Seismic nonlinear frequency-dependent inversion for viscoelastic media based on a modified scattering coefficient approximation.

Pre-stack seismic amplitude versus offset (AVO) inversion estimates reservoir properties by relying on the variation of seismic data amplitude with offset, usually ignoring the role of frequency information in seismic data, which leads to its limited and incomplete use of seismic data. Because the reservoir medium has certain viscoelastic properties, its seismic response changes with the offset, and also has a frequency variation response. The traditional linearized PP wave reflection coefficient of viscoelastic medium is only suitable for small angle incidence, and the inversion based on this equation is difficult to meet the requirements of high-precision parameter estimation for complex media. In this paper, starting from the scattering matrix of a viscoelastic medium, a modified second-order viscoelastic scattering coefficient equation based on perturbation theory is derived, and a nonlinear inversion strategy based on the combination of amplitude and frequency is proposed. The new equation has a wider application range because it considers the viscoelasticity and nonlinear approximation of the medium at the same time, and includes five model parameters to be inverted, namely the P-wave velocity, S-wave velocity, density, quality factor (Q-factor) of P-wave, and Q-factor of S-wave. Accuracy analysis shows that the newly proposed approximation equation is more accurate than the traditional linear approximation equation. In addition, a pre-stack seismic nonlinear frequency-dependent inversion method for viscoelastic media is proposed, which realizes the simultaneous estimation of five model parameters. Synthetic example and field seismic data set are also used to demonstrate the stability and accuracy of the proposed method. • A second-order viscoelastic scattering coefficient equation of P -wave reflectivity is derived based on perturbation theory. • A high-order pre-stack seismic nonlinear frequency-dependent inversion method (FANNI) is proposed. • Through the proposed FANNI inversion method, the high-precision direct extraction of quality factors can be realized. [ABSTRACT FROM AUTHOR]