Geologic Modeling and Simulation Report for the Aquistore Project

The Plains CO2 Reduction (PCOR) Partnership, through the Energy & Environmental Research Center (EERC), in collaboration with the Petroleum Technology Research Centre (PTRC), has constructed static and dynamic geologic models to simulate and assess the potential carbon dioxide (CO2 ) storage at the Aquistore site. The Aquistore project is part of the world’s first commercial postcombustion carbon capture, utilization, and storage project from a coal-fired power-generating facility, the SaskPower Boundary Dam, located in Saskatchewan, Canada, and will be acting as a storage site for a portion of the captured CO2 from the Boundary Dam power plant. The Aquistore site includes one injection well and a 500-foot offset observation well. Both wells were drilled and completed in the Deadwood and Black Island Formations. At the time of this report, injection at the Aquistore site is anticipated to begin in mid- to late 2014. To better understand the storage implications of injecting at the Aquistore site, the EERC developed a geocellular model of the basal saline system for the dual purposes of determining the static CO2 storage capacity and as a basis to run detailed reservoir simulation to determine injectivity, dynamic storage capacity, and breakthrough time at the observation well. To compensate for a lack of well control locally, a regional-scale model was first constructed to determine the regional stratigraphic reservoir and nonreservoir zones. From this regional model, a fine-scale model was confined to the extent of the 13.1-square-mile PTRC 3-D seismic survey area, with higher structural resolution. Integration of the data derived from the regional model and the data from the 3-D seismic survey helped create a robust and heterogeneous model around the Aquistore injection well and the observation well. As a first pass, the detailed 13.1-square-mile model was used to estimate a static storage capacity employing the U.S. Department of Energy methodology and resulted in a range of static storage resource of approximately 8.4 to 27.1 Mt for the P10 to P90 confidence intervals, respectively. This result indicated that our model was probably big enough to model a short- to medium-duration injection of perhaps 5–30 years at 1 Mt/yr; however, it may be too small to adequately simulate a 50-year injection period. To further evaluate the targeted saline system, and thus its viability as a potential storage horizon for CO2 , the geocellular model was used as the framework for an assessment of the dynamic storage capacity of the system. Two scenarios were designed based on the static geologic model. The first investigated the injectivity of the system and the timing of CO2 breakthrough at the observation well in a 13.1-mi 2 area. The second scenario, which will be detailed in a subsequent report, encompasses a 3670-mi 2 area. As part of this investigation, core plug analysis and relative permeability studies were also conducted on samples provided from the injection well core. Information from these analyses was integrated into the construction of the geocelluar model and the dynamic simulations and will be provided in a subsequent report. A total of nine simulation cases were run to investigate factors such as boundary conditions, injection rates, and time length. The injection duration for these scenarios was set at 1, 5, and 50 years, and the injection rates were set at 1 Mt/yr and 0.3 Mt/yr. Although the maximum injection rate in the model was set as 1 Mt/yr, the maximum attained in the model was 0.73 Mt/yr because of bottomhole pressure limitations. The total mass of CO2 injected in the 50-year cases ranged from 1.5 to 33.6 Mt, with the large range in values a result of changing the boundary conditions from closed to open. CO2 storage values for the 5-year cases range from 1.5 to 3.6 Mt, and those for the 1-year cases range from 0.3 to 0.7 Mt. An important aspect of this investigation with regard to potential monitoring efforts is the timing of CO2 breakthrough at the observation well. The earliest breakthrough occurred between 10 and 15 days at the higher injection rate (0.73 Mt/yr), and the projected CO2 path follows the top reservoir zone of the Deadwood Formation. At the lower injection rate (0.3 Mt/yr), breakthrough happens between 25 and 30 days after injection and follows the same path. Overall, CO2 breakthrough in most of the reservoir zones happens in about 3 months for the low injection rate; this time is reduced to 45 days at the higher rate. Based on the information derived from the various simulation cases, the CO2 breakthrough will most likely happen in the first month of injection regardless of the injection rate and assumptions of relative permeability. Based on the simulation results, the storage of CO2 in the study area using the existing two-well configuration is feasible, depending on the volume of CO2 that need to be injected and stored from the neighboring Boundary Dam power plant. Generally, the maximum injectivity for the current injection well could reach 0.73 Mt/yr based on the geological characterization of the study area. However, this could be improved through optimization operations such as adding additional injection wells, utilizing formation water extraction wells, and/or the use of horizontal injection wells. All of these additional optimization techniques will be investigated in the next phase of work and reported on in a subsequent report. In addition, the larger regional-size model will be utilized to provide better insights with respect to a commercial-scale injection rate over a long period of time. Finally, future work will also include geomechanical, geochemical, and geothermal behaviors and integrate them throughout the entire modeling and simulation process to investigate the role these variables may play in CO2 storage at the Aquistore site.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.