Generation of Archean Granitoids and the Onset of Plate Tectonics.

点击翻译按钮获取中文摘要

The formation of the continental crust had a first order impact on the habitability of the Earth, being directly linked to the emergence of life and later allowing life to migrate onto land. Continents have a high elevation compared to oceans, because continental crustal rocks are granitic in composition and have a relatively low density compared to mantle rocks. It is generally accepted that these rocks must have formed by melting the mantle in a multi-step process that requires basalt and hydrous phases to be present. On the present-day Earth this occurs in subduction zones. Yet, in the Archean Earth, mantle temperatures were higher, more melt was produced in the mantle, and oceanic crust was significantly thicker. Models suggest that if mantle potential temperatures were 150-250K higher than present-day values, subduction could not have operated. Instead, parts of the crust may have dripped down into the mantle, a process known as sagduction. The question of whether plate tectonics operated in the Archean therefore depends on two key points: 1) that temperatures in the Archean upper mantle were so high that oceanic plates were unsubductable; 2) that Archean granitoids were produced by a process independent of subduction. How granites formed is incompletely understood, particularly for a warmer Earth. In this proposal, we will use petrological constraints to build a model of thick oceanic crust; one that is strongly differentiated with ultramafic cumulates in the lower half and hydrated basalt restricted to upper layers. We will also re-evaluate petrological constraints and cooling models that define temperatures in Archean ambient mantle. Then we will use 2D and 3D numerical models to simulate dehydration, melting and deformation within this crust and compare the results with available petrological/ geochemical constraints. By including recently developed thermodynamic melting models and the chemical evolution of melts in the thermomechanical numerical models, we can use model predictions to test whether sagduction or subduction is a more efficient process to create granites. In addition, we will derive parameterizations that describe the effect of both subduction and sagduction on the cooling of the Earth and include these in parameterized cooling models. This will give new insights in how such processes affect the thermal evolution of the Earth, how this affected crustal growth and on the timing of the onset of plate tectonics on our planet.