{"id":"96e08360-8787-460d-a2e1-c960b3f89eeb","arxiv_id":"2502.06538","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"ALMA's CO(3-2) detection in ADF22-QG1 provides the first direct measurement of molecular gas in a quiescent galaxy at z~3, with gas fraction and gas-to-dust ratio constraints.","lead":"Astronomers detected carbon monoxide in a massive, already-quiet galaxy from when the universe was about two billion years old, the first direct sighting of molecular gas in such an early quiescent galaxy. The find shows that a surprising amount of star-forming fuel can survive in a galaxy that has stopped forming stars, testing ideas about how galaxies shut down star formation.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"CO(3-2) association with ADF22-QG1 is not secured: the line is ~250 km/s from the Balmer systemic redshift and ADF22.A4, a bright DSFG at the phase center, lies ~6 arcsec away; the paper does not rule out contamination or a companion.","rationale":"The reader's weakest assumption identifies the same load-bearing concern: the physical association of the CO(3-2) line with ADF22-QG1. I agree with that assessment and add that the proximity of the bright DSFG ADF22.A4, which is the ALMA phase center, makes contamination a concrete, testable possibility rather than a generic worry. The paper's own statement that DSFG emission is velocity-offset from the other two QGs implies that DSFG lines are present in the band and can be near the QG frequencies; for QG1, the line redshift z_CO = 3.0889 falls close to the proto-cluster redshift, so A4 or another group member could contribute. The 5.6σ significance and spatial coincidence with QG1's stellar light are real evidence, but they do not settle the association given the ~250 km/s offset from the Balmer systemic redshift and the 3″ beam. Because the reader's verdict is already CONDITIONAL and this concern supports that conditional status, no verdict change is needed. The concrete uv-plane two-source test would resolve whether the molecular gas actually belongs to the quiescent galaxy or to a contaminating neighbor.","tokens_in":11058,"tokens_out":6327,"duration_ms":58050,"concrete_test":"Fit the CO(3-2) visibility data with a two-source model containing point sources at the positions of ADF22-QG1 and ADF22.A4, using a uv-tapered image with ~1.5″ resolution, and compare the Bayesian or χ² evidence against a single-source model at ADF22-QG1. If the QG1 component is not independently detected at >3σ after subtracting A4's contribution, or if the centroid shifts toward A4, the claimed association is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the 5.6σ CO(3-2) emission at z_CO = 3.0889 ± 0.0007 originates in ADF22-QG1. This is the least secure link. The systemic redshift from Balmer absorption is z_abs = 3.0922 (+0.0008/-0.0004), a ~250 km/s offset at roughly 3σ, and the target is only ~6 arcsec from the bright DSFG ADF22.A4, which was the ALMA phase center and lies well within the primary beam; the synthesized beam is 3.00″ × 2.46″. The paper does not report A4's spectroscopic redshift in the text, does not show a spectrum or moment map centered on A4, and does not quantify sidelobe contamination from this source. The statement that DSFG emission does not contaminate the QG2/QG3 upper limits because of a velocity offset shows that DSFG emission is present and can be close in frequency, but no equivalent check is presented for QG1. If the emission arises from A4, a companion, tidal debris, or a blend, the derived M_H2, f_gas, and δ_gdr do not apply to the quiescent galaxy, and the 'first direct detection' claim loses its foundation. The qualitative agreement in position with the stellar light of QG1 is suggestive but does not exclude a bright neighbor in a ~3″ beam, especially when the putative line is offset from the more reliable Balmer absorption redshift.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports ALMA Band 3 CO(3–2) observations of three quiescent galaxies in the AzTEC14 group at z≈3.09 in the SSA22 protocluster. A 5.6σ line detection is claimed for ADF22-QG1, yielding z_CO = 3.0889 ± 0.0007; the paper derives molecular gas masses, gas fractions, and gas-to-dust ratios under two CO-to-H2 conversion factors (α_CO = 4.4 and 0.8), places upper limits on QG2 and QG3, and constructs toy models of gas-consumption history to argue that DSFGs at z≈4–6 are plausible progenitors and that halted gas accretion, possibly supplemented by morphological quenching, explains the quiescent state. The central claim is that this is the first direct detection of molecular gas in a quiescent galaxy at z≈3.","tokens_in":11447,"tokens_out":8568,"duration_ms":72882,"significance":"If the CO(3–2) line is securely associated with ADF22-QG1, this is an important result: it would provide the first direct molecular gas measurement in a z∼3 quiescent galaxy, with direct implications for quenching mechanisms. The paper is transparent about the α_CO degeneracy, presents upper limits for two additional quiescent galaxies, and benefits from deep ALMA observations (17 hr on-source, 33 μJy beam−1 per 100 km s−1) and updated JWST/NIRCam photometry. The toy-model interpretation is explicitly labeled as such. However, the scientific impact of the paper hinges on the source association of the CO line and on whether the 'molecular gas reservoir' remains substantial under the allowed conversion-factor range.","major_comments":[{"comment":"The attribution of the 5.6σ CO(3–2) line to ADF22-QG1 is not yet secured. The line redshift, z_CO = 3.0889 ± 0.0007, is offset from the Balmer-absorption redshift, z_abs = 3.0922 (+0.0008/−0.0004), by about 250 km s−1, or roughly 3σ with the quoted uncertainties. The phase center is ADF22.A4, a bright DSFG located about 6 arcsec from QG1, and the synthesized beam is 3.00″ × 2.46″, so sidelobe or faint extended emission from A4 cannot be dismissed without a dedicated check. The paper gives no spectrum or moment map centered on A4 and does not report A4's redshift. The statement in §3 that DSFG emission does not contaminate the QG2/QG3 upper limits because of a velocity offset shows that DSFG emission is present in the field, but no equivalent check is presented for QG1. Because the derived M_H2, f_gas, and δ_gdr in Table 1 and the interpretive discussion in §4 all rest on this attribution, please add a quantitative test of the association (e.g., fitting the line at the QG1 position after excluding A4, a uv-plane analysis, or an independent redshift measurement for A4).","section":"§3, Figure 2"},{"comment":"The magnitude of the claimed molecular gas reservoir is not robust to the adopted conversion factor. With α_CO = 4.4, log M_H2 = 10.26 and f_gas ≈ 14%; with α_CO = 0.8, log M_H2 = 9.52 and f_gas ≈ 2.5%, the latter being comparable to the upper limits derived for QG2 and QG3 (3.1% and 2.4%). The manuscript does not justify why α_CO = 4.4 is the appropriate value for a quiescent, likely metal-enriched galaxy at z∼3, and the abstract and discussion repeatedly emphasize the larger value. Please either provide a literature-based justification for the adopted α_CO and r31 = 0.5 for this class of galaxy, or reframe the central claim so that it does not depend on the α_CO = 4.4 case.","section":"§3, Table 1"},{"comment":"Please clarify how S_CO ΔV = 45 ± 8 Jy km s−1 was measured. The text says 'the total line intensity ... was measured as peak flux density in the moment-0 map,' which is ambiguous: a total line intensity should be an integrated flux (sum over channels), not a peak flux density. If the reported value is the peak of the moment-0 map, the conversion to CO luminosity is not specified and the derived M_H2 would not follow from the stated formula. This is a central quantity for every derived gas mass and upper limit, so the measurement definition needs to be precise.","section":"§3, Table 1"}],"minor_comments":[{"comment":"The uncertainties on z_abs are inconsistent: the text gives +0.0008/−0.0004, while Table 1 gives +0.008/−0.004. Please correct the typographical factor of 10.","section":"§3 vs Table 1"},{"comment":"There is a typo: 'lough upper limit' should be 'rough upper limit'.","section":"§3"},{"comment":"The caption says contours are drawn at 2σ, 3σ, ..., 5σ, but the text reports a 5.6σ CO(3–2) detection; please clarify whether the contour levels stop at 5σ by construction or why the detection significance exceeds the highest contour.","section":"Figure 1 caption"},{"comment":"The definition of σ_gas = 180 ± 50 km s−1 should be stated explicitly (e.g., Gaussian σ of the CO line profile versus second moment of the spectrum); this matters for the comparison with the stellar velocity dispersion.","section":"§3"},{"comment":"The closed-box, no-inflow toy model is clearly labeled as a toy model, but the text should add a sentence noting that alternative gas histories (e.g., strong outflows or episodic gas accretion) could also reproduce the observed f_gas values; this would avoid over-interpreting the fine details of the model tracks.","section":"§4, Figure 4"},{"comment":"The dust mass upper limit is derived assuming T_d = 20 K, β = 2.0, and κ_850 = 0.05 m² kg⁻¹; because the derived δ_gdr lower limits scale with T_d, a brief sentence on the sensitivity of δ_gdr to the assumed dust temperature would be useful.","section":"§3"}],"recommendation":"major_revision","confidential_remarks":"The main risk to the paper is the source association: if the authors cannot demonstrate that the CO(3–2) emission originates from ADF22-QG1 rather than from ADF22.A4 or a companion, the central claim will not stand. The α_CO dependence of the gas mass is a secondary but important issue that could be addressed by reframing the conclusions. The observational work itself is solid, and the paper is well written; a major revision with additional association tests and a more cautious quantitative summary would make it suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First direct CO(3-2) detection in a z~3 quiescent galaxy, with the caveat that the line's association with the target is less solid than the text claims. The 5.6-sigma detection is credible, and the paper is honest about the alpha_CO and r31 sensitivity. But the ~250 km/s offset from the Balmer systemic redshift and the unaddressed possibility of contamination from the bright DSFG ADF22.A4 (the phase center, ~6\" away) are real soft spots that need to be addressed before the claim fully lands.\n\nThe paper does a good job with the data: the CO spectrum is clean, the spatial overlap with the stellar light is reassuring, and the upper limits for QG2 and QG3 are reported with the same assumptions. The first gas-to-dust ratio constraint for a z~3 QG is a bonus, even if it's only a lower limit. The toy models of gas consumption are clearly labeled as such and don't overstate their case.\n\nMy main concern is the association. The text says the CO redshift is 'broadly consistent' with the Balmer absorption, but the numbers give ~250 km/s at about 3-4 sigma significance. That is a non-trivial offset. Also, the paper explicitly checks that DSFG emission doesn't contaminate the QG2/QG3 upper limits, but no such check is shown for QG1, even though ADF22.A4 is the phase center and only ~6\" away. A spectrum or moment map centered on A4 would settle this.\n\nThe derived molecular gas mass spans a factor of ~5 depending on alpha_CO, and the paper presents both values without pretending the uncertainty is small. That's fair. The measurement description for the line intensity — 'peak flux density in the moment-0 map' — is ambiguous; it should be the integrated flux.\n\nThese issues are fixable. If the authors can confirm the line belongs to QG1 and quantify the redshift offset, this becomes a solid, citable result. I'd send it to a referee; it deserves a careful look, not a desk rejection.","headline":"First CO(3-2) detection in a z~3 quiescent galaxy, but the line's association with the target is not as secure as the text implies.","tokens_in":12027,"tokens_out":3393,"would_cite":true,"duration_ms":30264,"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":"The paper reports the first direct detection of a molecular gas reservoir in a massive quiescent galaxy at $z\\approx3$, implying that quenching can leave cold gas in place.","keywords":["quiescent galaxies","molecular gas","CO(3-2) emission","high-redshift galaxies","proto-clusters","SSA22","gas quenching","ALMA observations"],"falsifier":"A follow-up observation with a synthesized beam smaller than the current $3.00^{\\prime\\prime}\\times2.46^{\\prime\\prime}$ beam, targeting CO(3-2) or CO(1-0) toward ADF22-QG1, would settle the question: if the line centroid and velocity field do not coincide with the galaxy's stellar light and systemic redshift but instead match the position of a companion or of ADF22.A4, then the reservoir does not belong to the quiescent galaxy and the derived gas mass, gas fraction, and gas-to-dust ratio all collapse.","tokens_in":10831,"feed_emoji":"🔭","tokens_out":15976,"duration_ms":110017,"temperature":0.7,"pith_summary":"This paper reports the first direct detection of molecular gas in a quiescent galaxy from the early universe. ALMA observations of the CO(3-2) line toward three massive quiescent galaxies in the core of the $z=3.09$ SSA22 proto-cluster detect one galaxy, ADF22-QG1, at $5.6\\sigma$. Taking the Milky-Way-like conversion factor $\\alpha_{\\rm CO}=4.4$, the implied molecular gas mass is $\\log M_{\\rm H_2}/M_\\odot = 10.26\\pm0.07$, a gas mass fraction of roughly 14%, while the other two galaxies are not detected and yield upper limits of $f_{\\rm gas}\\approx13$ to $17\\%$. The authors argue that quiescence can coexist with a substantial cold-gas reservoir, so star formation must be suppressed by a mechanism that prevents gas collapse rather than by exhausting the gas supply. They also show that the galaxy's gas-consumption history, traced from its star-formation history, is consistent with a halt in cosmic-web gas accretion after a dusty starburst phase.","feed_headline":"First molecular gas spotted in a massive dead galaxy at z≈3","feed_subtitle":"A 5.6-sigma CO detection means quenching can leave a galaxy quiet but still full of star fuel.","key_machinery":"The measurement machinery is the CO(3-2) line and its conversion chain to molecular gas mass: observed line intensity gives $L'_{\\rm CO(3-2)}$, a brightness-temperature ratio $r_{31}=0.5$ converts this to $L'_{\\rm CO(1-0)}$, and a CO-to-${\\rm H_2}$ factor $\\alpha_{\\rm CO}=4.4$ (or $0.8$) gives $M_{\\rm H_2}$. This chain produces the gas masses, gas fractions, and gas-to-dust ratios that anchor all of the paper's comparisons. The interpretive machinery is a set of toy closed-box gas-consumption models (no net inflow or outflow) driven by the non-parametric star-formation history of the galaxy: they integrate the star-formation rate backward in time to estimate the gas fraction a progenitor must have had, which connects the observed $z\\approx3$ quiescent galaxies to dusty star-forming galaxies at $z\\approx4$ to $6$ and to an assumed shutoff of fresh gas accretion.","core_discovery":"The central claim is that a massive quiescent galaxy at $z\\approx3$ can still contain a large reservoir of molecular gas. The CO(3-2) emission detected at $z_{\\rm CO}=3.0889\\pm0.0007$ overlaps the stellar light of ADF22-QG1, and the measured line intensity, combined with $r_{31}=0.5$ and $\\alpha_{\\rm CO}=4.4$, gives $\\log M_{\\rm H_2}/M_\\odot = 10.26\\pm0.07$ ($9.52\\pm0.07$ if $\\alpha_{\\rm CO}=0.8$), corresponding to $f_{\\rm gas}\\approx14\\%$ (2.5%). The gas velocity dispersion of $180\\pm50\\ \\mathrm{km\\,s^{-1}}$ is smaller than the stellar dispersion of roughly $320\\ \\mathrm{km\\,s^{-1}}$, and the galaxy is not detected in the 1.1 mm dust continuum, yielding the first gas-to-dust ratio constraint for a high-redshift quiescent galaxy of $\\delta_{\\rm gdr}\\gtrsim170$ (or $\\gtrsim30$ for the lower conversion factor). The paper interprets this as evidence that gas starvation, most plausibly a cessation of accretion from cosmic-web filaments, operates alongside an internal mechanism such as morphological quenching, in which the stellar potential suppresses gas collapse, to keep the remaining gas from forming stars.","pith_inferences":["If the line truly belongs to ADF22-QG1, the result shifts the theoretical burden from gas exhaustion to gas stabilization at early epochs: models in which feedback removes or consumes all cold gas would predict $f_{\\rm gas}\\approx0$, whereas the data allow a reservoir that is simply not collapsing.","Because $\\alpha_{\\rm CO}$ changes the gas mass by a factor of about 5.5, an obvious next test is a CO(1-0) or optically thin isotopologue map, which would also show whether the gas is a rotating disk or a diffuse warm phase and would distinguish morphological quenching from turbulence or virial support.","The mixed result within one group, one gas-rich quiescent galaxy and two gas-poor ones, suggests the local environment, not just global accretion shutdown, controls whether a quiescent galaxy keeps its reservoir; a survey of several $z\\approx3$ proto-cluster cores would test this."],"forward_implications":["ADF22-QG1 becomes the first high-redshift benchmark for quiescent-galaxy gas fractions, with $f_{\\rm gas}\\approx14\\%$ at $\\alpha_{\\rm CO}=4.4$.","The low star-formation rate relative to the CO luminosity shows that a molecular reservoir does not guarantee star formation, so quiescence must be maintained by dynamical or structural suppression.","The non-detections of the other two quiescent galaxies place upper limits of $f_{\\rm gas}\\approx13\\%$ to $17\\%$, showing that gas-rich and gas-poor quiescent galaxies coexist in the same proto-cluster core.","Closed-box, no-inflow gas-consumption tracks based on the measured star-formation history can reproduce the observed $f_{\\rm gas}$, favoring dusty star-forming galaxies at $z\\approx4$ to $6$ as progenitors and a halt in filamentary gas accretion as a quenching ingredient.","The first high-redshift quiescent-galaxy gas-to-dust ratio constraint, $\\delta_{\\rm gdr}\\gtrsim170$ for $\\alpha_{\\rm CO}=4.4$, gives a target that deeper dust-continuum observations can test against the predicted $\\delta_{\\rm gdr}\\sim500$ to $1000$."],"supporting_citations":[{"why":"Confirmed ADF22-QG1 as a quiescent galaxy with a Balmer absorption redshift, stellar mass, and star-formation history used in the gas-consumption models.","marker":"Kubo et al. 2021"},{"why":"Provided the spectroscopic confirmation and stellar properties of ADF22-QG2 and ADF22-QG3 that anchor the non-detection upper limits.","marker":"Kubo et al. 2022"},{"why":"Supplies the brightness-temperature ratio $r_{31}=0.5$ used to convert the CO(3-2) luminosity to a CO(1-0) luminosity.","marker":"Bothwell et al. 2013"},{"why":"Defines the star-formation-rate-versus-CO-luminosity scaling that the paper uses to show ADF22-QG1's star formation is suppressed.","marker":"Tacconi et al. 2013"},{"why":"Previous CO measurements of $z\\sim1$ quiescent galaxies and the constant-depletion-timescale toy model with which the $z\\approx3$ gas fractions are compared.","marker":"Williams et al. 2021"},{"why":"The $z\\approx3$ quiescent galaxy GS-10578 comparison target and the gas-consumption model adopted for the paper's star-formation-history tracks.","marker":"Scholtz et al. 2024"},{"why":"Proposed morphological quenching, the mechanism the paper invokes to explain suppressed star formation despite abundant molecular gas.","marker":"Martig et al. 2009"},{"why":"Predicts high gas-to-dust ratios in high-redshift quiescent galaxies, providing the theoretical expectation against which the derived $\\delta_{\\rm gdr}$ is interpreted.","marker":"Whitaker et al. 2021b"},{"why":"Mapped the Ly$\\alpha$ cosmic-web filaments around the AzTEC14 group, supporting the claim that filamentary gas accretion has ceased.","marker":"Umehata et al. 2019"}],"fun_headline_variants":["Dead galaxy at z≈3 harbors cold molecular gas","ALMA finds gas in a quiescent galaxy at z≈3","Quenched galaxy at z≈3 still has star fuel","Massive dead galaxy hides gas reservoir at z≈3","Molecular gas detected in early dead galaxy"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the CO(3-2) emission detected at $5.6\\sigma$ really comes from ADF22-QG1 and not from a nearby companion, tidal debris, or the bright dusty star-forming galaxy ADF22.A4 a few arcseconds away; the roughly ${\\sim}250\\ \\mathrm{km\\,s^{-1}}$ offset between the CO redshift and the redshift measured from stellar absorption lines makes this association the point where the argument could break.","fun_headline_variants_meta":{"raw":{"variants":["Dead galaxy at z≈3 harbors cold molecular gas","ALMA finds gas in a quiescent galaxy at z≈3","Quenched galaxy at z≈3 still has star fuel","Massive dead galaxy hides gas reservoir at z≈3","Molecular gas detected in early dead galaxy"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000231,"raw_usage":{"total_tokens":1675,"prompt_tokens":1326,"completion_tokens":349,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":942,"completion_tokens_details":{"reasoning_tokens":269}},"tokens_in":942,"tokens_out":349,"duration_ms":5428,"temperature":1.0,"reasoning_tokens":269,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T15:07:40.007263+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A follow-up observation with a synthesized beam smaller than the current $3.00^{\\prime\\prime}\\times2.46^{\\prime\\prime}$ beam, targeting CO(3-2) or CO(1-0) toward ADF22-QG1, would settle the question: if the line centroid and velocity field do not coincide with the galaxy's stellar light and systemic redshift but instead match the position of a companion or of ADF22.A4, then the reservoir does not belong to the quiescent galaxy and the derived gas mass, gas fraction, and gas-to-dust ratio all collapse.","supporting_citations":[],"review_version":1}