The Feedback In Realistic Environments (FIRE) project explores feedback in cosmological galaxy formation simulations. Previous FIRE simulations used an identical source code ("FIRE-1") for consistency. Motivated by the development of more accurate numerics -including hydrodynamic solvers, gravitational softening, and supernova coupling algorithms -and exploration of new physics (e.g. magnetic fields), we introduce "FIRE-2", an updated numerical implementation of FIRE physics for the GIZMO code. We run a suite of simulations and compare against FIRE-1: overall, FIRE-2 improvements do not qualitatively change galaxy-scale properties. We pursue an extensive study of numerics versus physics. Details of the star-formation algorithm, cooling physics, and chemistry have weak effects, provided that we include metal-line cooling and star formation occurs at higher-than-mean densities. We present new resolution criteria for high-resolution galaxy simulations. Most galaxy-scale properties are robust to numerics we test, provided: (1) Toomre masses are resolved; (2) feedback coupling ensures conservation, and (3) individual supernovae are time-resolved. Stellar masses and profiles are most robust to resolution, followed by metal abundances and morphologies, followed by properties of winds and circum-galactic media (CGM). Central (∼kpc) mass concentrations in massive (> L * ) galaxies are sensitive to numerics (via trapping/recycling of winds in hot halos). Multiple feedback mechanisms play key roles: supernovae regulate stellar masses/winds; stellar mass-loss fuels late star formation; radiative feedback suppresses accretion onto dwarfs and instantaneous star formation in disks. We provide all initial conditions and numerical algorithms used.

ABSTRACT We present the implementation and the first results of cosmic ray (CR) feedback in the Feedback In Realistic Environments (FIRE) simulations. We investigate CR feedback in non-cosmological simulations of dwarf, sub-L⋆ starburst, and L⋆ galaxies with different propagation models, including advection, isotropic, and anisotropic diffusion, and streaming along field lines with different transport coefficients. We simulate CR diffusion and streaming simultaneously in galaxies with high resolution, using a two-moment method. We forward-model and compare to observations of γ-ray emission from nearby and starburst galaxies. We reproduce the γ-ray observations of dwarf and L⋆ galaxies with constant isotropic diffusion coefficient $\kappa \sim 3\times 10^{29}\, {\rm cm^{2}\, s^{-1}}$. Advection-only and streaming-only models produce order of magnitude too large γ-ray luminosities in dwarf and L⋆ galaxies. We show that in models that match the γ-ray observations, most CRs escape low-gas-density galaxies (e.g. dwarfs) before significant collisional losses, while starburst galaxies are CR proton calorimeters. While adiabatic losses can be significant, they occur only after CRs escape galaxies, so they are only of secondary importance for γ-ray emissivities. Models where CRs are ‘trapped’ in the star-forming disc have lower star formation efficiency, but these models are ruled out by γ-ray observations. For models with constant κ that match the γ-ray observations, CRs form extended haloes with scale heights of several kpc to several tens of kpc.

We present and study a large suite of high-resolution cosmological zoom-in simulations, using the FIRE-2 treatment of mechanical and radiative feedback from massive stars, together with explicit treatment of magnetic fields, anisotropic conduction and viscosity (accounting for saturation and limitation by plasma instabilities at high β), and cosmic rays (CRs) injected in supernovae shocks (including anisotropic diffusion, streaming, adiabatic, hadronic and Coulomb losses). We survey systems from ultrafaint dwarf ($M_{\ast }\sim 10^{4}\, \mathrm{M}_{\odot }$, $M_{\rm halo}\sim 10^{9}\, \mathrm{M}_{\odot }$) through Milky Way/Local Group (MW/LG) masses, systematically vary uncertain CR parameters (e.g. the diffusion coefficient κ and streaming velocity), and study a broad ensemble of galaxy properties [masses, star formation (SF) histories, mass profiles, phase structure, morphologies, etc.]. We confirm previous conclusions that magnetic fields, conduction, and viscosity on resolved ($\gtrsim 1\,$ pc) scales have only small effects on bulk galaxy properties. CRs have relatively weak effects on all galaxy properties studied in dwarfs ($M_{\ast } \ll 10^{10}\, \mathrm{M}_{\odot }$, $M_{\rm halo} \lesssim 10^{11}\, \mathrm{M}_{\odot }$), or at high redshifts (z ≳ 1–2), for any physically reasonable parameters. However, at higher masses ($M_{\rm halo} \gtrsim 10^{11}\, \mathrm{M}_{\odot }$) and z ≲ 1–2, CRs can suppress SF and stellar masses by factors ∼2–4, given reasonable injection efficiencies and relatively high effective diffusion coefficients $\kappa \gtrsim 3\times 10^{29}\, {\rm cm^{2}\, s^{-1}}$. At lower κ, CRs take too long to escape dense star-forming gas and lose their energy to collisional hadronic losses, producing negligible effects on galaxies and violating empirical constraints from spallation and γ-ray emission. At much higher κ CRs escape too efficiently to have appreciable effects even in the CGM. But around $\kappa \sim 3\times 10^{29}\, {\rm cm^{2}\, s^{-1}}$, CRs escape the galaxy and build up a CR-pressure-dominated halo which maintains approximate virial equilibrium and supports relatively dense, cool (T ≪ 106 K) gas that would otherwise rain on to the galaxy. CR ‘heating’ (from collisional and streaming losses) is never dominant.

Using high-resolution simulations with explicit treatment of stellar feedback physics based on the FIRE (Feedback in Realistic Environments) project, we study how galaxy formation and the interstellar medium (ISM) are affected by magnetic fields, anisotropic Spitzer-Braginskii conduction and viscosity, and sub-grid metal diffusion from unresolved turbulence. We consider controlled simulations of isolated (non-cosmological) galaxies but also a limited set of cosmological "zoom-in" simulations. Although simulations have shown significant effects from these physics with weak or absent stellar feedback, the effects are much weaker than those of stellar feedback when the latter is modeled explicitly. The additional physics have no systematic effect on galactic star formation rates (SFRs) . In contrast, removing stellar feedback leads to SFRs being over-predicted by factors of ∼ 10 − 100. Without feedback, neither galactic winds nor volume filling hot-phase gas exist, and discs tend to runaway collapse to ultra-thin scale-heights with unphysically dense clumps congregating at the galactic center. With stellar feedback, a multi-phase, turbulent medium with galactic fountains and winds is established. At currently achievable resolutions and for the investigated halo mass range 10 10 − 10 13 M , the additional physics investigated here (MHD, conduction, viscosity, metal diffusion) have only weak (∼ 10%-level) effects on regulating SFR and altering the balance of phases, outflows, or the energy in ISM turbulence, consistent with simple equipartition arguments. We conclude that galactic star formation and the ISM are primarily governed by a combination of turbulence, gravitational instabilities, and feedback. We add the caveat that AGN feedback is not included in the present work.

ABSTRACT We study the spatially resolved (sub-kpc) gas velocity dispersion (σ)–star formation rate (SFR) relation in the FIRE-2 (Feedback in Realistic Environments) cosmological simulations. We specifically focus on Milky Way-mass disc galaxies at late times (z ≈ 0). In agreement with observations, we find a relatively flat relationship, with σ ≈ 15–30 km s−1 in neutral gas across 3 dex in SFRs. We show that higher dense gas fractions (ratios of dense gas to neutral gas) and SFRs are correlated at constant σ. Similarly, lower gas fractions (ratios of gas to stellar mass) are correlated with higher σ at constant SFR. The limits of the σ–ΣSFR relation correspond to the onset of strong outflows. We see evidence of ‘on-off’ cycles of star formation in the simulations, corresponding to feedback injection time-scales of 10–100 Myr, where SFRs oscillate about equilibrium SFR predictions. Finally, SFRs and velocity dispersions in the simulations agree well with feedback-regulated and marginally stable gas disc (Toomre’s Q = 1) model predictions, and the simulation data effectively rule out models assuming that gas turns into stars at (low) constant efficiency (i.e. 1 per cent per free-fall time). And although the simulation data do not entirely exclude gas accretion/gravitationally powered turbulence as a driver of σ, it appears to be subdominant to stellar feedback in the simulated galaxy discs at z ≈ 0.