Multichannel Analysis of Ambient Noise Surface Waves Based on Semblance Phase-Shift Method

Ambient noise surface wave exploration is one of the fields of interest in geophysical research. Extracting dispersion curves and inverting the S-wave velocity structure from the dispersion characteristics is also of primary importance. The accuracy of dispersion curves has great significance for the subsequent inversion result and its interpretation. The phase-shift method is widely used in dispersion imaging of surface waves. This method possesses advantages on stability but also suffers a lot from low resolution and low noise resistance. Therefore, we propose an improved phase-shift method based on semblance coefficients. This method replaces linear stacking in the traditional phase-shift method by calculating semblance coefficients and, therefore, can effectively improve the resolution and noise resistance of surface wave dispersion spectrum imaging. Tests are implemented on both synthetic ambient noise data and field data recorded by a short-period dense seismic array located in the ChangbaiShan region to evaluate the proposed method. The dispersion spectrum imaging results of the model and field data show that the semblance phase-shift (SPS) method has better noise resistance and computational accuracy than the traditional phase-shift method. The inversion results indicate that it is possible to obtain a reasonable S-wave velocity structure by inverting the dispersion curves resulting from the semblance phase-shift method. By constructing a 3 km deep and 4.8 km long S-wave velocity image, the velocity structure and abnormal conditions beneath the array in the ChangbaiShan region are presented. The results indicate a significant low-velocity anomaly at a depth of 1 km. It is inferred that it may be a fluid-rich structure.