Simultaneous CO2 injection and water production to optimise aquifer storage capacity

https://doi.org/10.1016/j.ijggc.2010.09.002Get rights and content

Abstract

The estimates for geological CO2 storage capacity worldwide vary, but it is generally believed that the capacity in saline aquifers will be sufficient for the amounts of CO2 that will need to be stored. The effort required to select and qualify a geological storage site for safe storage will, however, be significant and storage capacity may be a limited resource regionally. Both from a economic and resource management perspective it is therefore important that potential storage sites are exploited to their full potential.

In static capacity estimates, where the maximum stored amount of CO2 is given as a fraction of the formation pore volume, typically arrive at efficiency factors in the range of a few per cents. Recent work has shown that when the dynamic behaviour of the injected CO2 is taken into account, the efficiency factor will be reduced because of the increase in pore pressure in the region around the injection well(s). The increase in pore pressure will propagate much further than the CO2. The EU directive on geological CO2 storage specifically addresses the restriction that will apply when different storage sites are interacting due to pressure communication. Consequently, the pore pressure increase at the boundary of the storage license area will be an important limiting factor for the amount of CO2 that can be injected.

One obvious method to control the pore pressure is to produce water from the aquifer at some distance from the CO2 injection wells. This paper discusses results from simulations of CO2 injection in two aquifers on the Norwegian Continental Shelf; the Johansen aquifer and the southern part of the Utsira aquifer. These aquifers are candidates for injection of CO2 shipped out via pipeline from the Norwegian West Coast. The injected amounts of CO2 over a period of 50 years are 0.518 Gtonne for the Johansen aquifer and 1.04 Gtonne for the Utsira aquifer.

Several design options for the injection operations are investigated: Injection of CO2 without water production; injection into several wells to distribute the injected fluids and reduce the local pressure increase around each injection well; and injection with simultaneous production of water from one or more wells. The boundaries of the aquifer formations are assumed closed in all simulations. The possible consequences of other types of boundary conditions (semi-closed or open) are briefly discussed.

Introduction

Previous estimates of geological storage resources have indicated that the global capacity should not be a limiting factor for CO2 capture and storage (CCS) as a mitigation option to reduce climate warming (Holloway et al., 1996, IPCC, 2005, Vangkilde-Pedersen et al., 2009). Even if these estimates assume that only a fraction of the formation pore volume can be utilised as effective storage space for dense CO2, these estimates can only be realised if the formation pore pressure is effectively managed during the injection period (Lindeberg et al., 2009). It is important to realise that the pore space underneath the seal of a prospective aquifer storage site is already filled with water which has low compressibility. If the formation is assumed to be a closed unit, the only space that injected CO2 can occupy will be from compression of this water and from expansion of the pore space by compression of the rock minerals (usually negligible) and by increasing the bulk volume of the formation, (accompanied by an increase in the pore pressure of the formation). The low compressibility of rock and water means that utilisation of more than 1–2% of the pore space will lead to a pressure increase that might compromise the integrity of the cap rock or activate faults. If CCS shall be applied as a large scale option to mitigate climate warming (Odenberger et al., 2008) a larger fraction of the pore space will need to be utilised.

Some relief may be found in the observation that fully closed boundaries are rarely found for aquifer units. If open or semi-closed boundary conditions (lateral and vertical) for the storage formation is assumed, the utilised pore space may be larger, since pore water may then flow into neighbouring formations and some of the pressure increase will be avoided (Birkholzer and Zhou, 2009, Birkholzer et al., 2009, Nicot, 2008). The simulated pressure increase shown in this paper can therefore be considered to be conservative. However, if the formation(s) that the pore water would flow into also are underground resources of some kind (potential CO2 storage sites, potable water resources, etc.), conflicts of interest would arise which could make this kind of pressure control unfeasible. Uncontrolled displacement of formation water, typically brine, upward into shallow aquifers or to the surface is not a desired option for pressure control. The EU directive on geological CO2 storage (EU Directive, 2009) even contains a provision that a storage operation should be managed in such a way that it will not degrade other storage sites in the same hydraulic unit. A conclusion from this is that it will be necessary to control the pore pressure in some way to avoid that the pressure increase propagates too far from the injection point.

This paper investigates one potentially efficient way to actively control pressure in the reservoir by producing formation water from dedicated water production wells. The method is investigated through model simulation of two potential saline aquifer storage sites on the continental shelf west of Norway.

The location of the two aquifer formations used in this study is illustrated in Fig. 1.

The Johansen formation is a promising candidate for permanent storage of CO2 conveniently located near the coast only 70 km from Mongstad, a major CO2 source at the west coast of Norway. The lower Jurassic Johansen formation of the Dunlin Group consists of an east-west dipping sand stone formation with several large vertical faults in the north-south direction, some with a throw of several hundred meters. The north-western part of the formation lies around 600 m below the oil and gas bearing formations in the Troll field. The main faults also penetrates the Troll oil and gas field where they are sealing (part of the trap) and it is therefore assumed that there is no lateral communication over the faults in the Johansen formation. The shallow part of the formation is less than 40 km from Mongstad and the formation deepen towards west to a depth of 3200 m over a distance of around 60 km. The thickness of the sandy part of the formation varies from a few meters in the eastern part to around 150 m in central and western parts.

The Utsira formation is one of the large aquifers in the North Sea (areal extent 24,000 km2). In the southern part, large areas of the formation have a thickness of more than 200 m. The formation thins out to the east. More than 10 million tonne of CO2 has already been injected into the southern part of the formation during the Sleipner injection project since 1996 (see e.g., Chadwick et al., 2006, Arts et al., 2008). The behaviour of the injected CO2 has been extensively monitored and this has revealed geological information about the internal structure and transport properties of the formation. The main capillary seal, the Nordland formation, is expected to provide safe storage, and all monitoring results to date support this assumption.

Section snippets

Compressibility and pressure increase

Assuming that the formation targeted for CO2 injection is a closed entity with a given initial pore volume, the injection of CO2 will result in a compression of the existing formation fluids and an increase in pore volume due to rock compressibility. The resulting average increase in pressure is proportional to the in situ volume of the injected phase. This can be expressed asΔp=VinjVporct,where Vinj is the in situ volume of injected phase, Vpor is the total pore volume and ct is the total

Simulation models

The formation water viscosity, CO2 solubility and formation volume factor used in the simulations are based on Numbere et al. (1977), Diamond and Akinfiev, 2003a, Diamond and Akinfiev, 2003b and Enick and Klara (1990). The injection gas is assumed to be pure CO2 with properties taken from correlations prepared by Span and Wagner (1996) and Vukalovich and Altunin (1969).

Johansen formation

During the injection period, the flow of gas is mainly dominated by the radial spread around the injection points, displacing formation water and only to a small extent dissolving in the contacted water. Fig. 5 shows the distribution of the injected CO2 after 50 years for the case with two injection wells and a water production well at the boundary of the maximum CO2 extent. The location of the production well in the case where it is located at the southern boundary of the model is also shown.

Discussion and conclusions

In both simulation cases local pressure increase under the seal is close to the estimated safe limit if no water production is employed. Increasing the number of injection wells will reduce the maximum pressure increase. Controlling formation pressure by allowing water to be produced from passive water production wells will further reduce the pressure build-up in particular in the near-well region but also in the formation as a whole.

Passive water production is considered in this study. While

Acknowledgements

This publication has been produced with support from the BIGCCS Centre, performed under the Norwegian research program Centres for Environment-friendly Energy Research (FME). The authors acknowledge the following partners for their contributions: Aker Solutions, ConocoPhilips, Det Norske Veritas AS, Gassco AS, Hydro Aluminium AS, Shell Technology AS, Statkraft Development AS, Statoil Petroleum AS, TOTAL E&P Norge AS, and the Research Council of Norway (193816/S60).

References (27)

  • Bergmo, P.E., Lindeberg, E., 2007. Near coastal sites for storing CO2 from Mongstad and Kårstø. SINTEF Petroleum...
  • Bradley, H.B., 1987. Petroleum Engineering Handbook, pp....
  • A. Chadwick et al.

    Geophysical monitoring of the CO2 plume at Sleipner, North Sea: an outline review

  • Cited by (122)

    View all citing articles on Scopus
    View full text