{"id":"7375c0ac-44d5-4f0e-8e0f-f7f90727aaac","arxiv_id":"2501.04295","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A systematic LAMOST search identifies 14 accreting M stars with disks that appear older than 10 Myr, nearly doubling the known Peter Pan disk census.","lead":"The authors searched LAMOST spectra of M stars and found 14 old stars that still have gas-rich disks and are still pulling in material. The result, if confirmed, nearly doubles the known census of rare 'Peter Pan' disks and challenges the idea that such disks only survive in nearby young associations.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 14 'Peter Pan' candidates hinge on PARSEC isochronal ages for individual M/K stars; the paper's own Section 5.1 shows this method demotes most known Peter Pan disks below 10 Myr when Baraffe tracks are used.","rationale":"The paper is a careful, systematic search with genuinely useful products: a well-defined working sample of 855 disked M/K stars, visual WISE inspection to remove spurious photometry, detailed SED fitting, and a coherent finding of lower accretion rates in the isochronally old population. The validation against literature Peter Pan disks in Section 5.1 is an honest test, and it is precisely what reveals the weakness: individual isochronal ages for low-mass stars are strongly model-dependent. The authors already cite Herczeg & Hillenbrand (2015) on mass-dependent age discrepancies, and they acknowledge both the Baraffe-track discrepancy and the edge-on degeneracy. The absence of Li detection, while not evidence of youth, means no independent age check is available; the paper asks the reader to accept PARSEC placement as sufficient. The edge-on concern is not merely formal: several candidates are classified as full/evolved disks with fitted inclinations near or above 60 degrees, and the SED fitting cannot fully break the inclination-luminosity degeneracy for unresolved disks. For these reasons the central 'older than 10 Myr' claim is not yet established. The paper remains valuable as a candidate catalog and as a methodological demonstration, and conditional acceptance with a requirement for independent age indicators or a clear statement of model dependence is the appropriate outcome.","tokens_in":22174,"tokens_out":2638,"duration_ms":30942,"concrete_test":"Re-derive ages for the 14 candidates from the same Teff and Lbol values using at least two independent pre-main-sequence model sets, e.g., Baraffe et al. (2015, BHAC15) and Siess et al. (2000) or Feiden (2016), and count how many remain below the 10 Myr isochrone under both alternative sets. If substantially fewer than 14 remain older than 10 Myr under any independent track set, the census claim is not robust and the paper should be revised to present the objects as candidates pending independent age confirmation.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim that 14 sources are genuinely older than 10 Myr, still disk-bearing and actively accreting rests on the assumption that PARSEC isochronal placement is reliable for individual late-K/M stars. Section 3.2 uses 'below the 10 Myr isochrone in both H-R diagrams' as the age criterion, but both diagrams are derived from PARSEC and the same SED fitting; the disk-scattering-corrected version is not an independent check. Section 5.1 exposes the fragility: when the same analysis is applied to known Peter Pan disks with Baraffe et al. (2015) tracks, only 1 of 10 remains older than 10 Myr. No age uncertainties are reported in Table 1, and Section 4.2 states that no Li I 6707 absorption is detected because the LAMOST spectra are too noisy, so there is no independent youth indicator. Section 5.2 explicitly concedes that edge-on young disks cannot be excluded; for a young disk viewed edge-on, scattered light can reduce apparent bolometric luminosity and mimic an old, low-luminosity star. Two of the 14 candidates are plausibly in the ~20 Myr 32 Ori association (Table 1 notes), but the remaining candidates rely entirely on isochronal ages. If the PARSEC ages are systematically old, or if several candidates are edge-on young disks, the 'nearly doubled census' claim loses its foundation. This does not invalidate the careful disk identification or the accretion analysis, but 'older than 10 Myr' is the least secure, load-bearing premise.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper searches the LAMOST M-star catalog for old, disk-bearing, actively accreting stars (Peter Pan disks). From 855 objects with infrared excess, the authors identify 526 as isochronally older than 10 Myr using PARSEC isochrones, and 14 of these as classical T Tauri stars. They derive stellar parameters, variability, accretion rates, and disk classifications, and argue that these 14 objects nearly double the known census of Peter Pan disks, with most located outside nearby associations and instead associated with open clusters or isolated. The accretion analysis finds lower accretion rates for old disks at a given stellar mass, while the age evolution of accretion rates is similar to young disks.","tokens_in":22582,"tokens_out":3142,"duration_ms":30471,"significance":"If correct, the paper would substantially expand the census of long-lived protoplanetary disks, extend them from nearby associations to open clusters and the field, and provide new constraints on disk evolution and accretion physics. The disk selection is carefully done: the authors visually inspect WISE images, use two separate SED-fitting approaches, and reclassify spectral types with veiling. The variability and accretion characterizations are standard and clearly described. However, the central age claim rests on isochronal ages for individual late-K/M stars, a method the authors themselves show is strongly model-dependent, and no independent youth indicator is available. The strength of the paper is its systematic census and the careful disk identification; the fragility is in the age certification that defines the sample.","major_comments":[{"comment":"The central age criterion is entirely isochronal and model-dependent. The paper's own validation in Section 5.1 shows that when the same isochronal analysis is applied to the ten literature Peter Pan disks using Baraffe et al. (2015) tracks, only one remains older than 10 Myr. This demonstrates that PARSEC-based isochronal ages are not robust for individual low-mass stars. Because both H-R diagrams in Section 3.2 are based on PARSEC, the requirement that a source lie below the 10 Myr isochrone in both diagrams does not provide an independent check of the age. Please provide age estimates with uncertainties from multiple evolutionary models, or otherwise justify why PARSEC is reliable for these late-K/M stars when it demotes most known Peter Pan disks to ages below 10 Myr.","section":"§3.2, §5.1"},{"comment":"No independent youth indicator is available for the sample. The Li I 6707 Å line is not detected because the LAMOST spectra are too noisy, and Section 5.2 explicitly concedes that edge-on young disks cannot be excluded. For a young disk viewed edge-on, scattered light can reduce the apparent bolometric luminosity and mimic an old, low-luminosity star. Thus, for the 12 candidates not associated with the 32 Ori association, the only evidence for ages beyond 10 Myr is the PARSEC isochronal placement. Please quantify how many of the 14 candidates would survive if Baraffe et al. (2015) tracks or other evolutionary models are adopted, and discuss the edge-on degeneracy more quantitatively using the inclination angles from the SED fitting reported in Figure 4.","section":"§4.2, §5.2"},{"comment":"The ages in Table 1 are quoted without uncertainties (e.g., 11.6 Myr, 32.9 Myr, >50 Myr). For individual low-mass stars, isochronal ages are notoriously uncertain, especially for M dwarfs near or below 0.2 M☉, as in source 3308700559817832576 with M⋆ = 0.16 M☉ and an age of 12.4 Myr. Without error bars or a multi-model comparison, it is impossible to assess whether any individual source is older than 10 Myr at a statistically significant level. Please report age uncertainties that propagate the Teff and Lbol uncertainties and, at least for the 14 candidates, the spread among different evolutionary tracks.","section":"Table 1, §4.2"}],"minor_comments":[{"comment":"The text contains the typo 'Perter Pan' in the first sentence; it should be 'Peter Pan'.","section":"§5.3"},{"comment":"The cluster name is misspelled as 'Columnba'; it should be 'Columba'.","section":"§5.1"},{"comment":"The disk type entry 'EVOL VED' contains a spurious space; it should read 'EVOLVED'.","section":"Table 1"},{"comment":"Equation (4) is written with a comma-separated formula that is ambiguous; it should be presented as a fraction with the weighted mean defined clearly.","section":"§4.2, Eq. (4)"},{"comment":"The first author's name appears as 'Xiao-Long W ang' with an extra space; this should be corrected to 'Xiao-Long Wang'.","section":"Author byline"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid observational census with careful disk identification, but the headline claim of 14 new Peter Pan disks hinges on PARSEC isochronal ages for individual M/K stars. Given the authors' own demonstration in Section 5.1 that a different but standard evolutionary model removes most of the comparison sample from the >10 Myr category, the age certification must be strengthened before the paper can be accepted. The authors may also consider re-framing the paper as a census of old disk candidates with a clearly quantified model dependence rather than as a definitive discovery claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things about this paper. First, it is a genuinely useful catalog effort: 14 new disk-bearing M/K stars with accretion signatures, selected from LAMOST and WISE with careful SED fitting and one-by-one visual image inspection. Second, the central claim that these objects are older than 10 Myr is weaker than the census language suggests, because it hinges entirely on PARSEC isochronal ages for individual low-mass stars. The paper itself shows in Section 5.1 that with Baraffe tracks only 1 of 10 known Peter Pan disks remains older than 10 Myr. That is an honest admission, but it means the new candidates should be treated as provisional until independent age indicators arrive.\n\nWhat is new and good: the blind search extends the Peter Pan disk phenomenon from nearby associations to open clusters and isolated field stars, which matters for disk dissipation timescales and late-stage planet formation. The accretion-rate result—long-lived disks accreting systematically below younger counterparts at the same mass—is coherent and worth following up. The selection pipeline is transparent, and the appendix demonstrates the WISE image inspection that guards against spurious infrared excesses. The catalog of 855 disked stars, with derived parameters, is a resource for the community.\n\nWhere the soft spots are: the isochronal age for an individual M dwarf is the load-bearing assumption, and it is not well tested here. No age uncertainties are reported in Table 1. The two H-R diagrams are not independent—both come from PARSEC, and the disk-scattering-corrected version is only a minor adjustment. No lithium is detected because the LAMOST spectra are too noisy, so there is no youth check. Edge-on young disks are explicitly conceded as a possibility in Section 5.2; for those, scattered light can mimic an old, low-luminosity star. These are real limitations, and the authors acknowledge most of them, but the abstract still states the age as fact. A referee should push for a systematic test across multiple isochrone sets and quantitative treatment of the edge-on degeneracy.\n\nWho this is for: people working on disk evolution and M-star youth. It deserves a serious referee—the search design and photometric analysis are solid, and even if some candidates turn out young, the working sample and the accretion comparison are valuable. Send it to peer review with a request for age-systematics analysis. The paper would be stronger if it explicitly framed the 14 as candidates pending confirmation via radial velocity, lithium, or [O I] 6300 observations.","headline":"A careful search that likely produces real disk-bearing M-star candidates, but the 'older than 10 Myr' label rests on one isochrone family and needs independent age checks before the Peter Pan claim carries.","tokens_in":23056,"tokens_out":1897,"would_cite":true,"duration_ms":21529,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["97.21.+a"],"model":"deepseek-v4-flash","headline":"LAMOST data reveal 14 M/K stars older than 10 million years that still harbor actively accreting protoplanetary disks, nearly doubling the census of the rare class known as Peter Pan disks.","keywords":["Peter Pan disks","protoplanetary disks","T Tauri stars","mass accretion rates","stellar isochronal ages","LAMOST","M dwarfs","infrared excess"],"falsifier":"High-resolution spectroscopy of the 14 candidates measuring Li I $\\lambda$6707 absorption and [O I] $\\lambda$6300 line profiles would settle the matter: strong lithium or near-edge-on disk geometry would mean the sources are young systems mis-aged by the isochrones, while re-deriving the ages with the Baraffe et al. (2015) tracks and finding most candidates above the 10 Myr isochrone would directly falsify the age certification.","tokens_in":21912,"feed_emoji":"🪐","tokens_out":14950,"duration_ms":120035,"temperature":0.7,"pith_summary":"Protoplanetary disks are supposed to be gone within about 10 million years, yet a few rare exceptions known as Peter Pan disks survive much longer. The paper reports a systematic, catalog-wide search of the LAMOST M-star archive that identifies 14 M/K stars older than 10 Myr that still carry a circumstellar disk and still accrete gas as classical T Tauri stars, nearly doubling the previously known census of about 15 such objects. To count as a Peter Pan disk, a source must lie below the 10 Myr isochrone in two separately constructed Hertzsprung-Russell diagrams, show infrared excess above its stellar photosphere, and have H$\\alpha$ emission strong enough to mark active accretion. Most of the new cases are not in the nearby associations that hosted the earlier finds; they are members of open clusters or isolated field stars. The paper finds that these aged disks accrete at systematically lower rates than young disks around stars of similar mass, while the decline of accretion rate with age continues the same trend seen in young disks.","feed_headline":"14 aged stars still feed their planet-forming disks","feed_subtitle":"LAMOST survey nearly doubles the census of Peter Pan disks, most of them in open clusters.","key_machinery":"The working machinery is a three-stage filter. First, infrared excess: sources are selected in $J-H$ versus $H-W1$, $H-W2$, and $H-W3$ color-color planes against the intrinsic colors of 5-30 Myr old stars, and every candidate's WISE image is inspected by eye to remove spurious photometry. Second, age: a source must fall below the 10 Myr PARSEC isochrone in the H-R diagram in both a photosphere-only SED fit and a full disk-inclusive SED fit, the doubled requirement guarding against scattered disk light mimicking an older star. Third, accretion: H$\\alpha$ equivalent widths above spectral-type-dependent thresholds classify a source as a classical T Tauri star, and the H$\\alpha$ luminosity is converted to a mass accretion rate through an empirical accretion-luminosity relation and the boundary-layer formula $\\dot{M}_{\\rm acc}=L_{\\rm acc}R_\\star/(GM_\\star(1-R_\\star/R_{\\rm in}))$. That combination — old isochronal age, persistent disk, ongoing accretion — is what defines a Peter Pan disk in this study.","core_discovery":"The central claim is that long-lived protoplanetary disks are common enough to be found by a blind search rather than only as anomalies in well-studied young associations. Applying three successive filters to the LAMOST M-star catalog — infrared excess in WISE bands, placement below the 10 Myr PARSEC isochrone in two H-R diagrams built from photosphere-only and disk-inclusive SED fits, and H$\\alpha$ emission above the classical T Tauri star threshold — the authors arrive at 14 previously unknown Peter Pan disks with isochronal ages from just over 10 Myr to beyond 50 Myr. These objects nearly double the known class. Their mass accretion rates, derived from H$\\alpha$ line luminosity, are systematically lower than those of young accreting disks with similar stellar masses, yet the accretion-rate decline with age follows the same trend as in young populations. Spatially, nearly all of them avoid nearby star-forming regions; most appear to belong to open clusters or to be isolated, with a few possibly ejected from their birth sites.","pith_inferences":["The paper's own cross-check shows the isochronal-age verdict flips largely when another evolutionary-track family is used, so part of the 14-object census may not survive better age indicators; lithium measurements or kinematic trace-back would separate confident members from borderline ones.","The systematically low accretion rates of these disks suggest a testable mechanism: disks that accrete quietly from the start may be the ones that survive longest, and millimeter interferometric gas-mass measurements of the 14 systems could test whether low turbulence or weak external radiation is the cause.","Because most new cases sit in open clusters, mining cluster catalogs built from astrometric surveys for infrared-excess members could uncover further long-lived disks far more efficiently than the association-by-association searches that produced the earlier sample."],"forward_implications":["Long-lived disks are not confined to nearby associations: most of the 14 new Peter Pan disks are open-cluster members or isolated field stars, so the phenomenon is not tied to a single environment.","The known census nearly doubles, from about 15 catalogued Peter Pan disks to 29, giving the first sample large enough to test statistically what governs disk survival.","Aged disks accrete at systematically lower rates than young disks of the same stellar mass, suggesting that low accretion activity goes together with long disk lifetime.","The decline of mass accretion rate with age in these old disks follows the same trend as in young accretors, meaning late disk evolution extends the normal sequence rather than breaking it.","Several objects sit well below the 50 Myr isochrone; if they are truly that old and not edge-on young disks, their existence would challenge current models of disk dispersal."],"supporting_citations":[{"why":"Tabulates the 15 previously known Peter Pan disks used as the comparison sample for the census and for the H-R diagram validity check.","marker":"Lee et al. 2020"},{"why":"Supplies the PARSEC evolutionary tracks and isochrones whose 10 Myr line certifies a source's age in the H-R diagram.","marker":"Bressan et al. 2012"},{"why":"Provides the alternative evolutionary tracks the authors use to test whether the isochronal age verdict survives a change of models.","marker":"Baraffe et al. 2015"},{"why":"Gives the spectral-type-dependent H-alpha equivalent width thresholds for classical T Tauri classification and the line-luminosity to accretion-luminosity relation.","marker":"Fang et al. 2009"},{"why":"Supplies the intrinsic colors of 5-30 Myr old stars against which infrared excess is defined in the color-color selection.","marker":"Pecaut & Mamajek 2013"},{"why":"Provides the extinction-corrected color-excess scheme used to classify disks as full, transitional, evolved, or debris.","marker":"Esplin et al. 2014, 2018"},{"why":"Supplies the empirical mass-accretion-rate relations with stellar mass and age against which the old disks are compared.","marker":"Hartmann et al. 2016"},{"why":"Catalog of open clusters used to assign membership for the newly discovered disks and to show that most are clustered.","marker":"Hunt & Reffert 2023"},{"why":"Defines the Peter Pan disk class that this paper's 14 classical T Tauri disks are assigned to.","marker":"Silverberg et al. 2020"}],"fun_headline_variants":["LAMOST spots 14 Peter Pan disks still feeding their stars","14 new 'Peter Pan' disks refuse to age, still accreting","LAMOST finds 14 new Peter Pan disks, nearly doubling census","Aged protoplanetary disks: 14 still accreting after 10 Myr"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that PARSEC isochronal ages for individual low-mass stars are trustworthy enough to certify ages beyond 10 Myr; the paper itself shows that with the Baraffe et al. (2015) tracks only 1 of 10 previously known Peter Pan disks stays older than 10 Myr, and no lithium absorption is available to independently confirm youth.","fun_headline_variants_meta":{"raw":{"variants":["LAMOST spots 14 Peter Pan disks still feeding their stars","14 new 'Peter Pan' disks refuse to age, still accreting","LAMOST finds 14 new Peter Pan disks, nearly doubling census","Aged protoplanetary disks: 14 still accreting after 10 Myr"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000808,"raw_usage":{"total_tokens":3544,"prompt_tokens":942,"completion_tokens":2602,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":558,"completion_tokens_details":{"reasoning_tokens":2531}},"tokens_in":558,"tokens_out":2602,"duration_ms":16790,"temperature":1.0,"reasoning_tokens":2531,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:36:09.709596+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"High-resolution spectroscopy of the 14 candidates measuring Li I $\\lambda$6707 absorption and [O I] $\\lambda$6300 line profiles would settle the matter: strong lithium or near-edge-on disk geometry would mean the sources are young systems mis-aged by the isochrones, while re-deriving the ages with the Baraffe et al. (2015) tracks and finding most candidates above the 10 Myr isochrone would directly falsify the age certification.","supporting_citations":[],"review_version":1}