{"id":"4f4c53a3-35ea-4820-9cda-307520ea921c","arxiv_id":"2502.02568","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Higher-order Balmer lines in M dwarfs show larger intrinsic short-term variability, and the photometric amplitude-activity relation is weak for all four lines.","lead":"This paper tracks how four hydrogen emission lines in 77 small, cool stars brighten and fade over tens of minutes, alongside brightness data from the TESS satellite. The work matters because M dwarfs are the most common stars and their magnetic activity can distort planet detection and habitability signals.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The higher-order Balmer variability gradient rests on Eq. 2's assumption that quoted EW uncertainties are complete; unquantified blue-side systematics (lower S/N, continuum placement, residual fringing) could plausibly produce the trend, so the central claim is not yet secured.","rationale":"The paper is a solid observational contribution with new data, new chi values, and a RADYN analysis; it is not internally inconsistent. The reader's weakest assumption is the right one, and the paper itself flags the S/N issue but does not quantify a systematic-error floor. Because the central result depends on the subtraction in Eq. 2, and because a plausible systematic could explain the trend, the conditional verdict is appropriate. I agree with the reader's identification; no verdict change is needed beyond the already-conditional status.","tokens_in":25014,"tokens_out":2831,"duration_ms":29176,"concrete_test":"For every star with >=6 spectra in §5.2, recompute sigma_intrinsic after replacing sigma_measurement with sqrt(sigma_measurement^2 + sigma_sys^2), where sigma_sys is the epoch-to-epoch RMS of the continuum-normalized flux in the adjacent line-free continuum regions (Table 5), scaled to EW units. If the median sigma_intrinsic no longer increases monotonically from H-alpha to H-delta, or if the H-gamma/H-delta values drop by more than ~30%, the claimed gradient is not robust. A complementary check is to repeat the analysis on a constant star or on repeated observations of the same star on a single night without flares.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim of §5.2—that H-beta, H-gamma, and H-delta show progressively larger intrinsic short-term variability than H-alpha—is inferred from sigma_intrinsic = sqrt(sigma_observed^2 - sigma_measurement^2)/|median| (Eq. 2), with sigma_observed computed as half the 16th-84th percentile spread of EW measurements across epochs. This subtraction isolates astrophysical variability only if sigma_measurement captures all non-intrinsic scatter. The text explicitly notes in the Figure 3 discussion that the apparent variability 'also includes the impact of decreased signal-to-noise towards the bluer wavelengths,' and fringing is excluded only by a qualitative per-star visual cut. For H-gamma and H-delta, the median EW is also smaller, so any fixed systematic error floor contributes more to sigma_intrinsic after normalization. If measurement uncertainties are underestimated—e.g., by continuum-region misplacement (Table 5 regions are narrow), imperfect fringing subtraction, or flux-calibration errors—the monotonic rise in Figures 4-5 would be an artifact. The decreasing star counts (61, 48, 36, 16) make the trend partly a statement about a shrinking subsample.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper analyzes short-term (≥15 min) variability in Hα, Hβ, Hγ, and Hδ equivalent widths for 77 M dwarfs, combining new OSMOS/ModSpec spectroscopy with TESS photometry. It presents new χ factors converting equivalent widths to Balmer luminosities, updates the Rvar–activity relation, quantifies intrinsic line variability via Eq. (2), reports flare behavior in three stars, phase-folds observations of TIC 283866910, and fits a RADYN model to a flare. The main conclusions are that the Rvar–activity relation is weak for all four Balmer lines, that higher-order Balmer lines show progressively larger intrinsic short-term variability, and that TIC 283866910 shows an anti-correlation between Hα emission and TESS flux consistent with dark spots.","tokens_in":25302,"tokens_out":6762,"duration_ms":63350,"significance":"If the intrinsic-variability gradient is real, it is a useful observational constraint on chromospheric heating and flare microphysics in M dwarfs, and the published χ factors plus equivalent-width time series will be a valuable community resource. The paper is careful in describing data reduction, provides machine-readable tables and software links, and makes the observational products reproducible. However, the central gradient currently rests on an unvalidated quadrature subtraction of measurement errors and on qualitative fringing cuts; the dark-spot interpretation for TIC 283866910 likewise depends on an assumed phase zero-point. These load-bearing points need additional support before the headline claims can be regarded as established.","major_comments":[{"comment":"The central claim that higher-order Balmer lines are intrinsically more variable than Hα rests on subtracting σ_measurement^2 from σ_observed^2 in quadrature. The text itself notes, in the discussion of Figure 3, that the apparent variability 'also includes the impact of decreased signal-to-noise towards the bluer wavelengths,' and fringing is excluded only by a per-star visual cut. If the quoted EW uncertainties undercount continuum-placement errors, residual fringing, or flux-calibration systematics—which are not quantified anywhere in the paper—the inferred σ_intrinsic gradient could be an artifact. Because the sample shrinks from 61 stars for Hα to 16 for Hδ, the trend is also partially a statement about a changing subsample. I request robustness tests: (i) show that a systematic error floor added to σ_measurement does not erase the gradient; (ii) recompute the gradient on the subset of stars with all four lines measured; and (iii) quantify how σ_intrinsic changes when the Table 5 continuum regions are perturbed.","section":"§5.2, Eq. (2)"},{"comment":"The denominator in Eq. (2) is the median EW, which is systematically smaller for Hγ and Hδ than for Hα. A fixed absolute systematic uncertainty in the EW therefore maps to a larger fractional σ_intrinsic for the higher-order lines even if the astrophysical variability is identical. This bias is separate from the completeness of σ_measurement. The paper should report absolute σ_intrinsic (in Å) alongside the fractional value, or demonstrate that the gradient survives when a constant absolute error of plausible size is added to all lines.","section":"§5.2, Eq. (2) normalization"},{"comment":"The dark-spot conclusion for TIC 283866910 depends on the phase-folding zero-point 2458423.7150, which the authors state is an 'assumption' for the epoch used by Medina et al. (2022a). They also report that they were unable to reproduce the Medina et al. figure with that zero-point. A different zero-point can change the sign of the correlation between Hα EW and TESS flux, so the anti-correlation shown in Figure 9 is not yet robust. The authors' own new data yield F-test = 3 with p = 0.08, only marginal. I ask that the authors justify the zero-point from the data or demonstrate that the anti-correlation persists over a plausible range of zero-points.","section":"§5.4"}],"minor_comments":[{"comment":"The caption says the intrinsic variability is 'calculated by Equation 5.2'; this cross-reference should be to Eq. (2).","section":"Figure 4 caption"},{"comment":"The text reports F-test = 33 for the re-analysis and F-test = 3 for the 2020 December data, but Table 6 lists F-test = 2.84 for the 2020 December row; the text and table should be reconciled.","section":"§5.4 and Table 6"},{"comment":"The line widths are described as visually selected for each star; please clarify whether the same integration windows are used for all epochs of a given star, since varying the window between epochs would inflate σ_observed independently of astrophysical variability.","section":"Table 5"},{"comment":"Please state how many stars were excluded by the fringing criterion for each Balmer line, so the reader can assess the impact of that qualitative cut on the reported sample sizes.","section":"§5.2"},{"comment":"The table heading appears as 'T able 1.Parameters'; this formatting typo should be corrected.","section":"Table 1 heading"}],"recommendation":"major_revision","confidential_remarks":"To the editor: this is a solid observational paper with valuable data products, but the headline intrinsic-variability gradient is not fully secured because Eq. (2) requires complete measurement uncertainties, which are not demonstrated. I am recommending major revision rather than rejection because the requested tests are within the manuscript's scope and the paper otherwise makes a clear empirical contribution. The paper fits the journal's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a straightforward, honest extension of the G23 program: 77 M dwarfs, four Balmer lines, new chi factors for H-beta/H-gamma/H-delta, a RADYN flare fit for one event, and a dark-spot interpretation for one star. The data publication is thorough, and the amplitude-activity analysis is a clean null result—weak positive correlations that don't reach significance. The RADYN section is a genuine extra: the grid search is transparent, and the fact that the favored low-energy beam predicts a TESS-band luminosity matching the observed flare is a good internal consistency check.\n\nThe main soft spot is the paper's central claim that higher-order Balmer lines show progressively larger intrinsic variability. That claim is built on Eq. 2, subtracting sigma_measurement in quadrature from the observed EW spread. If the quoted EW uncertainties miss continuum misplacement, residual fringing, or flux-calibration errors—all plausible for narrow continuum windows and visually selected line widths—the trend toward H-gamma and H-delta could be partly artifact. The paper itself admits in the Figure 3 discussion that the apparent variability \"also includes the impact of decreased signal-to-noise towards the bluer wavelengths,\" and the star count drops from 61 to 16 across the four lines. Reassuringly, the same trend was already reported by Duvvuri et al. (2023), so the paper is extending rather than originating this result, but the extension deserves a firmer statistical foundation.\n\nThe dark-spot detection for TIC 283866910 is weaker than the abstract suggests. The new data give an F-test of 3 with p = 0.08, marginally significant at best, and the literature re-analysis depends on an assumed ephemeris zero-point. The authors are candid that they cannot reproduce the Medina figure, which is honest but means the anti-correlation result should be framed as tentative.\n\nThe RADYN beam parameters are fit to the data rather than predicted, so that section is an interpretation, not a test. But the luminosity agreement is a nice cross-check.\n\nOverall: the data and the careful treatment of periods, EWs, and chi values make this a useful contribution. The central variability gradient needs a systematic-uncertainty analysis to be secure. I'd send it to a serious referee, with the request that they quantify how the trend responds to plausible floor errors in the bluer lines and to less aggressive visual fringing cuts.","headline":"Solid observational extension of the G23 Balmer-line program, but the headline intrinsic-variability gradient rests on an error-subtraction assumption that the paper itself flags as contaminated by blue-side S/N, and the dark-spot claim for TIC 283866910 is suggestive, not solid.","tokens_in":25840,"tokens_out":1902,"would_cite":true,"duration_ms":21577,"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":"In M dwarfs, the blue Balmer lines H-beta through H-delta are intrinsically more variable than H-alpha, and photometric spot amplitude does not predict line activity.","keywords":["M dwarfs","Balmer lines","chromospheric activity","stellar flares","starspots","stellar rotation","TESS photometry","equivalent width"],"falsifier":"Re-observe a subset of these stars with a fringing-free, high-signal-to-noise blue spectrograph at the same cadence and recompute $\\sigma_{\\rm intrinsic}$ for each line: if the median intrinsic variability of H$\\delta$ is no longer larger than that of H$\\alpha$ once the bluer lines are measured cleanly, the claimed gradient is a measurement artifact, whereas if the ordering H$\\alpha$ < H$\\beta$ < H$\\gamma$ < H$\\delta$ survives, the trend is astrophysical.","tokens_in":24830,"feed_emoji":"⭐","tokens_out":23748,"duration_ms":181359,"temperature":0.7,"pith_summary":"This paper analyzes 77 M dwarfs observed with ground-based optical spectroscopy combined with TESS photometry, and argues that short-term intrinsic variability grows systematically from H$\\alpha$ through H$\\beta$, H$\\gamma$, to H$\\delta$ on timescales of roughly 15 minutes to an hour, far shorter than a rotation period. It also re-examines the amplitude-activity relation and finds that the link between photometric spot amplitude and Balmer-line luminosity is weak and statistically insignificant for all four lines, and becomes more dispersed toward higher-order lines. The stakes are practical: M dwarfs host most small exoplanets, and the same magnetic activity that produces this line flicker contaminates exoplanet detection and characterization. The paper additionally reports Balmer-line flares that appear without corresponding TESS white-light flares, shallow Balmer decrements during flares, and evidence that one young M dwarf's active region is darker than its photosphere.","feed_headline":"77 M dwarfs: bluer Balmer lines flicker more than H-alpha","feed_subtitle":"Spot amplitude fails to predict the chromospheric flicker that contaminates exoplanet data.","key_machinery":"The load-bearing object is the fractional intrinsic variability parameter $\\sigma_{\\rm intrinsic} = \\sqrt{\\sigma_{\\rm observed}^2 - \\sigma_{\\rm measurement}^2} / |{\\rm Median(measurement)}|$, which takes the scatter of each star's measured equivalent widths, subtracts the reported per-measurement error in quadrature, and normalizes by the median line strength so that stars can be compared. The entire claim that higher-order lines are more intrinsically variable rests on this subtraction isolating astrophysical signal from noise. Two supporting tools carry the rest of the analysis: the $\\chi$ factor, the ratio of the continuum flux near a line to the bolometric flux, converts equivalent widths into $L_{\\rm line}/L_{\\rm bol}$ without flux calibration (recomputed here for all four lines from model spectra), and the Balmer decrement relative to H$\\beta$ tracks how relative line fluxes change during flares. For flare energetics, the paper uses RADYN, a one-dimensional non-local-thermodynamic-equilibrium radiative-hydrodynamic code, to forward-model the H$\\alpha$ and H$\\beta$ profiles of the strongest observed flare and identify the electron-beam parameters that best reproduce them.","core_discovery":"On its own terms, this paper establishes three connected results. First, using a fractional intrinsic variability parameter that subtracts reported measurement errors in quadrature from the observed scatter of equivalent widths, it finds that higher-order Balmer lines are progressively more variable: the distribution of $\\sigma_{\\rm intrinsic}$ shifts toward larger values from H$\\alpha$ to H$\\beta$ to H$\\gamma$ to H$\\delta$ across the 61 stars with more than six spectra. Second, the Spearman correlation between $R_{\\rm var}$ (the semi-amplitude of the TESS light curve) and $L_{\\rm line}/L_{\\rm bol}$ is weak and not statistically significant for H$\\alpha$ ($\\rho \\approx 0.39$), H$\\beta$ ($\\rho \\approx 0.43$), H$\\gamma$ ($\\rho \\approx 0.49$), or H$\\delta$ ($\\rho \\approx 0.36$), meaning photometric spot amplitude does not determine chromospheric line output. Third, for stars observed simultaneously in TESS and in Balmer lines, white-light flares and Balmer flares frequently do not coincide: TIC 415508270 shows strong, long Balmer flares during weak TESS activity, and Balmer-only flares appear in TIC 220044948. The paper also shows that the Balmer decrement (line flux relative to H$\\beta$) becomes shallower during Balmer-detected flares and differs between flares on the same star, and that in TIC 283866910 the H$\\alpha$ equivalent width is anti-correlated with TESS flux, indicating an active region darker than the photosphere.","pith_inferences":["If the bluer lines are both more variable and noisier, a decisive test of the gradient would be repeat observations at higher signal-to-noise with fringing-free blue coverage: if the H$\\gamma$/H$\\delta$ excess over H$\\alpha$ persists, the trend is astrophysical, and if it shrinks, part of it is instrumental (a test the authors' own caveat invites).","If Balmer-only, sub-white-light flares are as common as this sample suggests, then white-light flare statistics—and the UV irradiation doses used in exoplanet habitability studies—systematically undercount low-energy flaring on M dwarfs.","The anti-correlation of H$\\alpha$ with TESS flux for TIC 283866910 suggests a general diagnostic: the phase relationship between chromospheric lines and broadband photometry can reveal whether active regions are darker or brighter than the photosphere, and applying the same F-test to a larger sample with full rotation-phase coverage could map spot contrasts across spectral types.","The newly tabulated $\\chi$ values for H$\\beta$, H$\\gamma$, and H$\\delta$ let other groups convert published equivalent widths into line luminosities for stars down to 2300 K, which could homogenize activity measurements across surveys that use different lines."],"forward_implications":["Because H$\\gamma$ and H$\\delta$ carry the largest intrinsic variability per epoch, they are the most sensitive chromospheric tracers of short-term changes on ~15–60 minute timescales.","$R_{\\rm var}$ is not a dependable predictor of an individual star's chromospheric activity: the amplitude-activity relation is weak for H$\\alpha$ and statistically insignificant for all four lines, so photometric spot amplitude alone cannot rank activity levels.","Single-band flare surveys are incomplete: Balmer-line flares occur without detectable TESS white-light flares, and some TESS flares have no Balmer counterpart, so white-light monitoring alone underestimates the flare budget of M dwarfs.","The Balmer decrement shallows during Balmer-detected flares, consistent with higher-order lines brightening relative to lower-order lines as the latter become optically thick, and the decrement's shape differs between flares on the same star.","The RADYN fits attribute the largest observed flare to a modest electron beam (flux ≈ $10^{11}$ erg s$^{-1}$ cm$^{-2}$, electron energies 10–40 keV), providing concrete inputs for modeling the photochemical irradiation of M dwarf exoplanet atmospheres."],"supporting_citations":[{"why":"Supplies the original dataset and the H$\\alpha$ amplitude-activity relation that this paper extends to H$\\beta$, H$\\gamma$, and H$\\delta$.","marker":"García Soto et al. (2023)"},{"why":"Reported the pattern of increasing variability toward higher-order Balmer lines that this paper's $\\sigma_{\\rm intrinsic}$ result is consistent with.","marker":"Duvvuri et al. (2023)"},{"why":"Provides the H$\\alpha$ basal absorption correction, continuum-region conventions, and the prior M dwarf H$\\alpha$ activity framework used for equivalent-width measurement.","marker":"Newton et al. (2017)"},{"why":"Introduces the $\\chi$ factor that converts equivalent widths to $L_{\\rm line}/L_{\\rm bol}$, the method recomputed here for all four lines.","marker":"Walkowicz et al. (2004)"},{"why":"Supplies the model-template procedure for deriving $\\chi$ values that the paper follows, with differences attributed to continuum placement.","marker":"Núñez et al. (2024)"},{"why":"Provides the continuum regions for H$\\beta$, H$\\gamma$, and H$\\delta$ used in the equivalent-width measurements.","marker":"Walkowicz & Hawley (2009)"},{"why":"Establishes the electron-beam RADYN flare modeling and the expectation that the Balmer decrement shallows during flares.","marker":"Allred et al. (2006)"},{"why":"Reported the H$\\alpha$-rotation correlation for TIC 283866910 that this paper re-tests and does not reproduce.","marker":"Medina et al. (2022a)"},{"why":"Provides the grid of self-consistent RADYN flare models searched to fit the observed H$\\alpha$ and H$\\beta$ profiles.","marker":"Kowalski et al. (2024)"}],"fun_headline_variants":["Higher-order Balmer lines vary more in M dwarf survey","M dwarf chromospheric flicker: bluer lines vary more","Photometry can't predict M dwarf Balmer line variability","TESS misses some M dwarf flares seen in Balmer lines","Dark spot on TIC 283866910 hints at active region"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result that higher-order lines are more intrinsically variable assumes that the reported measurement uncertainties fully capture every non-stellar source of scatter, so that subtracting them in quadrature leaves only true astrophysical variability; if errors are underestimated for the bluer, noisier lines, or if fringing and continuum misplacement add scatter the error bars do not include, the apparent H$\\gamma$/H$\\delta$ excess could be partly instrumental.","fun_headline_variants_meta":{"raw":{"variants":["Higher-order Balmer lines vary more in M dwarf survey","M dwarf chromospheric flicker: bluer lines vary more","Photometry can't predict M dwarf Balmer line variability","TESS misses some M dwarf flares seen in Balmer lines","Dark spot on TIC 283866910 hints at active region"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000282,"raw_usage":{"total_tokens":1810,"prompt_tokens":1231,"completion_tokens":579,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":847,"completion_tokens_details":{"reasoning_tokens":494}},"tokens_in":847,"tokens_out":579,"duration_ms":5782,"temperature":1.0,"reasoning_tokens":494,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T11:42:40.625587+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-observe a subset of these stars with a fringing-free, high-signal-to-noise blue spectrograph at the same cadence and recompute $\\sigma_{\\rm intrinsic}$ for each line: if the median intrinsic variability of H$\\delta$ is no longer larger than that of H$\\alpha$ once the bluer lines are measured cleanly, the claimed gradient is a measurement artifact, whereas if the ordering H$\\alpha$ < H$\\beta$ < H$\\gamma$ < H$\\delta$ survives, the trend is astrophysical.","supporting_citations":[{"cited_title":"R., Douglas, S","cited_arxiv_id":null,"evidence_quote":"Supplies the original dataset and the H$\\alpha$ amplitude-activity relation that this paper extends to H$\\beta$, H$\\gamma$, and H$\\delta$."},{"cited_title":"C., Hawley, S","cited_arxiv_id":null,"evidence_quote":"Establishes the electron-beam RADYN flare modeling and the expectation that the Balmer decrement shallows during flares."}],"review_version":1}