Abstract The correct determination of the kinematic rotation axis in high-strain zones is essential to the study of the tectonic evolution of the Earth's crust. However, the common assumption that the kinematic rotation axis lies orthogonal to the XZ plane of the finite strain ellipse may be invalid in the case of general shear. Orientation data obtained by electron backscatter diffraction from calcite, deformed in the high-strain Gressoney Shear Zone of the Western Alps, has been investigated using orientation maps, bulk sample crystallographic orientation and misorientation analyses, and detailed intragrain misorientation and crystallographic dispersion analysis. The results demonstrate a strong geometrical coincidence amongst (1) the bulk macroscopic kinematic rotation axis, (2) the orientation of misorientation axes associated with low-angle boundaries, and (3) rotation axes associated with crystallographic dispersion at the intragrain scale. This coincidence is interpreted to reflect a geometric control of the kinematic framework of the high-strain zone on the activity of crystal slip systems. It is proposed that this relationship may be exploited as a new microstructural tool to determine the orientation of bulk kinematic rotation axes in high-strain zones without assuming a geometric link between kinematic rotation and XZ sections. Although further testing is required, application of the approach may lead to a significant advance in our understanding of natural general shear deformation.

Abstract Fracture and vein patterns in the brittle crust of the Earth contain information on the stress and strain field during deformation. Natural examples of fracture and vein patterns can have complex geometries including combinations of extension and conjugate shear fractures. Examples are conjugate joint systems that are oriented with a small angle to the principal stress axis and veins that show an oblique opening direction. We developed a discrete numerical model within the modelling environment ‘Elle’ to study the progressive development of fractures in two dimensions. Results show that pure shear deformation alone can produce complex patterns with combinations of extension and shear fractures. These patterns change in geometry and spacing depending on the Young's modulus of the deforming aggregate and the initial noise in the system. A complex deformation history, including primary uniaxial loading of the aggregate that is followed by tectonic strain, leads to conjugate shear fractures. During progressive deformation these conjugate shear fractures may accommodate extensional strain or may be followed by a secondary set of extension fractures. The numerical patterns are consistent with joint, fault and vein geometries found in natural examples. The study suggests that fracture patterns can record complex deformation histories that include primary uniaxial loading due to an overlying rock sequence followed by tectonic strain.

Abstract Fault segments belonging to a fault population can link and interact, eventually forming a single larger fault, and thus affecting the estimation of the maximum expected earthquake. We present throw distribution data along the Quaternary normal faults of the Colfiorito fault system (central Italy), which consists of four main fault segments and where a seismic sequence occurred in 1997–1998. Throw values along the two central overlapping, en-échelon segments (8.5–9.5 km long) were measured on a good stratigraphic marker, by constructing a set of closely spaced geological cross-sections, perpendicular to the fault strike. As these faults are commonly retained active and border Quaternary basins, we compare morphological and geological throws in order to verify the faults neotectonic activity. Geological and morphological throw distributions show good correlation, testifying that recent faulting affected the topographic surface and suggesting that the observed offset completely accumulated during the Quaternary. The throw distribution along the fault segments is asymmetric and reaches maximum values (500–550 m) within the zone of fault overlap, suggesting mechanical interaction between the studied faults. Maximum length-throw correlation suggests that the studied faults grew according to a linear scaling relationship.

Abstract Textural controls on the peak strength of dolomite are investigated through 23 triaxial deformation experiments, performed at confining pressures of 25, 50, and 100 MPa on texturally diverse dolomites. The mechanical data from these experiments are fit to an empirical failure criterion to elucidate the most significant parameters in dictating the peak strength of dolomite. Neither grain size nor porosity is required to quantitatively predict the peak strength of dolomite. Instead, only the effective Young's modulus, and the empirically predicted uniaxial compressive strength, along with experimentally controlled confining pressure, are required in order to explain the peak strength to and R 2 better than 0.89. A Hall-Petch relationship does not apply to this data set as a consequence of the variability in intragranular and grain boundary textures, which appear to overshadow the role of grain size. It is, therefore, essential that grain boundary textures and intragranular flaws be examined prior to making predictions regarding the relative peak strengths of chemically and mineralogically similar dolomites.

Abstract To investigate the role of texture on the brittle deformation of dolomite, 23 triaxial deformation experiments were performed at confining pressures of 25, 50, and 100 MPa, dry, at room temperature, on dolomite from three texturally distinct sample suites. The variations in the mechanical response of these mineralogically and chemically similar dolomites, and the ensuing microstructures, indicate that grain boundary textures promote or inhibit the ability of grains to shear and rotate with respect to one another, whereas the presence of intragranular flaws, such as cleavage, that act as weaknesses, promote intragranular deformation. In samples with porosities greater than c. 7%, inelastic pore collapse controls the transition from brittle faulting to extensive intragranular deformation and cataclastic flow. This porosity is much higher than has been observed at the onset of pore collapse in calcite, as a consequence of the inability of dolomite to deform by crystal plastic processes at room temperature. Combined, these textural features may dictate the transition from brittle faulting to cataclastic flow in brittle rocks in the upper crust.

Abstract Statistical properties of crack-seal veins are investigated with a view to assessing stress release fluctuations in crustal rocks. Crack-seal patterns correspond to sets of successive parallel fractures that are assumed to have propagated by a subcritical crack mechanism in the presence of a reactive fluid. They represent a time-sequence record of an aseismic and anelastic process of rock deformation. The statistical characteristics of several crack-seal patterns containing several hundreds of successive cracks have been studied. Samples were collected in three different areas, gold-bearing quartz veins from Abitibi in Canada, serpentine veins from the San Andreas system in California and calcite veins from the Apennine Mountains in Italy. Digitized pictures acquired from thin sections allow accurate measurement of crack-seal growth increments. All the samples show the same statistical behaviour regardless of their geological origin. The crack-seal statistical properties are described by an exponential distribution with a characteristic length scale and do not show any spatial correlation. They differ from other fracture patterns, such as earthquake data, which exhibit power-law correlations (Gutenberg-Richter relationship). Crack-seal series represent a natural fossil record of stress release variations (less than 50 bars) in the crust that show a characteristic length scale, associated with the resistance of rock to effective tension, and no correlation in time.

Abstract Pressure solution creep rates and interface structures have been measured by two methods on calcite single crystals. In the first kind of experiments, calcite monocrystals were indented at 40 °C for six weeks using ceramic indenters under stresses in the 50–200 MPa range in a saturated solution of calcite and in a calcite-saturated aqueous solution of NH 4 Cl. The deformation (depth of the hole below the indenter) is measured ex situ at the end of the experiment. In the second type of experiment, calcite monocrystals were indented by spherical glass indenters for 200 hours under stresses in the 0–100 MPa range at room temperature in a saturated aqueous solution of calcite. The displacement of the indenter was continuously recorded using a specially constructed differential dilatometer. The experiments conducted in a calcite-saturated aqueous solution of NH 4 Cl show an enhanced indentation rate owing to the fairly high solubility of calcite in this solution. In contrast, the experiments conducted in a calcite-saturated aqueous solution show moderate indentation rate and the dry control experiments did not show any measurable deformation. The rate of calcite indentation is found to be inversely proportional to the indenter diameter, thus indicating that the process is diffusion-controlled. The microcracks in the dissolution region under the indenter dramatically enhance the rate of calcite indentation by a significant reduction of the distance of solute transport in the trapped fluid phase. This result indicates that care should be taken in extrapolating the kinetic data of pressure solution creep from one mineral to another.

Abstract Deformation mechanisms of amphibole and plagioclase were investigated in two metagabbroic sheets (the eastern and western metagabbros from the Staré Město belt, eastern Bohemian Massif), using petrology, quantitative microstructural and electron back-scattered diffraction methods. After the gabbroic pyroxene was replaced by amphibole, both gabbroic bodies became progressively deformed. The eastern metagabbros were deformed under temperature of c. 650 °C and the western metagabbros under c. 750 °C. Subgrain rotation and dislocation creep, characterized by strong crystallographic and shape preferred orientations, operated in plagioclase of the eastern belt during the early stages of deformation. Subsequent randomizing of plagioclase crystallographic preferred orientation is interpreted to be due to grain boundary sliding in the mylonitic stage. Large (50–150 μm) grain sizes during the mylonitic stages are interpreted to be due to low strain rates. Amphibole is stronger and deforms cataclastically, leading to important grain size reduction when the bulk rock strength drops substantially. In the western belt, plagioclase deformed by dislocation creep accompanied by grain boundary migration (possibly chemically induced) while heterogeneous nucleation and syndeformational grain growth in conjunction with dislocation creep were typical for amphiboles.

Abstract The quantification of quartz shear stress and strain rate within a midcrustal shear zone provides a mechanical frame to describe the evolution from penetrative ductile deformation to localized deformation and the onset of brittle deformation. The quantification is based on the relationships between the quartz recrystallized grain size, the quartz shear stress (piezometric relation) and the strain rate (dislocation creep flow law). Increasing strain is accompanied by a general decrease of quartz recrystallized grain size and a decrease in grain size scattering. These are interpreted as a result of a complex loading history. The evolution from penetrative ductile deformation toward strain localization, marked by an increase of the strain rate by one order of magnitude, is inferred from grain size memory. Brittle deformation is triggered for quartz shear stress of the order of 70 MPa and strain rate close to 10 −12 s −1 . This relative low value of the quartz shear stress necessary to trigger faulting implies a less important strength for the midcrust compared with strengths predicted by classical rheological envelopes.

Abstract The effects of dynamic recrystallization on the deformation mechanisms and rheology of olivine aggregates in the laboratory and the lithosphere are reviewed in this paper. The low-strain rheology of olivine is well documented; however, deformation in the lithosphere often involves large strains. Large strain experiments show that recrystallization can result in both hardening and softening during deformation. Moderate strain softening in experimental shear and torsion can be explained by the operation of dislocation-accommodated grain boundary sliding in bands of fine recrystallized grains. Data on the temperature dependence of recrystallized grain size are needed to extrapolate the effects of dynamic recrystallization to the lithosphere. Theories of dynamic recrystallization suggest that grain size is strongly stress dependent and moderately temperature dependent. A re-analysis of experimental grain size data indicates that the recrystallized grain size is temperature independent for olivine aggregates with low water content (<300 ppm H/Si). Rheological regime maps have been constructed for the lithospheric mantle. The maps suggest that grain size sensitive power law creep, involving both grain boundary sliding and dislocation creep, will produce strong strain softening, greater than found so far in experimental studies, in dry and wet lithosphere shear zones.

Abstract Some exhumed complexes in collisional belts consist of continental basement containing slivers of mafic and ultramafic material showing evidence of UHP metamorphism ( P c. 3 GPa). Their PTt history can be interpreted in terms of subduction of continental material to depths ≥ 100 km and subsequent exhumation. This type of tectonic history is illustrated by the Late Palaeozoic evolution of the Ulten Unit, Tonale Nappe, Eastern Austroalpine. The upper crustal felsic component ( c. 80% by volume) incorporated mafic material at the trench, and peridotitic material at deeper levels in the subduction zone. The peridotites show evidence of a P -increasing, T -decreasing path before incorporation in the felsic material, compatible with flow in the mantle wedge above the subducting slab. After emplacement of the peridotites, which occurred at or near peak metamorphic conditions ( P ≥ 2.7 GPa, T ≥ 850 °C), the complex underwent a two-stage pre-Alpine exhumation path: a first, fast stage ( c. 0.1–1 cm a −1 ), lasting c. 30 Ma and bringing rocks from depths ≥ 100 km to approximately 25 km; and a second, slow stage ( c. 0.01–0.1 cm a −1 ), lasting c. 100 Ma and bringing rocks to depths <20 km. The subduction of felsic material to the required depths can be modelled by analysing the time-evolution of negative buoyancy, which confirms that relatively light continental upper crust can be subducted to depths > 200 km if attached to a mature oceanic slab that does not break-off during the early stages of continental subduction. The first exhumation stage can be accounted for by buoyancy-driven tectonic extrusion of continental slices along the subduction channel during continuing subduction. A force balance analysis shows that such a mechanism is compatible with the rheology of felsic and intermediate rocks at high temperature. The second exhumation stage is compatible with isostatic rebound and tectonic denudation following slab break-off. The conclusion that fast exhumation occurs during continuing subduction and before slab break-off is in accordance with the observed rates, which show fast movement of the rising slices with respect to the surrounding material. Slab break-off, on the other hand, generates a long-wavelength gentle upwarping of the overlying region, which is more compatible with later and slower exhumation rates.

Abstract The northwestern part of Holsnøy island, in the Bergen Arcs, Norway, consists of a granulite-facies protolith partially transformed at depth in eclogite (700 °C, > 19 kbars) and amphibolite (650 °C, 8–10 kbars) facies during the Caledonian orogenesis. Eclogitized zones are mainly planar objects (fractures with parallel reaction bands and cm-to-100 m-scale shear zones). Eclogitic zones are distributed in two sets of orientations and the associated deformation can be described as ‘bookshelf tectonics’. The major shear zones strike around N120 and dip to the North, and show consistent top-to-the-NE shear sense throughout the area. In the large-scale kinematic frame of Caledonian NW-dipping slab, eclogitic shear zones are interpreted as the way to detach crustal units from the subducting slab and to prevent their further sinking. As the retrograde amphibolitic deformation pattern is similar to the eclogitic one, the detached crustal units started their way up along these eclogitic shear zones. Radiometric ages of eclogitic and amphibolitic metamorphism and their comparison with the chronology of Caledonian orogenesis show that the deformation recorded on Holsnøy occurred in a convergent context. The mechanism we propose can thus account for the first steps of exhumation during collision.

Abstract In the Vannes and St. Nazaire regions of the central part of the Domaine Sud-Armoricain, Variscan belt, western France, the lowermost tectonic unit is exposed as structural culminations composed of supracrustal migmatites that were deformed and metamorphosed along a multistep clockwise P-T path during Carboniferous time; peak P-T was around 800 °C at 9 kbar. The strain field that emerged under subsolidus conditions during prograde metamorphism controlled the initial distribution of granite melt produced by suprasolidus mica breakdown; the limited retrograde reaction of peritectic garnet indicates that melt loss occurred around the metamorphic peak. A second episode of melt production occurred during the retrograde evolution due to a decompression event that led to interconnection of melt in a mesoscale network of deformation bands and formation of ductile opening-mode fractures, as evidenced by layer-parallel and transverse leucosomes linked with petrographic continuity to granite in dykes. The preservation of peritectic cordierite with only limited associated leucosome and the occurrence of pucker structures without leucosome both indicate that melt loss occurred during the second event. Dykes vary from centimetric (common) to hundreds of metres in width (rare), and exhibit scale-invariance over a limited range of measurements; larger dykes are inferred to have fed upper crustal plutons. Melt extraction may have been a self-organized critical phenomenon, but this remains to be demonstrated satisfactorily in nature. Fugitive melt was trapped in the vicinity of the brittle—ductile transition zone and emplaced laterally along horizons reactivated as extensional detachments. A feedback relation is postulated between dextral transtensive deformation, decompression melting and lower crustal doming, and between dome amplification, melt extraction and emplacement in developing extensional detachments and core complex formation.

Abstract In the Vannes and St. Nazaire regions of the central part of the Domaine Sud-Armoricain, Variscan belt, western France, the lowermost tectonic unit is exposed as structural culminations composed of supracrustal migmatites that were deformed and metamorphosed along a multistep clockwise P-T path during Carboniferous time; peak P-T was around 800 °C at 9 kbar. The strain field that emerged under subsolidus conditions during prograde metamorphism controlled the initial distribution of granite melt produced by suprasolidus mica breakdown; the limited retrograde reaction of peritectic garnet indicates that melt loss occurred around the metamorphic peak. A second episode of melt production occurred during the retrograde evolution due to a decompression event that led to interconnection of melt in a mesoscale network of deformation bands and formation of ductile opening-mode fractures, as evidenced by layer-parallel and transverse leucosomes linked with petrographic continuity to granite in dykes. The preservation of peritectic cordierite with only limited associated leucosome and the occurrence of pucker structures without leucosome both indicate that melt loss occurred during the second event. Dykes vary from centimetric (common) to hundreds of metres in width (rare), and exhibit scale-invariance over a limited range of measurements; larger dykes are inferred to have fed upper crustal plutons. Melt extraction may have been a self-organized critical phenomenon, but this remains to be demonstrated satisfactorily in nature. Fugitive melt was trapped in the vicinity of the brittle—ductile transition zone and emplaced laterally along horizons reactivated as extensional detachments. A feedback relation is postulated between dextral transtensive deformation, decompression melting and lower crustal doming, and between dome amplification, melt extraction and emplacement in developing extensional detachments and core complex formation.

Abstract A correlation procedure of scattered tectonic and metamorphic imprints in the reactivated crust is elaborated from recent analytical work in three Alpine metamorphic complexes. It consists of: interpretation of the time-sequence of tectonic fabrics and test of their kinematic coherence; determination of paragenetic compatibility among the mineralogical support of mesoscopic fabrics; cross-validation of mineral transformation over-prints; construction of P-T-d-t paths using a time-sequence of parageneses. The representation of structural and metamorphic information conveys the full tectono-metamorphic history on maps displaying combined tectonic and metamorphic effects. Shape and size definition of metamorphic units, now individuated mainly using their lithological homogeneity and dominant metamorphic imprint, is improved. The analysis of interaction between fabric and metamorphic imprint distributions, proposed in three Alpine examples, shows that the dominant metamorphic imprint does not coincide with T max - P T max of each inferred P-T-d-t loop; the dominant metamorphic imprint is that given by the mineralogical support of the most pervasive fabric. Different metamorphic imprints may dominate in adjacent areas of a single tectono-metamorphic unit (TMU), or equivalent metamorphic imprints may dominate in different TMUs. Therefore, lithostratigraphic setting and dominant metamorphic imprint are inefficient to contour TMUs in terrains with polyphase deformation and metamorphism, without considering multiscale heterogeneity of superposed synmetamorphic fabrics.

Abstract The Periadriatic fault system (PFS) is an array of late orogenic faults (35-15 Ma) in the retro-wedge of the Alpine orogen that accommodated dextral transpression during oblique indentation by the southern Alpine crust. Decoupling along the leading edges of the southern Alpine indenter occurred where inherited lithological and rheological contrasts were accentuated by lateral thermal gradients during emplacement of the warm orogenic retro-wedge next to the cold indenter. In contrast, decoupling within the core and retro-wedge of the orogen occurred in a network of folds and mylonitic faults. In the Eastern Alps, this network comprises conjugate sets of upright, constrictional folds, strike-slip faults and low-angle normal faults that accommodated nearly coaxial NNE-SSW shortening and E-W extensional exhumation of the Tauern thermal dome. The dextral shear component of oblique convergence was taken up by a discrete, brittle fault parallel to the indenter surface. In the Central and Western Alps, a steep mylonitic backthrust, upright folds, and low-angle normal faults effected transpressional exhumation of the Lepontine thermal dome. Mylonitic thrusting and dextral strike-slip shearing along the steep indenter surface are transitional along strike to low-angle normal faults that accommodated extension at the western termination of the PFS. The areal distribution of poles to mylonitic foliation and stretching lineation of these networked structures is related to the local shape and orientation of the southern Alpine indenter surface, supporting the interpretation of this surface as the macroscopic shearing plane for all mylonitic segments of the PFS. We propose that mylonitic faults nucleate as viscous instabilities induced by cooling, or more often, by folding and progressive rotation of pre-existing foliations into orientations that are optimal for simple shearing parallel to the eigenvectors of flow. The mechanical anisotropy of the viscous continental crust makes it a preferred site of decoupling and weakening. Networking of folds and mylonitic fault zones allow the viscous crust to maintain strain compatibility between the stronger brittle crust and upper mantle, while transmitting plate forces through the lithosphere. Decoupling within the continental lithosphere is therefore governed by the symmetry and kinematics of strain partitioning at, and below, the brittle-to-viscous transition.

Abstract Lithospheric-scale analogue experiments have been conducted to investigate the influence of strength heterogeneities on the distribution and mode of crustal-scale deformation, on the resulting geometry of the deformed area, and on its topographic expression. Strength heterogeneities were incorporated by varying the strength of the crust and upper mantle analogue layers and by implementing a weak plate or part-of-a-plate between two stronger ones. Three (brittle crust/viscous crust/strong viscous upper mantle) and four (brittle crust/viscous crust/brittle upper mantle/strong viscous upper mantle) layer models were confined by a weak silicone layer on one side in order to contain but not oppose lateral extrusion. Experimental results show that relative strength contrasts between converging plates and intervening weak plates control the location and the shape of deformation sites taken as ‘collision orogens’. If the contrast is small, internal deformation of the strong plates through fore- and backthrusting occurs early in the deformation history. However, the bulk system is dominated by buckling that nucleates on the weak plate whose antiformal topography is highest; model Moho of the bordering stronger plates is deepest under these conditions. If the contrast is large, deformation remains localized within the weak plate for a larger amount of shortening and develops a root zone below a narrow deformation belt, which coincides with the locus of maximum topography. Implementing a buoyant, low-viscosity layer above the model Moho of the weak plate favours the development of asymmetric model orogens notwithstanding the initial symmetric setup. Once the asymmetry is established strain remains localized in thrust faults and ductile shear zones documenting foreland directed displacement of the model orogen. Such laterally and vertically irregular configurations have applications in continent-continent collision settings such as the Eastern Alps. First-order mechanical boundary conditions recognized from modelling to be favourable to the post early Oligocene tectonics of the Eastern Alps include: (1) subtle rather than high-strength contrasts between the Adriatic indentor and the strongly deformed region comprising Penninic and Austroalpine units to the north of it; (2) decoupling of Penninic continental upper crust from its substratum to allow for crustal-scale buckling of the Tauern Window; (3) weak mechanical behaviour of the European lower crust during collision to account for its constant thickness along the TRANSALP deep seismic transect; and (4) the direct continuation of the basal detachment underlying the fold and thrust belt in the hangingwall of the European plate with a wide ductile shear zone in the core of the orogen, which separates the European from the Adriatic plate. The mechanical properties of the continental lithosphere are non-uniform in space and time ( Ranalli 1997 ). This heterogeneity is primarily due to changes in composition and thermal conditions expressed in the rheological stratification of the lithosphere with the Moho being the most important discontinuity (e.g. Ranally & Murphy 1987). Laterally, the rheology of the continental lithosphere may be modified because of tectonics, leading for example to the separation or collision of continents. Such processes may result in changes of composition (e.g. continental next to oceanic rheology) and lithospheric thicknesses, both in compression as well as extension, and are usually associated with a pronounced thermal perturbation, which influences the strength of the lithosphere transiently. In that way, the thermo-mechanical age of the lithosphere is reset, which emphasizes the strong time-dependence of lithospheric strength (Cloetingh & Burov 1996). Additionally, the lithosphere is affected by faulting and shearing producing a number of metastable rheological discontinuities that are prone to reactivation (Ranalli 2000). Subsequent deformation of the lithosphere will be steered by pre-existing lateral strength variations (e.g. Ziegler et al. 1998) with relative strength differences among deforming minerals (Handy 1990), rock layers (Hudleston & Lan 1993), crustal-scale layers (Gerbault & Willingshofer 2004), or lithospheric plates (e.g. Molnar & Tapponnier 1975) as controlling factor in terms of strain distribution, structure and style of deformation. Our study focuses on this strength contrast across plate boundaries. In particular, we investigate differences in the structural evolution of collision zones, their deep structure, the relationship to higher-level deformations and the resultant topographic expression for conditions of continental convergence as a function of the relative strength contrast of the colliding plates. For this purpose we use a fully mechanical approach, namely lithospheric-scale analogue modelling, which is not restricted by the amount of imposed strain and allows incorporating lateral material transfer. We subsequently discuss implications of our modelling results on aspects of the tectonic evolution of the Eastern Alps in Europe, from where a wealth of surface and subsurface data allow constraining the large-scale geometry of the mountain range as well as its evolution through time.

Abstract We interpret the strain and stress fields of the western/central Alpine arc on the basis of 2.5D finite element modelling and a recent seismotectonic synthesis. Models have fixed boundary forces and different crustal geometries, so that they respond to buoyancy forces (variations in gravitational potential energies). The seismotectonic regime, characterized by orogen-perpendicular extension in the high topographic core of the belt and local orogenperpendicular compressional/transpressional deformation in the external zones, appears to be very close to the modelled gravitational regime. Rotation of Apulia has a minor effect on the current strain or stress fields of the Alpine realm. Nevertheless, it could help to explain the orogen-parallel dextral faulting that is observed all along external zones, from the northern Valais to the Argentera external crystalline massif. Our results highlight the consequences for the Alpine realm of ongoing convergence between the African and European plates. Our interpretation is that collision is no longer ongoing and that buoyancy-driven stresses dominate the present-day geodynamics of the western/central Alps.