Subduction zone stresses: Constraints from mechanics and from topographic and geoid anomalies

There are, in principle, direct relations between several important phenomena associated with subduction zones: the depth of oceanic trenches, the magnitude of the net force from trenches acting on subducting plates, the distribution and fault plane orientations of earthquakes, the magnitude of stresses on subduction faults, the bathymetry of back-arc regions, and the magnitudes of gravity and geoid anomalies. These phenomena are related through the stresses transmitted through surface and subducted lithosphere, and are associated with the mass anomaly of the subducted lithosphere. Quantitative estimates suggest that observed trench depths imply a trench pull force on subducting plates which is comparable to the ridge push force but much less than the excess weight of the subducted lithospheric slab. It is further suggested that either the mass anomaly of subducted lithosphere is much less than would be expected on the basis of conventional thermal and compositional models or that (a) a large resistance acts on the upper part of slabs due to high-stress corner flow, and (b) the mass anomaly of the slab is 70–90% compensated either by a broad 1 km-deep back-arc depression or a low density mantle wedge above the slab or both.     

西藏及其周围地区地壳、地幔地震层析成像——印度板块大规模俯冲于西藏高原之下的证据

文中介绍有关西藏—喜马拉雅碰撞带的一项地震层析成像研究。根据一个用天然地震数据产生的全球波速模型 ,印度板块有可能以近水平状俯冲于整个西藏高原之下至 16 5～ 2 6 0km深度。西藏岩石圈具有低波速地壳和高波速下岩石圈 (75～ 12 0km深 )。在 12 0～ 16 5km深度范围 ,西藏岩石圈与俯冲的印度板块之间有一层低速软流圈物质。高原中部从地表到 310km深处有一低速体 ,说明地幔物质有可能穿过俯冲板块的脆弱部位上隆。这些结果以及野外实测的地壳缩短值说明高原的抬升得助于印度板块的近水平俯冲。我们推论俯冲印度板块的升温上浮以及上覆软流层的存在是造成西藏高原高海拔抬升以及内部地表仍相对平坦的主要原因。2 0 0 1年 1月 2 6日在印度西部发生的毁灭性大地震有可能是俯冲应力在印度板块后缘薄弱处引发的岩石圈大断裂。     

From oceanic plateaus to allochthonous terranes: Numerical modelling

Large segments of the continental crust are known to have formed through the amalgamation of oceanic plateaus and continental fragments. However, mechanisms responsible for terrane accretion remain poorly understood. We have therefore analysed the interactions of oceanic plateaus with the leading edge of the continental margin using a thermomechanical–petrological model of an oceanic-continental subduction zone with spontaneously moving plates. This model includes partial melting of crustal and mantle lithologies and accounts for complex rheological behaviour including viscous creep and plastic yielding. Our results indicate that oceanic plateaus may either be lost by subduction or accreted onto continental margins. Complete subduction of oceanic plateaus is common in models with old (> 40 Ma) oceanic lithosphere whereas models with younger lithosphere often result in terrane accretion. Three distinct modes of terrane accretion were identified depending on the rheological structure of the lower crust and oceanic cooling age: frontal plateau accretion, basal plateau accretion and underplating plateaus.Complete plateau subduction is associated with a sharp uplift of the forearc region and the formation of a basin further landward, followed by topographic relaxation. All crustal material is lost by subduction and crustal growth is solely attributed to partial melting of the mantle.Frontal plateau accretion leads to crustal thickening and the formation of thrust and fold belts, since oceanic plateaus are docked onto the continental margin. Strong deformation leads to slab break off, which eventually terminates subduction, shortly after the collisional stage has been reached. Crustal parts that have been sheared off during detachment melt at depth and modify the composition of the overlying continental crust.Basal plateau accretion scrapes oceanic plateaus off the downgoing slab, enabling the outward migration of the subduction zone. New incoming oceanic crust underthrusts the fractured terrane and forms a new subduction zone behind the accreted terrane. Subsequently, hot asthenosphere rises into the newly formed subduction zone and allows for extensive partial melting of crustal rocks, located at the slab interface, and only minor parts of the former oceanic plateau remain unmodified.Oceanic plateaus may also underplate the continental crust after being subducted to mantle depth. (U)HP terranes are formed with peak metamorphic temperatures of 400–700 °C prior to slab break off and subsequent exhumation. Rapid and coherent exhumation through the mantle along the former subduction zone at rates comparable to plate tectonic velocities is followed by somewhat slower rates at crustal levels, accompanied by crustal flow, structural reworking and syndeformational partial melting. Exhumation of these large crustal volumes leads to a sharp surface uplift.     

Mass anomalies and the structure of the earth

Details of the Earth's geoid and gravity fields are summarized and examined. A set of 9274 centerpoints of 5 ° cubes (referred to as bloblets) represents subducted slab locations. This set, developed from reconstructed plate history, was provided by the first author of Lithgow-Berttelloni et. al. [1998] and is the best available estimate of locations of subduction material in the Earth's mantle. Two global mass solutions offered here utilize 1) only those bloblets in the outer 800 km, and 2) only those bloblets in the outer 1400 km. Since each bloblet location represents the center of a 5-degree cube [a larger volume than appropriate for a fragment of subducted lithosphere] it was necessary in the 800 km depth limit model to reduce their density to 0.004 grams/cc, and by increasing bloblet density six times at 797.5 km depth to simulate the piling up of slab material beneath the 670 km boundary. The 1400 km depth limit model [commensurate with evidence of slab penetration into the lower mantle from seismic tomography] required estimating densities for the bloblets at nine different mantle depths. An additional four point-masses at 3000 km depth (to simulate CMB topography, unrelated to dynamic topography) completes the mass models. Both these models show reasonable agreement to patterns and magnitudes for degrees 2–10, 3–10, 4–10, 2–3, 3, and 2 geoid fields with both geometric and hydrostatic flattening. These models support an assessment that topography at the core mantle boundary (CMB) may be produced by processes within the core rather than from within the mantle. Possible causes for the CMB topography are discussed.     

Spatial gaps in arc volcanism: The effect of collision or subduction of oceanic plateaus

One of the major processes in the formation and deformation of continental lithosphere is the process of arc volcanism. The plate-tectonic theory predicts that a continuous chain of arc volcanoes lies parallel to any continuous subduction zone. However, the map pattern of active volcanoes shows at least 24 areas where there are major spatial gaps in the volcanic chains (> 200 km). A significant proportion (~ 30%) of oceanic crust is subducted at these gaps. All but three of these gaps coincide with the collision or subduction of a large aseismic plateau or ridge.The idea that the collision of such features may have a major tectonic impact on the arc lithosphere, including cessation of volcanism, is not new. However, it is not clear how the collision or subduction of an oceanic plateau perturbs the system to the extent of inhibiting arc volcanism. Three main factors necessary for arc volcanism are (1) source materials for the volcanics—either volatiles or melt from the subducting slab and/or melt from the overlying asthenospheric wedge, (2) a heat source, either for the dehydration or the melting of the slab, or the melting within the asthenosphere and (3) a favorable state of stress in the overlying lithosphere. The absence of any one of these features may cause a volcanic gap to form.There are several ways in which the collision or subduction of an oceanic plateau may affect arc volcanism. The clearest and most common cases considered are those where the feature completely resists subduction, causing local plate boundaries to reorganize. This includes the formation of new plate-bounding transform faults or a flip in subduction polarity. In these cases, subduction has slowed down or stopped and the lack of source material has created a volcanic gap.There are a few cases, most notably in Peru, Chile, and the Nankai trough, where the dip of subduction is so shallow that effectively no asthenospheric wedge exists to produce source material for volcanism. The shallow dip of the slab may be a buoyant effect of the plateau imbedded in the oceanic lithosphere.The cases which are the most enigmatic are those where subduction is continuous, the oceanic plateau is subducted along with the slab, and the dip of the slab is clearly steep enough to allow arc volcanism; yet a volcanic gap exists. In these areas, the subducted plateau may have a fundamental effect on the physical process of arc volcanism itself. The presence of a large topographic feature on the subducting plate may affect the stress state in the are by increasing the amount of decoupling between the two plates. Alternatively, the subduction of the plateau may change the chemical processes at depth if either the water-rich top of the plateau with accompanying sediments are scraped off during subduction or if the ridge is compositionally different.     

Rheological evolution of a Mediterranean subduction complex

We use field and microstructural observations, coupled to previously published P-T-time histories, to track the rheological evolution of an intracontinental subduction complex exposed in the Betic Cordillera in the western Mediterranean region. The body of rock we focus on, known as the Nevado-Filabride Complex (NFC), was originally part of the upper crust of the Iberian margin. It was subducted into hot asthenospheric mantle, then exhumed back toward the surface in two stages: an early stage of fast exhumation along the top of the subducting slab in a subduction channel, and a late stage of slower exhumation resulting from capture by a low-angle detachment fault rooted at the brittle-ductile transition. Each stage of deformation in the NFC was punctuated by changes in the dominant deformation mechanism. Deformation during initial subduction of the complex was accommodated by pressure-solution creep in the presence of a fluid phase – the grain sizes, stress magnitudes, and estimated strain rates for this stage are most consistent with a thin-film model for pressure solution in which the diffusion length scale is controlled by the grain size. During the early stages of exhumation within the subduction channel, deformation transitioned from pressure solution to dislocation creep due to increases in temperature, which resulted in increases in both water fugacity and grain size, each of which favor the dislocation creep mechanism. Differential stress magnitudes for this stage were ∼10 MPa, and are consistent with simple models of buoyancy-driven channel flow. With continuing subduction-channel exhumation, deformation remained within the dislocation creep field because sequestration of free water into hydrous, retrogressive minerals suppressed the pressure-solution mechanism. Differential stresses progressively increased to ∼100 MPa near the mouth of the channel during cooling as the rocks moved into mid-crustal levels. During the final, core-complex stage of exhumation, deformation was progressively concentrated into a narrow zone of highly localized strain beneath a mid-crustal detachment fault. Localization was promoted by a transition from dislocation creep to dislocation-creep-accommodated grain boundary sliding at temperatures of ∼350–380 °C, grain sizes of ∼4 μm and differential stress magnitudes of ∼200 MPa. Peak differential stress magnitudes of ∼200 MPa recorded just below the brittle-ductile transition are consistent with Byerlee's law in the upper crust assuming a vertical maximum principal stress and near-hydrostatic pore fluid pressures. Overall, the distribution of stress with temperature, coupled to independent constraints on strain rate from field observations and geochronology, indicate that the naturally calibrated Hirth et al. (2001) flow law for wet quartzite accurately predicts the rheological behavior of mid-crustal rocks deforming by dislocation creep.     

The physico-chemical behaviour of the descending lithosphere

The lithosphere is the cold conductive boundary layer formed by cooling of the oceanic crust and upper mantle as it is convected away from oceanic ridges. Although its rheological properties vary continuously with depth, the lithosphere is conveniently divided into an upper elastic layer and a lower plastic layer, the latter overlying a zone of viscous flow. Chemically the lithosphere is vertically zoned with its uppermost part formed by variously hydrated oceanic crust; at M this overlies highly depleted dunite or harzburgite passing downwards over 50 km or so into garnet lherzolite. The vertical variation in density, and thus the gravitational stability of the lithosphere, is controlled by interplay of compositional variation and temperature distribution.As it enters an oceanic trench the lithosphere flexures elastically and plunges downwards at an average inclination close to 45°. During its descent it undergoes dissipative heating at its upper surface. Initially this heating drives a series of prograde metamorphic reactions in the oceanic crust ; because these are largely endothermic, the descending lithosphere heats less rapidly than previously expected, an effect which may be enhanced by percolation of the water of dehydration.Although it is commonly assumed that dehydration water is released upwards, it is not clear that this is true in the presence of the strong negative temperature gradients at the top of the slab, and water may initially be driven downwards into the slab to be released later at much greater depth. The magmatic activity which is associated with the partial melting of the uppermost part of the slab and with partial fusion of diapiric masses in the mantle above it, is critically dependent on the behaviour of the water carried down by the subduction process.The slab itself undergoes a series of phase changes during its descent some of which make a major contribution to the body force during subduction. By the time it reaches 700 km the slab has undergone significant thermal erosion, but the major compositional inhomogeneities within it are retained by the mantle into which it merges.     

Lithospheric structure beneath the south-eastern Zagros Mountains,Iran: recent slab break-off?

Integrated lithospheric modelling, based on the combined interpretation of gravity, geoid and topography data sets, highlights a previously undocumented lithospheric thinning beneath the Zagros collisional belt (Iran), which we propose to relate to recent slab break-off at the continent–ocean transitional lithosphere. Recent published data on the distribution of seismicity at depth support this interpretation. In agreement with other published models for the Zagros Mountains, the overlying crust exhibits, by contrast, a noticeable thickening, reaching a maximum of 52 km. The consequent thermal uplift expected from slab break-off is suggested to have modified the Zagros wedge taper and triggered the recently documented switch from thin-skinned to thick-skinned deformation in the Zagros Fold–Thrust Belt.     

Tomographic images of the upper mantle below central Europe and the Mediterranean

Results from delay time tomography of the European-Mediterranean upper mantle are discussed and where possible interpreted in terms of geodynamic processes. Slab-like positive velocity anomalies of which the locations correlate well with deeper seismicity are found beneath Spain, the Tyrrhenian basin, and the Aegean. These structures are interpreted as images of subducted slabs. Large aseismic regions with positive velocity anomalies are found beneath the Western Mediterranean, Italy, the Alps, Dinarides, the Pannonian basin, northern Greece, and the Aegean. These anomalies can also be linked to subducted lithosphere. From the anomaly patterns it is deduced that subduction occurred below the Western Mediterranean and along both sides of the Adriatic micro-plate. Beneath the Dinarides and northern Greece the velocity structures suggest detachment of the slab from the surface.     

Geochemical response of magmas to Neogene–Quaternary continental collision in the Carpathian–Pannonian region: A review

The Carpathian–Pannonian Region contains Neogene to Quaternary magmatic rocks of highly diverse composition (calc-alkaline, shoshonitic and mafic alkalic) that were generated in response to complex microplate tectonics including subduction followed by roll-back, collision, subducted slab break-off, rotations and extension. Major element, trace element and isotopic geochemical data of representative parental lavas and mantle xenoliths suggests that subduction components were preserved in the mantle following the cessation of subduction, and were reactivated by asthenosphere uprise via subduction roll-back, slab detachment, slab-break-off or slab-tearing. Changes in the composition of the mantle through time are evident in the geochemistry, supporting established geodynamic models.Magmatism occurred in a back-arc setting in the Western Carpathians and Pannonian Basin (Western Segment), producing felsic volcaniclastic rocks between 21 to 18 Ma ago, followed by younger felsic and intermediate calc-alkaline lavas (18–8 Ma) and finished with alkalic-mafic basaltic volcanism (10–0.1 Ma). Volcanic rocks become younger in this segment towards the north. Geochemical data for the felsic and calc-alkaline rocks suggest a decrease in the subduction component through time and a change in source from a crustal one, through a mixed crustal/mantle source to a mantle source. Block rotation, subducted roll-back and continental collision triggered partial melting by either delamination and/or asthenosphere upwelling that also generated the younger alkalic-mafic magmatism.In the westernmost East Carpathians (Central Segment) calc-alkaline volcanism was simultaneously spread across ca. 100 km in several lineaments, parallel or perpendicular to the plane of continental collision, from 15 to 9 Ma. Geochemical studies indicate a heterogeneous mantle toward the back-arc with a larger degree of fluid-induced metasomatism, source enrichment and assimilation on moving north-eastward toward the presumed trench. Subduction-related roll-back may have triggered melting, although there may have been a role for back-arc extension and asthenosphere uprise related to slab break-off.Calc-alkaline and adakite-like magmas were erupted in the Apuseni Mountains volcanic area (Interior Segment) from15–9 Ma, without any apparent relationship with the coeval roll-back processes in the front of the orogen. Magmatic activity ended with OIB-like alkali basaltic (2.5 Ma) and shoshonitic magmatism (1.6 Ma). Lithosphere breakup may have been an important process during extreme block rotations (60°) between 14 and 12 Ma, leading to decompressional melting of the lithospheric and asthenospheric sources. Eruption of alkali basalts suggests decompressional melting of an OIB-source asthenosphere. Mixing of asthenospheric melts with melts from the metasomatized lithosphere along an east–west reactivated fault-system could be responsible for the generation of shoshonitic magmas during transtension and attenuation of the lithosphere.Voluminous calc-alkaline magmatism occurred in the Cãlimani-Gurghiu-Harghita volcanic area (South-eastern Segment) between 10 and 3.5 Ma. Activity continued south-eastwards into the South Harghita area, in which activity started (ca. 3.0–0.03 Ma, with contemporaneous eruption of calc-alkaline (some with adakite-like characteristics), shoshonitic and alkali basaltic magmas from 2 to 0.3 Ma. Along arc magma generation was related to progressive break-off of the subducted slab and asthenosphere uprise. For South Harghita, decompressional melting of an OIB-like asthenospheric mantle (producing alkali basalt magmas) coupled with fluid-dominated melting close to the subducted slab (generating adakite-like magmas) and mixing between slab-derived melts and asthenospheric melts (generating shoshonites) is suggested. Break-off and tearing of the subducted slab at shallow levels required explaining this situation.     

The role of boundary conditions in numerical models of subduction zone dynamics

Numerical studies of subduction zone dynamics on a regional scale usually use a limited spatial extent for their models and therefore need to define boundary conditions on all model edges. These boundary conditions imply a choice for the mechanical and thermal state of the surrounding regions which may influence the evolution of the model system. We investigate the role of the surface and lateral boundary conditions for simple mechanical subduction models using a standard finite element method. We find that subduction is only possible if the slab can decouple from the surface. This decoupling can be achieved by a true free surface, a free-slip surface with a weak crust for the subducting plate, or a free-slip surface with a layer of low viscosity, low density material (‘sticky air’) between the model surface and the crust. Models of slab dynamics that employ a free-slip surface reproduce trench migration, slab sinking velocities and slab geometry of models with a free surface, as long as they use either a weak crust, which can be viscous, viscoelastic and/or brittle, or a ‘sticky air’ layer. The vertical topography will however not be reproduced for free-slip models without a ‘sticky air’ layer. For ocean–ocean convergent models we find that the application of inflow boundary conditions at the edges of the subducting or overriding lithosphere controls trench motion and the geometry of the subducting slab. Inflow on the overriding side causes trench retreat and a slab resting on the lower mantle, whereas inflow restricted to the subducting side can cause trench advance and a slab which folds on the lower mantle.     

西藏南羌塘增生杂岩中俯冲反向驱动高压变质岩折返:来自猫耳山大型拆离断层的启示

大洋岩石圈俯冲增生过程中可能伴随着复杂的深部板片运动过程。高压变质岩无疑是记录这些深部过程的良好载体。最近的研究提出,在特定情况下,双向俯冲中占主导的俯冲板块拖曳另一侧板块发生反向运动,从而短板片可能被另一侧长板片拖出。该研究提示我们关注俯冲增生过程中这种可能的“不正常”的板片运动方式,从而客观而全面地剖析碰撞造山带。现有高压变质岩折返模式中,除了俯冲隧道流模式,其余模式均强调单次快速折返。然而,俯冲反向运动导致的折返过程有所不同:对单个高压变质岩来说仍是快速折返,但是对整体高压变质岩带来说,整个俯冲反向期间必然都存在高压变质岩折返,从而形成较长的折返过程持续时间。对上地壳层次的折返相关构造变形的研究有助于揭示上述过程。     

Surface Expressions of Rayleigh-Taylor Instability in Continental Interiors

Two-dimensional thermal-mechanical numerical models show that Rayleigh-Taylor-type （RT） gravitational removal of high-density lithosphere may produce significant surface deformation （vertical deflection 〉1000 m） in the interior of a continental plate.A reasonable range of crustal strengths and thicknesses,representing a variation from a stable continental interior to a hot orogen with a thick crust,is examined to study crustal deformation and the surface deflection in response to an RT instability.In general,three types of surface deflection are observed during the RT drip event：（1） subsidence and negative topography; （2） uplift and positive topography; （3） subsidence followed by uplift and inverted topography.One key factor that determines the surface expression is the crustal thickness.Models with a thin crust mainly show subsidence and develop a basin.In the thick crust models,surface expressions are more variable,depending on the crustal strength and depth of highdensity anomaly.With weak crust and a deep high-density anomaly,the RT drip is decoupled from the overlying crust,and the surface exhibits uplift or little deflection,as the RT drip induces contraction and thickening of the overlying crust.In contrast,with a strong crust and shallow anomaly,the surface is more strongly coupled with the drip and undergoes subsidence,followed by uplift.     

Subduction geometry controls on dynamic topography: implications for the Jurassic Surat Basin

The Late Jurassic Springbok unconformity in the Surat Basin remains to date at odds with our understanding of basin-forming processes. A previously proposed cause for the unconformity is uplift related to the rifting of the Argo landmass from Western Australia, but the great distance between the Surat Basin and the Argo rift-zone argues against this interpretation. A more likely scenario is that the uplift was related to a west-dipping subduction zone along the east Australian margin. Numerous examples around the world link subduction processes, such as tears and detaching segments of the slab, to variations in surface topography. We used geodynamic forward modelling to generate uplift in the overriding plate of a subduction zone, as well as topographic changes that occur as subduction is disrupted by slab breakoff or tears. Our findings suggest that slab tearing can produce distribution and magnitude of uplift consistent with the Jurassic uplift in the Surat Basin, which potentially adds new insights into the tectonic evolution of eastern Australia.     

Seismic anisotropy in the region of the Chile margin triple junction

We have measured shear wave splitting at three temporary three-component short period stations that were deployed in southern Chile above the subducted Chile Rise spreading centre (Taitao Peninsula and environs). Subduction of the Chile Rise has been occurring beneath South America for at least the past 14 m.y. Previously published models of the ridge subduction posit the existence of ‘slab windows’, asthenosphere-filled gaps between subducted lithosphere segments of the spreading ridge, through which mantle might flow. Our preliminary results include two consistent fast polarization directions of splitting in the study region. Delay times between fast and slow split shear waves average around 1.0 s for all phases (ScS, PcS, SKS, and SKKS) that we measured. Fast-axis azimuths vary systematically among the three stations: near the coast, fast axes are parallel to the spreading ridge segments of the Chile Rise (approximately N-trending). This splitting fast-axis direction probably reflects either along-axis asthenospheric flow or results from the preferential attenuation effects of aligned pockets of melt at the subducted ridge segment. At one inland station above the slab window, we find two splitting fast-axis directions, one parallel to the subducted Chile Rise ridge segments, and a second trending NW–SE. We infer that upper mantle deformation in the vicinity of a well developed slab window is complicated and probably involves two superposed directions of upper mantle deformation. One of these directions (NW–SE) may indicate anomalous flow of asthenospheric mantle in the vicinity of the slab window gap.     

Mineral inclusions in diamonds track the evolution of a Mesozoic subducted slab beneath West Gondwanaland

Three major suites of silicate inclusions in sublithospheric diamonds show evidence of formation at depths > 250 km, and for each suite there is evidence of their formation from subducted material. Two of these are the well known basic (majoritic garnet) and ultrabasic (MgSi-perovskite + ferropericlase) suites. The third, the recently recognised Ca-rich suite, is characterised by carbonate, Ca-Si-Ti minerals and some aluminous material. Carbon isotope ratios in the host diamonds and geochemical-petrological features of the inclusions themselves provide evidence for their derivation from subducted lithosphere materials. The diamonds hosting the basic and ultrabasic suites are suggested to form in fluids/melts resulting from the release of water caused by dehydration reactions affecting both the crustal and mantle portions of a subducting slab of ocean lithosphere. Conversely, the diamonds containing the Ca-rich suite are linked with the formation of carbonatitic melts. In the Juina kimberlite province of Brazil, all three suites have been found in close proximity. A model is presented whereby the formation of the suites occurs progressively during the subduction and stagnation of a single lithospheric slab, with all three suites being transported to the lithosphere by a plume with which the carbonatitic melts of the Ca-rich suite are associated. Nd-Sr isotopic data are presented for the Juina majoritic-garnet inclusions, which supports their formation from oceanic crust of Mesozoic age. In conjunction with published age data for a Ca-Si-Ti inclusion, the Juina (Brazil) sublithospheric inclusions document a series of events involving diamond formation during and following the emplacement of a subducted slab between ca 190 and 90 Ma beneath west Gondwanaland. This slab and related subducted slabs dating from the Palaeozoic at the Gondwanan margin may be the source of the widespread DUPAL geochemical anomaly in the South Atlantic and Indian Oceans. The kimberlites bringing the diamonds to the Earth's surface may have arisen from a superplume, developed from a graveyard of former Gondwanan stagnant slabs, at the Core-Mantle-Boundary.     

Subduction history reveals Cretaceous slab superflux as a possible cause for the mid-Cretaceous plume pulse and superswell events

Subduction is a fundamental mechanism of material exchange between the planetary interior and the surface. Despite its significance, our current understanding of fluctuating subducting plate area and slab volume flux has been limited to a range of proxy estimates. Here we present a new detailed quantification of subduction zone parameters from the Late Triassic to present day (230–0 Ma). We use a community plate motion model with evolving plate topologies to extract trench-normal convergence rates through time to compute subducting plate areas, and we use seafloor paleo-age grids to estimate the thickness of subducting lithosphere to derive the slab flux through time. Our results imply that slab flux doubled to values greater than 500 km³/yr from 180 Ma in the Jurassic to 130 Ma in the mid-Cretaceous, subsequently halving again towards the Cretaceous-Paleogene boundary, largely driven by subduction zones rimming the Pacific ocean basin. The 130 Ma spike can be attributed to a two-fold increase in mid-ocean ridge lengths following the break-up of Pangea, and a coincident increase in convergence rates, with average speeds exceeding 10 cm/yr. With one third of the total 230 - 0 Ma subducted volume entering the mantle during this short ∼50 Myr period, we suggest this slab superflux drove a surge in slab penetration into the lower mantle and an associated increase in the vigour of mantle return flow. This mid-Cretaceous event may have triggered, or at least contributed to, the formation of the Darwin Rise mantle superswell, dynamic uplift of the South African Plateau and the plume pulse that produced the Ontong-Java-Hikurangi-Manihiki and Kerguelen plateaus, among others.The models presented here contribute to an improved understanding of the time-evolving flux of material consumed by subduction, and suggest that slab superflux may be a general feature of continental dispersal following supercontinent breakup. These insights may be useful for better understanding how supercontinent cycles are related to transient episodes of Large Igneous Province and superswell formation, and the associated deep cycling of minerals and volatiles, as well as leading to a better understanding of tectonic drivers of long-term climate and icehouse-to-greenhouse transitions.     

Geodynamics of the divergent double subduction along the Bangong-Nujiang tethyan suture zone: Insights from late mesozoic intermediate-mafic rocks in central Tibet

Bangong-Nujiang Suture Zone (BNSZ) in central Tibet plays an important role in evaluating the formation and uplift mechanism of Tibetan Plateau. However, its Mesozoic tectonic evolution is ambiguous and intensely debated. In this study, Early Cretacesous adakites and sodium-rich arc rocks are identified in Western Qiangtang (WQ) and Northern Lhasa (NL) sub-terranes. Forty-four adakite samples from both WQ and NL have akin geochemical features, and are derived from partial melting of subducted oceanic crust with amphibole residual. Nineteen sodium-rich samples originated from a mixed source region between crustal sediment and enriched lithospheric mantle. These two parallel arc belts separated by the Bangong-Nujiang Suture Zone (BNSZ) represent the divergent double subduction of the Bangong-Nujiang Tethyan Ocean (BNTO). Combined with the previous studies, our new data suggest three significant magmatic flare-ups at ∼240–140 Ma, 135–105 Ma and 92–60 Ma in the WQ and BNSZ, and two at 135–105 Ma and 92–60 Ma in the NL. These asymmetrical magmatic activities indicate that the southern subduction may have commenced at about 135 Ma and experienced slab breakoff at the latest Early Cretaceous, and the northern subduction could trace back to L-Triassic (228 Ma) and experienced episodic low-angle subduction, slab rollback (190-140 Ma) and oceanic ridge subduction (135-100 Ma). The 100–92 Ma magmatic gap, 92–60 Ma magmatic flare-up and L-Cretaceous angular unconformities indicate that the double-sided subduction of the BNTO resulted in soft collision with oceanic lithosphere detachment.     

Drastic shift in lava geochemistry in the volcanic-front to rear-arc region of the Southern Kamchatkan subduction zone: Evidence for the transition from slab surface dehydration to sediment melting

The shift of lava geochemistry between volcanic front to rear-arc volcanoes in active subduction zones is a widespread phenomenon. It is somehow linked to an increase of the slab surface depth of the subducting oceanic lithosphere and increasing thickness of the mantle wedge and new constraints for its causes may improve our understanding of magma generation and element recycling in subduction zones in general. As a case study, this paper focuses on the geochemical composition of lavas from two adjacent volcanic centres from the volcanic front (VF) to rear-arc (RA) transition of the Southern Kamchatkan subduction zone, with the aim to examine whether the shift in lava geochemistry is associated with processes in the mantle wedge or in the subducted oceanic lithosphere or both. The trace element and O-Sr-Nd-Hf-Pb (double-spike)-isotopic composition of the mafic Mutnovsky (VF) and Gorely (RA) lavas in conjunction with geochemical modelling provides constraints for the degree of partial melting in the mantle wedge and the nature of their slab components. Degrees of partial melting are inferred to be significantly higher beneath Mutnovsky (∼18%) than Gorely (∼10%). The Mutnovsky (VF) slab component is dominated by hydrous fluids, derived from subducted sediments and altered oceanic crust, eventually containing minor but variable amounts of sediment melts. The composition of the Gorely slab component strongly points to a hydrous silicate melt, most likely mainly stemming from subducted sediments, although additional fluid-contribution from the underlying altered oceanic crust (AOC) is likely. Moreover, the Hf-Nd-isotope data combined with geochemical modelling suggest progressive break-down of accessory zircon in the melting metasediments. Therefore, the drastic VF to RA shift in basalt chemistry mainly arises from the transition of the nature of the slab component (from hydrous fluid to melt) in conjunction with decreasing degrees of partial melting within ∼15 km across-arc. Finally, systematic variations of key inter-element with high-precision Pb-isotope ratios provide geochemical evidence for a pollution of the Mutnovsky mantle source with Gorely melt components but not vice versa, most likely resulting from trench-ward mantle wedge corner flow. We also present a geodynamic model integrating the location of the Mutnovsky and Gorely volcanic centres and their lava geochemistry with the recently proposed thermal structure of the southern Kamchatkan arc and constraints about phase equilibria in subducted sediments and AOC. Herein, the slab surface hosting the subducted sediments suffers a transition from dehydration to melting above a continuously dehydrating layer of AOC. Wider implications of this study are that an onset of (flush-) sediment melting may ultimately be the main trigger for the VF to RA transition of lava geochemistry in subduction zones.     

Subduction, back-arc spreading and global mantle flow

The shapes and orientations of Benioff zones beneath island arcs, interpreted as marking the location of subducted lithosphere, provide the best presently available constraints on the global convective flow pattern associated with plate motions. This global flow influences the dynamics of subduction. Subduction zone phenomena therefore provide powerful tests for models of mantle flow. We compute global flow models which, while simple, include those features which are best constrained, namely the observed plate velocities, applied as boundary conditions, and the density contrasts given by thermal models of the lithosphere and subducted slabs. Two viscosity structures are used; for one, flow is confined to the upper mantle, while for the other, flow extends throughout the mantle.Instantaneous flow velocity vectors match observed Benioff zone dips and shapes for the model which allows mantle-wide flow but not for the upper mantle model, which has a highly contorted flow pattern. The effect of trench migration on particle trajectories is calculated; it is not important if subduction velocities are greater than migration rates. Two-dimensional finite element models show that including a coherent high viscosity slab does not change these conclusions. A coherent high viscosity slab extending deep into the upper mantle would significantly slow subduction if flow were confined to the upper mantle. The maximum earthquake magnitude, M_w, for island arcs correlates well with the age of the subducted slab and pressure gradient between the trench and back-arc region for the whole mantle, but not the upper mantle, flow model. The correlations with orientations of Benioff zones and seismic coupling strongly suggest that the global return flow associated with plate motions extends below 700 km. For both models, regions of back-arc spreading have asthenospheric shear pulling the back-arc toward the trench; regions without back-arc spreading have the opposite sense of shear, suggesting global flow strongly influences back-arc spreading.