Forecasting commercial-scale CO2 storage capacity in deep saline reservoirs: Case study of Buzzard's bench, Central Utah

Abstract

Successful implementation of Geological Carbon Sequestration (GCS) projects requires long-term storage capacity and security at selected fields. The purpose of this work is to evaluate and quantify potential storage capacity, trapping mechanisms and potential risks associated with commercial-scale (at least 50 MT CO2) injection and storage of CO2 in the subsurface. The specific case study evaluated is the Navajo Sandstone formation, a target GCS reservoir in Buzzard's Bench, Central Utah. Two-dimensional reactive transport models are designed for 1000-year simulations, with 50 MT CO2 injection. An uncertainty analysis with different reservoir porosity and permeability are conducted, because only a few samples of the reservoir and caprock near the proposed reservoir were collected. Specific objectives include forecasting the extent of an injected CO2 plume, competing roles of different trapping mechanisms, trapping capacity changes due to porosity changes, and forecasts of CO2 migration into adjacent caprock (the Carmel formation in this case). Resultssuggest that the Navajo formation may be a reliable CO2 sequestration reservoir, capable of trapping commercial volumes. The Jurassic Kayenta and Wingate formations may also store some injected CO2, with these and other clastic formations forming a “stacked storage” system. Storage efficiency decreases with distance away from an injection well, and the estimated storage efficiency for the case study simulations (Navajo storage only) are 2.3 ± 1% within the area of review (AoR) calculated by National Risk Assessment Partnership (NRAP) toolset. After 1000 years, about half of the injected CO2 may be sequestered in safe phases including residually-trapped CO2 via surface tension, aqueous and mineral phases. A small amount of total injected CO2 (∼3%) tends to migrate into the caprock, but is mostly stored in the sandstone reservoir. Simulated porosity enhancement caused by mineral alteration is negligible within 1000 years of the start of injection, with only ∼0.6% added to the total pore volume by the end of simulations. Future studies of detailed reservoir and caprock characteristics with in-situ samples may be helpful for further determining reservoir sequestration capacity and reliability.