Alexandre I. Chemenda

Subduction versus accretion of intra-oceanic volcanic arcs: insight from thermo-mechanical analogue experiments

We perform thermo-mechanical laboratory experiments designed to explore the behaviour of the volcanic arc during intra-oceanic arc^continent collision following oceanic subduction and subsequent back-arc opening. The overriding oceanic lithosphere is made of two layers representing the oceanic crust and the lithospheric mantle. This lithosphere carries a volcanic arc and is thinned and weakened beneath both the arc and the back-arc basin. The subducting plate contains three parts: one-layer oceanic and two-layer (crust and mantle) continental lithosphere with a continental margin between them. When the continental margin reaches the trench and starts subducting, the overriding plate undergoes growing horizontal compression and finally fails in the vicinity of the back-arc spreading centre, which is the weakest part of this plate. The failure can result in subduction of the whole arc plate comprised between the trench and the back-arc spreading centre. During subduction of the arc plate, the mantle part of this plate subducts completely, while the behaviour of the arc crust depends on its thickness and strength, which is a function of composition and temperature. We tested four cases with different arc crust thicknesses and composition (rheology), with total lithosphere thickness in the arc being constant. Three types of tectonic evolution have been obtained: complete arc subduction, complete arc accretion, and partial arc subduction/accretion. The result is largely controlled by the crustal thickness of the arc. A thin arc (thickness equivalent to V16 km in nature) made of the same strong material as the oceanic crust subducts completely without leaving any trace at the surface. On the contrary, a thick arc (equivalent to V26 km in nature) made of the same material is scraped off and accreted to the overriding plate. The lower crust of such an arc is hotter, therefore its strength at ‘Moho’ depth and coupling between crust and mantle are small. In addition, the thick arc has a high isostatic relief and hence a greater mechanical resistance to subduction. Therefore, the arc is scraped off. If the arc is made of a weaker ‘continental-like’ material or contains a weak layer/ low friction interface, it is completely or partially scraped off even if it is small. When there is no back-arc opening before collision (no thin and weak lithosphere in the rear of the arc), the overriding plate fails in the arc area, which may result in a complete fore-arc block subduction, with the volcanic arc remaining at the surface. The obtained models are compared with mountain belts with nearly no trace of arc activity (Oman), with accreted arc (Kohistan), and with small remnants of subducted arc (southern Tibet). : 2003 Elsevier Science B.V. All rights reserved.

A model of Palaeozoic obduction and exhumation of high-pressure/low-temperature rocks in the southern Urals

Abstract A new evolutionary geodynamic model of Devonian continental subduction and exhumation of high-pressure/low-temperature metamorphic rocks in the southern Urals is proposed based on the results of physical modelling, geological and geophysical data, and on the comparison of this belt with other orogens. The model includes the following principal phases. (1) Closure of the western Uralian ocean by eastward intraoceanic subduction associated with the Magnitogorsk volcanic arc. (2) Subduction of the European continental margin, causing failure of the overriding plate in the arc area along a fault dipping eastward under the arc. (3) Subduction into the mantle of the fore-arc block together with the underlying continental crust and sedimentary layer. The sedimentary cover, as well as the crustal and mantle fragments of this block, are scraped off and accreted in front of and under the arc. The continental crust, shielded from the hot mantle by the fore-arc block, subducts to depths of more than 150 km, remaining at a relatively low temperature. The crust then fails at depths of several tens of kilometres. (4) The subducted crustal slice starts to rapidly rise and intrude the interplate zone, scraping and pushing up the sediments and slivers of the oceanic fore-arc block previously dragged down to different depths. (5) Exhumation of the rising high-pressure/low-temperature rocks within the Uraltau dome, separating the subduction accretionary complex from the Magnitogorsk arc. The eastern border of the Uraltau dome coincides with the Main Uralian Fault. This fault represents a complex suture zone which at the stage of deep crustal subduction was associated with a major thrust (subduction) fault, and at the exhumation stage with a major normal fault corresponding to the upper surface of the rising crustal/sedimentary body.

Evolutionary model for the Taiwan collision based on physical modelling

Abstract 2-D and 3-D physical modelling of lithospheric convergence in the Luzon-Taiwan-Ryukyu region is performed with properly scaled laboratory models. The lithospheric model consists of two parts, continental (the Asian Plate, AP) and oceanic (the Philippine Sea Plate, PSP). The oceanic lithosphere has one layer, while the continental lithosphere includes both mantle and crustal layers. The continental margin is covered by sediments. A low-viscosity asthenosphere underlies the lithosphere. The opposing Luzon and Ryukyu subduction zones are initiated by inclined cuts made within the PSP. The subduction/collision is driven by a piston. Pre-collisional intraoceanic subduction along the Luzon and Ryukyu boundaries results in the formation of a transform zone between them, with two tear faults at the ends. The PSP undergoes strong compression along this zone. Subduction of the Chinese margin under the Luzon boundary further increases the compression. Compressive stresses reach the yield limit of the PSP in the arc area, which is a weak zone in the experiments. The plate fails at the western side of the arc along an eastward dipping fault, the Longitudinal Valley Fault. Underthrusting of the frontal wedge of the PSP along this fault results in the closure of the fore arc basin and is then blocked. The PSP fails at the opposite side of the Luzon arc along the westward dipping fault. The failure releases lithospheric compression in this region and results in the initiation of southward-propagating subduction of the PSP under northeastern Taiwan. The incipient subduction zone becomes part of the southeastward-retreating Ryukyu subduction zone, which allows the Okinawa back arc rift to propagate into Taiwan. The Taiwan collision thus includes the following succession of major processes over time, or from south to north: (1) an E-W shortening of the PSP in the Luzon arc; (2) a failure of this plate at the western side of the arc and the formation of the eastward-dipping Longitudinal Valley Fault (the transient plate boundary); (3) a closure of the fore arc basin and a rapid uplift of the orogen; (4) a failure of the PSP at the eastern side of the Luzon arc partly overthrusting the orogen, and the initiation of westward (WN-ward) subduction of the PSP; (5) and finally ‘back arc’ rifting in the rear of this incipient subduction zone (i.e. in northern Taiwan). All these processes commence with some delay with respect to the preceding ones and propagate southwards.

Devonian geodynamic evolution of the Variscan Belt, insights from the French Massif Central and Massif Armoricain

The Paleozoic French Variscan Belt in Massif Central and Massif Armoricain is a collision belt that provides a good example of a suture zone where ophiolites are rare, and the frontal (i.e., the magmatic arc) part of the upper plate is not present. In the lower plate (or Gondwana), the continental rocks are subdivided into an Upper Gneiss Unit (UGU) and a Lower Gneiss Unit (LGU). The UGU experienced a high-pressure (and likely ultra-high-pressure) metamorphism followed by crustal melting during their exhumation. New chemical U-Th-Pb monazite ages and ion-probe U-Pb zircon ages on migmatites allow us to constrain the P-T-t paths followed by the UGU and LGU. By comparison with thermomechanical experiments, a possible geodynamic evolution scenario can be proposed for the Variscan convergence. The high-compression regime of continental subduction developed during the initial subduction of the northern margin of Gondwana under Armorica in Silurian times. This induced the formation of a new subduction zone in the back-arc basin, which is the youngest, hottest, and thus mechanically the weakest part of the overriding plate. As a result, the arc-back-arc basin domain has been almost totally subducted below Armorica. Only a limited part of the back-arc basin rocks remains exposed in the Devonian St-Georges-sur-Loire Unit. Subsequently, the continental subduction of Gondwana resumed with a steeper dip associated with low-compression regime that in turn allowed the high-pressure rocks to be exhumed and partly melted in Late Devonian times. Such a scheme depicts quite well the complexity of the Variscan Belt.

Impact of arc-continent collision on the conditions of burial and exhumation of UHP/LT rocks: experimental and numerical modelling

Abstract A 2-D physical and finite-element numerical modelling of arc continent collision was performed to study the deformation and failure of the overriding lithosphere. The experimental technique allowed us to model the whole subduction/collision process from oceanic subduction to deep subduction of the continental crust. With the numerical approach we have modelled the deformation of the overriding plate only through initial stages of its failure and studied the influence of different parameters on this process. The results obtained by both techniques are coherent and mutually complementary. They show that the failure of the overriding plate is physically quite plausible or even inevitable during subduction. The conditions for such a failure (the weakening of this plate) are prepared during oceanic subduction. The weakening occurs due to the interaction between the subducting lithosphere and the asthenosphere in the mantle corner between the two plates, and due to back-arc spreading. In oceanic subduction zones with a compressional regime (no back-arc opening, thick and strong back-arc lithosphere), the weakest zone is volcanic arc area. When weakening becomes sufficient during subduction, the lithosphere fails in this area. The failure occurs along a fault dipping under the arc in either of two possible directions and results either in subduction reversal or subduction of the fore-arc. Almost half of the presently active subduction zones are characterised by a tensional subduction regime with back-arc spreading. In such subduction zones, the weakest place is not the volcanic arc but the back-arc spreading centre. When a subduction regime changes from tensional to compressional, failure occurs in the vicinity of the extinct spreading centre. This process can occur during oceanic subduction again along either a trench-vergent or trenchward-dipping fault, but the formation of a trench-verging fault is most likely. In this latter case, which is a principal subject of our study, the failure is followed by partial subduction of the arc plate. Complete subduction occurs during arc-continent collision (subduction of the continental margin) when tectonic compression of the lithosphere increases rapidly and becomes sufficient to push the arc plate into the mantle. The arc itself can be subducted completely or be partially or completely scraped-off and accreted. A deeply subducted material (including continental margin) is preserved at relatively low temperatures between the lithospheric mantle and the “cold” subducted arc plate to about 150-km depth. Subduction of the arc plate is a major phenomenon, which affects all processes associated with continental subduction from deep burial and HP/LT metamorphism to exhumation of subducted material. Does this process occur in nature? Future investigations will allow us to answer this question. In this paper, we analyse the conditions of emplacement of a very young oceanic lithosphere (Samail ophiolite) on the continental crust in Oman in the late Cretaceous and argue that this lithosphere formed in a back-arc basin. It reached and overthrust the Arabian continent after complete subduction of the arc plate.

Numerical modelling of arc–continent collision: application to Taiwan

Two-dimensional finite element numerical modelling is applied to study the deformation and failure of the overriding plate during arc‐continent collision (continental margin subduction). This plate has elasto-plastic rheology with strain weakening and contains a ‘volcanic arc’ with thinned and weakened lithosphere. The overriding plate deforms due to the normal and tangential stresses applied along the interplate surface. These boundary conditions represent the friction and pressure between the plates. The latter is due to the flexural rigidity of the subducting lithosphere and the buoyancy force generated by the subducting continental crust. The modelling shows that continental margin subduction results in increasing compression and failure of the overriding plate, which occurs along the surface dipping under the arc in either of two possible directions. The failure mode is largely controlled by the two competitive factors: the rigidity of the subducting plate and thickness gradient of the subducted continental crust. A high rigidity favors failure along an ocean-vergent fault, which is followed by a subduction reversal, while a high thickness gradient favors failure in the opposite direction, which is followed by a fore-arc block underthrusting beneath the arc. Both scenarios seem to have natural analogs. We consider one of them, the ongoing arc‐continent collision in Taiwan, and argue that this process occurs according to the second scenario corresponding to the fore arc underthrusting. Wavelet statistical analysis of the seismicity distribution to the south of Taiwan has clearly displayed a shallow (0‐40 km) zone of high density, coherently distributed seismicity beneath the Luzon Arc. This zone, interpreted as a lithospheric-scale fault, dips from the forearc basin to the east and corresponds to the initiation of the forearc block subduction. A self-consistent, combined mechanical‐gravity‐topography model is used to see whether failure of the overriding plate to the south of Taiwan can be ‘captured’ by this model. By ‘tuning’ diVerent controlling parameters, we did not succeed in obtaining realistic topography and gravity field in a model where failure of the overriding plate was not allowed. Introduction of this failure and underthrusting of the forearc block under the Luzon Arc allowed us to fit both topography and gravity data.

Three-dimensional laboratory modelling of rifting: application to the Baikal Rift, Russia

Abstract Continental rifting is treated as a mechanical instability developing under horizontal tectonic tension. The instability results in strain localization and the formation of a neck, which is interpreted as a rift zone. At the scale of the whole lithospheric plate, this process occurs in plane-stress conditions and can therefore be modelled to a first approximation by a one-layer plate whose properties represent integral (over thickness) properties of the real lithosphere. We have designed a scaled experimental single-layer lithosphere model having elasto-plastic rheology and lying upon a liquid substratum to study its behaviour under axial horizontal tension. In a homogeneous plate, the instability develops along a linear zone oriented at an angle of ∼60° to the tension axis. This orientation is preserved even when the divergent displacement of the plate boundaries is not plain-parallel but rotational. In the latter case, the strain localization zone is rapidly propagating. When the plate length to width ratio is less than ∼2.5, the necking develops along two branches conjugated at an angle of about 120°, which is frequently observed in actual rift systems. If the model contains a local weak zone (hot spot or fault zone), the rift junction is located at this zone. In the lithospheric models comprising strong (cratonic) and weak segments, strain localization depends on the configuration of the boundary between different lithospheres. The necking starts to form within the weak segment in the vicinity of the cratonic promontories and propagates in opposite directions again at an angle of ca. 60° to the tension axis. In the models containing both a strong lithosphere and local weak zones, the rift configuration depends on their shape and relative positions, with necking always going through the weak zones. In a set of models, we have reproduced the geometry of the boundary between the Siberian craton and the thermally much younger (∼100 Ma) Sayan–Baikal lithosphere in the Baikal rift area. In these models, we were able to obtain the well-known three-branch configuration of the Baikal rift system only by introducing a weak zone in the area of Lake Baikal. Such a zone simulates the Paleozoic suture existing in this area. As in nature, two wide outer branches (eastern and western) are oblique to the regional tension axis, whereas the central one is narrow and orthogonal to the tension direction. In nature and in the model, rifting starts in the central branch corresponding to Lake Baikal. The modelling also predicts the formation of a fourth oblique ∼NS-trending branch to the south of Baikal. Although poorly expressed in the field, this branch has some seismotectonic and magmatic manifestations. The orientations of all four branches with respect to each other and with respect to the regional tension direction are remarkably similar in nature and in the model.

Physical Modeling of Arc–Continent Collision: A Review of 2D, 3D, Purely Mechanical and Thermo‐Mechanical Experimental Models

In this chapter we present a review of 2D, and 3D, purely mechanical and thermo‐mechanical experimental models of arc–continent collision obtained using the modeling technique pioneered by A. Chemenda. Also presented are earlier models of oceanic and continental subduction which led to the development of the arc–continent collision experiments. Physical models of continental subduction revealed the existence of two principal subduction regimes defined by the interplate pressure, which is inversely proportional to the slab pull‐force. In both high and low compression regimes, the subduction of a continental passive margin generates a horizontal compression of the overriding plate that can produce failure of the overriding plate in the arc area or near the back‐arc basin spreading center if the arc–continent collision was preceded by oceanic subduction in the extension regime. Failure of the overriding plate can lead to subduction reversal or the subduction of either the fore‐arc block or the entire arc plate. Evolutionary scenarios including subduction of the fore‐arc block have been proposed for Taiwan and the Urals, where the fore‐arc block is presently subducting or is missing. The scenario including the subduction of the arc plate with total or partial subduction/accretion of the arc crust fits the geological data of the Oman Mountains, the western Variscan belt and Kohistan‐Ladakh arc in western Himalaya. Although these modeling results correspond well to the geological data, it was purely mechanical and did not consider any change in the mechanical properties during subduction. In nature, however, both pressure and temperature increase with depth causing the strength of the subducting crust and mantle to be reduced by about one order of magnitude when reaching 100 km‐depth. Thermo‐mechanical laboratory experiments revealed that such strong change deeply affect the subduction and exhumation processes. In the low compression regime, subduction of the continental passive margin does not produce failure of the overriding plate in the arc area and the continental crust can only subduct to ~120 km‐depth in the asthenosphere. By then, it has become too hot and weak and undergoes large deformation, including upward ductile flow of the deeply subducted portions and a localized failure of the upper crust at depth of a few tens of kilometers allowing the buoyancy‐driven exhumation of a crustal slice in between the plates. In the high compression regime, the subducted continental crust reaches greater depth (~150–200 km), remaining relatively cold due to the subduction of the fore‐arc block or the arc plate that occurs in this regime. However, the exhumation of the deeply subducted continental crust that reached UHP/LT conditions does not occur in 2D models in the high compression subduction regime. Such exhumation has been obtained in 3D thermo‐mechanical laboratory experiments where the geometry of the interplate zone causes a local reduction of the interplate pressure, which in turn allows a local buoyancy‐driven exhumation of UHP/LT material.

Mechanical and field studies of deformation localization in rocks

Refining Constitutive Characterization of Geomaterials and Identifying Their Rupture Mechanisms in Natural Conditions; Montpellier, France, 7–8 September 2011 Results from rock testing and geological observations of different types of deformation bands and fractures highlight the problem of understanding the underlying formation mechanisms from granular (micro) to outcrop scale. Particularly puzzling are recently discovered compaction and dilatancy bands that are completely antipodal features. Both are narrow (several grain diameters thick), generally not planar but rather irregular and zigzagging (depending on the observation scale) bands that can form in the uniformly stressed rocks. The difference between them is that the compaction bands form at a relatively high pressure (mean stress) normal to the maximal compressive stress. The material within the compaction bands undergoes volume reduction due to the more dense grain packing and/ or crushing. Conversely, the dilatancy bands form at low pressure and normal to the least compressive stress. The material within them undergoes a volume increase due to grain rearrangements and/or formation of intragranular and intergranular microfractures. The dilatancy bands can become opened fractures with further deformation.

Field, experimental, and numerical studies of deformation localization bands

Characterization and Formation Mechanisms of Deformation Localization Bands(Shear, Compaction, and Dilatancy): Field and Experimental Data,Theoretical Analysis and Numerical Models;Montpellier, France, 9–10 May 2011. The origin of fractures and discontinuities in rocks has been for decades a subject of common interest in both the academic world and industry, as these features influence hydrocarbon reservoir production. Fractures (notably joints) and deformation bands are usually analyzed on a separate basis. Joints have commonly been interpreted as mode I fractures susing fracture mechanics, whereas the bifurcation theory has been used for deformation bands. This diversity was called into question by the research carried on by the Geo-FracNet consortium sponsored by Total and Shell International. New findings (A. I. Chemenda et al., J. Geophys. Res., 116, B04401, doi:10.1029/2010JB008104, 2011, and references therein) suggest that at least a part of joints could have been formed as dilatancy (dilation) bands, i.e., due not to the strong effective tensile stress concentration at the opening fracture tip but to the porosity increase and decohesion (perigranular and/ or intragranular grain breakage) of the material within a band several grains thick. The band (material damage and dilatancy within it) can propagate along the strike without opening and can eventually be opened.