Modeling Melting Ages and Pseudo-Isochron-Ages in Whole Mantle Convection

Abstract:

Many outstanding problems in Earth science relate to the geodynamical explanation of geochemical observations. Extensive geochemical databases of surface observations exist, but satisfying explanations of underlying processes are lacking. One way to address these problems is through numerical modeling. Although recently much progress on thermal-chemical convection modeling has been made, analytical solutions remain sparse. This makes testing and benchmarking the ever more complex models challenging.

We implemented tracking of chemical information in the mantle convection code TERRA (3D spherical), using particles. One value on each particle represents bulk composition; it can be interpreted as the proportion of basalt component. In our model, chemical separation happens at self-consistent, evolving melting zones. Molten material is transported to the surface, thereby increasing the basalt proportion close to the surface. On melting, we also record the time such that we can track the melting age of each particle. The system is set up to track the abundance of the 7 isotopes of the U-Th-Pb system.

We use a simplified setup of incompressible convection, without complexities such as phase changes and elastic/plastic deformation. We will show: 1: The evolution of bulk composition, showing the build up of oceanic crust (via melting induced chemical separation in bulk composition); i.e. a basalt-rich layer at the surface overlying a thin layer of depleted material (Harzburgite), and the transportation of these heterogeneities through the mantle. 2: A fit between the evolution of the pseudo isochron based on relative Pb-isotope abundances along the surface of the model; and the evolution of the isochron age based on the distribution of melting ages throughout the mantle. A good fit to analytical solutions (Rudge, 2006) is obtained which shows that our implementation is operating in a self-consistent way, plus it enables us to compare surface observables to internal dynamics.