Facies‐based full‐waveform inversion for anisotropic media: a North Sea case study

Abstract

Elastic full-waveform inversion can increase the resolution of reservoir characterization using seismic data. However, full-waveform inversion for realistic anisotropic media suffers from the trade-offs between the medium parameters and strongly relies on the accuracy of the initial model. Here, we employ a regularization methodology that utilizes geologically consistent information to reduce the inversion non-linearity and crosstalk between the parameters. The geologic constraints are obtained from well logs and interpolated along major horizons in the migrated image. The algorithm, designed for transversely isotropic media with a tilted symmetry axis, is applied to ocean-bottom data acquired at Volve field in the North Sea. The facies-based constraints help build high-resolution velocity fields and accurately image the reservoir region. In particular, the developed algorithm increases the resolution of the Pand S-wave symmetry-direction velocities and other parameters at the reservoir level. The facies-based inversion also provides robust estimates of density, which is inverted simultaneously with the velocity fields. Overall, even relatively sparse prior information proves to be sufficient for the proposed methodology to achieve a much higher spatial resolution than the unconstrained full-waveform inversion.

Two‐dimensional SH‐wave and acoustic P‐wave full waveform inversion: A Midland Basin case study

“…FWI aims at obtaining a model by iteratively minimizing the misfit between the observed and predicted waveform data generated with forward modelling and has great potential to estimate the seismic velocities by utilizing full waveform information contained in seismic data (Guo & Alkhalifah, 2017; Pratt et al., 1996; Romdhane & Querendez, 2014; Tarantola, 1986; Virieux & Operto, 2009). It can improve the resolution of background velocity model derived from tomography or from migration velocity analysis (Romdhane et al., 2014; Singh et al., 2021). An extensive review of FWI can be found in Virieux and Operto (2009).…”