Fault seal mechanisms

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. 

The prediction of fault seal capacity requires the evaluation of each fault seal mechanism and its possible effect on fault sealing. 

The fault sealing mechanisms considered are those which occur as a direct result of the faulting process, i.e., those due to either across-fault juxtapositions of reservoir and non-reservoir units or to the presence of sealing fault rocks, i.e., membrane seals. The diage-netic contribution to seals (Knipe, 1992) is not considered. 

This paper has highlighted that a number of components, important to fault seal analysis, are often either not included or not quantified in sufficient detail to allow a low risk seal evaluation. The main components which are not always considered in detail are (i) the errors in throw patterns which arise from seismic resolution and fault damage zone structures (ii) the assumption that juxtaposition of reservoir against low permeability units and shale smear are the only sealing mechanisms and (iii) that fault seal data from anywhere is directly applicable to any other sealing problem, i.e., that the geohistory is not critical... 

Despite the common assumption of fault sealing in hydrocarbon fields, very few faults have been characterised in the detail needed which allows identification of the sealing mechanism or controls. Without the construction of a robust set of case histories from such analysis, future seal evaluation will remain a high risk venture. These case histories are also needed to integrate seal behaviour with pressure test, production and in situ stress analysis. The paper has highlighted the importance of an integrated approach from micro to macro and stressed the value of core-based studies to quantify fault rock properties, sub-seismic fault populations and sealing mechanisms. 

Fig. 6. Brittle fault seal analysis strategy. The strategy aims to quantify the sealing capacity of brittle faults by first predicting the deformation mechanism. Particulate flow faults are treated as non-sealing. Cataclastic faults have variable sealing properties according to fault throw and matrix properties. The chart in the lower left of this figure is reproduced at a larger scale in Fig. 8.

During early 1994 a study was undertaken to provide geometric descriptions of the faults and their likely sealing mechanisms. A total of 16 block-bounding and internal faults (Fig. 3) were selected and analysed. 

Four mechanisms have been suggested to explain how faults provide seals. The most frequent case is that of clay smear and juxtaposition (Fig. 5.8)... 

Results and conclusions The fault zones at Frechen contain clay fillings of up to 1 m in thickness, derived from extremely plastic shale source beds and smeared out over distances as much as 70 m in dip direction. The generation of substantial smears requires slow fault displacement rates and sufficient shale ductility. When a thick shale source bed is traversed by a normal fault, it is first flexed and eventually disrupted by a pull-apart mechanism that creates room for the emplacement of thick clay smears. Simple theoretical considerations suggest that the source bed thickness to some power n + 1 > 2 may be a key parameter in the ranking of seal quality. The length of continuous smears increases with source bed thickness, but will ultimately be controlled by the smearing process. The latter remains to be investigated. 

The prediction of the sealing effect of brittle deformation in reservoir sandstones involves several steps (Figs. 2 and 6). At present, predictions are limited to clean sandstones. The initial step assesses the potential for and extent of cataclastic deformation. If cataclastic deformation can be shown to be the likely deformation mechanism, then the fault properties can... 

Lehner, F.K. and Pilaar, W.F. 1997. On a mechanism of clay smear emplacement in synsedimentary normal faults. In P. Mpller-Pedersen and A.G. Koesiter (Editors), Hydrocarbon Seals Importance for Exploration and Production, NPF Special Publication 7. Elsevier, Singapore, pp. 39-50. 

The bifurcation mechanisms for formation of multi-slip fault zones suggest that maximum fault zone thickness will often correspond to the strike-normal distance between the traces of two overlapping slip surfaces (Fig. 2c). Fault overlaps and their breached equivalents occur on faults of all sizes as do, by implication, paired and multi-slip surface fault zones. Complex and paired slip surface fault zone structures will occur on scales below that resolvable by even high quality seismic data (lateral resolution is no better than 50-100 m at North Sea reservoir depths). The possible impact of sub-seismic complexity and paired slip surfaces on connectivity and sealing across faults offsetting an Upper Brent type sequence are briefly considered below. 

Diagenetic analysis has proved to be useful in order to constrain the depths at which faulting occurs, and may help to predict deformation mechanisms and the related fault characteristics. The diagenetic processes are also important after the faults have been generated, as modifications of fault structures may occur during deeper burial. Understanding the deformation mechanisms is important in order to perform valid evaluations of the sealing potential of faults in reservoirs. 

Most seals in clastic sequences are membrane seals (Watts, 1987). The dominant control on seal failure is the capillary entry pressure of the seal-rock, that is, the pressure required for hydrocarbons to enter the largest interconnected pore throat of the seal. A number of mechanisms have been recognised whereby fault planes can act as a membrane seal (e.g.. Watts, 1987 Knipe, 1992) ... 

This paper seeks to contribute to a clarification of the overall perspective. The influence of a single layer of clay in a sandstone sequence on the basic mechanics of fault development and associated seal formation are examined by numerical comparisons with shear band and fault development in a homogeneous sandstone. The potential sealing capacity of shear bands, formed in the absence of clay layers, is re-evaluated by reference to pore physics models relating permeability and capillary pressure, supported by field data. By this means it is hoped to expedite resolution of apparently ongoing industry misconceptions of the potential sealing capacity of such structures. 

The normal stress acting on such a fault plane approaches the lowest values of the stress tensor. In the writers experience, the least effective stress in North Sea fields is typically less than 10 MPa. The hydraulic conductivity of many open natural fractures is reduced to that of the unfractured rock at approximately 10 MPa (see, for example, data presented by Gale (1982)). Consequently, we may readily infer that at least any unfilled fractures which are subparallel to the main fault plane, within or adjacent to that plane, may be conductive. Although fractures may not always be present (e.g., if the fault plane is filled with salt), this provides one mechanism which is broadly consistent with the observation of a minimum of sealing faults striking parallel to the maximum horizontal stress. 

A proper understanding of the above items, including maturation and filling history, formation pressure distributions, intra-reservoir communications, fault and top seal potentials, and leakage mechanisms, is considered essential for resource assessment, safe drilling of further explora-tion/delineation and production wells, and for reservoir management and production planning of the Njord Field. 

---