Understanding the impact of the level of characterization on long-term performance predictions at geologic CO2 sequestration sites

The estimates on geologic CO2 storage capacity suggest that deep saline reservoirs have the largest potential CO2 sequestration capacity. On the other hand, deep saline reservoirs have not been as well characterized as oil and gas reservoirs. While detailed characterization of proposed sequestration sites will be necessary, it is important to assess the impact of detailed information on predictions of a site’s long-term performance assessment and subsequently risk assessment. In this paper we discuss this issue through application of a systems level performance assessment model to a CO2-EOR (enhanced oil recovery) site, SACROC.
SACROC is the oldest CO2- EOR field in the United States, in operation for over 35 years. As part of the U.S. Department of Energy’s Regional Partnership program, a sequestration pilot test is planned at SACROC. As a part of this project characterization data such as, seismic surveys, well logs, production histories, geochemical information, and other data have been collected. The characterization data has been used to construct a geologic model of the site which will be used to predict the movement of CO2 and other fluids (such as reservoir brine) in the subsurface.
The Los Alamos National Laboratory (LANL) and the National Energy Technology Laboratory (NETL) are collaborating on developing approaches to geologic CO2 sequestration risk assessment. As part of this collaboration we are applying CO2-PENS, a CO2 sequestration site performance assessment model developed by LANL, to multiple sequestration pilot tests. CO2-PENS is a coupled process-systems model that integrates field/laboratory observations with numerical models and abstractions to predict long-term performance of a geologic CO2 sequestration site. The model accounts for CO2 migration in the primary reservoir and beyond through potential leakage pathways such as wellbores, faults etc.
For this work, we describe how we apply CO2-PENS to the SACROC site. We describe incorporation of potential leakage pathways in CO2-PENS, and how the probability of possible CO2 and brine leaks at the site is quantified. We discuss how results of predictions of CO2 plume migration through the geologic model described above are incorporated in CO2-PENS. Finally we demonstrate what impact a detailed geologic characterization has on the predicted CO2 migration through leakage pathways. This is done by generating geologic models with varying level of characterization data and incorporating the predicted CO2 plume migration results in CO2-PENS.