Bob Harrison FEI
Petroleum Engineer: Reserves Auditor: Project Manager: Carbon storage: Geothermal development: Repurposing infrastructure: Training Provider

When is a CO2 leak from a subsurface storage site not a leak? The answer is when it is a migration of CO2 into the surrounding storage complex. At least, that is what the European Council & Parliament suggests, and they should know as the definitions outlined in their DIRECTIVE 2009/31/EC for geological storage of CO2 have been adopted and adapted by countries that operate in the North Sea. Yet, in the case of a saline aquifer which has not undergone CO2 injection, how do we derive the size of a storage complex? Will the size change over time? And will this affect monitoring of the injected CO2 volume?

Table 1. Definitions of storage site, complex, and leakage from published regulations for geological storage of CO2

Table 1 shows the definitions of storage site, storage complex, and leakage used within Europe and the North Sea. The published guidance describes "storage site" as a volume area within a geological formation used to store CO2 with associated surface and injection facilities. The larger "storage complex" includes the storage site AND surrounding geological domain which can influence the overall storage integrity and security. The dreaded term, "leakage", is defined as any release of CO2 from the storage complex.

From these definitions, one can state with impunity that the CO2 plume which has escaped from the anticlinal structure of the Sleipner storage site, which probably began prior to 2020, has not “leaked”, rather it has “migrated” into the surrounding Sleipner storage complex. And if the latter’s area is defined to be very large, then it is very unlikely that one would ever have to report a “leak”.  

Yet, the regulations are vague on the setting of the boundaries of a storage complex, and this vagueness will inevitably introduce subjectivity into whether CO2 has leaked or not. One might think that the larger the storage complex negotiated between Operator and Regulator, the bigger the area that will need to be monitored, and that this would increase the cost and frequency of surveys, but is this the case and is it a cause for concern?

Does it matter?

There is a generally held belief amongst many in academia and the energy industry that if we can inject CO2 into the subsurface, whether it be a saline aquifer or a depleted gas field, then it will stay there forever, and we don’t need to worry about it anymore.

Their faith is reflected by the Intergovernmental Panel on Climate Change, who have reported that the fraction of CO2 retained in “appropriately selected and managed” storage sites is very likely to exceed 99% over 100 years and is likely to exceed 99% over 1000 years (IPCC, 2005). This view is echoed by the Zero Emissions Platform, who advise the EU on CCS matters, and who claim that, for a typical North Sea storage site, over 99.99% of injected CO2 is expected to remain stored deep underground for at least 500 years. They also state (from a proprietary report) that the chance of lateral migration from a storage site is only 3% (ZEP, 2019). But what do the Norwegians, who are the acknowledged world leaders in carbon geo-sequestration offshore, think of all this?

I was fortunate enough to discuss the topic with knowledgeable people from Equinor, the Operator of Sleipner, and from the Norwegian Petroleum Directorate, who are the country’s regulator. I thank them for their candour in answering my questions and their willingness to share information with me.

Operator’s view on Sleipner CO2 plume “migration”

Figure 1 shows the Top Utsira Formation in two-way time (TWT) in the Sleipner area. The Utsira is the sandstone aquifer in which CO2 is stored. The areal extent of the Sleipner CO2 storage site AND the storage complex is given by the red polygon. Note that the site and the complex are regarded having the same lateral definition and are therefore considered to be the same by Equinor.

Equinor’s defined area of the site/complex is approved by the NPD according to the Norwegian regulations for CO2 storage, so the storage site is not limited to the small local anticline above the injection well that tends to be shown in the numerous papers on the Sleipner project. According to Equinor, the current (western) lateral extent of the CO2 plume is up to 3 km from the injection point. They state that this is in line with predictions from the basis of design of the Sleipner CO2 storage project as modelled by Bakild et al (1996). In other words, the CO2 was always expected to migrate within the Utsira Formation away from the small anticline. In addition, Equinor believes that the CO2 plume will not expand "forever" as its long-term fate is for the CO2 to be dissolved in the Utsira Formation’s brine as predicted by numerical simulations by Lindeberg and Bergmo (2003). Equinor stated that the historical CO2 plume expansion rate has been slow (up to 100m per year) and that over 95% of the planned volume of CO2 to be stored has already been injected. Therefore, they are confident that there is no possibility of CO2 leakage from the Sleipner storage complex.

Figure 1. The background map is the top structure of the Utsira Sandstone in TWT, not in depth, where green denotes highs and blue denotes lows. The Sleipner CO2 storage complex area, as approved by the NPD, is defined by the solid red polygon. The 2020 Sleipner seismic survey area is defined by the dashed black polygon. The CO2 plume is outlined in black (map kindly provided by Equinor, 2023)

It should be noted that the maximum CO2 plume speed was stated to be 250m/year (in the NNE direction) by Torp (2007) and reportedly had to be increased threefold in the current interpretation model to match the latest seismic survey results (PGS, 2022). With Sleipner injecting around 1 Mt of CO2 per year, this implies that several Mt of CO2 have migrated out of the storage anticlinal structure into the surrounding Utsira Formation and are continuing to do so.

With regards to surveillance, the area defined by the black dashed lines in Figure 1 is the approximate area and shape of the 2020 seismic acquisition grid used for time-lapse monitoring purposes. The survey grid size is about one quarter of storage complex area.

Figure 2 reveals that the grid area of 1994 was shrunk in 2008 and then subsequently enlarged in 2020, which implies that existing Norwegian regulations allow one to change and adapt the seismic monitoring grid throughout the life of a CO2 storage project and for the grid to be smaller than the storage complex area. Equinor stated that, in their view, a seismic monitoring outline can be adjusted to accommodate future CO2 plume expansion, particularly in the case of Sleipner, as there are available seismic baseline data for the entire storage complex (defined by the red polygon).

Figure 2. Sleipner seismic survey areas over time with CO2 plume footprint from time-lapse analysis of 2020 data. Migration of CO2 out of the Sleipner anticline is shown in the west primarily, but also in the north and in the south (after PGS, 2020)

Regulator’s view on Sleipner CO2 plume “migration”

The NPD believe that the “migration” of CO2 which has been injected into the Utsira Sandstone aquifer is “as expected and in line with the storage permit”. They opine that the seismic data from the monitoring survey of 2020 indicate that CO2 was not migrating out of the Sleipner storage anticline, but they acknowledge that CO2 may migrate out of the structure to the west in the future (NPD, 2023). Note that this view is opposite to that of Sleipner project’s seismic contractor, who stated that CO2 has migrated out from the storage anticline (PGS, 2022).

Even with CO2 escaping from the storage site, the NPD are confident that it will migrate to and be stored in another nearby anticline, so they are unconcerned that CO2 will move spatially across large distances as they believe it will remain contained in the Utsira Formation.

The map shown in Figure 3 illustrates the NPD view that any CO2 migrating westwards from the Sleipner storage anticline will move preferentially into the structural high outlined in blue, which lies to the to the north of the Sleipner facility and its wells. Most production wells are drilled from the Sleipner-A facility and are situated within the cluster of black squares shown in Figure 3. Also, there are three wells drilled from Sleipner-D facility, in the northern part of the field. The wells are deviated, but since the Utsira Formation is much shallower than the reservoir, the deviation from the platforms is quite small in the storage formation.

Figure 3. Top structure of the Utsira Formation in the Greater Sleipner area map in TWT with a contour interval of 5 msec. The structures, outlined in white, blue, and red, would look very similar if the map were shown in depth. The injection well is drilled from the Sleipner-A facility and ends close to well 15/9-13. The black squares show the position of installations, some of which are at the sea floor. Most production wells are drilled from the Sleipner-A facility and are situated within the cluster of black squares (map kindly provided by NPD, 2023)

The NPD do not believe that CO2 will come into contact with the wells of the Sleipner-A facility as they penetrate the Utsira formation down dip from the mapped structural highs in the north, south and west. Also, as the Sleipner licence has a storage permit of 25 Mt of CO2 in total, the NPD claim that this is an insufficient volume to fill up the mapped structural highs to allow CO2 to reach the wells. Yet, Hansen and Haugen (2007) believed there was a “medium risk” of CO2 migration along the Utsira Formation to adjacent Sleipner wells, which would introduce corrosive “wet CO2” to the casings. Due to the age of the Sleipner wells, it seems likely that the casing strings over the Utsira Formation interval and their associated cement sheaths are not CO2-resistant.

So, how do the NPD view the risk of containment of a CO2 volume within a structural closure compared with a mobile CO2 volume in an open saline aquifer?

What is medium risk?

Regarding risk estimation, the NPD admitted they would estimate the risk of CO2 storage in an open aquifer differently to CO2 stored in a structural trap, but they were adamant that any CO2 spillage from the Sleipner storage structure will migrate to and be stored in another nearby anticline. Hence, the NPD considers aquifer storage in the case of Sleipner to be somewhat irrelevant as they are confident that all CO2 will remain trapped within the Utsira Formation.

When asked about the Operator’s assertion that there was a “medium risk” of the migrating CO2 plume contacting Sleipner wells and what probability value does this correspond to, the NPD admitted that they did not make any risk estimates for their CO2 Atlas (which is an excellent document by the way) as it was in the early stages of production by Halland et al (2011).

So, what probability could we assign to the “medium risk” estimate made by the Operator in 2007?

When debating global warming, researchers are advised to treat uncertainties consistently when developing their expert judgments and in evaluating and communicating the uncertainty in their findings. Table 2 shows the “calibrated language” of the IPCC which provides qualitative descriptions of the confidence in one’s research results which have a corresponding quantitative probabilistic measure to express the degree of certainty in one’s findings.

Table 2. Defined calibrated language for developing expert judgments and communicating certainty in climate change research results (IPCC, 2005)

Hence, the IPCC lexicon could describe “medium risk” as ranging from a 1-in-3 chance to something as high as 2-in-3, so maybe 50:50 is a fair reflection of the likelihood that some CO2 may reach the Sleipner well sections that have penetrated the Utsira Sandstone.

Summing up

The published regulations relating to the subsurface storage of CO2 in the North Sea are based largely on the DIRECTIVE/2009/31/EC, which is more of a guideline and is vague on how to define a storage site vs. a storage complex. Any regulations that are based on the Directive will inherit the same subjectivity. For example, the "geological surroundings" of a storage complex could be defined as being close to the structural spill point (as in the case of the Endurance structure in the UK Southern North Sea) or they could be large in relation to the storage structure (as in the case of Sleipner).

Academic studies emphasise that structural trapping is the most important factor in the early stages of CO2 geo-sequestration and implies a higher likelihood of containment. Yet, many are comfortable in regarding CO2 that has escaped from storage structures into the surrounding saline aquifer formation as having the same likelihood of being contained forever. This probably stems from the Directive’s assertion that a storage site is simply part of a greater storage complex.

Equinor and the NPD have always accepted that the CO2 injected into the Sleipner structure would escape eventually, and both are very confident that it will be captured subsequently by one of the other structures in the surrounding Utsira Formation. Yet, 15 years ago, Statoil (the forerunner of Equinor) stated that there was a "medium chance" that escaped CO2 could migrate towards the Sleipner production wells and pose a potential threat of creating a future leakage pathway. Has our confidence in the seismic modelling of the growth and distribution of the Sleipner CO2 plume grown so much in the intervening years that we believe the likelihood of lateral containment has improved significantly, from around 50% to almost 100%?

Furthermore, can we ever be sure of permanent lateral containment if no further wells are planned to evaluate the other structures identified in a storage complex?

Equinor’s assertion that the surveillance grid area can be made smaller or bigger depending on the currently perceived CO2 plume size and distribution is debateable. If the plume moves to an area where seismic coverage is more sparse and much older (which usually means of inferior data quality), then the accuracy of the time-lapse modelling is likely to be impaired. Maybe we need specific criteria on required spatial coverage, noise-to-signal level, allowable age of the baseline seismic, etc. needed for plume surveillance to be added to the published CO2 storage regulations, i.e. more rules and less guidance.

The disagreement between Regulator and Contractor over whether CO2 has escaped from the Sleipner structure should be a cause for concern. It suggests that we need more appraisal wells to fill gaps in our knowledge with respect to saline aquifer characterisation. As the aquifers being drilled through are rarely cored or tested (such as the Utsira Formation) any mapping of their spatial properties such as permeability will have inherent uncertainty, and thus mobile CO2 plumes in these aquifers may go to places that the models did not predict.

The combination of subjective regulations and unshakeable faith in subsurface models of poorly characterised saline aquifers does not seem the most solid ground from which to claim that injected CO2 volumes will be stored forever.

References

• Baklid et al (1996) “Sleipner Vest CO2 disposal, CO2 injection into a shallow underground aquifer”, paper SPE-36600-MS presented at the SPE ATCE, Denver, Colorado.
• Equinor (2023) personal correspondence with author – gratefully acknowledged.
• Halland et al (2011) “CO2 Storage Atlas, Norwegian North Sea”.
• Hansen & Haugen (2007) “Risk evaluation - Sleipner CO2 storage”, presentation by Statoil at IEAGHG risk workshop.
• IPCC (2005) “Carbon dioxide capture and storage”, special report by the Intergovernmental Panel on Climate Change.
• Lindeberg & Bergmo (2003) “The long-term fate of CO2 injected into an aquifer”, presented at Greenhouse gas control technologies conference.
• NPD (2023) personal correspondence with author – gratefully acknowledged.
• PGS (2022) “Broadband processing improves CO2 monitoring - Sleipner, North Sea”, case study from the seismic contractor’s website.
• Torp (2007) “10 years of CO2 storage”, presentation by Statoil Research Centre at Response to Climate Change workshop in Zagreb.
• ZEP (2019) “CO2 storage safety in the North Sea: implications of the CO2 Storage Directive”, report by Zero Emissions Platform.