Uncertainty in Regional-scale Evaluation of CO2 Geologic Storage Resources—comparison of the Illinois Basin (USA) and the Ordos Basin (China)

Abstract

To meet the ambitious goals of both the United States and China for advancing the deployment of Carbon Capture Utilization and Storage technology will require an improved understanding of the magnitude and geographical distribution of carbon geologic storage (CGS) resources. Evaluation of CGS resources and the fractional component of practical storage capacity is thus a major focus of research in both nations. In this paper, our purpose is to carefully evaluate the sources of uncertainty that propagate into regional-scale CGS resource estimates and assess the extent to which uncertainty may be reduced by applying increasingly informed levels of geologic characterization. We achieve this objective by focusing on two geologic basins of great similarity but with varying amounts of data the Illinois Basin in the United States and the Ordos Basin in China. Our investigation focuses on storage resource in deep saline formations because large-scale greenhouse gas mitigation is expected to require the use of this CGS resource. To accurately compare results from different basins and different nations requires applying a common methodology for estimating CGS resources. In this study we follow the methodology published by the U.S. Department of Energy.Results demonstrate that in both the Illinois and Ordos Basins, review of the open literature is adequate for identifying the saline formations that should be considered as potential targets for geologic storage (i.e., regionally extensive porous formations having an overlying low permeability seal at a minimum depth of 800 meters). Analysis of such prior work allows for an initial, simple quantification of CGS resources at regional scales by applying probabilistic-based storage efficiency factors to generalized maps of bulk formation characteristics. Resource characterization may be improved through more advanced analysis when data are available, however, demonstrating that enhanced characterization leads to a quantifiable reduction in uncertainty appears problematic. This difficulty in demonstrating reduced uncertainty led us to identify some key issues in applying the published Department of Energy (DOE) methodology for resource estimation.Our primary finding is that the DOE methodology underestimates storage resource uncertainty because it does not account for error in the total formation bulk rock pore volume. The methodology uses storage efficiency factors that account only for uncertainty in the fraction of this total formation pore volume that can effectively store CO2. Thus the uncertainty range for formation storage resource is based on just a single realization of the total formation pore volume. To evaluate the significance of this impact we use data from the Mount Simon Sandstone formation in the Illinois Basin to account for this additional uncertainty in a probabilistic manner. The resulting storage resource estimates at the 10th and 90th percentile probability range from 18 to 313 gigatonnes, compared to 24 and 253 gigatonnes when following the standard published methodology. Also noteworthy from our analysis of the Mount Simon Sandstone is the fact that our results are significantly larger than the resource estimates published in the DOE Sequestration Atlas (11 and 151 gigatonnes for the 10th and 90th percentile range). It appears that the range in the published resource estimate values for the Mount Simon Sandstone may be significantly underestimated not only because of the formation pore volume issue discussed above, but also as a result of over discounting the resource by applying the published total storage efficiency factor values to a formation area that was already reduced to a net area. Our results provide insight on some important issues and challenges to applying the DOE methodology in a broader range of CGS resource characterization analyses. We conclude that further refinement of the DOE methodology is necessary to provide a more robust assessment of the uncertainty that exists in regional-scale CGS resource estimates.