Time-lapse AVO inversion: Application to synthetic data

Abdul-Nassir Saeed, Laurence R. Lines, Gary F. Margrave

A reservoir characterization workflow for time-lapse studies requires integrating seismic data of different vintages and well logs information into a single consistent model to delineate changes of reservoir parameters.

In this study, we implemented three different time-lapse AVO inversion algorithms (total inversion of the differences, inversion of seismic difference only and sequential reflectivity-constrained inversion) using synthetic data that simulate a time-lapse model of a heavy oil reservoir. Elastic physical parameters of the time-lapse model were chosen to represent reservoir conditions at pre-production and post-production periods after reservoir depletion.

The time-lapse AVO inversion schemes simultaneously invert baseline and monitor seismic data to estimate the change of model parameters. The proposed algorithms have proved their robustness in terms of computation time as well as stability in presence of noise to ensure smooth changes in estimating reservoir attributes from time-lapse inversion.