Search

This paper presents and discusses the results of phase diagram (Perple_X) and diffusion modelling (CZGM, or Compositional Zoning and its Modification by diffusion) to constrain the P–T path of metamorphism. The approach is based on the best fits between the zoning profile in measured garnet and that obtained by the intersections of garnet isopleths calculated by phase diagram modelling using whole rock bulk composition. The model was applied to garnets in natural rocks of various metamorphic grades, which were formed within different geotectonic environments. To compare the sequence of compositional change during Barrovian-type metamorphism, well-studied pelitic rocks from garnet–staurolite, kyanite–sillimanite, and sillimanite-K-feldspar metamorphic zones were selected. Garnets with two-stepped core and rim profiles that were formed during two different metamorphic stages or events were used for pressure–temperature (P–T) path constraint of each stage or event. For high-grade rocks, in which the original zoning profile in garnet was severely modified, the diffusion of the initial zoning profile was quantified to estimate the timescale of the metamorphic event. These rocks include high- to ultra-high-pressure rocks, which were subjected to thermal overprinting during collisional orogenesis. The results of the application of this approach allow for deciphering the reason why the calculated profile by phase diagram modelling does not fit with that of the measured garnet from low-grade rocks, in which garnet has preserved the original compositional zoning. This includes garnets whose nucleation was shifted from the garnet-in boundary to higher temperatures and pressures, as well as garnet crystallised during different metamorphic stages or events. Finally, the P–T paths in high-grade rocks were constrained after the multicomponent diffusion in garnet was quantified, and this was used for further P–T-time path constraint of metamorphism in the rocks.

In their discussion of our recent publication, Scoon and Mitchell (2020) put forward a number of arguments against the hydromagmatic model of Bushveld Complex formation that we present. Their criti...

Crystallization of the 2.06 Ga Bushveld magma formed a 9 km (maximum) sequence of ultramafic and mafic rocks that generated a large volume of country fluid as it thermally metamorphosed a 3+ km sec...

We investigate how a conceptual hydrodynamic model consisting of fluid localization and stagnation by thermally activated compaction can explain low-resistivity anomalies observed in the lower crust (>20 km depth). Electrical resistivity models, derived from magnetotelluric data collected across the intracontinental Bulnay region, a subset of a larger regional array across central Mongolia, are generated. They reveal low-resistivity (3 - 30 Ωm) domains with a width of ~25 km and a vertical extent of Based on the observed thermal structure of the crust, and assuming the mean stress at the brittle-ductile transition is twice the vertical load, the hydrodynamic model predicts that fluids would collect in zones From the electrical resistivity models, the lower crustal viscous compaction-length is constrained to be ~25 km - in this region. Within the conceptual model, this length-scale is entirely consistent with independent estimates for the specific hydraulic and rheological properties of this region. In fact, this can be used to independently constrain acceptable ranges for the lower crustal effective viscosity, which is found to be low (on the order of 10^18 Pas). Accordingly, the results indicate that low-salinity fluids (likely 1 - 0.01 wt% NaCl), and correspondingly low porosities (likely 5 - 0.1 vol%), are the most plausible. These key findings suggest partial melts are not favoured to explain the anomalies. Overall, the results of this contribution imply that it is tectonic and compaction processes that control lower crustal fluid flow, rather than lithological or structural heterogeneity.

We present electrical resistivity models, derived from magnetotelluric data, of the crust beneath the Bulnay region, Mongolia. They reveal that the lower crust contains a pattern of discrete zones (width of ~25 km) of low resistivity (, Geophysical Research Letters, 47 (19), ISSN:0094-8276, ISSN:1944-8007

Subduction of crustal C governs the long-term global C cycling. The role of carbonates recycling in subduction zones and the related dissolution of C at various depths have been the subject of a large body of literature over the last decades. Much less is known about the contribution of carbonaceous material (CM) to the deep C cycling in subduction zones. This paper presents natural evidence for intense fluid-mediated leaching of CM in pelitic schists at high-pressure/low-temperature conditions relevant to the forearc region of subducting slabs. Manifestations of such process were identified along fluid pathways at various scales in the blueschist-facies subduction complexes of both Alpine Corsica and the Western Alps. Microstructural, whole-rock and Raman analyses across a selected metasomatic aureole were used to quantify the amount and mechanisms of C loss during fluid-rock interaction. In samples affected by intense fluid infiltration, >90% of the initial CM was removed from the rock. Microstructural and micro-Raman data indicate selective leaching of disordered CM relative to nearly crystalline graphite. The collected data allowed constraining the magnitude of fluid fluxes required to bleach the studied CM-bearing lithologies at different P-T-fO2 conditions, which corresponds to rather high time-integrated fluid fluxes in the order of ~106 m3/m2. In settings of large-scale fluid channelization, such as along regional-scale, lithological/tectonic boundaries or at the top of the subducted sedimentary pile, intense dissolution of subducted CM is expected. This process may thus exert a negative feedback on the sink of C phases into the deep mantle over the geological timescales and contribute to the release of isotopically light C from subducting slabs in forearc regions.

The number of solute species required to describe the thermodynamic behavior of electrolytic fluids is a hindrance to the incorporation of aqueous geochemistry in petrological Gibbs energy minimization procedures. An algorithm is developed to overcome this problem. Beginning from the solute-free limit, chemical potentials, and phase stability are determined by minimization, the solute speciation and bulk fluid properties consistent with these chemical potentials are then computed and the procedure repeated until the chemical potentials converge. Application of the algorithm to a model for metamorphism of subducted sediment shows that accounting for solute chemistry does not change the conclusion based on molecular fluid models that a pervasive water flux from the subjacent mantle is required to explain island-arc carbon emissions by fluid-mediated slab decarbonation. This putative flux would deplete the sediment in potassium, limiting the capacity of the slab to transport water to greater depth and rendering it refractory to melting.

Rocky planets and satellites form through impact and accretion processes that often involve silicate fluids at extreme temperatures. First-principles molecular dynamics (FPMD) simulations have been used to investigate the bulk thermodynamic properties of SiO 2 fluid at high temperatures (4000–6000 K) and low densities (500–2240 kg m −3 ), conditions which are relevant to protoplanetary disc condensation. Liquid SiO 2 is highly networked at the upper end of this density range, but depolymerises with increasing temperature and volume, in a process characterised by the formation of oxygen–oxygen (O O) pairs. The onset of vaporisation is closely associated with the depolymerisation process, and is likely to be non-stoichiometric at high temperature, initiated via the exsolution of O 2 molecules to leave a Si-enriched fluid. By 6000 K the simulated fluid is supercritical. A large anomaly in the constant-volume heat capacity occurs near the critical temperature. We present tabulated thermodynamic properties for silica fluid that reconcile observations from FPMD simulations with current knowledge of the SiO 2 melting curve and experimental Hugoniot curves.

The chemistry of aqueous fluids controls the transport and exchange-the cycles-of metals and volatile elements on Earth. Subduction zones, where oceanic plates sink into the Earth's interior, are the most important geodynamic setting for this fluid-mediated chemical exchange. Characterizing the ionic speciation and pH of fluids equilibrated with rocks at subduction zone conditions has long been a major challenge in Earth science. Here we report thermodynamic predictions of fluid-rock equilibria that tie together models of the thermal structure, mineralogy and fluid speciation of subduction zones. We find that the pH of fluids in subducted crustal lithologies is confined to a mildly alkaline range, modulated by rock volatile and chlorine contents. Cold subduction typical of the Phanerozoic eon favours the preservation of oxidized carbon in subducting slabs. In contrast, the pH of mantle wedge fluids is very sensitive to minor variations in rock composition. These variations may be caused by intramantle differentiation, or by infiltration of fluids enriched in alkali components extracted from the subducted crust. The sensitivity of pH to soluble elements in low abundance in the host rocks, such as carbon, alkali metals and halogens, illustrates a feedback between the chemistry of the Earth's atmosphere-ocean system and the speciation of subduction zone fluids via the composition of the seawater-altered oceanic lithosphere. Our findings provide a perspective on the controlling reactions that have coupled metal and volatile cycles in subduction zones for more than 3 billion years7.

This work explores the use of several van der Waals (vW)-type equations of state (EoS) for predicting vaporous phase relations and speciation in the Si-O system, with emphasis on the azeotropic boiling curve of SiO2-rich liquid. Comparison with the observed Rb and Hg boiling curves demonstrates that prediction accuracy is improved if the a-parameter of the EoS, which characterizes vW forces, is constrained by ambient pressure heat capacities. All EoS considered accurately reproduce metal boiling curve trajectories, but absent knowledge of the true critical compressibility factor, critical temperatures remain uncertain by ~500 K. The EoS plausibly represent the termination of the azeotropic boiling curve of silica-rich liquid by a critical point across which the dominant Si oxidation state changes abruptly from the tetravalent state characteristic of the liquid to the divalent state characteristic of the vapor. The azeotropic composition diverges from silica toward metal-rich compositions with increasing temperature. Consequently, silica boiling is divariant and atmospheric loss after a giant impact would enrich residual silicate liquids in reduced silicon. Two major sources of uncertainty in the boiling curve prediction are the heat capacity of silica liquid, which may decay during depolymerization from the near-Dulong-Petit limit heat capacity of the ionic liquid to value characteristic of the molecular liquid, and the unknown liquid affinity of silicon monoxide. Extremal scenarios for these uncertainties yield critical temperatures and compositions of 5200–6200 K and Si1.1O2–Si1.4O2. The lowest critical temperatures are marginally consistent with shock experiments and are therefore considered more probable.

Phase equilibrium models are used routinely to predict geophysically relevant mantle properties. A limitation of this approach is that nonlinearity of the phase equilibrium problem precludes direct assessment of the resultant uncertainties. To overcome this obstacle, we stochastically assess uncertainties along self-consistent mantle adiabats for pyrolitic and basaltic bulk compositions to 2000 km depth. The dominant components of the uncertainty are the identity, composition and elastic properties of the minerals. For P wave speed and density, the latter components vary little, whereas the first is confined to the upper mantle. Consequently, P wave speeds, densities, and adiabatic temperatures and pressures predicted by phase equilibrium models are more uncertain in the upper mantle than in the lower mantle. In contrast, uncertainties in S wave speeds are dominated by the uncertainty in shear moduli and are approximately constant throughout the model depth range.

Contributions to Mineralogy and Petrology, 159 (1), ISSN:0010-7999, ISSN:1432-0967

Fluids exert a key control on the mobility of elements at high pressure and temperature in the crust and mantle. However, the prediction of fluid composition and speciation in compositionally complex fluid–rock systems, typically present in subduction zones, has been hampered by multiple challenges. We develop a computational framework to study the role of phase equilibria and complex solid-solutions on aqueous fluid speciation in equilibrium with rocks to 900 °C and 3 GPa. This is accomplished by merging conventional phase-equilibrium modeling involving electrolyte-free molecular fluids, with an electrostatic approach to model solute–solute and solute–solvent interactions in the fluid phase. This framework is applied to constrain the activity ratios, composition of aqueous solutes, and p H of a fluid in equilibrium with a pelite lithology. Two solvent compositions are considered: pure H2O, and a COH fluid generated by equilibration of H2O and graphite. In both cases, we find that the p H is alkaline. Disparities between the predicted peralkalinity of our fluid ( [ Na ] + [ K ] ) / [ Al ] ∼ 6 to 12 and results from independent mineral solubility experiments (∼2) point to the presence of Na–K–Al–Si polymers representing ca. 60 to 85% of the total K and Al content of the fluid at 600 °C and 2.2 GPa, and to an important fraction of dissolved Ca and Mg not accounted for in present speciation models. The addition of graphite to the system reduces the relative permittivity by ca. 40% at elevated T and low P, triggers the formation of C-bearing anions, and brings the p H closer to neutrality by up to 0.6 units at low T. This ionic C pool represents up to 45 mol% of the fluid ligands at elevated P, and is dominant at low P despite the low ionic strength of the fluid ( p H dependent processes that govern volatile, major and trace element partitioning between rocks and fluids in experimental or natural systems.

We explore the influence of major elements chemistry and H2O-content on the density and seismic velocity of crustal rocks by computing stable and metastable crustal mineralogy and elastic properties as a function of pressure and temperature (P-T). Proposed average compositions of continental crust result in significantly different properties, for example a difference in computed density of ∼ 4 % is obtained at a given P-T. Phase transformations affect crustal properties at the point that crustal seismic discontinuities can be explained with mineral reactions rather than chemical stratification. H2O, even if introduced in small amount in the chemical system, has an effect on physical properties comparable to that attributed to variations in major elements composition. Thermodynamical relationships between physical properties differ significantly from commonly used empirical relationships. Density models obtained by inverting CRUST 1.0 compressional wave velocity are different from CRUST 1.0 density and translate into variations in isostatic topography and gravitational field that ranges ±600 m and ±150 mGal respectively. Inferred temperatures are higher than reference geotherms in the upper crust and in the deeper portions of thick orogenic crust, consistently with presence of metastable rocks. Our results highlight interconnections/dependencies among chemistry, pressure, temperature, seismic velocities and density that need to be addressed to better understand the crustal thermo-chemical state.

Buoyancy-driven flow in deformable porous media is important for understanding sedimentary compaction as well as magmatic and metamorphic differentiation processes. Here mathematical analysis of the viscoplastic compaction equations is used to develop an understanding of the porosity wave instability and its sensitivity to the choice of rheological model. The conditions of propagation, size, speed, and shape of the porosity waves depend strongly on the properties of the solid rock frame. Whereas most of the previous studies on porosity waves were focused on viscous or viscoelastic mode, here we consider the ability of a solid matrix to undergo simultaneous plastic (rate-independent) and viscous (rate-dependent) deformation in parallel. Plastic yielding is identified as a cause of compaction-decompaction asymmetry in porous media—this is known to lead to a strong focusing of porous flow. Speed and amplitude of a porosity wave are given as functions of material parameters and a volume of a source region. Formulation is applicable to fluid flow in sedimentary rocks where viscous deformation is due to pressure solution as well as in deep crustal or upper mantle rocks deforming in a semibrittle regime.

The siderite (FeCO3) melting curve is determined through multi-anvil experiments at 6–20 GPa, and 1300–1870 °C. The experiments define a melting curve with a Clapeyron slope steepening from 45 to 18 °C/GPa but without backbend at upper mantle conditions, i.e., siderite is denser than FeCO3-melt (FeCO3L). The melting curve fits Tm = 1037(44) + 70.0(88) * P − 1.43(37) * P2 (valid from 5 to 20 GPa) where pressure is in GPa and temperature in °C. Siderite melting is not stoichiometric, minor quench magnetite was always observed and is interpreted as the result of partial redox dissociation of FeCO3L leading to dissolved Fe3 + and CO2 in the carbonate melt. At pressures below ~ 6.8 GPa, siderite does not melt but decomposes through an auto redox dissociation reaction to magnetite, a carbon polymorph and CO2. From the experimental determination of the pure siderite melting curve, we calculate thermodynamic properties of the FeCO3L end-member, which reproduce the siderite melting curve in P–T space better than 10 °C. The metastable 1 atm melting temperature is calculated to 1012 °C. Siderite has the lowest melting temperature of the Ca–Mg–Fe carbonates, its melting curve may cross the mantle geotherm at transition zone pressures. The stability of siderite is not only dependent on pressure and temperature but also strongly on oxygen fugacity (fO2). Model calculations in P–T-fO2 space in a Fe–C–O2 system show that the siderite or FeCO3-melt maximum stability is always reached at conditions of the CCO buffer. Our experimental and thermodynamic data constitute a cornerstone to model carbonate melting in the deep Earth, necessary to understanding the deep carbon cycle.

Gibbs energy minimization is the means by which the stable state of a system can be computed as a function of pressure, temperature and chemical composition from thermodynamic data. In this context, state implies knowledge of the identity, amount, and composition of the various phases of matter in heterogeneous systems. For seismic phenomena, which occur on time-scales that are short compared to the timescales of intra-phase equilibration, the Gibbs energy functions of the individual phases are equations of state that can be used to recover seismic wave speeds. Thermodynamic properties relevant to modelling of slower geodynamic processes are recovered by numeric differentiation of the Gibbs energy function of the system obtained by minimization. Gibbs energy minimization algorithms are categorized by whether they solve the non-linear optimization problem directly or solve a linearized formulation. The former express the objective function, the total Gibbs energy of the system, indirectly in terms of the partial molar Gibbs energies of phase species rather than directly in terms of the Gibbs energies of the possible phases. The indirect formulation of the objective function has the consequence that although these algorithms are capable of attaining high precision they have no generic means of treating phase separation and expertise is required to avoid local minima. In contrast, the solution of the fully linearized problem is completely robust, but offers limited resolution. Algorithms that iteratively refine linearized solutions offer a compromise between robustness and precision that is well suited to the demands of geophysical modeling.

Understanding the physical–chemical mechanisms and pathways of geochemical transport in subduction zones remains a long-standing goal of subduction-related research. In this study, we perform fully coupled geochemical–thermo-mechanical (GcTM) numerical simulations to investigate Pb isotopic signatures of the two key “outputs” of subduction zones: (A) serpentinite melanges and (B) arc basalts. With this approach we analyze three different geodynamic regimes of intra-oceanic subduction systems: (1) retreating subduction with backarc spreading, (2) stable subduction with high fluid-related weakening, and (3) stable subduction with low fluid-related weakening. Numerical results suggest a three-stage Pb geochemical transport in subduction zones: (I) from subducting sediments and oceanic crust to serpentinite melanges, (II) from subducting serpentinite melanges to subarc asthenospheric wedge and (III) from the mantle wedge to arc volcanics. Mechanical mixing and fluid-assisted geochemical transport above slabs result in spatially and temporarily variable Pb concentrations in the serpentinized forearc mantle as well as in arc basalts. The Pb isotopic ratios are strongly heterogeneous and show five types of geochemical mixing trends: (i) binary mantle–MORB, (i) binary MORB–sediments, (iii) double binary MORB–mantle and MORB–sediments, (iv) double binary MORB–mantle and mantle–sediments and (v) triple MORB–sediment–mantle. Double binary and triple mixing trends are transient and characterize relatively early stages of subduction. In contrast, steady-state binary mantle–MORB and MORB–sediments trends are typical for mature subduction zones with respectively low and high intensity of sedimentary melange subduction. Predictions from our GcTM models are in agreement with Pb isotopic data from some natural subduction zones.

Porosity waves are a mechanism by which fluid generated by devolatilization and melting, or trapped during sedimentation, may be expelled from ductile rocks. The waves correspond to a steady-state solution to the coupled hydraulic and rheologic equations that govern flow of the fluid through the matrix and matrix deformation. This work presents an intuitive analytical formulation of this solution in one dimension that is general with respect to the constitutive relations used to define the viscous matrix rheology and permeability. This generality allows for the effects of nonlinear viscous matrix rheology and disaggregation. The solution combines the porosity dependence of the rheology and permeability in a single hydromechanical potential as a function of material properties and wave velocity. With the ansatz that there is a local balance between fluid production and transport, the solution permits prediction of the dynamic variations in permeability and pressure necessary to accommodate fluid production. The solution is used to construct a phase diagram that defines the conditions for smooth pervasive flow, wave-propagated flow, and matrix fluidization (disaggregation). The viscous porosity wave mechanism requires negative effective pressure to open the porosity in the leading half of a wave. In nature, negative effective pressure may induce hydrofracture, resulting in a viscoplastic compaction rheology. The tubelike porosity waves that form in such a rheology channelize fluid expulsion and are predicted by geometric argumentation from the one-dimensional viscous solitary wave solution.

Analysis of lunar laser ranging and seismic data has yielded evidence that has been interpreted to indicate a molten zone in the lowermost mantle overlying a fluid core. Such a zone provides strong constraints on models of lunar thermal evolution. Here we determine thermochemical and physical structure of the deep Moon by inverting lunar geophysical data (mean mass and moment of inertia, tidal Love number, and electromagnetic sounding data) in combination with phase-equilibrium computations. Specifically, we assess whether a molten layer is required by the geophysical data. The main conclusion drawn from this study is that a region with high dissipation located deep within the Moon is required to explain the geophysical data. This region is located within the mantle where the solidus is crossed at a depth of ∼1200 km (≥1600 ◦ C). Inverted compositions for the partially molten layer (150-200 km thick) are enriched in FeO and TiO2 relative to the surrounding mantle. The melt phase is neutrally buoyant at pressures of ∼4.5-4.6 GPa but contains less TiO2 (

Variable morphologies of subducting slabs are observed in tomographic images of the mantle transition zone (MTZ), where slabs appear to stagnate in the MTZ or enter the lower mantle. These contrasting morphologies of subducting slabs are dependent on the joint effects of various dynamic, kinematic and geometric factors. Force balance analysis indicates that the favorable conditions of slab stagnation in the MTZ include old/cold slab subducting into the mantle with large Clapeyron slopes at 410 km and 660 km discontinuities, as well as the significant trench retreat and shallow dip angle. However, these conditions are often not achieved together for specific subduction zones on the Earth, which thus require systematic studies to distinguish their relative effects. Here, we analyze the slab mode selection in the MTZ based on coupled petrological-thermomechanical numerical model. The model results indicate that (1) water activity and partial melting weaken the subduction channel and form a hot and weak mantle wedge beneath the island arc that affects slab dynamics. (2) The Clapeyron slope of phase transition at 660 km can significantly contribute to the slab stagnation in the MTZ, whereas the Clapeyron slope at 410 km does not change the general mode selection, but does affect the trench motion and further the length of flattened slab. (3) A sharp viscosity jump between the lower and upper mantles can promote slab stagnation in the MTZ, which has a similar effect with a strong viscosity-depth gradient. (4) Fast trench retreat is the most critical factor controlling slab stagnation, especially the long slab flattening in the MTZ. (5) The age/thickness of converging plates can also influence the slab/MTZ interaction by modifying the slab dip angle and trench motion. (6) A thin, weak layer at the bottom of MTZ does not play significant roles in the slab mode selection. The combined force balance analysis and numerical studies are compared with the comprehensive observations of natural subduction zones, which improve understanding of the dynamics of slab/MTZ interaction and the resulting variability of subducting slab morphologies.

A model for the incorporation of alumina in FeO–MgO–Al 2 O 3 –SiO 2 –H 2 O (FMASH) serpentinites has been developed by considering ideal Tschermak (Al 2 Mg − 1 Si − 1 ) solid solution in antigorite. The antigorite model has been calibrated by fitting the experimental conditions for the decomposition of antigorite to chlorite + olivine + orthopyroxene + fluid in the FMASH system. The antigorite Al-contents predicted with this model are in agreement with natural observations and suggest a maximum alumina solubility in antigorite of 3.6 wt.% Al 2 O 3 at 20 kbar–650 °C and of 4.5 wt.% Al 2 O 3 at 3 kbar–560 °C. In the assemblage antigorite–olivine–chlorite–fluid, the Al-content of antigorite is buffered and temperature sensitive. This temperature sensitivity is the basis for a serpentinite geothermometer at greenschist, amphibolite and eclogite facies conditions. The buffered assemblage is stable in harzburgite compositions for relatively moderate amounts of Al 2 O 3 (> 1.8 wt.%) and is widespread in lherzolites, where it occurs together with diopside or, in a narrow temperature field, with tremolite.

[1] Traditional inversion techniques applied to the problem of characterizing the thermal and compositional structure of the upper mantle are not well suited to deal with the nonlinearity of the problem, the trade-off between temperature and compositional effects on wave velocities, the nonuniqueness of the compositional space, and the dissimilar sensitivities of physical parameters to temperature and composition. Probabilistic inversions, on the other hand, offer a powerful formalism to cope with all these difficulties, while allowing for an adequate treatment of the intrinsic uncertainties associated with both data and physical theories. This paper presents a detailed analysis of the two most important elements controlling the outputs of probabilistic (Bayesian) inversions for temperature and composition of the Earth's mantle, namely the a priori information on model parameters, ρ(m), and the likelihood function, L(m). The former is mainly controlled by our current understanding of lithosphere and mantle composition, while the latter conveys information on the observed data, their uncertainties, and the physical theories used to relate model parameters to observed data. [2] The benefits of combining specific geophysical datasets (Rayleigh and Love dispersion curves, body wave tomography, magnetotelluric, geothermal, petrological, gravity, elevation, and geoid), and their effects on L(m), are demonstrated by analyzing their individual and combined sensitivities to composition and temperature as well as their observational uncertainties. The dependence of bulk density, electrical conductivity, and seismic velocities to major-element composition is systematically explored using Monte Carlo simulations. We show that the dominant source of uncertainty in the identification of compositional anomalies within the lithosphere is the intrinsic nonuniqueness in compositional space. A general strategy for defining ρ(m) is proposed based on statistical analyses of a large database of natural mantle samples collected from different tectonic settings (xenoliths, abyssal peridotites, ophiolite samples, etc.). This strategy relaxes more typical and restrictive assumptions such as the use of local/limited xenolith data or compositional regionalizations based on age-composition relations. We demonstrate that the combination of our ρ(m) with a L(m) that exploits the differential sensitivities of specific geophysical observables provides a general and robust inference platform to address the thermochemical structure of the lithosphere and sublithospheric upper mantle. An accompanying paper deals with the integration of these two functions into a general 3-D multiobservable Bayesian inversion method and its computational implementation.

Here we present a 3-D multi-observable probabilistic inversion method, particularly designed for high-resolution (regional) thermal and compositional mapping of the lithosphere and sub-lithospheric upper mantle that circumvents the problems associated with traditional inversion methods. The key aspects of the method are as follows: (a) it exploits the increasing amount and quality of geophysical datasets; (b) it combines multiple geophysical observables (Rayleigh and Love dispersion curves, body-wave tomography, magnetotelluric, geothermal, petrological, gravity, elevation, and geoid) with different sensitivities to deep/shallow, thermal/compositional anomalies into a single thermodynamic- geophysical framework; (c) it uses a general probabilistic (Bayesian) formulation to appraise the data; (d) no initial model is needed; (e) compositional a priori information relies on robust statistical analyses of a large database of natural mantle samples; and (f) it provides a natural platform to estimate realistic uncertainties. In addition, the modular nature of the method/algorithm allows for incorporating or isolating specific forward operators according to available data. The strengths and limitations of the method are thoroughly explored with synthetic models. It is shown that the a posteriori probability density function (i.e., solution to the inverse problem) satisfactorily captures spatial variations in bulk composition and temperature with high resolution, as well as sharp discontinuities in these fields. Our results indicate that only temperature anomalies of ΔT a 150°C and large compositional anomalies of ΔMg# > 3 (or bulk ΔAl 2O3 > 1.5) can be expected to be resolved simultaneously when combining high-quality geophysical data. This resolving power is sufficient to explore some long-standing problems regarding the nature and evolution of the lithosphere (e.g., vertical stratification of cratonic mantle, compositional versus temperature signatures in seismic velocities, etc) and offers new opportunities for joint studies of the structure of the upper mantle with unprecedented resolution. © 2013. American Geophysical Union. All Rights Reserved.

Porosity waves are a mechanism by which fluid generated by devolatilization and melting, or trapped during sedimentation, may be expelled from ductile rocks. The waves correspond to a steady-state solution to the coupled hydraulic and rheologic equations that govern flow of the fluid through the matrix and matrix deformation. This work presents an intuitive analytical formulation of this solution in one dimension that is general with respect to the constitutive relations used to define the viscous matrix rheology and permeability. This generality allows for the effects of nonlinear viscous matrix rheology and disaggregation. The solution combines the porosity dependence of the rheology and permeability in a single hydromechanical potential as a function of material properties and wave velocity. With the ansatz that there is a local balance between fluid production and transport, the solution permits prediction of the dynamic variations in permeability and pressure necessary to accommodate fluid production. The solution is used to construct a phase diagram that defines the conditions for smooth pervasive flow, wave-propagated flow, and matrix fluidization (disaggregation). The viscous porosity wave mechanism requires negative effective pressure to open the porosity in the leading half of a wave. In nature, negative effective pressure may induce hydrofracture, resulting in a viscoplastic compaction rheology. The tubelike porosity waves that form in such a rheology channelize fluid expulsion and are predicted by geometric argumentation from the one-dimensional viscous solitary wave solution.

To constrain the thermodynamics and melting relations of the siderite–magnesite (FeCO3–MgCO3) system, 27 piston cylinder experiments were conducted at 3.5 GPa and 1170–1575 °C. Fe-rich compositions were also investigated with 13 multi-anvil experiments at 10, 13.6 and 20 GPa, 1500–1890 °C. At 3.5 GPa, the solid solution siderite–magnesite coexists with melt over a compositional range of X Mg (=Mg/(Mg + Fetot)) = 0.38–1.0, while at ≥10 GPa solid solution appears to be complete. At 3.5 GPa, the system is pseudo-binary because of the limited stability of siderite or liquid FeCO3, Fe-rich carbonates decomposing at subsolidus conditions to magnetite–magnesioferrite solid solution, graphite and CO2. Similar reactions also occur with liquid FeCO3 resulting in melt species with ferric iron components, but the decomposition of the liquid decreases in importance with pressure. At 3.5 GPa, the metastable melting temperature of pure siderite is located at 1264 °C, whereas pure magnesite melts at 1629 °C. The melting loop is non-ideal on the Fe side where the dissociation reaction resulting in Fe3+ in the melt depresses melting temperatures and causes a minimum. Over the pressure range of 3.5–20 GPa, this minimum is 20–35 °C lower than the (metastable) siderite melting temperature. By merging all present and previous experimental data, standard state (298.15 K, 1 bar) thermodynamic properties of the magnesite melt (MgCO3L) end member are calculated and the properties of (Fe,Mg)CO3 melt fit by a regular solution model with an interaction parameter of −7600 J/mol. The solution model reproduces the asymmetric melting loop and predicts the thermal minimum at 1240 °C near the siderite side at X Mg = 0.2 (3.5 GPa). The solution model is applicable to pressures reaching to the bottom of the upper mantle and allows calculation of phase relations in the FeO–MgO–O2–C system.

Natural moissanite (SiC) is reported from mantle-derived samples ranging from lithospheric mantle keel diamonds to serpentinites to podiform chromitites in ophiolites related to suprasubduction zone settings (Luobusa, Dongqiao, Semail, and Ray-Iz). To simulate ultra-reducing conditions and the formation of moissanite, we compiled thermodynamic data for alloys (Fe–Si–C and Fe–Cr), carbides (Fe3C, Fe7C3, SiC), and Fe-silicides; these data were augmented by commonly used thermodynamic data for silicates and oxides. Computed phase diagram sections then constrain the P–T–fO2 conditions of SiC stability in the upper mantle. Our results demonstrate that: Moissanite only occurs at oxygen fugacities 6.5–7.5 log units below the iron–wustite buffer; moissanite and chromite cannot stably coexist; increasing pressure does not lead to the stability of this mineral pair; and silicates that coexist with moissanite have X Mg > 0.99. At upper mantle conditions, chromite reduces to Fe–Cr alloy at fO2 values 3.7–5.3 log units above the moissanite-olivine-(ortho)pyroxene-carbon (graphite or diamond) buffer (MOOC). The occurrence of SiC in chromitites and the absence of domains with almost Fe-free silicates suggest that ultra-reducing conditions allowing for SiC are confined to grain scale microenvironments. In contrast to previous ultra-high-pressure and/or temperature hypotheses for SiC origin, we postulate a low to moderate temperature mechanism, which operates via ultra-reducing fluids. In this model, graphite-/diamond-saturated moderately reducing fluids evolve in chemical isolation from the bulk rock to ultra-reducing methane-dominated fluids by sequestering H2O into hydrous phases (serpentine, brucite, phase A). Carbon isotope compositions of moissanite are consistent with an origin of such fluids from sediments originally rich in organic compounds. Findings of SiC within rocks mostly comprised by hydrous phases (serpentine + brucite) support this model. Both the hydrous phases and the limited diffusive equilibration of SiC with most minerals in the rocks indicate temperatures below 700–800 °C. Moissanite from mantle environments is hence a mineral that does not inform on pressure but on a low to moderate temperature environment involving ultra-reduced fluids. Any mineral in equilibrium with SiC could only contain traces of Fe2+ or Cr3+.

Alpine deformation in the Grimsel granodiorite (Aar massif, Central Alps) at greenschist facies conditions (6.5 ± 1 kbar for 450� C±2 5� C) is characterized by the development of a network of centimetre to decametre localized shear zones that surround lenses of undeformed granodiorite. Localization of deformation is assumed to be the result of a first stage of extreme localization on brittle precursors (nucleation stage) followed by a transition to ductile deformation and lateral propagation into the weakly deformed granodiorite (widening stage). A paradox of this model is that the development of the ductile shear zone is accompanied by the crystallization of large amounts of phyllosilicates (white mica and chlorite) that maintains a weak rheology in the localized shear zone relative to the host rock so that deformation is localized and prevents shear zone widening. We suggest that chemical processes, and more particularly, the metamorphic reactions and metasomatism occurring during re-equilibration of the metastable magmatic assemblage induced shear zone widening at these P-T-X conditions. These processes (reactions and mass transfer) were driven by the chemical potential gradients that developed between the thermodynamically metastable magmatic assemblage at the edge of the shear zone and the stable white mica and chlorite rich ultramylonite formed during the first stage of shear zone due to localized fluid infiltration metasomatism. P-T and chemical potential projections and sections show that the process of equilibration of the wall rocks (l-l path) occurs via the reactions: kf + cz + ab + bio + MgO + H2O = mu + q + CaO + Na2O and cz + ab + bio + MgO + H2O = chl + mu + q + CaO + Na2O. Computed phase diagram and mass balance calculations predict that these reactions induce relative losses of CaO and Na2 Oo f� 100% and � 40% respectively, coupled with hydration and a gain of � 140% for MgO. Intermediate rocks within the strain gradient (ultramylonite, mylonite and orthogneiss) reflect various degrees of re-equilibration and metasomatism. The softening reaction involved may have reduced the strength at the edge of the shear zone and therefore promoted shear zone widening. Chemical potential phase diagram sections also indicate that the re-equilibration process has a strong influence on equilibrium mineral compositions. For instance, the decrease in Si-content of phengite from 3.29 to 3.14 p.f.u, when white mica is in equilibrium with the chlorite-bearing assemblage, may be misinterpreted as the result of decompression during shear zone development while it is due only to syn-deformation metasomatism at the peak metamorphic condition. The results of this study suggest that it is critical to consider chemical processes in the formation of shear zones particularly when deformation affects metastable assemblages and mass transfer are involved.

Knowledge of melt transfer within the mantle is primarily from the study of mid-ocean ridge settings, leaving elusive the process of sub-arc melt transfer. The sub-arc mantle section of the Sapat Complex (Kohistan-Pakistan) exposes coarse-grained dunites that contain clinopyroxene-enriched zones, which, in turn, contain gabbroic lenses. Structural, petrological, and chemical relationships indicate that the clinopyroxene-enriched zones reflect a continuum of melt transport mechanisms between pervasive percolation and fully segregated melt flow. This spectrum of mechanisms and the petrographic features at Sapat are explained by mechanical flow instabilities (porosity waves) that cause channeling of pervasively distributed melt.

The origin of the ancient martian crustal dichotomy and the massive magmatic province of Tharsis remains an open problem. Here, we explore numerically a hypothesis for the origin of these two features involving both exogenic and endogenic processes. We propose a giant impact event during the late stage of planetary formation as the source of the southern highland crust. In a second stage, the extraction of excess heat by vigorous mantle convection on the impacted hemisphere leads to massive magmatism, forming a distinct Tharsis-like volcanic region. By coupling short-term and long-term numerical simulations, we are able to investigate both the early formation as well as the 4.5 Gyr evolution of the martian crust. We demonstrate numerically that this exogenic–endogenic hypothesis is in agreement with observational data from Mars.

SUMMARY Geodynamic models incorporating metamorphic phase transformations almost invariably assume the validity of the Boussinesq approximation that violates conservation of mass. In such models metamorphic density changes take place without volumetric effects. We assess the impact of the Boussinesq approximation by comparing models of orogeny accompanied by lower crustal eclogitization both with and without the approximation. Our results demonstrate that the approximation may cause errors approaching 100 per cent in characteristic measures of orogenic shape. Mass conservation errors in Boussinesq models amplify with model time. Mass conservative models of metamorphism are therefore essential to understand long-term tectonic evolution and to assess the importance of the different geodynamic processes.

Experiments on a synthetic, foliated, quartz-muscovite rock were undertaken to evaluate and explain the influence of shear deformation on partial melting at 750 °C and 300 MPa. Torsion experiments were conducted at a constant strain rate of 3 × 10−4 s−1 with the initial foliation parallel and orthogonal to the direction of applied angular displacement. The non-equilibrium melting reaction forms melt, sillimanite, biotite, and K-feldspar at the expense of muscovite and quartz. In both static and torsion experiments, for the first 1.5 h the rate of melting is identical, thereafter melting accelerates in the torsion experiments. The rate of melting is highest in torsion experiments with the sample foliation orthogonal to the shear plane. First order models for shear heating, surface and strain energy, and mean stress effects suggest that none of these effects explain the difference in melting rate between static and torsion experiments. We conclude that the most probable explanation for the correlation between reaction rate and shear deformation is that shear deformation lowers the effective viscosity of the rock matrix. The reduction in effective viscosity facilitates the elimination of the grain-scale mean-stress perturbations caused by the volume change of the melting reaction. As these perturbations inhibit reaction, their elimination by ductile dilational deformation accelerates the melting process.

Large volumes of greenhouse gases such as CH4 and CO2 form by contact metamorphism of organic-rich sediments in aureoles around sill intrusions in sedimentary basins. Thermogenic gas generation and dehydration reactions in shale are treated numerically in order to quantify basin-scale devolatilization. We show that aureole thicknesses, defined as the zone of elevated metamorphism relative to the background level, vary within 30–250% of the sill thickness, depending on the temperature of the host-rock and intrusion, besides the sill thickness. In shales with total organic carbon content of >5 wt.%, CH4 is the dominant volatile (85–135 kg/m3) generated through organic cracking, relative to H2O-generation from dehydration reactions (30–110 kg/m3). Even using conservative estimates of melt volumes, extrapolation of our results to the scale of sill complexes in a sedimentary basin indicates that devolatilization can have generated ∼2700–16200 Gt CH4 in the Karoo Basin (South Africa), and ∼600–3500 Gt CH4 in the Voring and More basins (offshore Norway). The generation of volatiles is occurring on a time-scale of 10–1000 years within an aureole of a single sill, which makes the rate of sill emplacement the time-constraining factor on a basin-scale. This study demonstrates that thousands of gigatons of potent greenhouse gases like methane can be generated during emplacement of Large Igneous Provinces in sedimentary basins.

Three-dimensional thermo-chemical mantle convection simulations with mineral assemblages self-consistently calculated using free energy minimization are used to check the sensitivity of model behavior to the assumed compositions of mid-ocean ridge basalt (MORB) and harzburgite. In addition to five-oxide CaO–FeO–MgO–Al 2 O 3 –SiO 2 (CFMAS) compositions, we test the effect of a more realistic compositional model by adding a sixth oxide Na 2 O (NCFMAS) with three compositions. Results indicate that thermo-chemical structures are quite sensitive to variations in MORB composition of the order 1–2% oxide fraction, particularly FeO and Al 2 O 3 . Differences occur in (i) the amount of compositional stratification around 660 km depth caused by the inversion of the MORB-harzburgite density difference between 660 and 740 km depth, which is different in magnitude and depth extent between the different tested compositions, and (ii) in the degree of MORB segregation above the CMB, which is related to differences in the MORB-harzburgite density difference in the deep mantle. While improving the realism of the model by including Na 2 O tends to reduce the MORB-harzburgite density difference at most pressure and temperature conditions, the differences in behavior among three NCFMAS compositions are at least as large as between either NCFMAS and the CFMAS composition, and are also related to differences in the (pressure and temperature) stability range of the post-perovskite phase between the different compositions. Comparing model spectra to those of seismic tomography using spectral heterogeneity maps, NCFMAS compositions provide a better match to seismic tomography but in all cases there is too much heterogeneity at mid lower mantle depths compared to typical seismic tomographic models, which implies that less CMB basalt segregation occurs in Earth than in the convection models.

Metamorphic devolatilization generates fluid and grain-scale porosity. Evidence for high fluid pressure indicates that devolatilization occurs under poorly drained conditions. Under such conditions, fluid expulsion is limited by the capacity of the reacted rocks to resist compaction or by the rate at which deformation modifies the permeability of the overlying rocks. In the former case, the compaction timescale must be greater than the metamorphic timescale, and flow patterns are dictated by details of rock permeability. The alternative is that compaction processes are fast relative to metamorphism. In this case, flow is compaction driven and accomplished by waves of fluid-filled porosity.

We analyze the mechanical behavior of a two-phase system consisting of rigid grains and an interconnected pore fluid. For this purpose we use 2D direct numerical simulations on the spatial scale of individual grains for Newtonian and non-Newtonian fluid rheology. By using the stress–strain rate relation we derive scaling laws for effective viscosity of two-phase particle suspensions. We demonstrate that the effective rheology of the assemblage is non-Newtonian only if the fluid has a non-Newtonian rheology. At small fluid fraction, inter-granular strain rates are up to 3 orders of magnitude higher than the applied background strain rate. We suggest that this effect explains the experimentally observed change at higher strain rates in rheology, from Newtonian to non-Newtonian aggregate rheology. To establish the conditions at which the fluid–solid aggregate deforms coherently as a consequence of Rayleigh–Taylor instabilities we studied flow patterns of particle suspensions and characterized them as a function of fluid fraction, viscosity, density, shape and size of the grains. From initial conditions with homogeneously distributed grains and interstitial fluid above a layer of pure fluid, our results show that the Rayleigh–Taylor instability dominates for moderate to large fluid fractions. At large fluid fractions, we observed a transition to a Stokes suspension mode, in which grains do not interact but sink independently. An analytical expression is derived that predicts the transition from Rayleigh–Taylor instability to Stokes suspension mode. The transition is a function of fluid fraction, radius of the grains, height of the interface and initial amplitude. Systematic numerical simulations are in good agreement with the analytical predictions.

Slab detachment or breakoff is directly associated with phenomena like morphological orogenesis,occurrenceofearthquakesandmagmatism.Atdepththedetachmentprocessisslowand characterized by viscous rheolgy, whereas closer to the surface the process is relatively fast and plastic. Using a 2D mantle model 1500 km deep and 4000 km wide we investigated, with finite- difference and marker-in-cell numerical techniques, the impact of slab age, convergence rate and phase transitions on the viscous mode of slab detachment. In contrast to previous studies exploring simplifiedbreakoffmodelsinwhichtheblockageresponsibleforinducingbreakoffiskinematically prescribed, we constructed a fully dynamic coupled petrological-thermomechanical model of viscousslabbreakoff.Inthismodel,forcedsubductionofa700 km-longoceanicplatewasfollowed bycollisionoftwocontinentalplatesandspontaneousslabblockingresultingfromthebuoyancyof thecontinentalcrustonceithadbeensubductedtoadepthof100-124 km.Typically,fivephasesof model development can be distinguished: (a) oceanic slab subduction and bending; (b) continental collision initiation followed by the spontaneous slab blocking, thermal relaxation and unbending - in experiments with old oceanic plates in this phase slab roll-back occurs; (c) slab stretching and necking; (d) slab breakoff and accelerated sinking; and (e) post-breakoff relaxation. Our experiments confirm a correlation between slab age and the time of spontaneous viscous breakoff as previously identified in simplified breakoff models. The results also demonstrate a non-linear dependence of the duration of the breakoff event on slab age: a positive correlation being characteristic of young (,50 Ma) slabs while for older slabs the correlation is negative. The increasing duration of the breakoff with slab age in young slabs is attributed to the slab thermal thickness, which increases both the slab thermal relaxation time and duration of the neckingprocess.Inolderslabsthistendencyiscounteractedbynegativeslabbuoyancy,whichgen- eratehigherstressesthatfacilitateslabneckingandbreakoff.Apredictionfromourbreakoffmodels isthattheolivine-wadsleyitetransitionplaysanimportantroleinlocalizingviscousslabbreakoffat depths of 410-510 km due to the buoyancy effects of the transition.

Simple polythermal extensions to two widely used isothermal equations of state, the Murnaghan and the Birch-Murnaghan, can lead to non-physical material behavior without proper parameterization: the thermal expansivity at high pressure can become negative. We show how this arises and propose a remedy using an approximation to the thermal relaxation of the bulk modulus. Using the revised equation of state for thermodynamic equilibrium calculations leads to low-pressure and -temperature behavior indistinguishable from the unmodified equation of state, yet extrapolates to high pressure and temperature without non-physical behavior.

article i nfo A new approach is presented for modelling Cr-rich peridotite compositions and garnet-spinel compositions found in diamonds and xenoliths at conditions relevant to the deep continental Earth. Using recent experimental data, it is now possible to calculate phase relations and mineral compositions relevant to pressures, temperatures, and compositions of the deep lithospheric Earth using free energy minimization techniques. Here we present calculated phase relations in Cr-rich mantle compositions from pressures of 20- 60 kbar, and temperatures 800-1400 °C. The model is successful at modelling a wide range of natural mineral compositions which are found as xenoliths in diamond-bearing kimberlites from South Africa, and is illustrated using suites of Cr-rich xenoliths from near Kimberley, South Africa. The model can explain and quantify instances of garnet zonation in naturally occurring mantle rocks as a consequence of pressure- temperature re-equilibration; and shows that not all garnet zonations result from metasomatic processes. This sheds further light on peridotitic diamond inclusions and their abundances, and allows further quantitative constraints to be applied to diamond-indicator mineral chemistry.

The Palghat–Cauvery suture zone in southern India separates Archaean crustal blocks to the north and the Proterozoic Madurai block to the south. Here we present the first detailed study of a partially retrogressed eclogite (from within the Sittampundi anorthositic complex in the suture zone) that occurs as a 20-cm wide layer in a garnet gabbro layer in anorthosite. The eclogite largely consists of an assemblage of coexisting porphyroblasts of almandine–pyrope garnet and augitic clinopyroxene. However, a few garnets contain inclusions of omphacite. Rims and symplectites composed of Na–Ca amphibole and plagioclase form a retrograde assemblage. Petrographic analysis and calculated phase equilibria indicate that garnet–omphacite–rutile–melt was the peak metamorphic assemblage and that it formed at ca . 20 kbar and above 1000 °C. The eclogite was exhumed on a very tight hairpin-type, anticlockwise P–T path, which we relate to subduction and exhumation in the Palghat–Cauvery suture zone. The REE composition of the minerals suggests a basaltic oceanic crustal protolith metamorphosed in a subduction regime. Geological–structural relations combined with geophysical data from the Palghat–Cauvery suture zone suggest that the eclogite facies metamorphism was related to formation of the suture zone. Closure of the Mozambique Ocean led to development of the suture zone and to its western extension in the Betsimisaraka suture of Madagascar.

Double subduction is a geodynamic process in which two plates following each other are synchronously subducted. Double subductions are known for both modern (Izu-Bonin-Marianas and Ryukyu arcs) and ancient (West Himalaya collision zone) plate tectonics. However, our knowledge about this process is limited to conceptual schemes and some restricted analogue experiments. In order to fill this gap we performed 2D numerical experiments using a coupled petrological–thermomechanical approach based on finite differences and marker-in-cell techniques combined with thermodynamic database for the mantle. We investigated the influence of convergence rate, intermediate plate length, activation volume of the mantle dislocation creep and age of the lithosphere. Based on these experiments we conclude that: (A) Subduction rates at two zones running in parallel differ and vary in time even when the total convergence rate remains constant. Supremacy of either subduction zone depends on physical parameters such as (i) relative rates of the plates, (ii) slab ages and (iii) length of the middle plate. (B) Subduction dynamics of the double subduction system involves several processes unknown in simple subduction systems, such as (i) eduction (i.e. “un-subduction”), (ii) subduction re-initiation, (iii) subduction flip triggered by shallow slab breakoff and (iv) turn-over of detached slabs to up-side-down attitudes. (C) Simulated tomographic structures related to slab propagation account for both penetration and non-penetration of the 660 km discontinuity. Non-penetration is favored by (i) low convergence rate, (ii) faster relative movement of the overriding plate, (iii) young age of the subducting slab and (iv) up-side-down turn-over of detached slab.

We investigate crustal growth processes on the basis of a 2D coupled geochemical–petrological–thermomechanical numerical model of retreating intraoceanic subduction. The model includes spontaneous slab retreat and bending, subducted crust dehydration, aqueous fluid transport, mantle wedge melting, and melt extraction resulting in crustal growth. Our numerical experiments show that the rate of plate retreat influences both the rate of crust formation and composition of newly formed crust. The rate of trench retreat, which is a manifestation of subduction rate, strongly varies with time: retreat rates slow (from 7 cm/a to 1 cm/a) shortly (in a few Ma) after the beginning of subduction and then increase (up to 4 cm/a). Subsequently two different scenarios can be distinguished: (1) subduction rate decay that leads ultimately to cessation of subduction, (2) subduction rate acceleration (up to 12 cm/a), which stabilizes subduction. The rate of crust formation positively correlates with rate of trench retreat. Modelled average rates of crustal growth (30–50 km3/km/Ma) do not include effects of dry mantle melting and are close to the lower edge of the observed range of rates for real arcs (40–180 km3/(km Ma)). The composition of new crust depends strongly on the evolution of subduction. Four major magmatic sources can contribute to the formation of the crust: (1) hydrated partially molten peridotite of the mantle wedge, (2) melted subducted sediments, (3) melted subducted basalts, (4) melted subducted gabbro. Crust produced from the first source is always predominant. In all studied cases it appears shortly after beginning of subduction and is a persistent component so long as subduction remains active. Significant amount of crust produced from other three sources appear (i) in the beginning of subduction due to the melting of the slab “nose” and (ii) at later stages when subduction velocity is low (

We present a tool for coupling thermochemistry with mechanics. Thermodynamic potential functions are used to calculate reversible material properties such as thermal expansion coefficient, specific heat, elastic shear modulus, bulk modulus and density. These material properties are thermodynamically self consistent. Transport properties such as thermal conductivity (diffusivity) and melt viscosity are also included, but these are derived from laboratory experiments. The transport properties are included to provide a reference database as a common standard of material properties necessary for comparing geological, geodynamic and geotechnical calculations. We validate the chemically derived elastic material properties by comparing computed seismic velocities for a pyrolitic composition to the seismic models PREM and ak135.

Garnet‐clinopyroxene‐quartz granulites of the central Highland Complex of Sri Lanka preserve textural and compositional features indicative of high‐pressure, ultrahigh‐temperature (HP‐UHT) crustal metamorphism and multistage retrogression. Grains of the peak metamorphic assemblage, garnet‐clinopyroxene‐quartz, are commonly separated and embayed by late orthopyroxene‐plagioclase symplectites; however, in some domains, rare grain‐to‐grain associations of the peak assemblages are still preserved. Thermodynamic modeling in the CaO‐Na2O‐K2O‐FeO‐MgO‐Al2O3‐SiO2 system indicates peak metamorphic conditions of 12.5 kbar at 925°C. The temperature estimates using garnet and clinopyroxene core compositions are in the range 844°–982°C, in agreement with the thermodynamic modeling. In conclusion, the textural, geochemical, and thermodynamic modeling and thermobarometric data indicate a multistage decompression after HP‐UHT metamorphism. U‐Pb zircon (laser ablation–inductively coupled plasma mass spectrometry) ...

We use a coupled petrological–thermomechanical model of subduction with spontaneous slab bending to investigate magmatic productivity at active continental margins. The model is designed to simulate fossil Pacific-type margins that have a broad well-developed fore-arc accretionary wedge system. The degree of plate coupling strongly depends on the dimensionless ratio ( R H 2 O ) between the plate convergence rate and the water propagation velocity. Delamination of the slab from the overriding plate followed by trench retreat is common for models with relatively slow convergence rates ( R H 2 O 4 ). In contrast, higher convergence rates ( R H 2 O > 4 ) result in continuous plate coupling. The slab bending curvature increases with the length of the subducted plate. Periodic variations of the slab angle with time are observed at the later stages of subduction and become more conspicuous with depth. These variations are favoured by slower subduction rates and stronger oceanic crust. Two fundamentally different regimes of melt productivity are observed in numerical experiments and correlated with natural observations: (1) during continuous convergence with coupled plates (as in the Late Paleozoic margin of central Chile) the largest melt productivity occurs at the onset of subduction due to temporary steepening of the slab, productivity then decays rapidly with time due to flattening of the slab inclination angle, thereby precluding further formation of partially molten mantle wedge plumes. (2) Highest melt productivity is obtained in simulations associated with slab delamination and trench retreat. As a result an extension occurs with decoupling of plates that finally leads to the formation of a pronounced back-arc basin (similar to the Mesozoic margin of southernmost Chile). In this case melt production increases with time due to stabilization of slab inclination associated with upward asthenospheric mantle flow toward the spreading centre, thereby favouring the generation of hydrous partially molten plumes from the slab.