode with degree one, but someprevious studies considered not. In this paper, based on carefully investigating these two methods, it is found that MSE can successfully strip the triplet of a degree one mode if only the effects of the self-coupling are considered~ while SHS cannot strip the triplet of a degree one mode in practical cases, but it can significantly enhance the relative amplitude strength of a target singlet mode. In addition, we explain why MSE can strip the triplet of a degree one mode for the first time, and we note that since MSE does not take into account the full coupling effects, it may only be suitable for the modes below 1 mHz. Taking into account the fact that the superconducting gravimeters (SGs) deployed in the frame of the Global Geodynamics Project (GGP) can outmatch the best seismometers in the frequency band below 1 mHz, in order to verify the conclusions obtained in this article, SHS and MSE are both used to detect a synthetic mutiplet signal and 3S1 mode, based on both the simulation datasets and practical SG observations. The results clearly verify our conclusions, and suggest that some previous conclusions should be reevaluated. Moreover, from the result of using three different methods and 13 SG records we obtain a clear spectral peak corresponding to m= 0 of 3S1 and the splitting width ratios of 1. 008, 1. 000 and 1. 001 respectively, which are much less than 1.5, hence we may conclude that 3S1 is normal split.