Saline Aquifer CO2 Storage (SACS2). Final report, geological characterisation of the Utsira Sand reservoir and caprocks (Work Area 1)

This report summarises the results and highlights the main findings of SACS Work Area 1, the geological and reservoir characterisation of the Utsira Sand and its caprock. For more detailed technical information on each topic, the reader is directed to the relevant SACS Technical Reports and, in particular, two earlier Work Area 1 interim reports, Holloway et al. (1999) and Chadwick et al. (2000). The Utsira Sand comprises a basinally-restricted deposit of Mio-Pliocene age forming a clearly defined seismic unit, pinching out to east and west, and seismically distinct from overlying and underlying strata.The reservoir is highly elongated, extending for more than 400 km from north to south and between 50 and 100 km from east to west, with an area of some 26100 km2. Its eastern and western limits are defined by stratigraphical lap-out, to the southwest it passes laterally into shaly sediments, and to the north it occupies a narrow channel deepening towards the More Basin. Locally, particularly in the north, depositional patterns are quite complex with some isolated depocentres, and lesser areas of non-deposition within the main depocentre. The top Utsira Sand surface generally varies relatively smoothly, mainly in the range 550 to 1500 m, but mostly from 700 to 1000 m. The base of the sand is more irregular, disturbed by diapirism of the underlying shales. Isopachs of the reservoir sand show two main depocentres. One is in the south, around Sleipner, where thicknesses range up to more than 300 m. The second depocentre lies some 200 km to the north of Sleipner. Here the Utsira Sand is locally 200 m thick, with an underlying sandy unit adding further to the total reservoir thickness. Macroscopic and microscopic analysis of core and cuttings samples of the Utsira Sand show that it consists of a largely uncemented fine-grained sand, with medium and occasional coarse grains. The grains are predominantly angular to sub-angular and consist primarily of quartz with some feldspar and shell fragments. Sheet silicates are present in small amounts (a few percent). The sand is interpreted as being deposited by mass flows in a marine environment in water depths of 100 m or more. The porosity of the Utsira Sand core ranges generally from 27% to 31%, but reaches values as high as 42% Regional log porosities are quite uniform, in the range 35 to 40% over much of the reservoir. Geophysical logs show a number of peaks on the -ray, sonic and neutron density logs, and also on some induction and resistivity logs. These are interpreted as mostly marking thin (~1m thick) intrareservoir shale layers. The shale layers constitute important permeability barriers within the reservoir sand, and have proved to have a significant effect on CO2 migration through, and entrapment within, the reservoir. The proportion of clean sand in the total reservoir thickness varies generally from about 0.7 to nearly 1.0. The caprock succession overlying the Utsira reservoir is rather variable, and can be divided into three main units. The Lower Seal forms a shaly basin-restricted unit, some 50 to 100 m thick. The Middle Seal mostly comprises prograding sediment wedges of Pliocene age, dominantly shaly in the basin centre, but coarsening into a sandier facies both upwards and towards the basin margins. The Upper Seal comprises Quaternary strata, mostly glacio-marine clays and glacial tills. The Lower Seal extends well beyond the area currently occupied by the CO2 injected at Sleipner and seems to be providing an effective seal at the present time. Cuttings samples comprise dominantly grey clay silts or silty clays. Most are massive although some show a weak sedimentary lamination. XRD analysis typically reveal quartz (30%), undifferentiated mica (30%), kaolinite (14%), K-feldspar (5%), calcite (4%), smectite (4%), albite (2%), chlorite (1%), pyrite (1%) and gypsum (1%) together with traces of drilling mud contamination. The clay fraction is generally dominated by illite with minor kaolinite and traces of chlorite and smectite. The cuttings samples are classified as non-organic mudshales and mudstones. Although the presence of small quantities of smectite may invalidate its predictions, XRD-determined quartz contents suggest displacement pore throat diameters in the range 14 to 40 nm. Such displacement pore throat diameters are consistent with capillary entry pressures of between about 2 and 5.5 MPa capable of trapping a CO2 column several hundred metres high. In addition, the predominant clay fabric with limited grain support resembles caprocks which are stated in the literature to be capable of supporting a column of 35 API oil greater than 150 m in height. Empirically, therefore, the caprock samples suggest the presence of an effective seal at Sleipner, with capillary leakage of CO2 unlikely to occur. Around and east of the injection point, a layer of sand, 0 - 50 m thick, lies close to the base of the Lower Seal and is termed the Sand-wedge. The geometry of this unit is likely to prove important in determining the long-term migration behaviour of the CO2. Fluid flow in the Utsira Sand, based on limited pressure measurements and basin-modelling, is likely to be low, in the range 0.3 – 4 metres per year, depending on assumed permeabilities. The total pore-space within the Utsira Sand is estimated at 6.05 x 1011 m3. However not all of this can necessarily be utilised for CO2 storage. The simplest assumption is that long-term storage of CO2 can only be accomplished in structural traps at the top of the reservoir. A detailed study around Sleipner indicates that 0.3% of the reservoir porosity is actually situated within structural closures such as this. In practical terms moreover, with a small number of injection wells, it is unlikely that all of the small traps could be utilised in any case. Around Sleipner the most realistic estimate of the pore-space situated within accessible closed structures is just 0.11% of the total pore-volume. On the other hand, trapping of CO2 beneath the intra-reservoir shales could significantly increase realisable storage volumes, particularly if it encouraged dissolution of CO2 into the groundwater. Similarly trapping of CO2 in the Sand-wedge, as well as beneath the top of the Utsira Sand, will increase the overall storage capacity significantly. In conclusion, the theoretical storage capacity of the Utsira Sand is very high, but how much of this can be utilised in reality is uncertain, and a function of several complex parameters. Migration models have been constructed with 30 x 106 m3 of CO2, injected into the Utsira Sand (approximating to the expected final injected mass of 20 million tonnes). They show that if the CO2 is trapped at the top of the Utsira Sand it will migrate generally northwestward, reaching a maximum distance from the injection site of about 12 km. However, if the CO2 is trapped within the Sand-wedge, migration is less well constrained, being northwards then northeastwards. Data limitations to the east of the injection point preclude quantitative estimates of the maximum migration distance in this case.