{"id":"fac43a9b-b883-4901-8add-5d322f1df722","arxiv_id":"2507.19249","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"New K-band spectra of six evolved massive stars in M31 and M33 revise or support classifications, with two B[e] supergiants showing CO emission and one candidate binary that needs confirmation.","lead":"Astronomers took new near-infrared spectra of six very bright, evolved stars in two nearby galaxies and used the shapes of carbon monoxide and hydrogen lines to sort them into known classes of unstable massive stars. The work adds new classifications and points to one candidate binary system that, if confirmed, would be the first of its kind.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Binary claim for J013242 rests on sparse radial velocities that overlap the known RSG pulsation timescale; 'most likely binary' is stronger than the current data support.","rationale":"The reader's CONDITIONAL verdict is sound and I do not want to overturn it. I prioritize the binary hypothesis as the load-bearing point: the post-RSG isotope claim for J004320 is also fragile, but the paper already labels it 'most likely' and asks for a better spectrum, and even if that ratio were wrong the object would likely remain a B[e]SG, preserving part of the diagnostic message. The binary claim is the paper's only 'first of its kind' result and depends on distinguishing a ~9 km/s orbital signal from RSG pulsation. Table 4 is sparse, includes one unquoted-uncertainty epoch from a different instrument, and the inferred minimum orbital period (532 d) overlaps the expected pulsation period (500-1000 d). The authors themselves flag the pulsation alternative, but the conclusion wording 'most likely a binary' outruns the evidence. The proposed Bayesian comparison on existing RVs tests whether the current data favor an orbit at all; if not, the correct status is 'candidate,' not 'most likely.' This does not reject the observational contribution; it sharpens the conditionality.","tokens_in":20685,"tokens_out":17744,"duration_ms":166923,"concrete_test":"Run a Bayesian model comparison on the 14 RVs in Table 4: model A is a Keplerian orbit with P in [532, 5000] d and free semi-amplitude; model B is an RSG pulsation model with P=500-1000 d (Kiss et al. 2006 period-luminosity relation for M_K~-11) and K<=10 km/s, represented as a quasi-periodic Gaussian process or multi-mode sinusoid. If the log-evidence difference in favor of A over B is not greater than 5, or the comparison is inconclusive, the current data do not support 'most likely binary'; the object should be labeled a binary candidate and the conclusion softened. The decisive observational follow-up is then a dense RV campaign over >2000 d on the RSG plus blue-optical RVs of the hot component to test for antiphase orbital motion.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The flagship claim - J013242.26+302114.1 as the first LBV/B[e]SG+RSG binary - rests on the radial velocities in Table 4 being orbital. The largest excursion (-121.6 km/s) comes from a single 2012 epoch by Drout et al. with no quoted uncertainty, while the 2017-2019 clusters span only about -110.8 to -100.0 km/s; the 'semi-amplitude of at least ~9 km/s' is therefore not a phase-resolved orbital amplitude. The expected RSG radial-pulsation period for M_K~-11 is 500-1000 d (Kiss et al. 2006; Chatys et al. 2019), RSG RVs reach ~10 km/s, and the paper's own minimum orbital period estimate (532 d) falls inside that window. The authors acknowledge in Section 4.2.2 that the RV variations might have a different origin, yet the conclusions still call the system 'most likely a binary.' Without a coherent RV curve over at least one full cycle and an RV for the hot component, the orbital interpretation is not favored over pulsation. If this concern lands, the 'first of its kind' result reduces to a candidate, substantially lowering the paper's significance.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents new GNIRS K-band spectra (R~5900, 2.2-2.4 microns) for six evolved massive stars in M31 and M33, five of which had previously ambiguous optical classifications. For each target the authors model the detected CO band emission (two M31 objects) or hydrogen Pfund emission (two objects) and combine the results with literature photometry, SED modeling, and radial velocities. They classify J004320.97+414039.6 and J004621.08+421308.2 as B[e] supergiants, argue that J004320 is most likely in a post-red supergiant stage based on 13C enrichment (12CO/13CO ~ 3), propose J004415.00+420156.2 as a B[e] supergiant, confirm Var 83 as an LBV, find J004229.87+410551.8 featureless and possibly an LBV in an S Dor cycle, and argue that J013242.26+302114.1 is most likely a binary system comprising a hot B-type emission-line object and a red supergiant, potentially the first of its kind.","tokens_in":20960,"tokens_out":5257,"duration_ms":55882,"significance":"If the classifications hold, the paper demonstrates that K-band CO and Pfund diagnostics can resolve optical ambiguities in extragalactic evolved massive stars and identifies a rare post-RSG B[e]SG candidate and a possibly first LBV/B[e]SG+RSG binary. The observations are new and useful: four objects receive their first near-infrared spectra, the model fits are quantitative and reported with parameters, and the analysis makes good use of archival APOGEE radial velocities and long-term photometry. The paper also contains explicit caveats about the weakest data. However, the two headline claims--the post-RSG status of J004320 and the binary nature of J013242--rest on fragile evidence, so the significance of the work is conditional on additional data or a more cautious presentation.","major_comments":[{"comment":"The binary interpretation of J013242.26+302114.1 is not yet supported by the radial-velocity data. The claimed semi-amplitude of at least ~9 km/s is set by the single 2012 measurement (-121.6 km/s, no quoted uncertainty) against the APOGEE/GNIRS cluster near -100 to -110.8 km/s; the APOGEE values alone span only about 7.5 km/s. The authors themselves note that RSG radial pulsations reach ~10 km/s with periods of 500-1000 d for M_K ~ -11, and their own minimum orbital period estimate of 532 d lies inside that range. With no phase-resolved orbit and no radial-velocity measurement of the hot component, the data do not favor orbital motion over intrinsic pulsation. The abstract and conclusions should state \"candidate binary\" rather than \"most likely a binary\" unless additional epochs or a coherent radial-velocity curve are provided.","section":"§4.2.2, Table 4, Fig. 7"},{"comment":"The post-RSG classification of J004320.97+414039.6 rests entirely on the fitted 12CO/13CO ratio of 3 +/- 1 in a spectrum with S/N = 16 and telluric remnants in the red portion. The modeling assumes LTE and a particular (Keplerian or equatorial-outflow) geometry, and the mapping to the stellar surface 12C/13C at the time of ejection depends on single-star rotating evolutionary tracks. No systematic-error analysis is given for the telluric correction, continuum placement, or optical depth, so it is not demonstrated that the ratio is robustly below the RSG threshold of 5. Since this is the load-bearing evidence for the \"post-red supergiant\" statement in the abstract, the paper should either add a quantitative robustness test (e.g., varying the continuum, line-formation geometry, or optical depth) or downgrade the claim to provisional pending higher-S/N data.","section":"§3.1, Table 2, Fig. 2"},{"comment":"The argument that a featureless K-band spectrum \"tentatively excludes\" a B[e]SG classification is weakened by the paper's own statement that only about 50% of B[e]SGs show CO band emission, and by the possibility that Pfund emission is weak or absent in some B[e]SGs. The conclusion that J004229.87+410551.8 is not a B[e]SG therefore relies more on the literature-based concern about IR excess contamination than on the new spectrum alone, and the wording should reflect this asymmetry.","section":"§4.1.1"}],"minor_comments":[{"comment":"The caption contains a corrupted text fragment (\"20 MO •\") and the color-bar label \"12C/13C\" should be explicitly defined as the carbon isotope abundance ratio.","section":"Fig. 2 caption"},{"comment":"The Drout et al. (2012) radial-velocity measurement is listed without an uncertainty; please add an estimated uncertainty or state explicitly why none is available.","section":"Table 4"},{"comment":"The statement that the single-star ages of the two components \"have at least the same order of magnitude\" is presented as supporting binarity, but a 3-4 Myr age difference is substantial; this argument should be removed or explicitly framed as non-constraining.","section":"§4.2.2"},{"comment":"Please clarify how the rest-wavelength shift was determined for objects with no intrinsic spectral features, given that for J004229.87+410551.8 no shift is applied.","section":"§2"},{"comment":"The abbreviation \"cLBV\" is used without definition at first occurrence; please expand it in the table caption or in Section 1.","section":"Table 1"}],"recommendation":"major_revision","confidential_remarks":"The paper is a useful observational contribution to the classification of extragalactic evolved massive stars, and the authors' internal caveats are appropriate. My main concern is that the abstract and conclusions state the binary and post-RSG claims more strongly than the data warrant. The binary claim in particular is not favored over RSG pulsation with the current sparse radial-velocity sampling, and the post-RSG claim rests on a single low-S/N spectrum. I recommend major revision to soften the headline claims and add quantitative robustness checks where feasible."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague — this is a workmanlike NIR paper with real value and a pair of headline claims that overreach a little. The new K-band spectra of six evolved massive stars in M31 and M33, four of them without any published NIR spectrum, are genuinely new and cleanly presented. The CO detection in J004621 and the Pfund emission in J004415 are solid; the reclassification of J004621 as B[e]SG is well argued from the persistent CO bands, and the featureless J004229 object is handled with appropriate caution. The paper also does the field a service by spelling out its own caveats in Section 4.2.2.\n\nThe two claims that carry the abstract's excitement are both softer than the language used. The post-RSG classification of J004320 rests on a 12CO/13CO ratio measured from a S/N~16 spectrum with telluric remnants. The authors themselves say better spectra are needed, but the conclusion still says 'most likely in a post-red supergiant stage,' and that's a stretch. The binary claim for J013242 is the bigger issue. The RV points are sparse, the -121.6 km/s point is a single 2012 epoch with no quoted uncertainty, and the remaining points span about 10 km/s with clusters that look consistent with the 500-1000 day pulsation timescale the authors cite. The minimum period of 532d falls right inside that window. The paper's own text acknowledges the pulsation alternative, yet the conclusions call it 'most likely a binary.' A candidate, yes; 'most likely,' not yet. This is a wording problem that a revision should fix, not a fatal flaw.\n\nThe modeling is what it is: LTE CO disk code and standard SED fitting. The fits don't presuppose the classifications, and the external diagnostics (optical spectra, variability, IR colors) are used properly. I don't see a circularity problem. The citation pattern is appropriate, with the authors' own prior work prominent but not exclusionary.\n\nBottom line: if you work on evolved massive stars, this is worth a read and a cite for the NIR data. It deserves a serious referee; the referee just needs to push the authors to soften the two claims to match the evidence. Send it to review, but demand the wording changes.","headline":"Honest, useful NIR classifications of six evolved massive stars; the two headline claims (post-RSG status, binary) are both weaker than the abstract implies, but the authors flag the gaps themselves.","tokens_in":21517,"tokens_out":2047,"would_cite":true,"duration_ms":20656,"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":"Near-infrared K-band spectroscopy of six evolved massive stars in M31 and M33 detects CO band emission that reclassifies ambiguous optical objects as B[e] supergiants, places one in a post-red-supergiant stage, and points to a…","keywords":["massive stars","B[e] supergiants","luminous blue variables","red supergiants","near-infrared spectroscopy","CO band emission","carbon isotope ratio","M31 and M33"],"falsifier":"A higher-quality K-band spectrum of J004320.97+414039.6 with $^{12}$CO/$^{13}$CO above 5 would falsify the post-red-supergiant classification; alternatively, a completed radial-velocity curve for J013242.26+302114.1 that follows 500–1000 day red-supergiant pulsations rather than a roughly 530-day orbit would falsify the binary claim.","tokens_in":20460,"feed_emoji":"🔭","tokens_out":8525,"duration_ms":76805,"temperature":0.7,"pith_summary":"This paper claims that near-infrared K-band spectra can break classification deadlocks among evolved massive stars that optical spectra alone cannot resolve. Observing six stars in M31 and M33, the authors detect CO band emission around two M31 objects and reclassify them as B[e] supergiants—hot evolved stars with dense circumstellar disks—with one most likely caught in the post-red-supergiant phase. They also argue that a third M31 star is a B[e] supergiant, that a featureless M31 object is likely an LBV in an S Dor cycle, and that an M33 star is a possible binary pairing a hot LBV/B[e] supergiant with a red supergiant, a combination never before confirmed. If right, the work would show that molecular and hydrogen-recombination diagnostics in the near-infrared are decisive where optical classification is ambiguous, and it would add a rare evolutionary stepping-stone to the map of massive-star lives.","feed_headline":"First candidate LBV/B[e] supergiant binary with red supergiant","feed_subtitle":"K-band spectra and radial velocity shifts point to the pair; confirming it would make it the first such system known.","key_machinery":"The load-bearing diagnostic is the CO first-overtone band spectrum in the K band: emission from the $^{12}$CO and $^{13}$CO band heads, modeled with a Keplerian-rotating-disk code under LTE, yields the gas temperature, column density, projected rotation velocity, and the $^{12}$CO/$^{13}$CO abundance ratio. That ratio is the evolutionary clock—it tracks the surface $^{12}$C/$^{13}$C ratio, which drops from the interstellar value near 90 toward values below 5 only after the red-supergiant phase. Pfund-series hydrogen emission supplies the complementary probe of ionized-wind density in hotter objects, and CO absorption identifies cool late-type components, as in the candidate binary.","core_discovery":"The central discovery is that the near-infrared K-band region, through CO first-overtone band emission and hydrogen Pfund-series emission, exposes the nature of evolved massive stars that optical spectra leave ambiguous. CO band emission indicates dense, warm (roughly 1500–2200 K) circumstellar rings, and the modeled $^{12}$CO/$^{13}$CO ratio reads the star's surface carbon-isotope enrichment at the time the gas was ejected. For J004320.97+414039.6 the ratio is about 3, which on rotating 20 solar-mass evolutionary tracks is reached only after the red-supergiant phase, placing the star in a post-red-supergiant stage. For J013242.26+302114.1 in M33, pure CO absorption reveals a red supergiant whose radial velocity wanders by at least 9 km/s; together with a spectral energy distribution that requires both a hot 22,000–24,000 K star and a cool 4,000 K component, this supports a binary of a hot LBV or B[e] supergiant with a red supergiant secondary, which, if confirmed, would be the first such system known.","pith_inferences":["A direct prediction is that higher-signal-to-noise K-band spectroscopy of J004320.97+414039.6 will keep $^{12}$CO/$^{13}$CO below about 5; if the measured ratio instead rises above that, the post-red-supergiant reading would be an artifact of noise or telluric residuals.","The same $^{12}$CO/$^{13}$CO method could be applied to other extragalactic B[e] supergiant candidates with ambiguous optical classifications, turning the technique into a population-level census of post-red-supergiant stars.","If the binary is real, its roughly 9 km/s semi-amplitude implies a long period, and a completed radial-velocity curve should show a period above about 530 days—the Roche-lobe limit derived in the paper—allowing the pulsation alternative to be excluded.","The paper's assumption that the CO ring is in Keplerian rotation is not directly tested; a non-Keplerian outflow geometry would change derived column densities and isotope ratios, so time-resolved or interferometric measurements of the ring kinematics would strengthen the classification."],"forward_implications":["J004320.97+414039.6 joins the small set of post-red-supergiant B[e] supergiants, marking the evolutionary transition from the red-supergiant phase back to a hot phase.","CO band emission should be treated as a decisive discriminator between B[e] supergiants and LBV candidates in extragalactic surveys, breaking the optical degeneracy between these classes.","The candidate binary J013242.26+302114.1, if confirmed with a complete orbit, would be the first known red-supergiant binary with an evolved massive primary, providing a testbed for binary evolution and mass transfer.","For J004229.87+410551.8, the featureless K-band spectrum plus brightening and reddening since 2002 supports an ongoing S Dor cycle, meaning its classification as an LBV should be tested with fresh optical spectroscopy.","The persistent CO emission and Na I doublet in J004621.08+421308.2 confirm that its molecular disk has remained stable over roughly eight years."],"supporting_citations":[{"why":"Provides the code that computes $^{12}$CO emission from a Keplerian rotating disk, the model used to fit the observed CO bands.","marker":"Kraus et al. (2000)"},{"why":"Establishes CO first-overtone emission as a diagnostic of dense warm circumstellar gas and the $^{12}$C/$^{13}$C ratio as an evolutionary indicator.","marker":"Kraus (2009)"},{"why":"Surveys the near-infrared properties of YHGs, LBVs, and B[e]SGs, providing the comparative behavior of CO and Pfund lines used for classification.","marker":"Oksala et al. (2013)"},{"why":"Proposes CO emission as a complementary criterion to classify B[e]SGs rather than LBV candidates.","marker":"Kraus et al. (2014)"},{"why":"Supplies the rotating evolutionary tracks that convert the measured carbon isotope ratio into an evolutionary phase, showing values below 5 occur only after the red-supergiant phase.","marker":"Ekström et al. (2012)"},{"why":"Provides updated energy levels and Einstein transition coefficients for both $^{12}$CO and $^{13}$CO used in the band modeling.","marker":"Li et al. (2015)"},{"why":"The original red-spectrum classification of J013242.26+302114.1 as a red supergiant, and the first radial velocity measurement.","marker":"Drout et al. (2012)"},{"why":"Provides the optical Fe II classifications and stellar luminosities for the M31 and M33 sample used in the HR diagram placement.","marker":"Humphreys et al. (2014)"}],"fun_headline_variants":["K-band spectra reveal first candidate hot+red supergiant binary","CO bands expose post-red supergiant B[e] star in Andromeda","Near-IR picks out B[e] supergiants optical missed in M31, M33","Radial velocity hints at LBV + red supergiant binary in M33","GNIRS K-band spectra identify evolved massive stars in M31, M33"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The post-red-supergiant classification of J004320.97+414039.6 stands on the modeled $^{12}$CO/$^{13}$CO ratio of about 3 being the true surface carbon-isotope ratio of the ejected gas, measured from a low-signal spectrum.","fun_headline_variants_meta":{"raw":{"variants":["K-band spectra reveal first candidate hot+red supergiant binary","CO bands expose post-red supergiant B[e] star in Andromeda","Near-IR picks out B[e] supergiants optical missed in M31, M33","Radial velocity hints at LBV + red supergiant binary in M33","GNIRS K-band spectra identify evolved massive stars in M31, M33"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000898,"raw_usage":{"total_tokens":3984,"prompt_tokens":1176,"completion_tokens":2808,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":792,"completion_tokens_details":{"reasoning_tokens":2704}},"tokens_in":792,"tokens_out":2808,"duration_ms":19764,"temperature":1.0,"reasoning_tokens":2704,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T17:57:16.726648+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A higher-quality K-band spectrum of J004320.97+414039.6 with $^{12}$CO/$^{13}$CO above 5 would falsify the post-red-supergiant classification; alternatively, a completed radial-velocity curve for J013242.26+302114.1 that follows 500–1000 day red-supergiant pulsations rather than a roughly 530-day orbit would falsify the binary claim.","supporting_citations":[{"cited_title":"E., Kraus M., Cidale L","cited_arxiv_id":null,"evidence_quote":"Surveys the near-infrared properties of YHGs, LBVs, and B[e]SGs, providing the comparative behavior of CO and Pfund lines used for classification."},{"cited_title":"S., Arias M","cited_arxiv_id":null,"evidence_quote":"Proposes CO emission as a complementary criterion to classify B[e]SGs rather than LBV candidates."},{"cited_title":"M., Weis K., Davidson K., Bomans D","cited_arxiv_id":null,"evidence_quote":"Provides the optical Fe II classifications and stellar luminosities for the M31 and M33 sample used in the HR diagram placement."}],"review_version":1}