Driving Solar Giant Cells Through the Self-Organization of Near-Surface Plumes

Friday, March 24, 2017

Global 3D simulations of solar giant-cell convection have provided significant insight into the processes which yield the Sun’s observed differential rotation and cyclic dynamo action.

Self-organization of downflow plumes image
Self-organization of downflow plumes. (a) Snapshot of radial velocity on the outer boundary of a simulation over a 30 by 30 degree patch centered on the equator with north pointing up and east pointing to the right. This patch contains 25 imposed plumes. Plumes are numbered from 1 to 25 based on their central longitude from west to east. The downflow portion of each plume is seeded with 50 streamlines randomly initiated at between 0.973 and 0.983R. Streamlines are then integrated downward into the simulation domain. (b) 3D rendering of the streamlines initiated in the boundary plumes colored by the assigned plume number. Perspective is looking west. The upper and lower boundaries of the simulation domain have been added as black surfaces for visual clarity. (c) Same as (b) but now looking north. Almost all of the 25 plumes coalesce into a large rotationally-aligned downflow lane which is part of a strong giant cell in the bottom half of the domain.

However, as we move to higher resolution simulations a variety of codes have encountered what has been termed the convective conundrum. As these simulations increase in resolution and hence the level of turbulence achieved, they tend to produce weak or even anti-solar differential rotation patterns associated with a weak rotational influence (high Rossby number) due to large convective velocities. One potential culprit for this convective conundrum is the upper boundary condition applied in most simulations which is generally impenetrable. Here we present an alternative stochastic plume boundary condition which imposes small-scale convective plumes designed to mimic near-surface convective downflows, thus allowing convection to carry the majority of the outward solar energy flux up to and through our simulated upper boundary. The use of a plume boundary condition leads to significant changes in the convective driving realized in the simulated domain and thus to the convective energy transport, the dominant scale of the convective enthalpy flux, and the relative strength of the strongest downflows, the downflow network, and the convective upflows. These changes are present even far from the upper boundary layer. Additionally, we demonstrate that in spite of significant changes, giant cell morphology in the convective patterns is still achieved with self-organization of the imposed boundary plumes into downflow lanes, cellular patterns, and even rotationally-aligned banana cells in equatorial regions. This plume boundary presents an alternative pathway for 3D global convection simulations where driving is non-local and may provide a new approach towards addressing the so-called convection conundrum.

Submitted to Astrophysical Journal.