Spectral decomposition applied to time-lapse multicomponent seismic interpretation at Postle Field, Texas County, Oklahoma

Abstract

The Reservoir Characterization Project (RCP) at the Colorado School of Mines acquired three 9-C, 3-D time-lapse surveys over an area of 6.25 square miles in the Hovey Morrow Unit of Postle Field, Texas County, Oklahoma from March 2008 to February 2010. The investigation focuses on the use of multicomponent seismic data to characterize the Morrow A sandstone in the Postle Field, Oklahoma. In the Hovey Morrow Unit (HMU) of the field, the Morrow A sandstone varies from 0 ft to 70 ft thickness with a mean of 35 ft, and is located approximately 6000 ft deep. The 3-D multicomponent seismic data available in the area include P-, S11- and S22-waves from three different acquisition campaigns (baseline, monitor 1 and monitor 2). The thin reservoir and lack of acoustic impedance between reservoir and encasing shale make the Morrow A sandstone very difficult to detect. Spectral decomposition improves the detection of the Morrow A sandstone enabling the monitoring production and injection changes in the reservoir. The exponential matching-pursuit decomposition method was used to obtain the frequency spectrum of the seismic data. The peak amplitude and peak frequency are the spectral decomposition attributes used to conduct the time-lapse seismic interpretation. The time-lapse study shows that both P-wave and S11-wave time-lapse peak amplitude changes have a significant correlation with the pressure change map from reservoir simulation, with the S11-wave data having the best correlation. P-wave is sensitive to both pressure change and saturation change while S-wave is only sensitive to pressure change. Combining the P-wave and S-wave data together we may be able to separate the saturation change from pressure change.