Fault rocks

Inferences from San Gabriel fault rock geochemistry and microstructure. Journal of Geophysical Research, 100, 13007-13020. 

Fault seal prediction in hydrocarbon reservoirs requires an understanding of fault seal mechanisms, fault rock petrophysical properties, the spatial distribution of seals, and seal stability. The properties and evolution of seals within fault zones can be evaluated using the combined results of structural core logging, microstmctural and physical property characterisation, together with information on fault populations from seismic and outcrop studies and well test data. 

Successful seal analyses depends upon the amalgamation of data from the miero-scale to the macro-scale. This review demonstrates that improvements in fault seal risk evaluation are possible. The future direetions for improving fault seal risk evaluation are also discussed. The most critical of these are characterisation of the internal structure of fault zones, generation of a database for fault rock petrophysical properties and incorporation of the impact of realistic fault zone geometries into reservoir modelling programs. 

Without a detailed understanding of the fundamental processes which control the evolution of fault rocks and their properties, the prediction of sealing capacity and the evaluation of the behaviour of a faulted reservoir will never be anything more than speculative. 

Critical information required for fault-rock property evaluation... 

Surprising little has been published on the detailed physical properties or microstructural evolution of fault rocks in hydrocarbon reservoirs. Recent papers which have begun to address this aspect of sealing include analysis of clay smears (Knipe 1992a, 1994 Berg and Avery, 1995), cataclasites from pure sandstones (Pitman, 1981 Underhill and Woodcock, 1987 Antonellini and Aydin, 1994 Fowles and Burley, 1994), and deformation in impure sandstones... 

The terminology and classification of fault-rocks and seal types is not yet universally agreed (Knipe, 1992a Knott, 1993). The classification presented below is based on identification of the main process responsible for the reduction in permeability associated with the faults. Mechanistic terms have been combined with textural descriptive terms to provide a more expansive nomenclature system which covers the most common fault rocks and seat types. The fault seal types and associated fault-rock types can be divided into two broad categories ... 

Framework - phyllosilicate/microcrystalline quartz fault rocks This class of fault rock is introduced here to describe fault rocks which form in sediments with concentrations (>20%) of dissolvable sponge spicules and varying amounts (15-40%) of phyllosilicates (Fig. 2d). Although not common, sediments with high sponge spicules contents are prone to the development of fault rocks by the cataclasis, dissolution and reprecipitation of silica. The process may be initiated by the collapse of secondary pores created by spicule dissolution and induce the redistribution of more soluble material by mixing. 

Properties of different fault rock/seal types... 

Fundamental to a successful fault seal analysis is quantification of the petrophysical properties of the different fault rocks present in the hydrocarbon field under investigation. The critical properties which require quantification are permeabilities, capillary entry pressures, transmissibility, fault-rock thickness and the strength of the fault rocks. One of the reasons why fault seal analysis and reservoir modelling has proved difficult has been the absence of data on these properties. Analysis of the petrophysical properties of... 

Fig. 3, Porosity-permeability plot of cataclastic fault rocks developed from sandstones with low (<6%) clay contents. Note the values extend over six orders of magnitude.

Each of the points made in the last section above highlights the complex interaction of variables which can control the final properties of fault rocks. The list also demonstrates that although a difficult problem, understanding of fault rock behaviour is possible if systematic investigations of fault rock evolution are... 

Fig. 4. Capillary pressure curves for two fault rocks. The data emphasise that while a single entry pressure does describe the properties of material with a simple pore structure, e.g., (a), the complex mixing of grain and pore sizes in the cataclastic shown in (b) generates a more complex pressure curve where a single characteristic entry pressure is more difficult to define, i.e., reflects more of a membrane behaviour.

Accurate permeability measurements of fault rocks as well as characterisation of capillary entry pressure data of fault rocks requires different equipment and procedures than those applicable to... 

The critical elements of fault damage zones which are needed for fault seal evaluation and for input into reservoir behaviour simulation include (i) the dimensions of the damage zone (ii) the fault clustering characteristics (iii) the fault offset populations, which can control the distribution of fault rocks and juxtapositions (iv) the orientation distributions of deformation features present within damage zones and (v) the total thickness of fault-rocks. Each of these aspects are reviewed below, where the data presented are part of a large database collected from the structural analysis of -90 wells, (-25 km of core) from the North Sea area (see example in Fig. 7). The final part of this section presents a simple model which demonstrates the impact of damage zone structures on flow. 

Most fault offset population analysis (see Cowie et al., 1996) have concentrated on prediction of the number of sub-seismic over large areas (>1 km ) faults rather than the distribution of the faults within the fault zones. In many cases a uniform distribution of faults across an area is assumed. The data from the structural logging of North Sea wells illustrate that the characteristics of faults found on a field scale are also present within individual fault zones identified on seismic or from well data. Because the population of small faults around larger structures will control the distribution of juxtapositions and fault rocks, detailed characterisation of the offsets is important to seal analysis. Fig. 11 illustrates the population characteristics of three fault zones with different offsets. 

The volume and continuity of fault rock in the damage zone... 

The thickness of the low permeability fault rocks is an important variable in evaluation of cross-fault flow behaviour. Because fault zones are usually composed of complex arrays of intersecting sub-seismic faults and fractures and because the permeability reductions associated with faulting develop after low amounts ( cm) of displacement incorporation of the impact of complex fault zones (as opposed to single faults ) is not a simple procedure. 

Two variables are fundamental to assessing the flow across complex fault zones. The first variable is the cumulative fault-rock thickness across the fault zone, i.e., the total thickness of fault-rock from all faults along the flow path. This depends upon the fault frequency along the flow path and is not equivalent to the fault damage zone thickness (cf. Knott, 1993) unless the fault zone is invaded by cements. The second variable is the connectivity of the faults or deformation features with low permeabilities in the fault zone. In the case of a completely connected array with no windows of undeformed material along possible flow paths, the flow is controlled by the permeability of the fault rocks. Where a more open network of faults is present then the flow will depend upon the tortuosity associated with flow around the low permeability zones and the ratio of matrix to fault-rock permeability. The interaction of these two factors will control the effective transmissivity of the zone. We have constructed a database on... 

Fig. 16. Numerical analysis evaluating the connectivity of faults with a damage zone. The analysis is aimed at identifying the critical density of structural features needed to separate domains with (a) fault-rock controlled flow behaviour and (b) tortuosity controlled flow behaviour. In the example shown, a transition from an open array of faults with tortuosity flow to a connected array occurs when the structural frequency is equivalent to between 200 features/100 m of core and 400 features/100 m of core.

The basic requirement for mapping fault seals is the generation of a realistic, maximum probability map of sealing capacities along individual fault zones. This involves evaluation of the possible juxtaposition patterns within the zone as well as an assessment of the variance of fault rock properties. 

---