The long-term evolution and present-day structure of the Earth's mantle

The thermo-chemical structure of the Earth interior is controlled by the long-term evolution of mantle convection and surface tectonics. However, our knowledge of present-day mantle structure is restricted by the limited resolution of seismic-tomography models. Depending on the history of the coupled plate-mantle system as well as mantle-rock properties, various styles of present-day mantle structure are predicted by geodynamic models. Thus, better constraints on mantle structure can help to constrain Earth formation and differentiation, the onset age of plate tectonics, and mantle-atmosphere co-evolution with implications for planetary habitability.

In this project, we will explore numerical models of long-term mantle convection and mixing to compare their predictions with seismic data. Comparison of geodynamic model predictions with seismic tomography is routinely performed but subject to intrinsic limitations; so, we will focus on direct comparison with data. This approach will involve forward-modeling of synthetic waves through geodynamically predicted mantle structures, and comparison of these waveforms with real seismograms using artificial intelligence (AI). Such an integrated geodynamic-seismological effort will help to distinguish various styles of mantle evolution and present-day structure.