|
Paul Tackley, ETH Zurich (Switzerland)
Takashi Nakagawa, Kyushu University (Japan)
Frederic Deschamps, ETH Zurich (Switzerland)
James Connolly, ETH Zurich (Switzerland)
|
|
Mineralogy and petrology play an important role in mantle dynamics, because high pressure and temperature experiments and calculations of the properties of mantle minerals show that many different mineral phases exist as a function of pressure, temperature and composition, and that these have a first-order influence on properties such as density and elastic moduli (hence seismic velocity).
Numerical models of mantle convection have, however, typically used a simple approximation to treat these complex variations in material properties. Here, in order to get closer to a realistic mineralogy, we calculate composition-dependent mineral assemblages and their physical properties using the code PERPLEX, which minimizes free energy for a given combination of oxides as a function of temperature and pressure, and use this in numerical models of thermo-chemical mantle convection in a three-dimensional spherical shell, to calculate three-dimensionally-varying physical properties. The numerical models treat the evolution of a planet over billions of years, including self-consistent plate tectonics arising from plastic yielding, melting-induced differentiation, and a parameterised model of core evolution based on heat extracted by mantle convection.
With time, an extremely heterogeneous mantle builds up, including an accumulation of subducted crust at the core-mantle boundary, with the amount depending on the density contrast. Models in which this segregated crust covers only part of the CMB are most consistent with seismic observations and with geodynamo evolution scenarios. Crustal enrichment is observed above the 660 km discontinuity and depletion below this discontinuity, because of the composition-dependent depth of the transition to lower mantle mineralogy (mainly perovskite).
In the rest of the mantle, heterogeneity is observed at all scales. The model results are analysed in terms of (1) the time evolution of upper mantle composition, which is generally depleted relative to the bulk composition, (2) the role of crustal recycling in the geochemical evolution of the mantle,(3) the balance between homogenisation of heterogeneities by convection and generation of larger heterogeneities by crustal settling or transition zone processes, (4) the sensitivity of these to uncertainties in mineral physics parameters.
|
