Meridional Circulation From Differential Rotation in an Adiabatically Stratified Solar/Stellar Convection Zone

TL;DR

This paper demonstrates that differential rotation, even when constant on cylinders, can drive meridional circulation in stellar convection zones due to baroclinic effects, challenging conventional assumptions and highlighting the role of additional forces.

Contribution

It shows that differential rotation can induce meridional flow through baroclinic effects, supported by analytical and numerical evidence, with implications for stellar convection models.

Findings

Differential rotation on cylinders can drive meridional circulation.

Meridional flow is sensitive to departures from cylindrical rotation.

Additional forces like buoyancy are needed to match observed solar circulation.

Abstract

Meridional circulation in stellar convection zones is not generally well observed, but may be critical for MHD dynamos. Coriolis forces from differential rotation (DR) play a large role in determining what the meridional circulation is. Here we consider whether a stellar DR that is constant on cylinders concentric with the rotation axis can drive a meridional circulation.Conventional wisdom says that it can not. Using two related forms of governing equations that respectively estimate the longitudinal components of the curl of meridional mass flux and the vorticity, we show that such DR will drive a meridional flow. This is because to satisfy anelastic mass conservation, non-spherically symmetric pressure contours must be present for all DRs, not just ones that depart from constancy on cylinders concentric with the rotation axis. Therefore the fluid is always baroclinic if DR is present, because, in anelastic systems, the perturbation pressure must satisfy a Poisson type equation, as well as an equation of state and a thermodynamic equation. We support our qualitative reasoning with numerical examples, and show that meridional circulation is sensitive to the magnitude and form of departures from rotation constant on cylinders. The effect should be present in 3D global anelastic convection simulations, particularly those for which the DR driven by global convection is nearly cylindrical in profile. For solar-like DR, Coriolis forces generally drive a two-celled circulation in each hemisphere, with a second, reversed flow at high latitudes. For solar like turbulent viscosities, the meridional circulation produced by Coriolis forces is much larger than observed on the Sun. Therefore there must be at least one additional force, probably a buoyancy force, which opposes the meridional flow to bring its amplitude down to observed values.