Mantle Convection Models Research Articles

Crustal shear wave velocity and radial anisotropy beneath the Rio Grande rift from ambient noise tomography

AbstractShear wave velocity and radial anisotropy beneath New Mexico are obtained from ambient seismic noise tomography using data from the Transportable Array. Besides the distinct seismic structure imaged across the Rio Grande rift from the Colorado Plateau to the Great Plains, both velocity and anisotropy models also reveal significant variations along the rift. The rift at Albuquerque is characterized by remarkably low velocity in the shallow crust, high velocity and strong positive anisotropy in the middle and lower crust, and low velocity in the upper mantle. These observations can be interpreted as magma accumulation in the shallow crust and significant mafic underplating in the lower crust with abundant melt supply from the hot mantle. We propose that the Albuquerque region has recently been experiencing the most vigorous extensional deformation in the rift. Positive anisotropy with Vsh > Vsv appears in the central and southern rifts with a stronger anisotropy beneath younger volcanoes, reflecting layering of magma intrusion due to past and recent rifting activities. The low velocities in the uppermost mantle are observed under high‐elevation places, the Jemez Lineament, northern rift, and east rift boundary, implying that the buoyancy of hot mantle largely compensates the local high topography. Low mantle velocities appear at the boundary of the southern rift, corresponding to the large lithosphere thickness change, instead of the rift center, consistent with the prediction from the small‐scale, edge‐driven mantle convection model. We conclude that the edge‐driven upper mantle convection is probably the dominant mechanism for the recent and current rifting and uplift in the Rio Grande rift.

A regime diagram of mobile lid convection with plate-like behavior

Terrestrial planetary bodies that undergo solid-state convection can exhibit a variety of tectonic styles, from stagnant lid one-plate planets to those with a mobilized lithosphere. For modeled planetary convection the de facto mode of recycling of lithosphere into the planetary interior is typically achieved through 2-sided and symmetric downwelling flows. However, lithosphere recycling on Earth occurs in a distinctly 1-sided mode known as subduction. Using numerical models of mantle convection in which the viscosity of planetary mantle material is strongly temperature-dependent, yet maintains a finite material strength as dictated by its yield stress, we investigate the continuum of mobile lid convection with plate-like behavior. The models span a parameter space of Rayleigh number and plate strength, and explore convective systems with low yield stresses resulting in weak subduction hinges that bend easily and highly deformable subducted slabs. Three distinct modes are found to occur in convective systems with weak plates: the stagnant lid mode, 2-sided downwelling mode, and a mode that alternates between 1-sided subduction and 2-sided downwellings. We classify the style of convective downwelling for a range of models and show that mode selection strongly depends on the combination of surface mobility and strength of the downgoing plate. Using these measurements, we have developed a regime diagram that can predict whether a particular system will be in one of those three modes.

The history of supercontinents and oceans from the standpoint of thermochemical mantle convection

A thermochemical model of mantle convection reproducing both the supercontinental and oceanic history of the Earth is formulated. The model takes into account (i) the stratification of the mantle into the upper and lower layers, (ii) the differentiation, which lead to the formation of light material in D” layer and (iii) the eclogitization, which results in the production of heavy material in the subduction zones. The numerical experiments are carried out in the Boussinesq approximation with isoviscous rheology.It is found that the high degree of spherical symmetry of the young planet promotes the cubic shape of the mantle convection. The hot initial state and rich chemical potential are reflected in active differentiation of the mantle material and formation of the bulk of the continental crust at the early stage of evolution.It is established that after losing a significant part of thermochemical potential, the planet passes into megacyclic regime of evolution, where long periods of predominantly two-layer convection are interrupted by mantle overturns. The overturns render extensive fast mass exchange between the two mantles: part of the cold substance from the upper mantle subsides and hot material lifts up from the lower mantle to fill its place. The numerical experiments demonstrate a remarkable phenomenon of self-organization of a single global convective sink during the mantle overturn. Such a collective sinking of the upper mantle substance is energetically more favorable. The planetary-scale flow, which converges to the global downwelling flow (sink), assembles the supercontinent.Our modeling of the geodynamic evolution of the Earth reproduces five mega-cycles. It is found that the configurations of the overturns tend to a dipole form and the position of the global sinks is being stabilized. Thus, the spherical mantle becomes slightly different: the hemisphere where the global sink is pulsating hosts the continents and the alternating Atlantic-type oceans, while the other hemisphere marked by the superplumes activity is dominated by the formation of the giant floor of the Pacific Ocean. As a result, the Earth acquires an asymmetrical appearance with the continental and oceanic hemispheres.

A Cenozoic uplift history of Mexico and its surroundings from longitudinal river profiles

AbstractGeodynamic models of mantle convection predict that Mexico and western North America share a history of dynamic support. We calculate admittance between gravity and topography, which indicates that the elastic thickness of the plate in Mexico is 11 km and in western North America it is 12 km. Admittance at wavelengths > 500 km in these regions suggests that topography is partly supported by subcrustal processes. These results corroborate estimates of residual topography from isostatic calculations and suggest that the amount of North American topography supported by the mantle may exceed 1 km. The Cenozoic history of magmatism, sedimentary flux, thermochronometric denudation estimates, and uplifted marine terraces imply that North American lithosphere was uplifted and eroded during the last 30 Ma. We jointly invert 533 Mexican and North American longitudinal river profiles to reconstruct a continent‐scale rock uplift rate history. Uplift rate is permitted to vary in space and time. Erosional parameters are calibrated using incision rate data in southwest Mexico and the Colorado Plateau. Calculated rock uplift rates were 0.15–0.2 mm/yr between 25 and10 Ma. Central Mexico experienced the highest uplift rates. Central and southern Mexico continued to uplift at 0.1 mm/yr until recent times. This uplift history is corroborated by independent constraints. We predict clastic flux to the Gulf of Mexico and compare it to independent estimates. We tentatively suggest that the loop between uplift, erosion, and deposition can be closed here. Mexico's staged uplift history suggests that its dynamic support has changed during the last 30 Ma.

Assimilating lithosphere and slab history in 4-D Earth models

We develop methods to incorporate paleogeographical constraints into numerical models of mantle convection. Through the solution of the convection equations, the models honor geophysical and geological data near the surface while predicting mantle flow and structure at depth and associated surface deformation. The methods consist of four constraints determined a priori from a plate history model: (1) plate velocities, (2) thermal structure of the lithosphere, (3) thermal structure of slabs in the upper mantle, and (4) velocity of slabs in the upper mantle. These constraints are implemented as temporally- and spatially-dependent conditions that are blended with the solution of the convection equations at each time step. We construct Earth-like regional models with oceanic and continental lithosphere, trench migration, oblique subduction, and asymmetric subduction to test the robustness of the methods by computing the temperature, velocity, and buoyancy flux of the lithosphere and slab. Full sphere convection models demonstrate how the methods can determine the flow associated with specific tectonic environments (e.g., back-arc basins, intraoceanic subduction zones) to address geological questions and compare with independent data, both at present-day and in the geological past (e.g., seismology, residual topography, stratigraphy). Using global models with paleogeographical constraints we demonstrate (1) subduction initiation at the Izu–Bonin–Mariana convergent margin and flat slab subduction beneath North America, (2) enhanced correlation of model slabs and fast anomalies in seismic tomography beneath North and South America, and (3) comparable amplitude of dynamic and residual topography in addition to improved spatial correlation of dynamic and residual topography lows.

A positive feedback between magmatism and mantle upwelling in terrestrial planets: Implications for the Moon

AbstractA series of two‐dimensional numerical models of magmatism and mantle convection in small planetary bodies are presented to discuss how the small size of the Moon exerts control over its mantle evolution. Mantle convection is modeled by a solid‐state convection of internally heated materials with temperature‐dependent viscosity. Magmatism is modeled as upward permeable flow of basaltic magma generated by decompression melting of upwelling mantle materials. Migration of the generated magma causes compaction/expansion of the coexisting matrix. The volume change of matrix and the buoyancy of magma induced by magmatism enhance the upwelling flow and hence the magmatism itself, when the Rayleigh number exceeds a threshold that is around the critical Rayleigh number for the onset of thermal convection. This positive feedback makes magmatism vigorous and episodic and causes efficient cooling and stirring of the mantle in large planets like Venus and the Earth. In small planetary bodies where the Rayleigh number is lower than the threshold, however, magmatism occurs more continuously with a characteristic time of several hundred million years, cooling and convective stirring of the mantle caused by magmatism are less important, and a compositionally stratified structure develops in the mantle. These features of magmatism and mantle structure fit in with the observations on the Moon. The positive feedback is a key to understanding why the evolution of the lunar mantle is so different from that of lager planets like the Earth and Venus.

Tomography-based mantle flow beneath Mongolia-Baikal area

Recent progress in seismic tomography of Asia allows us to explore and understand more clearly the mantle flow below the Mongolia-Baikal area. We present a tomography-based model of mantle convection that provides a good match to the residual topography. The model provides predictions on the present-day mantle flow and flow-induced asthenospheric deformation which give us new insights on the mantle dynamics in the Mongolia-Baikal area. The predicted mantle flow takes on a very similar pattern at the depths shallower or deeper than 400km and almost opposite flow directions between the upper (shallower than 400km) and lower (deeper than 400km) parts. The flow pattern could be divided into the ‘simple’ eastern region and the ‘complex’ western region in the Mongolia. The upwelling originating from about 350km depth beneath Baikal rift zone is an important possible drive force to the rifting. The seismic anisotropy cannot be simply related with asthenospheric flow and flow-induced deformation in the entire Mongolia-Baikal area, but they could be considered as an important contributor to the seismic anisotropy in the eastern region of Mongolia and around and in Sayan-Baikal orogenic belt.

On the sensitivity of 3-D thermal convection codes to numerical discretization: a model intercomparison

Abstract. Fully 3-D numerical simulations of thermal convection in a spherical shell have become a standard for studying the dynamics of pattern formation and its stability under perturbations to various parameter values. The question arises as to how the discretization of the governing equations affects the outcome and thus any physical interpretation. This work demonstrates the impact of numerical discretization on the observed patterns, the value at which symmetry is broken, and how stability and stationary behavior is dependent upon it. Motivated by numerical simulations of convection in the Earth's mantle, we consider isoviscous Rayleigh–Bénard convection at infinite Prandtl number, where the aspect ratio between the inner and outer shell is 0.55. We show that the subtleties involved in developing mantle convection models are considerably more delicate than has been previously appreciated, due to the rich dynamical behavior of the system. Two codes with different numerical discretization schemes – an established, community-developed, and benchmarked finite-element code (CitcomS) and a novel spectral method that combines Chebyshev polynomials with radial basis functions (RBFs) – are compared. A full numerical study is investigated for the following three cases. The first case is based on the cubic (or octahedral) initial condition (spherical harmonics of degree ℓ = 4). How this pattern varies to perturbations in the initial condition and Rayleigh number is studied. The second case investigates the stability of the dodecahedral (or icosahedral) initial condition (spherical harmonics of degree ℓ = 6). Although both methods first converge to the same pattern, this structure is ultimately unstable and systematically degenerates to cubic or tetrahedral symmetries, depending on the code used. Lastly, a new steady-state pattern is presented as a combination of third- and fourth-order spherical harmonics leading to a five-cell or hexahedral pattern and stable up to 70 times the critical Rayleigh number. This pattern can provide the basis for a new accuracy benchmark for 3-D spherical mantle convection codes.

The adjoint method in geodynamics: derivation from a general operator formulation and application to the initial condition problem in a high resolution mantle circulation model

The adjoint method is a computationally efficient way to compute the gradient of a physical observable or an associated objective function relative to its parameters. In geodynamics the observable can be thought of as a representation of the present day heterogeneity structure in the Earth’s mantle, inferred in some form through seismic imaging, while a crucial derivative of interest is that relative to an earlier convective system state. Since mantle convection is governed by coupled, non-linear conservation equations for mass, momentum and energy, computation of the derivative consists of iterative solutions to the forward and the adjoint problem, rendering the approach superior to finite difference approximations, which become impractical at the resolution of modern geodynamic models. Moreover, similarities in the forward and adjoint equations allow one to apply existing numerical codes that solve the forward problem to the adjoint equations with little adaptation. Bunge et al. (Geophys J Int 152(2):280–301 (2003)), have derived the adjoint equations for mantle convection using the concept of Lagrangian multipliers. Here we introduce a more general approach using an operator formulation in Hilbert spaces, in order to connect to recent work in seismology (Fichtner et al. Phys Earth Planet Int 157(1–2):86–104 (2006a)), where the approach was used to derive the adjoint equations for the scalar wave equation. We demonstrate the practicality of the method for use in a high resolution mantle circulation model with more than 80 million finite elements by restoring a representation of present day mantle heterogeneity derived from the global seismic shear wave study of Grand et al. (GSA Today 7(4):1–7 1997) back in time for the past 40 million years. An important result is our finding of a strong global minimum for the unknown initial condition, regardless of the assumed first guess for the initial heterogeneity structure, which we attribute to the uniqueness theorem by Serrin. Paleo mantle convection modelling will improve our ability to test assumptions about the internal structure and dynamics of the Earth’s mantle against the geologic record.

Spontaneous development of arcuate single‐sided subduction in global 3‐D mantle convection models with a free surface

AbstractWe present temporally evolving 3‐D global mantle convection models with single‐sided subduction and a free surface in both 3‐D Cartesian and fully spherical geometry. Special focus is given to the spontaneous development of three‐dimensional structures at the surface and in the upper mantle. We find that an arcuate shape is the natural form for trenches and slabs. Cartesian models are used first to study the dynamic evolution of subduction zones, spreading ridges, and interconnected transform features. These experiments highlight the strong variation of spontaneously developing, arcuate slab curvature and subduction polarity along the trench strike. The spontaneous development of spreading ridges leads to lateral offsets between separated segments that are characterized by normal transform motion. Spherical models then allow insights into the evolution of plate tectonics on a sphere. Investigated are the spontaneous evolution of slab geometry, trench motion, and subduction‐induced mantle flow. Two new dynamical features are discovered: “back‐slab spiral flow” and “slab tunneling.” © 2014. American Geophysical Union. All Rights Reserved.

A comparison of mantle convection models featuring plates

AbstractOceanic plates are an integral part of the Earth's mantle and thus play an important role in its dynamics and evolution. To allow plate behavior to arise naturally in numerical mantle convection models, self‐consistent plate generation methods apply a fully rheological approach (featuring a temperature‐, pressure‐ and stress‐dependent viscosity). However, due to the extreme local viscosity changes that the self‐generation of model plates entails, their computational requirements are demanding. Alternative plate modeling methods specify the existence of plates explicitly but can also obtain dynamically determined velocities (e.g., by employing a force‐balance method). Here we present modifications to a force‐balance model by utilizing a rheology‐dependent viscosity profile. Accordingly, plate viscosity and plate thickness are no longer prescribed by the modeler but now follow as a dynamic consequence of the temperature and stress dependence of the viscosity and the model's evolution. We describe the new method and present benchmark results for a rheologically self‐consistent mantle convection model and the modified force‐balance plate model. Our results show that both plate modeling methods lead to the same system behavior for a wide range of system parameters making the new method a powerful tool to also achieve plate‐like surface motion naturally.

Atmosphere/mantle coupling and feedbacks on Venus

AbstractWe investigate the coupled evolution of the atmosphere and mantle on Venus. Here we focus on mechanisms that deplete or replenish the atmosphere: atmospheric escape to space and volcanic degassing of the mantle. These processes are linked to obtain a coupled model of mantle convection and atmospheric evolution, including feedback of the atmosphere on the mantle via the surface temperature. During early atmospheric evolution, hydrodynamic escape is dominant, while for later evolution we focus on nonthermal escape, as observed by the Analyzer of Space Plasma and Energetic Atoms instrument on the Venus Express Mission. The atmosphere is replenished by volcanic degassing from the mantle, using mantle convection simulations based on those of Armann and Tackley [2012], and include episodic lithospheric overturn. The evolving surface temperature is calculated from the amount of CO2 and water in the atmosphere using a gray radiative‐convective atmosphere model. This surface temperature in turn acts as a boundary condition for the mantle convection model. We obtain a Venus‐like behavior (episodic lid) for the solid planet and an atmospheric evolution leading to the present conditions. CO2 pressure is unlikely to vary much over the history of the planet, with only a 0.25–20% postmagma‐ocean buildup. In contrast, atmospheric water vapor pressure is strongly sensitive to volcanic activity, leading to variations in surface temperatures of up to 200 K, which have an effect on volcanic activity and mantle convection. Low surface temperatures trigger a mobile lid regime that stops once surface temperatures rise again, making way to stagnant lid convection that insulates the mantle.

Restoring past mantle convection structure through fluid dynamic inverse theory: regularisation through surface velocity boundary conditions

Mantle convection is governed by coupled, non-linear conservation equations for mass, momentum and energy. In the forward problem the equations relate an initial to a final state of the convective system, while one can pose an inverse problem to restore earlier structure from a final state. Although the ability to restore earlier mantle structure is essential in geodynamics, allowing one to test uncertain parameters of mantle convection models against constraints derived from the geologic record, the convergence properties of the inverse problem are not well understood. Here we show that knowledge of the surface velocity field over the restoration period is crucial for the convergence rate of the restoration problem. With simple mantle convection models, and assuming for reference a given initial and final state, we explore the restoration problem in cases where knowledge of the surface velocity field over the restoration period is available relative to those where it is not. We find convergence of the inverse problem for time periods of $$\sim $$ 1/3 of a transit time and corresponding to about 50 million years in the Earth’s mantle, if the surface velocity field is known, while it diverges otherwise. For the Earth’s mantle the history of the surface velocity field is known for time periods of $$\sim $$ 1/2 transit times from reconstructions of past plate motion. Our results suggest that this constraint is of key importance in any attempt to restore past mantle structure.

The gravity field and crustal thickness of Venus

The gravity and topography of Venus obtained from observations of the Magellan mission, as well as the gravity and topography from our numerical mantle convection model, are discussed in this paper. We used the hypothesis that the geoid of degrees 2–40 is produced by sublithospheric mantle density anomalies that are associated with dynamical process within the mantle. We obtained the model dynamical admittance (the geoid topography ratio based on a convection model) by a numerical simulation of the Venusian mantle convection, and used it to correct the dynamical effect in the calculation of crustal thickness. After deducting the dynamical effect, the thickness of the Venusian crust is presented. The results show that the gravity and topography are strongly correlated with the Venusian mantle convection and the Venusian crust has a significant influence on the topography. The Venusian crustal thickness varies from 28 to 70 km. Ishtar Terra, and Ovda Regio and Thetis Regio in western Aphrodite Terra have the highest crustal thickness (larger than 50 km). The high topography of these areas is thought to be supported by crustal compensation and our results are consistent with the hypothesis that these areas are remnants of ancient continents. The crustal thickness in the Beta, Themis, Dione, Eistla, Bell, and Lada regiones is thin and shows less correlation with the topography, especially in the Atla and Imdr regiones in the eastern part of Aphrodite Terra. This is consistent with the hypothesis that these highlands are mainly supported by mantle plumes. Compared with the crustal thickness calculated with the dynamical effect, our results are more consistent with the crust evolution and internal dynamical process of Venus.

Effects of various lithospheric yield stresses and different mantle-heating modes on the breakup of the Pangea supercontinent

Numerical simulations of three-dimensional spherical mantle convection were performed to investigate the effects of various lithospheric yield stresses and two different mantle-heating modes (i.e., mixed heating from the bottom and interior and purely internal heating) on the breakup of the Pangea supercontinent and the subsequent continental drift for the past 200 Myr. Results show that the continental breakup and subsequent continental drift are accomplished in mantle convection models with two different heating modes. This implies that active upwelling plumes from the core-mantle boundary are not necessarily required for the breakup of Pangea. In addition, I found that the continental breakup is only realized when choosing a moderate value of the yield stress (~120 MPa). The bound on the yield stress derived in the present study will potentially enable the self-consistent reconstruction of continental breakup and drift as well as the accompanying pattern on mantle convection since 200 Ma.

A wet, heterogeneous lunar interior: Lower mantle and core dynamo evolution

While recent analyses of lunar samples indicate the Moon had a core dynamo from at least 4.2–3.56 Ga, mantle convection models of the Moon yield inadequate heat flux at the core-mantle boundary to sustain thermal core convection for such a long time. Past investigations of lunar dynamos have focused on a generally homogeneous, relatively dry Moon, while an initial compositionally stratified mantle is the expected consequence of a postaccretionary lunar magma ocean. Furthermore, recent re-examination of Apollo samples and geophysical data suggests that the Moon contains at least some regions with high water content. Using a finite element model, we investigate the possible consequences of a heterogeneously wet, compositionally stratified interior for the evolution of the Moon. We find that a postoverturn model of mantle cumulates could result in a core heat flux sufficiently high to sustain a dynamo through 2.5 Ga and a maximum surface, dipolar magnetic field strength of less than 1 μT for a 350-km core and near ∼2 μT for a 450-km core. We find that if water was transported or retained preferentially in the deep interior, it would have played a significant role in transporting heat out of the deep interior and reducing the lower mantle temperature. Thus, water, if enriched in the lower mantle, could have influenced core dynamo timing by over 1.0 Gyr and enhanced the vigor of a lunar core dynamo. Our results demonstrate the plausibility of a convective lunar core dynamo even beyond the period currently indicated by the Apollo samples.

A new conceptual model for whole mantle convection and the origin of hotspot plumes

A new conceptual model of mantle convection is constructed for consideration of the origin of hotspot plumes, using recent evidence from seismology, high-pressure experiments, geodynamic modeling, geoid inversion studies, and post-glacial rebound analyses. This conceptual model delivers several key points. Firstly, some of the small-scale mantle upwellings observed as hotspots on the Earth's surface originate at the base of the mantle transition zone (MTZ), in which the Archean granitic continental material crust (TTG; tonalite-trondhjemite-granodiorite) with abundant radiogenic elements is accumulated. Secondly, the TTG crust and the subducted oceanic crust that have accumulated at the base of MTZ could act as thermal or mechanical insulators, leading to the formation of a hot and less viscous layer just beneath the MTZ; which may enhance the instability of plume generation at the base of the MTZ. Thirdly, the origin of some hotspot plumes is isolated from the large low shear-wave velocity provinces (LLSVPs) under Africa and the South Pacific. I consider that the conceptual model explains why almost all the hotspots around Africa are located above the margins of the African LLSVP. Because a planetary-scale trench system surrounding a “Pangean cell” has been spatially stable throughout the Phanerozoic, a large amount of the oceanic crustal layer is likely to be trapped in the MTZ under the Pangean cell. Therefore, under Africa, almost all of the hotspot plumes originate from the base of the MTZ, where a large amount of TTG and/or oceanic crusts has accumulated. This conceptual model may explain the fact that almost all the hotspots around Africa are located on margins above the African LLSVP. It is also considered that some of the hotspot plumes under the South Pacific thread through the TTG/oceanic crusts accumulated around the bottom of the MTZ, and some have their roots in the South Pacific LLSVP while others originate from the MTZ. The numerical simulations of mantle convection also speculate that the Earth's mantle convection is not thermally double-layered at the ringwoodite to perovskite+magnesiowüstite (Rw→Pv+Mw) phase boundary, because of its gentle negative Clapeyron slope. This is in contrast with some traditional images of mantle convection that have independent convection cells between the upper and lower mantle. These numerical studies speculate that the generation of stagnant slab at the base of the MTZ (as seismically observed globally) may not be due to the negative Clapeyron slope, and may instead be related to a viscosity increase (i.e., a viscosity jump) at the Rw→Pv+Mw phase boundary, or to a chemically stratified boundary between the upper and the lower mantle, as suggested by a recent high-pressure experiment.

Statistical cyclicity of the supercontinent cycle

Supercontinents like Pangea impose a first-order control on Earth's evolution as they modulate global heat loss, sea level, climate, and biodiversity. In a traditional view, supercontinents form and break up in a regular, perhaps periodic, manner in a cycle lasting several 100 Myr as reflected in the assembly times of Earth's major continental aggregations: Columbia, Rodinia, and Pangea. However, modern views of the supercontinent cycle propose a more irregular evolution on the basis of an improved understanding of the Precambrian geologic record. Here we use fully dynamic spherical mantle convection models featuring plate-like behavior and continental drift to investigate supercontinent formation and breakup. We further dismiss the concept of regularity but suggest a statistical cyclicity in which the supercontinent cycle may have a characteristic period imposed by mantle and lithosphere properties, but this is hidden in immense fluctuations between different cycles that arise from the chaotic nature of mantle flow.

Mantle evolution in Venus due to magmatism and phase transitions: From punctuated layered convection to whole‐mantle convection

AbstractA series of numerical models of magmatism and mantle convection with a stagnant lithosphere are developed to understand the mantle evolution in Venus. Magmatism is modeled as a permeable flow of basaltic magma generated by decompression melting, and the solid‐state convection of mantle materials with temperature‐dependent Newtonian rheology is affected by the garnet‐perovskite transition and the postspinel transition. In our preferred models, the mantle evolves in two stages: The earlier stage is characterized by layered mantle convection punctuated by repeated bursts of hot material from the deep mantle to the surface. Mantle bursts induce vigorous magmatism and also cause the basaltic crust, enriched in heat‐producing elements (HPEs), to recycle into the mantle. A part of the recycled basaltic crusts accumulates along the postspinel boundary to form a barrier, and this basalt barrier causes mantle convection to become layered. At a later stage, when the HPEs have already decayed, in contrast, the basalt barrier disappears and whole‐mantle convection occurs more steadily. Mild magmatism is induced by small‐scale partial melting at the base of the crust and hot plumes from the deep mantle. The internal heating by the HPEs that recycled into the mantle in the earlier stage allows the magmatism of the later stage to continue throughout the calculated history of mantle evolution. The two stages arise when the barrier effect of the postspinel transition is weak and the lithosphere is mechanically strong enough. The two‐stage evolution model meshes with the observed history of magmatism and the lithosphere on Venus.

Impact heating and coupled core cooling and mantle dynamics on Mars

AbstractSeveral giant impact basins of mid‐Noachian age have been identified on Mars, and the global magnetic field appears to have vanished at about the same time. The impacts that formed these basins delivered a large amount of heat to the planetary interior, modified the pattern of mantle convection, and suppressed core cooling, potentially contributing to the cessation of dynamo activity. Here we investigate the thermal evolution of Mars in response to the largest basin‐forming impacts, using a new method of coupling models of mantle convection with parameterized core cooling. We find that heating by a large impact generates a strong hemispheric upwelling in the mantle, which quickly spreads into a warm layer beneath the stagnant lid. The impact heating of the core leads to spherically symmetric stratification of the core; the outermost layers are strongly heated. This acts as a thermal “blanket” that prevents cooling of the interior and shuts down core convection. While the hottest part of the thermal blanket in the outermost core disappears relatively quickly, dynamo activity does not restart for ~100 Myr, and the core does not return to a fully convective state for ~1 Gyr following the impact.