Search

Improving the prediction of sub-seismic structures and their petrophysical properties is essential for realistic characterization of deformed sandstone reservoirs. In the present paper, we describe permeability contrasts induced by cataclastic deformation bands and faults in porous sandstones (766 data synthesized from field examples and the literature). We also discuss the influence of several factors, including tectonic regime, presence of a fault, burial depth, host sandstone porosity, and grain size and sorting for their initiation and permeability. This analysis confirms that permeability decrease is as a function of grain-crushing intensity in bands. Permeability reduction ranges from very limited in crush-microbreccia of compaction bands to high permeability reduction in cataclasites and ultracataclasites of shear-dominated bands, band clusters and faults. Tectonic regime, and especially normal-fault regime, with its tendency to localize strain and generate faults, is identified as the most important factor, leading to the formation of cataclastic bands with high permeability contrasts. Moreover, moderate burial depth (1–3 km) favors cataclastic bands with high permeability contrasts with respect to the host sandstone. High porosity, coarse-grain size and good grain sorting can slightly amplify the permeability reductions recorded in bands. [ABSTRACT FROM AUTHOR]

Deformation within the upper crust (elastico-frictional regime) is largely accommodated by fractures and conjugate faults. The Coulomb fracture criterion leads us to expect that the average dihedral angle of conjugate-fault sets is expected to be ∼60°. Experiments, however, reveal a significant amount of scatter from this 60° average. The confining pressure under which these rocks are deformed is a contributing factor to this scatter. The Canyon Range syncline, Sevier fold-thrust belt (USA) and the Jebel Bani, Anti-Atlas fold-belt (Morocco) both folded under different depths, within the elastico-frictional regime, by cataclastic flow. Conjugate-fault sets assisted deformation by cataclastic flow. The Canyon Range syncline and the Jebel Bani are used here as natural examples to test the relationship between the dihedral angle of conjugate-faults and confining pressure. Variations is confining pressure are modeled by the difference in depth of deformation and position within the folds. Results from this study show that the dihedral angle increases with an increase in depth and within the hinge regions of folds, where space problems commonly occur. Moreover, the shortening directions based on the acute bisectors of conjugate-faults may not be accurately determined if the dihedral angles are unusually large or small, leading to incorrect kinematic analyses. [ABSTRACT FROM AUTHOR]

Slip along faults generates wear products such as gouge layers and cataclasite zones that range in thickness from sub-millimeter to tens of meters. The properties of these zones apparently control fault strength and slip stability. Here we present a new model of wear in a three-body configuration that utilizes the damage rheology approach and considers the process as a microfracturing or damage front propagating from the gouge zone into the solid rock. The derivations for steady-state conditions lead to a scaling relation for the damage front velocity considered as the wear-rate. The model predicts that the wear-rate is a function of the shear-stress and may vanish when the shear-stress drops below the microfracturing strength of the fault host rock. The simulated results successfully fit the measured friction and wear during shear experiments along faults made of carbonate and tonalite. The model is also valid for relatively large confining pressures, small damage-induced change of the bulk modulus and significant degradation of the shear modulus, which are assumed for seismogenic zones of earthquake faults. The presented formulation indicates that wear dynamics in brittle materials in general and in natural faults in particular can be understood by the concept of a 'propagating damage front' and the evolution of a third-body layer. [ABSTRACT FROM AUTHOR]

Abstract: In this study three different sites along the ENE-trending, sinistral Salzach–Ennstal–Mariazell–Puchberg [SEMP] fault zone were investigated with respect to brittle fault zone evolution and fault re-activation. All sites crop out in Triassic carbonates (Ladinian Wetterstein limestone/–dolomite). Simultaneously (re-) activated faults were investigated with focus on fault-slip data and structural inventory of each individual fault zone. Configuration of (internal) structural elements, fault core thickness, strike direction and slip sense in addition to particle analysis of fault core cataclasites add up to three different fault types (Fault Types I, II and III). Fault Type I is classified by a complex internal fault core structure with thicknesses up to several 10s of meters and generally evolve in a strike direction of maximum shear stress (τmax). Type II faults, characterized by cataclastic fault cores with thicknesses up to 1m, as well as Type III faults (thin solitary cataclastic layers) evolve sub-parallel to the main fault direction and in orientation according to R, R' or X shear fractures with variable (σn/τ) ratio. Progressive development from Type III to Type II and Type I faults is consistent with increasing displacement and increasing fault core width. Fault type classification and related paleostress analysis provide evidence from field observation compared to theoretical and analog models of Mohr–Coulomb fracture evolution. [Copyright &y& Elsevier]

Abstract: Recent work from portions of the Sevier fold-thrust belt that have deformed primarily within the elastico-frictional regime, demonstrates that cataclastic flow can be subdivided into two types: matrix- and block supported. The two types may operate simultaneously within the same deforming material. However, their activity can vary spatially, temporally and across scales. Although block-supported cataclastic flow is a critical process in upper crustal deformation, it continues to be largely ignored and/or misunderstood, primarily because established concepts and definitions for cataclastic flow are chiefly based on matrix-supported cataclastic flow. Here, block-supported cataclastic flow is examined to better understand cataclastic flow in general and to explore its relationship with matrix-supported cataclastic flow. [Copyright &y& Elsevier]

The Hailar-Tamtsag Basin is a typical rift basin where two sets of regional caprocks are developed, i.e., mudstone caprocks (containing a small amount of tuff) developed in strong rifting stage and fault-depression stage, respectively. The caprocks have a cumulative thickness of 50-120 m in general, and a single-layer thickness of 20-50 m, interbedded with sandstone about 1-2 m thick. The large set of mudstone is distributed continuously, as pure mudstone caprock. Forty-three mudstone and tuff samples were taken to perform displacement pressure test with our independently developed displacement pressure tester. Test result shows that the displacement pressure is 0.04-10.00 MPa, which is equivalent to 0.09-20.01 MPa after being corrected to the actual burial depth. As the burial depth increases, the displacement pressure rises gradually, and is 1-10 times greater than that of oil-bearing sandstone or conglomerate reservoir at the same burial depth. The difference between displacement pressure of mudstone caprock and sandstone or conglomerate reservoir increases greatly if the burial depth exceeds 1000 m. Because of the displacement pressure difference between caprock and reservoir, the maximum height of hydrocarbon column sealed by the caprock ranges from 300 m to 2000 m, much higher than the height of the trap closure at the same burial depth. No hydrocarbons will leak through such caprocks. Analysis of the reservoir GOI, homogeneous temperature, and crude maturity of both deep and shallow reservoirs in Well Wu-20 in the Wuerxun Depression shows that Damoguaihe II Member reservoir is a secondary reservoir developed during adjustment of early reservoir. It is mainly distributed in inverted structures. Faulting is one of the main factors breaking the integrity of caprocks. Three kinds of faults are often related to the migration of hydrocarbons across regional caprocks: the first type refers to the positive inverted faults breaking the early sealing conditions and making the hydrocarbons accumulated in the early period re-migration; the second type refers to the faults with shear mudstone smear structures which lose their continuity and open vertically when the fault throws exceed 5 times of the mudstone thickness; the third type refers to the normal faults developed in the structural inversion stage, which have a typical cataclastic structure and are easy to cause vertical migration of hydrocarbons. Hydrocarbon enriched horizons of the rift basins are co-controlled by 'fault-caprock'. Specifically, 95% of the geological reserve is controlled by regional caprocks, and only 5% is controlled by open faults, secondary oil/gas reservoirs are developed over regional caprocks. [ABSTRACT FROM AUTHOR]

The 2004-2008 eruption of Mount St. Helens produced seven dacite spines mantled by cataclastic fault rocks, comprising an outer fault core and an inner damage zone. These fault rocks provide remarkable insights into the mechanical processes that accompany extrusion of degassed magma, insights that are useful in forecasting dome-forming eruptions. The outermost part of the fault core consists of finely comminuted fault gouge that is host to 1- to 3-mm-thick layers of extremely fine-grained slickenside-bearing ultracataclasite. Interior to the fault core, there is an ~2-m-thick damage zone composed of cataclastic breccia and sheared dacite, and interior to the damage zone, there is massive to flow-banded dacite lava of the spine interior. Structures and microtextures indicate entirely brittle deformation, including rock breakage, tensional dilation, shearing, grain flow, and micro faulting, as well as gas and fluid migration through intergranular pores and fractures in the damage zone. Slickenside lineations and consistent orientations of Riedel shears indicate upward shear of the extruding spines against adjacent conduit wall rocks. Paleomagnetic directions, demagnetization paths, oxide mineralogy, and petrology indicate that cataclasis took place within dacite in a solidified steeply dipping volcanic conduit at temperatures above 500 °C. Low water content of matrix glass is consistent with brittle behavior at these relatively high temperatures, and the presence of tridymite indicates solidification depths of <1 km. Cataclasis was coincident with the eruption's seismogenic zone at <1.5 km. More than a million small and low-frequency "drumbeat" earthquakes with coda magnitudes (Md) <2.0 and frequencies <5 Hz occurred during the 2004-2008 eruption. Our field data provide a means with which to estimate slip-patch dimensions for shear planes and to compare these with estimates of slip patches based on seismic moments and shear moduli for dacite rock and granular fault gouge. Based on these comparisons, we find that aseismic creep is achieved by micron-scale displacements on Riedel shears and by granular flow, whereas the drumbeat earthquakes require millimeter to centimeter displacements on relatively large (e.g., ~1000 m²) slip patches, possibly along observed extensive principal shear zones within the fault core but probably not along the smaller Riedel shears. Although our field and structural data are compatible with stick-slip models, they do not rule out seismic and infrasound models that call on resonance of steam-filled fractures to generate the drumbeat earthquakes. We suggest that stick-slip and gas release processes may be coupled, and that regardless of the source mechanism, the distinctive drumbeat earthquakes are proving to be an effective precursor for dome-forming eruptions. Our data document a continuous cycle of deformation along the conduit margins beginning with episodes of fracture in the damage zone and followed by transfer of motion to the fault core. We illustrate the cycle of deformation using a hypothetical cross section of the Mount St. Helens conduit, extending from the surface to the depth of magmatic solidification. [ABSTRACT FROM AUTHOR]

Abstract: Field investigations, combined with meso- and microstructural analyses, reveal that numerous ultracataclastic veins are widely developed within a fault zone (<150 m wide) as simple veins, complex lenses, and networks, along active faults of the Arima–Takatsuki Tectonic Line, southwest Japan. These veins comprise mainly pseudotachylyte-like vein and weakly consolidated to unconsolidated fault gouge that is black, dark-brown, brown, gray, and brownish-red in color. Meso- and microstructural features show that these pseudotachylyte-like and fault gouge veins and networks formed during multiple stages, as earlier veins are generally cut and overprinted by younger veins, indicating that the vein-forming events occurred repeatedly and that ultracataclastic material was injected into networks of faults and fractures in the fault zone. The pseudotachylyte-like and fault gouge veins are characterized by an ultrafine- to fine-grained matrix and angular to subangular fragments of host granitic rocks of various sizes, ranging from submicron to millimeters. SEM–EDS (Scanning Electronic Microscope-Energy Dispersive X-ray) and powder X-ray diffraction analyses show that all the ultracataclastic veins are characterized by crystalline materials composed mainly of quartz and feldspar, similar to the host granitic rocks. The present results support the existing hypothesis that ultrafine- to fine-grained materials formed by comminution can be fluidized and injected rapidly into fracture networks located far from the source fault plane in a solid–fluid–gas system during seismic slip; therefore, such materials provide a record of paleoseismic faulting events that occurred repeatedly within the seismogenic fault zone. [Copyright &y& Elsevier]

Abstract: Conjugate cataclastic deformation bands cut unconsolidated sand and gravel at McKinleyville, California, and dip shallowly towards the north-northeast and south-southwest. The acute dihedral angle between the two sets of deformation bands is 47° and is bisected by the sub-horizontal, north-northeast directed incremental and finite shortening directions. Trishear models of fault propagation folding above the McKinleyville fault predict two sets of LNFE (lines of no finite elongation) that plunge steeply and shallowly to the south and north. These predictions are inconsistent with deformation band orientations and suggest that deformation bands did not form parallel to these LNFE. During plane strain, zero extension directions with acute dihedral angles of 47° develop when the dilatancy rate (dV/dε 1) is −4.3. Experimental dilatancy rates for Vosges sandstone (cohesion > 0) and unconsolidated Hostun sand suggest the deformation bands either developed parallel to zero extension directions or in accordance with the Mohr–Coulomb criterion, assuming initial porosities of 22% and 39%, respectively. An empirical relationship between dV/dε 1, relative density and mean stress suggests that dilatancy rates for Vosges sandstone overestimate dV/dε 1 at McKinleyville. Deformation bands at McKinleyville likely developed either in a Mohr–Coulomb orientation, or an intermediate orientation bounded by the Mohr–Coulomb (θ C ) and Roscoe (θ R ) angles. [Copyright &y& Elsevier]

Abstract: With optical microscopy, scanning electron microscopy (SEM), transmission electron microscopy (TEM), and synchrotron X-ray diffraction measurements, we analyzed clay microfabrics in ultracataclastic/gouge and cataclastic core samples obtained from the main bore hole of the San Andreas Fault observatory at depth (SAFOD). The analysis reveals a significant contrast between weak clay fabrics observed in the core samples with synchrotron X-ray fabric measurements and strong degree of preferred alignment for clay particles documented with the optical microscope. TEM and SEM observations also show distinct zones of locally aligned and randomly oriented clay minerals. The lack of a strong fabric may be attributed to randomly oriented matrix sheet silicates dominating the fault rocks. The presence of weak fabrics in intensely strained ultracataclasites/fault gouges is attributed to 1) newly formed clay minerals that grew in many orientations, 2) folded and kinked clay minerals, and 3) clay particles that are wrapped around grains. In addition, the locally aligned clay particles may act as barriers to fluid flow, which in turn decrease porosity, expel intergranular pore fluids, and consequently, may increase fluid pressure. [Copyright &y& Elsevier]

Abstract: Understanding the impact of faults on fluid flow in the subsurface is important for the extraction of oil, gas and groundwater as well as the geological storage of waste products. We address two problems present in current industry-standard workflows for fault seal analysis that may lead to fault rocks not being represented adequately in computational fluid flow models. Firstly, fluid flow properties of fault rocks are often measured only for small-scale faults with throws not exceeding a few centimetres. Large seismic-scale faults (throws >20 m) are likely to act as baffles or conduits to flow but they are seldom recovered from subsurface cores and consequently fault rock data for them is sparse. Secondly, experimental two-phase fluid flow data is lacking for fault rocks and, consequently, uncertainties exist when modelling flow across faults in the presence of two or more immiscible phases. We present a data set encompassing both single- and two-phase fluid flow properties of fault and host rocks from the 90-Fathom fault and its damage zone at Cullercoats Bay, NE England. Measurements were made on low-throw single and zones of deformation bands as well as on slip-surface cataclasites present along the ∼120 m throw main fault. Samples were analysed using SEM and X-ray tomography prior to petrophysical measurements. We show that single deformation bands, deformation band zones and slip-surface cataclasites exhibit dissimilar single- and two-phase fluid flow properties. This is due to grain-size reduction being more pronounced in slip-surface cataclasites and changes in microstructure being fault-parallel for deformation bands but mostly fault-perpendicular for slip-surface cataclasites. A trend of fault rocks with low absolute permeabilities exhibiting lower relative permeabilities than more permeable rocks at the same capillary pressure is evident. [Copyright &y& Elsevier]

Fault rocks can function as barriers to subsurface fluid flow and affect the storage of CO2 in geological structures. Even though flow across faults often involves more than one fluid phase, it is typically modeled using only single-phase functions due to a lack of fault rock relative permeability data and complexities in incorporating two-phase flow properties into flow simulations. Here we present two-phase fluid flow data for cataclastic fault rocks in porous sandstone from the 90-Fathom fault (northeast England). The study area represents a field analogue for North Sea saline aquifers of Permian-Triassic age that are currently being considered for CO2 storage. We use the experimental data to populate a synthetic model of a faulted saline aquifer to assess the impact of these fault rocks on CO2 injection. We show that even fault rocks with low clay contents and very limited quartz cementation can act as major baffles to the flow of a non-wetting phase if realistic two-phase properties are taken into account. Consequently, pressure may increase far more rapidly in the storage compartment during CO2 injection than anticipated based on models that only incorporate absolute fault rock permeabilities. To avoid high pressures, which may lead to hydrofracturing and CO2 leakage, either more complex injection strategies need to be adopted or seismic data acquired to ensure the absence of faults in aquifers selected for CO2 storage. [ABSTRACT FROM AUTHOR]

Abstract: Previous studies reveal that while compact carbonate rocks display exclusively dilatancy under compressive deformation, compaction may be observed in their more porous counterparts. In their compactive behavior the porous carbonate rocks are more akin to e.g., sandstone. Whereas the micromechanics of brittle faulting and cataclastic flow in sandstone has been studied extensively, little is known about these processes in a porous limestone. To investigate both failure modes we deformed samples of Tavel limestone with porosity 10–14% to various stages of deformation in conventional triaxial configuration at confining pressures corresponding to brittle faulting and cataclastic flow and described the microstructures associated with the damage evolution using optical and electron microscopy. In this porous micritic limestone cataclasis is the dominant mechanism of deformation. The microcracks initiate as pore-emanated and while all pores contribute to microcrack initiation, it is the large pores that drive crack propagation and coalescence leading to failure. In brittle faulting, dilatancy arises from microcracks growing parallel to maximum principal stress, with their coalescence leading to shear localization. In cataclastic flow microcracks do not have preferred orientation. Interplay between pore collapse and formation of new microcracks determines the compactive versus dilatant character of the cataclastic flow. [ABSTRACT FROM AUTHOR]

Abstract: The factors controlling the development of different types of fault rock assemblages and, more specifically, the formation of friction melts are still not fully understood. In this study we compared two exhumed strike–slip faults in the Adamello batholith (Southern Alps): the Gole Larghe and the Passo Cercen fault zones, active at 9–11km depth and temperatures of 250–300°C. Each fault zone consists of hundreds of sub-parallel strands exploiting pre-existing joints. The Gole Larghe fault strikes N105±5° and is dextral; the fault rocks are cataclasites and widespread, centimetre-thick pseudotachylytes. The Passo Cercen fault strikes on average N130° and is formed by multiple fault horizons: fault segments striking N105°–N130° are mainly dextral, whereas faults striking N135°–N140° are mainly sinistral. Microstructural, mineralogical and geochemical investigations show that the fault rocks are cataclasites associated with thick epidote+K-feldspar+quartz veins and rare, millimetre-thick pseudotachylytes. Field evidence suggests that in both fault zones, the direction of the maximum horizontal stress σ 1 was N135°. The Gole Larghe fault strikes at about 30° to σ 1 and is favourably oriented for reactivation. By contrast, the Passo Cercen fault strikes at low angles to σ 1 and is unfavourably oriented for reactivation, therefore requiring the development of high pore pressures, as suggested by the occurrence of extensive epidote veining and hydraulic breccias. It is proposed that frictional melting in the Passo Cercen fault zone was inhibited by the development of high pore pressures and low effective normal stresses. [Copyright &y& Elsevier]

Abstract: In the Sub-Pelagonian and Pelagonian Zones, kinematic analyses on faults and cataclastic shear zones document an overall displacement vector towards NNW and WSW/SSW for Attica and western Macedonia, respectively. This tectonic episode is interpreted as being triggered by the kinematically Oligocene–Miocene detachment of the Sub-Pelagonian and Pelagonian Zones from the high-P rocks. During this episode the Sub-Pelagonian and Pelagonian Zones were decoupled along detachment zones, fragmentised by brittle mechanisms and horizontally stretched in km-scale. The strain record of the cataclastically-deformed lithologies of the upper plates reveals a complex kinematic pattern with displacement partitioning in an overall non-coaxial regime. These structural processes are considered to be kinematically linked to the exhumation of the high-P rocks, deforming the Sub-Pelagonian and Pelagonian Zones. [Copyright &y& Elsevier]

Abstract: Field observations and interpretations of satellite images reveal that the westernmost segment of the Altyn Tagh Fault (called Karakax Fault Zone) striking WNW located in the northwestern margin of the Tibetan Plateau has distinctive geomorphic and tectonic features indicative of right-lateral strike-slip fault in the Late Quaternary. South-flowing gullies and N–S-trending ridges are systematically deflected and offset by up to ~1250 m, and Late Pleistocene–Holocene alluvial fans and small gullies that incise south-sloping fans record dextral offset up to ~150 m along the fault zone. Fault scarps developed on alluvial fans vary in height from 1 to 24 m. Riedel composite fabrics of foliated cataclastic rocks including cataclasite and fault gouge developed in the shear zone indicate a principal right-lateral shear sense with a thrust component. Based on offset Late Quaternary alluvial fans, 14C ages and composite fabrics of cataclastic fault rocks, it is inferred that the average right-lateral strike-slip rate along the Karakax Fault Zone is ~9 mm/a in the Late Quaternary, with a vertical component of ~2 mm/a, and that a M 7.5 morphogenic earthquake occurred along this fault in 1902. We suggest that right-lateral slip in the Late Quaternary along the WNW-trending Karakax Fault Zone is caused by escape tectonics that accommodate north–south shortening of the western Tibetan Plateau due to ongoing northward penetration of the Indian plate into the Eurasian plate. [Copyright &y& Elsevier]

Abstract: It is becoming increasingly common practice to model the impact of faults on fluid flow within petroleum reservoirs by applying transmissibility multipliers, calculated from the single-phase permeability of fault rocks, to the grid-blocks adjacent to faults in production simulations. The multi-phase flow properties (e.g. relative permeability and capillary pressure) of fault rocks are not considered because special core analysis has never previously been conducted on fault rock samples. Here, we partially fill this knowledge gap by presenting data from the first experiments that have measured the gas relative permeability (krg ) of cataclastic fault rocks. The cataclastic faults were collected from an outcrop of Permo-Triassic sandstone in the Moray Firth, Scotland; the fault rocks are similar to those found within Rotliegend gas reservoirs in the UK southern North Sea. The relative permeability measurements were made using a gas pulse-decay technique on samples whose water saturation was varied using vapour chambers. The measurements indicate that if the same fault rocks were present in gas reservoirs from the southern Permian Basin they would have krg values of <0.02. Failure to take into account relative permeability effects could therefore lead to an overestimation of the transmissibility of faults within gas reservoirs by several orders of magnitude. Incorporation of these new results into a simplified production simulation model can explain the pressure evolution from a compartmentalised Rotliegend gas reservoir from the southern North Sea, offshore Netherlands, which could not easily be explained using only single-phase permeability data from fault rocks. [Copyright &y& Elsevier]

Abstract: To assess the role of the fault thickness on its mechanical behavior, we first present the results of an experimental modeling of a thick fault core. Our laboratory setup consists in an annular simple shear apparatus in which we can apply very large shear displacements (50 m) to 100 particle thick granular samples. Thanks to a window in the apparatus, pictures of the microstructures can be continuously taken during shear. We observe from a Correlation Image Velocimetry technique that a significant strain field exists outside of the observable shear band. This strain field, though of small magnitude compared to that existing inside the shear band, is very structured and extends in a region much wider than expected from individual static observations (i.e. wider than the directly observable shear band). Moreover, this strain field controls most of the evolution of the shear strength of the fault. We then propose plausible comparisons of our experimental results to geological observations of fault cores in the region of Aigion (Corinth Gulf, Greece). The studied faults indeed display spectacular indurated fault planes lying on weakly cohesive material. Signatures of cementation, clay mineral distribution and porosity profile of one of the studied fault cores are included and discussed in the light of the experimental results. Our observations suggest that the maximum shear strain during earthquakes might occur not in the center, but on the border of the fault cores. It is presumably localized in a transition zone which exhibits a significant cementation owing to a process of mechanical smearing by fine particles. This zone may also act as a very low permeability layer responsible for a channeling of the fluid flow. Such a scheme of progressive multi sub-localizations, is different from classical descriptions of faults and consistent with a layering of the core consisting of separated zones of high strains or large cataclastic flows. [Copyright &y& Elsevier]

We report the results of permeability measurements of fault gouge and tonalitic cataclasite from the fault zone of the Median Tectonic Line, Ohshika, central Japan, carried out during triaxial compression tests. The experiments revealed marked effects of deformation on the permeability of the specimens. Permeability of fault gouge decreases rapidly by about two orders of magnitude during initial loading and continues to decrease slowly during further inelastic deformation. The drop in permeability during initial loading is much smaller for cataclasite than for gouge, followed by abrupt increase upon failure, and the overall change in permeability correlates well with change in volumetric strain, i.e., initial, nearly elastic contraction followed by dilatancy upon the initiation of inelastic deformation towards specimen failure. If cemented cataclasite suffers deformation prior to or during an earthquake, a cataclasite zone may change into a conduit for fluid flow. Fault gouge zones, however, are unlikely to switch to very permeable zones upon the initiation of fault slip. Thus, overall permeability structure of a fault may change abruptly prior to or during earthquakes and during the interseismic period. Fault gouge and cataclasite have internal angles of friction of about 36° and 45°, respectively, as is typical for brittle rocks. [Copyright &y& Elsevier]