High-resolution characterization of a CO2 plume using crosswell seismic tomography: Cranfield, MS, USA

• Published in: International journal of greenhouse gas control Vol. 18; pp. 497 - 509
• Main Authors: Ajo-Franklin, J.B., Peterson, J., Doetsch, J., Daley, T.M.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2013
• Subjects: Borehole seismic methods; Crosswell seismic imaging; Geologic CO2 storage; Rock physics; Seismic monitoring methods; USA; USA, Alabama, Tuscaloosa; Geologic CO2 storage; Crosswell seismic imaging; Borehole seismic methods; Rock physics; Seismic monitoring methods
• ISSN: 1750-5836; 1878-0148
• DOI: 10.1016/j.ijggc.2012.12.018

► Multiple crosswell seismic surveys were acquired at the Cranfield pilot to monitor CO2 migration and saturations. ► Baseline survey provided a detailed view of reservoir structure in the interwell region. ► Timelapse analysis revealed significant P-wave velocity reduction due to CO2 injection. ► Conversion of seismic velocity changes to CO2 saturation was accomplished using White's model for patchy saturation. ► Resulting CO2 distribution estimate suggests flow localization in two smaller regions of the Tuscaloosa D/E unit. We present the results of a high-resolution time-lapse crosswell seismic survey carried out at a large-scale CO2 injection pilot located in Cranfield, MS, USA. This dataset, spanning an injector and two monitoring wells, provided a detailed view of the boundaries and internal structure of the injection unit, the Tuscaloosa D/E sand. Time-lapse tomographic processing of one well pair revealed the signature of the injected plume, two zones of decreased P-wave velocity which spatially correspond to higher permeability sections of the reservoir unit. We then used White's model for patchy saturation, combined with secondary information from core and well-log measurements, to convert the imaged changes in P-wave velocity into estimates of supercritical (sc) CO2 saturation in the interwell region. Our estimate is relatively consistent with existing lithologic and core data as well as independent measurements of scCO2 saturation derived from pulsed neutron logging. While effective in mapping the zone of CO2 invasion in the interwell region, several sources of uncertainty still exist in quantifying saturation using crosswell seismic methods, including (but not limited to) the degree of mesoscale mixing or “patchiness” at scales below a seismic wavelength.