{"id":"92ec5029-93b9-4f3a-80d2-2be9e0311212","arxiv_id":"1908.03025","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A Gaia DR2 sample of 15,274 M-dwarfs yields empirical temperature and luminosity-radius relations showing 3 to 7 percent radius inflation with intrinsic scatter below 1 to 2 percent.","lead":"Using Gaia distances and multi-band photometry, the authors measured temperatures, luminosities, and radii for more than 15,000 M-dwarf stars. They find M-dwarf radii are inflated 3 to 7 percent relative to standard stellar models, but lie on a strikingly tight sequence, which challenges magnetic inflation theories.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 1–2% intrinsic-spread claim depends on an unvalidated model-derived metallicity correction, and Eq. 4 has a sign error.","rationale":"The reader's weakest assumption—that BT-Settl CIFIST model atmospheres correctly predict broadband fluxes—captures the foundation of both the radius scale and the tight-sequence claim. My pass sharpens this to a specific sub-mechanism: the Section 4.2.5 metallicity correction is computed from the same model grids, so any systematic model error enters twice, once in the raw radii and again in the correction applied to them. The post-correction scatter of 2.4% is almost exactly the quadrature sum of the quoted uncertainties (1.6% radius and 1.7% metallicity), leaving no headroom for intrinsic spread. If F(LSED) is biased, the final scatter could be artificially small. The sign error in Eq. 4 is a concrete, checkable flaw in the manuscript's equations; even if the code is correct, it signals that the analytical derivation has not been independently verified. These issues do not invalidate the central measurement—the inflation signal appears robust across comparisons with DEBs and interferometry—but they make the paper's most striking claim (intrinsic spread <1–2%) conditional on model fidelity and on a correct metallicity correction. The reader's CONDITIONAL verdict remains appropriate, and I recommend no change.","tokens_in":27186,"tokens_out":13039,"duration_ms":135079,"concrete_test":"Recompute F(LSED) and the Section 4.2.5 correction using an independent model atmosphere grid (e.g., Phoenix-2012 or MARCS) and apply it to the same Terrien+Gaidos sub-sample. If the corrected radius scatter changes by more than 0.5% or the median inflation shifts by more than 1%, the <1–2% intrinsic-spread claim is not robust to the model atmosphere assumption. Additionally, verify Eq. 4 by re-deriving the analytical minimisation; if the printed sign is incorrect, issue a correction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's strongest claim—that M-dwarfs have <1–2% intrinsic radius spread—is established only for a sub-sample with measured [Fe/H] (Section 4.2.5), after correcting radii using F(LSED), a metallicity-dependent correction computed from the same BT-Settl/AGSS2009 model atmospheres used for all fits. If those atmospheres have systematic flux errors (for example the 4000 K grid discontinuity the authors themselves identify, or molecular opacity errors in the optical/NIR), then F(LSED) is biased and the corrected radii are not reliable. The post-correction scatter is 2.4%, nearly identical to the quadrature sum of the assumed 1.6% radius uncertainty and 1.7% metallicity-induced uncertainty, so there is no budget for intrinsic spread; any underestimate of these uncertainties or any bias in F(LSED) would make the tight-sequence conclusion an artifact. In addition, Equation (4) as printed has the wrong sign for the analytical radius minimisation: minimizing chi^2 = sum((m_i - Z_i + x)^2/sigma_i^2) over x = 5 log10(R/d) gives x = + sum((Z_i - m_i)/sigma_i^2)/sum(1/sigma_i^2), so log10(R^2/d^2) should have a positive, not negative, sign; if the code followed the printed equation the radii would be systematically inverted. This suggests the manuscript's equations have not been independently checked. The model-atmosphere premise is therefore doubly load-bearing: it sets the radius scale and defines the correction that produces the headline result.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper develops a spectral energy distribution (SED) fitting method to measure effective temperatures and radii for 15,274 M-dwarfs within 100 pc using Gaia DR2 parallaxes and multi-band photometry (Gaia BP/RP, 2MASS, WISE). The method uses BT-Settl CIFIST model atmospheres to fit the shape of the SED and the dilution factor, with a log(g) prior from Baraffe et al. isochrones. The authors derive empirical TSED-R and LSED-R relations, find radii inflated by 3-7% relative to theoretical isochrones, argue that the intrinsic scatter in the inflated sequence is at most 1-2%, and conclude that magnetic activity is unlikely to explain the inflation. They also provide metallicity-dependent corrections and discuss practical strategies for measuring accurate M-dwarf radii.","tokens_in":27536,"tokens_out":7489,"duration_ms":72389,"significance":"The paper provides the largest homogeneous sample of M-dwarf radii to date and offers a method that can be applied to any star with photometry and a parallax. The derived empirical relations are valuable for exoplanet host star characterization and for testing stellar structure models. The paper is transparent about many systematic checks (spot simulations, contamination, activity correlations) and makes the full catalogue publicly available, which is a strength. The main claims, if confirmed, would significantly constrain the radius inflation mechanism and provide falsifiable predictions that can be tested with future data.","major_comments":[{"comment":"Equation (4) has a sign error. Minimizing χ² = Σ((m_i - Z_i + x)²/σ_i²) over x = 5 log10(R/d) gives x = +Σ((Z_i - m_i)/σ_i²)/Σ(1/σ_i²), so log10(R²/d²) = +0.4 Σ((Z_i - m_i)/σ_i²)/Σ(1/σ_i²). The printed equation has a minus sign. If the analysis code followed the printed equation, the fitted dilution factors—and hence all radii—would be systematically inverted. Please verify the sign in the code and correct Equation (4).","section":"2.4, Eq. (4)"},{"comment":"The claim of a tight intrinsic sequence with scatter below 1-2% is an upper limit rather than a measured value. After applying the metallicity correction, the residual scatter is 2.4%, which equals the quadrature sum of the 1.6% radius uncertainty and the 1.7% metallicity-induced uncertainty, leaving no statistical budget for an intrinsic spread. The paper should present this as an upper limit (as it does in the text 'at most 1-2%') and avoid stating in the abstract and conclusions that the spread is 'no more than' a measured quantity. In addition, F(LSED) in Equation (16) is computed from the same BT-Settl AGSS2009 atmosphere family used for the fits; if those atmospheres have systematic flux errors, the correction and the resulting tight-sequence conclusion are biased. A sensitivity test with an independent atmosphere grid would significantly strengthen this claim.","section":"4.2.5"},{"comment":"The correlations with activity indicators are performed on radius residuals [R−Rfit(LSED)]/R that have not been corrected for metallicity. Figure 18 shows that these uncorrected residuals span ±6% as a function of [Fe/H], so the metallicity-induced scatter can dilute any genuine activity-radius correlation. The authors should re-run these correlation tests on metallicity-corrected radii (Equation 16) or include [Fe/H] as a covariate in the regression before concluding that no correlation exists.","section":"4.2.4"},{"comment":"The uncertainty estimates are based on the full 3D grid search for only 158 stars (1% of the sample), and this characteristic uncertainty is then applied to all stars. While this is a reasonable approximation, the paper should explicitly state that this assumes the uncertainty distribution of the 1% subsample is representative of the entire sample, and it should provide some evidence (e.g., a comparison of the distributions of photometric quality or fitted parameters) that this representative assumption holds.","section":"2.3"}],"minor_comments":[{"comment":"The phrase 'no more than a 1-2% intrinsic spread' should be phrased as 'an upper limit of 1-2%' to reflect the fact that the intrinsic spread is not directly measured but rather bounded by the uncertainty budget.","section":"Abstract"},{"comment":"The paper switches between TSED and Teff in the text and figures; for clarity, define TSED as the SED-derived effective temperature and use it consistently, and note explicitly that TSED is assumed equal to the physical Teff.","section":"2.2 and throughout"},{"comment":"The caption should note that the 68% confidence contours are based on the randomly selected 1% subsample, as stated in the text, so that readers do not infer that these are per-star uncertainties for the full sample.","section":"Figure 6"},{"comment":"The spot simulation adopts a fixed spot temperature ratio Tspot = 0.8 Timac; the conclusion that spots can reproduce the observed scatter may depend on this choice, and a brief discussion of how varying this ratio affects the results would be useful.","section":"4.2.3"}],"recommendation":"major_revision","confidential_remarks":"The sign error in Eq. (4) is the most serious issue; if the code is correct, it is a typographical error, but the authors must confirm by checking the code. The overstatement of the intrinsic-spread claim and the uncorrected activity correlations are fixable with revision. The paper's core method and dataset are valuable, and the public catalogue is a significant community resource. I recommend major revision rather than rejection, and I would be willing to review a revised version."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this paper gives the largest homogeneous M-dwarf radius sample to date (15,274 stars) from Gaia DR2 distances plus broadband photometry, using a modified SED-fitting method. The 3–7% inflation relative to theoretical models is consistent with earlier work; the more striking claim is that the intrinsic radius spread is <1–2%. That second claim is the one to scrutinize.\n\nThe method itself is sound in outline: they fit synthetic photometry from BT-Settl atmospheres to Gaia, 2MASS, and WISE bands, and convert the fitted dilution factor to a radius using geometric distances. They benchmark against DEBs and interferometric radii, and the overall scale agrees. They also ship the catalog and code, which is genuinely useful. The empirical T-R and L-R relations will get cited for exoplanet host work. And they make a nice point that fitting a metallicity-spread sample with solar-metallicity atmospheres creates a spurious radius–metallicity correlation—that is a real diagnostic contribution.\n\nNow the soft spots, in proportion. First, the sign error in Eq. 4 is real. Minimizing their Eq. 3 with respect to x = 5 log10(R/d) gives x = + sum((Z_i - m_i)/sigma_i^2)/sum(1/sigma_i^2), so log10(R^2/d^2) should have a positive sign, not negative. The published equation would produce radii that are systematically inverted; presumably the code uses the correct sign, since the results make sense. Still, an uncaught sign error in a key equation suggests the manuscript was not carefully proofread, and it should be fixed before publication.\n\nSecond, the <1–2% intrinsic spread claim is only as strong as the metallicity correction F(LSED), which is computed from the same model atmospheres used for the fits. The authors themselves note a 4000 K grid discontinuity in BT-Settl. If those atmospheres have systematic flux errors in the optical/NIR, both the radius scale and the metallicity correction shift. The post-correction scatter is 2.4%, almost exactly the quadrature sum of their 1.6% radius uncertainty and 1.7% metallicity uncertainty. That leaves no room for intrinsic spread, which is fine as an upper limit, but it means the claim is really an upper limit whose tightness depends on those error estimates being complete. The authors should state this more carefully.\n\nThird, minor: the abstract says “accuracy of 2.4%” where they mean precision. And the magnetic-activity samples are modest (189 rotation periods, 95 X-ray, 573 H-alpha); the conclusion that magnetism cannot explain inflation is reasonable but not as ironclad as the text sometimes suggests.\n\nOverall: the core measurement is well-posed and externally validated, and the paper is a real contribution. The soft spots are addressable. This deserves a serious referee, and I would bring it to the reading group—the metallicity-correction discussion alone is worth the hour.\n\nRecommendation: send it to review, but with a referee who will check the equations and press on the model dependence of the intrinsic-spread claim.","headline":"Large homogeneous M-dwarf radius sample with a solid core; the tight intrinsic-spread claim leans on a model-dependent correction, and the printed Eq. 4 has a sign error that needs fixing.","tokens_in":28040,"tokens_out":2693,"would_cite":true,"duration_ms":28749,"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":"M-dwarf radii are inflated by 3–7% relative to theoretical models, and the inflated sequence is intrinsically tight, with less than 1–2% spread.","keywords":["M-dwarfs","radius inflation","spectral energy distribution fitting","Gaia DR2","stellar models","metallicity","magnetic activity","low-mass stars"],"falsifier":"Interferometric radii for the same stars, derived from Gaia distances and compared band-by-band with the SED-fitted radii, would settle the claim: if the two radius scales disagree by more than the quoted 2.4% in a temperature-dependent way, the model-atmosphere temperature scale is implicated rather than a physical inflation.","tokens_in":27000,"feed_emoji":"🌟","tokens_out":8346,"duration_ms":83274,"temperature":0.7,"pith_summary":"This paper adds a fourth route to M-dwarf radii: fit the whole spectral energy distribution to model-atmosphere synthetic photometry, then combine the fitted angular scale with a Gaia DR2 geometric distance. Applied to 15,274 nearby stars, the method returns temperatures, luminosities, and radii that show a 3–7% inflation relative to purely theoretical isochrones below roughly 4000 K. The paper argues that the inflated sequence is intrinsically very tight—after correcting for metallicity, the intrinsic spread is at most 1–2%—and that this tightness, together with the absence of activity–radius correlations, rules out magnetic inflation as the explanation. If true, this gives exoplanet and stellar-modelling communities a large homogeneous sample and empirical radius relations anchored in archival photometry alone.","feed_headline":"M-dwarf radii run 3–7% larger than models","feed_subtitle":"A 15,000-star sample shows the inflation is intrinsic, with magnetism ruled out and metallicity precision the limit.","key_machinery":"The load-bearing object is the modified spectral energy distribution fit. Synthetic photometry is produced by folding the BT-Settl CIFIST model-atmosphere grid through the Gaia, 2MASS, and AllWISE passbands; the shape of the eight-band SED fixes $T_{\\rm SED}$ and $\\log(g)$, while the radius is obtained analytically from the dilution factor that minimises $\\chi^2$: $\\log_{10}(R^2/d^2) = -0.4\\,(\\sum_i (Z_i-m_i)/\\sigma_i^2)/(\\sum_i 1/\\sigma_i^2)$, with $d$ from Bailer-Jones et al. (2018). Because the fit acts only on the photosphere, it does not assume an interior model, which is what allows the paper to confront interior models with data. Two additional interpolated metallicity grids ($[{\\rm M/H}]=\\pm 0.25$) provide the correction $F(L_{\\rm SED})\\,[{\\rm Fe/H}]$ that removes the apparent metallicity–radius correlation and reduces the radius scatter to 2.4%.","core_discovery":"The central claim is that main-sequence M-dwarfs do not match the radii of purely theoretical models: at fixed luminosity, the measured radii are larger by 3–7%, while the empirical PARSEC 1.2S models, which adopt an observationally calibrated temperature–optical-depth relation, trace the inflated sequence. The paper also claims that the inflated sequence is remarkably coherent, with an intrinsic scatter no larger than 1–2%, after accounting for a ~1.7% radius scatter introduced by metallicity measurement uncertainties and the ~1.6% fitting uncertainty. This tightness, plus the lack of correlation between radius residual and rotation period, Rossby number, X-ray luminosity, or H-$\\alpha$ activity, leads the authors to conclude that stellar magnetism is currently unable to explain the inflation. The fitted sample yields empirical $R(T_{\\rm SED})$ and $R(L_{\\rm SED})$ relations, with the luminosity–radius relation expressed as a correction to the Dotter et al. (2008) solar-metallicity isochrone.","pith_inferences":["A reader could test the atmosphere dependence directly: cross-calibrating these SED-fitted radii against interferometric radii for the same stars would expose any temperature-scale bias hidden in the 3–7% inflation claim.","The 4000 K discontinuity in the CIFIST grid could be used as a natural experiment—comparing bolometric corrections across that gap with independent spectrophotometry would tell whether the gap is a model artifact or a real spectral feature.","The paper's tight-sequence result implies that spot coverage must be strikingly uniform across M-dwarfs; measuring spot filling factors from rotational light curves of stars spanning the same $T_{\\rm SED}$ range would test whether that homogeneity is real.","The metallicity-correction calibration, built here from two literature samples, could be extended to larger spectroscopic surveys with higher-precision metallicities, which would show whether the 1.7% metallicity-limited scatter is a floor or an artefact of those catalogues."],"forward_implications":["Exoplanet transit radii around M-dwarfs could be measured to roughly 2% accuracy from archival photometry plus a metallicity measurement, without new spectra or eclipses.","Stellar evolution codes that predict smaller radii below 4000 K are missing physics; the empirical $R(L_{\\rm SED})$ relation gives them a quantitative target.","Magnetic inflation models, which predict a spread of radii at fixed mass or luminosity, are constrained to saturate at rotation rates slower than the slowest rotators in the sample if they are to survive.","The previously reported correlation between M-dwarf radius and metallicity at fixed luminosity is reinterpreted as a fitting artifact of solar-metallicity atmospheres, not a physical structural effect.","The dominant source of error in M-dwarf radii becomes the precision of metallicity measurements, redirecting effort toward better stellar metallicities."],"supporting_citations":[{"why":"Supplies the geometric distances used to convert the fitted R/d ratio into radius and to compute luminosity.","marker":"Bailer-Jones et al. (2018)"},{"why":"Provides the BT-Settl CIFIST model atmosphere grid from which the synthetic photometry is generated.","marker":"Allard et al. (2012)"},{"why":"Provides the log(g) prior through isochrone matching and serves as one of the theoretical comparison isochrones.","marker":"Baraffe et al. (2015)"},{"why":"Serves as the reference isochrone against which the radius inflation and luminosity–radius correction are measured.","marker":"Dotter et al. (2008)"},{"why":"Supplies the spectroscopic-temperature comparison sample used to test the radius inflation measured by different methods.","marker":"Mann et al. (2015)"},{"why":"Supplies the interferometric radii comparison sample used to check the inflation level.","marker":"Boyajian et al. (2012)"},{"why":"Supplies the detached eclipsing binary radii and metallicity constraints used for method comparison.","marker":"Parsons et al. (2018)"},{"why":"Provides the Gaia DR2 astrometry and photometry that anchor the distances and blue-optical SED points.","marker":"Gaia Collaboration et al. (2018)"},{"why":"Supplies the 2MASS JHKs photometry sampling the peak of the M-dwarf SED.","marker":"Skrutskie et al. (2006)"},{"why":"Supplies the AllWISE W1-W3 photometry that constrains the Rayleigh-Jeans tail of the SED.","marker":"Wright et al. (2010)"}],"fun_headline_variants":["M-dwarf radii 3–7% over model predictions","15,000 M-dwarfs: radius inflation is intrinsic","M-dwarf inflation: magnetism can't explain it","Gaia DR2: M-dwarf radii inflated by 3–7%","M-dwarf radii: models miss by 3–7%"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The method trusts that the synthetic model atmospheres predict correct relative fluxes across the eight photometric bands; if they do not—and the paper itself flags a 4000 K discontinuity in the CIFIST grid—the fitted temperatures and radii, and hence the inflation and tightness claims, would shift.","fun_headline_variants_meta":{"raw":{"variants":["M-dwarf radii 3–7% over model predictions","15,000 M-dwarfs: radius inflation is intrinsic","M-dwarf inflation: magnetism can't explain it","Gaia DR2: M-dwarf radii inflated by 3–7%","M-dwarf radii: models miss by 3–7%"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000543,"raw_usage":{"total_tokens":2613,"prompt_tokens":972,"completion_tokens":1641,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":588,"completion_tokens_details":{"reasoning_tokens":1552}},"tokens_in":588,"tokens_out":1641,"duration_ms":14718,"temperature":1.0,"reasoning_tokens":1552,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:27:41.840468+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Interferometric radii for the same stars, derived from Gaia distances and compared band-by-band with the SED-fitted radii, would settle the claim: if the two radius scales disagree by more than the quoted 2.4% in a temperature-dependent way, the model-atmosphere temperature scale is implicated rather than a physical inflation.","supporting_citations":[],"review_version":1}