Geological Survey of Denmark and Greenland Bulletin 17, 2009 Carbon capture and storage (CCS) is increasingly considered to be a tool that can significantly reduce the emission of CO2. It is viewed as a technology that can contribute to a substan- tial, global reduction of emitted CO2 within the timeframe that seems available for mitigating the effects of present and continued emission. In order to develop the CCS method the European Union (EU) has supported research programmes for more than a decade, which focus on capture techniques, transport and geological storage. The results of the numerous research projects on geological storage are summarised in a comprehensive best practice manual outlining guidelines for storage in saline aquifers (Chadwick et al. 2008). A detailed directive for geological storage is under implementation (European Commission 2009), and the EU has furthermore established a programme for supporting the development of more than ten large-scale demonstration plants throughout Europe. Geological investigations show that suitable storage sites are present in most European countries. In Denmark initial investigations conducted by the Geological Survey of Denmark and Greenland and private companies indicate that there is significant storage potential at several locations in the subsurface. The Danish perspective in storage capacity The ten largest point sources of CO2 emission in Denmark account for 21 mega-tonnes per year (Mt/year). From pre- liminary investigations of the Danish subsurface the CO2 storage capacity in selected subsurface structures is estimated to 2500 Mt (GeoCapacity 2009a). This corresponds to more than 100 years of storage from the ten largest emission point sources. The critical parameters of this analysis are the size of the structure, thickness, continuity and quality of the reser- voir and the amount of formation water that may be dis- placed by the injected CO2. The estimate is calculated assuming a surrounding aquifer volume displacement of for- mation water, which is limited to 50 times the trap volume (GeoCapacity 2009b). These estimates of storage capacity are uncertain and have not yet been tested in real physical storage operation. Therefore it is difficult to evaluate to what degree the volume calculations are realistic. When a specific structure is selected for storage, a number of investigative steps are necessary, including acquisition and interpretation of new 2-D or 3-D seismic data, drilling of new wells, geological and reservoir modelling and flow simulation studies. For each step of incorporating new geological data, the site model of the reser- voir is updated in the process of maturing the structure towards a storage site. This stepwise approach to site charac- terisation gradually leads to a research-based and relatively certain capacity estimate and an evaluation of the safety and behaviour of the site under simulated conditions, including the uncertainties of the estimates. Assessment of geological and environmental risks can be carried out at various stages in the process. Similarly, estab- lishment of baseline studies and monitoring strategies need to be considered along with the progress of the characterisa- © GEUS, 2009. Geological Survey of Denmark and Greenland Bulletin 17, 13–16. Available at: www.geus.dk/publications/bull The potential for large-scale, subsurface geological CO2 storage in Denmark Peter Frykman, Lars Henrik Nielsen, Thomas Vangkilde-Pedersen and Karen Lyng Anthonsen 13 �� �� �� �� �� �� �� �� �� �� Vedsted structure Stenlille structure Jylland Fyn Sjælland Ålborg Large point sources Structures �� 50 km Fig. 1. Map of Denmark showing the most important point sources of CO2 emission and prospective structures for geological storage of CO2. The Stenlille structure is presently used for storage of natural gas; it serves to moderate seasonal fluctuations in consumption. The Vedsted structure is currently investigated for possible storage of CO2. ROSA_2008:ROSA-2008 01/07/09 15:47 Side 13 tion of the storage site in order to make sure that the neces- sary background information is obtained before storage is ini- tiated. In order to be reliable and operational, the baseline studies should preferably focus on measurement of condi- tions and properties that are stable and only show limited sea- sonal variations. Such studies may include groundwater-flow models and groundwater chemistry, pore water and pore gas analyses from deep wells, surface topography and natural seismicity. A site study Site investigations have recently been initiated of the Vedsted structure by Vattenfall A/S, with the intention of using the structure for storage of CO2 from a nearby coal-fired power plant in Ålborg (Fig. 1; Sørensen et al. 2009). Existing data from oil exploration activities in the 1950s include one well in the centre of the structure and sparse 2-D seismic line data. The main target layer is the Triassic–Jurassic Gassum For - mation at around 1800 m depth. The formation is widely distributed in the Danish Basin and has good reservoir prop- erties (Fig. 2). It is currently used for storage of natural gas in the Stenlille structure on Sjælland and for geothermal energy in the Thisted area in northern Jylland. Detailed sedimentological and sequence stratigraphic interpretations and correlations of the well logs and cores have established a robust stratigraphic framework for the Upper Triassic – Jurassic succession (Nielsen 2003). This framework forms the basis for the interpretation of the Vedsted-1 well section as well as predictions regarding the lithology of the potential reservoirs and seals in the Vedsted area (Fig. 3). The process has also underlined the necessity of acquisition of new data and more detailed modelling at sev- eral different scales. At site scale, the optimal positioning of injection wells, as well as injectivity and capacity can be modelled and analysed, and the coupling between the operation of the power plant and the capture facility can be studied. The specific geologi- cal properties of the storage reservoir layers have conse- quences for the propagation and distribution of the injected CO2 and for the storage mechanisms in the specific reservoir. Most reservoirs show both vertical and horizontal hetero- geneities that will influence the distribution of the CO2. The preliminary reservoir model for the Vedsted structure has been investigated by simulating an injection well on its south-eastern flank and using injection rates realistic for power-plant supply rates (Frykman et al. 2009). After ten years of constant injection, the CO2 distribution is as seen in Fig. 4, which clearly shows the subdivision of the migrating front into several sub-layers due to intraformational sealing layers with low permeability that also have high capillary entry pressures. The layering in the model has maximum lat- eral continuity, which probably overestimates the segregation to be found in real cases, but any intra-reservoir sealing layers will have such an effect on the distribution. Since this filling pattern influences the capacity, it is necessary to analyse fur- ther the properties and the continuity of the intraformational sealing layers. CO2 can be trapped by several mechanisms, including structural trapping under an overlying sealing formation, dis- solution of CO2 in formation water, capillary trapping in the pore network and mineral trapping by reactions between CO2 and mineral phases in the reservoir rock. These trapping mechanisms work on different scales both in space and time and need to be studied by designing appropriate models and experiments. For large-scale injection of CO2 displacing saline porewa- ter, the propagation of the pressure field during injection out- side the immediate site area is of interest. Modelling of this pressure distribution will serve to predict the amount of over- pressure building up locally within the storage site, and can be used to suggest possible means of management. 14 50 km Fig. 2. Distribution of the Triassic–Jurassic Gassum Formation in the sub- surface at depths between 800 m and 2400 m (yellow), the depth inter- val in which CO2 exists as a supercritical phase and where burial dia- genesis has not yet provoked significantly lowered porosity. At the supercritical phase the volume of CO2 is much less than that of the CO2 gas at the surface. ROSA_2008:ROSA-2008 01/07/09 15:47 Side 14 The scale of the challenge and future perspective The first detailed pan-European assessment of CO2 storage capacity in the framework of the EU research project Geo - Capacity has resulted in a geographic information system (GIS) database of CO2 emissions, storage capacity estimates and geological information. The database includes informa- tion on reservoirs with a total storage capacity of 360 000 Mt CO2, with 326 000 Mt in deep saline aquifers, 32 000 Mt in depleted hydrocarbon fields and 2000 Mt in unmineable coal beds: 116 000 Mt are onshore, and 244 000 Mt offshore (GeoCapacity 2009a). Some of the estimated storage capac- ity is associated with structural traps, but a very large part is in regional deep saline aquifers without identified specific traps. Almost 200 000 Mt of the total storage capacity in the database are located offshore Norway. These estimates date back to 2003 and have not been updated within the GeoCapacity project. An attempt to provide a more cautious and conservative European estimate has yielded a storage ca - pacity of 117 000 Mt with 96 000 Mt in deep saline aquifers, 20 000 Mt in depleted hydrocarbon fields and 1000 Mt in coal beds, and with approximately 25% located offshore Norway. This must be compared to a total of 2000 Mt of CO2 emission from large point sources, i.e. point sources emitting more than 0.1 Mt/year within Europe. In order to illustrate the scale of the technology and infra- structure that has to be established if CCS is to become an active industry, we can look at the amount of CO2 produced by the ten largest point sources in Denmark. There are 43 large point sources emitting 28 Mt CO2/year, the ten largest of which are responsible for 21 Mt/year. At surface condi- 15 Fig. 3. SW–NE-oriented cross-section across the Danish Basin, the Sorgenfrei–Tornquist Zone and the Skagerrak–Kattegat Platform (red line on the index map). The panel shows the lower part of the Gassum reservoir which comprises fluvial, estuarine and shallow-marine deposits interbedded with off- shore mudstones and some lacustrine mudstones. Shoreline fluctuations have caused interfingering of these different facies types and given rise to pro- nounced vertical variability. Informations from the four wells in the section about sedimentary facies have been interpreted and correlated into a sequence-stratigraphic framework. At a local site, this framework must be confirmed from detailed investigations of material from new wells drilled, and supplemented with new seismic data. Modified from Nielsen (2003). 4 CE RL RL Skagerrak–Kattegat Platform Sorgenfrei–Tornquist Zone 37 km 5 km 41 km 30 km 20 km Børglum Fault ation PT PT TS 7 Danish Basin Sæby-1 Vedsted-1Rødding-1 SP Hyllebjerg-1Farsø-1 GR Flyvbjerg-1 NESW 32 50 m 50 km Sorgenfrei–Tornquist Zone Danish Basin Skagerrak–Kattegat Platform Fyn High Ringkøbing – Fluvial Estuarine Lacustrine Lagoonal Shoreface Offshore Depositional environments ROSA_2008:ROSA-2008 01/07/09 15:47 Side 15 tions this corresponds to 11 billion (11 × 109) m3 CO2 gas. The annual production of natural gas from the Danish part of the North Sea amounts to 10 billion (10 × 109) m3 (Danish Energy Agency 2008), which is transported in pipe lines and tankers and processed at plants and refineries. The compara- ble size of the potential volume of CO2, to be moved around at surface and injected into the subsurface (although com- pressed to smaller volumes at depth), points to the large scale at which a CCS-related processing and transporting industry has to be established. Concluding remarks The CCS activities described here related to large-scale stor- age operations will involve significant physical resources and manpower. Fortunately, the work with storage-related items does not have to begin from first principles, because much of the experience already exists in the oil and gas industry, which can provide methods and tools for immediate use. The skills of geoscientists and engineers are needed in the investigation and characterisation of the sites and the subsurface condi- tions for storage of CO2, and a whole new infrastructure and industry may be established. Geoscience and geo-engineering will play a major role in the analysis of the geological foun- dation, the assessment of site performance, and will be criti- cal in securing the safety of the operations. Initial investigations of the Danish subsurface indicate that suitable structural traps with a significant storage poten- tial are present at several locations, and that the structures can accommodate the CO2 produced from several or most of the large Danish point sources. Thus, geological storage of CO2 may contribute considerably to the reduction of the Danish CO2 emission, if we can be assured about safety issues, and if political and public acceptance can be obtained. References Chadwick, A., Arts, R., Bernstone, C., May, F., Thibeau, S. & Zweigel, P. 2008: Best practice for the storage of CO2 in saline aquifers – obser- vations and guidelines from the SACS and CO2STORE projects. British Geological Survey Occasional Publication 14, 267 pp. Danish Energy Agency 2008: Oil and gas production in Denmark 2007, 98 pp. Copenhagen: Danish Energy Agency. European Commission 2009: Directive of the European Parliament and of the Council on the geological storage of carbon dioxide, 84 pp. Brussels: European Union. Frykman, P., Bech, N., Sørensen, A.T., Nielsen, L.H., Nielsen, C.M., Kristensen, L. & Bidstrup, T. 2009: Geological modelling and dynamic flow analysis as initial site investigation for large-scale CO2 injection at the Vedsted structure, NW Denmark. 9th International Conference on Greenhouse Gas Control Technologies, Washington D.C., 16–20 No - vember, 2008. GHGT9 Energy Procedia 1, 2975–2982. GeoCapacity 2009a: GeoCapacity WP 2 Report, Storage capacity. EU GeoCapacity deliverable D16, 162 pp. EU GeoCapacity Consortium, Brussels. GeoCapacity 2009b: GeoCapacity WP 4 Report, Capacity standards and site selection criteria. EU GeoCapacity deliverable D26, 45 pp. EU GeoCapacity Consortium, Brussels. Nielsen, L.H. 2003: Late Triassic – Jurassic development of the Danish Basin and the Fennoscandian Border Zone, southern Scandinavia. In: Ineson, J.R. & Surlyk, F. (eds): The Jurassic of Denmark and Green - land. Geological Survey of Denmark and Greenland Bulletin 1, 459–526. Sørensen, A.T., Klinkby, L., Christensen, N.P., Dalhoff, F., Biede, O. & Noer. M. 2009: Danish development of a full-scale CCS demonstration plant in a saline aquifer. First Break 27, 79–83. 16 Authors’ address Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark. E-mail: pfr@geus.dk Fig. 4. Vertical NW–SE section in the Gassum reservoir model through the injection well, showing CO2 saturation Sg (free gas-phase supercritical CO2) after 10 years of injection. Although the model is constructed in a fairly coarse grid, the intra-reservoir sealing layers are clearly reflected and influence the spatial distribution of the injected CO2. The sedimen- tary layering causes filling of the individual layers of porous sand with the injected CO2, whereas the interbedded mudstone layers with much lower reservoir quality are not filled and also limit the vertical movement of CO2. Model thickness 300 m, length 4800 m, vertical exaggeration 5 times. (Modified from Frykman et al. 2008). 0.55 Sg 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 ROSA_2008:ROSA-2008 01/07/09 15:47 Side 16