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

Abstract

We present STARFORGE (STAR FORmation in Gaseous Environments): a new numerical framework for 3D radiation MHD simulations of star formation that simultaneously follow the formation, accretion, evolution, and dynamics of individual stars in massive giant molecular clouds (GMCs) while accounting for stellar feedback, including jets, radiative heating and momentum, stellar winds, and supernovae. We use the GIZMO code with the MFM mesh-free Lagrangian MHD method, augmented with new algorithms for gravity, timestepping, sink particle formation and accretion, stellar dynamics, and feedback coupling. We survey a wide range of numerical parameters/prescriptions for sink formation and accretion and find very small variations in star formation history and the IMF (except for intentionally-unphysical variations). Modules for mass-injecting feedback (winds, SNe, and jets) inject new gas elements on-the-fly, eliminating the lack of resolution in diffuse feedback cavities otherwise inherent in Lagrangian methods. The treatment of radiation uses GIZMO’s radiative transfer solver to track 5 frequency bands (IR, optical, NUV, FUV, ionizing), coupling direct stellar emission and dust emission with gas heating and radiation pressure terms. We demonstrate accurate solutions for SNe, winds, and radiation in problems with known similarity solutions, and show that our jet module is robust to resolution and numerical details, and agrees well with previous AMR simulations. STARFORGE can scale up to massive (>105M⊙) GMCs on current supercomputers while predicting the stellar (≳ 0.1M⊙) range of the IMF, permitting simulations of both high- and low-mass cluster formation in a wide range of conditions.

Cosmological Simulations of Quasar Fueling to Subparsec Scales Using Lagrangian Hyper-refinement

“…The same feedback implementation has been successfully used in cosmological zoom-in simulations with mass resolution ranging from m b ∼ 20 M e in dwarfs (Wheeler et al 2019) to m b ∼ 3 × 10 4 M e in massive galaxies (Anglés-Alcázar et al 2017c). At the highest resolution reached here (m b ∼ 15 M e ), individual star particles are still massive enough to accommodate a full SN II mass ejection, but the same algorithm performs well down to significantly lower particle masses, conserving mass in a time-average sense (Grudić et al 2021). Stellar winds follow a similar implementation but with continuous rather than impulsive injection of mass, momentum, energy, and metals into surrounding gas particles.…”

The dynamics and outcome of star formation with jets, radiation, winds, and supernovae in concert

“…We perform a 3D radiation MHD simulation of star cluster formation in a GMC with initial mass 𝑀 0 = 2 × 10 4 𝑀 and radius 𝑅 0 = 10pc using the STARFORGE numerical framework implemented in the GIZMO code (Grudić et al 2021a;Hopkins 2015), with all implemented feedback physics enabled: protostellar jets, radiation, winds, and core-collapse supernovae. The numerical implementation and tests of the STARFORGE modules are detailed fully in Paper I, so here we only summarize them briefly.…”

“…the same mesh-free volume discretization as used for MHD solver ( §2.1). To make the calculation tractable, we assume a reduced speed of light c = 30km s −1 , sufficient to capture the dynamics of D-type HII region expansion (Geen et al 2015;Grudić et al 2021a).…”

“…We found these models inevitably predicted excessively-high star formation efficiency and an extreme excess of massive stars in Milky Way-like conditions, implying that additional mechanisms are important for star formation, and in particular some form of feedback must moderate stellar accretion. In Grudić et al (2021a) (hereafter Paper I) we introduced the more-advanced STARFORGE 1 framework for the GIZMO code, combining modules for gravity, N-body dynamics, MHD, radiative transfer, cooling and chemical physics, (proto-)stellar evolution, and feedback in the form of accretion-and fusion-powered radiation from stars and protostars, stellar winds, protostellar jets, and core-collapse supernovae. And in Guszejnov et al (2021) (hereafter Paper II) we used STAR-FORGE to re-run our GMC models with the addition of realistic ISM cooling/heating physics and protostellar jet feedback, finding that jet feedback in particular is crucial for moderating the growth of individual stars and recovering a realistic IMF, in agreement with other IMF studies with jet feedback (Hansen et al 2012;Krumholz et al 2012;Myers et al 2013;Federrath et al 2014;Mathew & Federrath 2021).…”

A model for the formation of stellar associations and clusters from giant molecular clouds

“…However, the advent of massively scalable, Lagrangian codes with fast, accurate MHD methods and well-developed feedback coupling techniques will soon make this possible: in Guszejnov et al (2020a) we simulated GMCs up to 2 × 10 6 M with individually resolved collapsing stellar cores without feedback, at a mass resolution 200 times finer than the present work. In future work we will present results with a full accounting of both protostellar and stellar feedback (Grudić et al 2021;.…”