Abstract Field characteristics of crustal extrusion zones include: high-grade metamorphism flanked by lower-grade rocks; broadly coeval flanking shear zones with opposing senses of shear; early ductile fabrics successively overprinted by semi-brittle and brittle structures; and localization of strain to give a more extensive deformation history within the extrasion zone relative to the flanking regions. Crustal extrusion, involving a combination of pure and simple shear, is a regular consequence of bulk orogenic thickening and contraction during continental collision. Extrusion can occur in response to different tectonic settings, and need not necessarily imply a driving force linked to mid-crustal channel flow. In most situations, field criteria alone are unlikely to be sufficient to determine the driving causes of extrusion. This is illustrated with examples from the Nanga Parbat–Haramosh Massif in the Pakistan Himalaya, and the Wing Pond Shear Zone in Newfoundland.

Abstract Faults and their deeper-level equivalents, shear zones are localized regions of higher strain which effectively accommodate differential movement in the Earth’s crust and mantle during deformation of the lithosphere. Shear zones may be more precisely defined as approximately tabular regions of concentrated deformation and flow across which adjacent relatively undeformed rock units are offset. They are recognized at all sizes from micro to plate boundary scale (Ramsay 1980; Sornette et al. 1990) ( Figs 1 & 2 ). Faults and shear zones are therefore important examples of the heterogeneous nature of deformation in natural rocks, and profoundly influence the location, architecture and evolution of a broad range of geological phenomena (e.g. Rutter et al. 2001 ). The topography and bathymetry of the Earth’s surface is marked by mountain belts and sedimentary basins which are controlled by faults and shear zones. In addition, faults and shear zones control fluid migration and transport, including hydrothermal fluids and hydrocarbons of economic significance (e.g. McCaig 1997). Magma transport, emplacement and eruption are also frequently controlled by faults and shear zones, as are earthquakes. Once faults and shear zones are established, they are often long-lived features prone to multiple reactivation over very large time-scales (e.g. Holdsworth et al. 1997 ). Fig. 1. Banded orthogneiss with darker amphibolite layers displaying dextral offset across minor shear zones in West Greenland. Note the attenuation of layering, variable displacement and minor melt component along the shears. Pocket knife for scale. Fig. 2. Aeromagnetic map of part of the 180 km long Nordre Strφmfjord shear zone in the Nagssugtoqidian belt of West Greenland. Magnetic patterns clearly define the ENE–WSW-trending shear zones which are marked by pronounced swings in the trend of the regional magnetic signature (see Rasmussen & Van Gool 2000 and references therein). Faults and shear zones are typically arranged into complex interlinked networks that permit 3D strain in response to plate tectonic forces ( Dewey et al. 1986 ). However, analysis of ductile shear zones is complicated as they are only directly accessible to geoscientists after exhumation to the Earth’s surface. In such cases, the relationships between the observed finite deformation patterns, the preserved microstructures at any given location, and the deformation path and strain rate history are potentially difficult to resolve (e.g. Knipe 1989). It is therefore necessary to first consider the bulk deformation behaviour of the lithosphere and the nature and strength of deformed rocks at depth within shear zones.

Abstract There is abundant field and microstructural evidence for localization of deformation in alpine- and ophiolite-type mantle massifs. On the basis of field relationships and microstructures we recognize two types of tectonite shear zones (medium- to coarse- and fine-grained), as well as two types of mylonitic shear zones (anhydrous and hydrous peridotite mylonites). In tectonite shear zones, softening processes responsible for localization are probably melt-related weakening in the medium to coarse tectonites and a change in limiting slip system in the fine-grained tectonites. In peridotite mylonites, the most likely cause for softening and localization is a change in dominant deformation mechanism from dislocation to grain size sensitive creep. Microstructural and petrological study of mylonite rocks reveals that reactions, either continuous net-transfer reactions (anhydrous and hydrous) or melt-rock reactions, play a key role in the formation of fine-grained material that promotes grain size sensitive creep. These reactions occur over a broad range of pressure-temperature conditions encompassing a large part of the lithospheric upper mantle. We conclude that mantle shear zones are widespread and that they reduce the (bulk) strength of the lithosphere significantly.

Abstract Blocks of granulite from within the megabreccia at Clarke Head, Nova Scotia, Canada contain extremely well preserved mylonitic and ultramylonitic textures developed in mineral assemblages for which thermobarometic calculations have indicated temperatures and pressures between 700–860 °C and 750–950 MPa. Deformation within these rocks is characterized by localization at several discrete length scales associated with the development of new microstructures comprising finer-grained material. Mylonitized granulite exhibits dislocation creep microstructures, with development of intense S–C fabrics and shear bands during the transition to ultramylonite. Dynamically recrystallization of plagioclase can be followed through progressive grain size reduction to about 5 μm, but there remain extensive zones with grains less than 1 μm in diameter. Localization of the these finest-grained ultramylonites occurs by transient frictional events associated with the introduction of partial igneous melts and formation of pseudotachylyte which produces abrupt decreases in grain size that cannot arise during dislocation mediated grain size reduction. The heterogeneous response of these rocks demonstrates the importance of considering characteristic length scales when assigning evidence from the rock record (e.g. palaeopiezometry) to bulk behaviour of the lithosphere. Associated with the localization of strain and subsequent strain softening is the observation that microstructures formed during the event that initiated the instability can be an obliterated by ductile flow. In instances where critical components of the microstructural evolution are known to have been largely overprinted, it becomes possible to reconcile contradictions in the rock record, such as production of ultra-fine-grained superplastic aggregates in what otherwise appears to be a dominantly dislocation creep regime.

Abstract SEM/EBSD-based orientation and misorientation analyses are described for a lower amphibolite facies simple shear zone (Torridon, NW Scotland). It is shown that as well as conventional crystal-slip processes (i.e. basal- a , prism- a , rhomb- a and negative second order rhomb- a slip), dauphine twinning also plays a role in both microstructural and petrofabric evolution. Twinning assists in the initial grain size comminution processes, including dynamic recrystallization, from originally coarse wall rock grains to a typical mylonitic microstructure in the centre of the shear zone. Subsequently, twinning helps to accommodate high shear strains in the mylonite whilst maintaining a stable microstructure and constant ‘single crystal’ petrofabric. The role of dauphine twinning appears to be to allow efficient switching between relatively ‘soft’ and relatively ‘hard’ slip directions that possibly exploit a distinction between negative and positive crystal forms. Misorientation analysis emphasizes the relationships between crystal-slip systems and grain boundary network, including dauphine twin planes, and suggests that the mylonitic microstructure contains preferred orientations of both tilt and twist boundaries that help to explain shear zone microstructural evolution and stability.

Abstract The ability to compare, integrate and knit together multidisciplinary datasets in terms of subject, space and scale is critical to the recognition of geological patterns. In this contribution, we show that the use of Geographic Information Systems (GIS) is extremely valuable in detecting patterns associated with broad zones of deformation in high grade terrains. The GIS methodology facilitates the geological interpretation and development of models as it permits an easy and quick investigation of several geoscientific datasets by subject, space and scale. The GIS-based integration of structural, metamorphic, fabric type and aeromagnetic datasets collected in west Greenland shows that patterns seen within one dataset coincide with patterns observed in other datasets. Consequently, two major domains are recognized that are separated by a broad boundary zone. The southern block is characterized by a distinct, irregular magnetic signal coupled with granulite facies metamorphism and dominant S-type fabrics. The map scale geometry of this block controls the patterns observed within the amphibolite facies domain further north. Foliation and lineation patterns form an arcuate swing in strike about the southern block. Fabric types vary both around the strike swing and across strike. An indentor model that incorporates a rigid, cooled granulite block in the south bounded to the north by a rheologically weaker amphibolite facies domain can explain these patterns. The preserved metamorphic grade governs the rheology of the different, but essentially authochthonous blocks with the amphibolite facies domain being plastered and ‘moulded around’ the rigid granulite indentor. As patterns of remote geophysical and geological data closely correspond with one another, greater confidence may be placed in the application of remote geophysics in areas which lack abundant ground-based data.

Abstract A global model is presented to account for the specific rheology of a two-phase material. Examples of observations are taken from a crystallizing magma and these are applied to a partially molten rock, plastic deformation and soil liquefaction. The general behaviour of the viscosity is drawn as a function of the strain rate and the amount of solid phase. It constitutes a 3D diagram developing a cubic surface. The cubic equation is justified by thermodynamic considerations. It results from the mixing of a Newtonian ( n = 1) and a power law ( n = 3) type of deformation. The diagram shows two types of rheological response. At high strain rate values, the viscosity contrast between the two phases is the lowest. The resulting en masse behaviour is observed during tectonic activity. It manifests itself by homogeneous transport of magma during emplacement and fabric development. An equivalent medium, with average viscosity is a good proxy. Conversely, at low strain rate values, the viscosity contrast between the two phases is the highest. The two end members behave according to their respective rheology. In between, a transitional state develops, in which instability occurs depending on the strain rate and stress conditions. In the 3D diagram it appears as a cusp shape. Rheology presents continuous jumps between the liquid-like and the solid-like rheology. They result in strain localization or phase segregation. The latter preferentially develops during magma crystallization. Deformation under a constant amount of each phase is also possible, resulting in pressure dissolution-like processes. A bifurcation in the solution plane of the equation of viscous motion causes instability. It is comparable with strain softening. A similar situation should develop when mixing Newtonian and power law rheology, for example during diffusion and dislocation creep, or water-saturated sediment deformation. Owing to the continual jumps between the two types of rheology, hysteresis or memory effect may develop. Rapid cyclic deformation may drive strain to extreme straining. The effect of simple shear seems much more effective than pure shear (compaction) to separate the weak phase from its strong matrix. The development of instabilities and continuous jumps from one rheology to the other lead to discontinuous motion of the weak phase. In a molten region, it corresponds to discontinuous bursts of magma that are extracted.

Abstract During the past 18 Ma extensional tectonism has migrated from the Tyrrhenian sea eastward into the Northern Apennines of Italy. The extension is due in part to lowangle east-dipping normal faults, that are now exhumed in the Tyrrhenian islands and Tuscany, while additional extension is still occurring in the Apennine chain (Umbria region, c. 200 km eastward). This tectonic framework is an example where active extensional processes affecting the Umbria region can be studied in exhumed faults that are no longer active. Here a comparison between the Zuccale Fault (ZF), cropping out in the Isle of Elba, and the Altotiberina Fault (ATF), revealed by geophysical data, seismology and seismic profiles crossing the Umbria region, provide insights into the processes affecting low-angle normal fault development and evolution. Recorded microseismicity suggests that the ATF is presently active under a vertical σ 1 . Structural analysis of the ZF depict a comparable scenario with fluid involvement during the activity. The comparison of these two structures suggests movements with fluid involvement along gently dipping planes under a vertical σ 1 , implying that these faults are mechanically weak.

Abstract Experimental tests on simulated clay gouges and data from shear zones developed in pelitic media at convergent plate margins provide contrasting evidence regarding the hydraulic characteristics and, in consequence, the frictional properties of sheared clays. The natural shear zone analysed in this work indicates that shear strain can induce mineralogical changes in smectite-bearing sediments that imply loss of water from the smectite minerals and their replacement with anhydrous illite minerals. The extreme localization of the illitization process and its geometric characteristics allow us to argue that the reaction is initiated by stress concentration along the shear zone and, once discrete shears develop, it is accelerated by both cataclasis and the frictional dehydration of smectites. This process would generate fresh water from within the shear zone, leading to fluid overpressure build up, and can account for the observed hydraulic circulation and salinity anomalies in modern accretionary prisms.

Abstract The mega-faults between actively converging plates have recently been penetrated by the Ocean Drilling Program at three plate margins: Barbados, Costa Rica and Nankai. Cores, downhole instrumentation and detailed seismic imagery provide data which may be helpful in interpreting ancient examples of shear zones. The mega-faults, developed in poorly lithified sediments, separate major lithospheric plates yet are merely tens of metres in thickness. They respond to ongoing strain by intensifying inwards rather than propagating outward splays and can grow thinner because of continuing compaction. Surprisingly, lithological influence on the localization of fault propagation seems slight, but lithology determines the deformation style within the faults. The resulting structures show asymmetric distributions within the zones but, in these flat-lying structures, tend to show a downward increase in strain. Upper margins are typically gradational whereas lower boundaries can be strikingly abrupt. The fluid-transport behaviours are complex. In some situations the horizontal flux is very diffuse but centred around the fault. Some faults can efficiently channelize fluids — for distances of tens of kilometres — while at the same curbing flow across them. The fluid transport is clearly episodic and heterogeneous. Fingers of pressured fluid migrate within the fault zone, in patterns that constantly change through time.

Abstract Geochemical data from three profiles crossing the West Fissure Zone in northern Chile were used to describe the chemical effects of fluids on faulting processes. The results document considerable differences of fluid–rock interactions between the profiles (A–C). Within the area of profile A and C fluid activities have neither led to intensive exchange reactions between fluids and rocks nor to notable changes in whole rock and mineral composition of fault rocks relative to their undeformed host rock. Investigations of stable isotopes (δ 13 C, δ 18 O) and fluid inclusions indicate infiltration of predominantly descending (meteoric) fluids and only subordinate involvement of ascending hydrothermal fluids. In the case of profile B, fluid-enhanced weakening mechanisms are dominant. Dissolution transfer has led to the formation of an alteration zone up to 400 m wide. Along profile B we have no indications for warm fluids derived from a deep source. Rather, a number of lines of evidence (e.g. δ 18 O and δD) show that fault rock alteration took place due to the infiltration of low temperature meteoric water. For this profile, all geochemical data indicate an open fluid system. Moreover, the chemical variations across the West Fissure Zone agree well with the structural variability observed at this traverse. The application of a variety of geochemical analyses of fault rocks shows the heterogeneity of large-scale continental fault zones.

Abstract On the island of Sikinos in the Cyclades a schistose carapace separates a marble-blueschist cover sequence from underlying basement rocks. The basement ‘core’, comprising metapelitic gneisses, biotite-bearing granodiorites and aplites, becomes increasingly strained towards the carapace with progressive obliteration of earlier structures and intensification of a mylonitic foliation that becomes pervasive within the carapace. Granodiorites in the ‘core’ can be traced into microcline schists within the carapace, whereas metapelitic gneisses are converted to garnet–mica schists. The carapace is therefore a simple shear zone comprising basement rocks mylonitized during overthrusting of the cover. Biotite clusters in granodiorites of the basement ‘core’ are partially altered to phengite, whereas plagioclase shows incipient sericitization. In more strained rocks these hydration reactions show enhanced progress, until biotite and plagioclase are finally eliminated in the carapace. Rare glaucophane and chloritoid inclusions within garnets of metapelitic gneisses adjacent to the carapace are also attributed to hydration reactions. The association of higher strain with increased hydration in the basement suggests localization of fluids in the strained carapace zone, with limited percolation into underlying rocks. The restricted availability of water outside the carapace may be responsible for preservation of pre-Alpine assemblages in large parts of the Cycladic basement.

Abstract Deformation within shear zones can be both temporally and spatially variable, resulting in multiple generations of folds which display a range of scales and overprinting relationships in mylonitic rocks associated with high strain zones. Despite such complexities, two main fold associations are broadly recognized in many shear zone settings: early tight to isoclinal sheath folds, often with mylonitic limbs that are post-dated by one or more local generations of synshearing folds which are preserved within, or root downwards into mylonitic high strain zones. These latter structures locally fold the mylonitic foliation and lineation whilst displaying geometric characteristics that are kinematically compatible with the movement regime of the major shear zone. Using examples related to ductile thrusting in Moine metasediments of north Scotland, we show that both types of fold display predictable geometric patterns on fabric topology plots. Fold axes and axial surfaces display consistent changes in asymmetry and sense of obliquity relative to local, transport-parallel mineral lineations that can be used to map out a series of culminations and depression zones. The sheath folds preserve more acute, but almost identical geometric patterns compared to the later synshearing folds, with culmination and depression zones often coinciding in location and scale. Detailed analysis also demonstrates that the distribution of finite strain is systematically linked to the architecture of all folds and that clear and predictable relationships exist between the fabric topologies of both the sheath folds and synshearing folds. These consistent topological relationships could be explained in terms of a fold evolution model , where sheath folds represent a more highly deformed and evolved variety of synshearing folds originally generated during perturbations in ductile flow. However, an alternative fold inheritance model predicts that the gross structural architecture generated during sheath folding may subsequently control the geometry and govern the orientation of the synshearing folds. Both models may be widely applicable in a broad range of shear zone environments.

Abstract The Minas fault system is an ENE-WSW trending transpressional boundary between the Appalachian Meguma and Avalon tectono-stratigraphic terranes of Nova Scotia, Canada. Along this boundary there is large-scale partitioning of deformation into distinct external (contractional) and internal (shear) zones. With the increase in strain from external to internal zones there is progressive localization of deformation, culminating in the discrete shear band domain. Deformation has produced materially, temporally and spatially distinct folds and faults throughout the fault system history. Ductile structures are generally composite features derived from multiple transposition of pre-existing layers. The partitioning of deformation found amongst fault rock units can in turn be associated with contrasting deformation micromechanisms. The distinctive variation in mechanical response and microstructures provides an insight into the role of localization, partitioning and distribution of deformation. Kinematic analysis has demonstrated that the Minas fault system segment examined here is a thinning deformation zone, in which strain is accommodated within progressively narrower volumes of rock. Deformation can be summarized as a broad, initially diffuse zone of triclinic transpression that has evolved, with the accumulation of finite strain, into zones of distinct structural style and variation in finite strain. It is not possible to demarcate ‘deformed shear zone’ and ‘undeformed host rocks’. Instead, the Minas fault system is described in terms of discontinuous transitions in finite strain and deformation style within a large scale movement picture.

Abstract The Coimbra–Cordoba shear zone (Iberian Massif), characterized by simple-shear dominated sinistral transpression, exposes several outcrops of strongly sheared peralkaline gneisses surrounded by mica schists and amphibolites. These gneisses are included in the Arronches Tectonic Unit, a thick unit of mylonitic rocks with a steep foliation and an associated gently plunging stretching lineation parallel to the fold axes. Strain partitioning is testified by widely spaced anastomosing shear bands around less-strained domains and by the existence of different shearing domains ranging from relatively ‘less-strained’ and coarse-grained mylonites to highly strained and fine-grained ultramylonites. Three shearing domains defined by textural and structural changes resulted from progressive deformation and increasing strain, which leads to increased mylonitization of gneisses. This is revealed by the increased modal percentage of the matrix and the decreased percentage of porphyroclasts, accompanied by evolution from orthorhombic to monoclinic fabrics: Conjugate Shearing Domain ( CSD ), Intermediate Sinistral Domain ( ISD ), and Sinistral Domain ( SD ). This contribution shows that in a simple-shear sinistral dominated transpression zone with a well-developed and widespread monoclinic fabric, it is possible to find mechanical conditions to produce local orthorhombic fabrics. In the Arronches gneisses a local strain regime exists in apparent contradiction with the bulk deformation regime.

Abstract Detailed mapping of four areas representing different geological units with varying formation histories within the crustal-scale Errabiddy Shear Zone shows an apparently simple temporal progression from foliation and mineral lineation development to folding and then to brittle deformation across the shear zone. However, in detail the structural evolution of the shear zone shows considerable complexity. The dominant foliation throughout the shear zone was formed in the upper greenschist to amphibolite facies during the 2000–1960 Ma Glenburgh Orogeny, which involved the accretion of the Archaean to Palaeoproterozoic Glenburgh Terrane onto the Archaean Yilgarn Craton and the subsequent formation of the Errabiddy Shear Zone. Orthorhombic kinematic indicators formed during the Glenburgh Orogeny as did the widespread mineral lineation. These fabrics were overprinted by a greenschist facies deformation and metamorphic event during the 1830–1780 Ma Capricorn Orogeny. During the Capricorn Orogeny mineral lineation development was rare, and mostly took place in high-Capricorn strain zones in areas where a pre-existing Glenburgh-aged mineral lineation was present. Such mineral lineations trend parallel to Capricorn-aged fold hinges. Regardless of the presence or absence of Capricornaged mineral lineations, dextral strike-slip kinematics and simple shear indicated by delta and sigma porphyroclasts, and displacement along detachment faults, are prevalent close to discrete shear zone boundaries, within the Errabiddy Shear Zone. However, between shear zone boundaries flattening and coaxial strain dominated during the Capricorn Orogeny. This difference in Capricorn Orogeny kinematics throughout the shear zone is caused by strain partitioning – although progressive deformation throughout the shear zone with dextral strike-slip faults overprinting older structures formed by pure shear also took place. These results suggest that analyses of small parts of shear zones may not give the complete history of an evolving transpressional shear zone because of the presence of strain partitioning and strain localization over time.

Abstract Strain and vorticity analysis of two Late Palaeozoic high-strain zones from the southern Appalachian Piedmont indicates that these zones experienced general shear transpression with a monoclinic to triclinic symmetry. Granitic rocks in the Brookneal high-strain zone from the southwestern Virginia Piedmont were transformed into mylonites under greenschist facies conditions. Sectional strains, estimated from quartz grain shapes, in mylonites range from three to ten and three-dimensional fabrics record flattening strains. The mean vorticity number ( W m ) estimated with the R s /θ method ranges from 0.3 to 0.95. In the central Virginia Piedmont, lower amphibolite facies deformation in the Spotsylvania high-strain zone affected biotite gneisses, amphibolites, and granitic pegmatites. Minimum sectional strains, estimated from folded and boudinaged pegmatite dykes, of 8–20 are common and three-dimensional strains are dominantly constrictional. Porphyroclast hyperbolic distribution analysis of ultramylonites yields W n values from 0.4 to 0.8. The kinematic significance of these transpressional high-strain zones is threefold: they record tens to hundreds of kilometres of strike-slip offset; 40 to 70% contraction normal to the zone; and significant orogen-parallel material elongation.

Abstract We develop a method for constraining the kinematics and finite strain of deformation in shear zones based on a three-dimensional numerical model of the rotation populations of rigid clasts. The results of the model are characterized in terms of a fabric ellipsoid, which is directly measurable from field data. Fabric ellipsoids measured from populations of prolate clasts have anisotropies that increase steadily and plateau; the shape of the fabric ellipsoid becomes increasingly more prolate with progressive deformation. The behaviour of populations of oblate clasts is much more complex because the stability of individual oblate clasts depends on their aspect ratio and the vorticity of deformation. Populations of oblate clasts may produce fabric ellipsoids with oscillating anisotropies and shapes if their aspect ratio is low enough for a continuous rotation. For either prolate or oblate clasts, the maximum anisotropy that a fabric ellipsoid will reach is governed by the aspect ratio of the individual clasts of that population. The theoretical maximum anisotropy is achieved when all of the clasts are perfectly aligned. The shape of the fabric ellipsoid, in conjuncture with the anisotropy, can be used to constrain the vorticity and finite strain of deformation. The numerical model suggests that there is no consistent relationship between the asymmetrical orientation of a population of rigid markers and the simple shear component of deformation. Therefore, the asymmetrical alignment of a population of porphyroclasts is not a reliable shear sense indicator. Additionally, there is no direct correlation between the fabric ellipsoid and the strain ellipsoid. Model results are applied to shape preferred orientation data collected from a feldspar megacrystic granite in the western Idaho shear zone (USA). Three-dimensional fabric ellipsoids are calculated from two-dimensional sectional measurements of oblate-shaped, unmantled, potassium feldspar porphyroclasts. Comparison of these data with the results of the numerical model suggests that transpressional deformation had an intermediate angle of oblique convergence (30°–60°). This implies that deformation in the western Idaho shear zone was characterized by a large component of convergent motion.

Abstract Central Brittany comprises a major shear belt formed during the Hercynian Orogeny. In this paper, we present a model for the restoration of the belt, based on an exhaustive compilation of available structural data. The model uses a geostatistical analysis of cleavage orientation data. The analysis shows that the eastern part of Central Brittany has been deformed by bulk dextral strike-slip, a feature that validates previous structural interpretations. It also provides explanations for: the strain patterns observed within and around the restored domain; the nature of domain boundaries and of associated kinematics; and the occurrence of local superimposed deformations.