ANALYZING THE CO2 SEQESTRATION POTENTIAL OF THE MIDDLE DEVONIAN SYLVANIA SANDSTONE USING NUMERICAL SIMULATIONS

The Middle Devonian Sylvania Sandstone is a high quality brine reservoir in the Michigan Basin on the basis of decades of commercial brine mining and waste water disposal activities initiated in the early 1900's and comprises complex heterolithic, interbedded sandstone, limestone, sucrosic dolostone, and tripolitic chert lithofacies. Previous regional geological characterization studies indicate that the primary reservoir rock types are calcareous sandstone and recrystallized, sucrosic dolostone and that the Sylvania has large CO₂ storage resource potential ranging from 1.85-7.81 Gt, depending on storage efficiency assumptions. Large, estimated CO₂ storage resource potential in the Sylvania Sandstone in spatial proximity to large stationary emissions sources in Michigan justifies high resolution static reservoir and injection simulation modeling studies.

Static reservoir models of the porosity, permeability, and lithology and numerical simulation models using Schlumberger's Petrel-2011 and GEM, respectively, were used for higher resolution evaluation of the CO₂ sequestration potential of the Sylvania Sandstone. The static models were generated from conventional core analysis data from 38 wells and core-based sedimentary facies analysis from 4 wells in Midland County, MI. These data serve as the basis for grids used in transient GEM simulations.

A primary goal of the numerical simulations is to evaluate the influence of lithologic heterogeneity on CO₂ migration and capillary entrapment of CO2. Relative permeability analyses from sandstone and dolostone were incorporated in flow models to constrain multiphase fluid flow properties. Sensitivity analyses of the base transient model was conducted to test the effects on plume migration from variations in the horizontal-to-vertical permeability ratio, the CO₂ injection rate, the number and spacing of injection wells, and the geologic model. The sensitivity analysis also aids in the optimization of storage efficiency in available pore space. These local scale transient models can be compared to regional storage resource potential estimated using traditional geological approaches. Storage resource estimates for most of Midland County can be evaluated by simulating injection of CO₂ into the Sylvania Sandstone until steady state is attained.