arxiv: 2604.14273 · v1 · submitted 2026-04-15 · 🌌 astro-ph.GA

Recognition: unknown

Cold vs. Hot Gas Accretion and Angular Momentum in FIRE Simulations: From Halo to Galaxy Scales

Imran Sultan , Claude-Andr\'e Faucher-Gigu\`ere , Jonathan Stern , Guochao Sun

Authors on Pith no claims yet

Pith reviewed 2026-05-10 12:10 UTC · model grok-4.3

classification 🌌 astro-ph.GA

keywords gas accretionhalo virializationcold inflowshot inflowsangular momentumstar formation timescalesgalaxy diskssurrounding gas

0 comments

The pith

Gas accretion switches from cold to hot as halos virialize, setting whether stars form in bursts or steadily with disk rotation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper analyzes gas inflows in cosmological simulations across halo masses near the 10^12 solar mass scale where galaxies shift from bursty to steady star formation and develop thin disks. It establishes that inflows remain almost entirely cold before the inner surrounding gas reaches virial equilibrium and become hot afterward, with the hot mode allowing cooling to occur at the same time as the gas settles into circular orbits. This switch matters because cold inflows in low-mass systems turn into stars rapidly without building rotation, while hot inflows in massive systems linger longer, cool, and produce stars with disk-like motions. A reader would care because the result connects the thermal state of halo gas directly to the observed change in how galaxies grow and form stars over time.

Core claim

The authors find that the temperature of gas accreting onto galaxies correlates directly with whether the inner gas reservoir around the galaxy has virialized. In pre-virialized halos inflows are almost entirely cold while hot inflows dominate once virialization sets in, and in the hot case cooling proceeds together with circularization at galaxy radii. Cold inflows carry higher specific angular momentum than hot gas or dark matter, yet in bursty low-mass galaxies this gas forms stars before circularizing, whereas in steady massive galaxies the hot inflows circularize, cool, and form stars with disk-like kinematics. Accreted gas forms stars after fewer than about five galaxy free-fall times,

What carries the argument

The correlation between the temperature of accreting gas and the virialization state of the inner gas reservoir around the galaxy, which controls whether inflows circularize before forming stars and how long the gas persists before turning into stars.

If this is right

Halos below the virialization threshold grow via rapid star formation from cold gas that never settles into stable rotation.
Halos above the threshold sustain steady star formation because hot gas has time to gain circular motion and cool at galaxy scales.
The higher angular momentum carried by cold inflows does not produce early disks in low-mass galaxies because star formation occurs too quickly.
Even at high redshift where cold streams can still exist, hot inflows can dominate circularization and disk formation in massive halos.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Surveys could check the prediction by comparing the orbital motions of the youngest stars in low-mass versus high-mass galaxies.
The residence-time difference implies that chemical enrichment and feedback efficiency should vary systematically with galaxy mass and star-formation steadiness.
If the virialization threshold shifts with redshift, models should expect a corresponding shift in the mass at which disks emerge.
Higher-resolution runs could test whether the reported free-fall time ratios remain stable when cooling and mixing are treated more finely.

Load-bearing premise

The simulations correctly capture cooling, heating, and angular-momentum changes in the gas around galaxies without being dominated by resolution limits or model choices.

What would settle it

Observations of real galaxies showing no systematic difference in the rotation of young stars or in the time between gas arrival and star formation between low-mass bursty systems and high-mass steady systems.

Figures

Figures reproduced from arXiv: 2604.14273 by Claude-Andr\'e Faucher-Gigu\`ere, Guochao Sun, Imran Sultan, Jonathan Stern.

**Figure 1.** Figure 1: Gas temperature maps for two representative FIRE halos in our analysis set: m12i (top) and m13A4 (bottom). Projections of log 𝑇 weighted by mass through a 30 comoving kpc slice perpendicular to the galactic plane are shown, giving an edge-on view of each galaxy. The galactic plane is defined as the plane orthogonal to the rotation axis of the galaxy, which we compute from the total angular momentum of all … view at source ↗

**Figure 2.** Figure 2: Star formation histories for low-redshift (left) and high-redshift (right) halos in our sample. The top panels show the instantaneous SFR normalized by the SFR averaged in a moving 300 Myr window. The bottom panels show the standard deviation of the SFR averaged in a moving 300 Myr window, 𝜎300 Myr (log SFR). The results are plotted as a function of cosmic time relative to 𝑡bursty, which is defined as the … view at source ↗

**Figure 3.** Figure 3: Time evolution of the ratio of the average azimuthal velocity 𝑣𝜙 of HI gas (calculated at 𝑟 < 0.05𝑅vir) to the dispersion of 𝑣𝜙 of HI gas. The upper panel shows the low-redshift MW-mass halos we analyze, while the lower panel shows the high-redshift more massive halos. The results are plotted as a function of cosmic time relative to 𝑡bursty, the time when the SFR becomes steady. At early times relative to … view at source ↗

**Figure 4.** Figure 4: The hot gas inflow rate fractions in the CGM as a function of redshift. For the seven halos in our analysis set, ratios of the hot (𝑇 > 105 K) CGM mass inflow rate with respect to the total CGM inflow rate, 𝑀¤ hot/𝑀¤ tot𝑣𝑟 <0 are shown in three radial shells (indicated by different line colors). The halos are sorted by ascending order in 𝑀vir (𝑧 = 1). The results are averaged over a moving 400 Myr window.… view at source ↗

**Figure 5.** Figure 5: The magnitude of the total specific angular momentum of hot and cold inflowing gas in the CGM as a function of redshift. For the seven halos in our analysis set, | ®𝑗tot | = | 𝐽® tot/𝑀tot | 𝑣𝑟 <0 is shown for hot (𝑇 ≥ 105 K; thick lines) and cold (𝑇 < 105 K; thin lines) CGM gas with 𝑣𝑟 < 0. The thick brown line shows the specific angular momentum of all dark matter particles in the halo. The results are av… view at source ↗

**Figure 6.** Figure 6: The coherence of the specific angular momentum of hot and cold inflowing gas in the CGM as a function of redshift. Similar to [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗

**Figure 7.** Figure 7: Tracks of azimuthal velocity of accreted gas particles, shown for particles that are initially in the hot phase and are accreted during two time bins subsequent to ICV. Tracks are shown with respect to 𝑡 105 K , which is the time of the first snapshot in the time bin for which 𝑇 < 105 K. The solid lines indicate median tracks, while the shaded bands represent the 16- to 84- percentiles. The dot-dashed line… view at source ↗

**Figure 8.** Figure 8: Tracks of the radius of accreted gas particles, shown for particles that are initially in the hot phase. Tracks are shown for hot gas accreted during two time bins subsequent to ICV, as in [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗

**Figure 9.** Figure 9: Tracks of azimuthal velocity of accreted gas particles, shown for particles that are initially in the cold phase and are accreted during an early time bin prior to ICV. Tracks are shown with respect to 𝑡SF, defined as the time of the first snapshot in which a gas particle forms a star. We select cold gas particles at 0.3𝑅vir < 𝑟 < 0.5𝑅vir at 4.5 Gyr (1.25 Gyr) before ICV for m12 (m13) halos, track them for… view at source ↗

**Figure 10.** Figure 10: Tracks of azimuthal velocity of accreted gas particles, shown for particles that are initially in the hot phase and are accreted during a late time bin following ICV. Tracks are shown with respect to 𝑡SF, defined as the time of the first snapshot in which a gas particle forms a star. We select hot gas particles at 0.3𝑅vir < 𝑟 < 0.5𝑅vir at 1 Gyr after ICV, track them for 2 Gyr (1 Gyr) for the m12 (m13) hal… view at source ↗

**Figure 11.** Figure 11: Comparison of the circularity of stars at birth to the circularity of their accreted gas progenitors, for stars that form in bursty galaxies with pre-ICV CGM. Only stars that formed in situ at early times (i.e., stars that formed at 1.5 < 𝑧 < 2.5 for the m12, and from 500 Myr before 𝑧ICV to 𝑧 = 5 for the m13 halos) are shown. 𝜖gas(0.1𝑅vir) is the circularity of a gas particle at the last time it crosses 0… view at source ↗

**Figure 12.** Figure 12: Comparison of the circularity of stars at birth to the circularity of their accreted gas progenitors, for stars that form in steady galaxies with virialized CGM. Similar to [PITH_FULL_IMAGE:figures/full_fig_p013_12.png] view at source ↗

**Figure 13.** Figure 13: Comparison of stellar circularity at birth with the temperature of their progenitor gas at the time of accretion onto 0.1𝑅vir. The left panel shows the two-dimensional distribution stacked across seven galaxies in both low-mass bursty phases ( [PITH_FULL_IMAGE:figures/full_fig_p014_13.png] view at source ↗

**Figure 14.** Figure 14: Distributions of the time gas accreted onto 𝑟 < 0.05𝑅vir spends in bursty galaxies before forming stars (𝑡SF − 𝑡enter) or being ejected from the ISM (𝑡exit − 𝑡enter). Gas accreted during a 1 Gyr time interval when galaxies are bursty (starting at 𝑧 = 2.5 for m12 halos and 𝑧 = 5 for m13 halos) is tracked to the end of the simulation. For the 𝑡exit − 𝑡enter distributions, we include only particles that are … view at source ↗

**Figure 15.** Figure 15: Distributions of the time gas accreted onto 𝑟 < 0.05𝑅vir spends in steady galaxies before forming stars (𝑡SF − 𝑡enter) or being ejected from the ISM (𝑡exit − 𝑡enter). Similar to [PITH_FULL_IMAGE:figures/full_fig_p015_15.png] view at source ↗

read the original abstract

We present a systematic study of gas accretion and angular momentum in the circumgalactic medium (CGM) using high-resolution FIRE cosmological simulations. Our analysis includes halos spanning the critical $\sim 10^{12}\ \mathrm{M}_{\odot}$ scale where several transitions have been identified, including inner CGM virialization, the transition from bursty to steady star formation, and the emergence of thin disks. We find that the temperature of inflowing gas is correlated with the virialization of the inner CGM. CGM inflows are almost entirely cold ($T < 10^5$ K) in pre-virialized halos, while hot inflows ($T > 10^5$ K) dominate in virialized halos. When hot inflows dominate, cooling generally occurs simultaneously with circularization at galaxy radii. The dominance of hot inflows onto massive galaxies persists even at high redshift, where cold streams may coexist. Consistent with previous studies, cold inflows have higher specific angular momentum than dark matter and hot gas. However, in bursty, low-mass galaxies, cold inflows do not circularize prior to star formation, while in steady, massive galaxies, hot inflows circularize, cool, and form stars with disk-like kinematics. We additionally find that in bursty galaxies, accreted gas typically forms stars after residing in the galaxy for less than $\sim 5$ galaxy free-fall times, while in steady galaxies, gas can persist for up to $\sim 25$ free-fall times before forming stars. This highlights a key difference between star formation in bursty galaxies fed by cold accretion and steady equilibrium disks fed by hot accretion.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

FIRE runs tie the cold-to-hot accretion switch at 10^12 Msun to longer gas residence times and disk formation in the steady regime.

read the letter

This paper uses FIRE simulations to track how gas accretion changes from cold to hot around the 10^12 solar mass halo scale, and links that shift to the move from bursty star formation to steady disks. What stands out is the direct measurement that cold inflows stay cold in pre-virialized halos while hot inflows take over once the inner CGM virializes. They also show that in the hot case, cooling happens at the same time as the gas circularizes at galaxy radii. The residence time result is useful: gas in bursty systems forms stars in under 5 free-fall times, but can last up to 25 in steady ones. Cold gas carries more angular momentum than hot or dark matter, as expected. The work is solid on the simulation side because it pulls numbers straight from the snapshots rather than fitting models. It extends previous FIRE papers by adding the angular momentum and timescale analysis across the transition mass. The main limitation is that these trends depend on the FIRE subgrid physics and whether the inner CGM is resolved well enough for cooling and angular momentum transport. The abstract gives no convergence checks, so if the resolution misses key mixing or cooling lengths, the temperature split and the residence time gap might not hold up exactly. Still, the overall picture matches what other groups have seen. This is worth reading for anyone modeling galaxy assembly or interpreting CGM observations. It gives testable predictions on kinematics and gas phases. I would send it to referees because the measurements are clear and the mass range is the right one for current debates.

Referee Report

2 major / 2 minor

Summary. The manuscript analyzes gas accretion and angular momentum in the CGM using high-resolution FIRE cosmological zoom-in simulations for halos spanning the ~10^12 M_sun scale. It reports that CGM inflows are almost entirely cold (T < 10^5 K) in pre-virialized halos but hot inflows (T > 10^5 K) dominate once the inner CGM virializes; when hot inflows dominate, cooling occurs simultaneously with circularization at galaxy radii. Cold inflows carry higher specific angular momentum than dark matter or hot gas. In bursty low-mass galaxies, accreted gas forms stars after residing less than ~5 galaxy free-fall times, while in steady massive galaxies gas persists up to ~25 free-fall times before forming stars with disk-like kinematics.

Significance. If the numerical results hold, the work provides a direct link between the cold-to-hot accretion transition, inner-CGM virialization, and the shift from bursty to steady star formation, with quantitative residence-time statistics in free-fall units. The systematic halo sample and tracking of angular momentum from halo to galaxy scales offer concrete predictions for CGM observations and constraints on subgrid models. The simulation-based approach supplies falsifiable, resolution-dependent measurements rather than fitted parameters.

major comments (2)

[Methods and §4] Methods and §4 (Results on temperature and timescales): No resolution convergence tests or subgrid variations are presented for the inner CGM (r ≲ 0.1 R_vir), where the cold/hot inflow fractions, simultaneous cooling-circularization, and <5 vs ~25 free-fall-time statistics are measured. These quantities are sensitive to metal-line cooling, thermal conduction, and torque resolution; without demonstrated convergence, the reported dichotomy and bursty-vs-steady contrast remain vulnerable to numerical artifacts.
[§3 and temperature analysis] §3 (virialization definition) and temperature analysis: The correlation between inflow temperature and inner-CGM virialization is central, yet the manuscript does not quantify how the adopted T = 10^5 K threshold or the precise virialization criterion (e.g., t_cool/t_ff or entropy) affects the reported dominance fractions; small shifts in these definitions could weaken or strengthen the claimed transition.

minor comments (2)

[Abstract and §2] Abstract and §2: The specific FIRE version (FIRE-2 or FIRE-3) and the mass resolution in the inner CGM should be stated explicitly so readers can assess the applicability of the results.
[Figures] Figure captions: Panels showing inflow properties should list the halo mass, redshift, and virialization state for each example to improve traceability of the trends.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive evaluation of our work's significance and for the detailed comments. We address each major comment below, agreeing on the need for additional tests and planning revisions accordingly.

read point-by-point responses

Referee: [Methods and §4] Methods and §4 (Results on temperature and timescales): No resolution convergence tests or subgrid variations are presented for the inner CGM (r ≲ 0.1 R_vir), where the cold/hot inflow fractions, simultaneous cooling-circularization, and <5 vs ~25 free-fall-time statistics are measured. These quantities are sensitive to metal-line cooling, thermal conduction, and torque resolution; without demonstrated convergence, the reported dichotomy and bursty-vs-steady contrast remain vulnerable to numerical artifacts.

Authors: We acknowledge the importance of demonstrating numerical convergence for these key quantities in the inner CGM. While the FIRE simulations have been validated for convergence in global galaxy properties across multiple resolutions in previous studies (e.g., Hopkins et al. 2018), we did not explicitly test the CGM inflow metrics in this work. In the revised manuscript, we will add an appendix with resolution convergence tests using available lower-resolution counterparts from the FIRE suite, focusing on the cold/hot inflow fractions and residence time distributions at r < 0.1 R_vir. Regarding subgrid variations, we note that FIRE uses a fixed subgrid model, but we will discuss the potential impact of metal-line cooling and conduction based on sensitivity tests from related FIRE papers. This will be a partial revision as full subgrid variations would require new simulations. revision: partial
Referee: [§3 and temperature analysis] §3 (virialization definition) and temperature analysis: The correlation between inflow temperature and inner-CGM virialization is central, yet the manuscript does not quantify how the adopted T = 10^5 K threshold or the precise virialization criterion (e.g., t_cool/t_ff or entropy) affects the reported dominance fractions; small shifts in these definitions could weaken or strengthen the claimed transition.

Authors: We agree that quantifying the sensitivity to these definitions would improve the robustness of our claims. The T = 10^5 K threshold is motivated by the sharp drop in the cooling function below this temperature, but we will revise §3 to include a brief sensitivity analysis. Specifically, we will present the hot inflow fraction as a function of threshold temperature (varying from 5×10^4 K to 2×10^5 K) and show that the transition at the virialization mass scale remains qualitatively unchanged. For the virialization criterion, we use the condition where the inner CGM has t_cool/t_ff > 1 (or similar entropy-based definition as in the manuscript), and we will add a note or small test showing the effect of alternative thresholds. These additions will be included in the revised version. revision: yes

Circularity Check

0 steps flagged

No significant circularity; central results are direct measurements from simulation snapshots.

full rationale

The paper presents findings from post-processing of FIRE cosmological simulation outputs, including temperature distributions of inflows, correlations with inner CGM virialization, angular momentum comparisons, and gas residence times before star formation. These are reported as empirical results extracted from snapshots across halo masses and redshifts, without any fitted parameters that are then relabeled as predictions, self-definitional equations, or load-bearing self-citations that reduce the claims to prior unverified work. The abstract and described analysis contain no equations or derivations that loop back to their own inputs by construction. Self-citations to prior FIRE work are present but serve only as context for the simulation suite; the new measurements stand independently. This matches the default expectation for simulation-based papers whose value lies in the numerical data rather than closed-form theory.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Claims rest on outputs of cosmological hydrodynamical simulations whose sub-grid star-formation and feedback prescriptions are taken as given from the FIRE project; no new physical axioms or entities are introduced.

axioms (2)

standard math Standard Lambda-CDM cosmology and initial conditions
Implicit in all cosmological simulations; invoked throughout the halo selection and evolution.
domain assumption FIRE sub-grid prescriptions for cooling, star formation, and stellar feedback accurately represent unresolved physics
Central to interpreting the temperature and angular-momentum results; stated as the simulation framework used.

pith-pipeline@v0.9.0 · 5622 in / 1400 out tokens · 39758 ms · 2026-05-10T12:10:38.119130+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

1 extracted references · 1 canonical work pages · 1 internal anchor

[1]

Ade P., et al., 2019, Journal of Cosmology and Astroparticle Physics, 2019, 056 Agertz O., Kravtsov A. V., 2016, The Astrophysical Journal, 824, 79 Anglés-AlcázarD.,Faucher-GiguèreC.-A.,KerešD.,HopkinsP.F.,Quataert E., Murray N., 2017a, Monthly Notices of the Royal Astronomical Soci- ety, 470, 4698 Anglés-Alcázar D., Faucher-GiguèreC.-A., Quataert E., Hop...

work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1907.07648 2019