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.

We use cosmological simulations to study the effects of self-interacting dark matter (SIDM) on the density profiles and substructure counts of dark matter halos from the scales of spiral galaxies to galaxy clusters, focusing explicitly on models with cross sections over dark matter particle mass σ/m = 1 and 0.1 cm 2 /g. Our simulations rely on a new SIDM N-body algorithm that is derived self-consistently from the Boltzmann equation and that reproduces analytic expectations in controlled numerical experiments. We find that well-resolved SIDM halos have constant-density cores, with significantly lower central densities than their CDM counterparts. In contrast, the subhalo content of SIDM halos is only modestly reduced compared to CDM, with the suppression greatest for large hosts and small halo-centric distances. Moreover, the large-scale clustering and halo circular velocity functions in SIDM are effectively identical to CDM, meaning that all of the large-scale successes of CDM are equally well matched by SIDM. From our largest cross section runs we are able to extract scaling relations for core sizes and central densities over a range of halo sizes and find a strong correlation between the core radius of an SIDM halo and the NFW scale radius of its CDM counterpart. We construct a simple analytic model, based on CDM scaling relations, that captures all aspects of the scaling relations for SIDM halos. Our results show that halo core densities in σ/m = 1 cm 2 /g models are too low to match observations of galaxy clusters, low surface brightness spirals (LSBs), and dwarf spheroidal galaxies. However, SIDM with σ/m ≃ 0.1 cm 2 /g appears capable of reproducing reported core sizes and central densities of dwarfs, LSBs, and galaxy clusters without the need for velocity dependence. Higher resolution simulations over a wider range of masses will be required to confirm this expectation. We discuss constraints arising from the Bullet cluster observations, measurements of dark matter density on small-scales and subhalo survival requirements, and show that SIDM models with σ/m ≃ 0.1 cm 2 /g ≃ 0.2 barn/GeV are consistent with all observational constraints.

We introduce a set of high-resolution dissipationless simulations that model the Local Group (LG) in a cosmological context: Exploring the Local Volume in Simulations (ELVIS). The suite contains 48 Galaxy-size haloes, each within high-resolution volumes that span 2 − 5 Mpc in size, and each resolving thousands of systems with masses below the atomic cooling limit. Half of the ELVIS galaxy haloes are in paired configurations similar to the Milky Way (MW) and M31; the other half are isolated, mass-matched analogs. We find no difference in the abundance or kinematics of substructure within the virial radii of isolated versus paired hosts. On Mpc scales, however, LG-like pairs average almost twice as many companions and the velocity field is kinematically hotter and more complex. We present a refined abundance matching relation between stellar mass and halo mass that reproduces the observed satellite stellar mass functions of the MW and M31 down to the regime where incompleteness is an issue, M ∼ 5 × 10 5 M . Within a larger region spanning approximately 3 Mpc, the same relation predicts that there should be ∼ 1000 galaxies with M > 10 3 M awaiting discovery. We show that up to 50% of haloes within 1 Mpc of the MW or M31 could be systems that have previously been within the virial radius of either giant. By associating never-accreted haloes with gas-rich dwarfs, we show that there are plausibly 50 undiscovered dwarf galaxies with HI masses > 10 5 M within the Local Volume. The radial velocity distribution of these predicted gas-rich dwarfs can be used to inform follow-up searches based on ultra-compact high-velocity clouds found in the ALFALFA survey.

Among the most important goals in cosmology is detecting and quantifying small (M halo 10 6−9 M ) dark matter (DM) subhalos. Current probes around the Milky Way (MW) are most sensitive to such substructure within ∼ 20 kpc of the halo center, where the galaxy contributes significantly to the potential. We explore the effects of baryons on subhalo populations in ΛCDM using cosmological zoom-in baryonic simulations of MW-mass halos from the Latte simulation suite, part of the Feedback In Realistic Environments (FIRE) project. Specifically, we compare simulations of the same two halos run using (1) DM-only (DMO), (2) full baryonic physics, and (3) DM with an embedded disk potential grown to match the FIRE simulation. Relative to baryonic simulations, DMO simulations contain ∼ 2× as many subhalos within 100 kpc of the halo center; this excess is 5× within 25 kpc. At z = 0, the baryonic simulations are completely devoid of subhalos down to 3 × 10 6 M within 15 kpc of the MW-mass galaxy, and fewer than 20 surviving subhalos have orbital pericenters < 20 kpc. Despite the complexities of baryonic physics, the simple addition of an embedded central disk potential to DMO simulations reproduces this subhalo depletion, including trends with radius, remarkably well. Thus, the additional tidal field from the central galaxy is the primary cause of subhalo depletion. Subhalos on radial orbits that pass close to the central galaxy are preferentially destroyed, causing the surviving population to have tangentially biased orbits compared to DMO predictions. Our method of embedding a potential in DMO simulations provides a fast and accurate alternative to full baryonic simulations, thus enabling suites of cosmological simulations that can provide accurate and statistical predictions of substructure populations.

We show that the canonical oscillation-based (non-resonant) production of sterile neutrino dark matter is inconsistent at > 99% confidence with observations of galaxies in the Local Group. We set lower limits on the non-resonant sterile neutrino mass of 2.5 keV (equivalent to 0.7 keV thermal mass) using phase-space densities derived for dwarf satellite galaxies of the Milky Way, as well as limits of 8.8 keV (equivalent to 1.8 keV thermal mass) based on subhalo counts of N -body simulations of M 31 analogues. Combined with improved upper mass limits derived from significantly deeper X-ray data of M 31 with full consideration for background variations, we show that there remains little room for non-resonant production if sterile neutrinos are to explain 100% of the dark matter abundance. Resonant and non-oscillation sterile neutrino production remain viable mechanisms for generating sufficient dark matter sterile neutrinos.