Capstone: CO₂ sequestration, Sleipner revisited

The Sleipner field in the Norwegian North Sea has injected CO₂ into the Utsira Sand since 1996. It is the world's longest-running industrial carbon capture and storage (CCS) project and the most-published 4D monitoring case. A 1994 baseline and repeated monitor surveys (from 1999, every one to three years) track the CO₂ plume and provide the evidence for containment claims.

Project setup

About 1 Mt of CO₂ a year injected at roughly 1010 m depth into the Utsira Sand, a saline aquifer about 200 to 300 m thick, sealed by the Nordland shale. Repeat 3D surveys: a 1994 baseline and monitors in 1999, 2001, 2002, 2004, 2006, 2008, 2010, 2013 and 2016, with later surveys continuing. Primary deliverable: a 4D difference volume per vintage showing the plume's extent and shape. Secondary: per-layer saturation maps through rock-physics inversion.

The pipeline

As delivered, the figure's overburden NRMS is 35 %, above the 15 % target, and every event under the plume is up to 32 ms late, so below it the difference shows misalignment as much as CO₂. After the full flow the overburden NRMS is 11 % and the reservoir window's difference is 55 times the overburden's. Step the stages and watch in (d) which of them buy the drop, then test the stringer above the caprock with Leave out set to each stage.

Why joint processing is the central theme

The CO₂ signal at Sleipner is large: bright reflections from thin CO₂ layers and tens of milliseconds of velocity pushdown beneath them. The faint parts that matter for containment, the thin plume edges and anything above the caprock, must still stand above the non-repeatability of surveys decades apart. Any difference between surveys that is not geological (wavelet mismatch, drift in processing parameters, legacy versus modern acquisition) contaminates the plume map and can lead to a wrong call on containment. Joint processing with shared parameters, together with tide and water-velocity statics, cross-equalisation and time-shift warping, is what keeps the normalised RMS difference, $\\mathrm{NRMS}=200\\,\\mathrm{RMS}(m-b)/(\\mathrm{RMS}(m)+\\mathrm{RMS}(b))$, low and consistent from vintage to vintage.

Regulatory dimension

• Containment evidence. The operator must demonstrate that injected CO₂ stays in the approved storage complex. 4D difference volumes are the primary evidence; NRMS QC certifies that the observed plume shape is real.
• Mass reconciliation. The CO₂ mass inferred from 4D (volume times saturation times CO₂ density) should be consistent with the metered injected mass within its sizeable uncertainty. A persistent mismatch triggers investigation; evidence of leakage is a 4D anomaly above the caprock or outside the licensed complex that exceeds the local repeatability noise.
• Temporal consistency. Each monitor must be comparable with all previous monitors. The joint processing and 4D matching are not optional: they are the audit trail.

What Sleipner actually showed

Nine bright reflections mark CO₂ pooled beneath thin (1 to 2 m) mudstone layers inside the Utsira Sand. The mudstones are leaky baffles, not seals: CO₂ reached the top of the sand within about three years, and the uppermost layer has kept spreading laterally beneath the caprock. Quantitative studies find the seismically mapped CO₂ broadly consistent with the metered injected mass, within uncertainties of tens of percent set by saturation, layer thickness and CO₂ density, and no leakage through the caprock has been detected.

Where this goes next

Section 10.5 changes scale entirely: an ultra-high-frequency near-surface survey for engineering geotechnics, where depths and wavelengths are one to two orders of magnitude smaller than in marine exploration.

References

• Yilmaz, Ö. (2001). Seismic Data Analysis (2 vols.). SEG.
• 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, 1383.
• Chadwick, R. A., Williams, G., Delepine, N., Clochard, V., Labat, K., Sturton, S., Buddensiek, M.-L., Dillen, M., Nickel, M., Lima, A. L., Arts, R., Neele, F., Rossi, G. (2010). Quantitative analysis of time-lapse seismic monitoring data at the Sleipner CO₂ storage operation. The Leading Edge, 29, 170.
• Sheriff, R. E., Geldart, L. P. (1995). Exploration Seismology (2nd ed.). Cambridge UP.