4D petroelastic model calibration using time-lapse seismic signal

Abstract

In the last two decades, 4D seismic monitoring has become a widely used technique for oil and gas field production. Modeling studies are a standard for defining reservoir monitoring plans, optimizing survey design, and justifying the expense of data acquisition. Discrepancies between 4D seismic data and synthetic results can be analyzed through petroelastic modeling of reservoir simulations. However, assuming that a history match is available and that the reservoir model and fluid-flow simulation results can be trusted, characterization of pressure and fluid changes in the field remain challenging. A workflow is proposed to adjust the 4D petroelastic model (PEM) to better fit 4D seismic attributes with the dynamic behavior of the reservoir. The input data for 4D inversion consist of multiple broadband 4D-compliant processed base and monitor surveys recorded in a highly depleted clastic field offshore Africa. The broadband inversion results greatly reduce the background noise level, enhance the signal-to-noise ratio, and improve the definition of 4D signals. Due to various production effects all over the field, a new global calibration workflow to speed up the 4D petroelastic model adjustment is proposed. The combination of good 4D seismic inversions and a well-calibrated PEM is expected to have a significant impact on the reservoir monitoring. During the calibration process, reservoir model discrepancies with 4D seismic attributes can be identified, suggesting some updates of the reservoir model. In addition, when further monitors are considered, the calibrated 4D PEM provides more reliable predictability.