Dynamic localized turbulent diffusion and its impact on the galactic ecosystem

Abstract

Modelling the turbulent diffusion of thermal energy, momentum, and metals is required in all galaxy evolution simulations due to the ubiquity of turbulence in galactic environments. The most commonly employed diffusion model, the Smagorinsky model, is known to be over-diffusive due to its strong dependence on the fluid velocity shear. We present a method for dynamically calculating a more accurate, locally appropriate, turbulent diffusivity: the dynamic localised Smagorinsky model. We investigate a set of standard astrophysically-relevant hydrodynamical tests, and demonstrate that the dynamic model curbs over-diffusion in non-turbulent shear flows and improves the density contrast in our driven turbulence experiments. In galactic discs, we find that the dynamic model maintains the stability of the disc by preventing excessive angular momentum transport, and increases the metal-mixing timescale in the interstellar medium. In both our isolated Milky Way-like galaxies and cosmological simulations, we find that the interstellar and circumgalactic media are particularly sensitive to the treatment of turbulent diffusion. We also examined the global gas enrichment fractions in our cosmological simulations, to gauge the potential effect on the formation sites and population statistics of Population III stars and supermassive black holes, since they are theorised to be sensitive to the metallicity of the gas out of which they form. The dynamic model is, however, not for galaxy evolution studies only. It can be applied to all astrophysical hydrodynamics simulations, including those modelling stellar interiors, planetary formation, and star formation.

STARFORGE: Towards a comprehensive numerical model of star cluster formation and feedback

“…We do note these tests have demonstrated that our default implementation can simultaneously accurately evolve phenomena including gas in regular or warped Keplerian discs, strong interacting shocks, current sheets and flux tubes, supersonic and subsonic turbulence, fluid mixing instabilities (Kelvin-Helmholz, Rayleigh Taylor, etc. ), multifluid dustgas dynamics, collisional+collisionless gravitational dynamics, and reproduces the correct linear growth rates of the magnetorotational instability (MRI) and non-ideal Hall MRI and anisotropic MHD instabilities (magnetothermal, heat-flux-bouyancy) (Hopkins 2015;Hopkins & Raives 2016;Zhu & Li 2016;Lupi, Volonteri & Silk 2017;Deng et al 2019a, b;Moseley et al 2019;Rennehan et al 2019;Hu & Chiang 2020;Panuelos, Wadsley & Kevlahan 2020). Tests of idealized problems involving self-gravitating MHD including the Evrard (1988) problem (spherical collapse of a self-gravitating polytrope), the MHD Zel'dovich (1970) pancake (self-gravitating collapse of an initially linear density perturbation along one dimension in a 3D Hubble flow) demonstrate that the MFM/MFV methods in GIZMO (as well as related moving-mesh methods) converge much more rapidly than popular AMR or SPH methods applied to the same problem (Hopkins 2015;Hopkins & Raives 2016;Hubber, Rosotti & Booth 2018).…”

3D gas-phase elemental abundances across the formation histories of Milky Way-mass galaxies in the FIRE simulations: initial conditions for chemical tagging

“…where Zi is the mass fraction of a metal in gas element i, and D is the diffusion coefficient. While there is some uncertainty in the exact value to choose for this coefficient, our fiducial value is physically motivated based on tests of the metal diffusion implementation in FIRE-2 on idealized, converged turbulent box simulations by Colbrook et al (2017) and other more extensive studies by Rennehan et al (2019).…”

Rapid early coeval star formation and assembly of the most-massive galaxies in the Universe

“…We can model a small comoving volume of the universe and resolve the galaxies in this volume (e.g. Schaye et al 2015;Pillepich et al 2018;Davé et al 2019) but such volumes generally do not contain the rare massive clusters we would expect to host an object such as SPT2349-56. Or, we can sacrifice resolution in favour of large volumes but then galaxy formation must be introduced in an ad hoc fashion (Ruszkowski & Springel 2009), which can introduce biases.…”