Frio Brine Pilot: Field Validation of Numerical Simulation of CO₂ Storage*

By

Susan D. Hovorka ¹, Christine Doughty ², Shinichi Sakurai ³, and Mark Holtz ³
 

Search and Discovery Article #40165 (2005)

Posted August 20, 2005

*Oral presentation at AAPG Annual convention, Calgary, Alberta, June 19-22, 2005.

Click to view presentation in PDF format.

¹Bureau of Economic Geology, Jackson School of Geosciences, Austin, TX ([email protected])

²Lawrence Berkeley National Laboratory

³The University of Texas at Austin, Austin, TX

Abstract 

For the Frio Brine Pilot we injected 1,600 tons of supercritical CO₂ into a brine-saturated formation 1,500 m below surface and measured transport between two wells 30 meters apart through a moderately homogeneous, high permeability (2.5 Darcy) sandstone. Modeling using TOUGH2 with parameters based on detailed study of petrophysics predicted that breakthough of CO₂ to the observation well would occur in 2.5 to 6 days. During the injection phase of the field experiment, breakthough of CO₂ occurred 30 percent earlier than predicted. Logging using the Schlumberger reservoir saturation tool determined that, by the fourth day after the initiation of injection, the plume thickness was half what had been predicted and that saturation in preferred flow zones was higher than had been predicted by the model showing that factors favoring rapid flow such as high permeability zones or gravity override were larger than predicted. As injection continued for 9 days, plume thickness increased by addition of CO₂ in the lower part of the plume, culminating with an observed saturation profile similar to that predicted by the model. CO₂ saturation was variable but was estimated to be in excess of 50% of porosity.

Post injection saturation changes observed as CO₂ migrated though the steeply dipping sandstone under gravitational forces appears to reverse the trend observed during injection. Continued observation will yield information about saturation history significant to understanding CO₂ trapping mechanisms. Downhole pressure and temperature proved to be sensitive indicators of plume behavior, showing changes in plume as CO₂ saturation changed before and after breakthrough.

Initial Step: Site Search

Locating a high-permeability, high-volume sandstone representative of a broad area that is an ultimate target for large-volume sequestration

Conclusions

•      Plume front during injection: 30% faster and thinner than predicted

•      Stable plume at 10 days similar to modeled

•      Plume evolution during postinjection migration under gravity saturation controlled by decreasing relative
    permeability to CO₂ as saturation decreases