Using oxygen isotope ratios to quantitatively assess trapping mechanisms during CO₂ injection into geological reservoirs: The Pembina case study

Carbon Capture and Storage (CCS) is considered a viable option for reducing CO₂ emissions into the atmosphere from point sources such as coal-fired power plants. Monitoring of CO₂ storage sites is widely considered necessary for safety reasons and for verification of injected CO₂ in the reservoir. The latter is crucially dependent on the ability to determine CO₂ trapping mechanisms and to assess pore-space saturation of CO₂. Thus far, attempts to determine CO₂ pore-space saturations at CO₂ injection sites have had limited success. Here, we present data from the Pembina Cardium CO₂ Monitoring Project in Alberta, Canada, that demonstrate that changes in the oxygen isotope ratios (δ¹⁸O) of reservoir water can be indicative of the extent of pore-space saturation with CO₂. The δ¹⁸O value of injected CO₂ at the injection site was +28.6‰ (V-SMOW) and δ¹⁸O values of reservoir water at eight observation wells varied between −13.5 and −17.1‰ (V-SMOW) before CO₂ injection. After commencement of CO₂ injection the δ¹⁸O values of reservoir water at three observation wells increased between 1.1 and 3.9‰ due to the presence of large quantities of injected CO₂ and equilibrium isotope exchange between water and CO₂. Our calculations revealed that reservoir water fully saturated with CO₂ would only result in increases of δ¹⁸OH₂O values of 0.4‰. Hence the observed larger increases in δ¹⁸O values of reservoir water indicate free phase CO₂ with estimated pore-space saturations ranging from 12% (corresponding to a δ¹⁸O increase of 1.1‰) to 64% (δ¹⁸O increase 3.9‰) of the non-oil saturated pore-space. Contributions to oxygen in the system from mineral dissolution were calculated to be less than 0.01% of total oxygen and therefore did not alter the δ¹⁸O value of the reservoir water significantly. Hence we conclude that changes in the δ¹⁸O values of reservoir water caused by the presence of injected CO₂ can be used as a tracer for CO₂ plume migration in the subsurface provided that the injected CO₂ is isotopically distinct. Furthermore, we submit that the extent of the changes in the δ¹⁸O values of the reservoir water provides a quantitative assessment of CO₂ stored in dissolved form (solubility trapping), assuming no density driven convective overturn, and as free-phase CO₂ (structural, stratigraphic and hydrodynamic trapping) thereby elucidating the trapping mechanisms within the reservoir.