ABSTRACT: In many rotating fluids, the lowest order force balance is between gravity, pressure, and rotational acceleration (‘GPR’ balance). Terrestrial GPR balance takes the form of geostrophy and hydrostasy, which together yield the terrestrial thermal wind equation. By contrast, stellar GPR balance is an oblateness equation, which determines the departures of the thermal variables from spherical symmetry; its curl yields the ‘stellar thermal wind equation’. In this sense, the stellar thermal wind should be viewed not as a consequence of geostrophy, but of baroclinicity in the oblateness. Here, we treat the full stellar oblateness, including the thermal wind, using pressure coordinates. We derive the generalized stellar thermal wind equation and identify the parameter regime for which it holds. In the case of the Sun, not considering the full oblateness has resulted in conflicting calculations of the theoretical aspherical temperature anomaly. We provide new calculation here and find that the baroclinic anomaly is ∼3–60 times smaller than the barotropic anomaly. Thus, the anomaly from the thermal wind may not be measurable helioseismically, but if measurement were possible, this would potentially yield a new way to bracket the depth of the solar tachocline.

Abstract: Meridional circulation regulates the Sun’s interior dynamics and magnetism. While it is well accepted that meridional flows are poleward at the Sun’s surface, helioseismic observations have yet to provide a definitive answer for the depth at which those flows return to the equator, or the number of circulation cells in depth. Here, we explore the observability of multiple circulation cells stacked in radius. Specifically, we examine the seismic signature of several meridional flow profiles by convolving time–distance averaging kernels with mean flows obtained from a suite of 3D hydrodynamic simulations. At mid and high latitudes, we find that weak flow structures in the deep convection zone can be obscured by signals from the much stronger surface flows. This contamination of 1–2 m s−1 is caused by extended side lobes in the averaging kernels, which produce a spurious equatorward signal with flow speeds that are 1 order of magnitude stronger than the original flow speeds in the simulations. At low latitudes, the flows in the deep layers of the simulations are stronger (>2 m s−1) and multiple cells across the convection zone can produce a sufficiently strong signal to survive the convolution process. Now that meridional flows can be measured over two decades of data, the uncertainties arising from convective noise have fallen to a level where they are comparable in magnitude to the systematic biases caused by nonlocal features in the averaging kernels. Hence, these systematic errors are beginning to influence current helioseismic deductions and need broader consideration.