Global numerical simulations of solar convection and differential rotation

Guerrero Gustavo, gag@stanford.edu, Stanford University, United States
Smolarkiewicz Piotr, smolar@ucar.edu, UCAR
Kosovichev Alexander, sasha@sun.stanford.edu, Stanford University
Mansour Nagi, nagi.n.mansour@nasa.gov, NASA

Abstract
Large-scale plasma flows are of great importance for astrophysical dynamos. Large-scale shear is one of the source terms in the induction equation of the mean magnetic field. Meridional flows and/or winds transport magnetic flux and could determine the migration pattern of an oscillatory dynamo. In addition, both shear and advective flows could play an important role in the evolution of magnetic helicity fluxes, a key ingredient for stellar dynamos. In the case of the Sun, global helioseismology has been able to infer in great detail the subsurface profile of the solar rotation. The meridional flow is observed at the surface and in subsurface layers, but its profile in the deep convection zone remains unknown. On the other hand, numerical simulations of solar convection have difficulties in reproducing the observed differential rotation profile. In this work we present new results of global numerical simulation of solar rotation performed with the EULAG (Eulerian or Lagrangian frameworks) code in an anelastic approximation. The numerical method (MPDATA) introduces a minimal amount of numerical viscosity such that the numerical stability of the model is guaranteed, and the Reynolds number is maximized for any given resolution. We investigate the relationship between the structure of large-scale convection and the differential rotation patterns for the simulations with various grid resolutions. To better understand the mechanism determining the rotation pattern, we compute the turbulent fluxes of heat and angular momentum in models with different rotation rates and resolutions. We present the results of this parametric analysis and discuss the physics of the differential rotation based on our simulations.