Watching CO2

The Same Physics, a Different Purpose

Carbon storage turns the rock physics of this course toward a monitoring question: after CO2 is injected into a saline aquifer, can a repeated seismic survey see it, and can it say how much is there? The answer is a clean split, yes to the first and barely to the second, and both halves come straight from the pore-fluid mixing law. Take the standard soft brine sand at porosity 0.25 and mix in supercritical CO2 at representative storage values, a bulk modulus near 0.08 GPa and a density near 0.65 g per cc. These are representative figures for typical storage conditions; CO2 properties vary strongly with depth and temperature, so a real project measures them for its own reservoir.

The brine sand carries an impedance of 6.47. Displace 30 percent of the brine with CO2 and the impedance drops to 5.11, a fall of 21.1 percent. That is an enormous, unmistakable time-lapse signal: any monitoring survey with a detection threshold of a few percent will light up the CO2 the moment it arrives. Detection is loud.

The Curve That Saturates

Now push the saturation higher and watch what does not happen. At 60 percent CO2 the impedance drop is 22.9 percent; at 100 percent it is 24.1 percent. The whole journey from 30 percent to fully saturated moves the impedance only another 3 points, while the first 30 percent moved it 21. The response saturates almost immediately. This is the Wood average of Part 3.5 again: a uniform brine-CO2 mixture takes its fluid modulus from the reciprocal average, which collapses toward the soft CO2 value as soon as any CO2 is present, and then flattens.

That flat curve is the quantification problem in one picture. Because 30 percent and 100 percent saturation give nearly the same impedance, a measured amplitude cannot tell them apart. Mapping where the CO2 is, the plume outline, is easy, because the edge is a sharp step from zero to a large change. Measuring how much is at any point inside the plume is hard, because the amplitude is nearly blind to saturation above a few tens of percent. A monitoring program can draw the plume boundary with confidence and still argue about the tonnes it contains.

The Green Jacket on an Old Problem

Look back at the first section of this part and the shape is identical. The fizz-water trap of 10.1 was exactly this curve: a little gas makes almost the whole amplitude change, so amplitude cannot read saturation. Watching CO2 is the same Wood-collapse physics wearing a green jacket, the storage-monitoring twin of the DHI-risking problem. The detect-easy, quantify-hard asymmetry is not a limitation of a particular survey; it is written into the pore-fluid mixing law, and every partial-saturation problem in the course inherits it. For the seismic side of the same monitoring story, the CO2 capstone of the Synthetic Seismic Modeling course carries this rock into a time-lapse difference volume. And that closes the interpretation capstones. All that remains of this course is the review and exam of Part 11, which consolidate what you have built, and then the Lab of Part 12, where you drive every model in this course yourself.

References

• Arts, R., Eiken, O., Chadwick, A., Zweigel, P., van der Meer, L., & Zinszner, B. (2004). Monitoring of CO2 injected at Sleipner using time-lapse seismic data. Energy, 29(9-10), 1383-1392.
• Mavko, G., Mukerji, T., & Dvorkin, J. (2009). The Rock Physics Handbook (2nd ed.). Cambridge University Press.