Time-lapse seismic monitoring for enhanced oil recovery and carbon capture and storage field site at Cranfield field, Mississippi

Abstract

The Cranfield field, located in southwest Mississippi, is an enhanced oil recovery and carbon sequestration project that has been under a continuous supercritical CO₂ injection by Denbury Onshore LLC since 2008. Two 3D seismic surveys were collected in 2007, pre-CO₂ injection, and in 2010 after > 2 million tons of CO₂ was injected into the subsurface. The goal of this study is to characterize a time-lapse response between two seismic surveys to understand where injected CO₂ is migrating and to map the injected CO₂ plume edge. In order to characterize a time-lapse response, the seismic surveys were cross equalized using a trace-by-trace time shift. A normalized root-mean-square (NRMS) difference value was then calculated to determine the repeatability of the data. The data were considered to have “good repeatability,” so a difference volume was calculated and showed a coherent seismic amplitude anomaly located through the area of interest. A coherent seismic amplitude anomaly was also present below the area of interest, so a time delay analysis was performed and calculated a significant added velocity change. A Gassmann-Wood fluid substitution workflow was then performed at two well locations to predict a saturation profile and observe post-injection expected changes in compressional velocity values at variable CO₂ saturations. Finally, acoustic impedance inversions were performed on the two seismic surveys and an acoustic impedance difference volume was calculated to compare with the fluid substitution results. The Gassmann-Wood fluid substitution results predicted smaller changes in acoustic impedance than those observed from acoustic impedance inversions. At the Cranfield field, time-lapse seismic analysis was successful in mapping and quantifying the acoustic impedance change for some seismic amplitude anomalies associated with injected CO₂. Additional well log data and refinement of the fluid substitution workflow and the model-based inversion performed is necessary to obtain more accurate impedance changes throughout the field instead of at a single well location.