{"id":"01a78941-f62b-4352-9b55-cdf5e7d2dd72","arxiv_id":"2506.05094","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Non-solar metallicity 1D-3D coupled stellar models produce temperature shifts under 30 K for a 20% mixing-length change and broadly match eclipsing binary observations.","lead":"This paper extends a hybrid stellar modeling method that couples cheap 1D evolution models with realistic 3D surface convection simulations, now covering stars with metal content from about 0.001 to 3 times solar. The authors show the method makes predicted stellar temperatures nearly independent of the uncertain mixing-length parameter, which could improve stellar age estimates.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Metallicity interpolation in [M/H] spans a discontinuous +0.4 dex α-enhancement at [Fe/H] = -1; the outer 3D boundary for -1 < [Fe/H] < -0.5 is compositionally inconsistent, and the KIC 10001167 validation target lies in this mismatch region.","rationale":"The reader's weakest assumption was the general fidelity of the interpolated mean Stagger-grid models across the full metallicity range. I narrow this to a specific, testable composition discontinuity: the +0.4 dex α-enhancement applied only for [Fe/H] ≤ -1 means the interpolation in [M/H] blends chemically distinct models in the range -1 < [Fe/H] < -0.5. The KIC 10001167 validation target, with [Fe/H] = -0.73 and [α/Fe] = 0.37, sits exactly in this problematic regime, and its 3D boundary condition is not α-enhanced. This is a strong candidate explanation for the paper's own unexplained systematic mass and mean-density overestimates in both Kepler giants (Tables 3 and 4). The concern does not overturn the paper's core technical achievement: the α_MLT insensitivity is convincingly demonstrated at solar metallicity (Fig. 4), and the paper honestly discloses the kinks and mass offsets. But the new metallicity-extension claim depends on the interpolation integrity over a discontinuous composition grid. The proposed test would settle whether this mismatch is the source of the offsets. If confirmed, the method would require α-enhanced 3D models at all metallicities or a more careful interpolation across the threshold. I therefore keep the reader's CONDITIONAL verdict: the concern is concrete but currently unverified, and it does not invalidate the method's potential.","tokens_in":30732,"tokens_out":7116,"duration_ms":80702,"concrete_test":"Recompute the KIC 10001167 model of Sect. 4.3 using outer boundary conditions interpolated from an α-enhanced Stagger-grid composition at [Fe/H] = -0.7, e.g., by interpolating between the α-enhanced [Fe/H] = -1.0 models and a newly computed α-enhanced [Fe/H] = -0.5 model (or by rescaling the α-enhanced [Fe/H] = -1.0 models to the target [M/H]). If the inferred mass moves from 0.998 M_sun toward the dynamical 0.934 ± 0.008 M_sun, or if Δν and ν_max shift toward the observed values, the composition mismatch is the cause. In parallel, evaluate the interpolated boundary pressure P_m and temperature T_m as functions of [M/H] in a fine sweep across [Fe/H] = -1; a jump or non-monotonic derivative would confirm that the cubic monotonic interpolation is inadequate across the α-enhancement discontinuity.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that a fixed solar-calibrated α_MLT works across -3 < [Fe/H] < 0.5 requires the interpolated mean 3D models to be a smooth, faithful function of [M/H]. In Sect. 2.2, the Stagger-grid applies a fixed +0.4 dex α-enhancement only for [Fe/H] ≤ -1, so the composition jumps discontinuously at [Fe/H] = -1. The metallicity interpolation (Steffen cubic monotonic in [M/H]) treats grid values as a smooth scalar field, but the underlying models switch chemical composition abruptly. For -1 < [Fe/H] < -0.5, the interpolation blends α-enhanced and non-enhanced structures, producing a boundary condition that corresponds to no real stellar abundance mixture. The paper validates the method on KIC 10001167, an α-enhanced giant with [Fe/H] = -0.73 and [α/Fe] = 0.37 (Sect. 4.3), but the Stagger-grid models at this metallicity are not α-enhanced; the α-enhanced models begin only at [Fe/H] = -1. The model therefore uses a 3D outer boundary whose composition differs from the star's. The paper reports a systematic overestimate of mass and mean density for both Kepler giants (Tables 3 and 4) and states the cause is not understood. The α-enhancement inconsistency is a concrete, testable candidate for that bias and directly affects the validity of the fixed-α_MLT claim at intermediate metallicities. Unless the interpolation is shown to be continuous across the α-enhancement threshold and consistent with the abundances of the validation stars, the claim that the method reproduces observations for all target stars is not yet established.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper extends the on-the-fly 1D-3D coupling method (GARSTEC + Stagger-grid mean atmospheres) from solar metallicity to a broad metallicity range, implementing interpolation over [M/H] for the 3D outer boundary condition. The central claims are that placing the matching point in the near-adiabatic layer makes stellar evolution tracks far less sensitive to alpha_MLT (a 20% change shifts Teff by less than 30 K, versus more than 200 K in standard models), and that with a fixed solar-calibrated alpha_MLT the method reproduces most observational constraints for stars spanning roughly -3 < [Fe/H] < 0.5. The method is validated against the eclipsing binary AI Phe and the asteroseismic binaries KIC 9970396 and KIC 10001167. The paper also discusses unphysical kinks in the tracks, interpolation errors, and remaining systematic offsets, but concludes that alpha_MLT can be regarded as a constant fixed by solar calibration.","tokens_in":31173,"tokens_out":8156,"duration_ms":86309,"significance":"If the central claim holds, this is a significant step toward reducing a major systematic uncertainty in 1D stellar modeling: the mixing-length parameter and the choice of atmospheric boundary conditions. The alpha_MLT-insensitivity result is cleanly demonstrated in Fig. 4 and is conceptually solid because the matching point sits where the temperature gradient is nearly adiabatic. The method is anchored to external 3D hydrodynamics (the Stagger-grid) and tested against independent binary observations, so it is not circular in its validation strategy. The paper is also unusually honest: it explicitly reports interpolation errors up to about 9% in warm dwarfs (Appendix A), unphysical track kinks (Sect. 3.2), and systematic mass and mean-density offsets that it cannot explain (Sect. 4.3). These strengths make the paper a valuable contribution even though the validation is not as complete as the abstract's broad claim suggests.","major_comments":[{"comment":"The Stagger-grid applies a +0.4 dex alpha-element enhancement only to models with [Fe/H] <= -1, while the metallicity interpolation is performed in [M/H] as a smooth scalar field. KIC 10001167, used as a validation target in Section 4.3, has [Fe/H] = -0.73 and [alpha/Fe] = 0.37, placing it in the interval -1 < [Fe/H] < -0.5 where no alpha-enhanced Stagger-grid model exists. The interpolated 3D outer boundary for this star is therefore a blend of non-enhanced models near [Fe/H] = -0.5 and alpha-enhanced models near [Fe/H] = -1, a mixture that corresponds to no real stellar abundance pattern. The paper does not test or discuss this composition discontinuity, and it is a concrete candidate for the unexplained systematic overestimate of mass and mean density reported in Section 4.3. This issue is load-bearing because the central claim of a fixed alpha_MLT across -3 < [Fe/H] < 0.5 requires the interpolated mean 3D models to be a faithful, continuous function of chemical composition.","section":"Sections 2.2, 2.3, 4.3"},{"comment":"The validation shows offsets well outside the quoted uncertainties: the modeled [Fe/H] difference between the AI Phe components is 0.12 dex versus 0.04 dex observed; the best-fit masses are 1.229 Msun versus 1.178 +/- 0.015 Msun for KIC 9970396 and 0.998 Msun versus 0.934 +/- 0.008 Msun for KIC 10001167. The paper acknowledges that the mass and mean-density offset is not understood (Section 4.3). These systematic discrepancies do not by themselves invalidate the method, but they weaken the abstract and conclusion statements that the method 'successfully reproduces most observational constraints for all target stars.' The authors should either resolve these offsets, propagate them into the quoted model uncertainties, or explicitly moderate the claim to reflect the observed level of agreement.","section":"Sections 4.1-4.3, Tables 2-4"},{"comment":"The evolutionary tracks contain unphysical zigzags and kinks that the authors trace to discontinuities in partial derivatives from the interpolation scheme. Such kinks appear in parameter regions relevant for the validation grids, and the paper does not quantify how these numerical artifacts affect the fitted stellar parameters or the quoted posterior distributions. Since the conclusions present the method as a tool for asteroseismic characterization and cluster-age determination, the authors should quantify the impact of these track irregularities on derived parameters, or at least restrict the claim to regions where tracks are smooth.","section":"Section 3.2 and Fig. 7"},{"comment":"The interpolation test for the warm dwarf t62g43m00 shows relative temperature errors up to about 9% in the deep layers below the density inflection, and the paper cautions that systematic errors are likely larger in the high-Teff, high-log g region. The validation targets do not cover this region (Tables 2-4), so the abstract's range '-3 < [Fe/H] < 0.5' and 'FGK-type stars' is broader than the evidence presented. The authors should either add a test in that region or restrict the stated range of applicability.","section":"Appendix A, Fig. A2"}],"minor_comments":[{"comment":"The text contains a duplicated word in 'Fig.7 show shows surface boundary conditions affect the evolution'; please correct this typo.","section":"Section 3.2"},{"comment":"The statement 'A0.4dex enhancement of alpha-element abundances was applied to metal-poor models with [Fe/H]<=-1' should specify whether the +0.4 dex applies to [alpha/Fe] relative to the Asplund et al. (2009) mixture or to the total metal mass fraction; this matters for converting grid entries to [M/H].","section":"Section 2.2"},{"comment":"The observed [Fe/H] values are listed in the tables but the inferred posterior [Fe/H] values are not directly compared to them in the text; adding an explicit comparison would help the reader assess the accuracy of the metallicity interpolation.","section":"Sections 4.2-4.3, Tables 3-4"},{"comment":"Please specify whether rho_CZ and M_CZ in the turbulent diffusion coefficient are evaluated at the current model time step or at an initial reference structure; this affects reproducibility.","section":"Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":"The paper is honest about its limitations, but the composition discontinuity at [Fe/H] = -1 is not addressed even though the KIC 10001167 validation target lies in the affected interval. The systematic mass offsets are mentioned but not resolved; in a methods paper this is acceptable if the claims are scoped accordingly, but the current abstract and conclusions claim more than the validation supports. The alpha-insensitivity result is solid and well illustrated; the main work needed is to address the composition gap and to moderate or better support the generality claims."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe two things you should know: this is a real step forward for the 1D-3D coupling method, and the paper's own validation results are weaker than the abstract claims.\n\nWhat is genuinely new: the implementation of on-the-fly interpolation in [M/H] for the mean Stagger-grid models inside GARSTEC, which extends papers I and II beyond solar metallicity. The alpha_MLT insensitivity result (Fig. 4) is clean and convincing — a 20% change in alpha_MLT shifts Teff by under 30 K because the matching point sits in the near-adiabatic layer. That part of the central claim holds. The validation against AI Phe is also solid; the eclipsing binary masses and radii are well reproduced, and the 50 K Teff offsets are within the systematic noise.\n\nNow the soft spots, in proportion. The stress-test note is right: the Stagger-grid applies a fixed +0.4 dex alpha enhancement only at [Fe/H] ≤ -1. The [M/H] interpolation treats the grid as a smooth scalar field, so for -1 < [Fe/H] < -0.5 the boundary condition is a blend of alpha-enhanced and non-enhanced models that corresponds to no real star. KIC 10001167, with [Fe/H] = -0.73 and [α/Fe] = 0.37, sits exactly in that gap. The paper uses alpha-enhanced interior models for this star but a non-enhanced 3D boundary. The authors report a systematic overestimate of mass and mean density for both Kepler giants and say the cause is not understood. The alpha mismatch is a concrete, testable candidate. It may not be the whole story, but it is a genuine unaddressed issue that directly affects the validity of the fixed-alpha_MLT claim at intermediate metallicity.\n\nThe paper is honest about other weaknesses: the interpolation errors up to 9% for warm dwarfs (Appendix A), the unphysical kinks in the tracks (Sect. 3.2), and the lack of a full explanation for the RGB offsets. That candor is good. The absence of public code and data is a real limitation, especially since the method is fiddly and the interpolation details matter.\n\nBottom line: this is a serious paper with a clean core result and an important, but possibly patchable, flaw in the validation at intermediate metallicity. It deserves a serious referee. I would accept it for review, but ask the authors to address the alpha-enhancement discontinuity explicitly, re-examine KIC 10001167 with a consistent composition boundary, and release code/data. For a stellar evolution or asteroseismology audience, it belongs on the reading list. I would cite it for the method and the alpha_MLT insensitivity demonstration, with a caveat about the compositional inconsistency.\n\nRegards,","headline":"A real extension of the 1D-3D coupling method with a clean alpha_MLT insensitivity result, but the validation at intermediate metallicity is undercut by an unaddressed alpha-enhancement inconsistency.","tokens_in":31817,"tokens_out":5585,"would_cite":true,"duration_ms":55871,"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":"Coupling 1D stellar models to 3D atmospheres makes evolutionary tracks nearly independent of the mixing-length parameter.","keywords":["1D-3D coupling","mixing-length parameter","stellar evolution","3D hydrodynamical simulations","Stagger-grid","asteroseismology","eclipsing binaries","stellar atmospheres"],"falsifier":"Take a detached eclipsing binary at $[{\\rm Fe/H}]\\approx-1.5$ with masses, radii, and effective temperatures known to about 1%; if the 1D-3D coupled track with solar-calibrated $\\alpha_{\\rm MLT}$ cannot match both radii and temperatures within roughly 30 K while a standard model with tuned $\\alpha_{\\rm MLT}$ can, the claimed insensitivity to the mixing-length parameter is falsified in that metallicity regime.","tokens_in":30552,"feed_emoji":"🌟","tokens_out":11187,"duration_ms":105296,"temperature":0.7,"pith_summary":"This paper extends the 1D-3D coupling method, which replaces the near-surface layers of a one-dimensional stellar evolution model with a horizontal- and time-averaged three-dimensional hydrodynamical atmosphere at every time step, from solar metallicity to the full range $-3<[{\\rm Fe/H}]<0.5$. Its central result is that when the outer boundary of the interior model sits well below the photosphere in the near-adiabatic layer, the mixing-length parameter $\\alpha_{\\rm MLT}$ becomes almost irrelevant: changing it by 20% shifts the effective temperature by under 30 K, versus more than 200 K in standard evolution calculations. The paper argues that $\\alpha_{\\rm MLT}$ can therefore be fixed to its solar-calibrated value, and validates this against both components of the eclipsing binary AI Phe and two oscillating red giants in Kepler binaries, reproducing radii, temperatures, and oscillation frequencies without empirical surface corrections. If correct, the method removes the most uncertain tunable parameter from low-mass stellar modeling and sharpens ages for metal-poor stars of interest to galactic archaeology.","feed_headline":"1D-3D coupling makes the mixing-length parameter nearly irrelevant","feed_subtitle":"A 20% change in the mixing-length parameter shifts tracks by under 30 K instead of over 200 K.","key_machinery":"The mean 3D model, the horizontal- and time-averaged Stagger-grid simulation, supplies the outer boundary conditions. The load-bearing identity is the density inflection region: each model is trimmed to its minimum superadiabatic temperature gradient and scaled by quantities at the density inflection (the local minimum of $\\partial\\ln\\rho/\\partial\\ln P$), making the scaled structure nearly universal and enabling robust linear or cubic interpolation in $(T_{\\rm eff},\\log g)$ and monotonic cubic interpolation in $[{\\rm M/H}]$. The matching point is placed at $P=10^{1.2}P_{\\rm di}$, about 15.8 times the density-inflection pressure, deep enough that $\\nabla_T-\\nabla_{\\rm ad}$ is an order of magnitude below its surface peak; the pressure and luminosity boundary conditions therefore come from the 3D model, and the mixing-length parameter has almost no lever arm on the track.","core_discovery":"The paper's claim is that the 1D-3D coupling method, extended across metallicity by interpolating mean 3D Stagger-grid models on the fly, makes stellar evolution tracks insensitive to the mixing-length parameter. By placing the matching point between the 1D interior and the 3D atmosphere in the near-adiabatic convective region, the temperature structure there is set by the 3D simulation rather than by mixing-length theory, so $\\alpha_{\\rm MLT}$ only governs heat transport in layers where convection is nearly adiabatic and the superadiabatic gradient is small. With $\\alpha_{\\rm MLT}$ fixed to the solar-calibrated value, the method reproduces the observed properties of stars in detached eclipsing binaries and of asteroseismic giants across $-3<[{\\rm Fe/H}]<0.5$. The authors' stated conclusion is that $\\alpha_{\\rm MLT}$ can be regarded as a constant determined by solar calibration.","pith_inferences":["If constant $\\alpha_{\\rm MLT}$ holds, the observationally inferred growth of $\\alpha_{\\rm MLT}$ with [Fe/H] in standard models may be an artifact of the near-surface boundary treatment rather than a real metallicity dependence of convection efficiency.","The grid's fixed +0.4 dex $\\alpha$-element enhancement for $[{\\rm Fe/H}]\\le-1$ implies specific surface abundance patterns; spectroscopic measurement of $\\alpha$-elements in metal-poor eclipsing binaries would test this boundary assumption directly.","The systematic overestimate of mean density for the two Kepler giants suggests a residual, possibly interpolation-driven surface term; a dedicated study of surface effects in coupled red-giant models could separate numerical from physical causes.","Applying the same coupling to stars with $T_{\\rm eff}\\gtrsim6300$ K requires handling diffusion-induced surface composition changes, since the metallicity interpolation is based on $Z/X$ and could otherwise drift outside the grid."],"forward_implications":["The mixing-length parameter can be fixed at its solar-calibrated value for all covered metallicities, removing a major free parameter from stellar evolution calculations.","Model oscillation frequencies for red giants agree with observed radial and mixed modes closely enough that empirical surface-effect corrections are largely unnecessary.","Stellar ages, which are strongly degenerate with $\\alpha_{\\rm MLT}$ and $Y_{\\rm init}$ for red giants, can be constrained more tightly, with direct consequences for globular-cluster ages.","The $\\nu_{\\rm max}$ scaling relation becomes a testable prediction rather than a calibration output, since $T_{\\rm eff}$ and radius are insensitive to $\\alpha_{\\rm MLT}$.","The method turns pre-computed 3D simulation grids into on-the-fly boundary conditions for evolutionary modeling of field and cluster stars across a wide metallicity range."],"supporting_citations":[{"why":"Develops the interpolation technique based on the density inflection region that the metallicity extension relies on.","marker":"Jørgensen et al. (2017)"},{"why":"Paper I establishes the on-the-fly 1D-3D coupling scheme for the solar model.","marker":"Jørgensen et al. (2018)"},{"why":"Paper II applies the coupling to evolutionary tracks and oscillation frequencies at solar metallicity.","marker":"Mosumgaard et al. (2020)"},{"why":"Presents the Stagger-grid of 3D surface convection simulations used for the boundary conditions.","marker":"Magic et al. (2013a)"},{"why":"Provides the refined and extended Stagger-grid covering the metallicities used here.","marker":"Rodríguez Díaz et al. (2024)"},{"why":"Supplies precise masses and radii for the AI Phe eclipsing binary used in validation.","marker":"Maxted et al. (2020)"},{"why":"Gives the dynamical mass and radius of KIC 9970396 used to select evolutionary models.","marker":"Brogaard et al. (2018)"},{"why":"Extracts the oscillation frequencies for KIC 9970396 that the coupled models must reproduce.","marker":"Li et al. (2018)"},{"why":"Provides mass, radius, and metallicity of KIC 10001167 used as a low-metallicity validation target.","marker":"Thomsen et al. (2025)"},{"why":"Supplies the measured frequencies and the metallicity-α_MLT context for KIC 10001167.","marker":"Li et al. (2024b)"}],"fun_headline_variants":["Stellar evolution tracks nearly immune to mixing-length choice","Mixing-length parameter becomes a detail in new 1D-3D stellar models","1D-3D star models: 20% mixing-length change shifts tracks by under 30 K","New stellar models make convection parameter almost vanish","Metallicity-spanning 1D-3D stellar evolution tames mixing-length parameter"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The interpolated mean 3D Stagger-grid models faithfully represent the near-surface stratification of real FGK stars across $-3<[{\\rm Fe/H}]<0.5$, including the fixed +0.4 dex $\\alpha$-enhancement for metal-poor models; if the grid or the interpolation is biased in any region, every evolutionary track and validation conclusion inherits that bias.","fun_headline_variants_meta":{"raw":{"variants":["Stellar evolution tracks nearly immune to mixing-length choice","Mixing-length parameter becomes a detail in new 1D-3D stellar models","1D-3D star models: 20% mixing-length change shifts tracks by under 30 K","New stellar models make convection parameter almost vanish","Metallicity-spanning 1D-3D stellar evolution tames mixing-length parameter"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0007,"raw_usage":{"total_tokens":3227,"prompt_tokens":1075,"completion_tokens":2152,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":691,"completion_tokens_details":{"reasoning_tokens":2062}},"tokens_in":691,"tokens_out":2152,"duration_ms":17049,"temperature":1.0,"reasoning_tokens":2062,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T10:25:11.635750+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a detached eclipsing binary at $[{\\rm Fe/H}]\\approx-1.5$ with masses, radii, and effective temperatures known to about 1%; if the 1D-3D coupled track with solar-calibrated $\\alpha_{\\rm MLT}$ cannot match both radii and temperatures within roughly 30 K while a standard model with tuned $\\alpha_{\\rm MLT}$ can, the claimed insensitivity to the mixing-length parameter is falsified in that metallicity regime.","supporting_citations":[{"cited_title":"R., J rgensen A","cited_arxiv_id":null,"evidence_quote":"Paper II applies the coupling to evolutionary tracks and oscillation frequencies at solar metallicity."},{"cited_title":"Advancing accuracy in age determinations of old-disk stars using an oscillating red giant in an eclipsing binary","cited_arxiv_id":"2504.17853","evidence_quote":"Provides mass, radius, and metallicity of KIC 10001167 used as a low-metallicity validation target."}],"review_version":1}