High-resolution impedance estimation using multiparameter viscoelastic FWI in a Gulf of Mexico setting

Abstract Traditional seismic imaging methods rely on assumptions (e.g., single scattering), which become less reliable in high-impedance-contrast scenarios. Such scenarios are more prevalent as oil production moves into increasingly complex geologic settings. Full-waveform inversion (FWI) offers an alternative by directly inverting for earth properties while simulating complex wave phenomena and handling uneven illumination through iterative processes. Recent advancements in computing power and data acquisition have enabled high-frequency FWI, including elastic FWI, which can produce high-resolution velocity models for structural interpretation. However, full quantitative interpretation requires multiparameter FWI, which remains challenging due to the crosstalk between different parameters. We present a multiparameter viscoelastic FWI workflow that attempts to mitigate crosstalk by separating the problem into smaller pieces, focusing on different properties with specific portions of the data as well as better-suited cost functions. The workflow first improves the P-wave background velocity and attenuation models, then it updates acoustic and shear impedance using reflection data amplitude. Application to a field data set from the Gulf of Mexico, characterized by complex salt geometry with variable thickness, shows enhanced subsalt structural imaging and sensible quantitative information, with a reasonable match to well logs and consistent amplitude variation with angle behavior compared to least-squares reverse time migration.