{"id":"e5020090-f771-4a08-95ee-bcb23cf06271","arxiv_id":"1908.01795","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Maximum-light period-luminosity relations for Miras have up to 30% smaller scatter than mean-light relations and give a smaller-uncertainty LMC-SMC distance modulus.","lead":"Mira variable stars in the Magellanic Clouds show tighter period-luminosity relations when measured at maximum brightness instead of average brightness, across optical, Gaia, and near-infrared data. If the peak brightness is also stable from cycle to cycle, Miras could become more precise and efficient distance indicators.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Max-light definition may co-select stable-phased epochs; synthetic light-curve injection test needed to confirm the 30% scatter reduction is physical.","rationale":"The reader's weakest assumption is exactly the operational definition of max-light, and the strongest claim depends on that definition being a faithful probe of the stellar property. However, the paper does contain a partial robustness check, the minimum-numerical-magnitude definition, which shows the I-band PLR dispersion reduction survives an alternative definition. The concern is real but does not clearly invalidate the claim; it is a testable assumption rather than a demonstrated failure. The F-tests and the consistency across OGLE VI, Gaia G/BP/RP, and NIR also give independent weight. Therefore the verdict should remain CONDITIONAL, not ACCEPT (because the cycle-to-cycle stability claim is based on a small Kepler sample and sparse OGLE max-light epochs) and not REJECT (because the alternative-definition robustness check and multi-band consistency support the core result).","tokens_in":22608,"tokens_out":1815,"duration_ms":17138,"concrete_test":"Run an injection test on simulated Mira light curves with realistic cycle-to-cycle amplitude and phase-jitter, sampled exactly with the OGLE/Kepler cadence. Construct two ensembles with identical cycle-to-cycle behavior but with (a) max-light magnitudes drawn with the same scatter as the mean-light magnitudes, and (b) max-light magnitudes drawn with the reduced scatter claimed by the paper. Apply the paper's decade definition of max-light, fitting with the same GPR procedure, and fit the PL/PLC relations. If the measured scatter reduction in case (a) is substantially smaller than the claimed ~30%, the definition itself is not the sole driver; if case (a) reproduces the claimed reduction, then the headline scatter reduction cannot be attributed to physical stability.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim is that max-light PL and PLC relations are intrinsically tighter (up to ~30% lower scatter) because Miras are physically more stable at their warmest phase, and that this makes max-light a better distance indicator. The load-bearing step is the operational definition of max-light: the mean of all epochs within the brightest 10% of the peak-to-peak amplitude (Section 2). This is applied to real, irregularly sampled light curves, and the same GPR fit is used to assign I-band magnitudes at V-band epochs. Several non-physical mechanisms could inflate the apparent scatter reduction without any true cycle-to-cycle stability: (1) An average over the brightest decile is a truncated mean; if the light-curve is steeper near maximum or if phase coverage varies with period, the truncation selects a narrower magnitude range at max-light, mechanically compressing the dispersion relative to a full-cycle mean. (2) Measurement errors and GPR residuals are heteroscedastic in phase; the brightest epochs may be precisely the epochs where the GPR fit has the smallest uncertainty, so the reported max-light uncertainty is not actually the cycle-to-cycle stability. The paper's robustness check (using the single minimum numerical magnitude, Section 3.1) reproduces the I-band result, which partially counters the 'truncated mean' concern, but it is only applied to one band and one relation. The claim that cycle-to-cycle max-light magnitudes are more stable (Section 4.1) is based on a small Kepler sample and on OGLE data where max-light epochs are often sparse. The central physical interpretation therefore rests on a narrow methodological base.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies Mira variables in the Large and Small Magellanic Clouds using OGLE-III optical light curves, Gaia DR2 photometry, NIR data from Yuan et al. (2017) and Ita et al. (2018), and Kepler/LAMOST observations for a small sample. The authors define maximum-light magnitude as the mean of all epochs within the brightest 10% of the peak-to-peak amplitude, and derive Period-Luminosity and Period-Luminosity-Color relations at mean-, maximum-, and minimum-light. They report that the maximum-light relations have up to ~30% smaller scatter than the mean-light relations at optical wavelengths, that individual Miras show smaller cycle-to-cycle variations at maximum than at minimum light, and that spectra show weaker TiO bands at maximum light. They also report a break period near 300 days in the optical, shifting to ~350 days in the near-infrared, and derive a relative LMC-SMC distance modulus of Delta(mu) = 0.48 +/- 0.08 mag from maximum-light relations.","tokens_in":22961,"tokens_out":5168,"duration_ms":58182,"significance":"If the claimed effect is real, maximum-light Mira relations could provide an empirical distance indicator with smaller scatter than mean-light relations, and the apparent stability of maximum-light magnitudes would be a useful constraint on Mira atmosphere models. The paper makes good use of multiple independent public datasets and includes a robustness check using the single minimum numerical magnitude in the I-band. The multiwavelength comparison and the spectroscopic evidence from LAMOST are valuable. However, the central quantitative claims rely on an operational maximum-light definition whose statistical behavior is not fully controlled, and the cycle-to-cycle stability analysis is not described with enough detail to be reproduced. These issues are load-bearing for the paper's main conclusions.","major_comments":[{"comment":"The maximum-light magnitude is defined as the mean of all epochs within the brightest 10% of the peak-to-peak amplitude. This is a truncated average that restricts the phase window by construction, so a reduction in dispersion relative to the phase-averaged mean is expected even if the intrinsic cycle-to-cycle dispersion is unchanged. The robustness check in Section 3.1 using the single minimum numerical I-band magnitude removes the averaging effect but retains the phase-selection effect and is reported only for the I-band PLR. No analogous test is shown for the V-band, the Gaia-band relations, or the NIR relations. The authors should add a synthetic light-curve injection test that preserves the actual sampling patterns and known cycle-to-cycle scatter, reporting the expected dispersion reduction under the null hypothesis that maximum light is no more stable than mean light.","section":"Section 2, maximum-light definition"},{"comment":"The claim that individual Miras have more stable maximum-light magnitudes over multiple pulsation cycles is central to the physical interpretation, but the methodology is not described. The text does not state how individual cycles are segmented, how the per-cycle maximum, minimum, and median magnitudes are estimated from the GPR predictions, how many cycles are used for each star, or how the criterion that the time coverage is equivalent to the period is applied. Without this information, Figure 9 cannot be reproduced and the comparison conflates the adopted maximum-light estimator with true cycle-to-cycle stability. The authors should specify the algorithm in detail and provide per-cycle statistics for at least the Kepler sample.","section":"Section 4.1 and Figure 9"},{"comment":"The F-test is used to establish that the dispersion reduction is statistically significant, but it is applied after iterative 3-sigma outlier rejection and the final sample sizes differ between the mean- and maximum-light fits (Table 2, e.g., N_f = 440 vs 441 and 438 vs 439). The F distribution is not the correct null distribution for variance ratios of residuals from data-dependent clipped fits with different retained samples. A permutation or bootstrap test that applies the identical clipping procedure to each resample should be used to confirm that p < 0.05 is not an artifact of the clipping algorithm or of sample-size differences.","section":"Section 3.1, F-test paragraph"},{"comment":"The relative distance modulus Delta(mu) = 0.48 +/- 0.08 mag is obtained by combining three band estimates that differ by more than their formal errors: 0.45 +/- 0.08 (optical PLC), 0.46 +/- 0.08 (J), and 0.53 +/- 0.07 (K). The SMC's substantial line-of-sight depth and any metallicity-dependent differences in the color coefficients are not propagated into the quoted uncertainty. The statement that maximum-light relations provide a smaller statistical uncertainty is formally supported, but the improvement in precision is not demonstrated to survive these systematic uncertainties. The authors should either include these systematics in a combined uncertainty or restrict the claim to statistical precision only.","section":"Section 5.2 and Table 6"}],"minor_comments":[{"comment":"The phrase \"smaller than 0.05 dex\" should read \"smaller than 0.05\" or \"p < 0.05\"; dex is a logarithmic unit and is not appropriate for a probability.","section":"Section 3.1, final sentence of the F-test paragraph"},{"comment":"The vertical dotted lines are described as showing V- and I-band magnitudes at the same epoch, but the figure would benefit from a legend or labels distinguishing the V and I magnitudes and the shaded ten-percentile region.","section":"Figure 1 caption"},{"comment":"The abstract and conclusions state that maximum-light relations show up to 30% smaller scatter, but the NIR results in Table 5 show comparable or slightly larger scatter at maximum light in the J band (sigma = 0.15 mean vs 0.17 max for the quadratic fit). The wording should clearly restrict the 30% claim to optical wavelengths.","section":"Section 5.1 and Tables 4-5"},{"comment":"The selection of Kepler Mira candidates from Banyai et al. (2013) and the criterion for calling a star a Mira candidate should be stated explicitly, since only two example stars are shown and the reader cannot assess how representative they are.","section":"Section 4.2"},{"comment":"The paper says the median E(V-I) is taken when the reddening map does not provide a value within 2 arcsec, but it does not state how many stars are affected by this substitution; a brief statement of the sample fraction would be useful.","section":"Section 2, extinction corrections"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and addresses a question of interest for Mira-based distance work. The main concerns are methodological rather than a fundamental flaw in the empirical analysis. In particular, the maximum-light definition and the per-cycle stability analysis need additional validation and documentation. I would be willing to review a revised version."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nYou can safely take this paper as the modern, multi-wavelength confirmation of an old idea: Kanbur et al. (1997) showed that Mira PLRs at maximum light are tighter than at mean light, but only for 48 LMC stars in JHK. This group extends that to the full OGLE-III LMC/SMC samples, adds Gaia DR2 photometry, a GPR-based light-curve model, an explicit cycle-to-cycle stability check with Kepler and LAMOST spectra, and uses the max-light relations to get an LMC–SMC distance modulus with smaller uncertainty. The result is coherent and the analysis is careful; the 30% scatter reduction appears consistently across optical bands, and the robustness check using the single minimum numerical magnitude in I band reproduces it. So the central claim is probably real, not a fitting artifact.\n\nThe main soft spot is exactly the one you would guess from the methods: max-light is defined as the mean of the brightest 10% of epochs. That is a truncated mean, and it could mechanically compress the scatter unless the light-curve shape cooperates. The authors address this only partially—the min-magnitude check is for one band and one relation. A synthetic light-curve injection test would settle whether the scatter reduction is an artifact of phase selection, and the paper is missing that. The cycle-to-cycle stability section rests on a small Kepler sample (a dozen stars) and sparse OGLE coverage, though the LAMOST spectra showing strong Balmer emission and weak TiO at max are a nice, independent piece of evidence for the physical interpretation. The SMC samples are small, so the SMC results are suggestive rather than decisive.\n\nMinor points: the iterative 3-sigma clipping can inflate the apparent significance of the F-test if the clipping removes more outliers from the max-light sample; the paper should show results with fixed outlier thresholds or a robust regression. The break-period selection is standard but not fully model-independent; no big deal. The code and derived tables are not in the arXiv version, which would help reproduction.\n\nWho is this for? Anyone working on Mira distance scale, AGB pulsation, or LSST-era variable star surveys. It is a useful subfield advance, not a branch reshuffling. The paper is honest and the literature context is well handled. I would send it to a serious referee; the referee should ask for the synthetic light-curve test and a more robust treatment of outliers, but the core result deserves to be published.","headline":"Solid multiwavelength extension of the old max-light Mira PLR result; the top-decile definition deserves a synthetic test, but the core scatter reduction is likely real.","tokens_in":23492,"tokens_out":3698,"would_cite":true,"duration_ms":37879,"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":"Maximum-light measurements of Mira variables produce period-luminosity relations with up to 30% less scatter than mean-light relations, offering a more precise distance indicator.","keywords":["Mira variables","period-luminosity relation","maximum light","Magellanic Clouds","distance scale","asymptotic giant branch","stellar pulsation","molecular bands"],"falsifier":"Recompute the maximum-light relations using an independent estimator of the true light-curve peak—for instance, the maximum of a Gaussian-process or Fourier fit, or the mean over a fixed phase window—and compare the dispersions with the ten-percentile results. If the up-to-30% scatter reduction disappears or diminishes substantially, the paper's central claim is an artifact of the chosen maximum-light definition. Alternatively, use continuous high-cadence photometry (as from a space telescope) to measure the rms scatter of maximum-light magnitudes over many tens of cycles; if that scatter is not smaller than the scatter at minimum light, the stability claim fails.","tokens_in":22383,"feed_emoji":"🌟","tokens_out":9627,"duration_ms":86822,"temperature":0.7,"pith_summary":"Maximum-light observations of Mira variables produce period-luminosity and period-luminosity-color relations with up to about 30% less scatter than their mean-light counterparts at optical wavelengths. The paper also shows that individual Miras vary less from cycle to cycle at maximum light than at minimum light, and that this stability likely arises because molecular absorption bands are weaker when the star is warmest. If these results hold, maximum-light measurements could serve as a simpler and more precise distance indicator for galaxies, and the paper demonstrates this by deriving a relative Magellanic Clouds distance modulus of 0.48 ± 0.08 mag with smaller uncertainty than mean-light relations.","feed_headline":"Mira brightness peaks yield 30% tighter distance relations","feed_subtitle":"Maximum-light Mira relations cut scatter in period-luminosity curves, giving a new way to measure cosmic distances.","key_machinery":"The key operational object is the maximum-light magnitude, defined as the mean of all observed epochs within the brightest ten percent of the peak-to-peak amplitude of each light curve in each band. To make simultaneous colors, the paper fits a Gaussian-process regression to the well-sampled I-band light curve and predicts the I-band magnitude at the epoch of each V-band observation. This phase-matched photometry, combined with the ten-percentile maximum-light definition, is what produces tighter PL and PLC relations. The physical mechanism invoked is phase-dependent molecular absorption: at maximum light the star is hottest, TiO and water-vapor bands are weak, and the observed brightness is therefore less sensitive to temperature and layer variations that inflate scatter at fainter phases.","core_discovery":"The central claim is that the empirical Period-Luminosity (PL) and Period-Luminosity-Color (PLC) relations for Mira variables are intrinsically tighter when evaluated at maximum light rather than at mean light. Using hundreds of oxygen-rich Miras in the Large Magellanic Cloud and smaller samples in the Small Magellanic Cloud, the paper measures dispersions that are up to roughly 30% smaller at maximum light in optical bands, with comparable or slightly smaller dispersions in the near-infrared. The paper further demonstrates that maximum-light magnitudes are more stable over multiple pulsation cycles than minimum-light magnitudes, and that maximum-light spectra show strong Balmer emission lines with weak TiO absorption, consistent with reduced molecular-band sensitivity at the warmest phase. On this basis, the paper locates a kink in the oxygen-rich Mira PL relation at about 300 days in the optical, shifting to about 350 days in the near-infrared, and derives a relative distance modulus of 0.48 ± 0.08 mag between the Large and Small Magellanic Clouds from maximum-light relations.","pith_inferences":["If the molecular-band explanation is correct, the scatter reduction should strengthen toward bluer bands and weaken toward the infrared, a gradient that could be tested with more extensive near- and mid-infrared time series.","The same maximum-light logic may apply to other large-amplitude pulsators, such as semi-regular variables or red supergiants, whose light curves also modulate via molecule formation; this would extend the method beyond Miras.","A critical check the paper does not perform is to vary the ten-percentile window (e.g., top 5% or 20%) and the required number of epochs; if the scatter reduction is robust to those choices, the effect is more likely physical than procedural.","If maximum-light stability holds in the Milky Way and in external galaxies with different metallicities, Mira maximum-light relations could become a backbone of the local distance ladder that does not depend on full phase coverage of each star."],"forward_implications":["Maximum-light Mira PL/PLC relations can serve as a distance indicator with statistical precision competitive with classical Cepheids, even in optical bands where mean-light Mira relations are notoriously scattered.","Because maximum-light magnitudes are stable across cycles, a single well-timed epoch near maximum could be sufficient to place a Mira on the relation, dramatically reducing the observing time needed for distance work.","The reported tightenings predict that the color term in the PLC relation becomes smaller and more physically interpretable at maximum light, offering a clean test for stellar atmosphere models.","The kink at 300–350 days gives an empirical anchor for when hot bottom burning begins to dominate in intermediate-mass AGB stars, which pulsation models can now be required to reproduce.","The smaller statistical uncertainty in the LMC–SMC distance modulus (0.48 ± 0.08 mag) suggests that maximum-light Mira relations could help arbitrate between conflicting distance estimates to the Magellanic Clouds and beyond."],"supporting_citations":[{"why":"Prior study showing smaller PLR scatter at maximum light for a small sample of LMC Miras; the direct predecessor this paper extends with modern multiwavelength data.","marker":"Kanbur et al. 1997"},{"why":"OGLE-III catalog of long-period variables in the LMC, providing the period, classification, and time-series photometry for the primary LMC sample.","marker":"Soszyński et al. 2009"},{"why":"OGLE-III catalog for the SMC, supplying the corresponding SMC Mira sample.","marker":"Soszyński et al. 2011"},{"why":"Near-infrared PLRs and template-based maximum-light magnitudes for LMC Miras, used for the NIR comparison and distance calibration.","marker":"Yuan et al. 2017"},{"why":"Gaussian-process regression model used to fit light curves and to predict simultaneous-band magnitudes, enabling phase-matched colors.","marker":"He et al. 2016"},{"why":"Earlier identification of a kink in Mira PLRs around 400 days, which this paper locates more precisely at 300–350 days in maximum-light relations.","marker":"Ita & Matsunaga 2011"},{"why":"Demonstrates the phase dependence of Balmer emission and TiO band strengths that underpins the molecular-band explanation for maximum-light stability.","marker":"Yao et al. 2017"},{"why":"Spectro-interferometric measurements showing photospheric radius and water-vapor effects vary from maximum to minimum light, supporting the proposed physical mechanism.","marker":"Wittkowski et al. 2018"}],"fun_headline_variants":["Mira maximum-light relations cut scatter by 30%","Peak-light Mira data sharpen distance scale","Mira peaks stabilize, sharpen period-luminosity links","Maximum-light Miras reveal tighter PL relations","Mira max-light curves yield sharper cosmic distances"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the operational definition of maximum light as the mean of the brightest ten percent of the peak-to-peak amplitude; if the top decile of epochs is a narrower range by construction (due to sampling or light-curve shape), the measured scatter reduction and cycle-to-cycle stability could be partly artifacts of that definition.","fun_headline_variants_meta":{"raw":{"variants":["Mira maximum-light relations cut scatter by 30%","Peak-light Mira data sharpen distance scale","Mira peaks stabilize, sharpen period-luminosity links","Maximum-light Miras reveal tighter PL relations","Mira max-light curves yield sharper cosmic distances"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000575,"raw_usage":{"total_tokens":2782,"prompt_tokens":1078,"completion_tokens":1704,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":694,"completion_tokens_details":{"reasoning_tokens":1631}},"tokens_in":694,"tokens_out":1704,"duration_ms":12153,"temperature":1.0,"reasoning_tokens":1631,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:03:08.995625+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the maximum-light relations using an independent estimator of the true light-curve peak—for instance, the maximum of a Gaussian-process or Fourier fit, or the mean over a fixed phase window—and compare the dispersions with the ten-percentile results. If the up-to-30% scatter reduction disappears or diminishes substantially, the paper's central claim is an artifact of the chosen maximum-light definition. Alternatively, use continuous high-cadence photometry (as from a space telescope) to measure the rms scatter of maximum-light magnitudes over many tens of cycles; if that scatter is not smaller than the scatter at minimum light, the stability claim fails.","supporting_citations":[{"cited_title":"M., Hendry, M","cited_arxiv_id":null,"evidence_quote":"Prior study showing smaller PLR scatter at maximum light for a small sample of LMC Miras; the direct predecessor this paper extends with modern multiwavelength data."},{"cited_title":"M., He, S., et al","cited_arxiv_id":null,"evidence_quote":"Near-infrared PLRs and template-based maximum-light magnitudes for LMC Miras, used for the NIR comparison and distance calibration."},{"cited_title":"2017, ApJS, 232, 16","cited_arxiv_id":null,"evidence_quote":"Demonstrates the phase dependence of Balmer emission and TiO band strengths that underpins the molecular-band explanation for maximum-light stability."}],"review_version":1}