Modelling large-scale carbon dioxide injection into the Bunter Sandstone in the UK Southern North Sea

https://doi.org/10.1016/j.ijggc.2012.03.011Get rights and content

Abstract

The Bunter Sandstone in the UK sector of the Southern North Sea Basin is a reservoir rock that is typically 200 m or more thick and has variable but commonly fair to good porosity and permeability. East of the Dowsing Fault Zone it is folded into a number of large periclines as a result of post-depositional halokinesis in the underlying Zechstein salt. It is sealed by the overlying Haisborough Group and younger fine-grained strata and is underlain by the Bunter Shale and Zechstein Group. As such it appears to be an attractive target for industrial-scale CO2 storage. However, the very large masses of CO2 that would have to be injected and stored if CCS is to be an effective greenhouse gas mitigation option are likely to cause (a) significant pore fluid pressure rise and (b) displacement of formation brines from the reservoir. A series of reservoir flow simulations of large-scale CO2 injection was carried out to investigate these effects. A simple, 3D geocellular model of the Bunter Sandstone in the NE part of the UK sector of the Southern North Sea was constructed in the TOUGH2 reservoir simulator in which porosity and both horizontal and vertical permeability could be varied. The injection of CO2 at various rates into the model through a variable number of wells for 50 years was simulated and the model was then run forward for up to 3000 years to see how pore fluid pressures, brine displacement and CO2 distribution evolved. The simulations suggest that provided there is good connectivity within the reservoir, and 12 optimally distributed injection locations are used, 15–20 million tonnes of CO2 per year could be stored in the modelled area without the reservoir pore pressure exceeding 75% of the lithostatic pressure anywhere within the model. However, significant fluxes of the native pore fluid (saline brine) to the sea occurred at a point where the Bunter Sandstone crops out at the seabed. This suggests that the potential environmental impacts of brine displacement to the sea floor should be investigated. The injected CO2 fills only up to about 1% of the total pore space within the model. This indicates that pore fluid pressure rise may be a greater constraint on CO2 storage capacity than physical containment within the storage reservoir.

Highlights

► Static and dynamic modelling of CO2 storage in the Bunter Sandstone. ► Reservoir pressure modelled as a fraction of the lithostatic pressure. ► Reservoir pressure kept below 75% of lithostatic pressure during injection. ► CO2 storage capacity and rate estimated in the modelled area. ► Pore fluid displacement to sea bed modelled.

Introduction

Carbon capture and storage (CCS) is a greenhouse gas mitigation option in which CO2 generated from large scale fossil fuel combustion, e.g. in power stations, is captured and permanently stored in geological reservoirs (IPCC, 2005). If CCS is to be adopted, very large quantities of CO2 would have to be permanently stored, even from a single power plant. A new, pulverized coal-fired, supercritical power plant with a net electrical capacity of 1 GWe and 90% CO2 capture could send about 6 million tonnes (Mt) of CO2 p.a. for storage. By comparison, a modern gas-fired power station of 1 GW net electrical capacity with 90% capture could send around 2.5 Mt CO2 p.a. for storage.

One of the main strategic issues for large-scale CO2 injection and storage in saline water-bearing reservoir formations is the determination of their ultimate CO2 storage capacity. This is a key parameter for policymakers because, potentially, it could determine whether or not CCS is a viable greenhouse gas mitigation option in a particular jurisdiction. Estimates of the ultimate CO2 storage capacity of a saline water-bearing reservoir need to take into account not only the pore space available for CO2 storage but also the pore fluid pressure rise induced by the displacement of in situ pore fluids by CO2 around injection wells and, for large-scale CO2 injection, more generally within the storage formation (Bachu et al., 2007, NETL, 2008, Van der Meer and Egberts, 2008). When large-scale CO2 injection takes place, the reservoir rock's native pore fluids will be compressed and/or mobilised. Mobilisation of the native pore fluids could result in migration of pore fluid to the ground surface or seabed through onshore or seabed outcrops of the reservoir formation. Alternatively the displaced fluids could move into adjacent formations, which will become overpressured and their native pore fluids may in turn be displaced. Direct displacement of (potentially highly saline) brines to the water table/ground surface or seabed could have ecological impacts, which would need to be assessed on a case-by-case basis. These potential dynamic effects are investigated below, using the TOUGH2 reservoir simulator (Pruess et al., 1999, Pruess, 2005) to model large-scale CO2 injection into a simple model of the Bunter Sandstone Formation in the NW part of the UKSNS.

Section snippets

The Bunter Sandstone

The Bunter Sandstone is the offshore part of a major sandstone reservoir rock unit that onshore in Eastern England is known as the Sherwood Sandstone Group. Fig. 1 shows the outcrop and subsurface distribution of this rock unit, and eight gas fields that have Bunter Sandstone reservoirs. The Dowsing Fault Zone and Central Fracture Zone (Cameron et al., 1992) separate the shallower Eastern England Shelf from the deeper basin to the east. In the latter area, depth to top Bunter Sandstone is

Geological model of the Bunter Sandstone

A simplified 3D static geological model of part of the Bunter Sandstone in the NE of the UKSNS was built from isopach and depth maps (Fig. 5). These maps are based on the interpretation of a loose grid of wells and 2D seismic profiles and should be regarded as only broadly indicative of thickness and structure. The model spans some 130 km E–W by 116 km N–S and uses a grid of 152 × 131 × 15 elements, sized 800 m × 800 m over the central portion and expanding to 2 km × 2 km near the boundaries. Element

Pressure control considerations for the dynamic modelling

A set of Leak-Off Test (LOT, squares) and formation fluid pressure data (triangles) for the Southern North Sea are displayed in Fig. 6 relative to sea floor, a hydrostatic pressure gradient of 10.07 MPa/km (blue line) and a lithostatic pressure gradient of 22.5 MPa/km (red line).

The LOT and formation fluid pressures are adjusted to a sea floor depth and pressure reference. No data from depths greater than 3000 m are displayed as 3000 m is assumed to be the maximum depth appropriate for the Bunter

Dynamic modelling of CO2 injection into the Bunter Sandstone

Flow simulations were run to simulate the effects of large-scale CO2 injection into the Bunter Sandstone for a range of reservoir properties. In all simulations each injection point consists of a vertically stacked set of single cells that comprise the entire thickness of the Bunter Sandstone. CO2 was injected through these sets of cells at a constant annual rate for 50 years. The model was then allowed to run forward for a further 950 years (1000 years from the start of injection). Dissolution

Discussion

The modelling provides only a very simplified representation of the NW part of the Bunter Sandstone Formation in the UKSNS. In particular, the model does not include reservoir heterogeneity. Moreover, the reservoir pressure observed in the simulations is likely to be sensitive to the model's lateral boundary conditions, being significantly lower if the model's lateral boundary conditions are fully or partially open (Smith et al., 2011). Nevertheless it is clear that pore fluid pressure rise

Acknowledgements

The authors would like to thank IHS for permission to show pressure-related information from the Southern North Sea and David Hughes for much useful discussion of the Bunter Sandstone as a potential reservoir. This publication has been produced with support from the BIGCCS Centre. The BIGCCS Centre is part of the Norwegian research programme Centres for Environment-friendly Energy Research (FME) and is funded by the following partners: Aker Solutions, ConocoPhillips, Det Norske Veritas, Gassco,

References (25)

  • S. Bachu et al.

    CO2 storage capacity estimation: methodology and gaps

    International Journal of Greenhouse Gas Control

    (2007)
  • D.J. Smith et al.

    The impact of boundary conditions on CO2 storage capacity estimation in aquifers

    Energy Procedia

    (2011)
  • D.B. Bennion et al.

    Dependence on Temperature, Pressure and Salinity of the IFT and Relative Permeability Displacement Characteristics of CO2 Injected in Deep Saline Aquifers. SPE 102138

    (2006)
  • R. Bifani

    Esmond gas complex

  • T.D.J. Cameron et al.

    United Kingdom Offshore Regional Report: The Geology of the Southern North Sea

    (1992)
  • R.A. Chadwick et al.

    Flow processes and pressure evolution in aquifers during the injection of supercritical CO2 as a greenhouse gas mitigation measure

    Petroleum Geoscience

    (2009)
  • P. Cooke-Yarborough et al.

    The Hewett fields: Blocks 48/28a, 48/29, 48/30, 52/4a, 52/5a, UK North Sea: Hewitt, Big Dotty, Little Dotty, Della, Dawn and Delilah fields

  • M.J. Fisher

    Triassic

  • M.C. Geluk et al.

    High-resolution sequence stratigraphy of the Lower Triassic ‘Buntsandstein’ in the Netherlands and northwestern Germany

    Geologie en Mijnbouw

    (1997)
  • J.F. Harrington et al.

    Laboratory study of gas and water flow in the Nordland Shale

  • S. Holloway et al.

    Top-down and bottom-up estimates of CO2 storage capacity in the UK sector of the Southern North Sea Basin

    Environmental Geoscience

    (2006)
  • IPCC, 2005. IPCC Special Report on Carbon Dioxide Capture and Storage. Prepared by Working Group III of the...
  • Cited by (63)

    View all citing articles on Scopus
    View full text