EMEvangelos Moulas

Evangelos Moulas

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Evangelos Moulas has a total of 119 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Comment on egusphere-2026-3963

    by Jacob Frasunkiewicz, Christian Schuler, Evangelos Moulas, Boris J. P. Kaus

    Material parameter values used in geoscientific simulations are often poorly constrained due to the sparse nature of the observations. In addition, the constitutive relationships used to describe natural processes typically involve nonlinearities, which results in large changes in simulation output when material parameters or initial conditions change. It is therefore important to have a tool that can efficiently determine the quantitative effect of parameter changes on the simulation output. The adjoint gradient method addresses this by delivering the gradient of an observation, here the result of the forward simulation, with respect to all material parameters at a computational cost independent of their number. Deriving the adjoint gradient analytically can be labor-intensive and in some cases must be re-derived when the constitutive relationships change. Here, we present a framework that avoids the re-derivation by applying automatic differentiation directly to the code, which is used for the forward simulation. The residual calculation contains all discrete operations needed to assemble the Jacobian of the forward problem and the parameter sensitivity matrix. The full gradient with respect to all spatially varying parameters then follows from a single solution of the (linear) adjoint problem, regardless of the dimension of the parameters space. We additionally give a geometric interpretation of the discrete calculation of the adjoint gradients, making its computational structure explicit to guide implementation in other codes. The framework is implemented in Julia and demonstrated on four problems of increasing complexity: transient heat diffusion in a petrological experimental setup, coupled fluid flow and heat transport in a Central European geothermal reservoir, thermo-visco-elasto-plastic Stokes flow applied to a rising mantle plume, and to localizing shear bands. We deliberately emphasize flexibility and transparency over computational efficiency, providing a compact and reproducible basis for sensitivity analysis.

  • Open Access

    Automatic adjoints without the algebra: flexible sensitivity analysis of nonlinear problems in the geosciences

    by Jacob Frasunkiewicz, Christian Schuler, Evangelos Moulas, Boris J. P. Kaus

    Material parameter values used in geoscientific simulations are often poorly constrained due to the sparse nature of the observations. In addition, the constitutive relationships used to describe natural processes typically involve nonlinearities, which results in large changes in simulation output when material parameters or initial conditions change. It is therefore important to have a tool that can efficiently determine the quantitative effect of parameter changes on the simulation output. The adjoint gradient method addresses this by delivering the gradient of an observation, here the result of the forward simulation, with respect to all material parameters at a computational cost independent of their number. Deriving the adjoint gradient analytically can be labor-intensive and in some cases must be re-derived when the constitutive relationships change. Here, we present a framework that avoids the re-derivation by applying automatic differentiation directly to the code, which is used for the forward simulation. The residual calculation contains all discrete operations needed to assemble the Jacobian of the forward problem and the parameter sensitivity matrix. The full gradient with respect to all spatially varying parameters then follows from a single solution of the (linear) adjoint problem, regardless of the dimension of the parameters space. We additionally give a geometric interpretation of the discrete calculation of the adjoint gradients, making its computational structure explicit to guide implementation in other codes. The framework is implemented in Julia and demonstrated on four problems of increasing complexity: transient heat diffusion in a petrological experimental setup, coupled fluid flow and heat transport in a Central European geothermal reservoir, thermo-visco-elasto-plastic Stokes flow applied to a rising mantle plume, and to localizing shear bands. We deliberately emphasize flexibility and transparency over computational efficiency, providing a compact and reproducible basis for sensitivity analysis.

  • Open Access

    Metamorphic evidence of rapid heat production during geological information

    by Evangelos Moulas

    The advent of inverse diffusion modelling in minerals has allowed the quantification of effective timescales and cooling rates of metamorphic assemblages. Such information can be proven very useful if combined with information from conventional thermobarometry. This is because, cooling rates at various geologic conditions can provide essential constraints on the temporal character of geodynamic models. Constraining the temporal character of metamorphic processes from natural data provides us with non-unique solutions because of the nature of the data examined. However, considering additional constraints from geophysical/geodynamic models helps reducing the uncertainty of the results since it allows the incorporation of additional (and independent) constraints. Additional constraints include, but are not limited to, erosion rates, maximum topography and maximum surface heat flux.In this work I present a systematic study from thermo-kinematic and thermo-mechanical models that aims in identifying the key parameters responsible for the preservation of steep compositional gradients in minerals. Results show that in active tectonic environments, fast advection is essential for the rapid transient cooling of metamorphic assemblages. However, for the same type of data, less extreme velocity values can be used if the effects of dissipative heating are included. Therefore, apart from the thermodynamic consistency, the consideration of dissipative heating during geological deformation allows the use of more realistic parameters in geodynamic models.

  • Open Access

    Cooling vs. heating histories of mantle peridotites revealed by multiple Fe-Mg exchange thermometers

    by Vasileios Giatros, Dimitrios Kostopoulos, Dimitrios Moutzouris, Evangelos Moulas, Panagiotis Pomonis, Argyrios Papadopoulos

    Mineral geothermometry is based on cation exchange between minerals and effectively reflects the closure temperature through kinetic processes. Each geothermometer is designed for a specific purpose and has its own merits. The use of different thermometers that employ cations of different diffusivities is extremely important in capturing “snapshots” of thermal events, which facilitate decoding cooling/heating pulses, ascent rates and residence times.Here we compare the results of Fe-Mg exchange thermometers between coexisting mineral pairs in spinel peridotites from different geotectonic environments and discuss the most important implications that stem from this. Three mineral pairs were considered, olivine-spinel (ol-spl), orthopyroxene-spinel (opx-spl) and orthopyroxene-clinopyroxene (opx-cpx) which show progressively higher Fe-Mg closure temperatures in the order listed. The first two thermometers constitute new calibrations constructed by us whilst the third is the formulation of Brey & Köhler, 1990. Application of the above thermometers to abyssal peridotites (abyssal) and peridotites exposed in oceanic forearc and backarc regions shows undisturbed cooling patterns for each setting, with higher mean opx-cpx temperatures followed by opx-spl and then by ol-spl temperatures. When these patterns are compared to those obtained for ophiolitic peridotites it becomes immediately apparent that temperature distributions in mantle peridotites from oceanic forearcs and oceanic backarcs exhibit great similarities with those from ophiolites for all 3 thermometers. Abyssal peridotites have distinctly higher mean temperatures suggesting that ophiolitic massifs have not been formed in major ocean basins but rather in oceanic forearc or backarc settings. Such a conclusion is also strongly supported by trace-element geochemistry of ophiolitic volcanic rocks.The above order of the three thermometers is, nonetheless, reversed when they are applied to peridotite xenoliths found in volcanic rocks along potential continental rift zones. The subcontinental lithospheric mantle is expected to cool normally through time hence display a cooling pattern like that observed for abyssal peridotites. In the case of the xenoliths however, where mantle pieces are collected and transported by hot magma at temperatures much higher than the ambient geotherm, their exposure to high temperatures reactivates diffusion so that the thermometer containing the mineral pair with the fastest Fe-Mg diffusion (i.e., ol-spl) slides uphill faster and records the highest temperatures. The very fact that this (counter-intuitive) reverse order of temperature distributions has been preserved, places further constraints on the time scales of xenolith transport as protracted times of magma storage and fractionation in crustal chambers would have led to subsequent cooling and obliteration of the temperature patterns observed.G. P. Brey & T. Köhler, 1990. Geothermobarometry in Four-phase Lherzolites II. New Thermobarometers, and Practical Assessment of Existing Thermobarometers. Journal of Petrology, Vol. 31, Part 6, pp. 1353-1378. DOI:10.1093/petrology/31.6.1353

  • Open Access

    Intense magmatic heating of oceanic lithosphere revealed by Hawaiian xenoliths

    by Dimitrios Kostopoulos, Evangelos Moulas, Panagiotis Pomonis, Argyrios Papadopoulos

    Mantle peridotite thermobarometry has been extensively used to establish the thermal state of the lithosphere. The pressure and temperature (P-T) information from these rocks is routinely compared to model geotherms and the goodness of fit can be used to identify critical parameters of the thermal model. This method implies that mineral thermobarometers do not have sufficient time to re-equilibrate during their transport to the surface and therefore have preserved the ambient geotherm of the source region.In this work we performed a systematic analysis on garnet peridotite xenoliths from Hawai’i. We employed well-established and new thermometers based on Ca-Mg, Cr-Al, Fe-Mg exchange reactions between opx-cpx, opx-grt and cpx-grt mineral pairs. Pressure was determined using the Al solubility between opx and grt. Our results reveal that the Hawaiian xenoliths fit oceanic geotherms that span from 45 to 25 Ma. This apparent fit is at odds with the well-established age of the oceanic lithosphere in the region (90 Ma).The discrepancy between the two age groups can be perfectly explained by the fact that the lithosphere beneath Hawai’i has experienced intense thermal perturbation during the ascent of the magmas to the surface. Our interpretation is in agreement with published plate flexure models that call on magma-assisted flexural weakening of the lithosphere.

  • Open Access

    Experimental constraints on the kinetics of the albite = jadeite + quartz reaction depending on grain size and reaction overstepping

    by Simon Schorn, Zongpeng Yang, Friedrich Hawemann, Stephan Buhre, Roman Botcharnikov, Evangelos Moulas

    Densification and deformation during eclogitization govern the strength and buoyancy of orogenic roots and the stability of mountain ranges over geological timespans. The breakdown of albite to jadeite + quartz represents a key end-member reaction that is associated with densification of about 20%; shear stresses induced by such volumetric changes may cause brittle failure and have been linked to intermediate-depth seismicity (Yamato et al., 2022). Eclogitization is a kinetically sluggish process that requires significant reaction overstepping and may proceed far beyond the equilibrium pressure–temperature conditions, and/or remain largely incomplete – particularly in fluid-deficient felsic crust as evidenced by field observations (Palin et al., 2017) and geophysical constraints (Hetényi et al., 2021). However, the kinetics of the albite = jadeite + quartz reaction is poorly constrained, especially regarding the roles of grain size, pressure–temperature overstepping, and reaction duration. To address this gap, we conducted high-pressure experiments using a piston-cylinder apparatus at the Institute of Geosciences, JGU Mainz. Natural albite crystals were crushed and sieved into grain size fractions between 50 and 500 µm, loaded into Au-capsules, and separated by Au-foils. A subset of experiments involved furnace-drying (~500°C) of the starting materials followed by hot-welding of the capsules to minimize atmospheric moisture contamination. In experimental stage I, pressure was initially set just below (~1 kbar) the albite = jadeite + quartz reaction boundary (Holland, 1980), followed by heating to target temperature. In stage II, pressure was increased at constant temperature to variable target pressures above the reaction to systematically explore the effect of reaction overstepping. Samples where quenched by power shutdown, and reaction progress was quantified using scanning electron microscopy (backscattered electron imaging and cathodoluminescence) based on the relative fractions of reactant albite and products jadeite–quartz. Preliminary results reveal highly variable degrees of reaction progress. Where present, jadeite–quartz occur as finely intergrown symplectites, typically decorating albite grains at the rims, as well as forming within larger albite grains. The latter textures indicate complications arising from fluid inclusions in the starting material. By combining constraints on P(T) overstep, grain size, and experimental run duration, we determine effective reaction rates for albite breakdown. These results provide end-member kinetic constraints on high-pressure transformation in fluid-deficient, coarse-grained felsic rocks, which constitute the bulk of many well-known (U)HP terranes such as the Western Gneiss Region (Norway) and the Dabie–Sulu belt (China).ReferencesHetényi, G. et al. (2021). Metamorphic transformation rate over large spatial and temporal scales constrained by geophysical data and coupled modelling. Journal of Metamorphic Geology, 39(9), 1131–1143.Holland, T. J. (1980). The reaction albite = jadeite + quartz determined experimentally in the range 600–1200°C. American Mineralogist, 65(1-2), 129–134.Palin, R. et al. (2017). Subduction metamorphism in the Himalayan ultrahigh-pressure Tso Morari massif: an integrated geodynamic and petrological modelling approach. Earth and Planetary Science Letters, 467, 108–119.Yamato, P. et al. (2022). Reaction-induced volume change triggers brittle failure at eclogite facies conditions. Earth and Planetary Science Letters, 584, 117520.

  • Open Access

    Thermodynamic Modelling of Mineral Growth – Implication for Metamorphic Rates

    by Evangelos Moulas

    Over the recent years, there has been a significant development in the numerical tools that are used in the thermodynamic and kinetic modelling of metamorphic assemblages. Inspired by the materials science community, different approaches of mineral growth have been suggested. Out of the multitude of approaches, two stand out as the most widely used: i) the phase field method and ii) the sharp interphase method. Both approaches have their advantages and disadvantages and can be used to tackle various problems.In this presentation, I will present the main methods used in mineral growth modelling along with their distinctive features. I will focus on the sharp-interface method since it allows direct comparison with mineral chemistry and thermodynamic data. Recent development on that field (e.g. Stroh et al., 2025) allows the forward modelling of growing/consuming crystals along with their diffusional response. This approach allows the hypothesis testing for various geothermobarometry and isotope chronology systems. The direct comparison of the measured mineral compositions together with their mineral equilibria modelling offers a self-consistent framework of the timescales of metamorphic processes. Using this framework, various scenarios are investigated and the potential pitfalls are discussed. ReferencesStroh, A., Aellig, P.S., Moulas, E., 2025. Numerical modelling of diffusion-limited mineral growth for geospeedometry applications. Geosci. Model Dev. 18, 10203–10220. https://doi.org/10.5194/gmd-18-10203-2025

  • Open Access

    Multiphase сompressible mantle convection – model formulation

    by Leonid Aranovich, Evangelos Moulas, Liudmila Khakimova, Yury Podladchikov

    Mantle convection governs the thermal and mechanical evolution of terrestrial planets and provides the long-wavelength engine for plate tectonics and upper-mantle deformation. While early geodynamic models treated the mantle as an incompressible, isoviscous fluid, realistic predictions require non-linear rheology and, in many settings, an explicit treatment of compressibility. Nevertheless, the Boussinesq approximation remains widely used because it neglects volumetric deformation and retains density variations only in the buoyancy term, simplifying the solution procedure. Recent studies indicate that Boussinesq models can reproduce first-order flow speeds in simple cases, but may introduce non-negligible errors in temperature and energetics, motivating the use of compressible formulations. In parallel, Earth’s mantle is inherently multiphase and reactive: mineral phase transitions and solid-solid reactions exchange latent heat and volume, modify buoyancy forces and dissipation patterns, and can alter the style and vigor of convection. Incorporating reaction effects into compressible mantle convection remains challenging, because governing equations are well established for single-phase fluids, but their upscaling and closure assumptions are not straightforward for polymineralic rocks.Here we present a model formulation for the convection of a multiphase material with applications to the Earth’s mantle. Starting from classical compressible thermomechanical balance laws, we explicitly state the assumptions leading to several compressible approximations used in geodynamics. We then derive the additional closures required to extend the system to reactive polymineralic rocks under a single-velocity approximation (i.e., without explicit melt/fluid percolation), enabling inclusion of both reaction enthalpy and reaction-induced volume change effects.We compute new two-dimensional solutions of fully compressible convection on Cartesian grids using a pseudo-transient iterative strategy that stabilizes the strongly coupled, highly non-linear system. The simulations confirm the strongly localized nature of adiabatic and dissipative heating and show that mineral reactions can further amplify this localization. Endothermic reactions generally damp convective vigor and can promote transient layering; however, for realistic mantle compositions, layering tends to be intermittent rather than persistent over long times.Finally, the same framework can be extended to coupled models of multicomponent aqueous-fluid migration with (de)hydration reactions, where fluid-rock interactions within vein networks are tracked together with density and composition changes of the coexisting phases; thermodynamic calculations show that fluid SiO₂ strongly controls the reacting mineral assemblage, and, for example, decompression from 2.5 to 0.2 GPa at 700 °C can shift a six-mineral system to a three-phase assemblage, increasing the fluid Si/O ratio and pre-conditioning the mantle protolith for felsic melt generation.

  • Open Access

    Modelling volcanic eruptions from the volcano to the atmosphere

    by Hugo Dominguez, Boris Kaus, Hendrik Ranocha, Evangelos Moulas, Ivan Utkin

    Volcanic eruptions are complex processes involving multiple interacting phases, such as ascending magma, exsolved gases, deformation of the host rock and atmospheric dynamics. Typically, numerical models treat the sub-aerial eruptive column and the subsurface rock deformation as distinct domains due to the different timescales and material properties involved. This study presents a 2D numerical framework that couples the propagation of atmospheric waves with the elastic deformation of the host rock via a unified formulation. Using a finite volume method to solve the conservative form of the mass and momentum equations on a staggered grid, we demonstrate that this formulation can correctly predict the localisation of shock waves in the atmosphere, as well as the propagation of elastic waves in the host rock. Furthermore, we show that a single discretisation can capture both the conversion of acoustic waves into elastic waves from the atmosphere to the host rock, and the reverse process. This provides a foundation for fully coupled models of explosive volcanic events to potentially offers new insights into the interaction between the subsurface and the atmosphere during these processes.

  • Open Access

    Hamiltonian Monte Carlo applied to inverse petrological problems

    by Dimitrios Moutzouris, Annalena Stroh, Simon Schorn, Evangelos Moulas

    The petrogenesis of rocks can be investigated by integrating multiple datasets and numerical methods. Petrology benefits from the abundance of field, petrographic, geochronological and geochemical data. Moreover, the growing number of numerical methods opens new opportunities for inversion. Inversion allows the quantification of useful parameters, offering deeper insights into natural processes. However, an increasing number of parameters to invert for is accompanied by significant computational costs. Therefore, relevant algorithms are needed to perform inversion and uncertainty quantification in high-dimensional spaces. Here, we demonstrate the use of Hamiltonian Monte Carlo for inverse diffusion modeling in a petrological framework. Hamiltonian Monte Carlo is a gradient-based method that efficiently explores high-dimensional parameter spaces. We implemented it using the Turing.jl package in Julia which makes use of Automatic Differentiation to efficiently explore the parameter space. Our analysis focused on the calculation of the initial cooling rate, equilibration temperature and effective mineral grain size. We fit compositional garnet data and Ar-muscovite geochronological data from the Pindos metamorphic sole (Greece) by using two different forward diffusion models. Our joint inversion shows an initial equilibration temperature of 632.3 ± 9.3 °C and a cooling rate of 202.6 ± 72.0 °C/Myr. These values reproduce not only thermobarometric and geochronological observations but also fit the garnet composition profiles and the 40Ar/39Ar age of muscovite. We finally aim to highlight the potential of Hamiltonian Monte Carlo as a robust method to perform high-dimensional inversion and constrain complex petrological processes.