A Method for Preliminarily Assessing the CO2 Sequestration Potential of Deep Saline Aquifer Horizons

Monday, April 20, 2009: 3:50 p.m.
Canyon Suites I/II (Hilton Tucson El Conquistador Golf & Tennis Resort )
Jason W. Deardorff , U.S. EPA, Denver, CO
John McCray, Ph.D. , Environmental Science and Engineering, Colorado School of Mines, Golden, CO
Dag Nummedal, Ph.D , Colorado Energy Research Institute - Colorado School of Mines, Golden, CO
            In the case of a large-scale deployment of carbon capture and sequestration, deep saline aquifers will play an important role in supplying the required capacity for CO2 storage.  Few existing well penetrations make these formations especially attractive sequestration targets but also present the problem of little or no subsurface data for CO2 injection modeling.  An established analytical solution for modeling CO2 injection was adapted within Microsoft Excel by adding algorithms and a database for the rapid determination of fluid properties at known or expected reservoir conditions.  Monte Carlo simulation software was used to allow rapid computation of up to 32,000 injection simulations with random number generation of uncertain input parameters from expected ranges and probability distributions.  The adapted analytical model outputs a cumulative probability chart for maximum plume radius during the injection period allowing the determination of minimum, median, and maximum plume size resulting from injection into a given formation and the probability of realizing any single outcome as a function of the combined uncertainty of all input parameters.  The model and a screening methodology were applied to 16 known or expected aquifer horizons at the proposed site of a coal-fired power plant with carbon capture technology and used to rank these aquifers by their potential to sequester 40 years of emissions.  Five formations were determined to be inadequate, 5 formations were determined to be likely candidates due to favorable modeling results but highly uncertain subsurface data, and 6 formations were determined to have high potential for the sequestration of CO2 from the power plant based on expected plume size and the number of expected intersections with existing well infrastructure.  Statistical aspects of determining “aquifer average” model input parameters based on actually occurring or expected populations will be discussed.