Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

The UV-bright galaxy excess at z>9 traces a dust-origin transition: supernova-produced, reverse-shock-processed dust with intrinsically low far-UV opacity, not dust-free gas or extreme star-formation efficiency.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-02 14:25 UTC pith:AQLEUWVQ

load-bearing objection A plausible and honest extension of the stardust scenario with new attenuation machinery; the central κ_UV constraint is visually calibrated and needs a censored fit before I'd call it established, but it deserves a serious referee. the 3 major comments →

arxiv 2605.09829 v3 pith:AQLEUWVQ submitted 2026-05-11 astro-ph.GA astro-ph.CO

Stardust Galaxies at z>9: A Dust-Origin Transition Behind the Excess of UV-Bright Galaxies

D. Burgarella , V. Buat , A.K. Inoue , T.T. Takeuchi , C. Aurin , J.-C. Bouret , P. Dayal , T. Dewachter
show 4 more authors
M. Dickinson C. Kobayashi G. P. Nikopoulos R. S. Somerville
This is my paper
classification astro-ph.GA astro-ph.CO
keywords high-redshift galaxiesdust attenuationsupernova dustGELDAsUV luminosity functionJWSTporous geometrydust opacity
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper argues that the surprisingly abundant UV-bright galaxies seen by JWST at z>9 are not powered by unusually efficient star formation or by dust-free interstellar media. Instead, their low far-UV attenuation reflects a dust-origin transition: at these early times, dust comes predominantly from supernova ejecta whose small grains have been destroyed by reverse shocks, leaving large grains with intrinsically low UV opacity. Using a physically motivated attenuation framework that combines SN-dust extinction laws, metallicity- and dust-to-metal-dependent scalings, and porous star-dust geometries, the authors reproduce the observed attenuation-stellar-mass relation and the dominance of galaxies with extremely low dust attenuation (GELDAs). Applied to intrinsic UV luminosity functions, the SN-dominated and hybrid prescriptions suppress only the brightest sources, reconciling theory with JWST measurements. The central claim is that the UV excess is a signature of how and where dust forms in the earliest galaxies, not of exotic stellar populations or feedback physics.

Core claim

The observed A_FUV-Mstar relation at z>9 is best reproduced for an intrinsic FUV dust opacity κ_UV(dust) ≈ 10^3–10^4 cm2/g, a value characteristic of low-opacity supernova dust that has undergone reverse-shock processing. This low opacity, combined with mass-conserving porous geometries (leaky screen, leaky mixed, and Poisson clumps), naturally yields very low attenuation even in gas-rich, metal-poor galaxies. The framework reproduces the population of galaxies with extremely low dust attenuation (GELDAs), which dominate samples at z≥9 (84% of the z≥9 objects in the authors' ultra-high-redshift sample), and explains the UV-bright galaxy excess without requiring dust-free ISM or extreme star-

What carries the argument

The central object is a hybrid attenuation framework that combines (i) extinction laws for reverse-shock-processed SN dust (with flat UV slopes and reduced small-grain content), (ii) metallicity- and dust-to-metal-dependent opacity scalings, and (iii) porous radiative-transfer geometries allowing partial UV-photon leakage. The key identity is the effective opacity κ_eff = (1−ωg)κ_ext, where ω is the single-scattering albedo and g the asymmetry parameter, which reduces the impact of forward-peaked scattering. Attenuation is computed from optical depth τ = κ Σ via three mass-conserving transmission kernels: leaky screen T=(1−f_cov)+f_cov exp[−(1−ωg)τ/f_cov], leaky mixed T=(1−f_cov)+f_cov[1−exp

Load-bearing premise

The result rests on the assumption that the A_FUV values and metallicities derived from CIGALE SED fits (many of which are upper limits, e.g., GS-z13-1-LA ≤0.014) accurately measure attenuation independently of the very attenuation physics being tested, and on the adopted size-mass relation that sets the dust surface density.

What would settle it

Measure dust masses and far-infrared luminosities of a sample of z>9 GELDAs with ALMA/NOEMA and JWST/MIRI: if the dust masses are as high as the ISM grain-growth branch predicts (rather than the SN-only branch) while A_FUV remains very low, the low-opacity interpretation fails and geometry or dust removal must dominate; alternatively, if Balmer-decrement or IRX-β-based attenuations are systematically higher than the SED-derived A_FUV values, the result would weaken.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • If the central claim is correct, GELDAs at z≥9 are not dust-free but contain SN-produced dust with large grains and low effective opacity, so gas reservoirs can remain intact without invoking large-scale dust expulsion.
  • The UV-bright galaxy excess does not require extreme star-formation efficiencies, top-heavy IMFs, or AGN contamination; it is a natural consequence of a dust-origin transition at early times.
  • The transition to ISM grain growth at Z_crit≈0.1 Z_sun implies that attenuation should rise sharply above stellar mass log(Mstar/Msun)≈8–9, a prediction testable with larger samples at z>9.
  • A minority of GELDAs may be direct descendants of Population III supernovae, whose dust 'ashes' are observable even if the stars themselves are not.
  • The framework predicts low dust-to-metal ratios and flat UV extinction curves for z>9 galaxies, providing specific targets for MIRI PAH searches and ALMA/NOEMA dust-continuum observations.

Where Pith is reading between the lines

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

  • A testable extension the paper leaves implicit: if independent attenuation tracers (e.g., Balmer decrements, IRX-β, or submillimeter dust continuum) yield systematically higher A_FUV for the same galaxies, the low-opacity SN-dust interpretation would need revision in favor of stronger geometric porosity or dust removal.
  • The same porosity-plus-low-opacity machinery could be applied to local extremely metal-poor dwarfs and to z∼4–8 'blue monsters' to test whether a unified stardust→ISM-growth transition explains the diversity of dust attenuation across cosmic time.
  • The adopted size-mass relation (r=100–500 pc) sets the dust surface density; a robust measurement of the effective radii of z>9 GELDAs (e.g., via JWST morphology or ALMA) would directly shift the preferred κ_UV, so the inferred opacity range is conditional on the assumed compact sizes.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper proposes that the pervasive low UV attenuation and the excess of UV-bright galaxies at z > 9 are explained by a dust-origin transition: dust produced by supernovae and processed by reverse shocks has intrinsically low far-UV opacity (κ_UV ≈ 10^3–10^4 cm²/g), and porous star–dust geometries further reduce effective attenuation while conserving dust mass. The authors construct three attenuation prescriptions (leaky screen, leaky mixed, Poisson clumps) and a gas-based route, calibrate them to scaling relations (Mdust–Mstar, MZR, DTM) and to the observed A_FUV–Mstar relation, and then apply the prescriptions to the intrinsic UV luminosity functions of two models (ND24 and SCSAM). They report that low-opacity SNe dust plus high covering fractions reproduces the observed A_FUV–Mstar relation and the JWST UVLF, whereas standard ISM-like opacities overpredict attenuation. The paper explicitly acknowledges that many A_FUV measurements are upper limits and that no statistically robust object-by-object correlation is inferred.

Significance. If the central claim is established, the paper offers a physically plausible and economical resolution of the UV-bright galaxy excess at z > 9, connecting GELDAs to SN-dominated dust and possibly to Population III descendants, without requiring extreme SFE or dust-free ISM. The framework is genuinely useful: it introduces mass-conserving porous geometries, quantitatively separates geometry from intrinsic opacity, and grounds the low-opacity assumption in independent SN dust models (Hirashita et al. 2008; Maiolino et al. 2004; Nozawa et al. 2007). The UVLF comparison using two independent intrinsic models is a strength, as is the authors' explicit discussion of limitations. However, the central quantitative inference—that κ_UV ≈ 10^3–10^4 cm²/g is 'best reproduced' by the data—relies on visual comparison without a statistical fit, and is partly degenerate with the adopted size–mass relation and Mdust–Mstar normalizations. These issues are load-bearing and need to be addressed before the claim can be accepted at face value.

major comments (3)
  1. [Section 3.2, Fig. 2, Table 2; §4.2] The headline claim that the A_FUV–Mstar relation is 'best reproduced' for κ_UV = 10^3–10^4 cm²/g is not quantitatively supported. The comparison in Fig. 2 is visual, with no goodness-of-fit statistic, no censored likelihood, and no model comparison. Many of the key data points are upper limits (e.g., GS-z13-1-LA A_FUV ≤ 0.014, CEERS5_2 ≤ 0.094, GZ-z13-0 ≤ 0.033; Table 2), and §4.2 states that the authors 'do not attempt to infer a statistically robust object-by-object correlation.' An upper-limit-aware fit (e.g., survival analysis or a censored likelihood) could show that κ_UV = 10^4 cm²/g is also acceptable, weakening the claimed range. The phrase 'best reproduced' should either be replaced by 'consistent with' or supported by a proper statistical analysis that accounts for censoring and scatter.
  2. [Section 3.1 and Appendix F] The A_FUV values used to calibrate the model are derived from CIGALE SED fits adopting a modified-starburst attenuation law with fixed R_V = 3.1 and δ = 0 (Appendix F). This is a screen-like, uniform-geometry attenuation law, which is precisely the kind of geometry the paper argues is incorrect for z>9 galaxies. If the true geometry is porous and produces grey attenuation, CIGALE's derived A_FUV could be systematically biased. The paper should test the sensitivity of the central κ_UV inference to this assumption, for example by re-fitting a subset of the sample with a grey or porosity-motivated attenuation law, or by arguing quantitatively why the CIGALE-derived A_FUV values are robust to this misspecification. Without such a test, the calibration data and the model being tested are not fully independent.
  3. [Eq. (7) and Section 3.2] The inferred κ_UV is degenerate with the adopted size–mass relation and with the normalization of the Mdust–Mstar relations. Since τ ∝ Σ_dust = Mdust/(π R_e²), changing R_e from 100 pc to 500 pc shifts the predicted A_FUV by a factor of 25, and the Witstok et al. (2023) branch intercepts also set the dust mass. The paper qualitatively notes this degeneracy (Sec. 3.2, Appendix C) but does not marginalize over these parameters or provide a confidence interval for κ_UV. The headline range 10^3–10^4 cm²/g is therefore not a robust posterior range; it is conditional on the assumed size–mass relation and dust-mass scalings. The authors should either quantify how much κ_UV would change within plausible ranges of R_e and branch normalizations, or explicitly present the result as conditional on these assumptions.
minor comments (5)
  1. [Abstract / Introduction] Typo: 'cosntraints' in the Introduction should be 'constraints.'
  2. [Fig. 2 caption] GELDAs are described as 'UV-bright systems' in the caption, but GELDA stands for 'Galaxies with an Extremely Low Dust Attenuation.' The two are not interchangeable; UV-brightness is not the defining criterion.
  3. [Table 2] The GELDA classification (Y/N) is based on CIGALE-derived A_FUV, but the threshold A_FUV value used to define GELDAs is not stated in Table 2 or its caption. Please provide the criterion or a reference to the definition in Burgarella et al. (2025).
  4. [Appendix E, Fig. E.2 caption] The caption states 'All models assume a fixed covering fraction of f_cov=0.99 for the ISM dust and f_cov=0.50 for stardust,' which differs from the main text (Sec. 3.2) where f_cov=0.99 is used for all tracks. This inconsistency should be resolved so the reader knows which f_cov is actually applied.
  5. [Eq. (21)] The broken MZR is described as 'designed to reproduce the trends and normalizations inferred from recent JWST measurements.' This means the gas-route attenuation is calibrated to the same observations it is used to explain. Please clarify in the text that the gas-route consistency check is not fully independent of the data, even if the attenuation-route calibration is separate.

Circularity Check

2 steps flagged

κ inference and GELDA 'reproduction' are partly a fit to the same CIGALE A_FUV values; independent UVLF and grain-physics support limit severity.

specific steps
  1. fitted input called prediction [§3.2 (Eqs. 8–10, Fig. 2) and §3.1 (GELDA definition)]
    "Fig. 2, and Fig. E.2 for the gas-route scenario, show that κFUV=10^3 - 10^4 cm2 g−1 matches quite nicely the data. ... GELDAs are defined as objects whose dust attenuation, AFUV, estimated by CIGALE, are very low."

    The central κ range is selected by comparing model tracks to the same CIGALE-derived A_FUV values that define the GELDA class. Because A_FUV = −2.5 log T(κ Σ_dust) is a monotone function of κ, choosing κ to match the low A_FUV systems guarantees that the model 'reproduces' GELDAs. This is a manual parameter fit rather than an independent prediction; no censored likelihood or ΔBIC is computed, and most GELDA A_FUV entries are upper limits (e.g., GS-z13-1-LA ≤0.014), so the claimed 'best reproduced' is not statistically forced by independent evidence.

  2. other [Appendix F (Table F.1, CIGALE parameters) and §3.1–3.2]
    "Attenuation law — dustatt_modified_starburst ... Power-law slope δ 0.0 RV RV 3.1 ... GELDAs are defined as objects whose dust attenuation, AFUV, estimated by CIGALE, are very low."

    The 'observed' A_FUV values used to infer κ are not direct observations but CIGALE outputs produced under a fixed modified-starburst attenuation law. The paper's own thesis is that the emergent attenuation is porosity-driven and greyer; if that is correct, CIGALE's fixed R_V=3.1, δ=0 law may misconvert reddening into A_FUV, biasing the very data used to constrain κ. Thus the inferred 'observational requirement' for low opacity is entangled with the attenuation ansatz used to build the data, weakening the independence of the test.

full rationale

The paper has real independent content: the UVLF comparison uses separate observational datasets, the low κ range is physically grounded in SN-dust reverse-shock models and Inoue et al. (2020) grain-opacity curves, and the Mdust-Mstar calibration comes from external data (Witstok et al. 2023). The main circularity concern is that the headline κ=10^3–10^4 cm2/g is inferred from CIGALE-derived A_FUV values that are simultaneously used to define GELDAs and to evaluate the model. The paper itself acknowledges the measurements are mostly upper limits and that no statistically robust object-by-object correlation is attempted, so 'best reproduced' is a visual fit rather than a formal inference. This does not make the derivation circular by equation identity, but it does mean the central 'prediction' of GELDAs is partly a restatement of the fitting choice. I therefore assign a moderate score of 4 rather than a higher one, because the UVLF application and external dust physics provide independent, non-circular support.

Axiom & Free-Parameter Ledger

10 free parameters · 9 axioms · 0 invented entities

The central claim rests on several empirical calibrations imported from prior work or fitted to the same JWST samples: the Mdust-Mstar branches, size-mass relation, MZR and DTM scalings, plus free geometry/opacity parameters. No new particles, forces, or physical media are introduced; the low-opacity SNe-dust hypothesis is drawn from existing dust formation models.

free parameters (10)
  • kappa_UV(dust) intrinsic FUV opacity = 10^3 cm2/g fiducial; 10^4, 10^5 explored
    Chosen because only low-opacity models reproduce the observed A_FUV-Mstar relation; not independently measured in this paper.
  • Covering fraction f_cov = 0.99 fiducial in Fig. 2; 0.2-1.0 grids; 0.5 for stardust in App. E
    Controls normalization of attenuation; degenerate with kappa and size; constrained visually against the observed data.
  • Mdust-Mstar SNe-branch intercept = a=-3.422, slope b=1.0
    Adopted from Witstok et al. (2023) Fig. 5; sets the dust column for the SNe-only branch and is central to the low-attenuation result.
  • Mdust-Mstar grain-growth-branch intercept = a=-1.296, slope b=1.0
    Adopted from Witstok et al. (2023); sets the upper dust-mass branch used for the ISM/hybrid tracks.
  • Size-mass relation parameters = r_lo=100 pc, r_hi=500 pc, mass turnover logM 7-9
    Prescribed from high-z size measurements; converts dust mass to surface density and directly scales the optical depth.
  • Gas fraction transition = f_g,max=0.98, f_g,min=0.85, turnover logM 8.5-9.0
    Chosen from Burgarella et al. (2025) gas-fraction constraints; affects the gas-route optical depth.
  • Broken MZR normalization and slopes = z9_low=0.08, z9_high=0.20, s_low=0.28, s_high=0.30, turnover 8.5-9.5
    Designed to reproduce JWST MZR measurements; drives the low metallicities that produce low gas opacity in the gas route.
  • DTM broken-power-law parameters = anchors 12+log(O/H)=7.4, 8.0; slopes 1.2 and -0.5; median normalization from Burgarella et al. (2025)
    Empirical interpolation of dust-to-metal ratio; central to the gas-route opacity calculation.
  • Hybrid transition Zcrit = 0.1 Zsun, corresponding to log Mstar 8-9
    Assumed location of the stardust-to-ISM grain-growth transition; supported by local-Universe data but not detected at z>0.
  • FUV scattering parameters = omega=0.35, g=0.60
    Adopted representative values for UV dust scattering; minor effect relative to kappa and f_cov.
axioms (9)
  • domain assumption Single young stellar population dominates the UV light and attenuation
    Sect. 2.1; needed to interpret A_FUV as attenuation of current star formation; older populations could change the dust budget and luminosity weighting.
  • domain assumption Dust surface density uses the same effective radius as stars; stars, gas, and dust share Re(Mstar)
    Sect. 2.3.1, Eqs. 6-7; acknowledged as an effective luminosity-weighted scale; changes the normalization of tau.
  • ad hoc to paper Effective opacity kappa_eff=(1-omega*g)*kappa_ext approximates anisotropic scattering in clumpy media
    Sect. 2.3.1; no full radiative transfer; forward-scattered photons are approximated as not removed and scattered-in photons are neglected.
  • standard math Natta & Panagia leaky screen/mixed and Poisson transmission kernels with mass conservation describe attenuation
    Appendix A/B; standard clumpy-media formalism, assumed applicable to high-z porous geometries.
  • domain assumption At z>9 dust is predominantly core-collapse SNe ejecta processed by reverse shocks, yielding large grains and low opacity
    Sect. 2.2 and App. D; key physical premise, supported by SN dust models but not directly observed in these galaxies.
  • domain assumption ISM grain growth becomes efficient only above Zcrit~0.1 Zsun
    Sect. 2.5; motivated by low-z data and dust models and explicitly noted as not detected at z>0.
  • domain assumption Gas masses from Tacconi/Kennicutt-Schmidt scaling and main-sequence SFR hold at z>9
    Sect. 2.4 and Table G.1; needed to compute the gas-route optical depth.
  • ad hoc to paper Broken power-law MZR and DTM relations represent the z~9 galaxy population
    Sect. 2.4, Eqs. 21-22; calibrated to JWST observations and then used to predict attenuation, so the gas route is partially circular.
  • ad hoc to paper Intrinsic UVLF models (ND24, SCSAM) plus modest SFE(z) evolution approximate the dust-free galaxy population
    Sect. 3.3; the SFE scaling is required to match the observed UVLF normalization, and the paper acknowledges the comparison does not uniquely validate these models.

pith-pipeline@v1.3.0-alltime-deepseek · 36785 in / 18934 out tokens · 193028 ms · 2026-08-02T14:25:22.511892+00:00 · methodology

0 comments
read the original abstract

Recent JWST observations suggest that galaxies at z > 9 may be dominated by low-opacity SNe-produced dust before efficient ISM grain growth is established. This transition in dust origin and opacity could explain both the prevalence of galaxies with extremely low dust attenuation and the excess of UV-bright galaxies relative to most pre-JWST predictions. We investigate whether this transition, combined with evolving star-formation efficiency, can reproduce these observed properties. We develop a physically motivated attenuation framework combining (i) extinction laws for reverse-shock-processed SNe dust, (ii) metallicity- and dust-to-metal-dependent opacity scalings, and (iii) porous radiative-transfer geometries allowing partial UV-photon leakage. Unlike outflow-driven scenarios requiring large-scale gas evacuation, our approach preserves gas reservoirs while reducing effective UV opacity through dust composition and geometry. We introduce extinction-based, gas-based, and hybrid attenuation prescriptions linking SNe-dominated and ISM grain-growth dust regimes. We find that the observed A_FUV-M_star relation at z > 9 is best reproduced for an intrinsic FUV dust opacity kappa_UV(dust)=10^3 - 10^4 cm2/g, characteristic of low-opacity SNe dust, naturally producing very low attenuation even in gas-rich galaxies. This regime reproduces galaxies with extremely low dust attenuation (GELDAs), which dominate observed samples at z > 9. Applied to intrinsic UV luminosity function models, our SNe-dominated and hybrid prescriptions mainly suppress the brightest galaxies, bringing predictions into agreement with JWST measurements without requiring extreme star-formation efficiencies or dust-free interstellar media. Our results suggest that the UV-bright galaxy excess at z > 9 reflects a transition in dust origin and opacity during the earliest phases of galaxy evolution.

Figures

Figures reproduced from arXiv: 2605.09829 by A.K. Inoue, C. Aurin, C. Kobayashi, D. Burgarella, G. P. Nikopoulos, J.-C. Bouret, M. Dickinson, P. Dayal, R. S. Somerville, T. Dewachter, T.T. Takeuchi, V. Buat.

Figure 1
Figure 1. Figure 1: Schematic illustration of the three dust–star geometries [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Predicted FUV attenuation, AFUV, as a function of stellar mass. This figure corresponds to a far-UV mass absorption coef￾ficient, κ0.15 = 103 cm2 g −1 . Rows correspond to different porous geometries: leaky screen and leaky mixed configurations (top), Poisson clump geometry (middle), and the gas–route attenuation model (bottom). For the leaky and Poisson cases, ISM-, SN-, and hybrid-type dust tracks are sh… view at source ↗
Figure 3
Figure 3. Figure 3: Observed rest-frame UV luminosity functions from the [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Same as Fig.3 but for dust-free UVLF predictions from [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: AFUV versus log10(Mstar/M⊙) for the sample presented in [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 5
Figure 5. Figure 5: AFUV versus log10(Mstar/M⊙) for the sample presented in [PITH_FULL_IMAGE:figures/full_fig_p013_5.png] view at source ↗

Review history (3 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. The first dusty galaxies across $6 \lesssim z \lesssim 14$: Blue monsters, red monsters, and the bimodality of dust content in early galaxies

    astro-ph.GA 2026-07 conditional novelty 6.0

    Mechanical venting of supernova dust by cloud blowout plus disk breakout can produce blue (βUV≲−2.4) and red (βUV≳−1.5) UV slopes in z≈6–14 galaxies.

  2. The Thesan-Zoom project: bursty star formation is incompatible with prolonged dust survival

    astro-ph.GA 2026-07 conditional novelty 6.0

    In Thesan-Zoom, bursty star formation prevents prolonged survival of large dust reservoirs (M_dust/M_star ≳ 10^{-3}), so it can explain z≳10 UV-bright galaxies only if it settles by z∼8.

Reference graph

Works this paper leans on

151 extracted references · 4 linked inside Pith · cited by 2 Pith papers

  1. [1]

    J., Conselice, C

    Adams, N. J., Conselice, C. J., Ferreira, L., et al.\ 2023, , 518, 3, 4755. doi:10.1093/mnras/stac3347

    doi:10.1093/mnras/stac3347 2023

  2. [2]

    J., Conselice, C

    Adams, N. J., Conselice, C. J., Austin, D., et al.\ 2024, , 965, 2, 169. doi:10.3847/1538-4357/ad2a7b

    doi:10.3847/1538-4357/ad2a7b 2024

  3. [3]

    L., et al.\ 2023, , 951, L22

    Arrabal Haro, P., Dickinson, M., Finkelstein, S. L., et al.\ 2023, , 951, L22. doi:10.3847/2041-8213/acdd54

    doi:10.3847/2041-8213/acdd54 2023

  4. [4]

    L., et al.\ 2023, , 622, 7984, 707

    Arrabal Haro, P., Dickinson, M., Finkelstein, S. L., et al.\ 2023, , 622, 7984, 707. doi:10.1038/s41586-023-06521-7

    doi:10.1038/s41586-023-06521-7 2023

  5. [5]

    S., Takeuchi, T

    Asano, R. S., Takeuchi, T. T., Hirashita, H., & Inoue, A. K.\ 2013, EPS, 65, 213

    2013

  6. [6]

    S., Takeuchi T

    Asano R. S., Takeuchi T. T., Hirashita H., Nozawa T., 2014, MNRAS, 440, 134. doi:10.1093/mnras/stu208

    doi:10.1093/mnras/stu208 2014

  7. [7]

    J., et al.\ 2023, , 519, 1, 1201

    Atek, H., Shuntov, M., Furtak, L. J., et al.\ 2023, , 519, 1, 1201. doi:10.1093/mnras/stac3144

    doi:10.1093/mnras/stac3144 2023

  8. [8]

    doi:10.1051/0004-6361/201834156

    Boquien, M., Burgarella, D., Roehlly, Y., et al.\ 2019, , 622, A103. doi:10.1051/0004-6361/201834156

    doi:10.1051/0004-6361/201834156 2019

  9. [9]

    doi:10.1093/mnras/stad1014

    Bouwens, R., Illingworth, G., Oesch, P., et al.\ 2023, , 523, 1, 1009. doi:10.1093/mnras/stad1014

    doi:10.1093/mnras/stad1014 2023

  10. [10]

    J., Illingworth, G

    Bouwens, R. J., Illingworth, G. D., Oesch, P. A., et al.\ 2023, , 954, 128. doi:10.3847/1538-4357/ace0d8

    doi:10.3847/1538-4357/ace0d8 2023

  11. [11]

    doi:10.1038/s41550-023-01950-2

    Boylan-Kolchin, M.\ 2023, Nature Astronomy, 7, 731. doi:10.1038/s41550-023-01950-2

    doi:10.1038/s41550-023-01950-2 2023

  12. [12]

    & Larson, R

    Bromm, V. & Larson, R. B.\ 2004, , 42, 79

    2004

  13. [13]

    doi:10.1088/0034-4885/76/11/112901

    Bromm, V.\ 2013, Reports on Progress in Physics, 76, 112901. doi:10.1088/0034-4885/76/11/112901

    doi:10.1088/0034-4885/76/11/112901 2013

  14. [14]

    doi:10.1146/annurev-astro-081710-102608

    Bromm, V., & Yoshida, N.\ 2011, , 49, 373. doi:10.1146/annurev-astro-081710-102608

    doi:10.1146/annurev-astro-081710-102608 2011

  15. [15]

    doi:10.1051/0004-6361/201219405

    Buat, V., Noll, S., Burgarella, D., et al.\ 2012, , 545, A141. doi:10.1051/0004-6361/201219405

    doi:10.1051/0004-6361/201219405 2012

  16. [16]

    doi:10.1051/0004-6361/202554231

    Burgarella, D., Buat, V., Theul \'e , P., et al.\ 2025, , 699, A336. doi:10.1051/0004-6361/202554231

    doi:10.1051/0004-6361/202554231 2025

  17. [17]

    C., et al.\ 2000, , 533, 682

    Calzetti, D., Armus, L., Bohlin, R. C., et al.\ 2000, , 533, 682. doi:10.1086/308692

    doi:10.1086/308692 2000

  18. [18]

    doi:10.3847/2041-8213/accea5

    Castellano, M., Fontana, A., Treu, T., et al.\ 2023, , 948, 2, L14. doi:10.3847/2041-8213/accea5

    doi:10.3847/2041-8213/accea5 2023

  19. [19]

    doi:10.3847/1538-4357/ad5f88

    Castellano, M., Napolitano, L., Fontana, A., et al.\ 2024, , 972, 2, 143. doi:10.3847/1538-4357/ad5f88

    doi:10.3847/1538-4357/ad5f88 2024

  20. [20]

    & Fall, S

    Charlot, S. & Fall, S. M.\ 2000, , 539, 718. doi:10.1086/309250

    doi:10.1086/309250 2000

  21. [21]

    Chemerynska, I., Atek, H., Dayal, P., et al.\ 2024, , 976, L15

    2024

  22. [22]

    Cherchneff, I., & Dwek, E.\ 2009, , 703, 642

    2009

  23. [23]

    doi:10.1093/mnras/stt318

    Chevallard, J., Charlot, S., Wandelt, B., & Wild, V.\ 2013, , 432, 2061. doi:10.1093/mnras/stt318

    doi:10.1093/mnras/stt318 2013

  24. [24]

    R., Kere s , D., Sandstrom, K

    Choban, C. R., Kere s , D., Sandstrom, K. M., et al.\ 2024, , 529, 2356. doi:10.1093/mnras/stae716

    doi:10.1093/mnras/stae716 2024

  25. [25]

    R., Hutter, A., Dayal, P., et al.\ 2024, , 686, A138

    Cueto, E. R., Hutter, A., Dayal, P., et al.\ 2024, , 686, A138. doi:10.1051/0004-6361/202349017

    doi:10.1051/0004-6361/202349017 2024

  26. [26]

    J., Donnan, C

    Cullen, F., McLure, R. J., Donnan, C. T., et al.\ 2023, , 520, 14. doi:10.1093/mnras/stad131

    doi:10.1093/mnras/stad131 2023

  27. [27]

    C., Scholte, D., et al.\ 2025, , 540, 3, 2176

    Cullen, F., Carnall, A. C., Scholte, D., et al.\ 2025, , 540, 3, 2176. doi:10.1093/mnras/staf838

    doi:10.1093/mnras/staf838 2025

  28. [28]

    doi:10.1093/mnras/stac2737

    Curti, M., D'Eugenio, F., Carniani, S., et al.\ 2023, , 518, 1, 425. doi:10.1093/mnras/stac2737

    doi:10.1093/mnras/stac2737 2023

  29. [29]

    doi:10.1093/mnras/stac537

    Dayal, P., Ferrara, A., Sommovigo, L., et al.\ 2022, , 512, 989. doi:10.1093/mnras/stac537

    doi:10.1093/mnras/stac537 2022

  30. [30]

    R., Ferrara, A., et al.\ 2024, , 529, 4221

    Dayal, P., Choudhury, T. R., Ferrara, A., et al.\ 2024, , 529, 4221. doi:10.1093/mnras/stae318

    doi:10.1093/mnras/stae318 2024

  31. [31]

    J., Swinyard, B

    De Looze, I., Barlow, M. J., Swinyard, B. M., et al.\ 2017, , 465, 3309

    2017

  32. [32]

    De Vis, P., Jones, A., Viaene, S., et al.\ 2019, , 623, A5

    2019

  33. [33]

    doi:10.1093/mnras/stac3702

    Di Cesare, C., Graziani, L., Schneider, R., et al.\ 2023, , 519, 4632. doi:10.1093/mnras/stac3702

    doi:10.1093/mnras/stac3702 2023

  34. [34]

    T., McLeod, D

    Donnan, C. T., McLeod, D. J., McLure, R. J., et al.\ 2023, , 518, 6011. doi:10.1093/mnras/stac3491

    doi:10.1093/mnras/stac3491 2023

  35. [35]

    T., McLure, R

    Donnan, C. T., McLure, R. J., Dunlop, J. S., et al.\ 2024, , 533, 3, 3222. doi:10.1093/mnras/stae2037

    doi:10.1093/mnras/stae2037 2024

  36. [36]

    T.\ 2003, , 41, 241

    Draine, B. T.\ 2003, , 41, 241

    2003

  37. [37]

    doi:10.1088/0004-637X/727/2/63

    Dwek, E., & Cherchneff, I.\ 2011, , 727, 63. doi:10.1088/0004-637X/727/2/63

    doi:10.1088/0004-637x/727/2/63 2011

  38. [38]

    Feldmann, R.\ 2015, , 449, 3274

    2015

  39. [39]

    Ferrara, A., Aiello, S., Ferrini, F., & Barsella, B.\ 1990, , 240, 259

    1990

  40. [40]

    doi:10.1093/mnras/stac620

    Ferrara, A.\ 2022, , 512, 1598. doi:10.1093/mnras/stac620

    doi:10.1093/mnras/stac620 2022

  41. [41]

    doi:10.1093/mnras/stad1095

    Ferrara, A., Pallottini, A., & Dayal, P.\ 2023, , 522, 3, 3986. doi:10.1093/mnras/stad1095

    doi:10.1093/mnras/stad1095 2023

  42. [42]

    Ferrara, A., & Carniani, S.\ 2025, , 694, A215

    2025

  43. [43]

    L., Bagley, M

    Finkelstein, S. L., Bagley, M. B., Arrabal Haro, P., et al.\ 2022, , 940, L55. doi:10.3847/2041-8213/acad00

    doi:10.3847/2041-8213/acad00 2022

  44. [44]

    L., Bagley, M

    Finkelstein, S. L., Bagley, M. B., Arrabal Haro, P., et al.\ 2023, , 946, L13. doi:10.3847/2041-8213/acb66d

    doi:10.3847/2041-8213/acb66d 2023

  45. [45]

    L., Bagley, M

    Finkelstein, S. L., Bagley, M. B., Arrabal Haro, P., et al.\ 2025, , 983, 1, L4. doi:10.3847/2041-8213/adbbd3

    doi:10.3847/2041-8213/adbbd3 2025

  46. [46]

    doi:10.3847/1538-4357/ad235c

    Fujimoto, S., Ouchi, M., Nakajima, K., et al.\ 2024, , 964, 2, 146. doi:10.3847/1538-4357/ad235c

    doi:10.3847/1538-4357/ad235c 2024

  47. [47]

    C., & Hjorth, J.\ 2011, , 19, 43

    Gall, C., Andersen, A. C., & Hjorth, J.\ 2011, , 19, 43

    2011

  48. [48]

    & Hjorth, J.\ 2018, , 868, 1, 62

    Gall, C. & Hjorth, J.\ 2018, , 868, 1, 62. doi:10.3847/1538-4357/aae520

    doi:10.3847/1538-4357/aae520 2018

  49. [49]

    Gallerani, S., Maiolino, R., Juarez, Y., et al.\ 2010, , 523, A85

    2010

  50. [50]

    Galliano, F., et al.\ 2022, , 30, 2

    2022

  51. [51]

    doi:10.48550/arXiv.2202.01868

    Galliano, F.\ 2022, Habilitation Thesis, 1. doi:10.48550/arXiv.2202.01868

    Pith/arXiv arXiv doi:10.48550/arxiv.2202.01868 2022

  52. [52]

    D., Clayton, G

    Gordon, K. D., Clayton, G. C., Misselt, K. A., et al.\ 2003, , 594, 279. doi:10.1086/376774

    doi:10.1086/376774 2003

  53. [53]

    doi:10.1093/mnras/staa796

    Graziani, L., Schneider, R., Ginolfi, M., et al.\ 2020, , 494, 1071. doi:10.1093/mnras/staa796

    doi:10.1093/mnras/staa796 2020

  54. [54]

    H., Springel, V., White, S

    Greif, T. H., Springel, V., White, S. D. M., et al.\ 2011, , 737, 75. doi:10.1088/0004-637X/737/2/75

    doi:10.1088/0004-637x/737/2/75 2011

  55. [55]

    K., Mawatari, K., et al.\ 2022, , 929, 1, 1

    Harikane, Y., Inoue, A. K., Mawatari, K., et al.\ 2022, , 929, 1, 1. doi:10.3847/1538-4357/ac53a9

    doi:10.3847/1538-4357/ac53a9 2022

  56. [56]

    doi:10.3847/1538-4365/acaaa9

    Harikane, Y., Ouchi, M., Oguri, M., et al.\ 2023, , 265, 5. doi:10.3847/1538-4365/acaaa9

    doi:10.3847/1538-4365/acaaa9 2023

  57. [57]

    2024, , 942, 64

    Harikane, Y., Ouchi, M., Nakajima, K., et al. 2024, , 942, 64

    2024

  58. [58]

    doi:10.3847/1538-4357/ad0b7e

    Harikane, Y., Nakajima, K., Ouchi, M., et al.\ 2024, , 960, 1, 56. doi:10.3847/1538-4357/ad0b7e

    doi:10.3847/1538-4357/ad0b7e 2024

  59. [59]

    E., Gim \'e nez-Arteaga, C., Fujimoto, S., et al.\ 2023, , 944, 2, L30

    Heintz, K. E., Gim \'e nez-Arteaga, C., Fujimoto, S., et al.\ 2023, , 944, 2, L30. doi:10.3847/2041-8213/acb2cf

    doi:10.3847/2041-8213/acb2cf 2023

  60. [60]

    E., Fudamoto, Y., Faisst, A

    Heintz, K. E., Fudamoto, Y., Faisst, A. L., et al.\ 2024, , 964, L15. doi:10.3847/2041-8213/ad2b25

    doi:10.3847/2041-8213/ad2b25 2024

  61. [61]

    Hirano, S., et al.\ 2015, , 448, 568

    2015

  62. [62]

    T., & Kozasa, T.\ 2008, , 384, 1725

    Hirashita, H., Nozawa, T., Takeuchi, T. T., & Kozasa, T.\ 2008, , 384, 1725. doi:10.1111/j.1365-2966.2007.12828.x

    arXiv 2008

  63. [63]

    K., et al.\ 2022, , 938, 2, L10

    Inayoshi, K., Harikane, Y., Inoue, A. K., et al.\ 2022, , 938, 2, L10. doi:10.3847/2041-8213/ac9310

    doi:10.3847/2041-8213/ac9310 2022

  64. [64]

    K.\ 2005, , 359, 171

    Inoue, A. K.\ 2005, , 359, 171. doi:10.1111/j.1365-2966.2005.08940.x

    arXiv 2005

  65. [65]

    K.\ 2011, Earth, Planets and Space, 63, 1027

    Inoue, A. K.\ 2011, Earth, Planets and Space, 63, 1027. doi:10.5047/eps.2011.02.013

    doi:10.5047/eps.2011.02.013 2011

  66. [66]

    K., Hashimoto, T., Chihara, H., et al.\ 2020, , 495, 1577

    Inoue, A. K., Hashimoto, T., Chihara, H., et al.\ 2020, , 495, 1577. doi:10.1093/mnras/staa1203

    doi:10.1093/mnras/staa1203 2020

  67. [67]

    G., Speagle, J

    Iyer, K. G., Speagle, J. S., Caplar, N., et al.\ 2024, , 961, 1, 53. doi:10.3847/1538-4357/acff64

    doi:10.3847/1538-4357/acff64 2024

  68. [68]

    L., & Bromm, V.\ 2019, , 488, 2, 2202

    Jaacks, J., Finkelstein, S. L., & Bromm, V.\ 2019, , 488, 2, 2202. doi:10.1093/mnras/stz1529

    doi:10.1093/mnras/stz1529 2019

  69. [69]

    doi:10.1051/0004-6361/201323199

    Kataoka, A., Okuzumi, S., Tanaka, H., et al.\ 2014, , 568, A42. doi:10.1051/0004-6361/201323199

    doi:10.1051/0004-6361/201323199 2014

  70. [70]

    Z., Redigolo, D., & Volansky, T.\ 2025, , 2025, 10, 047

    Katz, O. Z., Redigolo, D., & Volansky, T.\ 2025, , 2025, 10, 047. doi:10.1088/1475-7516/2025/10/047

    doi:10.1088/1475-7516/2025/10/047 2025

  71. [71]

    Kawamata, R., Ishigaki, M., Shimasaku, K., et al.\ 2018, , 855, 4

    2018

  72. [72]

    J., Priestley, F

    Kirchschlager, F., Barlow, M. J., Priestley, F. D., & Nozawa, T.\ 2019, , 487, 195. doi:10.1093/mnras/stz1203

    doi:10.1093/mnras/stz1203 2019

  73. [73]

    D., Barlow, M

    Kirchschlager, F., Schmidt, F. D., Barlow, M. J., et al.\ 2019, , 489, 4, 4465. doi:10.1093/mnras/stz2399

    doi:10.1093/mnras/stz2399 2019

  74. [74]

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

    doi:10.1146/annurev-astro-071221-053453 2023

  75. [75]

    D., Papovich, C., Trump, J

    Kocevski, D. D., Papovich, C., Trump, J. R., et al.\ 2023, , 946, L14. doi:10.3847/2041-8213/acb8b5

    doi:10.3847/2041-8213/acb8b5 2023

  76. [76]

    doi:10.3847/1538-4357/acdffd

    Langeroodi, D., Hjorth, J., Gall, C., & Zafar, T.\ 2023, , 952, 156. doi:10.3847/1538-4357/acdffd

    doi:10.3847/1538-4357/acdffd 2023

  77. [77]

    doi:10.48550/arXiv.2410.14671

    Langeroodi, D., Hjorth, J., Ferrara, A., et al.\ 2024, arXiv:2410.14671. doi:10.48550/arXiv.2410.14671

    Pith/arXiv arXiv doi:10.48550/arxiv.2410.14671 2024

  78. [78]

    Leung, G. C. K., Bagley, M. B., Finkelstein, S. L., et al.\ 2023, , 954, 2, L46. doi:10.3847/2041-8213/acf365

    doi:10.3847/2041-8213/acf365 2023

  79. [79]

    Lewis, J. S. W., Ocvirk, P., Dubois, Y., et al.\ 2023, , 519, 5987. doi:10.1093/mnras/stad081

    doi:10.1093/mnras/stad081 2023

  80. [80]

    doi:10.1093/mnras/stz2684

    Li, Q., Narayanan, D., & Dav \'e , R.\ 2019, , 490, 1425. doi:10.1093/mnras/stz2684

    doi:10.1093/mnras/stz2684 2019

Showing first 80 references.