ABSTRACT

CO2 geologic sequestration has been recognized as a potential greenhouse gas mitigation strategy. Regional CO2 geologic storage in deep saline formations will likely involve the injection of 10 to 100 million metric tons (11 to 110 million tons) of CO2 per year using a network of 10 to 50 wells over an area covering 10100 sq. miles (26259km2). Some of the wells will be injecting into closed volumes because of symmetry, thus providing the bounding case in terms of pressurization and brine efflux. This study describes a parametric analysis of the problem using characteristics typical of the Arches Province in the United States Midwest where Paleozoic rocks form broad arch and platform structures. Two-dimensional radial-cylindrical models developed with the numerical simulator STOMP (Subsurface Transport Over Multiple Phases) are utilized to investigate the impact of well spacing, injection depth, and reservoir characteristics of the injection zone (Mount Simon) and cap rock (Eau Claire) on system performance. Multiple linear regression analysis is then used to develop correlation equations between these design variables and performance metrics, such as cumulative CO2-mass injected and CO2-plume extent. The correlations are tested on new synthetic test sites, and are found to predict the performance metrics quite accurately. These results serve as a proxy simulator to quickly evaluate various design options, instead of having to run time-consuming numerical simulations, and can therefore be applied for developing optimal injection strategies for regional CO2 storage in the Arches Province.

You do not currently have access to this article.