April 2007
Special Report

United Kingdom: CO2 injection into North Sea fields will increase oil recovery


Substantial potential exists for improved oil recovery and CO2 sequestration using existing offshore facilities.

Vol. 228 No. 4  

Technology from Europe: United Kingdom

CO2 injection into North Sea oil fields will increase oil recovery.

 Substantial potential exists for improved oil recovery and CO2 sequestration using existing offshore facilities. 

Eugene Balbinski, Maggie Thompson and Frank O. Folorunso, RPS-Energy

There is presently a widespread acceptance of the possibility of major climate change due to increased levels of greenhouse gases such as CO2 in the atmosphere. Some governments and other organizations, including the UK, Norway and the EU, have perceived a need to take measures to reduce the magnitude of climate change. One technique is Carbon Capture and Storage (CCS), in which waste CO2 from sources such as power stations may be stored in geological structures, such as oil and gas reservoirs or aquifers. In 2005, the EU introduced Phase 1 of an Emissions Trading Scheme (EU-ETS) for carbon, which is expected eventually to incorporate CCS schemes. A North Sea basin Task Force, made up of governments and private and public organizations from the North Sea rim, was established with the aim of developing regulation and management of CO2 storage under the North Sea.

In 2006, the UK government conducted an energy review, which concluded that the UK had natural and commercial advantages to benefit from CCS technology.1 It will therefore continue to work with international partners to remove regulatory barriers and promote a commercial demonstration. The UK government is also committed to strengthening the EU-ETS, with the aim of developing a more efficient carbon market.

Two major groundbreaking North Sea CCS Improved Oil Recovery (IOR) projects have been proposed recently. In 2005, BP and its partners announced engineering studies on CO2 injection from a new power plant at Peterhead into the Miller reservoir for IOR and CO2 sequestration. In 2006, Shell and Statoil announced a large IOR and sequestration project injecting CO2 from a power station at Tjeldbergodden into the Norwegian Draugen and Heidrun fields.

This article reports on a recent update of estimates of CO2 injection potential for UKCS oil fields, for IOR and associated CO2 sequestration, prepared for the UK Department of Trade and Industry. It is important to note that the CO2 sequestration potential is only for CO2 available as a byproduct of IOR injection.

One of the many challenges in the offshore implementation of CO2 injection into oil reservoirs for IOR is the management of inevitable CO2 back production through producing wells. Most applications to date have been in North American onshore reservoirs, which offer greater reservoir management flexibility. Some results from detailed 3D, field-scale, multi-well modelling of Water Alternating Gas (WAG) CO2 injection are also reported, including conclusions from a simple economic model.


These estimates were made originally in 2001, based on a major screening exercise dating to the early 1990s, and were revised substantially in 2004. The results presented here have been updated for 2006 field Cessation Of Production (COP) estimates. This revision considered potential assuming, where possible, that reservoirs will be developed using down-dip WAG injection, rather than crestal injection. Although for some reservoirs, ideal implementation of crestal injection could maximize both IOR and CO2 sequestered, this will be delivered significantly later than for WAG, with additional costs if CO2 back-production can be managed effectively.

In economic regimes where most of the value is in IOR, rather than CO2 sequestration credits, WAG is likely to be more economic than crestal injection. Fortunately, WAG appears to be significantly more robust than crestal injection and is a more appropriate choice on which to base general estimates.

Estimates are field specific using systematic screening criteria and are restricted to UKCS oil reservoirs with STOIIP of at least 100 million bbl. They are informed by a number of exemplar detailed modelling studies, designed for the purpose, and earlier relevant modelling studies, including some on miscible hydrocarbon gas injection. Special estimates have been made for reservoirs into which hydrocarbon gas has already been injected. In some cases, analog reservoirs have been used to derive estimates.

A fairly conservative approach was adopted. Reservoir types with inherent CO2 injection difficulties were not included. In particular, viscous oils, oil rims, fractured reservoirs and gas-condensates were excluded. A review, done for the DTI, on CO2 injection into UKCS viscous oils suggested that potential IOR mechanisms, which may be operating for some non-UKCS reservoirs mentioned in the literature, would not apply.2 Oil rims would need to be examined on a case-by-case basis, which may not be appropriate at this stage. Gas condensates may well have some potential,2 but this is less certain than for oil reservoirs.

Figure 1 shows UKCS IOR and associated CO2 sequestration potential declining from 2006 to 2030, assuming this potential uses existing facilities. Such potential plots may be sensitive to COP estimates, since these were made in a high-oil-price regime. Present oil potential is estimated at about 2 billion bbl and associated CO2 potential at 1.2 billion t. As noted earlier, the sequestration potential is the CO2 available only as a byproduct of injection for IOR. It could be larger if additional CO2 was injected without IOR.

Fig. 1

Fig. 1. UKCS IOR and associated CO2 sequestration potential declines from 2006 to 2030. 


The model represents a typical heterogeneous North Sea sandstone reservoir and includes 21 producers and 12 peripheral injectors, Fig. 2. It was waterflooded maintaining average field pressure through voidage replacement for over 20 years, when a recovery factor of 40% had been achieved and the field oil rate had declined to about 10,000 bpd. The target production rate was set at 10% of the initial hydrocarbon pore volume per year, with a minimum field rate of 10,000 bpd. Producers were constrained to a maximum 90% watercut and minimum oil rate of 1,000 bpd.

Fig. 2

Fig. 2. The model represents a typical North Sea sandstone reservoir�s well positions (top) and permeability. 

The waterflood recovery is typical for a UKCS field of this type, especially as downdip gas injection normally would be started some time before the end of normal field life. Where water has flooded, remaining oil saturations are low, typically 25 to 35%, (Sorw = 25%). However, there is a significant volume of bypassed oil in disconnected pathways, Fig. 3.

Fig. 3

Fig. 3. Remaining oil saturations are low with bypassed oil in disconnected pathways, post-waterflood (top) and post-CO2 injection. 


CO2 injection was modeled from the end of the waterflood for six years by converting the water injectors into WAG or Continuous Gas Injection (CGI) injectors. Miscible CO2 injection was modelled using the Todd-Longstaff approach. The wells were controlled by maintaining the reservoir pressure at its initial value with a maximum gas production rate of 7 MMcfd, a maximum watercut of 95%, a minimum oil rate of 500 bpd and a minimum BHP of 1,200 psi.

Some effort was made to optimize CO2 injection, through both simulator controls and ad hoc adjustments. This included phasing of production wells, adjustment of CO2 injection rates to oil production and varying CO2 injection cycle length from three to nine months, each with a total cycle time of one year. This provided cases with WAG ratios of 1:3, 1:1 and 3:1.

CGI was also simulated (equivalent to a WAG ratio of zero). All production wells were reopened annually to equivalence the WAG cases. Other optimization methods were tried, such as closing high GOR completions, but these did not improve outcomes significantly.

The bottom of Fig. 3 shows an example of an oil saturation plot for the WAG 1:1 ratio case. Note that the residual oil saturation in the presence of CO2 was set to 5%. There is some reduction in the size of the >35% saturation region compared with the end of the waterflood. However, the lowest oil saturation range, 5 to 15%, is only achieved in a thin layer at the top of the formation. For WAG 3:1, this is quite small, and it is largest for the CGI case. If CO2 injection was continued longer, this region would have been larger.

The IOR is similar for all cases between 5% and just over 6% STOIIP, Fig. 4. Note that as CGI oil recovery tails off more quickly, it only gives the third highest IOR. This contrasts with the CO2 volume sequestered, for which CGI is highest at about 15% of the IHPV and WAG 3:1 is lowest with only about 6%.

Fig. 4

Fig. 4. Injection efficiency measures, (a) IOR and CO2 (top), CO2 utilization (middle) and flooding efficiency. 

The middle graph in Fig. 4 shows both the gross and net CO2 utilizations increasing as the WAG ratio decreases, indicating declining flood efficiency as a greater proportion of gas is injected. For example, the lower graph in Fig. 4 shows that the CO22 reservoir volume required to produce one IOR reservoir volume increases from 1.4, at a WAG ratio of 3, to 3.4 at zero. CGI stores more CO2 because it gives a less-efficient

The recycling ratio, the ratio of the gross to net CO2 utilization, is plotted at the top of Fig. 4. It is relatively low at about 1.4 and does not vary much with WAG ratio. This is because producing wells are closed if they start to back-produce too much CO2 and other optimizing assumptions. The recycling ratio is substantially lower for all these cases than for prior injector-producer sector models. This demonstrates that in a realistic field situation, as opposed to a sector model, excessive back-production may be managed effectively. The cost of this back-production, if managed effectively, only has a marginal effect on the economics.


A simple economic model was used incorporating the value of both IOR and CO2sequestration. Since the CO2 injection duration was limited to six years, discounting was not included. The model assumes prior and continuing waterflooding during WAG. The value of the IOR from WAG/CGI is the product of oil price and IOR volume. The CO2 sequestration value is the product of CO2 credit value and total CO2 sequestered. The WAG/CGI injection cost is the sum of the costs of capturing, transporting, injecting and back-producing CO2 plus water injection and base facility costs, Table 1.

TABLE 1. CO2 costs.
Tab 1

Cost estimates are general estimates with a fair degree of uncertainty and will change.3 These costs do, however, reflect the current expectation that capture and transport costs will dominate injection and production. The capture cost assumes a purpose-built, state-of-the-art power plant, such as coal-fueled Integrated Gasification Combined Cycle (IGCC).4 The capture cost with a standard gas turbine plant could double. Transport costs could be reduced, if offshore implementation were widespread. Field injection and production costs will be field specific, though different estimates did not give a wide range.5

Figure 5 shows the breakdown of the estimated cost per IOR bbl for each of the four WAG ratios considered. These all reflect the cost domination of importing CO22. Lower WAG ratios cost more as they inject more CO2 and more is likely to be back-produced.

Fig. 5

Fig. 5. A simplified economic model showing estimated cost breakdown (left) and net project values. 

In EU-ETS Phase 1, the non-compliance fine is 40 €/t ($51/t at $1.26/€), which sets a limit on the likely trading price. In June 2006, CO2 credits were trading under EU-ETS at about 15 €/t, after reaching a peak in 2005 of 30 €/t.1 From 2008 to 2012 in EU-ETS Phase 2, the non-compliance fine will be raised to 100 €/t.

With these factors in mind, Fig. 5 shows net project values calculated for oil prices of $20, $35, $50/bbl and CO2 credit values of $0, $25 and $50/t. All of these scenarios are substantially positive with the exception of the $20/bbl/scenario, which isn’t quite negative, and zero CO2 credit, which provides a reality check from earlier studies. In these price scenarios, the value of oil dominates that of CO2, though the CO2 value does affect which WAG scheme is the most valuable. In this particular example, it is a WAG ratio between 1:3 and 1:1, though this will also be field specific and depend on the relative prices.

At about $46/tonne, Fig. 6 shows the value-cost breakdown of the $35 oil, $50 CO2 scenario. Even with this apparently conservative oil price, the value of the oil dominates. The value of CO2 credits is comparable, but not quite sufficient to cover the estimated WAG injection costs. It should be noted that CO2 credit value is likely to rise in the longer term. WAG injection costs may also fall, if there is more implementation of CO22 injection. The bottom of Fig. 6 plots the net injection cost per IOR bbl against CO2 sequestration credit values for each WAG scheme modelled. At low CO22 credit values, CGI is most expensive, but as the CO2 credit value rises, it becomes the cheapest. At about $55/t, the CO2 credit value equals the estimated capture, transport and injection costs.

Fig. 6

Fig. 6. The value-cost breakdown of the $35 oil, $50 CO2 scenario shows that oil value dominates, value cost breakdown, (top) and injection cost/IOR bbl vs. CO2 credit value. 


The UKCS currently has substantial IOR and CO2 sequestration potential from CO2 injection into UKCS oil fields, but if this potential is to be realized using existing facilities, it declines with time. Field-scale modelling suggests that CO2 back-production may be managed effectively in North Sea reservoirs.

A simple economic model suggests that CO2 injection for IOR could be economic in some North Sea reservoirs. Under currently expected price scenarios for CO2 injection into North Sea oil reservoirs, the value of oil will dominate, but CO2 injection could contribute substantially to the value. WO  


The authors thank the UK Department of Trade and Industry for permission to publish this article, which was performed under the Oil and Gas Maximising Recovery Programme (OG-MRP)..


 1 UK Government Energy Review 2006, http://www.dti.gov.uk/energy/review/.
2 Dissemination section of DTI OG-MRP website: http://www.og-mrp.com/dissemination/d-co2.html.
3 Gozalpour, F., Ren S. and B. Tohidi, “CO2 injection for IOR and storage: Opportunities and challenges for the North Sea,” IORViews, issue 10, http://ior.rml.co.uk/issue10; Carbon dioxide capture and storage: A win-win option? (The Economic Case), May 2003, DTI/Pub URN 03/812: http://www.dti.gov.uk/files/file18798.pdf; Confidential internal DTI Reports.
4Hanstock, D., Progressive Energy Ltd., interviewed by author.
5 Gozalpour et al; “Carbodioxide capture and storage,” 2003; OTI Reports.


Eugene Balbinski is a principal reservoir engineer for RPS Energy with over 20 years’ experience performing reservoir engineering studies. He is a specialist in modelling gas injection processes, including CO2 injection with a particular interest in carbon capture and storage. He can be contacted at BalbinskiE@rpsgroup.com..


Maggie Thompson is a principal reservoir engineering advisor for RPS Energy with over 20 years’ experience. She has extensive knowledge and experience of enhanced oil recovery techniques including CO2 injection. She can be contacted at ThompsonM@rpsgroup.com.


Frank O. Folorunso is reservoir engineer for RPS Energy with experience in CO2 injection studies through MS degree research and commercial projects. He can be contacted at FolorunsoF@rpsgroup.com.


Related Articles
Connect with World Oil
Connect with World Oil, the upstream industry's most trusted source of forecast data, industry trends, and insights into operational and technological advances.