{"id":"deb22b0b-d871-4580-9f55-0123cf3931b4","arxiv_id":"2412.17980","paper_version":4,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"ACA low-resolution observations reveal a 75% CO(4-3) flux excess in SPT2349-56 relative to the sum of ALMA-detected galaxies, implying a large extended molecular gas reservoir.","lead":"Astronomers used a compact-array telescope with a wide beam and found 75% more carbon monoxide gas around the distant protocluster SPT2349-56 than sharp images of its member galaxies showed. The extra gas is spread out and dim, suggesting a reservoir that could keep the system's intense star formation burning for hundreds of millions of years.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'extended gas' interpretation rests on a luminosity-function extrapolation; if faint unresolved CO emitters are moderately more numerous than the Hill et al. (2020) constraint allows, the 75% excess requires no diffuse component.","rationale":"The reader's weakest_assumption (faint-end luminosity function no steeper than Hill et al. 2020) matches the single most load-bearing link in the argument. The empirical findings are otherwise well supported: the ACA-ALMA visibility amplitudes agree at similar uv distance (Appendix A), a 10'' taper of the ALMA data recovers a comparable excess (7.41+/-0.77 Jy km/s), and the APEX [CII] spectrum independently shows an excess in the same direction. These checks make a pure calibration or CLEANing artifact unlikely. What they do not rule out is a population of resolved-out discrete sources that are simply below the 12-m detection threshold; such a population would contaminate the ACA aperture and the APEX beam but not the high-resolution source sum. The paper's only quantitative argument against this scenario is the Hill et al. (2020) luminosity function, whose applicability to CO(4-3) at these faint fluxes is asserted rather than demonstrated, especially since the paper's own in-prep references (Sulzenauer et al.) acknowledge a 'considerable' faint-emitter contribution. A stacking and injection test would settle whether discrete sources can account for the missing flux. Because the authors have not yet provided that test, the conditional verdict is appropriate; no change to the reader's recommendation is needed.","tokens_in":19889,"tokens_out":7773,"duration_ms":81235,"concrete_test":"Stack the ALMA 12-m CO(4-3) data cube at the positions of all spectroscopically confirmed members and of all Hill et al. (2020) candidate line emitters below the original detection threshold, integrating over the same +/-3 sigma velocity window as the ACA spectrum, and compare the stacked-plus-detected flux with the ACA total. Separately, simulate a mock faint-emitter population drawn from the Hill et al. (2020) luminosity function with the faint-end slope varied over its 1 sigma uncertainty and a flux limit down to 0.01 Jy km/s, inject these as point sources into the 12-m visibilities, and re-extract the integrated flux. If the simulated discrete population accounts for more than ~50% of the 2.86 Jy km/s excess, the extended-gas interpretation is not required; if the observed excess survives, the central claim is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The ACA 7-m measurement integrates all CO(4-3) flux inside a ~30''x17'' aperture, while the ALMA 12-m comparison sums only sources above a finite detection threshold. The central claim, that the resulting ~2.9 Jy km/s excess is spatially extended molecular gas (CGM/proto-ICM), therefore requires that the population of fainter discrete emitters contribute far less than the excess. Section 4 dismisses this possibility by citing the 'flat faint end' of the Hill et al. (2020) luminosity function, but that constraint was set primarily by [CII]-selected emitters at S/N>=6.2 in high-resolution ALMA data; it is not demonstrated to constrain the CO(4-3) luminosity function at the low fluxes relevant here, and the conversion from [CII] to CO adds further uncertainty. The paper itself notes that 'a missing population of faint emitters might contribute a considerable amount' and that Sulzenauer et al. (in prep) finds faint emitters and streamers. A modestly steeper faint-end slope, or a population of CO-bright but [CII]-faint sources, would produce exactly the observed pattern: low-resolution ACA and single-dish APEX show excess flux while high-resolution ALMA resolves only the bright members. Thus, excluding a discrete faint-source population is the most load-bearing link between the robust empirical excess and the astrophysical interpretation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents deep Atacama Compact Array (ACA) Band-3 observations of the protocluster SPT2349-56 at z=4.3, measuring CO(4-3) and 3.2 mm dust continuum. The authors report a 75% excess in the velocity-integrated CO(4-3) flux measured with the low-resolution ACA 7-m array relative to the sum of individual galaxies detected in higher-resolution ALMA 12-m data, a result also seen when the ALMA data are tapered to 10 arcsec, and a 135% excess in [CII] emission from an archival APEX/FLASH spectrum relative to the co-added ALMA [CII] spectra. The 3.2 mm continuum does not show a significant excess. The authors argue that the missing CO and [CII] flux is unlikely to be due to undetected faint galaxies, and instead interpret it as spatially extended, low-surface-brightness molecular gas in the circumgalactic medium (CGM) or a pre-heated proto-intracluster medium (proto-ICM). They further use this excess to revise the gas depletion timescale to >400 Myr, reducing the need for synchronized ULIRG activity.","tokens_in":20172,"tokens_out":6006,"duration_ms":60649,"significance":"The empirical result is important: if confirmed, it demonstrates that high-resolution ALMA observations of protocluster cores can miss a substantial fraction (tens of percent) of the total line emission, with direct consequences for gas mass estimates and for our understanding of how extreme starbursts in protoclusters are sustained. The paper's strengths include the use of two independent low-resolution estimators (ACA 7-m and tapered ALMA), a single-dish [CII] spectrum from APEX, and a visibility-plane comparison that argues against a simple amplitude-calibration offset. The authors also explicitly acknowledge that their gas masses are lower limits and discuss possible systematic effects. However, the central astrophysical interpretation—that the missing flux is diffuse gas rather than a population of faint discrete sources—rests on a single, under-quantified argument, which is the load-bearing point for the CGM/proto-ICM conclusion.","major_comments":[{"comment":"The claim that the missing CO(4-3) flux cannot be attributed to faint unresolved emitters is not quantitatively supported. The text states that 'additional >100 faint line emitters with F_CO(4-3) ≳ 0.03 Jy km/s would be required' and that this 'cannot be accommodated by the current luminosity function in SPT2349-56, where the flat faint end is reasonably well constrained (Hill et al. 2020).' However, no integral of the luminosity function is presented, and the cited Hill et al. (2020) constraint is based on [CII]-selected emitters at S/N ≥ 6.2 in high-resolution ALMA data. The conversion from the [CII] luminosity function to a CO(4-3) luminosity function is not shown, and the possibility of a population of CO-bright but [CII]-faint sources is not addressed. Because this argument is the only quantitative basis for excluding the faint-source alternative, please provide a direct estimate of the expected contribution of sources below the ALMA detection threshold, including the effect of the CO/[CII] ratio uncertainty, or state the faint-end slope of the CO(4-3) luminosity function that would be required to explain the excess.","section":"Section 4, second paragraph"},{"comment":"The derived depletion timescale of >400 Myr and the corresponding relaxation of the synchronization requirement assume that the excess gas is a diffuse reservoir that can be accreted by the protocluster galaxies. If the excess instead originates from a population of faint, unresolved galaxies, the gas is already locked in galaxies and the reservoir interpretation—and the depletion argument—would not apply. The paper should either make this dependence explicit by presenting the depletion time as conditional on the diffuse-gas interpretation, or strengthen the evidence that the excess is not in discrete sources.","section":"Section 4, depletion timescale paragraph"},{"comment":"The visibility-plane comparison (Fig. 5) shows that the ACA and ALMA amplitudes are consistent at similar uv-distances, but the ACA data extend to shorter uv-distances than the ALMA 12-m data, and the excess appears at large angular scales (short uv). The paper interprets this as extended, low-surface-brightness gas, but a population of faint point sources distributed over the ~30 arcsec aperture would also contribute to the short-spacing flux. To distinguish these cases, the authors should model the uv-amplitude profile with both a diffuse component and a population of point sources, or use a stacking analysis of the faint [CII] emitters to estimate their total CO(4-3) contribution. As it stands, the evidence is consistent with extended gas but does not uniquely require it.","section":"Appendix A and Section 3.1"}],"minor_comments":[{"comment":"The entry '105.7.0 ±9.8' in the ALMA row appears to contain a typo; it should likely be '105.7 ±9.8' or '105.0 ±9.8'.","section":"Table 1"},{"comment":"The sentence 'Considering ≈ 200 dumps are included in ∼ 20 hours of observations (including calibrations), some effective beam size uncertainty may be introduced' is unclear; please clarify what 'dumps' refers to and how the number of dumps affects the beam size uncertainty.","section":"Section 2.2"},{"comment":"The caption refers to 'red curves' for the best-fit Gaussian profiles, but the figure description in the text indicates that the ACA spectrum is blue and the ALMA spectrum is red; please ensure the color references are consistent and unambiguous.","section":"Figure 2 caption"},{"comment":"The velocity integration range for the core is [−1373 km/s, 1429 km/s] while that for N1 is [−349 km/s, 1429 km/s]; the asymmetry and the different ranges are not fully explained, and a brief justification would help the reader trust the integrated fluxes.","section":"Section 3.1"},{"comment":"The sentence 'A missing population of faint emitters might contribute a considerable amount of [CII] and CO(4–3) emission (Sulzenauer et al. in prep)' is in tension with the later dismissal of that possibility; even though the later paragraph attempts to address it, the paper would benefit from a more explicit reconciliation of these statements, particularly given that Sulzenauer et al. is cited as in-preparation.","section":"Section 4, first paragraph"}],"recommendation":"major_revision","confidential_remarks":"The paper presents a robust empirical excess that should be of interest to the protocluster and ISM communities. However, the central astrophysical interpretation—that the missing flux is diffuse gas in the CGM or proto-ICM—depends on excluding a faint discrete source population, and that exclusion is currently not demonstrated with sufficient rigor. I would encourage the editor to require the authors to either provide a quantitative luminosity-function-based estimate of the faint-source contribution or to soften the abstract/summary claims to present the extended-reservoir scenario as one plausible explanation rather than the primary conclusion. Given that the excess itself is well supported, rejection is not warranted; the issues are addressable with additional analysis or a revised interpretation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The 75% CO(4-3) excess in the core of SPT2349-56 looks real. The ACA measurement, the 10\" tapered ALMA check, and the APEX [CII] spectrum agree with each other, and the visibility-plane comparison is a good control against calibration offsets. The dust continuum not showing a similar excess is a useful discriminant, even if the SZ explanation is only cited as in-prep. This is the cleanest demonstration yet that high-resolution ALMA observations of an extreme protocluster core miss a large fraction of the cold gas, and it has obvious implications for depletion timescales and the onset of the ICM.\n\nThe soft spot is the step from 'excess flux' to 'extended diffuse gas.' The paper rules out a population of faint unresolved galaxies by invoking the flat faint end of the luminosity function from Hill et al. (2020). But that constraint comes from [CII]-selected emitters at S/N > 6.2 in high-resolution ALMA data. It is not directly a constraint on the CO(4-3) luminosity function at low fluxes, and the [CII]-to-CO conversion adds uncertainty. A modestly steeper faint end, or a population of CO-bright but [CII]-faint sources, would produce exactly the observed pattern. The paper itself admits that 'a missing population of faint emitters might contribute a considerable amount' and that Sulzenauer et al. (in prep) finds such faint emitters and streamers. So the conclusion that the missing flux must be spatially extended rests on an assumption that is not as well tested as the opening of the paper suggests.\n\nThe other issues are minor in comparison. The curve-of-growth aperture choice could be pushed further, and the 'proto-ICM' interpretation is one of several plausible options (CGM, CLEANing artifacts, stripped gas). The gas masses and depletion times use standard conversion factors, and the authors are explicit that these may not apply to a diffuse or pristine phase. The reliance on two in-prep papers (Sulzenauer et al., Zhou et al.) for key supporting evidence is a bit frustrating, but not disqualifying.\n\nBottom line: this is a solid observational paper with a robust central measurement and an overreaching but clearly flagged interpretation. A referee should ask the authors to either provide the luminosity-function analysis in a form that directly constrains CO(4-3) at the relevant fluxes, or to soften the claims about excluding faint emitters. I would send it to review, and I would cite it for the flux-excess measurement.","headline":"The 75% CO(4-3) excess in SPT2349-56 is real and well-checked, but the claim that it must be extended diffuse gas outruns the luminosity-function constraint used to exclude faint discrete emitters.","tokens_in":20816,"tokens_out":3321,"would_cite":true,"duration_ms":32910,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Deep low-resolution maps reveal a diffuse molecular gas reservoir in SPT2349-56 that high-resolution ALMA misses.","keywords":["submillimeter astronomy","galaxy evolution","high-redshift galaxies","molecular gas","CO(4-3) emission","protoclusters","circumgalactic medium","intracluster medium"],"falsifier":"A very deep ACA-plus-total-power observation of the same field that reaches about 0.03 Jy km/s per synthesised beam would resolve the missing CO(4-3) into either discrete point sources or smooth extended emission: finding more than about 100 compact emitters would refute the extended-gas claim, while smooth diffuse emission would confirm it.","tokens_in":19688,"feed_emoji":"🌌","tokens_out":13498,"duration_ms":117548,"temperature":0.7,"pith_summary":"Deep, low-resolution radio observations of the most intensely star-forming protocluster core known at $z=4.3$ find that its CO(4-3) emission is 75% brighter than the sum of all galaxies individually detected in higher-resolution ALMA data, and [C ii] emission shows an even larger excess. The paper argues that this missing flux is not explained by faint undetected galaxies but by spatially extended, low-surface-brightness molecular gas in the circumgalactic medium or a pre-heated proto-intracluster medium. If true, high-resolution interferometric surveys systematically undercount the gas available to fuel star formation in dense early-universe environments. This extended reservoir could extend the gas consumption timescale to more than 400 Myr, easing the puzzle of how more than ten ultraluminous galaxies are all active at once, and may represent the cold beginnings of the intracluster medium.","feed_headline":"Protocluster hides 75% more gas than ALMA's sharpest view","feed_subtitle":"ACA low-resolution maps catch diffuse gas between galaxies, stretching the starburst fuel supply past 400 million years.","key_machinery":"The load-bearing comparison is between low-resolution and high-resolution views of the same gas. A single ACA 7-m pointing captures the full spatial extent of CO(4-3) and dust emission, while co-added ALMA 12-m spectra of individually detected galaxies capture only what the high-resolution array can see; re-imaging the ALMA data with a 10-arcsec taper, a processing step that down-weights long baselines to recover extended emission, acts as an intermediate cross-check. Three tracers are used in parallel, CO(4-3), 3.2 mm dust continuum, and [C ii], with visibility-plane amplitude comparisons to demonstrate that calibration differences do not produce the excess. The comparison isolates flux on angular scales that the 1-arcsecond ALMA observations resolve out, revealing a phase with high gas-to-dust ratio and low surface brightness that cannot belong to the galaxies' own ISM.","core_discovery":"Using roughly 30 hours of Atacama Compact Array (ACA) 7-m observations, the paper measures the CO(4-3) line and 3.2 mm continuum over the whole SPT2349-56 core with a roughly 21-by-12 arcsecond beam and compares the integrated line strength with the co-added spectra of all galaxies individually detected in high-resolution ALMA 12-m data. The ACA core spectrum contains 6.65 ± 0.30 Jy km/s, 75% more than the 3.79 ± 0.12 Jy km/s recovered by summing the 12-m detections, and the same excess appears when the ALMA data are re-imaged with a 10-arcsec taper and in the APEX [C ii] spectrum, which is 135% brighter than the summed ALMA emission. The 3.2 mm dust continuum does not show a corresponding excess, and visibility-plane amplitudes agree between the arrays, so the paper rules out flux calibration as the cause. It argues that no more than about 100 faint emitters below the detection threshold can account for the missing CO, because the faint end of the line luminosity function would need to be steeper than previous data allow. The conclusion is that high-resolution ALMA resolves out extended low-surface-brightness molecular gas, likely in the circumgalactic medium or a cold proto-intracluster medium, and that, if this reservoir feeds star formation, the core's depletion timescale is larger than 400 Myr.","pith_inferences":["Editorial extension: Detecting the same excess in CO(1-0), which is less dependent on excitation conditions than CO(4-3), would directly test whether the diffuse gas is truly cold and massive rather than warm and lower-density.","Editorial extension: Applying the same ACA-versus-high-resolution comparison to other extreme protoclusters would show whether this hidden reservoir is a general feature of overdense cores or special to SPT2349-56.","Editorial extension: The lack of a dust excess predicts a very high gas-to-dust ratio for the diffuse phase; a deep 850-micron continuum map of the same region could confirm or contradict that prediction."],"forward_implications":["The molecular gas mass of the core derived from CO(4-3) rises to about 1.9 trillion solar masses, making earlier high-resolution estimates lower limits.","The gas-depletion timescale of the core becomes at least 400 Myr, so the simultaneous ULIRG activity no longer requires a tightly synchronized, short-lived burst.","High-resolution ALMA fluxes of individual galaxies in similar systems should be treated as lower limits, and extended emission can be partially recovered by heavy uv-tapering, accurate CLEAN masks, or compact-array and total-power observations.","The excess emission may represent the cold, pre-virialized phase of the intracluster medium, meaning the ICM begins forming before the cluster becomes a hot, relaxed structure."],"supporting_citations":[{"why":"It supplies the co-added ALMA CO(4-3) and [C ii] spectra of the protocluster members and the faint-end luminosity function constraint that rules out a large population of undetected emitters.","marker":"Hill et al. 2020"},{"why":"It establishes the discovery and redshift of SPT2349-56 as a protocluster core and identifies the member galaxies.","marker":"Miller et al. 2018"},{"why":"It frames the ULIRG duty-cycle problem that the longer depletion timescale would relax.","marker":"Casey 2016"},{"why":"It provides the CLEANing method and the ACA 7-m flux-loss simulations used to treat the ACA measurements as lower limits.","marker":"Leroy et al. 2021"},{"why":"It supplies the standard formulas used to convert CO line strengths to line luminosities and molecular gas masses.","marker":"Solomon & Vanden Bout 2005"},{"why":"It provides the [C ii]-to-gas-mass conversion used to estimate the gas mass from the APEX spectrum.","marker":"Zanella et al. 2018"},{"why":"It supplies the dust-continuum gas-mass conversion used to compare the 3.2 mm continuum measurements.","marker":"Scoville et al. 2016"},{"why":"It provides earlier evidence for extended molecular gas around high-redshift protocluster galaxies that motivates the CGM/proto-ICM interpretation.","marker":"Emonts et al. 2016"}],"fun_headline_variants":["ACA reveals 75% extra CO that sharp ALMA misses in protocluster","Sharp vision loses 75% of protocluster gas; ACA wide beam finds it","Extended gas reservoir boosts CO by 75% in SPT2349 protocluster","Wide-angle CO map finds 75% more gas than sharp ALMA in cluster","Low-res ACA spots 75% more CO, hinting at proto-ICM gas"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion depends on the number of very faint, small galaxies below ALMA's detection limit being as small as earlier surveys suggest; if there are many more than about 100 of them, the excess could be galaxies rather than diffuse gas.","fun_headline_variants_meta":{"raw":{"variants":["ACA reveals 75% extra CO that sharp ALMA misses in protocluster","Sharp vision loses 75% of protocluster gas; ACA wide beam finds it","Extended gas reservoir boosts CO by 75% in SPT2349 protocluster","Wide-angle CO map finds 75% more gas than sharp ALMA in cluster","Low-res ACA spots 75% more CO, hinting at proto-ICM gas"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001136,"raw_usage":{"total_tokens":4868,"prompt_tokens":1246,"completion_tokens":3622,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":862,"completion_tokens_details":{"reasoning_tokens":3511}},"tokens_in":862,"tokens_out":3622,"duration_ms":23934,"temperature":1.0,"reasoning_tokens":3511,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T05:10:25.377577+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A very deep ACA-plus-total-power observation of the same field that reaches about 0.03 Jy km/s per synthesised beam would resolve the missing CO(4-3) into either discrete point sources or smooth extended emission: finding more than about 100 compact emitters would refute the extended-gas claim, while smooth diffuse emission would confirm it.","supporting_citations":[],"review_version":1}