{"id":"79794ff6-914c-4823-bfc9-8c175a8e6b15","arxiv_id":"2505.10614","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Spectroscopic metallicities of 6,240 DESI RR Lyrae stars reveal smooth period-metallicity correlations and a metallicity-dependent instability strip that shifts to cooler temperatures at lower [Fe/H].","lead":"Using 6,240 RR Lyrae stars observed by the DESI survey, this paper maps how pulsation periods, amplitudes, and effective temperatures depend on stellar iron abundance. It reports a first large-sample empirical map of the RR Lyrae instability strip edges and identifies a small group of metal-rich candidates.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The metal-poor blue-edge shift in Eqs. 4-5 may be a Teff/[Fe/H] calibration artifact; the paper's own admitted systematics are not ruled out by the non-phase-corrected comparison.","rationale":"The reader identified the same core risk: the strip-topology claim depends on the accuracy and representativeness of phase-corrected Teff and the absence of metallicity-dependent biases. The paper is careful and self-aware: it uses a large homogeneous sample, applies phase corrections, compares with non-phase-corrected estimates, and explicitly lists systematics and model disagreement as possible explanations for the metallicity trend. However, the central headline claim is not fully secured against the most plausible artifact, namely a Teff or [Fe/H] scale that drifts with metallicity. The specific slope at the blue edge is modest, the red edge does not independently confirm the shift, and the internal cross-checks do not use an independent temperature or metallicity scale. Therefore I do not move the verdict; the conditional recommendation is appropriate until the data release and independent Teff/[Fe/H] validation are available. I agree with the reader's weakest-assumption formulation rather than proposing a different concern.","tokens_in":37056,"tokens_out":7575,"duration_ms":85339,"concrete_test":"Recompute the Section 5.1 boundary fits using an independent Teff scale for the same stars: dereddened photometric effective temperatures from Gaia BP/RP plus Pan-STARRS/2MASS SED fitting (or the infrared flux method), while keeping the same RVS-[Fe/H] binning and the same 1σ/2σ percentile definitions. If the blue-edge slope in Eqs. 4-5 changes by more than ~50 K/dex, or becomes consistent with zero or with the Marconi-model sign, the claimed cooler-at-low-[Fe/H] strip shift is a Teff calibration artifact. As a complementary check, recompute Eqs. 4-5 with [Fe/H] from the ∆S method or the SP pipeline instead of RVS; a significant change in the blue-edge slope would demonstrate metallicity-scale dependence of the result.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on the slope of the RRc blue-edge boundary with RVS [Fe/H] in Eqs. 4-5: Teff(BE) = 7385 + 104[Fe/H] K (1σ) and 7473 + 83[Fe/H] K (2σ). Only the blue edge carries the 'cooler Teff with declining [Fe/H]' signal; the red-edge slopes are consistent with zero. The load-bearing assumption is that neither Teff nor [Fe/H] has a metallicity-dependent systematic error that imitates a ~100 K/dex shift over the fitted range [-2.79, -1.19] dex. The manuscript itself flags this possibility in Section 5.1 and the Summary: 'systematic lost of precision and accuracy for [Fe/H] and/or Teff in the very metal-poor regime' is listed as a potential explanation for the discrepancy with Marconi et al. (2015), whose models predict the opposite metallicity trend. Checking non-phase-corrected RVS/SP Teff only rules out the phase-correction step; it does not rule out shared PHOENIX/RVSpecFit fitting systematics, nor a metallicity-scale-dependent bias. In addition, the blue edge is defined by 84th/95th percentiles of the RRc Teff distribution, so any metallicity-dependent DESI target-selection or phase-sampling effect that changes the hot tail occupancy of RRc stars would shift the empirical edge without a physical change in the instability strip. With a slope of only ~83-104 K/dex, a Teff zero-point drift of ~200 K across the metallicity range, or an equivalent RVS [Fe/H] compression at the metal-poor end, could produce the entire claimed topological trend.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper uses the DESI Year 1 RR Lyrae catalog from the companion paper M25 (6,240 RRLs with homogeneously derived RVS/SP spectroscopic parameters and phase-corrected effective temperatures) to study (1) correlations between [Fe/H] and pulsation period/amplitude in the Bailey diagram, including the Oosterhoff dichotomy; (2) the metallicities and kinematics of HASP and SASP variables; (3) the period-ratio--[Fe/H] relation of double-mode RRd stars; (4) the claimed empirical, metallicity-dependent topology of the RR Lyrae instability strip; and (5) a small sample of metal-rich RRL candidates. The central claim, stated in the abstract and in Section 5.1, is that the instability strip moves toward cooler Teff with declining [Fe/H] while its width remains roughly constant, based on percentile-based blue and red edges fit as linear functions of [Fe/H] in Eqs. (4)-(5).","tokens_in":37386,"tokens_out":3975,"duration_ms":42929,"significance":"If established, the instability-strip topology result would be a genuinely new empirical constraint, using a large sample and pulsation-phase-corrected effective temperatures, and would be of interest to both stellar pulsation and Galactic archaeology communities. The paper also provides useful confirmation of period--[Fe/H] trends and an RRd period-ratio relation in agreement with Braga et al. (2022). The data product (the DESI Y1 RRL catalog with described fitting procedures and planned public release) is itself a contribution. However, the headline topology claim currently rests on percentile statistics and on the assumption that the [Fe/H] and Teff scales do not have metallicity-dependent systematics; the paper's own text acknowledges this as a possible explanation, and the existing checks do not exclude it. The supporting claims (Oosterhoff interpretation, HASP/SASP properties, RRd relation) are more robust and independently useful.","major_comments":[{"comment":"The central claim that the instability strip moves to cooler Teff with declining [Fe/H] is carried almost entirely by the blue edge. The reported red-edge slopes are 8 +/- 9 K/dex (1-sigma definition) and 40 +/- 27 K/dex (2-sigma definition), both consistent with zero at roughly the 1.5-sigma level. Consequently, Eqs. (4)-(5) do not show that the strip shifts as a whole. Moreover, the strip width implied by these fits is not constant: using the 1-sigma definitions, the BE - RE separation changes by roughly 96 K/dex (104 - 8), which is about 3.8 sigma from zero, while the 2-sigma separation changes by 43 +/- 34 K/dex, only marginally consistent with constant width. The abstract's statement 'an instability strip that moves towards cooler Teff with declining [Fe/H] with a width roughly consistent with stellar-evolution models' is therefore oversold relative to the fitted relations; the text should either qualify the claim as a blue-edge shift or provide a joint fit that explicitly tests and reports the width trend.","section":"Section 5.1, Eqs. (4)-(5)"},{"comment":"The blue edge is defined as the 84th or 95th percentile of the RRc Teff distribution, and the red edge as the 16th or 5th percentile of the RRab distribution. These are arbitrary statistical summaries of the observed Teff distributions, not physical boundaries derived from a pulsation model. Because DESI-MWS is not a complete or selection-function-corrected sample, a metallicity-dependent target-selection or phase-sampling effect that changes the occupancy of the hot tail of the RRc distribution would shift the derived blue edge without any change in the true instability strip. The paper does not quantify the selection function or test how the inferred slopes in Eqs. (4)-(5) respond to plausible incompleteness or to the choice of percentile definition (e.g., a fixed number of stars above a threshold, or a fit to the underlying distribution rather than percentiles). This is a load-bearing issue for the 'first empirical constraint' claim, because the entire metallicity shift is of order 100 K/dex over the fitted range, comparable to the width of the percentile tails being used.","section":"Section 5.1 and Figure 12"},{"comment":"The manuscript itself identifies 'systematic lost of precision and accuracy for [Fe/H] and/or Teff in the very metal-poor regime' as a potential explanation for the discrepancy with Marconi et al. (2015), whose models predict the opposite metallicity trend. The authors' check using non-phase-corrected RVS and SP temperatures rules out the phase-correction step as the sole cause, but it does not rule out a shared metallicity-dependent systematic in the RVSpecFit/PHOENIX Teff scale or a compression of the RVS [Fe/H] scale at low [Fe/H]. The fitted blue-edge slope of 83-104 K/dex over a range of about 1.6 dex corresponds to a total shift of roughly 130-170 K; a Teff zero-point drift of order 100-200 K across the metallicity range, or an equivalent [Fe/H] scale error, could produce the entire claimed trend. A concrete test would be to compare the M25 Teff against an independent photometric or spectroscopic temperature scale (e.g., period-color relations or high-resolution analyses of a calibration subset) and to repeat the edge fits using the SP and Delta-S [Fe/H] scales, showing that the slopes are stable. Without such a test, the central topology result remains vulnerable to the systematic uncertainty the authors themselves flag.","section":"Section 5.1 and Summary"}],"minor_comments":[{"comment":"The phrase 'Using a sample 6,240 RRLs' is missing 'of'; it should read 'Using a sample of 6,240 RRLs'.","section":"Abstract"},{"comment":"In the sentence 'Potential explanations for this shift include a systematic lost of precision and accuracy,' the word 'lost' should be 'loss'.","section":"Section 5.1"},{"comment":"In Figure 13's caption, 'S/N is < 3 at wavelengths < 5500 K' should refer to a wavelength in angstroms, not kelvin; likely '5500 A' is intended.","section":"Section 6.1"},{"comment":"The metal-rich candidate sample is very small (eight stars with good spectra, of which one is concordant across all three metallicity estimators and two are explicitly flagged as having poor RVS fits). The orbital classification into 'disk-like' versus 'halo-like' kinematics is therefore sensitive to individual measurement errors; the text should state more prominently that these are candidate-level findings and that no inference about the fraction of metal-rich RRLs from binary channels is made.","section":"Section 6 and Table 2"},{"comment":"The figure would benefit from showing the actual data points or binned medians for RRab and RRc together with the percentile fits, rather than only the fitted lines and the Marconi et al. (2015) shaded regions; this would help the reader assess how much of the blue-edge trend is driven by a few high-TeFF stars in the sparsely populated most metal-poor bins.","section":"Figure 12"}],"recommendation":"major_revision","confidential_remarks":"The paper is well within the scope of an astrophysics journal and the data set is valuable. My concern is not with the descriptive correlations in Sections 3-4, which are reasonable and well contextualized, but with the novelty claim in Section 5.1. The abstract's 'first time' claim and the topology statement go beyond what Eqs. (4)-(5) and the current systematic checks support. I would encourage the editor to treat the requested systematics tests as a required revision rather than a suggestion; if the authors can show that the blue-edge slope is robust to independent Teff/[Fe/H] scales and to selection-effect tests, the paper would be a strong contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know up front. This is the largest homogeneous sample yet — ~5,400 field RRLs with phase-corrected spectroscopic Teff and RVS [Fe/H] — and the instability-strip edge analysis is genuinely new, a clear step beyond Luo et al.'s ~400 LAMOST RRab. But the most interesting claim, that the strip moves cooler with declining [Fe/H], rests almost entirely on the blue edge and could be a calibration artifact. The paper knows this and says so; the systematics are named but not ruled out.\n\nWhat the paper does well: the RRd period-ratio vs [Fe/H] relation (107 stars) matches Braga et al. almost exactly using an independent metallicity scale — that is a strong cross-check and gives me real confidence in the underlying data quality. The period-metallicity and Oosterhoff results confirm Fabrizio et al. and strengthen them with better statistics. The eight metal-rich candidates are handled with appropriate caution; only one is metal-rich by all three methods, and two have bad RVS fits.\n\nSoft spots, in proportion. The blue-edge slope (83–104 K/dex) is the load-bearing number. A Teff zero-point drift of ~200 K across the fitted metallicity range, or a compression of RVS [Fe/H] at the metal-poor end, would manufacture the whole trend. Checking the raw, non-phase-corrected temperatures only rules out the phase-correction step; it does not rule out shared PHOENIX/RVSpecFit systematics. The edge definitions (16th/84th and 5th/95th percentiles) are arbitrary, and a metallicity-dependent selection or phase-sampling effect on the hot tail of the RRc distribution would shift the blue edge without any change in the physics. The red-edge slope being consistent with zero is a reminder that half the signal is missing. That said, the trend direction is corroborated by the Draco subsample and by the earlier hints from Sandage and Luo, so a pure artifact is not the most likely reading — it just is not excluded. Finally, the Zenodo link is a placeholder \"me/requests\" URL; a referee should insist on a public DOI.\n\nMy read: this is worth a serious referee. The fixes are clear — public data, a quantified treatment of Teff/[Fe/H] systematics, ideally an independent [Fe/H] or Teff check on the edge slopes. I would cite it for the period-metallicity relations and the RRd agreement while treating the blue-edge shift as provisional. Worth a reading-group discussion, because the calibration-artifact versus real-physics question is exactly what a good group should argue about.","headline":"Largest homogeneous sample yet for RRL instability-strip metallicity trends, with a genuinely new blue-edge result that is plausible but not yet safe from Teff/[Fe/H] calibration systematics; deserves a real referee.","tokens_in":38166,"tokens_out":6017,"would_cite":true,"duration_ms":53845,"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":"This paper claims that the RR Lyrae instability strip shifts to cooler temperatures as metallicity drops, while its width stays roughly constant, based on the first large spectroscopic sample with phase-corrected temperatures.","keywords":["RR Lyrae variable stars","Instability strip","Oosterhoff dichotomy","Stellar kinematics","Milky Way stellar halo","Globular star clusters","Dwarf galaxies","Spectroscopy"],"falsifier":"Measure $T_{\\rm eff}$ for the same stars with a method independent of the spectral fitting, such as multiband photometric temperatures or asteroseismic constraints, and check whether the blue-edge slope of roughly $+83$ K per dex in $[\\mathrm{Fe/H}]$ survives; if stars below $[\\mathrm{Fe/H}]\\approx-2.5$ are assigned temperatures that are systematically too cool, the strip shift disappears. A complete, selection-bias-free sample of low-metallicity RRc stars from deep wide-field photometry with follow-up spectra would also settle whether the blue edge truly moves to cooler temperatures.","tokens_in":36854,"feed_emoji":"⭐","tokens_out":8834,"duration_ms":74828,"temperature":0.7,"pith_summary":"The paper uses 6,240 RR Lyrae stars from the first year of the DESI survey, with homogeneous spectroscopic iron abundances and pulsation-corrected effective temperatures, to ask how metal content shapes where and how these stars pulsate. Its central empirical result is that the RR Lyrae instability strip moves toward cooler effective temperatures as $[\\mathrm{Fe/H}]$ declines, while its width stays roughly constant at about 1300–1400 K, consistent with stellar-evolution models. If this is right, it is the first large spectroscopic map of the range of temperatures over which RR Lyrae pulsation is possible as a function of metallicity, and it gives a quantitative explanation for the long-standing Oosterhoff dichotomy as a consequence of the scarcity of intermediate-metallicity globular clusters with sizable RR Lyrae populations.","feed_headline":"Metal-poor RR Lyrae stars pulsate at cooler temperatures","feed_subtitle":"Both edges shift together with [Fe/H], matching stellar-evolution models in a large spectroscopic sample.","key_machinery":"The load-bearing object is the DESI Year 1 RR Lyrae catalog of 6,240 stars, with one crucial refinement: the effective temperatures are phase-corrected, meaning each star's single-epoch spectrum is modeled against its pulsation cycle so the reported $T_{\\rm eff}$ represents the mean over the cycle rather than a random phase. The instability strip edges are defined operationally as percentile limits of the temperature distributions of RRab stars (cooler, fundamental-mode) and RRc stars (hotter, first-overtone) in equal-number metallicity bins; linear fits to the running 16th/84th and 5th/95th percentiles give the empirical red and blue edges. The same catalog's RVS-derived $[\\mathrm{Fe/H}]$ values and Gaia DR3 periods and subtype classifications carry the Bailey diagram, Petersen diagram, and metal-rich candidate analyses.","core_discovery":"The central claim is that the blue and red edges of the RR Lyrae instability strip shift toward cooler temperatures with declining $[\\mathrm{Fe/H}]$, with the strip's width staying roughly constant. For the 2-$\\sigma$ edges over $[\\mathrm{Fe/H}]$ from roughly -2.8 to -1.2 dex, the paper reports $T_{\\rm eff}^{\\rm RE}=6020(\\pm47)+40(\\pm27)[\\mathrm{Fe/H}]$ K and $T_{\\rm eff}^{\\rm BE}=7473(\\pm37)+83(\\pm20)[\\mathrm{Fe/H}]$ K, so the blue edge moves about 83 K per dex while the red edge moves much less. Alongside this, the paper reports that high-amplitude short-period and small-amplitude short-period RR Lyrae stars are comparatively metal-rich, with mean $[\\mathrm{Fe/H}]$ of $-1.39\\pm0.27$ and $-1.30\\pm0.28$; that double-mode RRd stars show metallicity declining smoothly with increasing fundamental-mode period; and that eight metal-rich candidates with $[\\mathrm{Fe/H}]>-0.5$ dex split roughly evenly between disk-like and halo-like orbits.","pith_inferences":["If the blue-edge shift is real, then period-luminosity-metallicity relations used to measure distances with RR Lyrae stars may need a metallicity-dependent temperature correction, because a cooler instability strip at low $[\\mathrm{Fe/H}]$ changes the expected pulsation properties at fixed luminosity.","The trend can be checked with independent, model-free temperature estimates, such as multiband colors or temperatures from eclipsing binary companions; if stars with $[\\mathrm{Fe/H}]<-2.5$ are not actually as cool as the spectroscopic fits say, the blue-edge slope would shrink or vanish.","The paper's own candidate explanations for the model discrepancy (small numbers near the edges at low metallicity, or decreasing accuracy in $[\\mathrm{Fe/H}]$ and $T_{\\rm eff}$ at the metal-poor end) suggest a targeted search for low-metallicity RRc stars in deep wide-field photometry could separate an intrinsic strip shift from a selection artifact.","If the Oosterhoff dichotomy is a selection effect, then globular cluster systems with continuous metallicity distributions should show continuous mean RRab periods rather than two clumps; applying the same DESI spectra to a larger cluster sample would test this directly."],"forward_implications":["The smooth anti-correlation between logarithmic period and $[\\mathrm{Fe/H}]$ for both RRab and RRc stars supports the view that the Oosterhoff dichotomy is not a fundamental bimodality of pulsation but follows from the lack of intermediate-metallicity globular clusters with large RR Lyrae samples.","High-amplitude short-period (HASP) and small-amplitude short-period (SASP) stars are metal-rich and sit on radial orbits associated with the Gaia-Sausage-Enceladus merger in large numbers, so they can serve as chemical tracers of massive accreted satellites that enriched early and were later disrupted.","The period-ratio versus $[\\mathrm{Fe/H}]$ relation for classical RRd stars turns the Petersen diagram into a spectroscopic metallicity indicator and, combined with existing models, places classical RRd masses above about $0.69\\,M_\\odot$ and anomalous RRd masses in a narrow $0.68$--$0.77\\,M_\\odot$ range.","Empirical red and blue edges with roughly constant width can replace theoretical assumptions in stellar population models that predict RRc-to-RRab ratios across metallicity.","The eight metal-rich candidates with $[\\mathrm{Fe/H}]>-0.5$ dex, about half on disk-like orbits, give concrete targets for testing whether some RR Lyrae stars form through binary mass-stripping channels rather than single-star evolution."],"supporting_citations":[{"why":"Supplies the DESI Year 1 RR Lyrae catalog with phase-corrected effective temperatures, systemic velocities, and RVS iron abundances used throughout the paper.","marker":"M25"},{"why":"Provides the Oosterhoff I and II fiducial lines in the Bailey diagram and the interpretation that the dichotomy reflects a scarcity of intermediate-metallicity globular clusters.","marker":"Fabrizio et al. 2019"},{"why":"Defines the prior period-amplitude-metallicity correlations and the equal-number metallicity binning method that this paper follows and compares against.","marker":"Fabrizio et al. 2021"},{"why":"Supplies the theoretical instability strip edges and the mass-metallicity-period-ratio relations used to compare empirical edges and to estimate RRd masses.","marker":"Marconi et al. 2015"},{"why":"Provides the Gaia DR3 RR Lyrae catalog with subtype classifications, periods, and amplitudes that define the RRab, RRc, and RRd subsamples.","marker":"Clementini et al. 2023"},{"why":"Provides the calibrated Delta S method for iron abundances used as a cross-check on the RVS metallicities.","marker":"Crestani et al. 2021a"},{"why":"Provides the previous period-ratio versus metallicity relation for classical RRd stars, used as the comparison and initial guess for the fit in this paper.","marker":"Braga et al. 2022"},{"why":"Provides the binary-evolution formation scenario for metal-rich RR Lyrae stars that motivates the metal-rich candidate analysis.","marker":"Bobrick et al. 2024"}],"fun_headline_variants":["RR Lyrae strip shifts cool with lower metallicity","Blue edge moves 83 K per dex as stars get metal-poor","Metal-rich RR Lyrae pulsate at shorter periods","RRd stars show smooth metallicity drop with period"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the DESI Year 1 RR Lyrae sample and its phase-corrected effective temperatures faithfully trace the true blue and red edges of the instability strip at every metallicity, with no metallicity-dependent bias in $T_{\\rm eff}$ or in which stars were selected for spectroscopy.","fun_headline_variants_meta":{"raw":{"variants":["RR Lyrae strip shifts cool with lower metallicity","Blue edge moves 83 K per dex as stars get metal-poor","Metal-rich RR Lyrae pulsate at shorter periods","RRd stars show smooth metallicity drop with period"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000657,"raw_usage":{"total_tokens":3106,"prompt_tokens":1143,"completion_tokens":1963,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":759,"completion_tokens_details":{"reasoning_tokens":1896}},"tokens_in":759,"tokens_out":1963,"duration_ms":13503,"temperature":1.0,"reasoning_tokens":1896,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T21:05:41.087565+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $T_{\\rm eff}$ for the same stars with a method independent of the spectral fitting, such as multiband photometric temperatures or asteroseismic constraints, and check whether the blue-edge slope of roughly $+83$ K per dex in $[\\mathrm{Fe/H}]$ survives; if stars below $[\\mathrm{Fe/H}]\\approx-2.5$ are assigned temperatures that are systematically too cool, the strip shift disappears. A complete, selection-bias-free sample of low-metallicity RRc stars from deep wide-field photometry with follow-up spectra would also settle whether the blue edge truly moves to cooler temperatures.","supporting_citations":[],"review_version":1}