CO2 STORAGE RESOURCE POTENTIAL OF A DEEP SALINE AQUIFER: ST. PETER SANDSTONE, MICHIGAN BASIN, USA

The St. Peter Sandstone is an aerially extensive, deep saline aquifer that occurs at about 1000 m (3200 ft) to 3700 m (12,000 ft) below the surface in the Michigan Basin, USA, and exceeds 335 m (1100’) in thickness near the basin center. The upper third of the St. Peter is dominated by sedimentary facies deposited in a normal marine, wave dominated shelf environment, while the lower two-thirds is less well understood, but was apparently deposited under a range of more restricted peritidal, marine conditions.

Preliminary CO₂ storage resource estimates (SRE) were calculated using data from hundreds of wells, mostly drilled for petroleum exploration and production purposes, that penetrate the entire thickness of the St. Peter Sandstone in the basin. Net porosity was estimated using well log porosity data and basin-wide grid maps were developed. Previously published SRE (DOE-NETL Carbon Sequestration Atlas III) using generalized, isopach and regional average porosity data, suggest 8.2 to 35.9 gigatons (Gt) of CO₂ storage capacity. The SRE methodology used in this study indicates 13.5 Gt to 58.7 Gt of CO₂ storage capacity in Michigan. This analysis also showed that the majority of storage capacity of the St. Peter Sandstone is in the lower two thirds of the formation. These results are the outcome of studies of new cores now available from the lower part of the formation. Reservoir characterization data, including sedimentary facies analysis and refined petrophysical characterization/calibration of well logs, has been generated from this new material.

Atlas III SRE methodology incorporates uncertainty referred to as the Storage Efficiency Factor (SEF). The SEF addresses uncertainty assigned to 1) net to total area, 2) net to gross thickness, 3) effective to total porosity, and 4) various fluid displacement mechanisms in the saline reservoir storage formations. The first three sources of uncertainty can be reduced or eliminated when well log and conventional core data from many uniformly distributed wells are available in a basin. Higher SRE generated in this study are a result of significantly refined reservoir characterization in the St. Peter Sandstone as a result of extraordinary availability of subsurface data and significantly reduced uncertainty and, therefore, increased SEF.