Pith. sign in

REVIEW 3 major objections 6 minor 102 references

Supersonic Turbulence in Primordial Halos: A Comparison With and Without The Stream Velocity

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The baryon–dark matter stream velocity reshapes turbulence inside primordial halos, boosting it below and suppressing it above a mass of roughly 10^6 solar masses.

desk verdict The stream-velocity turbulence bifurcation at ~10^6 M_sun is new and probably real, but the high-mass suppression is vulnerable to assembly-bias; deserves serious refereeing. read the letter →

arxiv 2507.12993 v1 pith:F4XQ6FYT submitted 2025-07-17 astro-ph.GA

Avi Chen , William Lake , Claire E. Williams , Blakesley Burkhart , Smadar Naoz , Shyam H. Menon , Federico Marinacci , Mark Vogelsberger
show 1 more author
Naoki Yoshida
This is my paper · ORCID
classification astro-ph.GA
keywords streamvelocitysupersonicturbulenceprimordialhalosPopulationIIIstarsfilteringmassaccretionshockscosmologicalsimulationsbaryon–darkmatterrelative
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper asks whether the relative velocity between baryons and dark matter left over from recombination—the 'stream velocity'—changes how turbulent the gas is inside the primordial halos that host the first stars. Using side-by-side cosmological simulations with and without a stream of $v_{\rm bc} = 2\sigma_{v_{\rm bc}}$, the authors find that at a fixed redshift the stream enhances turbulence in halos below roughly $10^6 \, \mathrm{M_\odot}$ and suppresses it in halos above that mass, with the turnover tracking the filtering mass. The low-mass boost is attributed to the stream's own kinetic energy being churned into random gas motions, while the high-mass suppression is attributed to weaker accretion-driven turbulence: streaming halos have shallower potential wells, less dense filaments, and fewer mergers. The result matters because the initial turbulent state inside a halo helps set whether and how it fragments, and therefore how the first generation of stars forms.

What carries the argument

The central quantity is the one-dimensional turbulent velocity $v_{\rm turb}$, defined by subtracting the bulk, radial, and azimuthal rotational motions from each gas cell's velocity and mass-weighting the residual (Equation 5, following the rotational-subtraction method of Safranek-Shrader et al. 2012b). The argument is carried by three comparisons across paired simulations: the mass at which the turbulent velocity difference changes sign, which is matched to the filtering mass computed from the perturbations of Tseliakhovich et al. (2011); the radial velocity and gravitational potential energy of halos, which quantify the energy available to accretion inflows; and the energy dissipation rate of shocked gas cells identified by the shock finder of Schaal & Springel (2015), which measures how much of that inflow energy is thermalized and available to stir the gas.

What would settle it

Match halos across the two simulations by formation time or assembly state (for example, the redshift at which half the final mass was assembled, or the virial ratio) instead of by total mass at a fixed redshift, and recompute the turbulent-velocity curves. If the high-mass suppression disappears or reverses in this matched comparison, then the paper's conclusion that the stream directly suppresses accretion-driven turbulence in high-mass halos would be undercut, leaving only the low-mass enhancement as a robust effect.

Watch

Extended reading notes

Core claim

At $z = 20$–$30$, halos below $M \sim 10^6 \, \mathrm{M_\odot}$ have higher turbulent velocities in the streaming run than in the no-streaming run, while halos above that mass have lower turbulent velocities; a two-sample Kolmogorov–Smirnov test confirms the difference is statistically significant on both sides of the divide. The paper identifies the turnover mass with the streaming-modified filtering mass of $\sim 1.07 \times 10^6 \, \mathrm{M_\odot}$ at $z = 20$, below which pressure forces including the stream dominate, and above which gravity wins. The enhancement in low-mass halos comes from residual kinetic energy that the stream velocity injects into the gas and that nonlinear interactions convert into turbulent and rotational motions. The suppression in high-mass halos is traced to accretion: without streaming, halos have deeper gravitational potentials, denser filamentary gas feeding them, higher radial inflow velocities, and higher shock energy dissipation rates (turning over near $M \sim 2 \times 10^6 \, \mathrm{M_\odot}$), all of which generate turbulence that is weaker when the stream is present. The authors therefore conclude that the stream velocity acts through two opposing channels, with the net effect changing sign near the filtering mass.

Load-bearing premise

The high-mass comparison assumes that a halo of a given total mass at a given redshift is dynamically comparable across the two runs, so that the lower turbulence in streaming halos reflects the stream's suppression of accretion rather than the fact that those halos assembled later and are less evolved.

Editorial extensions

If this is right

  • If the claim is right, the turbulent initial conditions inside primordial halos are not universal: at fixed redshift and fixed halo mass, whether the gas is more or less turbulent than it would be without streaming depends on the local stream velocity and on which side of $\sim 10^6 \, \mathrm{M_\odot}$ the halo sits.
  • In low-mass halos, stream-enhanced turbulence and rotation should promote fragmentation of the collapsing gas and alter the mass distribution of the first stars, consistent with earlier minihalo simulations and with reported enhancements of star formation in low-mass objects.
  • In high-mass halos, the stream suppresses accretion-driven turbulence and shock dissipation, so the gas that forms stars in these more massive halos is kinematically colder relative to a no-streaming halo of the same mass at the same redshift.
  • Because the stream velocity decays as $a(t)^{-1}$, the effect is strongest at the earliest epochs and has largely vanished by $z \sim 10$, which explains the earlier null result for streaming's impact on turbulent energy at that redshift.
  • The mass-dependent shock dissipation difference is a candidate observable: it may produce detectable non-thermal emission from accretion shocks (e.g., inverse Compton emission), although the paper leaves a quantitative prediction to future work.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • One consequence the paper does not develop: if the high-mass suppression is really an assembly offset, then comparing halos matched by formation redshift (or by the same peak height) rather than by total mass at a fixed redshift might weaken or relocate the turnover; this is a direct test of the causal story.
  • The low-mass boost appears to saturate at the smallest resolved halo masses in the higher-resolution run, hinting that there is an upper bound on how much turbulence streaming can inject; probing still lower masses could reveal whether the boost continues or plateaus.
  • The same logic could be extended to other agents that change the filtering mass, such as a Lyman–Werner or X-ray background: each should shift the turbulence turnover mass in the same direction as it shifts the filtering mass, a prediction that could be checked with the same simulation setup.
  • A testable extension would be to measure the turbulence in streaming halos as a function of the angle between the stream direction and the inflowing filament; the production of stream-driven turbulence should be largest where the flow is sheared most strongly against the accretion flow.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper uses two AREPO cosmological simulations of a 2 Mpc comoving box with 512^3 gas cells and 512^3 dark matter particles, evolved from z = 200 to z = 20, one run with a stream velocity vbc = 2σvbc and one without. After subtracting bulk, radial, and rotational motions, the authors measure the turbulent velocity and Mach number of gas inside dark matter halos as a function of halo mass. They report a mass-dependent bifurcation: streaming enhances turbulence in halos with M ≲ 10^6 Msun and suppresses it in halos with M ≳ 10^6 Msun, with the turnover near the filtering mass of Tseliakhovich et al. (2011). The enhancement is attributed to kinetic energy injected by the stream, while the suppression is attributed to weaker accretion-driven turbulence caused by shallower potential wells, lower filamentary gas densities, and fewer mergers in the streaming run. The paper also presents shock dissipation rates and interprets them as supporting the accretion-turbulence mechanism.

Significance. If the central result holds, this is the first systematic population-level demonstration that the stream velocity modulates the initial turbulent state of primordial halos in a mass-dependent way, with a turnover near the filtering mass. The implications for Population III fragmentation and the IMF are physically interesting, and the paper explicitly connects to recent simulations of Pop III star formation. The measurement itself is strengthened by being directly measured from the simulations with no parameter fitting, by being reproduced in a higher-resolution run, and by being corroborated through multiple diagnostics (turbulent velocity, RMS velocity, Mach number, radial velocity, potential energy, IGM density PDFs, and shock dissipation). The KS tests provide statistical support for the low- and high-mass branches separately. The principal weakness is that the causal interpretation of the high-mass suppression rests on an equivalence between streaming and no-streaming halos of equal mass and redshift that is asserted but not controlled for assembly history.

major comments (3)
  1. [§3.2, Figures 4–6, 8] The causal claim that streaming suppresses accretion-driven turbulence in high-mass halos is not uniquely supported by the presented evidence, because the comparison at fixed halo mass and redshift does not control for the assembly offset induced by streaming. Streaming delays structure formation, so a M ~ 10^7 Msun halo in the streaming run is systematically younger and less dynamically evolved than a no-streaming halo of the same mass: it has a shallower potential well (Figure 4), resides in a less dense filamentary environment (Figure 5), has lower radial velocities (Figure 1), and has likely experienced fewer mergers. The lower vturb and lower shock dissipation (Figure 8) in streaming halos are therefore equally consistent with the two halo populations being at different dynamical stages rather than with a direct suppression of accretion-driven turbulence by the stream. I request a control that separates these possibilities: for example, matching halos by formation redshift or assembly mass, by virial ratio or potential energy, or by number of merger progenitors, and then recomputing the velocity and dissipation trends on the matched samples. Without such a control, the conclusion in §5 that 'the suppression in high-mass halos likely results from weaker accretion-driven turbulence' is overstated.
  2. [§3.1, KS tests] The two-sample KS tests do not control for the mass distribution within the two bins, which matters because streaming suppresses the abundance of high-mass halos. If the M > 10^6 Msun bin in the streaming run is skewed toward lower masses relative to the no-streaming run, the KS statistic can reject identical turbulence distributions even when halos of exactly equal mass have statistically identical turbulence. The reported p-values of ~10^-67 and ~10^-9 are therefore not, by themselves, evidence that the turnover is a property of halos at fixed mass. Please repeat the KS tests on mass-matched subsamples (for instance, nearest-neighbor matching in Mvir or a quantile-regression approach with halo mass as a covariate), or report the mass distributions within the bins and show that the KS result is robust to the mass-skew.
  3. [§3.4, Eqs. (14)–(17)] The chain from higher shock dissipation to higher turbulence is not directly established. Equation (14) measures the thermal energy flux through shocks, and the text states that some of this energy is converted into non-thermal motions through post-shock Richtmyer–Meshkov and Kelvin–Helmholtz instabilities, but no diagnostic in the paper links the dissipation rate to the turbulent velocity within individual halos. As written, Figure 8 shows that the mean dissipation rate has a similar mass-dependent trend to vturb, but this could follow from the same assembly offset discussed above rather than from a causal connection between shocks and turbulence. A scatter plot or correlation of dissipation rate versus vturb at fixed halo mass, or a comparison of shocked gas mass fraction between the two runs, would make the proposed mechanism testable and would considerably strengthen the interpretation.
minor comments (6)
  1. [Title] The title contains a spacing typo: 'V elocity' should be 'Velocity'.
  2. [Appendix B, Figure 10] Figure 10 shows percent differences for vbc = 1, 2, and 3 σvbc without the shaded 1σ scatter bands used in the main figures; adding uncertainty bands would clarify whether the vbc-dependence is significant at the turnover and at high masses.
  3. [§2.2] The choice σ8 = 1.7 and the statement that it accelerates structure formation by a factor of sqrt(2) but 'does not otherwise affect the physics' would benefit from a more quantitative justification, since the mapping of results to physical redshifts depends on this normalization.
  4. [References] Several references appear duplicated or potentially mislabeled: Park et al. 2020a and 2020b have the same arXiv identifier, and Williams et al. 2025a and 2025b appear to refer to the same paper; these should be checked and corrected.
  5. [§4] The paragraph discussing hot-mode versus cold-mode accretion is somewhat disconnected from the quantitative shock analysis; connecting it explicitly to the dissipation rates in Figure 8 or shortening it would improve the flow of the discussion.
  6. [Reproducibility] The paper does not ship code, halo catalogs, or a data-availability statement; given that the central claim is based on population statistics rather than a single object, making the halo catalogs and analysis scripts available would substantially aid verification.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: turbulence is measured directly from the simulations, the turnover is compared post hoc to an independently computed filtering mass, and self-citations are contextual rather than load-bearing.

full rationale

The paper's central quantities are direct simulation outputs: vturb is defined by Equation 5 from the gas velocity field after subtracting bulk, radial, and rotational components, and the shock dissipation rates in Section 3.4 are computed with an external shock finder (Schaal & Springel 2015) applied to the simulated gas. No parameter is fitted to any target result, and no predicted ratio is built into a definition. The turnover near M ~ 10^6 M_sun is identified from the measured velocity curves in Figure 1 and then compared to the independently derived filtering mass of Tseliakhovich et al. (2011), an external result, not fitted to it. The causal language is explicitly tentative: the paper says the enhancement 'likely arises' and the suppression 'likely arises from a reduction in inflowing accretion-driven turbulence,' and the filtering-mass connection is presented as 'closely matches' rather than derived from it. Self-citations appear in the methods (e.g., the initial-condition equations from Naoz et al. 2013, which themselves build on Tseliakhovich et al. 2011 and Naoz & Barkana 2005) and in the interpretation sections as prior context or consistency checks, but the central measurement does not reduce to any self-cited result. The main scientific caveat, that the fixed-mass comparison at fixed redshift may be affected by assembly bias because streaming delays structure formation, is a validity concern about the interpretation and not a circular reduction in the paper's own derivation chain. No equation is identical by construction to an input, and no fitted parameter is renamed as a prediction. The paper is self-contained against external benchmarks for its measured turbulent velocities, so the circularity burden is minimal.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

All inputs to the central claim are standard cosmology, prior stream-velocity theory, and explicitly stated numerical choices (sigma8, vbc, resolution). No new entities are invented. The main hand-set inputs are listed as free parameters.

free parameters (3)
  • sigma8 normalization = 1.7
    Chosen by hand to accelerate structure formation in the 2 Mpc box; the paper assumes this only rescales formation time by sqrt(2) and does not otherwise affect the physics (Section 2.2). It affects the mass scale and redshift mapping of all results.
  • stream velocity magnitude = 2 sigma_vbc (plus 1 and 3 sigma_vbc in Appendix B)
    The main treatment variable; 2 sigma_vbc is chosen to realize the full streaming effect and match the local value estimate from Uysal and Hartwig (2023). Quantitative percent differences depend on this choice.
  • mass resolution limit = 300 DM particles and 100 gas cells per halo; initial gas mass 360 solar masses
    Limits the low-mass end; the resolution study uses 200 solar mass gas cells. The lowest-mass bins may be affected by resolution.
assumptions (5)
  • domain assumption Standard LCDM cosmology with Omega_L=0.73, Omega_m=0.27, Omega_b=0.044, h=0.71 and sigma8=1.7 provides a valid background for the stream velocity study.
    Stated at the end of Section 1 and in Section 2.2; the nonstandard sigma8 is load-bearing for the accelerated structure formation.
  • domain assumption The stream velocity is coherent on scales larger than the 2 Mpc box, so implementing it as a uniform velocity boost in the x-direction is valid.
    Section 2.2, based on Tseliakhovich and Hirata (2010). If the coherence scale were comparable to the box, the uniform boost would misrepresent the effect.
  • domain assumption Subtracting bulk, radial, and solid-body rotational velocities leaves a residual that is a faithful measure of turbulence.
    Equation 5, Section 3.1. Non-solid-body coherent flows could contaminate the residual.
  • domain assumption The absence of magnetic fields, star formation, and feedback does not change the relative stream/no-stream turbulence trends at the halo scale.
    Section 2.2 states these are not included; the paper argues they are secondary for the initial conditions studied.
  • domain assumption The time-rescaling equivalence for sigma8=1.7 (Park et al. 2020b) holds for turbulence statistics.
    Section 2.2: the paper claims this 'does not otherwise affect the physics'. This is the bridge from the accelerated box to real redshifts.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Supersonic Turbulence in Primordial Halos: A Comparison With and Without The Stream Velocity." pith.science (2026). https://pith.science/paper/F4XQ6FYT

@misc{pith2026250712993,
  author       = {Pith},
  title        = {Pith review of: Supersonic Turbulence in Primordial Halos: A Comparison With and Without The Stream Velocity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F4XQ6FYT}},
  note         = {Machine review of arXiv:2507.12993}
}
abstract

Turbulence plays a critical role in regulating star formation in molecular clouds and is also observed in simulations of primordial halos that host Population III (Pop III) stars. The relative velocity between baryons and dark matter at the time of recombination is thought to be a source of turbulence in the early universe. In this paper, we study how this stream velocity affects the turbulence inside primordial halos using high-resolution cosmological simulations across the redshift range of $z = 30$ to $z = 20$. We find that at a fixed redshift, the stream velocity enhances turbulence in low-mass halos ($M \lesssim 10^6 \ \mathrm{M_\odot}$) and suppresses it for high-mass halos ($M \gtrsim 10^6 \ \mathrm{M_\odot}$). The enhancement in low-mass halos likely arises from residual kinetic energy introduced by the stream velocity, while the suppression in high-mass halos likely arises from a reduction in inflowing accretion-driven turbulence. This mass-dependent modulation of turbulence suggests that the initial conditions inside primordial halos are altered in the presence of the stream velocity, potentially influencing their fragmentation and the resulting star formation.

Figures

Figures reproduced from arXiv: 2507.12993 by the authors.

Figure 1
Figure 1. Comparison of kinematic properties of gas within halos in streaming (vbc = 2σvbc, blue) and no-streaming (vbc = 0σvbc, red), simulations at z = 20. Panels from top to bottom show turbu￾lent velocity, RMS velocity , radial velocity, and rotational velocity, each plotted against total halo mass (gas+DM). Shaded color bands indicate the 1σ scatter around the mean values. Streaming enhances turbulent and RMS velocities … view at source ↗
Figure 2
Figure 2. Percent difference in the turbulence with and without streaming at different redshifts. The percent difference generally declines as a function of redshift as the stream velocity decays. The turnover at M ∼ 106 M⊙ also shifts slightly toward higher masses at earlier times and is marked by the gray shaded band [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Turbulent Mach number as a function of total halo mass with and without streaming at z = 20. Shaded color bands indicate the 1σ scatter around the mean values. A clear turnover appears near M ∼ 106 M⊙, marked by a gray band, mirroring the trend in the turbulent velocity ( [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Gravitational potential energy of halos as a function of total mass (gas + DM) at z = 20 with and without streaming. The top panel shows the halo’s gravitational potential energy, computed from gas–gas, DM–DM, and gas–DM interactions for all gas cells and dark matter p…
Figure 5
Figure 5. Figure 5: Projected gas density surrounding representative halos from two mass bins in simulations with streaming (bottom rows in each pair) and without streaming (top rows in each pair) at z = 20. Each panel shows a 1 kpc (physical) cutout centered on the halo’s center of mass.…
Figure 6
Figure 6. Figure 6: Probability density functions (PDFs) of gas velocity (left) and number density (right) in the intergalactic medium (IGM) at z = 20 with and without streaming. The IGM is defined as all gas not inside halos as identified by the FoF. Bottom subpanels show the ratio of th…
Figure 7
Figure 7. Figure 7: Shocked gas in an illustrative halo of mass M ≃ 107 M⊙ at z = 20. Top panel: All gas within the halo are shown in gray, while shocked gas are overplotted in blue. Red arrows indicate normalized velocity vectors of shocked gas outside the virial radius of the halo, and …
Figure 8
Figure 8. Figure 8: Average energy dissipation rate from shocked gas as a function of total halo mass at z = 20. Shaded color bands in￾dicate the 1σ scatter around the mean values. At high-mass ha￾los (M ≳ 106 M⊙), no-streaming halos exhibit higher dissipation rates, consistent with enhan…
Figure 10
Figure 10. Figure 10: shows the percent difference as a function of total halo mass. The vbc = 3σvbc run exhibits both a stronger enhancement in turbulence in low-mass halos and a shift of the turnover mass to higher values. In contrast, the vbc = 1σvbc run shows a more modest enhancement,…
Figure 9
Figure 9. Figure 9: RMS and turbulent velocities as a function of halo mass in the higher-resolution run. Shaded color bands indicate the 1σ scatter around the mean values. The turnover in turbulence remains at M ∼ 106 M⊙, marked by the gray shaded band, while the dif￾ference between stre…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

102 extracted references · 23 canonical work pages

  1. [1]

    D., & Yuan, S

    Ali-Ha¨ımoud, Y ., Meerburg, P. D., & Yuan, S. 2014, Phys. Rev. D, 89, 083506, doi: 10.1103/PhysRevD.89.083506

    doi:10.1103/physrevd.89.083506 2014

  2. [2]

    M., Burkhart, B., Semenov, V

    Appel, S. M., Burkhart, B., Semenov, V . A., et al. 2023, ApJ, 954, 93, doi: 10.3847/1538-4357/ace897

    doi:10.3847/1538-4357/ace897 2023

  3. [3]

    2016, PhRvD, 93, 023518, doi: 10.1103/PhysRevD.93.023518

    Asaba, S., Ichiki, K., & Tashiro, H. 2016, PhRvD, 93, 023518, doi: 10.1103/PhysRevD.93.023518

    doi:10.1103/physrevd.93.023518 2016

  4. [4]

    2003, MNRAS, 345, 349, doi: 10.1046/j.1365-8711.2003.06955.x

    Birnboim, Y ., & Dekel, A. 2003, MNRAS, 345, 349, doi: 10.1046/j.1365-8711.2003.06955.x

    arXiv 2003

  5. [5]

    A., Grassi, T., & Schleicher, D

    Bovino, S., Latif, M. A., Grassi, T., & Schleicher, D. R. G. 2014, MNRAS, 441, 2181, doi: 10.1093/mnras/stu714

    doi:10.1093/mnras/stu714 2014

  6. [6]

    2013, ApJ, 768, 70, doi: 10.1088/0004-637X/768/1/70

    Bovy, J., & Dvorkin, C. 2013, ApJ, 768, 70, doi: 10.1088/0004-637X/768/1/70

    doi:10.1088/0004-637x/768/1/70 2013

  7. [7]

    2018, ApJ, 863, 118, doi: 10.3847/1538-4357/aad002

    Burkhart, B. 2018, ApJ, 863, 118, doi: 10.3847/1538-4357/aad002

    doi:10.3847/1538-4357/aad002 2018

  8. [8]

    2018, ArXiv e-prints

    Burkhart, B., & Mocz, P. 2018, ArXiv e-prints. https://arxiv.org/abs/1805.11104

    arXiv 2018

Show all 102 references
  1. [9]

    2020, ApJ, 898, 168, doi: 10.3847/1538-4357/aba26a

    Cain, C., D’Aloisio, A., Irˇsiˇc, V ., McQuinn, M., & Trac, H. 2020, ApJ, 898, 168, doi: 10.3847/1538-4357/aba26a

    2020 doi

  2. [10]

    2025, arXiv e-prints, arXiv:2505.18964, doi: 10.48550/arXiv.2505.18964

    Chen, K.-J., Ho, M.-Y ., & Tung, P.-C. 2025, arXiv e-prints, arXiv:2505.18964, doi: 10.48550/arXiv.2505.18964

    2025 doi arXiv

  3. [11]

    Chiaki, G., & Wise, J. H. 2019, MNRAS, 482, 3933, doi: 10.1093/mnras/sty2984

    2019 doi

  4. [12]

    S., Naoz, S., Marinacci, F., & V ogelsberger, M

    Chiou, Y . S., Naoz, S., Marinacci, F., & V ogelsberger, M. 2018, MNRAS, 481, 3108, doi: 10.1093/mnras/sty2480

    2018 doi

  5. [13]

    C., Glover, S

    Clark, P. C., Glover, S. C. O., & Klessen, R. S. 2008, ApJ, 672, 757, doi: 10.1086/524187

    2008 doi

  6. [14]

    C., Glover, S

    Clark, P. C., Glover, S. C. O., Klessen, R. S., & Bromm, V . 2011, ApJ, 727, 110, doi: 10.1088/0004-637X/727/2/110

    2011 doi

  7. [15]

    2010, JCAP, 11, 7, doi: 10.1088/1475-7516/2010/11/007

    Dalal, N., Pen, U., & Seljak, U. 2010, JCAP, 11, 7, doi: 10.1088/1475-7516/2010/11/007

    2010 doi

  8. [16]

    Druschke, M., Schauer, A. T. P., Glover, S. C. O., & Klessen, R. S. 2018, MNRAS, 481, 3266, doi: 10.1093/mnras/sty2443 13

    2018 doi

  9. [17]

    2016, A&A, 585, A59, doi: 10.1051/0004-6361/201526747 Falceta-Gonc ¸alves, D., Kowal, G., Falgarone, E., & Chian, A

    Dutta, J. 2016, A&A, 585, A59, doi: 10.1051/0004-6361/201526747 Falceta-Gonc ¸alves, D., Kowal, G., Falgarone, E., & Chian, A. C. L. 2014, Nonlinear Processes in Geophysics, 21, 587, doi: 10.5194/npg-21-587-2014

    2016 doi

  10. [18]

    L., Haiman, Z., & Li, M

    Fernandez, R., Bryan, G. L., Haiman, Z., & Li, M. 2014, MNRAS, 439, 3798, doi: 10.1093/mnras/stu230

    2014 doi

  11. [19]

    2014, International Journal of Modern Physics D, 23, 1430017, doi: 10.1142/S0218271814300171

    Fialkov, A. 2014, International Journal of Modern Physics D, 23, 1430017, doi: 10.1142/S0218271814300171

    2014 doi

  12. [21]

    Gnedin, N. Y . 2000, ApJ, 542, 535, doi: 10.1086/317042

    2000 doi

  13. [23]

    H., Johnson, J

    Greif, T. H., Johnson, J. L., Klessen, R. S., & Bromm, V . 2008, Monthly Notices of the Royal Astronomical Society, 387, 1021, doi: 10.1111/j.1365-2966.2008.13326.x

    2008

  14. [24]

    H., White, S

    Greif, T. H., White, S. D. M., Klessen, R. S., & Springel, V . 2011a, ApJ, 736, 147, doi: 10.1088/0004-637X/736/2/147 —. 2011b, ApJ, 736, 147, doi: 10.1088/0004-637X/736/2/147

    doi

  15. [25]

    R., Millman, K

    Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357, doi: 10.1038/s41586-020-2649-2

    2020 doi

  16. [26]

    Hegde, S., & Furlanetto, S. R. 2023, MNRAS, 525, 428, doi: 10.1093/mnras/stad2308

    2023 doi

  17. [27]

    2021, ApJ, 915, 107, doi: 10.3847/1538-4357/ac01c7

    Higashi, S., Susa, H., & Chiaki, G. 2021, ApJ, 915, 107, doi: 10.3847/1538-4357/ac01c7

    2021 doi

  18. [28]

    2025, Formation of first star clusters under the supersonic gas flow – II

    Hirano, S. 2025, Formation of first star clusters under the supersonic gas flow – II. Critical halo mass and core mass function. https://arxiv.org/abs/2501.17491

    2025 arXiv

  19. [29]

    2017, Science, 357, 1375, doi: 10.1126/science.aai9119

    Hirano, S., Hosokawa, T., Yoshida, N., & Kuiper, R. 2017, Science, 357, 1375, doi: 10.1126/science.aai9119

    2017 doi

  20. [30]

    Hirano, S., Hosokawa, T., Yoshida, N., Omukai, K., & Yorke, H. W. 2015, MNRAS, 448, 568, doi: 10.1093/mnras/stv044

    2015 doi

  21. [31]

    2014, ApJ, 781, 60, doi: 10.1088/0004-637X/781/2/60

    Hirano, S., Hosokawa, T., Yoshida, N., et al. 2014, ApJ, 781, 60, doi: 10.1088/0004-637X/781/2/60

    2014 doi

  22. [32]

    Hirano, S., Yoshida, N., Sakurai, Y ., & Fujii, M. S. 2018, ApJ, 855, 17, doi: 10.3847/1538-4357/aaaaba

    2018 doi

  23. [33]

    W., Yuen, K

    Ho, K. W., Yuen, K. H., & Lazarian, A. 2023, MNRAS, 521, 230, doi: 10.1093/mnras/stad481

    2023 doi

  24. [34]

    2025, arXiv e-prints, arXiv:2505.23768, doi: 10.48550/arXiv.2505.23768

    Ho, M.-Y ., Chen, K.-J., & Tung, P.-C. 2025, arXiv e-prints, arXiv:2505.23768, doi: 10.48550/arXiv.2505.23768

    2025 doi arXiv

  25. [35]

    A., Stacy, A., & Bromm, V

    Hummel, J. A., Stacy, A., & Bromm, V . 2016, MNRAS, 460, 2432, doi: 10.1093/mnras/stw1127

    2016 doi

  26. [36]

    A., Stacy, A., Jeon, M., Oliveri, A., & Bromm, V

    Hummel, J. A., Stacy, A., Jeon, M., Oliveri, A., & Bromm, V . 2015, MNRAS, 453, 4136, doi: 10.1093/mnras/stv1902

    2015 doi

  27. [37]

    Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90, doi: 10.1109/MCSE.2007.55

    2007 doi

  28. [38]

    Jaura, O., Glover, S. C. O., Wollenberg, K. M. J., et al. 2022, MNRAS, 512, 116, doi: 10.1093/mnras/stac487

    2022 doi

  29. [40]

    2023, MNRAS, 523, 1496, doi: 10.1093/mnras/stad1484

    Kiyuna, M., Hosokawa, T., & Chon, S. 2023, MNRAS, 523, 1496, doi: 10.1093/mnras/stad1484

    2023 doi

  30. [41]

    S., & Glover, S

    Klessen, R. S., & Glover, S. C. O. 2023, ARA&A, 61, 65, doi: 10.1146/annurev-astro-071221-053453

    2023 doi

  31. [42]

    R., & McKee, C

    Krumholz, M. R., & McKee, C. F. 2005, ApJ, 630, 250, doi: 10.1086/431734

    2005 doi

  32. [43]

    Kulkarni, M., Visbal, E., & Bryan, G. L. 2021, ApJ, 917, 40, doi: 10.3847/1538-4357/ac08a3

    2021 doi

  33. [44]

    S., et al

    Lake, W., Naoz, S., Chiou, Y . S., et al. 2021, ApJ, 922, 86, doi: 10.3847/1538-4357/ac20d0

    2021 doi

  34. [45]

    2023a, ApJ, 943, 132, doi: 10.3847/1538-4357/acac8d

    Lake, W., Naoz, S., Burkhart, B., et al. 2023a, ApJ, 943, 132, doi: 10.3847/1538-4357/acac8d

    doi

  35. [46]

    2023b, ApJL, 956, L7, doi: 10.3847/2041-8213/acfa9b

    Lake, W., Naoz, S., Marinacci, F., et al. 2023b, ApJL, 956, L7, doi: 10.3847/2041-8213/acfa9b

    doi

  36. [47]

    E., Naoz, S., et al

    Lake, W., Williams, C. E., Naoz, S., et al. 2024, ApJ, 973, 115, doi: 10.3847/1538-4357/ad6762

    2024 doi

  37. [48]

    Y ., Naoz, S., et al

    Lake, W., Grudi´c, M. Y ., Naoz, S., et al. 2025, ApJL, 985, L6, doi: 10.3847/2041-8213/add347

    2025 doi

  38. [49]

    Larson, R. B. 1981, MNRAS, 194, 809, doi: 10.1093/mnras/194.4.809

    1981 doi

  39. [50]

    A., Niemeyer, J

    Latif, M. A., Niemeyer, J. C., & Schleicher, D. R. G. 2014a, MNRAS, 440, 2969, doi: 10.1093/mnras/stu489

    doi

  40. [51]

    A., Schleicher, D

    Latif, M. A., Schleicher, D. R. G., & Schmidt, W. 2014b, MNRAS, 440, 1551, doi: 10.1093/mnras/stu357

    doi

  41. [52]

    2006, MNRAS, 373, 1195, doi: 10.1111/j.1365-2966.2006.11093.x

    Lazarian, A., & Beresnyak, A. 2006, MNRAS, 373, 1195, doi: 10.1111/j.1365-2966.2006.11093.x

    2006

  42. [53]

    2023, Frontiers in Astronomy and Space Sciences, 10, 1154760, doi: 10.3389/fspas.2023.1154760

    Lazarian, A., Xu, S., & Hu, Y . 2023, Frontiers in Astronomy and Space Sciences, 10, 1154760, doi: 10.3389/fspas.2023.1154760

    2023

  43. [54]

    2024, MNRAS, 534, 290, doi: 10.1093/mnras/stae2066

    Liu, B., Gurian, J., Inayoshi, K., et al. 2024, MNRAS, 534, 290, doi: 10.1093/mnras/stae2066

    2024 doi

  44. [55]

    Loeb, A., & Furlanetto, S. R. 2013, The First Galaxies in the Universe (Princeton University Press)

    2013

  45. [56]

    J., & Hirata, C

    Long, H., Givans, J. J., & Hirata, C. M. 2022, MNRAS, 513, 117, doi: 10.1093/mnras/stac658

    2022 doi

  46. [57]

    Maio, U., Koopmans, L. V . E., & Ciardi, B. 2011, MNRAS, L197+, doi: 10.1111/j.1745-3933.2010.01001.x

    2011

  47. [58]

    McQuinn, M., & O’Leary, R. M. 2012, ApJ, 760, 3, doi: 10.1088/0004-637X/760/1/3 Mu˜noz, J. B. 2019, PhRvD, 100, 063538, doi: 10.1103/PhysRevD.100.063538

    2012 doi

  48. [59]

    2022, ApJL, 927, L12, doi: 10.3847/2041-8213/ac573e

    Nakazato, Y ., Chiaki, G., Yoshida, N., et al. 2022, ApJL, 927, L12, doi: 10.3847/2041-8213/ac573e

    2022 doi

  49. [60]

    2005, MNRAS, 362, 1047, doi: 10.1111/j.1365-2966.2005.09385.x —

    Naoz, S., & Barkana, R. 2005, MNRAS, 362, 1047, doi: 10.1111/j.1365-2966.2005.09385.x —. 2007, MNRAS, 377, 667, doi: 10.1111/j.1365-2966.2007.11636.x 14

    2005

  50. [61]

    2013, Physical Review Letters, 111, 051303, doi: 10.1103/PhysRevLett.111.051303 —

    Naoz, S., & Narayan, R. 2013, Physical Review Letters, 111, 051303, doi: 10.1103/PhysRevLett.111.051303 —. 2014, ApJL, 791, L8, doi: 10.1088/2041-8205/791/1/L8

    2013 doi

  51. [62]

    Naoz, S., Yoshida, N., & Gnedin, N. Y . 2012, ApJ, 747, 128, doi: 10.1088/0004-637X/747/2/128 —. 2013, ApJ, 763, 27, doi: 10.1088/0004-637X/763/1/27

    2012 doi

  52. [63]

    K., & Mellema, G

    Nebrin, O., Giri, S. K., & Mellema, G. 2023, MNRAS, 524, 2290, doi: 10.1093/mnras/stad1852 O’Leary, R. M., & McQuinn, M. 2012, ApJ, 760, 4, doi: 10.1088/0004-637X/760/1/4

    2023 doi

  53. [64]

    2020a, ApJ, 900, 30, doi: 10.3847/1538-4357/aba26e —

    Park, H., Ahn, K., Yoshida, N., & Hirano, S. 2020a, ApJ, 900, 30, doi: 10.3847/1538-4357/aba26e —. 2020b, ApJ, 900, 30, doi: 10.3847/1538-4357/aba26e

    doi

  54. [65]

    R., Ahn, K., Yoshida, N., & Hirano, S

    Park, H., Shapiro, P. R., Ahn, K., Yoshida, N., & Hirano, S. 2021, ApJ, 908, 96, doi: 10.3847/1538-4357/abd7f4

    2021 doi

  55. [66]

    2016, MNRAS, 460, 1625, doi: 10.1093/mnras/stw1045

    Popa, C., Naoz, S., Marinacci, F., & V ogelsberger, M. 2016, MNRAS, 460, 1625, doi: 10.1093/mnras/stw1045

    2016 doi

  56. [68]

    2011, ApJL, 731, L38, doi: 10.1088/2041-8205/731/2/L38

    Prieto, J., Padoan, P., Jimenez, R., & Infante, L. 2011, ApJL, 731, L38, doi: 10.1088/2041-8205/731/2/L38

    2011 doi

  57. [69]

    Riaz, R., Bovino, S., Vanaverbeke, S., & Schleicher, D. R. G. 2018, MNRAS, 479, 667, doi: 10.1093/mnras/sty1635

    2018 doi

  58. [70]

    Klessen, R. S. 2023, MNRAS, 518, 4895, doi: 10.1093/mnras/stac3310

    2023 doi

  59. [71]

    Richardson, M. L. A., Scannapieco, E., & Thacker, R. J. 2013, ApJ, 771, 81, doi: 10.1088/0004-637X/771/2/81

    2013 doi

  60. [72]

    2012a, MNRAS, 426, 1159, doi: 10.1111/j.1365-2966.2012.21852.x —

    Safranek-Shrader, C., Agarwal, M., Federrath, C., et al. 2012a, MNRAS, 426, 1159, doi: 10.1111/j.1365-2966.2012.21852.x —. 2012b, MNRAS, 426, 1159, doi: 10.1111/j.1365-2966.2012.21852.x

    2012

  61. [73]

    2015, MNRAS, 446, 3992, doi: 10.1093/mnras/stu2386

    Schaal, K., & Springel, V . 2015, MNRAS, 446, 3992, doi: 10.1093/mnras/stu2386

    2015 doi

  62. [74]

    2016, MNRAS, 461, 4441, doi: 10.1093/mnras/stw1587

    Schaal, K., Springel, V ., Pakmor, R., et al. 2016, MNRAS, 461, 4441, doi: 10.1093/mnras/stw1587

    2016 doi

  63. [75]

    Schauer, A. T. P., Glover, S. C. O., & Klessen, R. S. 2017, Mem. Soc. Astron. Italiana, 88, 702

    2017

  64. [76]

    Schauer, A. T. P., Glover, S. C. O., Klessen, R. S., & Ceverino, D. 2019, MNRAS, 484, 3510, doi: 10.1093/mnras/stz013

    2019 doi

  65. [77]

    Schauer, A. T. P., Glover, S. C. O., Klessen, R. S., & Clark, P. 2021, MNRAS, 507, 1775, doi: 10.1093/mnras/stab1953

    2021 doi

  66. [78]

    1996, ApJ, 469, 437, doi: 10.1086/177793

    Seljak, U., & Zaldarriaga, M. 1996, ApJ, 469, 437, doi: 10.1086/177793

    1996 doi

  67. [79]

    Sharda, P., Federrath, C., & Krumholz, M. R. 2020, MNRAS, 497, 336, doi: 10.1093/mnras/staa1926

    2020 doi

  68. [80]

    R., & Schleicher, D

    Sharda, P., Federrath, C., Krumholz, M. R., & Schleicher, D. R. G. 2021, MNRAS, 503, 2014, doi: 10.1093/mnras/stab531

    2021 doi

  69. [81]

    Sharda, P., & Menon, S. H. 2024, arXiv e-prints, arXiv:2405.18265, doi: 10.48550/arXiv.2405.18265

    2024 doi arXiv

  70. [82]

    Slepian, Z., & Eisenstein, D. J. 2015, Monthly Notices of the Royal Astronomical Society, 448, 9, doi: 10.1093/mnras/stu2627

    2015 doi

  71. [83]

    D., Bryan, G

    Smith, B. D., Bryan, G. L., Glover, S. C. O., et al. 2017, MNRAS, 466, 2217, doi: 10.1093/mnras/stw3291

    2017 doi

  72. [84]

    Springel, V ., White, S. D. M., Tormen, G., & Kauffmann, G. 2001, MNRAS, 328, 726, doi: 10.1046/j.1365-8711.2001.04912.x

    2001

  73. [85]

    2011, ApJL, 730, L1, doi: 10.1088/2041-8205/730/1/L1

    Stacy, A., Bromm, V ., & Loeb, A. 2011, ApJL, 730, L1, doi: 10.1088/2041-8205/730/1/L1

    2011 doi

  74. [86]

    H., & Bromm, V

    Stacy, A., Greif, T. H., & Bromm, V . 2010, MNRAS, 403, 45, doi: 10.1111/j.1365-2966.2009.16113.x —. 2012, MNRAS, 422, 290, doi: 10.1111/j.1365-2966.2012.20605.x

    2010

  75. [87]

    2023, ApJ, 959, 17, doi: 10.3847/1538-4357/ad02fc

    Omukai, K. 2023, ApJ, 959, 17, doi: 10.3847/1538-4357/ad02fc

    2023 doi

  76. [88]

    2014, ApJ, 792, 32, doi: 10.1088/0004-637X/792/1/32

    Susa, H., Hasegawa, K., & Tominaga, N. 2014, ApJ, 792, 32, doi: 10.1088/0004-637X/792/1/32

    2014 doi

  77. [89]

    L., & Li, M

    Tanaka, T. L., & Li, M. 2014, MNRAS, 439, 1092, doi: 10.1093/mnras/stu042

    2014 doi

  78. [90]

    Tseliakhovich, D., Barkana, R., & Hirata, C. M. 2011, MNRAS, 418, 906, doi: 10.1111/j.1365-2966.2011.19541.x

    2011

  79. [91]

    2010, PhRvD, 82, 083520, doi: 10.1103/PhysRevD.82.083520

    Tseliakhovich, D., & Hirata, C. 2010, PhRvD, 82, 083520, doi: 10.1103/PhysRevD.82.083520

    2010 doi

  80. [92]

    J., Smith, B

    Turk, M. J., Smith, B. D., Oishi, J. S., et al. 2011, ApJS, 192, 9, doi: 10.1088/0067-0049/192/1/9

    2011 doi

  81. [93]

    2023, MNRAS, 520, 3229, doi: 10.1093/mnras/stad350

    Uysal, B., & Hartwig, T. 2023, MNRAS, 520, 3229, doi: 10.1093/mnras/stad350

    2023 doi

  82. [94]

    E., et al

    Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods, 17, 261, doi: 10.1038/s41592-019-0686-2

    2020 doi

  83. [95]

    Visbal, E., Barkana, R., Fialkov, A., Tseliakhovich, D., & Hirata, C. M. 2012, Nature, 487, 70, doi: 10.1038/nature11177 V ogelsberger, M., Marinacci, F., Torrey, P., & Puchwein, E. 2020, Nature Reviews Physics, 2, 42, doi: 10.1038/s42254-019-0127-2

    2012 doi

  84. [96]

    2020, ApJS, 248, 32, doi: 10.3847/1538-4365/ab908c

    Weinberger, R., Springel, V ., & Pakmor, R. 2020, ApJS, 248, 32, doi: 10.3847/1538-4365/ab908c

    2020 doi

  85. [97]

    E., Naoz, S., Lake, W., et al

    Williams, C. E., Naoz, S., Lake, W., et al. 2023, ApJ, 945, 6, doi: 10.3847/1538-4357/acb820

    2023 doi

  86. [98]

    E., Lake, W., Naoz, S., et al

    Williams, C. E., Lake, W., Naoz, S., et al. 2024, The Astrophysical Journal Letters, 960, L16, doi: 10.3847/2041-8213/ad1491

    2024 doi

  87. [99]

    E., Naoz, S., Lake, W., et al

    Williams, C. E., Naoz, S., Lake, W., et al. 2025a, arXiv e-prints, arXiv:2502.17561, doi: 10.48550/arXiv.2502.17561 —. 2025b, arXiv e-prints, arXiv:2502.17561, doi: 10.48550/arXiv.2502.17561

    doi arXiv

  88. [100]

    H., & Abel, T

    Wise, J. H., & Abel, T. 2007, ApJ, 665, 899, doi: 10.1086/520036

    2007 doi

  89. [101]

    Wollenberg, K. M. J., Glover, S. C. O., Clark, P. C., & Klessen, R. S. 2020, MNRAS, 494, 1871, doi: 10.1093/mnras/staa289

    2020 doi

  90. [102]

    2011, JCAP, 7, 18, doi: 10.1088/1475-7516/2011/07/018

    Yoo, J., Dalal, N., & Seljak, U. 2011, JCAP, 7, 18, doi: 10.1088/1475-7516/2011/07/018

    2011 doi

  91. [103]

    2003, The Astrophysical Journal, 592, 645, doi: 10.1086/375810 15

    Yoshida, N., Abel, T., Hernquist, L., & Sugiyama, N. 2003, The Astrophysical Journal, 592, 645, doi: 10.1086/375810 15

    2003 doi

  92. [104]

    2007, ApJL, 667, L117, doi: 10.1086/522202

    Yoshida, N., Omukai, K., & Hernquist, L. 2007, ApJL, 667, L117, doi: 10.1086/522202

    2007 doi

  93. [105]

    2006, ApJ, 652, 6, doi: 10.1086/507978

    Yoshida, N., Omukai, K., Hernquist, L., & Abel, T. 2006, ApJ, 652, 6, doi: 10.1086/507978

    2006 doi

  94. [106]

    H., Ho, K

    Yuen, K. H., Ho, K. W., Law, C. Y ., & Chen, A. 2024, Reviews of Modern Plasma Physics, 8, 21, doi: 10.1007/s41614-024-00156-5 16 APPENDIX A. RESOLUTION STUDY To assess the impact of resolution on our velocity curves, we repeat the analysis using a higher-resolution setup: a 2...

    2024 doi

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.