{"id":"201aa9f0-8f78-452e-8cf8-a86204ef6ae9","arxiv_id":"1908.06670","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"New CN and CH photometric indices applied to NGC 6723 show no radial gradient between its two stellar populations and reveal a possible abundance anomaly in one AGB group.","lead":"This paper introduces two new photometric indices that measure the strengths of the CN and CH molecular bands in globular cluster stars, and uses them to re-examine the stellar populations of the cluster NGC 6723. It finds the two populations share the same radial distribution, contradicting an earlier study, and hints at unusual carbon chemistry in one AGB population.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Radial-gradient refutation relies on an incompleteness correction the paper itself calls inexact, and the Monte Carlo calibration in Table 4 gives P=4.5% rather than the expected 0.2% under the null.","rationale":"The main result that would remove a claimed counterexample to the standard formation scenario is the identical radial distributions. The reader's weakest assumption (Gaia DR2 membership and completeness) is exactly the place where this result is least secure. I agree with that, and add one quantitative observation: the paper's own Monte Carlo calibration, Table 4, reports P(p≤0.002)=4.46% for the 'Empirical' model, whereas under a correctly calibrated null the rate should be 0.2%. The text's conclusion that Lim et al.'s result is 'highly improbable' is therefore not supported by the cited number; the 4.46% entry is closer to a marginal fluctuation than to a strong refutation. Together with the explicit admission in Section 6.3.2 that the completeness correction was applied in 'not an exactly correct way,' this makes the no-gradient claim conditional rather than established. The paper does have real independent support for other results: the RGB number ratio agrees with Milone et al. (2017), and the cnJWL index correlates strongly with δS(3839) in M3. I see no reason to reject the paper outright, but the radial-gradient refutation should be accepted only after an independent reanalysis of the Lim et al. catalog with a proper completeness treatment. Therefore I leave the reader's CONDITIONAL verdict unchanged.","tokens_in":18835,"tokens_out":5752,"duration_ms":63401,"concrete_test":"Independently reanalyze the public Lim et al. (2016) catalog: assign CN-w/CN-s tags with the author's hkCTIO-calibrated classifier on the same stars, apply Gaia DR3 membership and a radial-binned completeness correction that is recomputed separately for CN-w and CN-s from artificial-star tests, then recompute the cumulative radial distributions and K-S p-values. If the two distributions remain indistinguishable after this treatment, the no-segregation claim stands; if the reversed gradient reappears, the refutation of Lim et al. is an artifact of the author's reduction, membership, or completeness choices.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the absence of radial segregation in NGC 6723, which is used to refute Lim et al. (2016). That claim is only as strong as the least secure link in the chain: the Gaia DR2 proper-motion membership selection and the completeness treatment for the dense, bulge-contaminated central field. Section 6.3.2 states openly that the completeness correction is approximate: 'our approach applying the completeness fractions to the observed radial distributions is not an exactly correct way to realize the effect of incomplete detection... the inverse process... delicate.' Because the cluster is at l=0.07, b=-17.30 with heavy field contamination, any differential incompleteness or misclassification as a function of radius between CN-w and CN-s would produce exactly the observed flat cumulative distributions even if a real reversed segregation existed. The paper does not demonstrate that the completeness fractions are identical for the two populations, since fcomplete is derived from artificial star tests on all stars, not separately for CN-w and CN-s. In addition, Table 4 is internally troubling: under the null hypothesis the probability of obtaining Lim et al.'s p-value ≤ 0.002 should be 0.2%, yet the 'Empirical' Monte Carlo gives 4.46%—more than twenty times larger. The text says the results show Lim's outcome is 'highly improbable,' but 4.46% is not a strong refutation and indicates the simulations are not calibrated to the true null distribution. The refutation also never re-analyzes Lim et al.'s published catalog; it relies on the author's own reduction, which can differ in membership, photometric zero points, and classification. Because the absence of a gradient is a null result, K-S tests cannot prove identical distributions, and the power of the test is not reported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper, the fifth in the author's series on homogeneous Ca-CN-CH photometry, introduces two new photometric indices, cn'_JWL = Ca_CTIO - Ca_JWL and ch_JWL = (JWL43 - b) - (b - y), intended as accurate measures of the CN lambda 3883 band and the CH G band. Using M5 RGB stars, the author shows that ||cn'_JWL correlates well with ||cn_JWL (rho = 0.918) and that the photometric CN-CH relation between the CN-w and CN-s populations is step-like rather than continuous. Applying the indices to NGC 6723, the paper reports an RGB number ratio n(CN-w):n(CN-s) = 35.5:64.5 +/- 2.8, no radial gradient between the two RGB populations (contradicting Lim et al. 2016), Delta Y = 0.012 +/- 0.012 for the CN-s population from the RGB bump, and discrete double AGB populations whose bright-star ratio is marginally consistent with the RGB ratio. Synthetic spectra reproduce the RGB sequences but leave a 0.035 mag discrepancy for the CN-w AGB, which is attributed to a possible mild nitrogen enhancement and/or a large 12C/13C decrement.","tokens_in":19147,"tokens_out":7262,"duration_ms":72702,"significance":"If the no-radial-gradient result holds, the paper removes an important claimed counterexample to the standard formation scenario for globular-cluster multiple populations and demonstrates that a ground-based Ca-CN-CH system can tag populations reliably. The paper's strengths include agreement of the RGB ratio with the independent HST measurement of Milone et al. (2017), consistency with the Milone et al. (2018) helium estimate, and an explicit, transparent discussion of the limitations of the completeness correction and of the adopted CNO abundances. The central refutation of Lim et al. (2016), however, rests on a Monte Carlo calibration that is internally inconsistent and on a completeness treatment that is not population-specific, so the quantitative support for the headline claim is currently weaker than the text states.","major_comments":[{"comment":"The calibration of the Monte Carlo test is internally inconsistent. Under the null hypothesis, a two-sample K-S p-value is uniform on [0,1], so P(p <= 0.002) should be 0.2%. The 'Empirical' entry of 4.46% is more than twenty times larger, while 'Empirical1' and 'Gaussian' give 0.18% and 0.20%. This indicates that the empirical simulation does not sample a correctly pooled null distribution, so the statement in Section 7 that Lim et al.'s result has less than 5% probability and is 'highly improbable' is not supported by the reported numbers. Please report the actual K-S statistics and p-values for the comparisons in Figure 13 and rerun the randomization under a properly defined null hypothesis.","section":"Section 6.3.2, Table 4"},{"comment":"The text states that K-S tests show the CN-w and CN-s populations are most likely drawn from the same parent distribution, but no test statistic or p-value is reported for any of the three color indices. The reader cannot distinguish p approximately 0.04 from p approximately 0.4, which matters because the paper's central claim is the absence of radial segregation. Please tabulate D and p for ||cn_JWL, ||cn'_JWL, and ||hk_CTIO, together with sample sizes and the radial ranges used.","section":"Section 6.3.2, Figure 13"},{"comment":"The completeness correction used to argue that incomplete detection cannot hide a radial gradient is not population-specific. The artificial-star completeness fractions f_complete are computed for the full sample, while the relevant question is whether CN-w and CN-s stars are detected and retained at different rates as a function of radius in this bulge-contaminated field (l = 0.07 deg, b = -17.30 deg) with Gaia DR2 proper-motion membership incomplete in the center. The paper itself notes that applying f_complete to the observed radial distributions is not exactly correct and that the inverse process is delicate. Please test the pipeline's ability to recover an artificially injected reversed gradient of the kind claimed by Lim et al. (2016), or provide per-population completeness and membership-retention curves with radial uncertainties.","section":"Sections 6.1 and 6.3.2"},{"comment":"The inferred need for extra nitrogen enhancement and/or a large 12C/13C decrement in the CN-w AGB is conditional on adopting the CNO abundances of M5 for NGC 6723 and on the specific ATLAS12/SYNTHE model assumptions. Since no CNO abundances for NGC 6723 are available, the 0.035 mag discrepancy could instead reflect incorrect input abundances, model atmospheres, or the adopted isochrones. The abstract states this as a result; please either soften the claim to a model-dependent speculation or obtain spectroscopic CNO constraints for at least a few NGC 6723 giants.","section":"Section 6.4.1, Figure 16"}],"minor_comments":[{"comment":"The phrase 'one of the most greatest achievements' contains a grammatical error; it should be 'one of the greatest achievements'.","section":"Section 1"},{"comment":"The fourth panel is labeled '(c)' twice; the label for the final panel should be '(d)'.","section":"Figure 14 caption"},{"comment":"'Mote Carlo simulations' should be 'Monte Carlo simulations'.","section":"Figure 14 caption"},{"comment":"'the second Gaia date release' should be 'the second Gaia data release'.","section":"Section 4"},{"comment":"The text adopts [Fe/H] approximately -1.0 dex but the footnote gives [Fe/H] = -0.93 +/- 0.05 from Crestani et al. (2019); please state explicitly which value is used in the synthetic models and whether the difference affects the conclusions.","section":"Section 6.4.1"},{"comment":"The in-text citation 'Zachairias et al. (2004)' is misspelled; it should be 'Zacharias et al. (2004)' to match the reference list.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"I have no concerns about novelty or citation behavior; the paper is a direct continuation of the author's series and makes appropriate use of external comparisons. The main risk is that the headline refutation of Lim et al. (2016) is currently supported by an uncalibrated Monte Carlo test and a non-population-specific completeness correction. If those issues are fixed, I would expect the paper to be publishable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a careful read, but the headline result is stronger than the statistics actually support. The central claim—that the two RGB populations in NGC 6723 have the same radial distribution—is credible and independently supported by Milone et al.'s HST number ratio, but the refutation of Lim et al. rests on the author's own reduction plus a Monte Carlo that returns P=4.5% for Lim's p-value under the null, not the 0.2% you'd expect from a well-calibrated test. \"Highly improbable\" is an overstatement.\n\nWhat's genuinely new and good: the cn'JWL index as a cheap surrogate for cnJWL, the chJWL index for the CH G band, and the photometric CN-CH step function in M5. These are useful additions to the population-tagging toolkit. The paper also does a real service by showing that hkCTIO—the exact index Lim used—is a poor CN tracer with continuous population transitions, which plausibly explains the earlier result. The synthetic RGB models match observations well, and the AGB discrepancy discussion is honest about adopting CNO abundances from M5 rather than measuring them.\n\nSoft spots, in proportion: (1) The actual KS p-values for the flat radial distributions are never reported, so \"most likely drawn from same parent\" is unquantified. (2) The Table 4 calibration problem is real; the author includes a better-calibrated \"Empirical1\" row (0.18%) but does not explain why the main empirical row gives 4.46%. (3) The completeness correction is openly described as not exactly correct and \"delicate,\" yet it is applied after the fact to both populations jointly; differential incompleteness by population and radius would not be captured. (4) The bright AGB number-ratio agreement uses a magnitude cut chosen after the full AGB sample disagreed. (5) No data or code, which limits independent reproducibility.\n\nThis is a paper for people working directly on globular cluster multiple populations, especially on photometric indices. It deserves peer review—the observations, external spectroscopic calibration, and the direct comparison to Lim's index make it legitimate—but a referee should ask for quantified KS results, a properly calibrated null, and more cautious wording in the abstract and summary.","headline":"A useful photometric-tool paper whose refutation of Lim et al. on NGC 6723 is probably correct but statistically oversold.","tokens_in":19805,"tokens_out":3501,"would_cite":true,"duration_ms":36483,"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":"This paper introduces a surrogate CN photometric index, cn′JWL, and uses it to show that NGC 6723's two red-giant populations share the same radial distribution — overturning a claimed counterexample to globular-cluster…","keywords":["Hertzsprung-Russell and C-M diagrams","stars: abundances","stars: evolution","stars: Population II","globular clusters: individual (NGC 6723)","multiple stellar populations","CN photometry","radial distributions"],"falsifier":"Reclassify NGC 6723 giants using cn′JWL photometry but with membership and completeness from HST proper motions, which cover the cluster center where Gaia DR2 is incomplete; if the CN-w population then appears more centrally concentrated than CN-s, the paper's refutation of the earlier counterexample fails. A cheaper check is high-resolution spectroscopy of stars tagged CN-w in the inner region: if their measured CN strengths disagree with the photometric tags, cn′JWL is not a reliable population tracer.","tokens_in":18570,"feed_emoji":"🔭","tokens_out":15124,"duration_ms":125735,"temperature":0.7,"pith_summary":"Star clusters born in a single burst should show one stellar population, yet most globular clusters show two or more; the usual explanation is that later-generation stars formed from ejecta enriched in nitrogen and helium. This paper introduces a new photometric index, $cn'_{\\rm JWL} = Ca_{\\rm CTIO} - Ca_{\\rm JWL}$, as a surrogate for the validated CN-band tracer $cn_{\\rm JWL}$, and uses it to re-examine NGC 6723, which a 2016 study had put forward as a counterexample because its CN-weak giants appeared more centrally concentrated than its CN-strong giants. With the new index, the two red-giant populations in NGC 6723 have statistically identical cumulative radial distributions and a number ratio $n({\\rm CN\\text{-}w}):n({\\rm CN\\text{-}s}) = 35.5:64.5 \\pm 2.8$. The paper concludes that the earlier counterexample was an artifact of a poor CN tracer, and that NGC 6723 behaves like normal globular clusters, with a slightly helium-enriched CN-strong population inferred from the red-giant-branch bump, $\\Delta Y = 0.012 \\pm 0.012$.","feed_headline":"A better CN index erases a globular-cluster counterexample","feed_subtitle":"Both RGB populations in NGC 6723 share the same radial distribution; the claimed reversal was a tagging artifact.","key_machinery":"The load-bearing instrument is the photometric index $cn'_{\\rm JWL} = Ca_{\\rm CTIO} - Ca_{\\rm JWL}$. Because the CTIO calcium filter's passband has aged into a shape close to the custom JWL39 filter, this difference reproduces the information in $cn_{\\rm JWL} = JWL39 - Ca_{\\rm JWL}$, making accurate CN-band ($\\lambda3883$) tagging possible without the custom filter. Its partner, $ch_{\\rm JWL} = (JWL43 - b) - (b - y)$, measures the CH G band at $\\lambda4250$, and the parallelized indices $\\|cn_{\\rm JWL}$ and $\\|cn'_{\\rm JWL}$ (sequences aligned to red and blue RGB fiducials) perform the actual population separation. Classification uses an expectation-maximization two-component Gaussian mixture, membership comes from Gaia DR2 proper motions, and completeness is estimated from artificial-star experiments. Synthetic spectra from ATLAS12/SYNTHE and MOOG are used to check whether abundance differences can account for the observed RGB and AGB colors.","core_discovery":"On its own terms, the paper's central discovery is that the color difference $cn'_{\\rm JWL} = Ca_{\\rm CTIO} - Ca_{\\rm JWL}$ is a reliable surrogate for the previously validated CN-band index $cn_{\\rm JWL} = JWL39 - Ca_{\\rm JWL}$, and that this index, combined with a new CH measure $ch_{\\rm JWL} = (JWL43 - b) - (b - y)$, cleanly separates the multiple populations of NGC 6723. For Gaia DR2 proper-motion members, the two RGB populations are classified with an expectation-maximization two-component Gaussian mixture, giving $n({\\rm CN\\text{-}w}):n({\\rm CN\\text{-}s}) = 35.5:64.5 \\pm 2.8$; their cumulative radial distributions are statistically indistinguishable. That result contradicts the earlier claim of a centrally concentrated CN-w population, which the paper traces to misclassification by the $hk_{\\rm CTIO}$ index, a poor CN tracer with weak CH contamination. The same data show discrete double AGB sequences, a CN-s RGB bump brighter by $0.031 \\pm 0.030$ mag (translating to $\\Delta Y = 0.012 \\pm 0.012$), and a step-like photometric CN--CH anticorrelation in M5. Synthetic spectra reproduce the RGB sequences but not the CN-w AGB, suggesting a mild nitrogen enhancement or a large drop in $^{12}$C/$^{13}$C in those AGB stars.","pith_inferences":["If the surrogate index passes independent validation, archival CaCTIO photometry of many clusters could be re-processed in the same way, turning claimed radial-segregation anomalies into a systematic survey without new observations.","The step-like photometric CN–CH relation in M5 suggests that the continuous abundance spreads reported in high-resolution Na–O studies may partly reflect measurement scatter; a joint photometric–spectroscopic analysis could test whether the underlying populations are genuinely discrete.","The RGB/AGB mismatch in synthetic colors points to phase-dependent abundance changes in first-generation stars that can be tested directly with high-resolution spectra of a handful of CN-w AGB stars in NGC 6723.","A center-complete reobservation with HST proper motions would decide whether Gaia DR2 incompleteness hid a real gradient; if the null result survives, other radial-segregation claims should be rechecked with the same index."],"forward_implications":["The NGC 6723 reversed radial distribution claimed in the 2016 study disappears when stars are tagged with a clean CN index, so this cluster no longer counts as an observational counterexample to standard multiple-population formation.","The measured RGB ratio of 35.5:64.5 agrees with the independent HST-based first-generation fraction of 0.363 ± 0.017, supporting a flat population ratio that does not change with radius.","The CN-s RGB bump is brighter by 0.031 ± 0.030 mag, implying a modest helium enhancement of ΔY = 0.012 ± 0.012 for the later generation.","The cnJWL and cn′JWL CMDs reveal discrete double AGB populations, and the bright AGB number ratio is marginally consistent with the RGB ratio; the deficit of faint CN-s AGB stars suggests helium-enhanced stars may evolve as AGB-manqué.","Because cn′JWL requires only the existing CTIO calcium filter, archival photometry of other globular clusters can be re-analyzed with the same population-tagging procedure."],"supporting_citations":[{"why":"The prior claim of a reversed radial distribution in NGC 6723 that this paper re-examines and refutes; the baseline counterexample.","marker":"Lim et al. 2016"},{"why":"Independent HST-based first-generation fraction of 0.363 ± 0.017 that matches the paper's RGB population ratio.","marker":"Milone et al. 2017"},{"why":"HST-based helium measurement and the membership-selection approach used for comparison.","marker":"Milone et al. 2018"},{"why":"Spectroscopic CH(4300) and δS(3839) measurements used to validate chJWL and to show cnJWL's tight correlation.","marker":"Smolinski et al. 2011"},{"why":"Defines the M5 CN–CH anticorrelation and supplies the CNO abundances adopted for synthetic-population models.","marker":"Cohen et al. 2002"},{"why":"Gaia DR2 astrometry used to select cluster members by proper motion.","marker":"Brown et al. 2018"},{"why":"Defines and validates the cnJWL index and the parallelization scheme on which the surrogate cn′JWL is based.","marker":"Lee 2019a"},{"why":"Provides the empirical Monte Carlo models used to show the radial distributions are statistically consistent with identical parent distributions.","marker":"Lee 2019b"}],"fun_headline_variants":["Surrogate CN index erases globular cluster counterexample","No radial gradient for NGC 6723's twin star populations","New CN surrogate index reveals clean split in NGC 6723","Globular cluster NGC 6723's odd radial split was a fluke","Double AGB populations confirmed in globular cluster NGC 6723"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that Gaia DR2 proper-motion membership and the artificial-star completeness corrections do not remove CN-w and CN-s stars at different rates as a function of radius; if the cluster center is incomplete in a population-dependent way, the observed identical radial distributions could be a sample artifact rather than a property of the cluster.","fun_headline_variants_meta":{"raw":{"variants":["Surrogate CN index erases globular cluster counterexample","No radial gradient for NGC 6723's twin star populations","New CN surrogate index reveals clean split in NGC 6723","Globular cluster NGC 6723's odd radial split was a fluke","Double AGB populations confirmed in globular cluster NGC 6723"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000739,"raw_usage":{"total_tokens":3451,"prompt_tokens":1249,"completion_tokens":2202,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":865,"completion_tokens_details":{"reasoning_tokens":2112}},"tokens_in":865,"tokens_out":2202,"duration_ms":15716,"temperature":1.0,"reasoning_tokens":2112,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:38:20.144578+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Reclassify NGC 6723 giants using cn′JWL photometry but with membership and completeness from HST proper motions, which cover the cluster center where Gaia DR2 is incomplete; if the CN-w population then appears more centrally concentrated than CN-s, the paper's refutation of the earlier counterexample fails. A cheaper check is high-resolution spectroscopy of stars tagged CN-w in the inner region: if their measured CN strengths disagree with the photometric tags, cn′JWL is not a reliable population tracer.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The prior claim of a reversed radial distribution in NGC 6723 that this paper re-examines and refutes; the baseline counterexample."},{"cited_title":"P ., Piotto, G., Renzini, A","cited_arxiv_id":null,"evidence_quote":"Independent HST-based first-generation fraction of 0.363 ± 0.017 that matches the paper's RGB population ratio."},{"cited_title":"P ., Marino., A","cited_arxiv_id":null,"evidence_quote":"HST-based helium measurement and the membership-selection approach used for comparison."},{"cited_title":"P ., Martell, S., Beers, T","cited_arxiv_id":null,"evidence_quote":"Spectroscopic CH(4300) and δS(3839) measurements used to validate chJWL and to show cnJWL's tight correlation."},{"cited_title":"G., Briley, M","cited_arxiv_id":null,"evidence_quote":"Defines the M5 CN–CH anticorrelation and supplies the CNO abundances adopted for synthetic-population models."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gaia DR2 astrometry used to select cluster members by proper motion."}],"review_version":1}