Exploring the interplay between continent formation, surface erosion, and the evolution of plate tectonics on Earth

In this work, we test the hypothesis of surface-erosion controlled plate tectonics preceded by plume-induced retreating subduction tectonic regime on Earth proposed by Sobolev and Brown (2019) using 2D global compressible convection models. To simulate the effect of increased sediment supply as a result of surface erosion after the emergence of continents in the late Archean and after the Neoproterozoic "snowball Earth" glaciation, we decrease the effective frictional strength of the oceanic lithosphere in models spanning the age of the Earth. These StagYY models self-consistently generate oceanic and continental crust while considering both plutonic and volcanic magmatism (Jain et al., 2019). Pressure-, temperature-, and composition-dependent water solubility maps calculated with Perplex (Connolly, 2009) are also utilised, which control the ingassing and outgassing of water between the mantle and surface (Jain et al., 2022). The core cools with time and different initial mantle potential temperature values are tested within the range of 1750-1900 K (Herzberg et al., 2010; Aulbach and Arndt, 2019).

Models that consider a more realistic upper mantle rheology (diffusion creep and dislocation creep proxy) show higher recycling of denser basaltic-eclogitic (oceanic) crust, efficient cooling of the planet, and higher mobilities (ratio of surface to mantle rms velocities) (Tackley (2000); Lourenço et al. (2020)). These models exhibit intermittent episodes of long-lasting mobile-lid regime and short-lived plutonic-squishy-lid regime in the Hadean and the early Archean accompanied by extensive subduction leading to rapid production and recycling of the continental crust. Models that consider adaptive frictional strength (to mimic sedimentation post glaciation and continental emergence) predict the transition to continuous plate tectonics in the late Archean, reproduce features of supercontinent cycles, and appear to be consistent with cooling history of the Earth inferred from petrological observations (Herzberg et al., 2010). 

The thermo-compositional evolution can vary between models due to the inherent randomness arising from the initial thermal perturbations and the initial positions of the tracers/particles. Accordingly, we intend to run multiple instances of every model considered in our parameter space to present statistically robust results. We also aim to test more realistic models where the lithospheric frictional strength adapts with the surface topography.