Vector Wavenumber Transformation Method To Extract Multimode Dispersion Curves From Ambient Seismic Noise

It's a hot topic for seismic study to obtain S-wave velocity structure through mi-crotremor analysis.Ambient seismic noise,which is also called microtremor in the field of geotechnique engineering,is a stochastic wave field generated by various pas-sive sources.There are two types of vibrations.One is low-frequency(<1Hz)ambient seismic noise generated by natural phenomena,such as tidal,water,rain,volcanic activ-ities.The other is high-frequency(>1Hz)ambient siesmic noise generated by human activities,such as traffic,mechanical vibrations,industrial activities.Ambient seismic noise survey method has become an important technology for both engineering geophysical exploration and ambient noise tomography because it is non-destructive,easy to operate,low cost and highly accurate.The new method is widely applied in areas where the geological conditions are complicated and it's diffi-cult to achieve by conventional geophysical methods,such as urban or areas unwanted to be damaged.Although the energy of ambient seismic noise is dominated by the fundamental mode surface wave,it also contains the energy of body waves,scattered waves,higher modes surface waves.These higher mode surface waves contain a lot of structural information.There are great uncertainty in the inversion only using the fundamental mode dispersion curve of surface wave.If the higher modes dispersion curves can be extracted,it will provide more constraint to reduce the non-uniqueness.Extracting dispersion curves from ambient noise is a key step in microtremor sur-vey and ambient seismic noise tomography.Several methods have been developed,namely,spatial autocorrelation(SPAC)method introduced by Aki(1957),frequency wavenumber(FK)method advanced by Capon(1969);Lacoss et al.(1969),phase velocity image-analysis proposed by Yao et al.(2006),multichannel analysis of sur-face wave(MASW)method proposed by Park et al(1998),and high-resolution linear Radon transform(LRT)method developed by Luo et al..However,the abovementioned methods cannot extract high-quality dispersion curves of surface waves,especially of the higher modes,from the ambient seismic noise data.Therefore,they do not meet the requirement for reducing non-uniqueness and increasing the accuracy of ambient seismic noise tomography.In this paper,We propose a new method called vector wavenumber transforma-tion method(VWTM)for extracting the dispersion curves of higher modes from ambi-ent seismic noise data,study the forwarding computations of dispersion curves of the Rayleigh wave,and research the signal analysis method and the observation array of ambient seismic noise.In the second chapter,we studied the computation of dispersion curves of the Rayleigh wave.In the third chapter,we introduced the computational procedure of the discrete wavenumber method.In the fourth chapter,the theory of the VWTM was given.To assess the accuracy,accuracy,and applicability of the VWTM,several numerical simulation studies were conducted in the fifth chapter.The param-eters for the VWTM and the observation system were also discussed in this chapter.In the sixth chapter,both the short-term and long-term ambient noise signals for real-world examples were given to evaluate the validation of the VWTM.The corresponding inversion problem of multi-models dispersion curves are presented in the last chapter.The new ambient seismic noise analysis method called the vector wavenumber transformation method(VWTM)can clearly extract the dispersion curves of multi-modes from ambient seismic noise data recorded by an observation array.It meets the requirement for reducing non-uniqueness and increasing the accuracy of ambient seismic noise tomography.