{"id":"184c7518-efa2-46af-95f3-b4189c0892f5","arxiv_id":"1908.02698","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"TESS observations of 149 M dwarfs show no correlation between flare occurrence and the rotational phase of the dominant starspot, confirming an earlier K2 result with a larger sample.","lead":"Using TESS two-minute photometry of 149 small, cool stars (M dwarfs), this paper finds that stellar flares happen at all rotation phases with no preference for the phase where the dominant starspot is most visible. The result challenges a solar-inspired expectation and matters for understanding M dwarf magnetic activity and for assessing flare impact on planets around these stars.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Phase-zero mapping is internally contradicted: Section 9 states the rotational modulation arises from multiple active regions, so the null phase result may reflect blurred spot labels rather than absence of flare-spot association.","rationale":"The reader's weakest assumption identifies the same load-bearing concern: the validity of φ0 as a proxy for the dominant spot longitude. The paper itself provides an internal contradiction: Sections 5 and 7 interpret the rotational modulation as evidence for a single large starspot, but Section 9 states that the sinusoidal pattern is produced by multiple active regions, with only one region responsible for the trough. This is a direct admission that φ0 may not isolate a single dominant spot. If multiple active regions are spread across the stellar surface, flares from those other regions will appear at all rotational phases, smoothing any phase–flare correlation and making the null result almost inevitable—regardless of whether the dominant spot actually drives flare production. The paper's attempt to address this by removing stars with obvious multiple-spot lightcurves is insufficient, because even sinusoidal lightcurves can be generated by several spots at similar longitudes or by a spot plus facular contributions. Thus, without an independent measurement of spot longitude (e.g., from Doppler imaging or a more detailed spot model), the central claim is not yet secure. The separate statistical error in §7.3—reversing the meaning of the KS and SW p-values—reinforces the need for caution but is not the primary reason for a conditional verdict. The χ² tests are the main quantitative evidence, and they appear correctly computed, but the phase-labeling ambiguity directly undermines their interpretation. A targeted test using Rayleigh statistics on high-count, single-spot candidates, or a comparison of phase concentration across modulation amplitude, would settle whether the null result is physical or an artifact of the phase definition.","tokens_in":14097,"tokens_out":5990,"duration_ms":69615,"concrete_test":"Select the subset of stars with a single, clearly sinusoidal lightcurve and ≥20 flares (e.g., TIC 231914259). Fit a one- or two-spot model to the out-of-flare lightcurve to derive the spot longitude and its uncertainty; then compute the Rayleigh statistic for the flare phases relative to both φ0 and the model spot longitude. If the Rayleigh test shows significant clustering at the model spot longitude but not at φ0, the null result is an artifact of phase mislabeling. Alternatively, split the 45 individually analysed stars into high- and low-modulation-amplitude groups; if the dominant-spot hypothesis holds, the high-amplitude group should show a stronger phase concentration, and a null result in both groups would support the paper's conclusion.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on φ0 (flux minimum) marking the phase of the dominant spot. Sections 5 and 7 assert the modulation is due to a 'large, dominant starspot,' but Section 9 states 'the sinusoidal pattern we do observe is not produced by a circular large starspot but in fact multiple active regions' with one region responsible for the trough. If multiple active regions contribute, φ0 is the phase of maximum summed spot coverage, not the longitude of a single dominant spot. Flares from other active regions at different longitudes will populate all phases, diluting any preference and biasing the χ² test toward uniformity. Removing the 23% of stars with obvious multi-spot lightcurves does not fix this because sinusoidal lightcurves can also arise from several spots at similar longitudes. Additionally, the KS/SW tests in §7.3 are misinterpreted: p-values of 0 and 7.3×10⁻²³ reject, not confirm, normality, and 'conforms to a normal distribution' is the opposite of what the tests show. While the χ² test is the primary evidence, this statistical inversion suggests the supporting analysis is unreliable. The load-bearing weakness is that without an independent spot longitude, the null result cannot distinguish 'no association with the dominant spot' from 'φ0 does not isolate the dominant spot.'","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper analyzes TESS 2-minute cadence photometry from Sectors 1–3 to test whether flares on M dwarfs occur preferentially at the rotational phase of a dominant starspot. From a sample of 167 M dwarfs, the authors retain 149 stars showing rotational modulation, derive rotation periods and phase zeros (defined as flux minimum), detect 1765 flares with FBEYE, and estimate flare energies in the TESS band-pass. Using reduced chi-squared tests on phase-binned flare counts, they find no significant phase preference for individual active stars (45 stars with ≥13 flares) or for the grouped remaining 104 stars, with χ2ν values of 1.25 (all), 0.57 (high-energy), and 0.57 (low-energy) after setting phase zero at flux minimum. The authors conclude that flare number is not correlated with the large, dominant starspot and discuss alternative scenarios including star–planet interactions, polar spots, and multiple active regions.","tokens_in":14312,"tokens_out":6441,"duration_ms":65572,"significance":"If the result holds, it is a valuable empirical constraint on the flare–spot connection for M dwarfs, extending the earlier K2 study of Doyle et al. (2018) to a larger TESS sample with higher cadence. The sample-level chi-squared analysis is straightforward and appropriate: rotation periods and phase zeros are determined before the phase test, and the null hypothesis of uniform phase distribution is not fitted to the data. The robustness checks (removing nearby-star contaminants, splitting by spectral type and period, excluding multi-spot lightcurves) are thoughtful and support a population-level null result. The main limitation is interpretive: the mapping from φ0 to a single dominant spot is not established, and the paper contains a clear statistical misstatement in the Kolmogorov–Smirnov/Shapiro–Wilk paragraph. With those points addressed, the paper would be a useful contribution.","major_comments":[{"comment":"The interpretation of the KS and SW tests is reversed. The reported p-values of 0 and 7.3×10⁻²³ mean the null hypothesis of normality is rejected with overwhelming significance; they do not indicate that the data 'conforms to a normal distribution.' Moreover, the invocation of the Central Limit Theorem is not a valid way to conclude that the phase distribution is random — the CLT describes the distribution of a sample mean, not the goodness of fit of the raw data to normality. This paragraph should be rewritten; the chi-squared test remains the primary evidence, but as written the supporting test actually argues against uniformity if taken at face value, and in any case is not evidence for uniformity.","section":"§7.3"},{"comment":"The interpretation of the null result as absence of a flare–dominant-spot correlation depends on φ0 marking the longitude of a single dominant spot. Section 5 defines φ0 as the flux minimum and attributes the modulation to a 'large, dominant starspot,' but Section 9 states that the sinusoidal pattern is not produced by a circular large starspot but by multiple active regions, with one region responsible for the trough. If several active regions at different longitudes contribute to the modulation, φ0 is the phase of maximum total spot coverage rather than a unique spot longitude. Flares from other active regions would then populate all phases, diluting any phase preference and biasing the chi-squared test toward uniformity. Removing the 23% of stars with obvious multi-spot lightcurves (Section 7.3) does not fully solve this, since a single sinusoid can also arise from a few spots at similar longitudes. The paper should either obtain an independent spot-longitude constraint or explicitly reframe the conclusion as a null result for the phase of maximum spot coverage, not for the dominant spot.","section":"§5, §7, §9"},{"comment":"The abstract claims 'none of the stars in our sample show any preference for rotational phase,' but only the 45 stars with 13 or more flares are tested individually; the other 104 stars are combined into a grouped analysis (Section 7.2). The grouped test supports a population-level null result but does not demonstrate that each star individually lacks a phase preference. The wording should be softened to match the actual analysis, e.g., 'we find no evidence for a rotational phase preference in the sample as a whole or in any of the 45 individually testable stars.'","section":"Abstract"}],"minor_comments":[{"comment":"Throughout these sections, the symbols '¿' and '¡' appear in place of '>' and '<'; please correct these and the missing superscripts in '10 33.5 erg' (e.g., §6, §7.1).","section":"§7.1, Figure 6, Figure 8"},{"comment":"The statement 'Phase zero, φ0, is also defined as the minimum of the flux of the rotational modulation which is initially determined by eye' would benefit from a clearer explanation of how the eye-determined φ0 was refined in the iterative period-fitting process.","section":"§5"},{"comment":"The reduced chi-squared values alone are reported without p-values or degrees of freedom; for 9 degrees of freedom, χ2ν = 1.25 gives χ2 = 11.25 (p ≈ 0.26), and χ2ν = 0.57 gives χ2 = 5.13 (p ≈ 0.82). Reporting p-values would make the uniformity claim easier to assess.","section":"§7.2"},{"comment":"The energy threshold is described as 'determined from a histogram distribution of all flares which levelled off at 10^33.5 erg'; please specify the exact criterion, since the low/high split is used throughout the analysis.","section":"§7.1"},{"comment":"'In principal' should be 'In principle.'","section":"Figure 3 caption"},{"comment":"'were we will discuss' should be 'where we will discuss.'","section":"§1"},{"comment":"The sentence 'Each Sector is observed for ∼ 27 days with 28 percent of the sample being observed in more than one sector' should clarify whether the 28 percent refers to 167 or 149 stars, and how multi-sector data were combined when computing phase.","section":"§4"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe short version: this is a useful extension of Paper I, not a new discovery. The main claim—no correlation between flare number and rotational phase for 149 M dwarfs observed with TESS—is supported by a simple, appropriate chi-squared test and is probably correct as a sample-level statement. The paper earns its keep by publishing 1765 flares and rotation periods for these stars, which other people will use.\n\nWhat is new: TESS 2-min cadence, a larger sample, and energies derived from Gaia and SkyMapper. The analysis follows Paper I cleanly, and the grouped chi-squared values (1.25, 0.57, 0.57) are consistent with uniform phase distributions. The checks after removing stars with nearby companions and stars with obvious multi-spot lightcurves are reassuring.\n\nThe soft spots are real but not fatal. Section 7.3 misinterprets the Kolmogorov-Smirnov and Shapiro-Wilk tests: p-values of 0 and 7.3e-23 reject normality, not confirm it. That paragraph should be rewritten, but the chi-squared test is the primary evidence, so the main result stands. More substantive: the paper defines φ0 as flux minimum and calls it the phase of the 'large, dominant starspot,' but Section 9 later says the sinusoidal modulation is produced by multiple active regions, not a single circular spot. If that is true, the test has less power to detect an association with any one active region, and the null result is weaker than the abstract suggests. The authors seem aware of this and list multiple spot locations as a preferred explanation, but the framing of the central claim should be softened.\n\nThe thresholds (flare-count cut at 12, energy cut at 10^33.5 erg) are post-hoc and should be justified or checked for sensitivity. Excluding the 18 non-modulated flare stars is understandable, but it means the conclusion applies only to the modulated subset.\n\nWho this is for: anyone working on M-dwarf flares, spot modeling, or exoplanet habitability. It is a legitimate observational paper, and the main null result deserves to be on the record. A serious referee should definitely engage with it, mostly to fix the statistical paragraph and sharpen the phase-zero discussion.\n\nRecommendation: send it to review; with modest revisions it should be a solid contribution.\n\nBest.","headline":"A clean confirmation of the Paper I null result on M-dwarf flare phases, with a larger TESS sample, but the paper overstates the cleanliness of the spot-phase mapping and contains one statistically wrong paragraph.","tokens_in":14932,"tokens_out":3113,"would_cite":true,"duration_ms":33126,"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":"This paper presents TESS observations showing that flares on 149 M dwarfs are uniformly distributed in rotational phase, meaning the dominant starspot is not the flare site.","keywords":["stellar flares","M dwarfs","starspots","rotational phase","TESS","flare energy","low-mass stars","magnetic activity"],"falsifier":"A direct test would be simultaneous Doppler imaging and high-cadence flare monitoring: if flares preferentially occur over the longitudes where the imaged surface spots are located, the null claim is refuted. Short of that, a star with a flat-bottomed, single-spot light curve and many flares should show a significant chi-squared excess at the phase of spot center if the spot is the flare site; looking for such an object in TESS data would settle the question.","tokens_in":13845,"feed_emoji":"🔭","tokens_out":8312,"duration_ms":85304,"temperature":0.7,"pith_summary":"Using TESS two-minute-cadence photometry of 149 rotationally modulated M dwarfs, this paper tests whether flares occur preferentially at the rotational phase when the dominant starspot is facing the observer. Across 1,765 detected flares, no star and no energy bin shows a significant phase preference: the reduced chi-squared values are 1.25, 0.57, and 0.57 for all, high-energy, and low-energy flares in the grouped 104-star sample. The paper therefore claims that the large spot that modulates the light curve is not the site of flare origin, contradicting the simple Sun-like picture in which flares emerge from the dominant active region. If correct, it means flare triggering on low-mass stars must be explained by a different mechanism, such as multiple active regions, polar spots, or unresolved magnetic configurations.","feed_headline":"M-dwarf flares show no phase preference for the big starspot","feed_subtitle":"Across 1,765 flares in 149 M dwarfs, flare times do not cluster when the dominant spot faces us.","key_machinery":"The load-bearing tool is the phase-folded flare distribution tested against uniformity with a $\\chi^2_\\nu$ statistic (the reduced chi-squared test), with phase zero $\\varphi_0$ defined as the minimum of the rotational light curve. That definition maps the phase of maximum spot visibility onto a common reference for all stars, so the test becomes a direct check of whether flare number clusters at the starspot phase. It is supplemented by Kolmogorov-Smirnov and Shapiro-Wilk tests and by binning at several phase widths, all of which return the same null result.","core_discovery":"The central discovery is a null result, stated as a positive claim: the rotational phase of flares on M dwarfs is statistically indistinguishable from uniform. Phase zero is defined as the minimum of the rotational modulation, the phase at which the inferred large spot is most visible; if flares came from that spot, they should pile up at this phase. Instead, a reduced chi-squared test on the full sample yields values consistent with randomness, and no individual active star deviates. The result survives when the sample is restricted to stars showing a single clean sinusoidal spot signal, when stars with nearby contaminating companions are removed, and when flares are split by energy around $10^{33.5}$ erg. The authors read this as evidence that the dominant starspot is not where flares originate, and discuss five scenarios, including multiple spot locations, polar spots, star-planet and star-star interactions, and magnetic configuration, that could produce flares across all phases.","pith_inferences":["Beyond the paper, if the phase-zero mapping at flux minimum does not pinpoint a single active region, the true flare-spot correlation could be diluted; Doppler imaging of spot longitudes would provide an unblurred phase label.","Beyond the paper, the uniform phase distribution hints that flare ignition on fully convective stars may be governed by small-scale or axisymmetric fields rather than longitude-localized activity; a coronal X-ray phase analysis could test whether the same holds for the highest-energy events.","Beyond the paper, because the TESS band-pass is red and M-dwarf flares peak toward the blue, low-energy flares may be missed; simultaneous blue or ultraviolet photometry of a subset could reveal phase structure that the red light curve hides."],"forward_implications":["The large spot that modulates the light curve is not the flare site, so models of M-dwarf flares must seek a different trigger geometry.","High- and low-energy flares, split at $10^{33.5}$ erg, are both phase-uniform, so the decoupling from the dominant spot holds across the energy range.","Removing stars with nearby companions and stars whose light curves show multiple spots does not change the null result, so the finding is not an artifact of those subsets.","The result confirms and strengthens the earlier K2-based null finding, extending the conclusion to TESS data and to a larger sample of low-mass stars."],"supporting_citations":[{"why":"Paper I supplies the phase-folding and reduced chi-squared method and reports the same null result in 34 K2 M dwarfs.","marker":"Doyle et al. (2018)"},{"why":"This is the Kepler long-cadence comparison that found lower-energy flares clustering at spot phases, which the present paper must beat and explains through sample and cadence differences.","marker":"Roettenbacher & Vida (2018)"},{"why":"Provides the flare-detection routine that scans for points above a 2.5-sigma threshold and is used to catalogue the 1,765 flares.","marker":"Davenport et al. 2014"},{"why":"Describes TESS and its two-minute cadence, the dataset and band-pass that define the flare sample and its energy estimates.","marker":"Ricker et al. 2015"},{"why":"Supplies the parallaxes used to derive distances and quiescent luminosities, which set the flare energy scale.","marker":"Gaia Collaboration 2018"},{"why":"Provides the TESS Input Catalog used to reject giant stars and obtain TESS magnitudes for the sample.","marker":"Stassun et al. 2018"},{"why":"Provides the survey photometry converted into template spectra for computing each star's quiescent flux in the TESS band.","marker":"Wolf et al. 2018"}],"fun_headline_variants":["M-dwarf flares don't care where their spots are","Starspot phase doesn't predict M-dwarf flares","Flares on M dwarfs: no preference for the dominant spot","M-dwarf flares are phase-blind to their starspots"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result rests on identifying the phase of minimum brightness with the phase at which the dominant starspot is most visible; if the rotational modulation actually combines several active regions, the phase label is blurred and the test cannot isolate a single flare site.","fun_headline_variants_meta":{"raw":{"variants":["M-dwarf flares don't care where their spots are","Starspot phase doesn't predict M-dwarf flares","Flares on M dwarfs: no preference for the dominant spot","M-dwarf flares are phase-blind to their starspots"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000629,"raw_usage":{"total_tokens":2916,"prompt_tokens":962,"completion_tokens":1954,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":1886}},"tokens_in":578,"tokens_out":1954,"duration_ms":16058,"temperature":1.0,"reasoning_tokens":1886,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:37:35.107300+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct test would be simultaneous Doppler imaging and high-cadence flare monitoring: if flares preferentially occur over the longitudes where the imaged surface spots are located, the null claim is refuted. Short of that, a star with a flat-bottomed, single-spot light curve and many flares should show a significant chi-squared excess at the phase of spot center if the spot is the flare site; looking for such an object in TESS data would settle the question.","supporting_citations":[{"cited_title":"M., Vida K., 2018, , 868, 3","cited_arxiv_id":null,"evidence_quote":"This is the Kepler long-cadence comparison that found lower-energy flares clustering at spot phases, which the present paper must beat and explains through sample and cadence differences."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the parallaxes used to derive distances and quiescent luminosities, which set the flare energy scale."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the survey photometry converted into template spectra for computing each star's quiescent flux in the TESS band."}],"review_version":1}