|
Articles |
1 Shell UK Limited, Altens Farm Road, Nigg, Aberdeen AB12 3FY, UK (e-mail: steve.jolley{at}shell.com)
2 BP Exploration, Burnside Drive, Farburn Industrial Estate, Dyce, Aberdeen AB21 7PB, UK
3 Fault Analysis Group, School of Geological Sciences, University College Dublin, Belfield, Dublin 4, Ireland
4 Rock Deformation Research Ltd, School of Earth Sciences, University of Leeds, Leeds LS2 9JT, UK
 |
Abstract |
|---|
 Top Abstract Structural complexityFault geometryDetection and prediction of...Fault compartmentalizationFractured reservoirsConcluding remarksReferences |
|---|
|
This introductory paper aims to provide context to this collected work (papers cited in bold are those presented in this book) and an entry point into the subject. We have selected a series of sub-themes for this overview, which include: (i) structural complexity and fault geometry; (ii) the detection and prediction of faults and fractures; (iii) the compartmentalizing effects of fault systems and complex siliciclastic reservoirs; and (iv) the critical controls that affect fractured reservoirs. A final section gives a brief comment on future directions and priority areas that emerge from the collected papers.
|
Structural complexity |
|---|
|
TopAbstractStructural complexityFault geometryDetection and prediction of...Fault compartmentalizationFractured reservoirsConcluding remarksReferences |
|---|
This volume presents a series of papers on the full range of these technical issues. The complete workflow of the reservoir structural geologist is well represented, with papers extending from the detection, mapping and prediction of faults through to fault property modelling and flow modelling of reservoir production. The running order of papers generally tracks this workflow, with the downstream side of it being essentially subdivided into faulted siliciclastic reservoirs and fractured reservoirs. This basic distinction is important, since it focuses the emphasis of studies which address the impacts of structural complexity on reservoir fluid flow.
In many siliciclastic reservoirs, the reservoir units which host the hydrocarbons have higher porosities and permeabilities than the faults that transect them. In these circumstances, the faults are detrimental to flow, acting as baffles or barriers within a generally more permeable host rock sedimentary sequence (e.g. Knipe 1993; Downey 1994; Knipe et al. 1997, Yielding et al. 1997; Jones et al. 1998; Manzocchi et al. 1999, 2002; Hesthammer & Fossen 2000; Fisher et al. 2001; Brown 2003; James et al. 2004). Issues such as the primary connectivity of reservoir units will also impact the behaviour of the flow system. In fractured reservoirs, the host rocks (which include limestone, chalks or granite/basement as well as siliciclastic rocks) generally have lower permeabilities than the faults and fractures that transect them (Reiss 1980; Plumb 1994; Nelson 2001; Lonergan et al. 2007). Fluid flow in such reservoirs incorporates the combined effects of pervasive fracture systems, including joints, combined with faults. In these circumstances, the faults will often represent the main flow pathways that tap into the host rock volume which provides the hydrocarbon storage capacity. In a low porosity rock such as granite, storage is primarily in the fracture system but the storage and flow domains may be separate (e.g. with most storage in joints but most flow in conductive faults). In an impermeable but porous rock like some chalks, storage is primarily in the matrix (host rock) but flow is in the fractures. Some siliciclastic fractured reservoirs have a similar split between the flow and storage domains, but in others the matrix permeability is high enough to provide a significant flow contribution, leading to particularly complex behaviour during hydrocarbon recovery. Some fractured limestone reservoirs similarly have zones of high matrix porosity, or of leaching (e.g. palaeokarsts) or secondary porosity (e.g. due to dolomitization).
This duality of faults and fractures as flow conduits or barriers is a fundamental property and provides the primary distinction between the two reservoir types. Faults within higher porosity-permeability reservoir units can represent baffles/barriers whereas faults in tighter porosity, lower permeability reservoir units, can act as conduits to fluid flow. This duality of behaviour is not drawn on geographical grounds, but can in fact occur on the same structure intersecting different rock sequences, or form on the same structure at different times, when the host rock properties and deformation conditions have changed over geological time. The convenience of defining end-member faulted reservoirs, nevertheless, provides a useful conceptual framework, even if individual reservoirs or faults may sometimes incorporate both types of flow behaviour. For example, faults within fractured reservoirs may act as both seals and conduits over different parts of the fault surface or even at different times during the production history of the field.
|
Fault geometry |
|---|
|
TopAbstractStructural complexityFault geometryDetection and prediction of...Fault compartmentalizationFractured reservoirsConcluding remarksReferences |
|---|
Tertois & Mallett show that newly developed methods of tetrahedral volume modelling are capable of modelling such complexly faulted reservoirs, and Matthäi et al. describe a hybrid meshing approach which, combined with innovative methods for the discretization of governing equations, could provide a comprehensive basis for future modelling efforts. Nevertheless, for the foreseeable future, most simulation of structurally complex reservoirs will take place in ‘conventional’ simulators operating on more-or-less regular grids of six-sided cuboid-shaped cells. More sophisticated approaches can and should be used to ‘ground-truth’ such simple models so that we understand the consequences of simplification and where it is or is not acceptable to do so.
A recurrent theme in the reservoir modelling of structurally complex reservoirs is that the technical limitations of reservoir modelling packages and the computing hardware which runs them will always be a constraining factor, and that their improvement will always lag behind our technical demands. Interpretation, mapping and visualization tools, by contrast, are now very refined, as are those permitting structural modelling of various types (e.g. displacement analysis, restoration, cross-fault juxtaposition analysis). These tools arise not only from the practical demands of reservoir studies but also from developments arising from research into the geometry and kinematics of faults. Indeed current models for many fundamental aspects of faulting either derive from, or were significantly advanced by, the analysis of seismic data at reservoir scales, including: (i) fault growth models (e.g. Walsh et al. 2002); (ii) polygonal faulting (e.g. Cartwright 1994; Watterson et al. 2000); (iii) salt-related deformation (e.g. Vendeville & Jackson 1992a, b; Jackson 1995); (iv) relays and segment linkage (e.g. Childs et al. 1995); and (v) fault populations (e.g. Yielding et al. 1996; Cowie et al. 1996). There are, of course, many outstanding technical issues relating to the geometry and kinematics of faults, some of which are considered in this volume.
Some reservoirs are characterized by very complex fault geometries with different modes of faulting developed at different times and with varying degrees of reactivation. These types of reservoir present a major challenge because although quantitative constraints on normal faults are relatively good, the characteristics of reverse fault systems and strike-slip fault systems, in particular, are less well understood and therefore much less predictable. Similarly, the nature and controls on the reactivation of earlier faults is not well understood, not just on geometrical grounds, but also in terms of flow. Nevertheless, Domínguez provides a comprehensive description of structural complexity arising from the interaction of two different fault trends and later fault reactivation in the Penguins field cluster, North Sea. Although this study shows how structurally complex reservoirs can arise from a relatively simple configuration of deformations, careful analysis is capable of unravelling the structural evolution of the fault arrays. Barr shows that contractional inversion of earlier rift-related normal faults in Southern North Sea gas fields has implications beyond the purely geometrical, with the breaching of earlier seals and creation of conductive fracture networks sometimes having a profound effect on reservoir flow.
Many of the studies in this volume have been conducted on reservoirs which include what might be considered relatively conventional tectonic normal fault systems and existing constraints on their geometry, at least in a generic sense, are generally good. High quality reservoir modelling demands the accurate characterization of fault geometries as a prelude to fault seal prediction (e.g. Dee et al.), juxtaposition analysis (e.g. Myers et al.) and fault property characterization and modelling (Fisher & Jolley; Zijlstra et al.), and therefore benefits from existing geometrical constraints. Similarly our current knowledge of the geometry and growth of gravity-slide normal fault systems, related either to the instability of delta slopes or salt, is now quite refined. The analysis and restoration of related structures of both tectonic and gravity-driven fault systems from 3D seismic data has provided excellent constraints on their kinematics (e.g. Jackson 1995). Physical modelling has made a significant contribution to our knowledge of fault array development (e.g. McClay et al. 2002) and the kinematics of gravity-driven fault systems, in particular (e.g. Vendeville & Jackson 1992a,b). The work of Krézsek et al. shows how recent technological innovations in imaging and quantifying deformation are capable of defining refined models for the kinematic evolution of margins characterized by salt-related thin-skinned tectonics. We anticipate that similar types of physical modelling studies will provide excellent constraints on the origin, geometry and growth of these types of fault system. Footwall collapse-related landslides are a type of gravity slide system which is less well understood, despite the fact that it is now well established that they dominate the structure of many reservoirs, including some in the North Sea. Although these reservoirs present challenges that are very different from other types of faulted reservoirs, they have not received a great deal of attention. Welbon et al. provide a comprehensive consideration of landslide structures and outline the challenges and opportunities provided by reservoirs in landslides and a new workflow for their characterization.
The foregoing discussion highlights some of the technical issues associated with fault systems that have different geometries, origins and multiple event histories. These issues are generally the subjects of the first phase of reservoir characterization, which is mainly conducted by seismic interpretation, sometimes supplemented by core analysis. The results of this type of analysis provide the essential backdrop for later phases of the structural geology and flow modelling workflow. This workflow involves a variety of technical components, the selection of which depends on the characteristics and type of reservoir concerned.
|
Detection and prediction of faults and fractures |
|---|
|
TopAbstractStructural complexityFault geometryDetection and prediction of...Fault compartmentalizationFractured reservoirsConcluding remarksReferences |
|---|
Fault and fracture prediction in the modern exploration and production industry typically begins with 2D or 3D seismic interpretation. The issues and pitfalls involved in representing the geometry of discrete, seismically mappable faults have been discussed in the previous section. In principle, every fault that can be identified on 3D seismic can be mapped in three dimensions and characterized for flow simulation purposes. In practice, of course, inherent limitations of seismic data resolution mean that only the largest faults can be mapped as discrete objects. For example, very high quality seismic may permit faults with throws down to c. 5–10 m (and lengths of hundreds of metres) to be mapped, but a decrease in seismic quality could mean that faults with throws of c. 30–40 m may not be resolvable.
In a general sense, the fault system can be subdivided into faults that are large enough to be visible and mappable from offset of reflectors, and the smaller faults that have more subtle seismic signatures that can sometimes be mapped from ‘attribute’ lineations on seismic reflectors (bedding horizons) using various amplitude variation and discontinuity detection techniques. Thus, in the absence of clear reflector offsets, the smaller faults are seen where amplitude becomes dimmed due to net destructive interference of diffracted seismic energy at the fault scarps (see Townsend et al. 1998 for discussion), and where horizon dip and dip azimuth changes sharply. Some of the latter attribute types assume that bedding is approximately planar or gently curved and that lines along which reflector dip changes rapidly are the ‘smeared out’ response to a fault too small to resolve as a discrete object. Others use wavelet correlation techniques to detect a change in seismic character, tracking along a single horizon or in time-slices or a 3D volume. In the first case the discontinuity detected is Horizon A – Fault – Horizon A; and in the second case, Horizon A – Horizon B. Most seismic workstations and many geo-modelling software packages have tools such as coherency (Bahorich & Farmer 1995), dip, edge-detection or semblance (Marfurt et al. 1998) available; indeed they are routinely used as interpretation aids in the early stages of defining seismically resolvable faults (e.g. Jones & Knipe 1996; Townsend et al. 1998). Care needs to be taken, however, to filter-out erroneous interpretations where seismic quality is sub-optimal, since some ‘semi-automated’ methods are capable of picking noise in addition to ‘real’ faults in the data (Hesthammer et al. 2001). It is a natural extension of this process to use these techniques to define subseismic lineaments.
Curvature of seismically mapped surfaces can also serve as an edge detection tool, with hanging wall and footwall cut-offs corresponding to parallel bands of negative and positive curvature respectively (e.g. Murray 1968; Lisle 1994; Stewart & Podolski 1998). It is advisable to filter the input data spatially, to separate gross structural configuration (long wavelength) from faults (short wavelength), and to enhance the signal:noise ratio and resolution at the target wavelength (Bergbauer et al. 2003). Other techniques to enhance the detection of small-scale faulting include shaded illumination, dip-azimuth and directional curvature (extracted along parallel, vertical observation planes). Some techniques are optimally sensitive to faults with particular azimuthal orientations, and care must be taken to ensure that preconceived ideas about the fault trend do not lead to a failure to detect faults with an unanticipated orientation.
Practical limitations on the time allocated for interpretation and the requirement to construct a computationally tractable model commonly lead to simplifications or omissions in the final model. Jolley et al. (2007) and Fisher & Jolley have shown how careful thought about the flow implications of a particular simplification can lead to better designed models, given the same data and model size restrictions. They also show that time spent in getting the starting model right is more than repaid by the reduced reservoir engineering effort required to generate a robust, history-matched flow simulation model.
Fields with only 2D seismic or outcrop data require much more interpolation and have increased ambiguities around fault geometry and linkage. Rules of thumb often come into play then, perhaps informed by analogue data from fields with 3D seismic, from well-exposed outcrops or from laboratory-scale models and kinematic or geomechanical modelling studies. The balanced cross-section approach (e.g. Dahlstrom 1969; Gibbs 1983) aims to constrain the structural interpretation to one which is kinematically self-consistent, by retro-deforming the current geometry to a plausible initial condition and then replicating the current geometry through a plausible forward deformation path. Software implementations of this approach are available in 2D and 3D modelling packages which link fairly seamlessly with seismic interpretation and simulation grid building packages. Typically, these assume that displacement is concentrated on the mappable faults, with intervening fault blocks deforming passively as they are carried on the faults. They also commonly assume constant displacement and slip directions on each fault segment, with discontinuous changes in displacement taking place at fault linkages or branch lines. This model is, of course, a simplification and an alternative view of the fault linkage in a system, which is supported by 3D seismic data and by closely spaced 2D observations in mines and quarries, is that displacement variation along fault planes is the norm and that displacement transfer can be accommodated by deformation of the intervening rock volume. These alternative models for fault linkage, referred to as hard- and soft-linkage, may generate a very different level of fault connectivity (Fig. 2; cf. Walsh & Watterson 1991). These two end members will display very different flow behaviour, with Figure 2a being more connected than Figure 2b if the faults are conductive, but less connected if they are sealing.
|
The possibility that fault mapping could unhinge a significant number of reservoir studies may reflect the inherent pressures on geologists and geophysicists, to create the definitive ‘top reservoir map’ rapidly, when a more measured and discriminating 3D mapping approach would represent the best means of defining the basic geometry of faulted reservoirs. Failure to recognize the importance of basic fault mapping in 3D can introduce spurious connections or offsets of stratigraphic models. Even where the reservoir geometry is mapped accurately it is also possible that it is not represented accurately in the 3D simulation grid. However, techniques are available in 3D modelling packages to reproduce important fault-related geometrical features faithfully, such as the 3D variations in fault displacement, the geometries of fault intersections and the presence of fault discontinuities (e.g. relays). If these are not applied consistently, they may introduce unavoidable computational penalties to the model and may compromise later fault property modelling (e.g. some geometrical solutions to so-called ‘Y-fault’ geometries involve severe discretization and aliasing of faults). Hoffman & Neave discuss the advantages and limitations of some 3D fault modelling approaches in common use today.
As well as being used to define discrete faults, seismic data, particularly 3D seismic, can be used to detect smaller fault and fracture distributions. Simple seismic attributes responsive to surface roughness at a scale too small to be resolved individually include coherency and its relatives (discrete faults give rise to linear features, whereas background faulting or fracturing generates a broad zone of low coherency) and reflection amplitude (small-scale faulting produces waveform interference and scattering of the seismic energy which dims the amplitude). By their nature these attributes are insensitive to whether the fractures are conductive or sealing. More sophisticated attributes are available with multicomponent and multidirectional seismic acquisition, which allows measurements of seismic anisotropy (Verwest 1994; Bouska & Johnston 2005). A directionally anisotropic seismic response is strongly suggestive of open fractures, because the seismic response of a closed fracture or granulation seam is only subtly different from that of matrix, but that of an open fracture is very different, especially if it is filled by a compressible fluid such as oil or gas.
The simplest measure is velocity anisotropy, where seismic rays travelling across an aligned fracture set are slower than those travelling parallel to the fractures (Crampin 1981; Hudson 1981; MacBeth 1995). Multidirectional or orthogonal fracture sets can give rise to an isotropic seismic response, but it may still be possible to infer fracture presence if the velocity is anomalously slow in all azimuths. Obviously that requires a meaningful definition of ‘anomalously low’, and may not be possible where there are large matrix velocity variations due to fluid or lithology effects. Fracture predictions based on seismic anisotropy and coherency-like attributes have been presented by, for example Bloch et al. (2003) and Barr et al. (2007). Calibration against core or image log data is advisable (e.g. Smith & McGarrity 2001) and proof-of-concept may be necessary before committing to a 4C OBC survey.
More sophisticated approaches involve shear wave splitting (e.g. Winsterstein 1989; Owen et al. 1998; Maultzsch et al. 2003) and seismic amplitude variation with offset and azimuth (AVOA; Lynn & Thomsen 1990; Hall & Kendall 2003). Kendall et al. show that the magnitude of the seismic response in the Clair field, west of Shetland, depends not only on the fracture anisotropy but also on the matrix anisotropy introduced by bedding and mineral grain alignment. The seismic rays at large source:receiver offset travel at an angle to bedding rather than perpendicular, and therefore ‘see’ a combination of fracture and matrix anisotropy. Kendall et al. therefore devoted considerable effort to matrix characterization, as a necessary precursor to seismic fracture prediction. They find an AVOA response between thin-bedded and massive sandstone units which is different in fractured and unfractured cases, even with the same fracture intensity in each unit, and thus they can distinguish between cases where the upper or lower unit is the more fractured.
The two end-member geometries in Figure 2 have very different strain distributions between and around the faults. This strain distribution is important because it provides a potential entry-point to the geomechanical prediction of subseismic faults and fractures, which can enhance or retard flow depending on whether they are more or less conductive than their host rock. Most faults are surrounded by a ‘damage zone’ tens of metres wide, comprising subsidiary faults and fractures (Antonellini & Aydin 1994; McGrath & Davison 1995; Caine et al. 1996; Knipe et al. 1997; Foxford et al. 1998; Beach et al. 1999; Hesthammer et al. 2000; Billi et al. 2003; Berg & Skar 2005). In this context damage refers to a change in the effective bulk properties of the rock caused by a swarm of small-scale structures and may cause flow enhancement or retardation (a drilling or production engineer will often restrict the term damage to flow retardation). In addition, the volumes between the faults have typically suffered volumetric or shear strain, perhaps related to bending during accommodation to the fault surface (e.g. Nicol et al. 2002; see also papers in McClay 2004) or to satisfy compatibility contraints as displacement dies out in a soft-linked fault array (e.g. Fig. 2b, Barnett et al. 1987; see also Hedland 1997; Soliva & Benedicto 2005). There may also be a systematic distribution of subseismic faults related to the large-scale structures. These can have a simple relationship to the major faults (e.g. antithetic in the hanging wall). If their distribution is unpredictable or poorly understood they can be modelled stochastically by assuming or observing a size, frequency relationship such as a fractal or power-law distribution and extrapolating downscale (Childs et al. 1990; Gauthier & Lake 1993; Mäkel). There are potential pitfalls in the analysis of such data (Heffer & Bevan 1990; Cowie et al. 1996; Yielding et al. 1996; Belfield 1998) and it is only meaningful within a genetically related range of structures. Subseismic faults might reasonably be extrapolated downscale from seismic faults but in circumstances where mechanical stratigraphy exercises a strong control on the fracturing process, fault or fracture systems may not have fractal geometries and the notion of extrapolation of fault size distributions could be flawed (see Nicol et al. 1996 and Soliva & Benedicto 2005 for discussion). This scenario almost certainly applies to stratabound joints even in circumstances where they have a component of shear displacement and may display some fractal characteristics (e.g. Odling et al. 1999; Barr). Finally, the mapped fault tip locations are themselves limited by seismic resolution. Where displacement:distance relationships are well defined (e.g. Walsh & Watterson 1987; Cowie & Scholz 1992) it is possible to extrapolate the fault beyond the seismically resolved location to a predicted tip-line, based on the observed rate of displacement loss where it is still seismically observable (Yielding et al. 1996; Rutten & Verschuren 2003). Similarly, the width and internal geometry of fault damage zones has been well characterized in numerous outcrop analogues (Antonellini & Aydin 1994; Hesthammer et al. 2000; Billi et al. 2003; Odling et al. 2005) and can be predicted by selecting the right analogue and control parameters such as fault displacement and host-rock lithology. Once the geometry and petrophysical properties of the small-scale features are defined (see below), their flow implications can be simulated by explicitly modelling the observed fractures or a stochastic representation of them (Heath et al. 1994; Manzocchi et al. 1998; Walsh et al. 1998; Harris et al. 1999, 2003, this volume).
Strain modelling and fracture prediction
Perhaps less widely acknowledged is the fact that two identical mapped fault networks can have different bulk strain distributions—although it has long been taught in structural geology textbooks that there are many potential paths to a particular deformed state (e.g. Hobbs et al. 1976, p. 32). If a record of the strain history during progressive deformation is preserved it may be possible to distinguish between alternative deformation paths (e.g. Ramsay 1967 p. 119–120). Lewis et al. demonstrate with a simple example how very different deformation paths resulting in very different internal strain distributions can produce identical-seeming fault and horizon geometries. Structural modelling software is available and under development which tracks the deformation history of the fault blocks (i.e. ‘kinematic restoration’) during the forward modelling step. Such models are non-unique, but they can yield valuable insights into potential subseismic deformation. In some software packages the non-uniqueness is exposed to the user, in that explicit kinematic choices have to be made about the fault-slip and inter-fault deformation mechanism; in that case it should be obvious that a matched result is a possible but not unique solution. But the more automated the process, developed perhaps in the interest of broadening the pool of potential practitioners beyond structural geology specialists, there is a greater risk of users assuming that the result must be right ‘because the model says so’.
Notwithstanding these limitations, considerable effort is currently devoted to developing techniques to predict fracture and small-scale fault distribution at a scale much less than seismic resolution and comparable to that of core or well logs. Typically these will be predictions of the ‘joint-like’ rather than ‘fault-like’ populations. The joint-like population includes shear fractures or granulation seams as well as tension fractures—essentially those features that are small and dispersed enough to form components of an effective medium at the scale of observation, rather than discrete entities. Long-wavelength curvature can be used as a strain predictor (e.g. Stewart & Podolski 1998) with the expectation that outer-arc extensional strains will be associated with open, tensile fractures. Use of this flexural beam model requires the correct choice of mechanical layering and identification of the neutral surface. Fractures would be predicted in synclines below the neutral surface as well as in anticlines above the neutral surface. Caution is advised in interpreting the output maps, as many existing approaches make potentially unacceptable simplifying assumptions (see Bergbauer & Pollard 2003 for review). Combinations of curvature attributes can be used to define a ‘shape curvature’ (Bergbauer et al. 2003), which classifies a mapped surface (at a particular wavelength) into anticlines, synclines, domes, basins etc. Shape curvature may be a predictor of fracture style rather than fracture intensity (e.g. orthogonal v. conjugate v. unidirectional, or shear v. tensile). Bergbauer describes an outcrop example of a fold where curvature shape and magnitude were poor predictors of fracture orientation and intensity but good predictors of fracture style. In this case, fractures developed within the relatively slab-like limbs are passively rotated and propagate along their axes, whereas additional strains at the curved fold hinge have reactivated fractures in shear. Indeed, Ferrill et al. describe the deformation within a monoclinal fault propagation fold, in which they find flexural shearing of well-bedded stratigraphy in the mid-limb has re-worked earlier formed fractures, such that in this case, fracture style is related to dip domain. Both papers emphasize the role of mechanical stratigraphy in governing the underlying deformation processes.
Geomechanical models, typically discretized using finite element or boundary element techniques (Crouch & Starfield 1983), go beyond simple curvature. These use elastic (e.g. Wilkins) or more complex, elastic–plastic, constitutive laws to predict the stress and strain distribution between mapped faults. The situation is at its simplest where the structure in question formed as a result of a single episode of progressive deformation. In that case it may be acceptable to approximate the driving stress or strain state by that required to deform an initial bedding configuration to the current field geometry. Predicted elastic strains are often scaled, to compensate for the fact that deformation was intermittent with some relaxation of elastic strain and accumulation of permanent strain between increments (e.g. Wilkins). More sophisticated constitutive laws such as the elastic–plastic one used by Lewis et al. attempt to model the progressive evolution of permanent strain through a deformation cycle. Refinements used include variable treatment of fault mechanics (purely frictional v. more complex formulations where some faults are effectively given finite strength) and the introduction of compaction, fluid pressure and gravity effects. An alternative approach where the deformation history is too complex to unravel or represent, is simply to model the present day stress state, treating the reservoir as a relatively homogeneous body of rock dissected by weak faults which locally perturb the far-field stress trajectory; an approach similar to that used in earthquake modelling. Stress perturbations and interactions around faults can be used to predict stress trajectories between faults (from which favoured open fracture orientations can be deduced), or to high-grade as potentially conductive those faults which are closest to a frictional failure criterion. Attempts are now being made to track strain evolution through complex deformation episodes (e.g. Dunbar & Cook 2003; Maerten & Maerten 2006), which in principle allows better control of fracture initiation and subsequent modification. The geomechanical models can be used to help define the displacement history (i.e. geomechanically based structural restoration, e.g. Maerten et al. 2006) or the displacement history can be extracted from a kinematic structural restoration and used as boundary condition inputs to a geomechanical model (e.g. Lewis et al. 2004).
The end-point of such modelling is rarely the prediction of individual fracture formation, except for those subseismic faults that are large enough to implement in a 3D reservoir simulation model. More typically, dilational strain is taken as an open fracture indicator and compactional strain as a closed fracture indicator. The magnitude and orientation of the principal stress or strain axes are used to constrain fracture orientation, type and intensity (Mäkel; Wilkins). The constraint can either be deterministic, via empirically defined lookup tables or correlations to well data; or stochastic where the geomechanical model outputs are used as constraints on a discrete fracture network which is then upscaled to effective flow properties in a reservoir simulator (e.g. Sabathier et al. 1998). An important modelling decision is whether to calibrate directly to static well and seismic observations, or to bypass those and calibrate directly to dynamic, reservoir engineering observations (e.g. Barr). An argument in favour of the latter approach is that the fracture description is a means to an end and that end is populating a reservoir simulation model. Ultimately, the fracture description has to be upscaled to effective cell properties (e.g. Bourbiaux et al. 1997), a step that introduces its own set of assumptions and uncertainties. In addition, the fractures in a vertical well may be unrepresentative even of the surrounding (say) 100 mx100 m grid cell area. A weakness of direct dynamic calibration is that it introduces an empirical step involving poorly defined or understood processes. If the next well fails to match predictions, there is little chance of understanding why, and it will likely be handled by making another poorly understood empirical adjustment. Hall & Lewis attempt to introduce some rigour into this process, by placing seismic and geomechanical attributes on a common descriptive footing, as an effective medium sampled at the target reservoir simulation scale.
|
Fault compartmentalization |
|---|
|
TopAbstractStructural complexityFault geometryDetection and prediction of...Fault compartmentalizationFractured reservoirsConcluding remarksReferences |
|---|
Natural fault zones are characterized by three-dimensionally complex juxtapositions and displaced lenses of host stratigraphy, and modified porosity-permeability variations within the suite of different fault rock products and fracture arrays that the zones usually contain (e.g. Knipe 1993, Childs et al. 1997, this volume; Knipe et al. 1997; Foxford et al. 1998; Fisher & Knipe 1998, 2001; Gibson 1998; Ingram & Urai 1999; Skerlac 1999; Sperrevik et al. 2000; Aydin & Eyal 2002; Jourde et al. 2002; Berg & Skar 2005; Eichhubl et al. 2005; Shipton et al. 2005; van der Zee & Urai 2005). However, there are significant hurdles to overcome in predicting the distribution of these lithological fragments and fault rock properties within a fault zone, from the often sparse information available to field appraisal and development teams. It is well known that clay content plays a significant role in reducing permeability in fault rocks developed within siliciclastic rocks (e.g. Fisher & Knipe 1998, 2001; Manzocchi et al. 1999; Sperrevik et al. 2002). Consequently algorithms have been developed which attempt to calculate the distribution of average clay content (and therefore the general permeability distribution) within fault zones—from the reservoir properties, clay content and shale bed distributions within the adjacent stratigraphy—at a similar scale to that of a typical simulation model fault (cf. Yielding et al. 1997; Doughty 2003). As discussed by Fisher & Jolley, at the modelling stage many of the predictive algorithms are based on the input of ‘clay contents’ derived from geophysical well log data—and care is therefore needed to account for uncertainties in petrophysical calculation of the ‘shale’ or ‘clay’ content measures (VShale, VClay), and to ensure that these results are compatible with independent measures of clay content (often taken from core samples). The three most commonly used algorithms, developed over a decade ago, use some of the basic processes which entrain clay minerals and discretely bedded shales into a fault zone (Fig. 3). Thus, Shale Gouge Ratio (SGR, Yielding et al. 1997) is based on mechanical mixture of shaley material within a fault gouge, assuming that the resulting fault rock/gouge clay content approximates to the average collective stratigraphic clay content which has been displaced past any given point on the fault. The Clay Smear Potential (CSP, Bouvier et al. 1989; Fulljames et al. 1997) predicts the length continuity of a shale bed plastically smeared into the fault from the fault throw and source shale bed thickness. The Shale Smear Factor (SSF, Lindsay et al. 1993) focuses on predicting the thickness continuity of smears caused by ductile smearing and abrasion of shales and also as a function of throw and source shale bed thickness.
|
The prevailing stress and temperature during deformation also controls the degree of grain breakage (cataclasis) and crystallization of some cementation types, and consequent permeability collapse within the fault zone (Knipe 1989; Zhang et al. 1990; Wong et al. 1997; Chester & Chester 1998; Fisher et al. 2000, 2003). Thus, despite the popular view from the usage of clay-content algorithms in fault seal analysis described above (that low clay content faults developed in sand-rich reservoirs do not seal), under the ‘right’ conditions and geohistories, sealing on a production timescale is also possible in low clay content fault rocks. Where there is a strong palaeotemperature gradient across a field or cluster of fields, this can lead to profound differences in fault compartmentalization and consequent production characteristics (e.g. Hesthammer et al. 2002). However, these low clay content cataclastic seal types can become brittle, damaged and leaky if a field is subsequently deformed under different stress and lower temperature conditions (e.g. Leveille et al. 1997). Barr gives a detailed description of the complex distribution of fault seal compartmentalization, and conductive faults and open fracture systems in the sand-rich aeolian reservoirs of the West Sole gas fields. He shows that sealing lithified cataclasites formed in fault zones during an early rifting phase at high pressures and temperatures with some influence from host sediment facies type on crystallization of certain cementation phases; and that open seal-breaching fracture systems developed during later contractional inversion and fault reactivation at lower pressures and temperatures. In general terms the sealing and open structures tend to be mutually exclusive. Although data resolution and a significant element of clustering or selectivity in the reactivation processes introduce technical difficulties, it has nevertheless been possible to use this model to detect reactivation and therefore predict the general distribution of intact and breached seals and/or open fracture networks within the fields.
Exploration projects commonly use a deterministic approach to ‘map’ fault zone properties, such as SGR, within a fault plane and thereby calculate the predicted hydrocarbon columns that can be accumulated and held by sealing faults over geological time (e.g. Yielding 2002). These methods are based on explicit modelling of reservoir and non-reservoir juxtapositions and the sealing properties of fault rocks. However, James et al. (2004) suggest that static fault seals are controlled exclusively by juxtaposition, and they describe a stochastic approach to fault seal analysis, which models variation of stratigraphic stacking across a fault to assess its hydrocarbon retention capacity. An energetic debate has since developed on the conference circuit between proponents of these two radically differing approaches. Dee et al. use the data presented by James et al. (2004), in order to compare and contrast the results that are achieved by these two radically different methods on the same dataset. They compare the stochastic analysis with a standard SGR-based approach and suggest that despite the conceptual differences between the deterministic and stochastic methods, the results are remarkably similar. This study therefore appears to provide another useful validation of a widely applied ‘rule of thumb’.
Production fault seal modelling
Under production conditions lower permeability fault zones will generally lead to the compartmentalization of pressure distributions, hydrocarbon saturations and contacts across the faults. Fault property modelling under these conditions therefore attempts to define the rate of fluid flow across the faults in order to quantify the connected petroleum volumes that can be accessed by any given well or group of wells. Numerical flow simulation models including the effects of fault properties are now routinely used to guide field development, production management and well planning decisions (e.g. Dake 2001). The reliability of the ‘prediction mode’ of a model is generally tested by comparing the match between actual historical production data and the simulated production history, the so called ‘history match’. In detail, the fluid flow between adjacent cells in a standard simulation model is expressed by the cell-cell ‘transmissibility’ (a function of the geometry and permeability of the cells). The reduced permeability of an intervening fault is accounted for by correcting the cell–cell transmissibility with a fault transmissibility multiplier, a function of fault rock thickness and permeability (Fig. 4; e.g. Knai & Knipe 1998). However, although the stratigraphic controls on field compartmentalization are routinely addressed within generally accepted geologically-rationalized tools and workflows, it has been an industry-wide experience that the flow retarding effects of the faults are treated in an ad hoc manner by the reservoir engineer, late in the workflow. It is possible to achieve a history match by using the production data to guide manual application of geologically unrealistic faults and fault properties to steer flow and pressures around the model in this way. However, this is likely to be an artificial compensation for other inadequacies and uncertainties in the model, which then become obscured by this activity (Fisher & Jolley). It follows that the more of these trial-and-error amendments there are in a model, the less likely it is that its simulated flow approximates to reality (despite the history match) and consequently the ‘prediction mode’ becomes unreliable. Such structural uncertainty can seriously impact field development planning and production management (e.g. Corrigan 1993; Lia et al. 1997). However, as described below, data, tools and methods have evolved in recent years to permit more valid, systematic incorporation of fault properties within simulation models.
|
The fault seal algorithms described above provide the building blocks for inclusion of fault properties within simulation models. For example, Manzocchi et al. (1999, 2002) provided robust methods with which it is possible to invoke these algorithms, to incorporate the flow-retarding effects of faults systematically. Their methods calculate geologically realistic fault transmissibility multipliers from the upscaled model geometry and geo-cellular properties (e.g. stratigraphic distribution of clay, porosity, permeability). Thus, fault rock clay content prediction methods (e.g. SGR/CSP/SSF and other similar algorithms) can be calculated from the reservoir geometries and properties implicit within the geo-cellular cornerpoint grid of a simulation model. The first-order sensitivity of a simulation to structural influence is caused by juxtaposition of flow units and non-flow units across the faults, since this affects the basic ‘plumbing’ within the model. Hoffman & Neave and Tertois & Mallet discuss some of the pitfalls, procedural limitations and potential solutions to the necessary compromises that are made in constructing a 3D fault model from seismic data and the subsequent simulation grid. Thus, geometric flaws introduced at that stage result in inappropriate layer juxtapositions in the cellular model, and also restrict the geologist's ability to properly represent fault zone properties. Constraints on simulation cell geometry, driven by a desire for simple flow formulae that can be solved by existing technology in a commercially useful timeframe, can introduce compromises in fault representation since it is typically the faults that contribute most geometrical complexity to the model. However the validity of flow simulation models (and therefore the reliability of their results) can be improved vastly through inclusion of geologically realistic faults, integrated with systematically calculated fault zone properties (e.g. Jolley et al. 2007; Fisher & Jolley). In this approach, SGR and/or an SGR/smear algorithm combination can be calculated and used as a proxy for fault rock clay content and integrated with fault rock permeability data to systematically assign fault transmissibility multipliers within simulation models (sensu Manzocchi et al. 1999). Myers et al. provide a case study from a North Sea fluvial reservoir in which the history matches obtained from simulation models are progressively improved by incorporating increasingly realistic fault geometries and stratigraphic architectures. This incremental approach helped to narrow the uncertainty range on fault properties applied to faults in the simulation. Jolley et al. (2007) found that provided a geologically valid model was transferred to the simulator, the best history matches were then achieved in a fraction of the usual project time by integrating fault rock property data acquired from drill cores obtained within and close to a given reservoir to calculate the multipliers. This was particularly the case where the sampled fault rocks had experienced a similar stress-temperature (burial) history to that of the study reservoir.
Modelling interaction between faults and stratigraphic complexities
Fisher & Jolley review the wide-range of uncertainties associated with the data acquisition and processing, interpretation and modelling phases of the fault property modelling workflow. Additionally, limitations within the generally available modelling technology, force a tension between efforts to capture and preserve the geological information which is critical to fluid storage and flow, and efforts to distil the geology down to more basic elements in order to satisfy a simulation model's computational memory budget. Care therefore needs to be exercised when characterizing and simplifying stratigraphic details to assign average properties to model cells, as this can disengage continuous depositional features and/or introduce erroneous connections between layers in a model (Myers et al.). For example, it is a common experience that actual flow connectivity within a reservoir is less than that implied by a geo-cellular model of the field. As Fisher & Jolley point out, these effects are frequently assumed to be caused by sealing faults, leading to erroneous ad hoc edits being applied to the structure of the model. There is an alternative, entirely logical explanation for these effects—since the averaging of thin shale beds into a net:gross value for each cell, and the stacking of these cells within the model, can overlook the compartmentalizing effects of relatively thin shale layers between sand bodies, unless these shale layers are explicitly modelled. Despite the obvious interdependence between faults and stratigraphic complexities in controlling compartmentalization, there have been few published attempts to characterize the interplay between these elements directly (e.g. Ainsworth 2006). Manzocchi et al. use a comprehensive suite of faulted and unfaulted models of sheet-like turbidite deposits, to examine the interplay between stratigraphic elements, faulting and fault zone properties using the PSSF method developed by Childs et al. Instead of using the more traditional cellular modelling methods and net:gross ratio to build their models of sand and shale distribution, Manzocchi et al. have developed a bed-scale modelling method which explicitly includes a measure of sand body connectivity, known as the amalgamation ratio. Compared to data collected from outcrops of similar turbidite deposits, this method was found to give a far more realistic stratigraphic architecture and interbed connectivity in the models. Describing their results in terms of percolation theory, they found that high net:gross sheet turbidite sequences can be very poorly amalgamated/connected. In those circumstances, the introduction of arrays of subseismic faults (< c. 5 m throw) only reduces the connectivity within models under a rare combination of circumstances and in some situations, the relative influence of faults and stratigraphic elements on flow connectivity in the models were indistinguishable.
Multi-phase flow properties of faults
Given the very small pore throats which characterize many fault rocks, water-wet faults (i.e. those having a water film coating all the grain surfaces in the fault rock and thus impinging on its available pore space) have such high water saturations close to the free water level (FWL) of a reservoir that they may have negligible relative permeability to hydrocarbons. At some distance above the FWL the buoyancy force in the hydrocarbon column may be sufficient to overcome the capillary threshold pressure of the fault rock, giving it a finite relative permeability to hydrocarbons, and thus permitting cross-fault flow of oil or gas (for discussion see Fisher et al. 2001; Fisher & Jolley). Traditionally, the multi-phase flow properties of faults have not been included during production simulation modelling, although several key publications have highlighted their importance (e.g. Manzocchi et al. 1998, 2002; Manzocchi 1999; Fisher & Knipe 2001; Rivenæs & Dart 2002; Al-Busafi et al. 2005). A recent innovation in fault handling within reservoir models is the development by Manzocchi et al. (2002) of a method for the inclusion of the two-phase flow properties of faults. Because two-phase properties, unlike single fluid phase properties, cannot be attached to the face of grid blocks in reservoir simulations, this method derives pseudo-relative permeability functions including the fault rock properties in the upstream grid block for cross-fault flow. This method incorporates the saturation and flow rate dependencies of two-phase flow and is a fairly comprehensive treatment of the problem. Zijlstra et al. present flow data from a number of faulted reservoirs suggesting that two-phase flow properties of the faults are important in controlling compartmentalization of fluid production. They support their conclusions by presenting the results of a method for fault property modelling which is easy to implement and accounts for some of the effects of fault-related single phase and multiphase flow (as described by Manzocchi et al. 1999, 2002).
Upscaling the flow effects of subseismic faults
The recognition that subseismic faults could have an impact on flow within siliciclastic reservoirs, has only recently been matched by the development of methods which provide a basis for their incorporation into flow simulation models. Typical approaches involve definition of the upscaled effective properties of subseismically faulted rock volumes, with their eventual implicit inclusion in reservoir simulations. Early work showed that for typical fault densities and geometries (including connectivities), subseismic fault arrays will generally only begin to have a significant impact on flow within reservoirs (i.e. decreasing effective permeabilities by more than c. 20%) when fault rock permeabilities are at least two orders of magnitude below those of the reservoir host rocks (e.g. Manzocchi et al. 1998; Walsh et al. 1998). For less permeable fault rocks, cross-fault flow decreases rapidly with an increase in flow tortuousity, until flow is dominated by the connectivity of sealing faults (Walsh et al. 1998).
Analysis of the impact of damage zones surrounding seismically imaged faults reveals similar features, with newly developed methods being capable of exploring the sensitivity of flow to the full range of geometric and scaling parameters associated with damage zones (see Harris et al. and references therein). Harris et al. (1999, 2005, this volume) extend these sensitivity studies into 3D, and make a strong case for the routine definition of damage zone effective properties and their implicit inclusion in reservoir flow simulations. Ma et al. show the extent to which modelling of small fault arrays is sensitive to the modelling methods used, including such issues as discretization and the preservation of fault connectivity. Other issues need to be resolved, such as the multi-scaling properties of faults, associated upscaling challenges and the incorporation and displacement of stratigraphic architectures. In that respect, the bed scale modelling of Manzocchi et al. highlights the complex flow sensitivity of faulted sedimentological sequences, a topic which we anticipate will be the subject of future research. Nevertheless, although adoption of associated modelling and upscaling approaches has, hitherto, been relatively slow (with most attention focused on larger seismically imaged faults), recent advances suggest that the incorporation of subseismic faults and damage zones should become relatively routine within the next decade.
|
Fractured reservoirs |
|---|
|
TopAbstractStructural complexityFault geometryDetection and prediction of...Fault compartmentalizationFractured reservoirsConcluding remarksReferences |
|---|
The fractured reservoir papers in this volume supplement a larger collection of papers on this topic, provided in Lonergan et al. (2007). More general reservoir engineering and geological aspects of fractured reservoirs are also covered by, for example, Aguilera (1995) and Nelson (2001).
Fractured reservoirs are traditionally classified according to the relative contributions of fracture and matrix permeability (e.g. Reiss 1980; Nelson 2001; Allan & Qing Sun 2003; Mäkel; Fig. 5). Because of their high conductivity and low pore volume, fractures typically make a large contribution to the flow domain but a small contribution to the storage domain. For a given well or reservoir, a ‘fracture index’ can be defined to represent the magnitude of the fracture v. matrix contribution. For example, the Fracture Productivity Index of Reiss (1980) ratios the well-test Kh (average permeability times the height of the tested interval) to the matrix Kh; it can be assumed that values much larger than unity reflect a significant fracture contribution. This leads to the interesting observation that ‘how fractured’ a field appears depends not just on the conductivity of its fracture network, but also on that of the matrix properties. For example, the West Sole field (Barr) behaves more like a ‘classical’ fractured reservoir than does the Clair field (Barr et al. 2007), despite having an effective fracture permeability an order of magnitude smaller. That is because it has two or three orders of magnitude less matrix permeability. Fractured reservoirs typically have very heterogeneous porosity and permeability distributions (Matthäi et al.), which result in characteristic patterns of well performance with most production coming from the best few wells (e.g. Nelson 2001; Barr). The worst wells have failed to intersect connected, conductive fractures and a large financial benefit would flow from an ability to target only the best well locations. In practice that requires a robust pre-drilling description of the effective fracture network, hence much industry and academic attention is focused on that objective.
|
The recognition of fractures in wells is also dealt with in detail by Mäkel. Core is the most definitive source of fracture information and procedures for describing, identifying and classifying them are well established (e.g. Kulander et al. 1990) but it is expensive to acquire and there are some data limitations to consider. Drilling, core recovery and sample preparation and handling can all modify natural fractures or create new fractures which must be screened out of the subsurface description. Purely drilling-induced fractures have characteristic features that make them easy to recognize (Kulander et al. 1990) but more subtle modifications to the geometry or aperture of pre-existing fractures can be harder to recognize. Open fractures are particularly vulnerable to disturbance because they weaken the rock and the most fractured part of a core may be recovered as uninterpretable rubble. Fractures are likely to have larger apertures at surface than in the subsurface, due to the reduction in effective closure stress, and if the fracture faces can be fitted together perfectly they may have had no subsurface aperture. Most fractures described in reservoirs are not simple planar breaks but have irregular faces and are partially propped or bridged by cementing minerals. Such partially cemented fractures may give the best indication of subsurface aperture and if a genetic link can be drawn between cemented and open fractures, and the cement can be shown to have grown in a single phase, vein widths can also provide a useful indicator (e.g. Mäkel). In fields produced by depletion drive, the effective stress acting to close a fracture will increase during production and the most hydraulically effective ones may be partially cemented fractures and shear (or shear-reactivated) fractures with mismatching walls (e.g. Barr).
Borehole image logging tools are available that can detect fractures with varying degrees of success (Mäkel). All image logging techniques have their limitations and ultimately benefit from calibration against overlapping core and all suffer to some degree from an inability to detect fractures subparallel to the wellbore (Mäkel). For those fractures that can be detected, a sampling correction may be made for the intersection angle between fracture and wellbore (Terzaghi 1965; Mäkel). Fractures perpendicular to the well will be over-represented relative to fractures oblique to the well and this must be corrected in order to model the 3D fracture density in the surrounding reservoir. However, well productivity may be more influenced by the number of fractures intersected (a well drilled perpendicular to fracture strike will be more productive than one drilled parallel to strike) and uncorrected data may be preferable if that is the objective of the study.
Generally, fracture description and prediction or interpolation in the petroleum industry is not carried out in isolation, but combined with integrated reservoir description and simulation modelling. The process of populating and upscaling a fracture model is conceptually similar to that for a matrix model but generally more difficult and less advanced. Techniques include simple interpolation between wells through conventional geostatistical techniques such as kriging (Olarewaju et al. 1997), neural networks (Ouenes & Hartley 2000) and methods which simultaneously incorporate static and dynamic data (Gauthier et al. 2002). Much modern effort is devoted to the construction and analysis of a discrete fracture network (DFN) model, in which a stochastically or (rarely) deterministically generated set of fractures is populated into a map or a 3D volume. Stochastic models are typically conditioned on seismic, geometric or geomechanical inputs (e.g. Bourbiaux et al. 2002; Maerten et al. 2006; Barr et al. 2007). Mäkel provides a worked example for a dataset comprising three wells. Traditionally, the fracture model has been built independently of the matrix, which can result in unwanted interactions, e.g. open fractures may be modelled in shales where observation shows they are absent. Most modelling packages now offer some ability to condition the DFN on a layer-cake or geo-cellular matrix model, enabling the modeller to control the effect of mechanical stratigraphy better (the influence of matrix lithology, typically some combination of rock-strength parameters, on fracture initiation and growth). Fracture models are typically built at multiple scales, to represent both discrete conductive faults or ‘fracture corridors’ and small, dispersed fractures or joints. Calibration can be at both the full-field scale and the scale of a well test (e.g. Rawnsley & Wei 2001).
Flow modelling and reservoir simulation
The typical situation whereby fractures dominate the flow domain, and matrix the storage domain; means that ideally, fractures and matrix should be kept independent of one another during flow simulation modelling, by use of explicit fracture and matrix cells. The complexity of fracture networks and the sheer number of simulation grid cells required mean that this is rarely done at scales larger than that of a well test, and even then complex models are difficult to represent fully (Basquet et al. 2005). The extreme aspect ratios, low pore volumes and large permeabilities of fractured cells also create computational difficulties for simulators optimized to solve problems involving more-or-less cuboidal cells having much the same pore volumes and permeabilities in each. In traditional simulators the simplest solution is to upscale the fractures to effective porosity and permeability at the matrix grid-cell size and merge them with the matrix description. That is successful only in the simplest of cases, typically involving single-phase flow in, for example, dry gas fields without an active aquifer. More complex cases are handled by the dual porosity and dual permeability formulations, in which parallel fracture and matrix descriptions are carried in two identical framework grids, with a transfer function controlling flow between the two domains (Warren & Root 1965; Kazemi et al. 1976). The dual porosity approach allows matrix to fracture fluid flow but not the reverse and is suited to cases where matrix permeability is sufficiently low to be neglected. The dual permeability approach allows flow in both directions and is suited to reservoirs with significant matrix permeability, but is much more computationally demanding. The limitations of these approaches are well known (see Matthäi et al. for a summary), particularly with respect to multiphase flow, and research is in progress to develop better framework descriptions (unstructured grids based on tetrahedral or polyhedral cells rather than cuboids, e.g. Matthäi et al.; Tertois & Mallet) and alternative solver approaches (e.g. Matthäi et al.).
Stress-sensitive reservoirs and critically stressed faults
Much fractured reservoir description takes a relatively static view of the fracture network and its flowing properties. Geomechanical models are used to predict fracture occurrence much more often than to represent production-induced changes in the fracture network. Where the matrix can be considered unchanging, the elastic response of pre-existing fractures to changing pressure and stress can be measured or modelled (e.g. Jones 1975; Bagheri & Settari 2005). Production-induced fluid pressure changes can do more than just change the aperture of existing fractures by elastic or plastic opening or closing. They can also reactivate pre-existing faults and fractures, create new fractures and deform the rock matrix. An extreme example of the latter effect is seen in the Valhall and Ekofisk fields offshore Norway (Agarwal et al. 1997; Zoback & Zinke 2002; Barkved et al. 2003; Toublanc et al. 2005), where overpressured and undercompacted chalk underwent dramatic production-induced changes in porosity and, particularly, permeability. In such stress-sensitive reservoirs it can be difficult to distinguish matrix from fracture response. Zhang et al. describe a scenario in which they simulate the geomechanical response of a faulted and fractured reservoir to hydrocarbon production and water injection. Many fractured reservoirs are geomechanically insensitive (or at least sufficiently so that it can be treated as a second-order effect) or have a sufficiently homogeneous and reversible response that they can be adequately modelled by introducing pressure-sensitive permeability modifiers to the flow simulation model. Others are geomechanically sensitive and require a coupled simulation modelling approach (e.g. Koutsabeloulis & Hope 1998; Maillot et al. 1999; Settari & Walters 1999; Bagheri & Settari 2005).
Acute sensitivity to stress or fluid pressure perturbations which are small relative to the total stress state are a characteristic of critically stressed geological systems. In a reservoir context the critically stressed elements are typically faults that are on the verge of frictional slip or failure (e.g. Zhang et al.). In a broader context the same phenomenon is seen with earthquakes, where small stress perturbations can lock or release previously unstable or stable fault segments (Scholz 1990; Harris 1998). Critically stressed faults can be considered to buffer the subsurface stress state; stress cannot be increased substantially without activating some faults and relieving the stress increase. It is not necessary that every fault is critically stressed, only that the system as a whole is locally near failure. Near-critically stressed faults are likely to have slipped in the recent past in response to tectonic and other stress perturbations; as brittle faulting typically causes dilation, such faults may act as fluid conduits (e.g. Barton et al. 1995; Sanderson & Zhang 2004). That forms the basis for some fracture modelling approaches, where the proximity to failure of each mapped fault, and of smaller features such as subseismic faults or an idealized Andersonian joint set, is used as a proxy for fracture conductivity. Micro-seismic recording of production or hydraulic-fracturing induced earthquakes (e.g. Raleigh et al. 1976; Shapiro et al. 1997, 1999; Segall & Fitzgerald 1998; Rutledge et al. 1998, 2004; Maxwell et al. 2006) suggests that some reservoirs are at least close to being critically stressed, although interpretation may be complicated by poro-elastic effects and/or matrix compaction (e.g. Zoback & Zinke 2002). Where there is observational evidence for fault reactivation, the original proximity of the fault to failure can be inferred if the magnitude of the pore pressure perturbation is known (e.g. Raleigh et al. 1976). In other cases independently determined stress and fluid pressure conditions may lie close to the frictional failure envelope for favourably oriented reservoir faults. A critical state can be inferred from the stress-buffer argument, although care must be taken to avoid circularity where some of the input parameters were estimated by assuming a critically stressed state. The extent to which reservoirs in general or only certain faults are critically stressed is debated, although it is strongly indicated in cases where in-situ stress or fracture orientations change across them (e.g. Finkbeiner et al. 1997). Heffer et al. (1995) and Heffer (2002) have documented directionality in flow and pressure transmission during water injection in nominally unfractured reservoirs. The direction is consistent with the predicted strike of those faults which were closest to failure, implying that they were already hydraulically conductive or became so as a result of the fluid pressure increase caused by injection. The inclusion or exclusion of such effects in field simulations can have a significant impact on waterflood sweep efficiency; an unrecognized flow directionality will probably mean that most of the injectors are in the wrong place. Critically or near-critically stressed faults can also pose a drilling hazard. Mud losses may occur if the fault is reactivated by drilling mud pressure, which is typically hundreds or thousands of psi higher than formation pressure; and a fault that has previously been reactivated by injection may be associated with a swarm of open fractures which takes losses even if the drilling mud pressure is too low to cause renewed activation. Conversely, such a fault may be responsible for a formation fluid inflow or ‘kick’ if it communicates directly with an active water injector.
A bridge between the categorization of structurally complex reservoirs into faulted and fractured types is provided by Main et al. who document the geomechanical response of a conventional faulted reservoir to water injection. Fault behaviour (as barriers or pathways for enhanced flow) changed during field production but in a complex manner related to stress release or transfer along and between faults. It is likely that most of the permeability increases were due to small-scale fracturing, which locally enhanced fluid flow without necessarily forming a widely connected fracture network. In a sense they describe a fractured reservoir, but perhaps one that was not a fractured reservoir prior to water injection and which might never have displayed such behaviour if produced by primary depletion only.
|
Concluding remarks |
|---|
|
TopAbstractStructural complexityFault geometryDetection and prediction of...Fault compartmentalizationFractured reservoirsConcluding remarksReferences |
|---|
|
References |
|---|
|
TopAbstractStructural complexityFault geometryDetection and prediction of...Fault compartmentalizationFractured reservoirsConcluding remarksReferences |
|---|
Full reference details for the papers presented in this book are provided in the prelim pages.
Ainsworth, R. B. 2006. Sequence stratigraphic-based analysis of reservoir connectivity: influence of sealing faults – a case study from a marginal marine depositional setting. Petroleum Geoscience, 12, 127–141.
Agarwal, B., Allen, L. R. & Farrell, H. E. 1997. Ekofisk Field reservoir characterization: mapping permeability through facies and fracture intensity. Society of Petroleum Engineers Formation Evaluation, 227–233, December 1997.
Aguilera, R. 1995. Naturally Fractured Reservoirs. Pennwell Publishing Company, Tulsa, Oklahoma.
Al–Busafi, B., Fisher, Q. J. & Harris, S. D. 2005. The importance of incorporating the multiphase flow properties of fault rocks into production simulation models. Marine and Petroleum Geology, 22, 365–374.[CrossRef][Web of Science][GeoRef]
Allan, J. & Qing Sun, S. 2003. Controls on recovery factor in fractured reservoirs: lessons learned from 100 fractured fields. Society of Petroleum Engineers Paper, 84590.
Allan, U. S. 1989. Model for hydrocarbon migration and entrapment within faulted structures. American Association of Petroleum Geologists Bulletin, 73, 803–811.[Abstract][GeoRef]
Antonellini, A. & Aydin, A. 1994. Effect of faulting on fluid flow in porous sandstones: petrophysical properties. American Association of Petroleum Geologists Bulletin, 78, 355–377.[Abstract][GeoRef]
Aydin, A. & Eyal, Y. 2002. Anatomy of a normal fault with shale smear: Implications for fault seal. American Association of Petroleum Geologists Bulletin, 86, 1367–1381.
Badley, M. E., Freeman, B. Roberts, A. M. Thatcher, J. S. Walsh, J. J. Watterson, J. & Yielding, G. 1990. Fault interpretation during seismic interpretation and reservoir evaluation. Proceedings of the 1st Archie Conference, Houston, Texas, 224–241. Association of American Petroleum Geologists The integration of geology, geophysics, petrophysics and petroleum engineering in reservoir delineation, description and management.
Bagheri, M. & Settari, A. 2005. Modelling of geomechanics in naturally fractured reservoirs. Society of Petroleum Engineers Paper, 93083.
Bahorich, M. S. & Farmer, S. L. 1995. 3–D Seismic coherency for faults and stratigraphic features; the coherence cube. The Leading Edge, 14, 1053–1058.[Abstract][CrossRef][GeoRef]
Barkved, O., Heavey, P. Kjelstadli, R. Kleppan, T. & Kristiansen, T. G. 2003. Valhall field – still on plateau after 20 years of production. Society of Petroleum Engineers Paper, 83957.
Barnett, J. A. M., Mortimer, J. Rippon, J. H. Walsh, J. J. & Watterson, J. 1987. Displacement geometry in the volume containing a single normal fault. American Association of Petroleum Geologists Bulletin, 71, 925–937.[Abstract][GeoRef]
Barr, D., Savory, K. E. Fowler, S. R. Arman, K. & McGarrity, J. P. 2007. Pre-development fracture modelling in the Clair Field, west of Shetland. In: Lonergan, L., Jolly, R. J. H., Rawnsley, K. & Sanderson, D. J. (eds) Fractured Reservoirs, Geological Society, London, Special Publications, 270, 205–225.
Barton, C. A., Zoback, M. D. & Moos, D. 1995. Fluid flow along potentially active faults in crystalline rock. Geology, 23, 683–686.
Basquet, R., Cohen, C. E. & Bourbiaux, B. 2005. Fracture flow property identification: an optimized implementation of discrete fracture network models. Society of Petroleum Engineers Paper, 93748.
Beach, A., Welbon, A. I. Brockbank, P. J. & McCallum, J. E. 1999. Reservoir damage around faults: outcrop examples from the Suez rift. Petroleum Geology, 5, 109–116.
Belfield, W. C. 1998. Incorporating spatial distribution into stochastic modelling of fractures: multifractals and Lévy-stable statistics. Journal of Structural Geology, 20, 473–486.[CrossRef][Web of Science][GeoRef]
Berg, S. S. & Skar, T. 2005. Controls on damage zone asymmetry of a normal fault zone: outcrop analyses of a segment of the Moab fault, SE Utah. Journal of Structural Geology, 27, 1803–1822.[CrossRef][Web of Science][GeoRef]
Bergbauer, S. & Pollard, D. D. 2003. How to calculate normal curvatures of sampled geological surfaces. Journal of Structural Geology, 25, 277–289.[CrossRef][Web of Science][GeoRef]
Bergbauer, S., Mukerji, T. & Hennings, P. 2003. Improving curvature analyses of deformed horizons using scale-dependent filtering techniques. American Association of Petroleum Geologists Bulletin, 87, 1255–1272.
Billi, A., Salvini, F. & Storti, F. 2003. The damage zone-fault core transition in carbonate rocks: implications for fault growth, structure and permeability. Journal of Structural Geology, 25, 1779–1794.[CrossRef][Web of Science][GeoRef]
Bloch, G., El Deeb, M. Badaam, H. Cailly, F. Lecante, G. Fonta, O. & Meunier, A. 2003. Seismic facies analysis for fracture detection: a powerful technique. Society of Petroleum Engineers Paper, 81526.
Bouska, J. & Johnston, R. 2005. The first 3D/4-C ocean bottom seismic surveys in the Caspian Sea: Acquisition design and processing strategy. The Leading Edge, 24, 910–921.
Bourbiaux, B. J., Cacas, M. -C. Sarda, S. & Sabathier, J. C. 1997. A fast and efficient methodology to convert fractured reservoir images into a dual-porosity model. Society of Petroleum Engineers Paper, 38907.
Bourbiaux, B., Basquet, R. Cacas, M.-C. Daniel, J.-M. & Sarda, S. 2002. An integrated workflow to account for multi-scale fractures in reservoir simulation models: implementation and benefits. Society of Petroleum Engineers Paper, 78489.
Bouvier, J. D., Kaars-Sijpesteijn, C. H. Kleusner, D. F. Onyejekwe, C. C. & van der Pal, R. C. 1989. Three-dimensional seismic interpretation, and fault sealing investigations, Nun River field, Nigeria. American Association of Petroleum Geologists Bulletin, 73, 1397–1414.[Abstract][GeoRef]
Brown, A. A. 2003. Capillary pressure and fault sealing. American Association of Petroleum Geologists Bulletin, 87, 381–396.
Caine, J. S., Evans, J. P. & Forster, C. B. 1996. Fault zone architecture and permeability structure. Geology, 24, 1025–1028.
Cartwright, J. A. 1994. Episodic basin-wide hydrofracturing of overpressured Early Cenozoic mudrock sequences in the North Sea Basin. Marine and Petroleum Geology, 11, 587–607.[CrossRef][Web of Science][GeoRef]
Chester, F. M. & Chester, J. S. 1998. Ultracataclasite structure and friction processes of the punchbowl fault, San Andreas System, California. Tectonophysics, 295, 199–221.[CrossRef][Web of Science][GeoRef]
Childs, C., Walsh, J. J. & Watterson, J. 1990. A method for estimation of the density of fault displacements below the limits of seismic resolution in reservoir formations. In: Buller, A. T., Berg, E., Hjelmeland, O., Kleppe, J., Torsæter, O. & Aasen, J. O. (eds) North Sea Oil and Gas Reservoirs II. Graham & Trotman, London, 193–203.
Childs, C., Watterson, J. & Walsh, J. J. 1995. Fault overlap zones within developing normal fault systems. Journal of the Geological Society, London, 152, 535–549.
Childs, C., Walsh, J. J. & Watterson, J. 1997. Complexity in fault zone structure and implications for fault seal prediction. In: Møller-Pedersen, P. & Koestler, A. G. (eds) Hydrocarbon seals - Importance for exploration and production, Norwegian Petroleum Society (NPF), Special Publication, 7, 61–72.[CrossRef]
Corrigan, A. F. 1993. Estimation of recoverable reserves: the geologists job. In: Parker, J. R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conference, 1473–1481. The Geological Society.
Coward, M. P., Daltaban, T. S. & Johnson, H. 1998. Structural geology in reservoir characterization, Geological Society, London, Special Publications, 127.
Cowie, P. A. & Scholz, C. H. 1992. Displacement–length scaling relationships for faults: data synthesis and discussion. Journal of Structural Geology, 14, 1149–1156.[CrossRef][Web of Science][GeoRef]
Cowie, P. A., Knipe, R. J. & Main, I. G. 1996. Scaling laws for fault and fracture populations - analyses and applications. Journal of Structural Geology, 18, v–xi.
Crampin, S. 1981. A review of wave motion in anisotropic and cracked elastic media. Wave Motion, 3, 343–391.[CrossRef][Web of Science][GeoRef]
Crouch, S. L. & Starfield, A. M. 1983. Boundary element methods in solid mechanics: with applications in rock mechanics and geological engineering. Allen and Unwin, Winchester, Massachusetts.
Dahlstrom, C. D. A. 1969. Balanced cross-sections. Canadian Journal of Earth Sciences, 6, 743–757.[GeoRef]
Dake, L. P. 2001. Practice of Reservoir Engineering, 36, Development in Petroleum Science, Elsevier, Amsterdam. 570 pp.
Doughty, P. T. 2003. Clay smear seals and fault sealing potential of an exhumed growth fault, Rio Grande rift, New Mexico. American Association of Petroleum Geologists Bulletin, 87, 427–444.
Downey, M. 1994. Hydrocarbon Seal Rocks. In: Magoon, L. B. & Dow, W. G. (eds) The Petroleum System - From Source to Trap, 60, American Association of Petroleum Geologists, Memoir, 159–164.
Dunbar, J. A. & Cook, R. W. 2003. Palinspastic reconstruction of structure maps: an automated finite element approach with heterogeneous strain. Journal of Structural Geology, 26, 1021–1036.[CrossRef]
Eichhubl, P., D'Onfro, P. S. Aydin, A. Waters, J. & McCarty, D. K. 2005. Structure, petrophysics, and diagenesis of shale entrained along a normal fault at Black Diamond Mines, California – Implications for fault seal. American Association of Petroleum Geologists Bulletin, 89, 1113–1137.
Finkbeiner, T., Barton, C. A. & Zoback, M. D. 1997. Relationships among in-situ stress, fractures and faults, and fluid flow: Monterey Formation, Santa Maria Basin, California. American Association of Petroleum Geologists Bulletin, 81, 1975–1999.[Abstract][GeoRef]
Fisher, Q. J. & Knipe, R. J. 1998. Fault sealing processes in siliciclastic sediments. In: Jones, G., Fisher, Q. J. & Knipe, R. J. (eds) Faulting, Fault Sealing and Fluid Flow in Hydrocarbon Reservoirs, Geological Society, London, Special Publications, 147, 117–134.
Fisher, Q. J. & Knipe, R. J. 2001. The permeability of faults within siliciclastic petroleum reservoirs of the North Sea and Norwegian Continental Shelf. Marine and Petroleum Geology, 18, 1063–1081.[CrossRef][Web of Science][GeoRef]
Fisher, Q. J., Knipe, R. J. & Worden, R. 2000. The microstructure of deformed and undeformed sandstones from the North Sea: its implications for the origin of quartz cement. In: Worden, R. H. & Morad, S. (eds) Quartz Cementation in Sandstones, International Association of Sedimentology, Special Publication, 29, 129–146.
Fisher, Q. J., Harris, S. D. McAllister, E. Knipe, R. J. & Bolton, A. J. 2001. Hydrocarbon flow across sealing faults: theoretical constraints. Marine and Petroleum Geology, 18, 251–257.[CrossRef][Web of Science][GeoRef]
Fisher, Q. J., Casey, M. Harris, S. D. & Knipe, R. J. 2003. The fluid flow properties of faults in sandstone: the importance of temperature history. Geology, 31, 965–968.
Foxford, K. A., Walsh, J. J. Watterson, J. Garden, I. R. Guscott, S. C. & Burley, S. D. 1998. Structure and content of the Moab Fault Zone, Utah, U.S.A. and its implications for fault seal prediction. In: Jones, G., Fisher, Q. J. & Knipe, R. J. (eds) Faulting, Fault Sealing and Fluid Flow in Hydrocarbon Reservoirs, Geological Society, London, Special Publications, 147, 87–103.
Fulljames, J. R., Zijerveld, L. J. J. & Franssen, R. C. M. W. 1997. Fault seal process: systematic analysis of fault seals over geological and production time scales. In: Møller-Pedersen, P. & Koestler, A. G. (eds) Hydrocarbon Seals: Importance for Exploration and Production, 7, Norwegian Petroleum Society, 51–60, Special Publication.[CrossRef]
Gauthier, B. D. M. & Lake, S. D. 1993. Probabilistic modelling of faults below the limit of seismic resolution in Pelican Field, North Sea. American Association of Petroleum Geologists Bulletin, 77, 761–777.[Abstract][GeoRef]
Gauthier, B. D. M., Garcia, M. & Daniel, J. -M. 2002. Integrated fractured reservoir characterization: a case study in a North Africa field. Society of Petroleum Engineers Paper, 79105.
Gibbs, A. D. 1983. Balanced cross section construction from seismic sections in areas of extensional tectonics. Journal of Structural Geology, 5, 153–160.[CrossRef][Web of Science][GeoRef]
Gibson, R. G. 1998. Physical character and fluid flow properties of sandstone-derived fault zones. In: Coward, M. P., Daltaban, T. S. & Johnson, H. (eds) Structural Geology in Reservoir Characterization, Geological Society, London, Special Publications, 127, 82–98.
Hall, S. A. & Kendall, J. M. 2003. Fracture characterization at Valhall: application of P-wave AVOA analysis to a 3D ocean-bottom data set. Geophysics, 68, 1150–1160.[CrossRef][Web of Science][GeoRef]
Harris, R. 1998. Introduction to special section: Stress triggers, stress shadows, and implications for seismic hazard. Journal of Geophysical Research, 103, 24347–24358.[CrossRef]
Harris, S. D., McAllister, E. Knipe, R. J. Elliott, L. & Ingham, D. B. 1999. Scaling of fluid flow associated with flow through fault damage zones and networks. In: Lippard, S. J., Næss, A. & Sinding-Larsen, R. (eds) Proceedings of the 5th Annual Conference, 711–716. International Association for Mathematical Geology.
Harris, S. D., McAllister, E. Knipe, R. J. & Odling, N. E. 2003. Predicting the three-dimensional population characteristics of fault damage zones: a study using stochastic models. Journal of Structural Geology, 25, 1281–1299.[CrossRef][Web of Science][GeoRef]
Harris, S. D., Fisher, Q. J. Karimi-Fard, M. Vaszi, A. Z. & Wu, K. 2005. Modelling the effects of faults and fractures on fluid flow in petroleum reservoirs. In: Ingham, D. B. & Pop, I. (eds) Transport Phenomena in Porous Medi, III, Elsevier, Netherlands, 441–476.
Heath, A. E., Walsh, J. J. & Watterson, J. 1994. Estimation of the effects of subseismic sealing faults in effective permeabilities in sandstone reservoirs. In: Aasen, J. O., Berg, E., Buller, A. T., Hjelmeland, O., Holt, R. M., Kleppe, J. & Torsæter, O. (eds) North Sea Oil and Gas Reservoirs – III. Kluwer Academic, Trondheim, Norway, 173–183.
Hedland, C. A. 1997. Fault-propagation, ductile strain, and displacement-distance relationships. Journal of Structural Geology, 19, 249–256.[CrossRef][Web of Science][GeoRef]
Heffer, K. J. 2002. Geomechanical influences in water injection projects: An overview. Oil & Gas Science and Technology Review, 57(5), 415–422, Institut Français du Pétrole.[CrossRef]
Heffer, K. J. & Bevan, T. 1990. Scaling relationships in natural fractures – Data, theory and applications. Society of Petroleum Engineers Paper, 20981.
Heffer, K. J., Fox, R. J. & McGill, C. A. 1995. Novel techniques show links between reservoir flow directionality, earth stress, fault structure and geomechanical changes in mature waterfloods. Society of Petroleum Engineers Paper, 30711.
Hesthammer, J. & Fossen, H. 2000. Uncertainties associated with fault sealing analysis. Petroleum Geoscience, 6, 37–45.[Web of Science][GeoRef]
Hesthammer, J., Johansen, T. E. S. & Watts, L. 2000. Spatial relationships within fault damage zones in sandstone. Marine and Petroleum Geology, 17, 873–893.[CrossRef][Web of Science][GeoRef]
Hesthammer, J., Landrø, M. & Fossen, H. 2001. Use and abuse of seismic data in reservoir characterization. Marine and Petroleum Geology, 18, 635–655.[CrossRef][Web of Science][GeoRef]
Hesthammer, J., Bjørkum, P. A. & Watts, L. 2002. The effect of temperature on sealing capacity of faults in sandstone reservoirs: examples from the Gullfaks and Gullfaks Sør fields, North Sea. American Association of Petroleum Geologists Bulletin, 86, 1733–1751.
Hobbs, B. E., Means, W. D. & Williams, P. F. 1976. An Outline of Structural Geology. Wiley, New York.
Hudson, J. A. 1981. Wave speeds and attenuation of elastic waves in materials containing cracks. Geophysical Journal of the Royal Astronomical Society, 64, 133–150.[Web of Science][GeoRef]
Ingram, G. M. & Urai, J. L. 1999. Top-seal leakage through faults and fractures: the role of mudrock properties. In: Aplin, A. C., Fleet, A. J. & Macquaker, J. H. S. (eds) Muds and Mudstones: Physical and Fluid Flow Properties, Geological Society, London, Special Publications, 158, 125–135.
Jackson, M. P. A. 1995. Retrospective salt tectonics. In: Jackson, M. P. A., Roberts, D. G. & Snelson, S. (eds) Salt Tectonics: a Global Perspective, 65, American Association of Petroleum Geologists, Memoir, 1–28.
James, W. R., Fairchild, L. H. Nakayama, G. P. Hippler, S. J. & Vrolijk, P. J. 2004. Fault-seal analysis using a stochastic multifault approach. American Association of Petroleum Geologists Bulletin, 88, 885–904.
Jolley, S. J., Dijk, H. Lamens, J. H. Fisher, Q. J. Manzocchi, T. Eikmans, H. & Huang, Y. 2007. Faulting and fault sealing in production simulation models: Brent Province, northern North Sea. Petroleum Geoscience, 13, 321–340.
Jones, F. O. 1975. A laboratory study of the effects of confining pressure on fracture flow and storage in carbonate rocks. Journal of Petroleum Technology, 4569, 21–27, Society of Petroleum Engineers Paper.
Jones, G. & Knipe, R. J. 1996. Seismic attribute maps: application to structural interpretation and fault seal analysis in the North Sea basin. First Break, 14, 449–461.[GeoRef]
Jones, G, Fisher, Q. J. & Knipe, R. J. 1998. Faulting, Fault Sealing and Fluid Flow in Hydrocarbon Reservoirs, Geological Society, London, Special Publications, 147.
Jourde, H., Flodin, E. A. Aydin, A. Durlofsky, L. J. & Wen, X. -H. 2002. Computing permeability of fault zones in eolian sandstones from outcrop measurements. American Association of Petroleum Geologists Bulletin, 86, 1187–1200.
Kazemi, H., Merril, L. S. Jr. Porterfield, K. L. & Zeman, P. R. 1976. Numerical simulation of water–oil flow in naturally fractured reservoirs. Society of Petroleum Engineers Paper, 5719.
Koutsabeloulis, N. C. & Hope, S. A. 1998. Coupled stress/fluid-thermal multi-phase reservoir simulation studies incorporating rock mechanics. Society of Petroleum Engineers Paper, 47393.
Knai, T. A. & Knipe, R. J. 1998. The impact of faults on fluid flow in the Heidrun Field. In: Jones, G., Fisher, Q. J. & Knipe, R. J. (eds) Faulting, Fault Sealing and Fluid Flow in Hydrocarbon Reservoirs, Geological Society, London, Special Publications, 147, 269–282.
Knipe, R. J. 1989. Deformation mechanisms - recognition from natural tectonites. Journal of Structural Geology, 11, 127–146.[CrossRef][Web of Science][GeoRef]
Knipe, R. J. 1993. The influence of fault zone processes and diagenesis on fluid flow. In: Horbury, A. D. & Robinson, A. D. (eds) Diagenesis and Basin Development, 36, American Association of Petroleum Geologists Studies in Geology, 135–154.
Knipe, R. J., Fisher, Q. F. Jones, G. et al. 1997. Fault seal analysis: successful methodologies, application and future directions. In: Møller-Pedersen, P. & Koestler, A. G. (eds) Hydrocarbon Seals: Importance for Exploration and Production, 7, Norwegian Petroleum Society, 15–40, Special Publication.[CrossRef]
Kulander, B. R., Dean, S. L. & Ward, B. J. 1990. Fractured Core Analysis: Interpretation, Logging and Use of Natural and Induced Fractures in Core, 8 American Association of Petroleum Geologists Methods in Exploration Series.
Leveille, G. P., Knipe, R. J. More, C. Ellis, D. Dudley, G. Jones, G. Fisher, Q. J. & Allinson, G. 1997. Compartmentalisation of Rotliegendes gas reservoirs by sealing faults, Jupiter Fields area, southern North Sea. In: Ziegler, K., Turner, P. & Daines, S. R. (eds) Petroleum Geology of the Southern North Sea; Future Potential, Geological Society, London, Special Publications, 123, 87–104.
Lewis, H., Guest, J, Hammond, L. & Hall, S. A. 2004. Kinematics to Geomechanics - an Exercise in Predicting Reservoir Deformation and Flow. American Association of Petroleum Geologists Abstracts with Programmes, Dallas, Texas. http://aapg.confex.com/aapg/da2004/techprogram/A87944.htm.
Lia, O., Omre, H. Tjelmeland, H. Holden, L. & Egeland, T. 1997. Uncertainties in reservoir production forecasts. American Association of Petroleum Geologists Bulletin, 81, 775–802.[Abstract][GeoRef]
Lindsay, N. G., Murphy, F. C. Walsh, J. J. & Watterson, J. 1993. Outcrop studies of shale smears on fault surfaces. In: Flint, S. T. & Bryant, A. D. (eds) The Geological Modelling of Hydrocarbon Reservoirs and Outcrop, 15, International Association of Sedimentology, 113–123, Special Publication.
Lisle, R. J. 1994. Detection of zones of abnormal strains in structures using Gaussian curvature analysis. American Association of Petroleum Geologists Bulletin, 78, 1811–1819.[Abstract][GeoRef]
Lonergan, L., Jolly, R. J. H., Rawnsley, K. & Sanderson, D. J. (eds) 2007. Fractured Reservoirs, Geological Society, London, Special Publications, 270.
Lynn, H. B. & Thomsen, L. A. 1990. Reflection shear-wave data collected near the principal axes of azimuthal anisotropy. Geophysics, 55, 147–156.[CrossRef][Web of Science][GeoRef]
MacBeth, C. 1995. How can anisotropy be used for reservoir characterization? First Break, 13, 31–37.[GeoRef]
Maerten, L. & Maerten, F. 2006. Chronologic modelling of faulted and fractured reservoirs using geomechanically based restoration: Technique and industry applications. American Association of Petroleum Geologists Bulletin, 90, 1201–1226.
Maerten, L., Gillespie, P. & Daniel, J -M. 2006. Three-dimensional geomechanical modelling for constraint of subseismic fault simulation. American Association of Petroleum Geologists Bulletin, 90, 1337–1358.
Maillot, B., Nielsen, S. & Main, I. 1999. Numerical simulation of seismicity due to fluid injection in a brittle poro-elastic medium. Geophysical Journal International, 139, 263–272.[CrossRef][Web of Science][GeoRef]
Manzocchi, T. 1999. Towards an improved fault representation in two-phase reservoir simulation. EAGE 61st conference and technical exhibition, Helsinki, Finland, 7–11 June 1999, 1, C-05.
Manzocchi, T., Ringrose, P. S. & Underhill, J. R. 1998. Flow through fault systems in high porosity sandstones. In: Coward, M. P., Johnson, H. & Daltaban, T. (eds) Structural Geology in Reservoir Characterization, Geological Society, London, Special Publications, 127, 65–82.
Manzocchi, T., Walsh, J. J. Nell, P. & Yielding, G. 1999. Fault transmissibility multipliers for flow simulation models. Petroleum Geoscience, 5, 53–63.
Manzocchi, T., Heath, A. E. Walsh, J. J. & Childs, C. 2002. The representation of two phase fault-rock properties in flow simulation models. Petroleum Geoscience, 8, 119–132.[Web of Science][GeoRef]
Marfurt, K. J., Kirlinz, R. L. Farmer, S. L. & Bahorich, M. S. 1998. 3-D seismic attributes using a semblance-based coherency algorithm. Geophysics, 63, 1150–1165.[CrossRef][Web of Science][GeoRef]
Maultzsch, S., Chapman, M. Liu, E. & Li, X-Y. 2003. The potential of measuring fracture sizes with frequency-dependent shear-wave splitting. First Break, 21, 45–51.[GeoRef]
Maxwell, S. C., Waltman, C. K. Warpinski, N. R. Mayerhofer, M. J. & Boroumand, N. 2006. Imaging seismic deformation induced by hydraulic fracture complexity. Society of Petroleum Engineers Paper, 102801.
McClay, K. R. (ed.) 2004. Thrust Tectonics and Hydrocarbon Systems, 82, American Association of Petroleum Geologists, Memoir.
McClay, K. R., Dooley, T. Whitehouse, P. & Mills, M. 2002. 4-D Evolution of Rift Systems: Insights from Scaled Physical Models. American Association of Petroleum Geologists Bulletin, 86, 935–960.
McGrath, A. G. & Davison, I. 1995. Damage zone geometry around fault tips. Journal of Structural Geology, 17, 1011–1024.[CrossRef][Web of Science][GeoRef]
Møller-Pedersen, P. & Koestler, A. G. (eds) 1997. Hydrocarbon Seals: Importance for Exploration and Production, 7, Norwegian Petroleum Society. Special Publication.
Murray, G. H. 1968. Quantitative fracture study - Spanish Pool, McKenzie County, North Dakota. American Association of Petroleum Geologists Bulletin, 52, 57–65.[Abstract][GeoRef]
Needham, D. T., Yielding, G. & Freeman, B. 1996. Analysis of fault geometry and displacement patterns. In: Buchanan, P. G. & Nieuwland, D. A. (eds) Modern Developments in Structural Interpretation, Validation and Modelling, Geological Society, London, Special Publications, 99, 189–199.
Nelson, R. A. 2001. Geologic Analysis of Naturally Fractured Reservoirs. Gulf Professional Publishing, Boston.
Nicol, A., Walsh, J. J. Watterson, J. & Gillespie, P. A. 1996. Fault size distributions - are they really power law? Journal of Structural Geology, 18, 191–197.[CrossRef][Web of Science][GeoRef]
Nicol, A., Gillespie, P. A. Childs, C. & Walsh, J. J. 2002. Relay zones between mesoscopic thrust faults in layered sedimentary sequences. Journal of Structural Geology, 24, 709–727.[CrossRef][Web of Science][GeoRef]
Odling, N. E., Gillespie, P. Bourgine, B. et al. 1999. Variations in fracture system geometry and their implications for fluid flow in fractured hydrocarbon reservoirs. Petroleum Geoscience, 5, 373–384.
Odling, N. E., Harris, S. D. Vaszi, A. Z. & Knipe, R. J. 2005. Properties of fault damage zones in siliclastic rocks: a modelling approach. In: Shaw, R. P. (ed.) Understanding the Micro to Macro Behaviour of Rock–Fluid Systems, Geological Society, London, Special Publications, 249, 43–59.
Olarewaju, J., Ghori, S. Fuseni, A. & Wajid, M. 1997. Stochastic Simulation of fracture density for permeability field estimation. Society of Petroleum Engineers Paper, 37962.
Ouenes, A. & Hartley, L. J. 2000. Integrated fractured reservoir modelling using both discrete and continuum approaches. Society of Petroleum Engineers Paper, 62939.
Owen, R. J., Li, X-Y. MacBeth, C. & Booth, D. C. 1998. Reservoir characterization: how can seismic anisotropy help? In: Coward, M. P., Daltaban, T. S. & Johnson, H. (eds) Structural Geology in Reservoir Characterization, Geological Society, London, Special Publications, 127, 115–119.
Plumb, R. A. 1994. Influence of composition and texture on the failure properities of clastic rocks. Society of Petroleum Engineers Paper, 28022.
Raleigh, C. B., Healy, J. H & Bredehoeft, J. D. 1976. An experiment in earthquake control at Rangely, Colorado. Science, 191, 1230–1237.
Ramsay, J. G. 1967. Folding and Fracturing of Rocks. Wiley, New York.
Rawnsley, K & Wei, L. 2001. Evaluation of a new method to build geological models of fractured reservoirs calibrated to production data. Petroleum Geoscience, 7, 23–33.[Web of Science][GeoRef]
Reiss, L. H. 1980. The Reservoir Engineering Aspects of Fractured Formations. Editions Technip Paris.
Rivenæs, J. C. & Dart, C. 2002. Reservoir compartmentalization by water-saturated faults - Is evaluation possible with todays tools? In: Koestler, A. G. & Hunsdale, R. (eds) Hydrocarbon seal Quantification, 11, Norsk Petroleumsforenig, 187–201, Special Publications.[CrossRef]
Rutledge, J. T., Phillips, W. S. & Schuessler, B. H. 1998. Reservoir characterization using oil-production-induced microseismicity, Clinton County, Kentucky. Tectonophysics, 289, 129–1532.[CrossRef][Web of Science][GeoRef]
Rutledge, J. T., Phillips, W. S. & Mayerhofer, M. J. 2004. Faulting induced by forced fluid injection and fluid flow forced by faulting: an interpretation of hydraulic-fracture microseismicity, Carthage Cotton Valley Gas Field, Texas. Bulletin of the Seismological Society of America, 94, 1817–1830.
Rutten, K. W. & Verschuren, M. A. J. 2003. Building and unfaulting fault-horizon networks. In: Nieuwland, D. A. (ed.) New Insights into Structural Interpretation and Modelling, Geological Society, London, Special Publications, 212, 39–57.
Sabathier, J. C., Bourbiaux, B. J. Cacas, M. C. & Sarda, S. 1998. A new approach of fractured reservoirs. Society of Petroleum Engineers Paper, 39825.
Sanderson, D. J. & Zhang, X. 2004. Stress-controlled localization of deformation and fluid flow in fractured rocks. In: Cosgrove, J. W. & Engelder, T. (eds) The Initiation, Propagation, and Arrest of Joints and Other Fractures, Geological Society, London, Special Publications, 231, 299–314.
Scholz, C. H. 1990. The Mechanics of Earthquakes and Faulting. Cambridge University Press, New York.
Segall, P. & Fitzgerald, S. D. 1998. A note on induced stress changes in hydrocarbon and geothermal reservoirs. Tectonophysics, 289, 117–128.[CrossRef][Web of Science][GeoRef]
Settari, A. & Walters, D. A. 1999. Advances in coupled geomechanical and reservoir modelling with applications to reservoir compaction. Society of Petroleum Engineers Paper, 51927.
Shapiro, S. A., Huenges, E. & Borm, G. 1997. Estimating the crust permeability from fluid-injected-induced seismic emission at the KTB site. Geophysical Journal International, 131, F15–F18.[CrossRef][Web of Science]
Shapiro, S. A., Audigane, P. & Royer, J.-J. 1999. Large-scale in-situ permeability tensor of rocks from induced microseismicity. Geophysical Journal International, 137, 207–213.[CrossRef][Web of Science][GeoRef]
Shipton, Z. K., Evans, J. P. & Thompson, L. B. 2005. The geometry and thickness of deformation-band fault core and its influence on sealing characteristics of deformation-band fault zones. In: Sorkhabi, R. & Tsuji, Y. (eds) Faults, Fluid Flow and Petroleum Traps, 85, American Association of Petroleum Geologists, Memoir, 181–195.
Skerlac, G. M. 1999. Evaluating top and fault seals. In: Beaumont, E. A & Foster, N. H. (eds) Handbook of Petroleum Geology: Exploring for Oil and Gas Traps, 82, American Association of Petroleum Geologists, Memoir.
Smith, R. L. & McGarrity, J. P. 2001. Cracking the fractures – seismic anisotropy in an offshore reservoir. The Leading Edge, 20, 18–26.
Soliva, R. & Benedicto, A. 2005. Geometry, scaling relations and spacing of vertically restricted normal faults. Journal of Structural Geology, 27, 317–325.[CrossRef][Web of Science][GeoRef]
Sorkhabi, R. & Tsuji, Y. (eds) 2005. Faults, Fluid Flow and Petroleum Traps, 85, American Association of Petroleum Geologists, Memoir.
Sperrevik, S., Færseth, R. B. & Gabrielsen, R. H. 2000. Experiments on clay smear formation along faults. Petroleum Geoscience, 6, 113–123.[Web of Science][GeoRef]
Sperrevik, S., Gillespie, P. A. Fisher, Q. J. Halvorsen, T. & Knipe, R. J. 2002. Empirical estimation of fault rock properties. In: Koestler, A. G. & Hunsdale, R. (eds) Hydrocarbon Seal Quantification, 11, Norwegian Petroleum Society, 109–125, Special Publication.[CrossRef]
Stewart, S. A & Podolski, R. 1998. Curvature analysis of gridded geological surfaces. In: Coward, M. P., Daltaban, T. S. & Johnson, H. (eds) Structural Geology in Reservoir Characterization, Geological Society, London, Special Publications, 127, 133–147.
Swennen, R., Roure, F. & Granath, J. W. (eds) 2004. Deformation, Fluid Flow, and Reservoir Appraisal in Foreland Fold and Thrust Belts, Hedberg Series, 1, American Association of Petroleum Geologists, 1–2.
Terzaghi, R. D. 1965. Sources of error in joint surveys. Geotechnique, 15, 287–304.[Medline]
Townsend, C., Firth, I. R. Westerman, R. Kirkevollen, L. Harde, M. & Andersen, T. 1998. Small-scale seismic fault identification and mapping. In: Jones, G., Fisher, Q. J. & Knipe, R. K. (eds) Faulting, Fault Sealing and Fluid Flow in Hydrocarbon Reservoirs, Geological Society, London, Special Publications, 147, 1–25.
Toublanc, A., Renaud, S. Sylte, J. E. Clausen, C. K. Eiben, T. & Nådland, G. 2005. Ekofisk Field: fracture permeability evaluation and implementation in the flow model. Petroleum Geoscience, 11, 321–330.[Web of Science][GeoRef]
van der Zee, W. & Urai, J. L. 2005. Processes of normal fault evolution in a siliciclastic sequence: a case study from Miri, Sarawak, Malaysia. Journal of Structural Geology, 27, 2281–2300.[CrossRef][Web of Science][GeoRef]
Vendeville, B. C. & Jackson, M. P.A. 1992a. The rise of diapirs during thin-skinned extension. Marine and Petroleum Geology, 9, 331–353.[CrossRef][Web of Science][GeoRef]
Vendeville, B. C. & Jackson, M. P.A. 1992b. The fall of diapirs during thin-skinned extension. Marine and Petroleum Geology, 9, 354–371.[CrossRef][Web of Science][GeoRef]
Verwest, B. J. 1994. Seismic migration in elliptically anisotropic media. Journal of Geophysical Prospecting, 37, 149–166.
Walsh, J. J. & Watterson, J. 1987. Distributions of cumulative displacement and seismic slip on a single normal fault surface. Journal of Structural Geology, 9, 1039–1046.[CrossRef][Web of Science][GeoRef]
Walsh, J. J. & Watterson, J. 1991. Geometric and kinematic coherence and scale effects in normal fault systems. In: Roberts, A. M., Yielding, G. & Freeman, B. (eds) The Geometry of Normal Faults, Geological Society, London, Special Publications, 56, 193–203.
Walsh, J. J., Watterson, J. Heath, A. E. & Childs, C. 1998. Representation and scaling of faults in fluid flow models. Petroleum Geoscience, 4, 241–251.
Walsh, J. J., Nicol, A. & Childs, C. 2002. An alternative model for the growth of faults. Journal of Structural Geology, 24, 1669–1675.[CrossRef][Web of Science][GeoRef]
Warren, J. E. & Root, P. J. 1963. The behaviour of naturally fractured reservoirs. Society of Petroleum Engineers Journal, September 1963, 245–255, (SPE426).
Watterson, J., Walsh, J. J. Nicol, A. Nell, P. A. R. & Bretan, P. 2000. Geometry and origin of a polygonal fault system. Journal of the Geological Society, London, 157, 151–162.
Winterstein, D. F. 1989. Velocity anisotropy terminology for geophysicists. Geophysics, 55, 1070–1088.[CrossRef][Web of Science]
Wong, T.-F., David, C. & Zhu, W. 1997. The transition from brittle faulting to cataclastic flow in porous sandstones: Mechanical deformation. Journal of Geophysical Research, 102,B2, 3009–3025.
Yielding, G. 2002. Shale gouge ratio – calibration by geohistory. In: Koestler, A. G. & Hunsdale, R. (eds) Hydrocarbon Seal Quantification, Norwegian Petroleum Society, Special Publication, 11, 1–15.[CrossRef]
Yielding, G., Needham, T. & Jones, H. 1996. Sampling of fault populations using sub-surface data: a review. Journal of Structural Geology, 18, 135–146.[CrossRef][Web of Science][GeoRef]
Yielding, G., Freeman, B. & Needham, D. T. 1997. Quantitative fault seal prediction. American Association of Petroleum Geologists Bulletin, 81, 897–917.[Abstract][GeoRef]
Zhang, J., Wong, T.-F. & Davis, D. M. 1990. Micromechanics of pressure-induced grain crushing in porous rocks. Journal of Geophysical Research, 95, 341–352.[GeoRef]
Zoback, M. D. & Zinke, J. C. 2002. Production-induced normal faulting in the Valhall and Ekofisk oil fields. Pure and Applied Geophysics, 159, 403–420.[CrossRef][Web of Science][GeoRef]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
