{"id":"0fd42d9b-9e2d-46f9-8ffe-0f0fcfea5c53","arxiv_id":"2502.09582","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"Asteroseismic ages of individual red giants in the same cluster scatter far more than expected and are systematically offset from isochrone-based cluster ages across all seven clusters studied.","lead":"A meta-analysis of published asteroseismic ages for red giants in seven star clusters finds that ages of individual stars in the same cluster disagree wildly and often differ from the cluster's independently measured age. The result suggests asteroseismic ages used for galactic archaeology may need corrections, or that binary stars are confusing the measurements.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Within-cluster scatter is compared to cluster-age precision, not to per-star asteroseismic age uncertainties, so the claimed tension is not actually tested.","rationale":"The reader's weakest assumption focused on heterogeneity across papers and the lack of error propagation. My concern is broader and more fundamental: even within a single homogeneous pipeline, the paper compares intra-cluster scatter to cluster-age precision (5.4% M92, 10% model scatter) rather than to the per-star asteroseismic age uncertainties, which are expected to be much larger. This is a load-bearing flaw because it directly undermines the central claim that the scatter 'far surpasses' the relevant uncertainty. The reader's verdict of CONDITIONAL already requires major revisions, including error propagation and comparison against per-star uncertainties. My concern reinforces that requirement without changing the verdict. I set UNCHANGED because the reader's recommendation (accept with major revision, conditional on a rigorous reanalysis) is the appropriate response; the paper's transparent data and useful call for homogeneous calibration are still valuable, but the central claim is not yet established. The proposed concrete test would settle the matter: if the observed scatter is consistent with propagated per-star uncertainties, the 'tension' evaporates; if not, the tension is real and the paper's suggestion of binary-induced mass alteration or additional corrections becomes credible.","tokens_in":20426,"tokens_out":6085,"duration_ms":64050,"concrete_test":"For a homogeneous subset (e.g., NGC 6791 and NGC 6819 from Pinsonneault et al. 2024), recover the per-star covariances of mass, logg, Teff, and [M/H], and propagate them through the same age grid via Monte Carlo (10^4 draws) to obtain per-star age distributions. Compute the observed dispersion of log10(age) from Table 2 and the expected dispersion from the Monte Carlo mixture, then test the null hypothesis that all stars share the cluster age using a chi-squared or likelihood-ratio statistic. Repeat for M4 and M9 using uncertainties from Howell et al. If the observed scatter is within 2 sigma of the propagated expectation for all clusters, the central claim fails; if significantly larger, the tension is real.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim in the abstract and Section 4 is that the scatter among asteroseismic ages for individual stars in any one cluster 'far surpasses' both the 5.4% absolute age uncertainty of M92 and the ~10% model-to-model isochrone scatter. But those baselines are not the expected dispersion of asteroseismic ages for individual stars within a cluster. The paper explicitly declines to propagate the per-star uncertainties (Section 2: 'we do not quote the individual uncertainties reported internally for each asteroseismic target here'; Section 3: 'we do not report uncertainties on each column here'). A red-giant age derived from mass, Teff, and logg via a grid is highly nonlinear: a 5-10% uncertainty in mass (typical for scaling-relation masses) can easily translate into a 20-50% uncertainty in age for low-mass, old stars. The Table 2 intra-cluster ranges (e.g., M4 ages 6.8-43 Gyr; M9 ages 4.0-34 Gyr) could therefore be entirely consistent with the current measurement errors. The comparison to 5.4% and 10% is misleading because those numbers describe the precision of cluster isochrone ages, not the precision of individual seismic ages. Until the per-star age uncertainties are propagated and compared to the observed scatter, the claim that there is a 'tension' between the asteroseismic and cluster age scales is unsupported. The alternative explanation that the scatter reflects underestimated per-star uncertainties or inter-paper method differences remains viable and is not ruled out by any test in the paper.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper compiles published asteroseismic masses, surface gravities, metallicities, and effective temperatures for first-ascent red giants in seven open and globular clusters (NGC 6819, NGC 6791, M67, M4, M19, M80, M9), converts these to individual stellar ages where needed using the Tayar et al. (2017) grid, and compares the resulting per-star ages and cluster medians with isochrone-based cluster ages from the literature. The authors report that the within-cluster dispersion of asteroseismic ages far exceeds both the absolute age uncertainty of the reference cluster M92 (5.4%) and a roughly 10% model-to-model isochrone systematic, and they argue that this points either to binary-driven mass changes or to missing corrections in the asteroseismic age scale. The paper also compares the age–metallicity coverage of clusters with seismic detections to that of the field population and calls for homogeneous future calibration efforts.","tokens_in":20627,"tokens_out":7005,"duration_ms":68994,"significance":"If established, the claimed broad inconsistency between asteroseismic ages of individual red giants and independent cluster isochrone ages would matter directly for the calibration of large asteroseismic surveys (Kepler, K2, TESS, PLATO). The compilation in Table 2 and the paper's emphasis on the heterogeneity of existing analyses are useful resources, and the call for a homogeneous reanalysis is well motivated. However, the central quantitative claim is not yet tested, because the observed scatter is compared with cluster-age precision rather than with the propagated per-star uncertainties of the seismic ages. A revision that includes realistic per-star age uncertainties, a formal statistical comparison, and independent cluster ages for all seven systems could make this a valuable contribution.","major_comments":[{"comment":"The central quantitative claim compares the observed scatter of asteroseismic ages within each cluster to the 5.4% absolute age uncertainty of M92 (Ying et al. 2023) and to a ~10% model-to-model isochrone systematic. These are not the appropriate baselines for the dispersion of individual asteroseismic ages. For low-mass, old red giants, a 5–10% uncertainty in the scaling-relation mass translates into a substantially larger, typically 20–50%, uncertainty in age, and the per-star age uncertainties are not propagated anywhere in the paper (Sections 2 and 3 explicitly decline to report them). Table 2 shows within-cluster spans such as 6.8–43 Gyr for M4 and 4.0–34 Gyr for M9; without an estimate of the expected per-star scatter, those spans cannot be shown to 'far surpass' the measurement noise. The claim of a tension between the asteroseismic and cluster age scales is therefore not actually tested.","section":"Section 4, Tables 1 and 2"},{"comment":"The conclusion that there are 'systematic offsets' between asteroseismic and cluster age scales is based on a visual comparison of medians to shaded bands of 6% and 10%, without a statistical test or the number of stars per cluster. For M67 and NGC 6791 the median offsets are stated to be consistent with the isochrone uncertainties, and for NGC 6819 the offset is driven in part by an extreme outlier (KIC 5024272, age 0.26 Gyr, mass 3.49 solar masses) that is very likely not a single first-ascent red giant. A quantitative analysis that removes or models such contaminants and propagates the method-to-method differences (which the authors themselves identify in Section 2 as potentially dominating) is needed before claiming a broad-scale discrepancy.","section":"Section 4"},{"comment":"The abstract and introduction describe the comparison as being against 'independent, isochrone-based' cluster age determinations. However, for three clusters — M80, M19, and M9 — the adopted cluster ages (13, 12, and 13 Gyr) are taken from the same papers (Howell et al. 2024; Howell et al. 2025) that provide the asteroseismic masses and gravities used to compute the individual seismic ages. The cluster-age estimates are therefore not independent of the seismic data for these objects, and the comparison is partially circular. Please replace these ages with independent isochrone-based determinations or explicitly restate the claim as a comparison with ages taken from the same source.","section":"Section 3, Table 1"},{"comment":"The ~10% model-to-model systematic uncertainty used as a baseline is attributed to 'L. Morales et al., in prep.' and to Pinsonneault et al. (2024). The in-prep reference is not publicly available, so this load-bearing quantitative assumption cannot be checked. Similarly, the 5.4% figure for M92 is used as a generic absolute age-uncertainty floor for all seven clusters, although M92 is a metal-poor globular cluster and is not in the sample. The paper should either provide a reproducible calculation of these baselines or adopt published, cluster-specific uncertainties.","section":"Sections 2 and 4"}],"minor_comments":[{"comment":"The caption says the orange band represents the 'asteroseismic age determination (Joyce et al. 2023; Pinsonneault et al. 2024)', but the text in Section 4 says the orange band represents model-to-model isochrone scatter; please make the two descriptions consistent.","section":"Figure 1 caption"},{"comment":"Table 2 includes two NGC 1817 stars even though Section 3 states that these stars were excluded from the figures; please clarify whether these entries are meant to be part of the analysis or are shown only for completeness.","section":"Table 2 and Section 3"},{"comment":"The M9 data are artificially shifted to 13.2 Gyr instead of the adopted cluster age of 13 Gyr; this should be flagged in the main text as well as the caption so readers do not interpret the shift as a data point.","section":"Figure 1"},{"comment":"Several entries have age values of -99 (e.g., M80 8000009, M4 40000229, M9 90000210 and 9000081); please define this placeholder for missing or invalid ages in the table notes.","section":"Table 2"},{"comment":"The statements that the asteroseismic dispersion exceeds the model-to-model systematics 'by a factor of at least 5' would be more informative if the number of stars per cluster and the sampling uncertainty of the 16th–84th percentile range were reported.","section":"Section 4"},{"comment":"The DOI for Poretti et al. (2015) appears truncated (10.1007/978-3-319-10993-011); please correct it.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper relies on an in-prep reference for a load-bearing quantitative baseline (the ~10% model-to-model systematic), and for three clusters it adopts cluster ages from the same papers that provide the seismic parameters. The editor may wish to verify the availability of the in-prep work and to consider whether the framing as a demonstrated 'tension' is too strong relative to the evidence presented in the manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis paper is worth a read as a cautionary meta-analysis, but the headline claim goes beyond what the data support. The authors assemble published asteroseismic ages for red giants in seven clusters and show that the scatter within clusters is large and that median ages sometimes sit far from isochrone-based cluster ages. That systematic demonstration across multiple clusters is new; previous work flagged individual clusters. The data table is transparent, and the galactic-context figure makes a fair point: the clusters we can calibrate on are sparse in the age-metallicity plane of the field population. The call for a homogeneous reanalysis with a single pipeline is reasonable and timely.\n\nThe soft spot is the quantitative comparison itself. The paper compares within-cluster scatter to the 5.4% absolute age uncertainty of M92 and to roughly 10% model-to-model isochrone scatter. Those are not the per-star asteroseismic age uncertainties. For a low-mass red giant, a few percent in mass can translate to tens of percent in age. The authors explicitly do not propagate the individual error bars (Sections 2 and 3), so the observed scatter (e.g., M4 ages from 6.8 to 43 Gyr) may be entirely consistent with the measurement errors. Until the per-star uncertainties are propagated and compared to the scatter, the claim that asteroseismic and cluster age scales are in tension is not actually tested. The stress-test note lands.\n\nTwo smaller issues. Two NGC 1817 stars with about 200% offsets are excluded on the grounds that the evolutionary state or the measurements must be wrong; that is post hoc and should either be defended with evidence or restored. And the analysis mixes ages from different pipelines, solar references, and model grids; the paper acknowledges this, but it means the scatter might be methodological rather than astrophysical.\n\nHaving said that, the paper is honest about its limitations and the field does need this kind of audit. With a revision that propagates uncertainties, uses the right baselines, and either includes or robustly justifies the excluded stars, this could be a solid contribution.\n\nRecommendation: send to peer review. The question is important, the compilation is useful, and the flaws are fixable in revision rather than fatal. I would not cite the current version for the tension claim, but I would want to cite a revised version that puts the numbers on firmer statistical ground.","headline":"A useful and transparent meta-analysis of published asteroseismic cluster ages, but the headline tension claim is not actually tested because within-cluster scatter is compared to cluster-age precision rather than per-star age uncertainties.","tokens_in":21217,"tokens_out":2168,"would_cite":false,"duration_ms":20348,"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":"Asteroseismic ages for red giants disagree with cluster ages by far more than uncertainties allow.","keywords":["asteroseismology","red giants","stellar ages","star clusters","isochrones","scaling relations","open clusters","globular clusters"],"falsifier":"A homogeneous reanalysis of the same cluster stars using a single asteroseismic pipeline with fixed solar reference values and one model grid would settle the matter: if the within-cluster age dispersion drops to the level of the quoted uncertainties, the reported tension is a methodological artifact rather than a fundamental offset between the asteroseismic and isochrone age scales. Alternatively, measuring asteroseismic ages for stars in a cluster with a precisely known geometric distance and independent mass constraints (e.g., an eclipsing binary) would test whether the mass–age relation underlying the ages is biased.","tokens_in":20144,"feed_emoji":"⭐","tokens_out":2246,"duration_ms":186729,"temperature":0.7,"pith_summary":"This paper compares asteroseismic ages of individual red-giant stars in seven well-studied open and globular clusters with the ages of their host clusters derived from isochrone fitting. It finds that the scatter among asteroseismic ages within any one cluster is far larger than both the best-case absolute age uncertainty for a cluster (about 5.4 percent) and the model-to-model systematic spread in isochrone ages (roughly 10 percent). For several clusters, the median asteroseismic age is also systematically offset from the isochrone-based cluster age. The authors conclude that the asteroseismic age scale for red giants is not consistent with cluster-based ages at the few-percent level, and that either binary processes are altering the masses of cluster stars or additional corrections are needed before asteroseismic ages can be used as reliable absolute ages.","feed_headline":"Star ages from pulsations clash with cluster ages","feed_subtitle":"Scatter among asteroseismic ages in the same cluster far exceeds the age scale's expected uncertainties.","key_machinery":"The central comparison is between two independent age scales: asteroseismic ages of individual red giants, derived from global oscillation properties via scaling relations and model grids, and isochrone-based cluster ages, derived from fitting color–magnitude diagrams. The paper evaluates the scatter (16th–84th percentile range) of asteroseismic ages within each cluster and the offset of the cluster median from the adopted isochrone age, against quoted uncertainty budgets for the cluster age scale and model-to-model systematics.","core_discovery":"Using ages for individual first-ascent red giants in NGC 6819, NGC 6791, M67, M4, M19, M80, and M9, the authors show that the dispersion among asteroseismic age estimates for stars in the same cluster exceeds the expected uncertainty of the cluster age scale by factors of at least five in all but one cluster. Median offsets between the asteroseismic and isochrone ages exceed these uncertainties for several clusters, most notably NGC 6819 and M4. The paper argues that this pattern reflects an underlying tension between the two age scales, not merely random noise, and that current field-star ages derived from asteroseismic scaling relations are therefore not on a well-calibrated absolute scale.","pith_inferences":["A testable extension would be to compute ages for the same cluster stars using two or more independent asteroseismic pipelines with identical assumptions; if the intragroup scatter collapses, the reported tension is largely methodological rather than astrophysical.","The paper's adopted cluster ages and metallicities come from heterogeneous sources; applying a uniform isochrone fit to all seven clusters might shift the median offsets and would provide a cleaner comparison.","The identified disconnect between the clusters with seismic data and the age–metallicity distribution of the Milky Way field population suggests that calibrating asteroseismic ages to clusters alone may leave systematic biases in the most common stellar types.","If the missing correction depends on mass, metallicity, or surface gravity, then the few clusters with data cover too narrow a parameter space to constrain it, and new cluster detections (e.g., from PLATO) would be needed to map the correction."],"forward_implications":["If the tension is real, absolute ages for the hundreds of thousands of field red giants from Kepler, K2, TESS, and future missions inherit a systematic uncertainty larger than currently assumed.","A correction to the asteroseismic age scale, analogous to existing corrections for radius and mass scaling relations, would be required to place field-star ages on the cluster-calibrated scale.","The finding motivates a homogeneous reanalysis of existing cluster seismic data with a single pipeline and consistent solar reference values, which would either confirm the offset or show it is an artifact of method differences.","If binaries are the cause, the implication is that a substantial fraction of red giants in these clusters have been affected by mass transfer or merger events, challenging single-star model assumptions."],"supporting_citations":[{"why":"Provides the asteroseismic ages for M67 red giants used in the comparison.","marker":"Stello et al. (2016)"},{"why":"Supplies the seismic ages for NGC 6791 and NGC 6819 and the model-to-model systematic uncertainty estimate of roughly 10 percent.","marker":"Pinsonneault et al. (2024)"},{"why":"Provides the masses, surface gravities, and temperatures for M4 red giants from which seismic ages are computed.","marker":"Howell et al. (2022)"},{"why":"Supplies the corresponding data for M80 red giants.","marker":"Howell et al. (2024)"},{"why":"Supplies the data and cluster ages for M9 and M19.","marker":"Howell et al. (2025)"},{"why":"Provides the model grid used to convert masses, gravities, and metallicities into ages when not published directly.","marker":"Tayar et al. (2017)"},{"why":"Establishes the 5.4 percent absolute age uncertainty for the reference cluster M92, the benchmark used for the cluster age scale.","marker":"Ying et al. (2023)"},{"why":"Supplies the adopted isochrone age of 3.95 Gyr for M67.","marker":"Reyes et al. (2024)"}],"fun_headline_variants":["Seismic ages of cluster stars clash with host cluster ages","Star pulsation ages in clusters don't match isochrone ages","Asteroseismic age scatter in clusters exceeds expected noise","Cluster star seismic ages show major inconsistency","Why do asteroseismic ages disagree in star clusters?"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper treats the spread of ages within each cluster as if it measures the same underlying quantity, even though the ages come from different papers, pipelines, solar reference values, and model grids, and the authors do not propagate the individual uncertainties from those sources.","fun_headline_variants_meta":{"raw":{"variants":["Seismic ages of cluster stars clash with host cluster ages","Star pulsation ages in clusters don't match isochrone ages","Asteroseismic age scatter in clusters exceeds expected noise","Cluster star seismic ages show major inconsistency","Why do asteroseismic ages disagree in star clusters?"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000334,"raw_usage":{"total_tokens":1809,"prompt_tokens":855,"completion_tokens":954,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":471,"completion_tokens_details":{"reasoning_tokens":876}},"tokens_in":471,"tokens_out":954,"duration_ms":8643,"temperature":1.0,"reasoning_tokens":876,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T20:56:38.156239+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A homogeneous reanalysis of the same cluster stars using a single asteroseismic pipeline with fixed solar reference values and one model grid would settle the matter: if the within-cluster age dispersion drops to the level of the quoted uncertainties, the reported tension is a methodological artifact rather than a fundamental offset between the asteroseismic and isochrone age scales. Alternatively, measuring asteroseismic ages for stars in a cluster with a precisely known geometric distance and independent mass constraints (e.g., an eclipsing binary) would test whether the mass–age relation underlying the ages is biased.","supporting_citations":[],"review_version":1}