5.2 Geomechanical site characterisation

Sufficiently representative and detailed characterisation of potential storage sites is essential for accurate simulation of their long-term storage performance. Geomechanical characterisation of the host and caprocks of the target reservoir and the assessment of the long-term behaviour of the overburden in a CO2 storage scenario require the determination of mechanical (elastic) properties of these rocks. The first step for geomechanical assessment involves the development of a static 3D geologic model of the storage site (e.g. using Petrel, EarthVision, etc.), which specifies stratigraphic contacts, structures, faults, well locations, and other salient features that have been identified from baseline well logs, seismic surveys, surface maps, etc. Within this geologic model, the distinct strata are populated with representative hydrological, geochemical, and geomechanical attribute data, which are typically obtained at sparsely-distributed locations through geophysical surveys, core/fluid sampling programs, and associated analytical studies, then extrapolated between imaging/sampling locations using geostatistical methods. Hydrological data would include temperature, pressure, porosity, permeability, and ambient flow gradients. Geochemical data may include detailed mineralogy and fluid-phase compositions/saturations, while geomechanical data may include in situ stress fields and fracture densities, apertures, and orientations (Johnson, 2009). Aarnes et al., 2010 gives a comprehensive list for the geomechanical data requirements, such as rock compressibility, Young's modulus, Poisson's ratio, compressive strength, tensile strength, in situ vertical total stress, in situ major horizontal and minor horizontal total stress, formation fracturing pressure, fault valving pressure, fault reactivation pressure, etc.

For geological storage of CO2, an important element of the model is whether conductive features exist within the caprock. Therefore, particular attention should be given to collecting data for the primary caprock and describing its properties relevant to storage containment. For storage sites that have a caprock that has contained hydrocarbons over geologic time scales, the task is focused on the characterisation of the geomechanical properties of the caprock and any pre-existing fault planes through the caprock. These can be used to estimate threshold reservoir pressures for creating new fractures through the caprock or activating existing fractures.

Caprocks consist typically of sediments from distal depositional environments, which are characterised by relatively uniform conditions over large areas. Caprock lithology, fluid-flow and geomechanical properties are therefore likely to vary much less than those of the reservoir rocks. Consequently, extrapolation of lithology-related caprock properties from a small number of wells over a large potential footprint area can be better constrained than extrapolation of reservoir properties. However, relevant caprock properties due to deformation (faults, joints) cannot easily be extrapolated but require detailed local assessment covering the whole footprint area. The regional seismic stratigraphy of the caprock should be discernible from 2D seismic data, as would major faults that cut it. Smaller structural features, for example 'polygonal' type minor faults that characterise some shale sequences, generally require good quality 3D seismic data for their proper identification. Very small structures, fractures and joints are usually below the limit of seismic detection resolution. Assessment of the presence of microfractures in the subsurface is challenging, because mechanical deformation and depressurisation during coring may induce microfractures in core samples that are difficult to distinguish from those that formed in situ. Consequently, careful coring and preservation of cores is a prerequisite for successful microfracture assessment (Chadwick et al., 2008).

The presence of fractures and their azimuth can be determined from core examination. Vertical in-situ stress can be obtained from density logs. Mechanical properties such as elastic properties for different formations can be obtained using dipole sonic logs as well.

Among the tests applied on core samples to measure rock geomechanical properties, uniaxial compression is the most widely performed method, where stress is applied only in one direction. This test is used to determine uniaxial or unconfined strength, Young's modulus and Poisson's ratio. On the other hand, stress is applied in all three directions in a triaxial compression test. From a triaxial compression test, the complete stress-strain curve may be obtained. A complete stress-strain curve records the core response to loading up to rock failure and beyond (post-failure). It provides information on the strength (maximum axial stress reached), yield stress point (marking the departure from linear elastic behaviour), and residual (post-failure) strength of the rock sample under a given confining pressure.

During geomechanical simulations, the initial stress state and stress/displacement boundary conditions for the model domain need to be defined.