Key Reports:

State-of-the-Art of Monitoring Methods to evaluate Storage Site Performance

2.4 Faults

Faults represent an important geological feature significantly influencing the CO2Carbon dioxide storage complex. Their role with respect to the storage reservoir can be twofold:

Permeable faults serve as preferred migration routes for both the injected CO2Carbon dioxide and the original reservoir fluids. In case they intersect the impermeable caprock, they may become migration pathways leading to leakage of CO2Carbon dioxide out of the reservoir.
Impermeable faults act as barriers for migration of CO2Carbon dioxide and reservoir fluids, often causing reservoir compartmentalisation. Consequences of this situation might be, e.g. reduced storage capacity and/or injectivity of the reservoir, rapid pressure increase around the injection well.

Determination of exact location and geometry of faults in the storage complex and investigation of their character belong to the most important goals of site characterisation. Usually, the first information on fault presence and geometry as well as their first-order characterisation are obtained from a 3D seismic survey. To assess the large-scale heterogeneity of the storage container, including its vertical and lateral compartmentalisation and the sealing nature of fractures and faults, pressure measurements may be employed, e.g. during well-tests or by Repeat Formation Tester (RFT) or Modular Formation Dynamics Tester (MDT) logging (Arts et al., 2009).

Mapping of faults from seismic data and building a fault model is an important activity when creating the three-dimensional static geological earth model of the storage complex. This commonly involves simplification or generalisation of fault planes. This initial fault model is usually refined by well correlation when sub-seismic faults may become apparent, especially in cases when stratigraphic sections are missing (normal faults) or doubled (reverse faults).

It is important to assess the sealing capacity of faults as this may lead to compartmentalisation of the storage reservoir and/or leakage through the caprock. compartmentalisation is a key input in the earth model, as it will lead to rapid pressure increase and demands for a relatively high number of injection wells (Arts et al., 2009). At the same time, knowledge of fault properties is also critical for geomechanical modelling of the reservoir.

Fault properties are also important for geochemical characterisation of the site and related geochemical modelling. The nature of fault-filling minerals must be assessed, since the sealing potential of faults cutting the reservoir or overburden may be reduced as a result of CO2Carbon dioxide induced dissolution. To avoid the risk of decreasing sealing potential in response to dissolution of carbonates, the carbonate content of sealing formations should be accurately studied and incorporated in the earth model, in particular if these formations are fractured or faulted (Arts et al., 2009).

Monitoring of faults

Monitoring of faults can be divided into two main issues:

monitoring of changes in the fault system in and around the reservoir, which especially comprises changes in the integrity and sealing properties of the faults;
monitoring of potential leakages of CO2Carbon dioxide from the storage reservoir via faults as preferential migration pathways.

The integrity of faults is part of the storage reservoir integrity as a whole, and the monitoring methods for these purposes are well established. Usually, the first indicator of a change in the reservoir behaviour is a sudden change in reservoir pressure.

Such a change needs to be detected as soon as possible since it normally indicates an unpredicted, and often undesired, event in the reservoir, which may (in the worst case) correspond to a significant irregularity according to the EU CCS Directive or even to a CO2Carbon dioxide leakage from the reservoir. Such a pressure change can have various causes, a change in fault integrity being one of them. Alternatively, the change in fault properties can be a slow process with slower changes in reservoir pressure. In both cases, however, reliable continuous reservoir pressure measurement is the main condition for timely detection of the change.

Permanent measurements of reservoir pressure are part of commonly used, permanent downhole monitoring technologies that have been used in the oil and gas industry for several decades. Measurement devices and downhole installation services are commercially available. The gauges are usually installed in the well casing to be able to read the formation pressure directly (see Fig. 2-14). The gauges are normally installed in monitoring wells penetrating the storage formation but they can also be placed in adjacent overburden layers, preferably on places where presence of faults has been detected or assumed. An example of a pressure monitoring record is shown in Fig. 2-15.

Fig. 2-14: Graphical illustration of a permanent downhole pressure gauge installation

Fig. 2-14: Graphical illustration of a permanent downhole pressure gauge installation. The gauge is cemented behind casing and operates in direct hydraulic communication with the formation (Alpak et al., 2004).

Fig. 2-15

Fig. 2-15: Example of reservoir pressure monitoring at the Cranfield CO2Carbon dioxide-EOR site in Mississippi, USA. The plot shows evolution of pressure within injection zone (green) and overlying monitoring zone (blue). Pressure in the injection zone has significantly increased since injection start, while monitoring zone pressure has stayed constant, indicating no communication between the two zones. Annotations are for various events during field injection. Data gaps are a result of data communication issues between downhole gauges and surface recording devices (Meckel et al., 2008).

Another technique able to detect changes in fault integrity is passive seismic monitoring. Re-activation of existing faults or formation of new faults and fractures caused by increased reservoir pressure produce seismic activity that may be monitored in the form of microseismic or even seismic events by passive seismic monitoring (see example in Fig. 2-16). For more details regarding this technique see Section 2.3.

Changes in fault extent, shape and properties can also be derived from time-lapse measurements of various kinds that are part of the site monitoring plan. This is especially valid for 3D seismics, well-logging in monitoring wells or surface deformation monitoring. The results of time-lapse measurements may lead to updates in both reservoir and geomechanical models and simulations, including the characteristics and role of the fault system (see also Sections 2.1 and 2.2).

Methods for monitoring of potential leakages of CO2Carbon dioxide from the reservoir via permeable faults are more or less identical with the methods suitable for monitoring of overlying and adjacent aquifers (Section 2.6), freshwater aquifers (Section 2.7) and near surface eco-compartments (Section 2.8). The known presence of faults leads to a special areal or spatial focus of monitoring layouts that can be concentrated to places/areas above or close to existing faults to allow early detection of a potential leakage.

These monitoring targets correspond to the segments of the subsurface where the CO2Carbon dioxide migrating from the storage reservoir can accumulate. The NETL Best Practices manual (2009) mentions two techniques particularly suitable for monitoring for leakage into overlying formations through faults or fractures:

pressure monitoring in the overlying formation (see Fig. 2-14 for example), and
monitoring for tracers (e.g. PFCs).

Fig. 2-16: Interpretation of microseismic events

Fig. 2-16: Interpretation of microseismic events recorded during water injection operations at the Groß Schönebeck geothermal research field, Germany. Yellow boxes show interpreted locations of low-magnitude microseismic events, probably occurring along an existing fault plane and indicating thus reactivation of an existing fault. The additional fluid pressure was between 20 and 24.5 MPa (Moeck et al., 2009). Similar events may potentially happen at CO2Carbon dioxide storage sites, if the reservoir pressure exceeds safety pressure limits.

Fig. 2-17: Results of seismic 4D modelling of the

Fig. 2-17: Results of seismic 4D modelling of the "leak" scenario at the Otway basin Pilot Project in Australia. In the modelled case, CO2Carbon dioxide migrates from the Naylor reservoir along the Naylor fault into the overlying Paaratte formation. Distribution of CO2Carbon dioxide for the model was extracted from reservoir simulation for various leakage quantities. The leaked CO2Carbon dioxide shows up as diffraction, in this case submerged into the background noise proportional in magnitude and frequency content to the actual one observed during field experiments (Urosevic et al., 2011).

Time-lapse measurements represent another group of techniques potentially capable of detecting leaking faults. Urosevic et al., 2011 modelled 4D seismic response of CO2Carbon dioxide leaking along a fault from the Naylor reservoir within the Otway basin Pilot Project in Australia. The authors demonstrated that small quantities of CO2Carbon dioxide are likely to produce very strong changes in the elastic properties of the host rock that would be readily detected by time lapse seismic monitoring (see Fig. 2-17).