Steady plumes produced by downwellings in Earth-like vigor spherical whole mantle convection models

[1] If mantle thermal upwellings ( plumes) are the cause of volcanic hot spots,'' then observations suggest that plumes are relatively fixed with nonuniform distribution and limited lifetimes. To date, fixity of upwellings has only been shown in models of convection at either low-vigor or with layering, though studies where the lower mantle has high viscosity do frequently show upwellings with much lower drift velocities than the surface velocity. Since more vigorous convection traditionally shows more time dependence, fixity of upwellings has not been expected for nonlayered convection at Earth-like vigor; rather, we might expect slow but increasing drift velocities. I have undertaken numerical models of whole mantle convection in three-dimensional spherical geometry at approaching Earth-like vigor. Surprisingly, these simulations show prominent steady, virtually fixed, plumes arising from self-organization controlled by the smaller but numerous cold downwellings. If downwelling ( subduction) dominates Earth's interior dynamics, then this work suggests that plume fixity need not require layering. The regular spacing and permanence of model upwellings contrast with observations of hot spots though. In these models, which do not simulate plates, the fixity and regular spacing result from the freedom of downwellings to occur everywhere except where plumes reach the surface. I suggest that on Earth this feedback is weakened by the presence of plates but that whole mantle thermal convection can still produce relatively fixed plumes, though we might expect them to be weaker, more mobile, and transient.