Nonlinear joint inversion of Rayleigh and Love wave dispersion curves based on Pearson correlation coefficient and thickness mean sharing

Abstract

The joint inversion of Rayleigh and Love waves plays a crucial role in mitigating the non-uniqueness of surface wave inversion results and enhancing the stability of these inversions. Existing approaches to the joint inversion of Rayleigh and Love wave dispersion curves, which rely on conventional objective functions, often struggle with complex stratigraphic configurations and yield results of limited accuracy. This study introduces two novel nonlinear joint inversion techniques for Rayleigh and Love waves: Pearson correlation coefficient and thickness mean sharing. The Pearson correlation coefficient approach employs the Pearson correlation coefficient and alternating iterative objective functions to synchronize the shear wave velocity structures derived from Rayleigh and Love waves, thereby enhancing the accuracy of the joint inversion. Conversely, the thickness mean sharing method computes an average of the thickness values obtained in each iteration of the inversion, utilizing the traditional joint inversion objective function. Tests on three distinct stratigraphic structures—characterized by increasing velocity, high-speed hard interlayers and low-speed soft interlayers—as well as on measured data, demonstrate that the proposed methods significantly improve the stability and accuracy of nonlinear joint inversion.