{"id":"9faf7462-5b85-4736-ab76-cd6cd04400e2","arxiv_id":"2608.08015","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"The Big Three Dragons galaxy pair at z=7.15 has hot dust (T_d ≈ 78 K) and is almost fully obscured (f_obs ≈ 0.94), implying that UV-bright galaxies can host major hidden starbursts in the early universe.","lead":"New ALMA observations in Bands 4 and 9 reveal that the Big Three Dragons, a pair of merging galaxies at z = 7.15, host very hot dust at roughly 78 kelvin, making it one of the most infrared-luminous systems known so early in the universe. The result matters because it suggests that a large fraction of star formation in bright early galaxies may be hidden by dust, which would change estimates of how fast the first galaxies built stars.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Hot-dust claim hinges on a free-β MBB fit: with β fixed to 2.0 the median temperature falls below the paper's own 60 K 'hot' threshold, and the new Band 9 fluxes are only ~2–3σ.","rationale":"The reader's weakest assumption is the optically thin MBB model, and that is a genuine limitation, especially for the Eastern clump where the paper itself flags possible optically thick dust (§5.2.3). My stress-test point is related but distinct: even under the optically thin assumption used in the fiducial analysis, the 'hot dust' label is not robust because the inferred temperature is strongly degenerate with β_IR, which is poorly constrained at only ~3.5σ on the Rayleigh-Jeans tail. The paper's own Appendix D fixed-β fit gives a median of 57 K, below the 60 K threshold, showing that the headline temperature is conditional on a fitted emissivity index that drifts below the standard value. The marginal significance of the Band 9 integrated fluxes further weakens the anchor. This does not invalidate the high IR luminosity or high obscured fraction, which survive in the β_IR = 2.0 fit (log L_IR ≈ 12.05, f_obs ≈ 0.89), but it does undercut the title-level and abstract-level claim of hot dust as a confirmed property. The paper is honest about the low S/N and provides the alternative fit, so the appropriate remedy is a re-analysis and a more cautious wording, not rejection. The concrete test is cheap, decisive, and can be run with the existing MCMC code.","tokens_in":27211,"tokens_out":6417,"duration_ms":79132,"concrete_test":"Re-run the global MBB fit in two controlled ways: (1) fix β_IR = 2.0 and compute the posterior probability P(T_d ≥ 60 K); (2) omit the Band 9 flux entirely and refit Bands 4, 6, 7, 8 plus the Herschel upper limits, reporting the resulting T_d posterior. If P(T_d ≥ 60 K) is below ~0.68 in either case, the abstract and conclusions should be reworded to state that hot dust is only marginally favored rather than confirmed; if the temperature stays above 60 K in both tests, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the Big Three Dragons hosts hot dust (T_d ≈ 78 K) is less secure than the abstract implies even within the optically thin MBB framework. The global Band 9 aperture flux is 1162 ± 452 μJy (Table 1), only ~2.6σ after the 20% calibration term is added, and the component Band 9 fluxes are 381 ± 191 μJy and 532 ± 191 μJy (Table 2), i.e. ~2.0σ and ~2.8σ. Because Band 9 does not fully sample the SED peak, separating T_d from β_IR relies on the Rayleigh-Jeans slope and on the prior. The free-β fit returns β_IR = 1.47 ± 0.46, lower than the prior mean of 1.8, and T_d = 78 K. Yet the paper's own Appendix D fit with the standard β_IR = 2.0 yields T_d = 57 ± 15 K, a median below the adopted 'hot' threshold of 60 K. The headline temperature is therefore not a parameter-free measurement; it is conditioned on an unconstrained emissivity index as well as on optically thin emission. If Band 9 flux is biased upward by noise or calibration, or if β_IR is closer to 2, the 'hot dust' conclusion weakens or disappears, although the system would remain IR-luminous and heavily obscured. The optically thick possibility discussed in §5.2.3 would push T_d upward, but that is an additional model assumption, not a confirmation of the fiducial value.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents new ALMA Band 4 and Band 9 continuum observations of the z=7.15 Lyman-break galaxy system 'Big Three Dragons', combines them with archival Band 3, 6, 7, and 8 data, and fits single-temperature optically thin modified blackbodies to the global and resolved East/West SEDs. From the fiducial free-β fit the authors infer a global dust temperature of T_d = 78^{+35}_{-23} K, a dust mass log(M_d/M_sun) = 6.85^{+0.33}_{-0.27}, and an infrared luminosity log(L_IR/L_sun) = 12.32^{+0.43}_{-0.41}, which implies a nearly complete obscured fraction f_obs = 0.94. The paper additionally places the two resolved components on the IRX-β_UV plane, discusses the T_d-Sigma_SFR relation, and makes a Monte-Carlo estimate of the contribution of such highly obscured UV-bright galaxies to the obscured star formation rate density at z~7.","tokens_in":27525,"tokens_out":8618,"duration_ms":84935,"significance":"The observations are novel and valuable: Band 9 (rest-frame ~55 micron) data for a z>7 LBG are rare, and the combination with Band 4 (rest-frame ~270 micron) data offers one of the few multi-band dust SED constraints at these redshifts. If the hot-dust interpretation is correct, the Big Three Dragons would be among the most IR-luminous and most obscured galaxies known at z>7, with direct implications for the obscured fraction of star formation in the epoch of reionization. The paper is honest in reporting alternative fits (Appendix D) and caveats about optically thick dust (Section 5.2.3), and the homogeneous reduction of all bands is a strength. However, as argued in the major comments, the headline quantitative claims are conditional on an unconstrained emissivity index and on the optically thin assumption, and the new detections themselves are of modest significance once calibration errors are included. The paper is therefore more useful as a demonstration of what is feasible and as a target for future follow-up than as a secure measurement of hot dust.","major_comments":[{"comment":"The claim that the Big Three Dragons hosts 'hot' dust (T_d ≈ 78 K) is not robust to the adopted emissivity index. With the standard β_IR = 2.0 fixed, the paper's own fit (Appendix D, Table D1) gives T_d = 57^{+15}_{-13} K, whose median falls below the paper's own definition of hot (T_d ≳ 60 K, footnote on p. 3). Since the free-β fit returns β_IR = 1.47^{+0.46}_{-0.35}, fully consistent with 2.0, the data cannot distinguish between a 78 K solution and a 57 K solution. The abstract's statement that the MBB fit 'robustly constrains' the dust temperature and the Section 5.1.1 wording that the observations 'strongly indicate' hot dust are therefore too strong. I recommend reporting the temperature as T_d ≈ 78^{+35}_{-23} K under a free-β model, with the β=2.0 solution presented as an equally viable interpretation, and moving the headline claim from 'hot dust' to a statement that the dust temperature is consistent with hot and with standard-cold solutions, while the IR luminosity is high either way.","section":"§4, Appendix D, Eq. (1)"},{"comment":"The significance of the new Band 9 detections is overstated. Table 1 lists a global Band 9 flux of 1162 ± 452 μJy from the aperture noise alone; when the 20% calibration error added in Section 3.1 is included in quadrature, the detection is at ≈2.6σ. The two resolved clumps (381 ± 191 and 532 ± 191 μJy in Table 2) fall to ≈2.0σ and ≈2.8σ, respectively. Because the rest-frame ~55 μm point is the only constraint on the Wien side of the SED, these marginal detections cannot carry the weight placed on them in the temperature and luminosity inference. Please report the calibration-inclusive significances in the abstract and body, and adjust the language in Section 2 ('clearly detected') accordingly.","section":"§2, Table 1, Table 2"},{"comment":"The fiducial physical parameters (T_d, M_d, L_IR, f_obs) are all derived under the optically thin MBB assumption of Eq. (1). The paper acknowledges in §3.2 that the optically thick case would yield higher temperatures, and in §5.2.3 that the Eastern clump could be optically thick, but the abstract and conclusions still present L_IR = 10^12.32 L_sun and f_obs = 0.94 as firm numbers. The statement in §5.2.3 that optically thick dust would 'not significantly affect the infrared luminosity' is not demonstrated with a fit; the luminosity is defined as the integral of the optically thin model. Since the data cannot constrain the optical depth (§3.2), the reported L_IR and f_obs should be explicitly labeled as optically thin-model values, and the optically thick alternative should be carried through the main discussion rather than being confined to a caveat.","section":"§3.2, §5.2.3"}],"minor_comments":[{"comment":"The conclusions bullet gives SFR_UV+IR = 267^{+419}_{-154} M_sun/yr, while Table 3 and the abstract give 267^{+418}_{-153}; this numerical inconsistency should be corrected.","section":"§6 bullet 2"},{"comment":"The statement that a single B3D-like galaxy accounts for 20–50% of all star formation at the bright end (M_UV < -21.5) is a definitional consequence of assigning f_obs = 0.95 to one object and zero to the rest; please add a caveat that this calculation does not constrain the true incidence rate, and that the two adopted UV LFs differ by a factor of ~3 in N_LBG, which is the dominant uncertainty in the shaded band.","section":"§5.3"},{"comment":"The tapered 0.8'' Band 9 image yields a peak S/N of 4.0σ, a value that is still modest; the main text should state this value alongside the natural-weighting significances, or the abstract should clarify that the quoted 3.7/3.5σ values do not include the tapering improvement.","section":"Appendix A"},{"comment":"The paper's 'hot' threshold is arbitrary and acknowledged as such in the footnote on p. 3, but later sections (e.g., §5.1.1, §6) use 'hot' as a factual category; consider consistently using 'hot (as defined here)' or replacing 'strongly indicate' with 'are consistent with' in the abstract and conclusions.","section":"§5.1.1"}],"recommendation":"major_revision","confidential_remarks":"For the editor: the manuscript is within the scope of MNRAS and contains unique new ALMA Band 9 data for a z>7 LBG. The central concern is not methodological sloppiness but a mismatch between the strength of the evidence and the strength of the claims. The authors are well placed to fix this by reframing the abstract and conclusions and by moving the β=2.0 and optically thick alternatives into the main analysis. I would support publication after such a revision. I did not see inappropriate citation inflation or duplication concerns."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The new ALMA Band 4 and 9 data are real and useful. Two extra points on the dust SED of the Big Three Dragons, including rest-frame ~55 micron coverage, make this the best-constrained dust SED of any z>7 LBG after MACS0416-Y1. The analysis is careful: Herschel upper limits are included, resolved clump fits are shown, and the paper openly tests the beta=2.0 case in Appendix D. This is not a sloppy paper.\n\nWhat is robust: under both the free-beta and fixed-beta fits, the system is IR-luminous (log L_IR ~12.1-12.3 Lsun) and heavily obscured (f_obs ~0.89-0.94). The total SFR exceeds what the UV alone would suggest, and the Western clump sits far above the IRX-beta_UV relation. Those conclusions survive the modeling choices.\n\nThe soft spot is the headline \"hot dust\" claim. The Band 9 detections are only ~2-3 sigma after calibration (1162 +- 452 uJy global; component fluxes ~2.0 and ~2.8 sigma), and Band 9 does not fully sample the SED peak. The free-beta fit returns beta_IR = 1.47 +- 0.46 and T_d = 78 K, but the paper's own Appendix D shows that fixing beta=2.0 gives T_d = 57 +- 15 K, below the 60 K threshold they define as hot. So \"hot\" is prior-dependent, not a measurement. The optically thick possibility discussed in Section 5.2.3 would push T_d up, but that is an additional assumption, not confirmation. The paper flags all this, yet the abstract and conclusions lean on the 78 K value. The resolved clump temperatures have similar fragility.\n\nThe SFRD implications are explicitly an extrapolation with an assumed incidence rate x_B3D. That section is transparent about its assumptions and does not overreach, so I would not penalize it heavily.\n\nFor peer review: this deserves a serious referee. The data are new, the analysis is reproducible in structure, and the interpretation is honest. The main revision should be to move the fixed-beta fit into the main text or soften the \"hot dust\" language, so that readers see the temperature is conditional on the emissivity index. I would cite this for the new fluxes and the SFRD framework, but not as proof of a hot dust population at z>7.\n\nRecommendation: send to review, with the expectation that the abstract be recalibrated to match the data's actual constraining power.","headline":"New bands, honest analysis, but the 'hot dust' label rests on a free-beta fit; the system is still IR-luminous and heavily obscured.","tokens_in":28273,"tokens_out":2350,"would_cite":true,"duration_ms":27184,"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 z=7.15 galaxy pair Big Three Dragons hosts dust heated to about 78 K, and a modified blackbody fit implies 94 percent of its star formation is obscured, making it one of the most luminous hidden starbursts known at z>7.","keywords":["galaxies: high-redshift","galaxies: evolution","submillimeter: galaxies","dust temperature","obscured star formation","Lyman-break galaxies","epoch of reionization","ALMA"],"falsifier":"Observe the Big Three Dragons in ALMA Band 10, probing rest wavelengths near 40 um, with enough sensitivity to measure or tightly limit its flux. The optically thin 78 K modified blackbody predicts a steeply rising SED into this band; a measurement well below that prediction, or a resolved turnover, would show that the dust becomes optically thick near the peak and that the quoted temperature is not the intrinsic dust temperature. A detection matching the extrapolation would support the hot-dust interpretation.","tokens_in":26985,"feed_emoji":"🔥","tokens_out":14861,"duration_ms":148051,"temperature":0.7,"pith_summary":"The paper argues that the Big Three Dragons, a merging pair of star-forming galaxies seen at redshift z=7.15, is a genuinely hot and heavily dust-obscured system: new ALMA observations at 0.45 mm and 2.2 mm, combined with earlier detections at 0.73-1.32 mm, imply a global dust temperature of $T_d=78^{+35}_{-23}$ K, an infrared luminosity of $\\log(L_{\\rm IR}/L_\\odot)=12.32^{+0.43}_{-0.41}$, and an obscured fraction of $f_{\\rm obs}=0.94^{+0.04}_{-0.09}$. If this is right, the bright end of the UV luminosity function at $z>7$ is not as clean a window on star formation as UV surveys assume: much of the star formation in at least some bright Lyman-break galaxies is hidden by warm dust. The result matters because cosmic star-formation histories are largely reconstructed from ultraviolet light, and dust corrections depend sensitively on the temperature assumed for the dust.","feed_headline":"Hot dust hides 94% of a galaxy pair's star formation at z=7.15","feed_subtitle":"Band 9 and 4 data make the pair one of the most infrared-luminous galaxies in the early Universe.","key_machinery":"The central object is a single-temperature modified blackbody (MBB) fit to the dust SED, $S_\\nu = \\frac{1+z}{d_L^2} M_d \\kappa_0 (\\nu/\\nu_0)^{\\beta_{\\rm IR}} [B_\\nu(T_{d,z}) - B_\\nu(T_{\\rm CMB,z})]$, with optically thin dust, $\\kappa_0=10.41$ cm$^2$/g at $\\nu_0=1900$ GHz, and CMB heating and attenuation included. The fit takes the ALMA flux densities in Bands 4, 6, 7, 8, and 9, plus Band 3 and Herschel upper limits, and constrains $T_d$, $M_d$, $\\beta_{\\rm IR}$, and the integrated $L_{\\rm IR}$ simultaneously through MCMC posterior sampling. The IRX-$\\beta_{\\rm UV}$ comparison and the $T_d$--$\\Sigma_{\\rm SFR}$ diagram then translate those fitted quantities into statements about dust geometry, possible optical thickness, and the incidence of such systems.","core_discovery":"The paper's central claim is that the Big Three Dragons is a genuinely hot, heavily obscured system at $z=7.15$. From a single-temperature optically thin modified blackbody fit to ALMA Bands 4, 6, 7, 8, and 9, the global dust temperature is $T_d = 78^{+35}_{-23}$ K, with $\\log(M_d/M_\\odot)=6.85^{+0.33}_{-0.27}$, $\\beta_{\\rm IR}=1.47^{+0.46}_{-0.35}$, and $\\log(L_{\\rm IR}/L_\\odot)=12.32^{+0.43}_{-0.41}$. This yields a total star formation rate $\\mathrm{SFR}_{\\rm UV+IR}=267^{+418}_{-153}\\,M_\\odot\\,\\mathrm{yr}^{-1}$ and an obscured fraction $f_{\\rm obs}=0.94^{+0.04}_{-0.09}$, placing the system well above the low-redshift obscured-fraction--stellar-mass relation. The resolved Bands 6, 8, and 9 data show both Lyman-break components are hot ($T_d\\approx67$--$84$ K) and heavily obscured ($f_{\\rm obs}\\approx0.88$--$0.95$); the Western component lies roughly one dex above the canonical IRX-$\\beta_{\\rm UV}$ relations, implying patchy dust, while the Eastern component is consistent with a Calzetti- or SMC-like foreground screen. If correct, the system is among the most infrared-luminous galaxies known at $z>7$, and a single such source already constitutes a substantial fraction of the star formation budget of UV-bright galaxies at $7<z<7.5$.","pith_inferences":["The paper does not state this, but its temperature-dependent luminosity argument implies that single-band high-redshift surveys fixing dust temperature near $\\sim35$--$50$ K will systematically underestimate infrared luminosities and obscured fractions for hot systems, so part of the apparent decline of obscured star formation beyond cosmic noon may be a temperature-selection effect.","The paper itself notes that the Eastern clump could be optically thick; if that is confirmed, the quoted $T_d$ would be a lower limit, while the infrared luminosity would remain roughly unchanged, so the qualitative conclusion of a luminous, heavily obscured starburst would survive.","A natural extension is to target compact, UV-bright major mergers at $z>7$ with dual-band millimeter photometry on the SED peak and Rayleigh-Jeans tail; the paper's Monte Carlo calculation implies that if more than about one in ten such systems is Big Three Dragons-like, obscured star formation would dominate the bright end of the UV luminosity function at that epoch."],"forward_implications":["A single system like this can account for roughly 20--50 percent of the star formation budget of UV-bright ($M_{\\rm UV}<-21.5$) galaxies at $7<z<7.5$ in the COSMOS field, depending on the adopted UV luminosity function.","The Big Three Dragons becomes a second, more luminous $z>7$ example alongside the similarly hot galaxy MACS0416-Y1, suggesting hot dust is a real phenomenon at these epochs rather than a unique object.","The extremely high obscured fraction, $f_{\\rm obs}\\approx0.94$, places the system far above the low-redshift obscured-fraction--stellar-mass relation, showing that compact merger-driven starbursts can hide almost all of their star formation.","The Western component's offset from the IRX-$\\beta_{\\rm UV}$ relation indicates that simple dust-screen geometries are insufficient; patchy or clumpy dust must be included when interpreting high-redshift dust attenuation.","If Big Three Dragons-like systems are not extremely rare, the obscured cosmic star-formation rate density at $z\\sim7$ could be substantially higher than UV-only estimates suggest."],"supporting_citations":[{"why":"supplied the ALMA Band 6 and 8 dust continuum and line measurements that anchor the short-wavelength side of the global SED.","marker":"Hashimoto et al. (2019)"},{"why":"added the Band 7 continuum detection and the earlier three-point SED estimate that the new data supersede.","marker":"Sugahara et al. (2021)"},{"why":"identified the Big Three Dragons as a z~7 Lyman-break galaxy in COSMOS, defining the selection volume used for the SFRD calculation.","marker":"Bowler et al. (2014)"},{"why":"provided the JWST/NIRCam resolved photometry, merger geometry, stellar masses, and IRX-beta values used in the comparison.","marker":"Sugahara et al. (2025)"},{"why":"provided effective radii, H-alpha maps, and UV luminosities used to compute SFR surface densities and obscured fractions.","marker":"Prieto-Jiménez et al. (2025)"},{"why":"supplied the z=8.31 MACS0416-Y1 hot dust measurement that anchors the comparison population.","marker":"Bakx et al. (2025)"},{"why":"provided the multi-band modified-blackbody fitting strategy and the comparison sample of z>6.5 Lyman-break galaxies.","marker":"Algera et al. (2024b)"},{"why":"supplied the dust opacity coefficient kappa0=10.41 cm^2/g adopted in the fit and the theoretical T_d-z framework.","marker":"Sommovigo et al. (2022a)"},{"why":"provided the UV luminosity function used to derive the incidence-rate implications for the obscured SFRD.","marker":"Harikane et al. (2025)"},{"why":"supplied the SFR_IR-to-L_IR conversion used to translate infrared luminosity into obscured star formation rate.","marker":"Inami et al. (2022)"}],"fun_headline_variants":["Hot dusty galaxy pair at z=7.15 hides 94% of star formation","z=7.15 galaxy pair: hot dust conceals 94% of star formation","94% of star formation hidden by hot dust in a z=7.15 merging pair","Merging galaxies at z=7.15: hot dust, 94% obscured star formation","At z=7.15, hot dust in a galaxy pair hides 94% of star formation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The dust is assumed to be optically thin at every observed wavelength, so if the compact Eastern component is actually optically thick, the quoted temperatures, luminosities, and obscured fractions would all shift.","fun_headline_variants_meta":{"raw":{"variants":["Hot dusty galaxy pair at z=7.15 hides 94% of star formation","z=7.15 galaxy pair: hot dust conceals 94% of star formation","94% of star formation hidden by hot dust in a z=7.15 merging pair","Merging galaxies at z=7.15: hot dust, 94% obscured star formation","At z=7.15, hot dust in a galaxy pair hides 94% of star formation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000816,"raw_usage":{"total_tokens":3820,"prompt_tokens":1436,"completion_tokens":2384,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":1052,"completion_tokens_details":{"reasoning_tokens":2265}},"tokens_in":1052,"tokens_out":2384,"duration_ms":16570,"temperature":1.0,"reasoning_tokens":2265,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T00:32:27.576297+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe the Big Three Dragons in ALMA Band 10, probing rest wavelengths near 40 um, with enough sensitivity to measure or tightly limit its flux. The optically thin 78 K modified blackbody predicts a steeply rising SED into this band; a measurement well below that prediction, or a resolved turnover, would show that the dust becomes optically thick near the peak and that the quoted temperature is not the intrinsic dust temperature. A detection matching the extrapolation would support the hot-dust interpretation.","supporting_citations":[],"review_version":1}