Single-well experimental design for studying residual trapping of supercritical carbon dioxide

• Published in: International journal of greenhouse gas control 2011, Vol.5 (1), p.88-98
• Main Authors: Zhang, Yingqi, Freifeld, Barry, Finsterle, Stefan, Leahy, Martin, Ennis-King, Jonathan, Paterson, Lincoln, Dance, Tess
• Format: Article
• Language: English
• Subjects: Geologic carbon sequestration; Residual CO 2 saturation; Single-well test
• ISSN: 1750-5836; 1878-0148
• DOI: 10.1016/j.ijggc.2010.06.011

The objective of our research is to design a single-well injection-withdrawal test to evaluate residual phase trapping at potential CO 2 geological storage sites. Given the significant depths targeted for CO 2 storage and the resulting high costs associated with drilling to those depths, it is attractive to develop a single-well test that can provide data to assess reservoir properties and reduce uncertainties in the appraisal phase of site investigation. The main challenges in a single-well test design include (1) difficulty in quantifying the amount of CO 2 that has dissolved into brine or migrated away from the borehole; (2) non-uniqueness and uncertainty in the estimate of the residual gas saturation ( S gr ) due to correlations among various parameters; and (3) the potential biased S gr estimate due to unaccounted heterogeneity of the geological medium. To address each of these challenges, we propose (1) to use a physical-based model to simulation test sequence and inverse modeling to analyze data information content and to quantify uncertainty; (2) to jointly use multiple data types generated from different kinds of tests to constrain the S gr estimate; and (3) to reduce the sensitivity of the designed tests to geological heterogeneity by conducting the same test sequence in both a water-saturated system and a system with residual gas saturation. To perform the design calculation, we build a synthetic model and conduct a formal analysis for sensitivity and uncertain quantification. Both parametric uncertainty and geological uncertainty are considered in the analysis. Results show (1) uncertainty in the estimation of S gr can be reduced by jointly using multiple data types and repeated tests; and (2) geological uncertainty is essential and needs to be accounted for in the estimation of S gr and its uncertainty. The proposed methodology is applied to the design of a CO 2 injection test at CO2CRC's Otway Project Site, Victoria, Australia.