{"id":"d6ebdb84-557f-4550-b7ee-4f44aa3d6635","arxiv_id":"2605.09829","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":10,"one_line_summary":"At z>9, the UV-bright galaxy excess can be explained by supernova-produced dust with low FUV opacity (κ≈10^3–10^4 cm^2/g) plus porous geometry and modest star-formation efficiency evolution.","lead":"This paper models dust attenuation in the earliest galaxies and argues that the surprisingly bright ultraviolet galaxies seen by JWST are explained by low-opacity dust freshly made by supernovae, not by dust-free gas or extreme star formation. The framework reproduces the observed attenuation-mass relation and leaves the bright end of the UV luminosity function largely intact, giving a physical route to the z>9 UV-bright excess.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central κ_UV inference depends on CIGALE-derived A_FUV values that are mostly upper limits and assume a modified-starburst law; no censored statistical fit is performed, so 'best reproduced' is not established.","rationale":"I read the paper as proposing a physically motivated forward model: SNe-only dust + porous geometry + modest SFE evolution explains low A_FUV and the UVLF bright end. The kernels are derived correctly from Natta & Panagia (1984), the mass-conservation limits are sensible, and the Appendix C grids do show that high ISM-like opacities fail at the quoted fcov values. The comparison with Inoue et al. (2020) large-grain opacities is an independent cross-check. No internal mathematical error was found.\n\nThe single load-bearing concern is that the central number κ_UV is calibrated against SED-fit attenuations that are not independent of the attenuation physics under test and that are largely upper limits. The paper explicitly does not perform a censored statistical analysis, yet the abstract/conclusion state a 'best reproduced' range. This is a strength-of-claim problem rather than a soundness problem. It is addressable by a survival-analysis fit that marginalizes over the size/dust-mass normalizations.\n\nThe reader's weakest_assumption pointed to the same CIGALE-attenuation-law and upper-limit issue, so I agree. Since the concern is addressable and the physics may well survive the check, the CONDITIONAL verdict should stand.","tokens_in":37167,"tokens_out":11200,"duration_ms":129243,"concrete_test":"Run a censored likelihood / nested-sampling fit of the leaky-screen and leaky-mixed models to Table 2 plus the CEERS sample, treating A_FUV values with '≤' as upper limits (survival analysis) and marginalizing over the normalization of the size-mass relation (Eq. 7) and the Mdust-Mstar offsets within their quoted scatter. Report the posterior on log10 κ_UV and the posterior probability that κ_UV ≥ 10^4 cm^2/g. If this probability is non-negligible, or if the best-fit κ_UV shifts by more than 0.3 dex when upper limits are handled correctly, the headline claim is not supported by the data.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim ('best reproduced for κ_UV=10^3–10^4') relies on a visual comparison of model tracks to A_FUV(Mstar) values from CIGALE SED fits (Table 2). Many entries are upper limits (GS-z13-1-LA ≤0.014, CEERS5_2 ≤0.094, GZ-z13-0 ≤0.033), and §4.2 explicitly disclaims a statistically robust object-by-object correlation. No censored likelihood, ΔBIC, or goodness-of-fit is computed; the phrase 'best reproduced' is therefore not quantitatively supported. If a proper upper-limit fit includes κ_UV=10^4 with acceptable likelihood, the headline range is not actually established. More fundamentally, the observed A_FUV are produced by CIGALE using the modified-starburst attenuation law with fixed R_V=3.1, δ=0 (Appendix F) and the GELDA classifications are derived from those same A_FUV values. The paper's own thesis is that the emergent attenuation is greyer and porosity-driven; if that is true, CIGALE's fixed law may misconvert reddening into A_FUV, biasing the very data used to calibrate κ_UV. The model also depends on the adopted size-mass relation (Eq. 7, r=100–500 pc) and on Mdust-Mstar branches from Witstok et al. (2023); κ_UV is degenerate with the normalizations of both, which are not marginalized in the comparison.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","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.","tokens_in":37637,"tokens_out":3402,"duration_ms":39456,"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":[{"comment":"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.","section":"Section 3.2, Fig. 2, Table 2; §4.2"},{"comment":"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.","section":"Section 3.1 and Appendix F"},{"comment":"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.","section":"Eq. (7) and Section 3.2"}],"minor_comments":[{"comment":"Typo: 'cosntraints' in the Introduction should be 'constraints.'","section":"Abstract / Introduction"},{"comment":"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.","section":"Fig. 2 caption"},{"comment":"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).","section":"Table 2"},{"comment":"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.","section":"Appendix E, Fig. E.2 caption"},{"comment":"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.","section":"Eq. (21)"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses an important question and offers a physically motivated framework that could be influential. The main issue is that the central quantitative claim—the specific κ_UV range—is not yet established rigorously: the comparison to A_FUV is visual, many data points are upper limits, and the inference is degenerate with size and dust-mass scalings. The authors are honest about some of these limitations, but the abstract and conclusions state the result more strongly than the analysis supports. A major revision that adds a censored statistical analysis (or weakens the claim appropriately), tests sensitivity to the CIGALE attenuation-law prior, and quantifies degeneracies would make the paper suitable for publication. I do not see a fatal flaw in the physical reasoning, and the UVLF comparison provides independent support for low attenuation, so rejection is not warranted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper you're asking about is Burgarella et al., a follow-up to their 2025 GELDA work. The genuinely new piece is the attenuation framework: leaky screen, leaky mixed, and Poisson clump geometries, run through SNe-only, ISM-growth, and hybrid dust tracks, plus a gas-route variant that ties opacity to metallicity and dust-to-metal ratio. Applied to two intrinsic UVLF models (ND24 and SCSAM), it gives a mechanism for the z>9 UV-bright excess that doesn't require dust-free ISM or extreme SFE. That part is solid, well-motivated, and the paper is honest about what it can't discriminate: the UVLF comparison is explicitly non-unique, and pure SNe versus hybrid produce nearly identical bright-end suppression.\n\nThe central numerical claim—that the A_FUV–Mstar relation at z>9 is best reproduced for κ_UV ≈ 10^3–10^4 cm^2/g—is the soft spot. The comparison is visual. Many of the observed A_FUV values in Table 2 are upper limits (GS-z13-1-LA ≤0.014, GZ-z13-0 ≤0.033), and there is no censored likelihood or goodness-of-fit. That matters, because a proper upper-limit fit could in principle accommodate κ_UV = 10^4, which would blur the sharp inference. There's also a genuine circularity concern: the A_FUV values come from CIGALE fits that assume a modified-starburst attenuation law with fixed R_V=3.1 and δ=0. If the true attenuation is greyer and porosity-driven, CIGALE's fixed law could bias the very data used to calibrate the model. The size–mass relation (Eq. 7) is another load-bearing input; changing radii shifts the normalization and thus the preferred κ. The gas route inherits the same issues via MZR/DTM scalings fitted to JWST data. None of this is fatal, and the paper actually acknowledges several of these limitations themselves, but the headline “best reproduced” is weaker than the analysis supports.\n\nWhat earns credit: the model is physically motivated, the assumptions are stated, the degeneracies are explored in appendices, and there is independent grounding in SN dust models and the Witstok et al. scaling relations. No internal contradiction or mathematical error surfaced. This is a genuine extension, not a restatement.\n\nWho gets value: anyone working on high-redshift dust attenuation, GELDAs, or JWST UVLF interpretation. I would send it to peer review—the framework deserves scrutiny and the claim needs the quantitative tightening a referee would demand. My own verdict is conditional, not reject.\n\nYes, I'd cite it if I work in this area; it's a useful reference for the stardust scenario and the attenuation prescriptions. Reading group maybe—good for a discussion of how far visual model-data comparison can be pushed.","headline":"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.","tokens_in":38161,"tokens_out":1030,"would_cite":true,"duration_ms":14236,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"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.","keywords":["high-redshift galaxies","dust attenuation","supernova dust","GELDAs","UV luminosity function","JWST","porous geometry","dust opacity"],"falsifier":"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.","tokens_in":37027,"feed_emoji":"🌌","tokens_out":2059,"duration_ms":25311,"temperature":0.7,"pith_summary":"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.","feed_headline":"Supernova dust explains the overbright early galaxies","feed_subtitle":"At z>9, reverse-shock-processed SN dust has low UV opacity, so gas-rich galaxies stay UV-bright without extreme star formation.","key_machinery":"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","core_discovery":"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-","pith_inferences":["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."],"forward_implications":["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."],"fun_headline_variants":["Supernova dust's low opacity powers z>9 UV-bright galaxies","Early galaxy brightness traced to supernova dust, not extreme stars","Low-opacity supernova dust explains z>9 UV-bright galaxy excess","Dust-origin shift behind the overbright early galaxies at z>9"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"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.","fun_headline_variants_meta":{"raw":{"variants":["Supernova dust's low opacity powers z>9 UV-bright galaxies","Early galaxy brightness traced to supernova dust, not extreme stars","Low-opacity supernova dust explains z>9 UV-bright galaxy excess","Dust-origin shift behind the overbright early galaxies at z>9"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000553,"raw_usage":{"total_tokens":2548,"prompt_tokens":892,"completion_tokens":1656,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":636,"completion_tokens_details":{"reasoning_tokens":1577}},"tokens_in":636,"tokens_out":1656,"duration_ms":11674,"temperature":1.0,"reasoning_tokens":1577,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T14:25:22.511892+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"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.","supporting_citations":[],"review_version":3}