{"id":"eacac858-484d-4811-b1c8-ed2d7d163215","arxiv_id":"2411.12722","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"At z~3.1, massive quiescent galaxies show short, uniform quenching timescales and no environmental dependence, suggesting internal mechanisms dominate.","lead":"This paper identifies 24 massive galaxies at redshift 3.1 whose star formation has shut off, and tests whether their locations in the cosmic web caused the shutdown. The galaxies seem to be quenched fast and evenly, regardless of how crowded their surroundings are, hinting that internal mechanisms like black hole feedback, not environmental interactions, are the main cause.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claim of uniformly short quenching timescales rests on BAGPIPES point estimates without propagated uncertainties; if the Tq posterior widths are large, the 'uniform <500 Myr' result and the environmental-independence conclusion are not established.","rationale":"The paper combines two lines of evidence for its central claim: (i) a null correlation between environment tracers and quenched fraction, quenching duration, or timing, and (ii) a set of uniformly short quenching timescales (300–500 Myr) inferred from SED fitting. The environment null results are useful and are supported by Monte Carlo sensitivity tests, but the uniform Tq result is the part that most directly motivates the 'internal mechanisms dominate' conclusion, and it is also the least secure. Table A.1 reports no Tq values or uncertainties, and Figure 3 shows no error bars. The corner plots in Appendix C show that the double-power-law parameters are often loosely constrained, and the enormous sSFR uncertainties for four galaxies indicate that the photometry does not pin down their recent star formation. If the Tq posteriors are wide, then the 300–500 Myr clustering could be a byproduct of the chosen parametric SFH and priors rather than a shared physical quenching process. This concern is exactly what the reader identified as the weakest assumption, so the conditional verdict should stand: the paper should publish the Tq posterior distributions and the MQG catalog before the uniformity claim is treated as established. No ad hominem is implied; this is a standard model-dependence and uncertainty-propagation issue.","tokens_in":43309,"tokens_out":5910,"duration_ms":65316,"concrete_test":"For each of the 24 MQGs, draw ~10^3 posterior samples from the BAGPIPES chains (or refit with the same priors and save the chains), compute Tq = tquench − tform for each sample using the definitions in Eq. 5, and report the median and 16th–84th percentile interval. Then determine whether all, or at least 19, of the intervals overlap a common value <0.5 Gyr and whether the intervals for the four poorly constrained sSFR galaxies are prior-dominated. If the credible intervals are wider than ~0.5 Gyr or do not overlap a common short value, the 'uniform rapid quenching' claim fails and the environmental independence of Tq cannot be assessed from these data.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline conclusion that high-redshift quenching is 'uniformly short' (<500 Myr) and therefore internally driven rests on the 19/24 MQGs with Tq = 300–500 Myr shown in Figure 3. However, Tq is never tabulated with uncertainties. It is derived from BAGPIPES double-power-law fits whose corner plots (Appendix C) show broad or asymmetric posteriors for the falling slope α and turnover time τ for several galaxies; four sources have log sSFR = −41±19, −31±33, −31.8±19.4, −37.3±21.6 (Table A.1), i.e., the SED does not constrain recent star formation. If Tq is computed from the maximum-a-posteriori SFH rather than marginalized over the posterior, the reported 300–500 Myr clustering may be an artifact of the prior and model rather than a physical property. Because the paper uses the uniformity of Tq to argue for a common internal mechanism and to interpret the null environment correlations (Figs. 6–7), the missing posterior propagation is the most load-bearing weakness. A quantitative check is to compute the full posterior distribution of Tq for every galaxy; if the 68% credible intervals are wide (≳300 Myr) or overlap the 0.9 Gyr outlier, the uniformity claim is not supported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper identifies 24 massive quiescent galaxies (MQGs) at z ≈ 3.1 in the COSMOS field, using COSMOS2020 photometry and BAGPIPES SED fitting with a double power-law star-formation history. It constructs Voronoi-based density maps from two independent tracers (photo-z selected galaxies and ODIN Lyman-alpha emitters) and compares the MQGs' quenching timescales, quenched fractions, and distances to protoclusters and filaments with those of massive star-forming galaxies. The main claims are that quenching timescales are uniformly short (300–500 Myr for 19 of 24 objects) and independent of environment, implying that internal processes such as AGN feedback dominate over environmental quenching at this epoch.","tokens_in":43587,"tokens_out":7280,"duration_ms":68213,"significance":"If the result holds, the paper provides an observational constraint on quenching mechanisms at z ≈ 3.1, complementing simulation-based studies such as IllustrisTNG and Magneticum and extending the study of quiescent galaxies to the cosmic-noon epoch. The authors are careful in several respects: they check FIR/sub-mm non-detections, visually inspect SEDs, reject contaminants, use two independent density tracers, and run Monte Carlo simulations to assess the power of the Spearman correlation test. However, the central quantitative claim of uniformly short quenching timescales is presented without propagated uncertainties, and a few sample members appear not to satisfy the stated quiescence criterion, so the conclusion is not yet established at the level claimed.","major_comments":[{"comment":"The central claim that 19 of 24 MQGs have uniformly short quenching timescales (300–500 Myr) is presented without any uncertainty on Tq. Tq is a deterministic function of the BAGPIPES parameters, but the paper quotes only point estimates. Appendix C shows broad or asymmetric posteriors for α, β, and τ for several objects (e.g., IDs 310229, 341682, 779869, 962569), and Table A.1 lists four galaxies with log sSFR uncertainties of ±19 to ±33 dex. The authors should compute the full posterior distribution of Tq for each galaxy and report 68% credible intervals. If those intervals are wide or overlap the 0.9 Gyr outlier, the uniformity claim, and hence the argument for a common internal quenching mechanism, is not supported.","section":"§3.1, Eq. (5), Fig. 3"},{"comment":"Three objects included in the MQG sample appear to violate the stated quiescence criterion (sSFR + σ_sSFR ≤ 0.2/t_age). For IDs 383298, 779869, and 961549, the quoted log sSFR + σ_sSFR values are approximately -9.90, -9.20, and -9.93, respectively, all above the -10.01 threshold. The authors should either correct the criterion, re-evaluate these objects, or explicitly justify their inclusion. The size and composition of the final sample is the basis for all subsequent environmental comparisons, so this inconsistency is load-bearing.","section":"§2.5, Eq. (2), Table A.1"},{"comment":"The conclusion of no density dependence in the quenched fraction is stronger than the statistics support. The bootstrap Spearman coefficient for the COSMOS2020 map is 0.55 ± 0.4, and the LAE map gives -0.55 ± 0.23, which is not obviously 'no correlation'. The Monte Carlo simulations in Appendix A show that a weakly correlated input with ~25 galaxies yields a median coefficient near 0.6, so the test has limited power to distinguish a weak correlation from no correlation. A quantitative upper limit (for example, a confidence interval on the Spearman coefficient or a model comparison) should be reported before concluding that the environment is irrelevant to the quenched fraction.","section":"§3.3, Fig. 8, Appendix A"},{"comment":"The Anderson-Darling test p-value of ~0.25 is used to conclude that MQGs and MSFGs have the same distribution with respect to protoclusters. With only 24 MQGs the test has limited power, and a non-significant p-value does not quantify the similarity of the two distributions. The authors should provide a power analysis or a confidence interval on the difference in median distance to protoclusters, parallel to the Monte Carlo exercise already performed for the Spearman test.","section":"§3.2, Fig. 5"}],"minor_comments":[{"comment":"The Abstract states quenching timescales of ≤400 Myr while the Results and Conclusions state <500 Myr and Fig. 3 shows a median of ~350 Myr; the numbers should be made consistent.","section":"Abstract and §5"},{"comment":"Section 3.2 says 6 of 24 MQGs (25%) are in protocluster candidates, while Section 4 says 20%; the correct fraction should be stated consistently.","section":"§3.2 and §4"},{"comment":"The text says 9 of 24 MQGs are within 5 cMpc of a filament but later refers to 25% of MQGs; 9/24 is 37.5%, so one of these statements is incorrect.","section":"§3.5"},{"comment":"The caption references 'Equation 3.1' for the quenching timescale definition, but the definition appears as Eq. (5); the cross-reference should be corrected.","section":"Fig. 3 caption"},{"comment":"The expression '0.2yr/t_age' is garbled and the units are unclear; the threshold should be written with explicit units for t_age and the derivation of 10^-10.01 shown.","section":"Eq. (2)"},{"comment":"No quenching timescale or its uncertainty is listed; adding a Tq column with credible intervals would make Fig. 3 reproducible and is directly relevant to the paper's main claim.","section":"Table A.1"},{"comment":"There is an unresolved citation '(Park et al. 2024; ?)' in the discussion of AGN outflows; the reference should be completed.","section":"§4"}],"recommendation":"major_revision","confidential_remarks":"The manuscript fits the scope of Astronomy & Astrophysics and the environmental test is not circular: the density maps are built from independent or nearly independent galaxy samples, and the null result does not assume the conclusion. The major concern is the missing propagation of SED-fitting uncertainties into the quenching timescale and the small-sample statistics behind the environmental nulls. These are fixable with additional analysis, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read the Singh et al. paper. The genuinely new thing is the environmental null result: 24 massive quiescent galaxies in COSMOS, with densities traced two independent ways (LAEs and photo-z galaxies), and no correlation between quenched fraction, quenching timescale, or SFR peak and local density. I don't think that explicit null is in the literature yet. That is worth having.\n\nThe paper is careful in the right places: they check FIR contamination, visually inspect SEDs, run Monte Carlo simulations to see how much correlation power a sample of 24 actually has, and use Anderson-Darling tests for protocluster and filament distances. They also admit that spectroscopic redshifts are needed. That honesty counts.\n\nThe soft spots are the usual ones for this kind of work, but they matter because the central claim leans on them. The 'uniformly short' quenching timescale is a BAGPIPES point estimate with no propagated uncertainties. Table A.1 shows four galaxies with log sSFR values like -41 ± 19, meaning the photometry does not pin down their recent star formation at all. The corner plots in Appendix C show broad posteriors for the falling slope and turnover time for several objects. If the Tq posterior widths are large, the 300–500 Myr clustering is partly a prior/model artifact. To their credit, they acknowledge model dependence in Section 3.1, but the abstract and conclusions do not carry that caveat. The conclusion that internal mechanisms dominate follows from the absence of an environmental correlation, which is an interpretation, not a detection. With 24 objects, a null result is weak, and the 2D projected densities are noisy. The Monte Carlo power tests help, but they only show you could detect a strong correlation, not that you are sensitive to a weak one.\n\nThere are small internal inconsistencies: the abstract and conclusions say ≤400 Myr while the text says 300–500 Myr for 19/24; the Spearman coefficient for quiescent fraction vs density is quoted as 0.55 in one place and as 0.43/−0.55 in the Figure 8 captions. These are fixable but should be fixed.\n\nBottom line: this deserves a serious referee. It is a solid observational study with a new null result, but the strong claims about uniform quenching timescales and internal quenching dominance need tempering. The authors should release a catalog with Tq uncertainties, or at least the full posteriors. I'd send it to review expecting major revisions, not desk reject.","headline":"A careful null result on environment and quenching at z~3.1, undermined by unpropagated Tq uncertainties and small-sample fragility.","tokens_in":44242,"tokens_out":3238,"would_cite":true,"duration_ms":32397,"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":"This paper claims that massive galaxies at z≈3.1 stop forming stars in a uniformly short time, 300–500 Myr, and that this happens independently of their environment, pointing to internal processes such as AGN feedback rather than…","keywords":["galaxy quenching","massive quiescent galaxies","high-redshift galaxies","cosmic noon","environmental density","Voronoi tessellation","SED fitting","AGN feedback"],"falsifier":"Take spectra of the 24 MQGs to pin down redshifts and star-formation rates, then refit the star-formation histories with a non-parametric model: if the quenching timescales spread out or begin to correlate with local density, the uniformity claim collapses. A simpler check is to remove the four galaxies with unconstrained sSFR values (log sSFR ≈ −31 to −41) and see whether the environment correlations stay null; if those four are actually dusty star-forming galaxies, the sample may be too small to support the conclusion.","tokens_in":43093,"feed_emoji":"🔭","tokens_out":9340,"duration_ms":82919,"temperature":0.7,"pith_summary":"This paper sets out to determine whether the shutdown of star formation in massive galaxies at z≈3.1—an epoch about 2 billion years after the Big Bang—is imposed by the surrounding environment or by processes inside the galaxies themselves. Using 24 massive quiescent galaxies selected from the COSMOS2020 catalogue and two independent density tracers, it finds that quenching timescales are short and strikingly uniform, with 19 of 24 galaxies quenching in 300–500 Myr. It also finds no correlation between environmental density and quenching duration, quenched fraction, or quenching timing, and no preference for quiescent galaxies to sit in protoclusters or filaments compared with equally massive star-forming galaxies. A sympathetic reader would take this as evidence that internal mechanisms—AGN feedback, stellar feedback, virial shock heating, or morphological quenching—dominate over environmental mechanisms at cosmic noon, and that external gas removal is not required to explain the early quiescent population.","feed_headline":"Massive galaxies at z≈3 quench in under 500 Myr","feed_subtitle":"Uniform 300–500 Myr quenching times and a flat quiescent fraction point to internal feedback, not surroundings.","key_machinery":"The argument is carried by the quenching timescale $T_q = t_{\\rm quench} - t_{\\rm form}$, where $t_{\\rm form}$ is the star-formation-weighted formation time and $t_{\\rm quench}$ is the moment when $t\\,{\\rm SFR}(t)/M_{\\rm formed}$ drops below 0.1, both extracted by fitting a double power-law star-formation history with the BAGPIPES SED-fitting code to 29-band photometry. This parameter does the work of linking photometry to mechanism, because the paper's simulations literature associates short timescales (≈0.1 Gyr) with stellar and AGN feedback and long timescales (≈1 Gyr) with merger-driven quenching. On the environment side, the machinery is a Voronoi Monte Carlo density map built independently from Lyman-$\\alpha$ emitters and from photo-$z$-selected galaxies, with protocluster and filament catalogues taken from that LAE map, and Anderson–Darling tests used to compare quiescent and star-forming galaxy distributions.","core_discovery":"The central claim is that at z≈3.1 massive quiescent galaxies are quenched rapidly and uniformly, and that this quenching is independent of environment. The 24 MQGs share a common star-formation history shape: an intense early starburst followed by a fast decline, with a median quenching timescale around 350 Myr and most galaxies in the 300–500 Myr range. Their quenched fraction is flat across local density, their neighbor counts match the general galaxy population, and their distances to protoclusters and filaments are statistically indistinguishable from those of massive star-forming galaxies. The paper concludes that environmental processes alone—mergers, interactions, ram-pressure stripping, strangulation—cannot account for quenching at this epoch, and that internal processes such as AGN feedback, stellar feedback, virial shock heating, or morphological quenching play the leading role.","pith_inferences":["Editorial inference: the tight clustering of $T_q$ could partly reflect the double power-law prior; a non-parametric star-formation history refit would reveal how much of the uniformity is imposed by the model.","Editorial inference: the two density tracers used here may miss bound group-scale environments at z≈3, so group-scale halo mass could still influence quenching even though large-scale density does not.","Editorial inference: if internal quenching dominates at z≈3.1, the same analysis at the other ODIN redshift slices (z≈2.4 and z≈4.5) should show the same environmental independence—a testable prediction of the paper's interpretation.","Editorial inference: dropping the four MQGs with unconstrained sSFR values and repeating the correlation analysis would show whether the null environmental result survives on the securely quiescent subsample."],"forward_implications":["If the claim holds, theoretical models of galaxy formation at z≈3 must make internal feedback—AGN, stellar, or virial shock heating—the primary quenching channel for massive galaxies, rather than relying on environmental processes.","Surveys that search for high-redshift quiescent galaxies inside protoclusters will systematically miss most of the population, since the quiescent fraction is flat across density.","Simulations now have a quantitative target: reproduce a quenching timescale of 300–500 Myr that is independent of environment at cosmic noon.","The presence of quenched galaxies near gas-rich filaments, with no sign of rejuvenation, implies gas heating rather than gas exhaustion, and motivates searches for heated or ionized gas around these galaxies."],"supporting_citations":[{"why":"It introduces BAGPIPES and validates that double power-law SFHs recover stellar mass and quenching timescales with small bias, and the paper's $T_q$ definition follows this work.","marker":"Carnall et al. 2018"},{"why":"It supplies the BAGPIPES SED-fitting setup, priors, and the sSFR-based quiescent selection criterion used to build the MQG sample.","marker":"Carnall et al. 2023"},{"why":"It provides the COSMOS2020 photometric catalogue and LePhare photo-zs from which the z≈3.1 galaxy sample and MQGs are selected.","marker":"Weaver et al. 2021"},{"why":"It describes the ODIN survey and its narrowband LAE selection, the source of one of the two environmental density tracers.","marker":"Lee et al. 2024"},{"why":"It details the LAE selection in COSMOS and the ~97% redshift classification accuracy that fixes LAEs at z≈3.1 for the density map.","marker":"Firestone et al. 2024"},{"why":"It publishes the LAE Voronoi density map and the protocluster and filament catalogues used to measure the environment of MQGs.","marker":"Ramakrishnan et al. 2023"},{"why":"It provides the simulation result that mergers do not distinguish quiescent from star-forming galaxies and that AGN feedback is required, supporting the paper's internal-quenching interpretation.","marker":"Kurinchi-Vendhan et al. 2024"},{"why":"It provides the FIR/sub-mm catalogue used to confirm that MQG candidates lack significant dust-obscured star formation, ruling out dusty contamination.","marker":"Jin et al. 2018"}],"fun_headline_variants":["Quenching at z~3 is fast, uniform, and internal","Rapid quenching at z~3 ignores surroundings","At z~3, galaxy quenching is environment-independent","Internal processes quench z~3 galaxies in under 500 Myr","Cosmic noon quenching: quick and environment-free"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the mathematical model used to fit the galaxies' light correctly measures when their star formation stopped, so the uniform 300–500 Myr quenching times are real; this is fragile because four of the 24 galaxies have essentially unconstrained star-formation rates in the fit.","fun_headline_variants_meta":{"raw":{"variants":["Quenching at z~3 is fast, uniform, and internal","Rapid quenching at z~3 ignores surroundings","At z~3, galaxy quenching is environment-independent","Internal processes quench z~3 galaxies in under 500 Myr","Cosmic noon quenching: quick and environment-free"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0007,"raw_usage":{"total_tokens":3138,"prompt_tokens":900,"completion_tokens":2238,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":516,"completion_tokens_details":{"reasoning_tokens":2157}},"tokens_in":516,"tokens_out":2238,"duration_ms":17416,"temperature":1.0,"reasoning_tokens":2157,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T17:13:40.083086+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take spectra of the 24 MQGs to pin down redshifts and star-formation rates, then refit the star-formation histories with a non-parametric model: if the quenching timescales spread out or begin to correlate with local density, the uniformity claim collapses. A simpler check is to remove the four galaxies with unconstrained sSFR values (log sSFR ≈ −31 to −41) and see whether the environment correlations stay null; if those four are actually dusty star-forming galaxies, the sample may be too small to support the conclusion.","supporting_citations":[{"cited_title":"R., Kauffmann , O., Shuntov , M., et al","cited_arxiv_id":null,"evidence_quote":"It provides the COSMOS2020 photometric catalogue and LePhare photo-zs from which the z≈3.1 galaxy sample and MQGs are selected."},{"cited_title":"M., Gawiser, E., Ramakrishnan, V., et al","cited_arxiv_id":null,"evidence_quote":"It details the LAE selection in COSMOS and the ~97% redshift classification accuracy that fixes LAEs at z≈3.1 for the density map."},{"cited_title":"H., et al","cited_arxiv_id":null,"evidence_quote":"It publishes the LAE Voronoi density map and the protocluster and filament catalogues used to measure the environment of MQGs."},{"cited_title":"2024, , 534, 3974","cited_arxiv_id":null,"evidence_quote":"It provides the simulation result that mergers do not distinguish quiescent from star-forming galaxies and that AGN feedback is required, supporting the paper's internal-quenching interpretation."},{"cited_title":"2018, , 864, 56","cited_arxiv_id":null,"evidence_quote":"It provides the FIR/sub-mm catalogue used to confirm that MQG candidates lack significant dust-obscured star formation, ruling out dusty contamination."}],"review_version":1}