The Effect of Plumes on the Dynamics of Supercontinents in a Self-Consistent Plate Tectonics Setting

Strong mantle plumes arising from the deep mantle can impose stresses on the continents, thereby facilitating continental rifting and disrupting the supercontinent cycle (Storey, Nature 1995; Santosh et al., Gondwana Research 2009). In recent years, several studies have characterized the relation between the location of the plumes and the continents, but with contradicting observations. While Heron and Lowman (GRL, 2010; Tectonophysics, 2011) propose regions where downwelling has ceased (irrespective of overlying plate) as the preferred location for plumes, O’Neill et al. (Gondwana Research, 2009) show an anti-correlation between the average positions of subducting slabs at continental margins, and mantle plumes at continental/oceanic interiors.

Extent of continental motion depends on the heat budget of the mantle (CMB heat flux, radiogenic heating, mantle cooling). CMB heat flux is not well defined; however, the recent determination of core's high thermal conductivity requires a CMB heat flow of at least 12 TW (de Koker et al., PNAS 2012; Pozzo et al., Nature 2012; Gomi et al., PEPI 2013), much higher than early estimates of 3-4 TW (Lay et al., Nature 2008). Thus, it is necessary to characterize the effect of increased CMB heat flux on mantle dynamics.

In almost all mantle convection simulations, the top boundary is treated as a free-slip surface whereas Earth's surface is a deformable free surface. Unlike free-slip, a free surface boundary condition allows for the development of topography and leads to realistic single-sided (asymmetric) subduction (Crameri et al., GJI 2012; Crameri et al., GRL 2012).

Using StagYY code (Tackley, PEPI 2008), we test (i) the impact of increased basal heating on mantle dynamics with continents and self-consistent plate tectonics, including whether plumes prefer to develop under continents; (ii) the influence of a free surface on continents using the ‘sticky air’ approach, in which a low density and a small viscosity fluid layer is added to the top of the model. The existing model from Rolf et al. (EPSL 2012) is developed further but with weaker continents.