Seismic Rayleigh Wave Velocity Inversion In The P-sv Wave Static Correction Applied Research

In multicomponent seismic exploration, P-SV wave static correction is one of themost challenging processing problems, and seriously affects the efficacy ofmulticomponent seismic exploration. It is theoretically and practically significant toresolve the problem of P-SV wave static correction.The crucial and difficult problem of the P-SV wave static correction is how toresolve its S-statics. At the same location, the magnitude of S-wave statics iscommonly two to ten times greater than P-wave statics. Generally, S-wave statics aresensitively affected by topographic relief, low velocity zone and fracture. The nearsurface thicknesses, as seen by a P-wave and an S-wave, can be different. The watertable serves as a boundary for the P-wave near surface layer, but not necessarily so forS-wave which is unaffected by the pore fluid composition. Hence, the staticsalgorithms for P-wave data can not be applied to P-SV wave data directly.This dissertation proposes to use seismic Rayleigh wave velocity inversion forthe near surface S-wave velocity structure and the P-SV wave S-statics.Firstly, the principles of Rayleigh wave, which are the foundations of Rayleighwave velocity inversion, are introduced. Experiments on the models show thatRayleigh wave in multi-layered media occurs dispersive, and the dispersion curvescan be theoretically computed by Knopoff method quickly and efficiently. S-wavevelocity is the main factor of Rayleigh dispersion curves.Secondly, the damping least square method and the genetic algorithm for inversecomputations of Rayleigh dispersion curves in multi-layered media are studied. Testson the theory Rayleigh dispersion curves show that the two methods can efficientlyinverse shallow S-wave velocity structures. The damping least square method is apartial linear algorithm with fast inversion. The genetic algorithm is a globalnon-linear algorithm with huge computation and slow inversion.Thirdly, in order to get regular shallow S-wave velocity profile subsequently, thegridding methods of linear interpolation, multiquadric, ordinary Kriging and inverseinterpolation for geophysical data are studied. Tests both on a synthetic data and on araw data show that inverse interpolation is a computationally efficient and robust 2-Dmethod with high accuracy, and its gridding results of different gridding orientationsare basically identical.Finally, seismic Rayleigh wave velocity inversion for P-SV wave S-statics isstudied, which is mainly including the flowing contents: the separation of Rayleighwave signals from seismic data, the pickup of Rayleigh dispersion curves, seismicRayleigh wave velocity inversion, and the principle and the processing flow ofresolving P-SV wave S-statics. The performance on the raw two-dimensional P-SVwave data shows that the method can predict the long wavelength trend in P-SV waveS-statics.