A-priori Validation of Subgrid-scale Models for Astrophysical Turbulence
Abstract
We perform a-priori validation tests of subgrid-scale (SGS) models for the turbulent transport of momentum, energy and passive scalars. To this end, we conduct two sets of high-resolution hydrodynamical simulations with a Lagrangian code: an isothermal turbulent box with rms Mach number of 0.3, 2 and 8, and the classical wind tunnel where a cold cloud traveling through a hot medium gradually dissolves due to fluid instabilities. Two SGS models are examined: the eddy diffusivity (ED) model wildly adopted in astrophysical simulations and the "gradient model" due to Clark et al. (1979). We find that both models predict the magnitude of the SGS terms equally well (correlation coefficient > 0.8). However, the gradient model provides excellent predictions on the orientation and shape of the SGS terms while the ED model predicts poorly on both, indicating that isotropic diffusion is a poor approximation of the instantaneous turbulent transport. The best-fit coefficient of the gradient model is in the range of [0.16, 0.21] for the momentum transport, and the turbulent Schmidt number and Prandtl number are both close to unity, in the range of [0.92, 1.15].
Turn this paper into a lesson
ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.