Exploring Mixing Thresholds in Asteroseismic Stellar Evolution Models

Lynn Buchele

REVIEW 3 major objections 6 minor 6 references

Exploring Mixing Thresholds in Asteroseismic Stellar Evolution Models

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper concludes that the numerical cutoff overshoot_D_min in MESA's exponential overshoot prescription should be set to 10^-2 cm^2/s or lower, because at the default 10^2 cm^2/s abundance profiles truncate abruptly and asteroseismic…

desk verdict A clean, reproducible numerical study showing that MESA's default overshoot_D_min truncation creates noisy asteroseismic frequencies, but the absolute 1e-2 recommendation lacks the resolution tests needed to be robust. read the letter →

arxiv 2507.23275 v1 pith:T7SKOWNH submitted 2025-07-31 astro-ph.SR

Lynn Buchele This is my paper

classification astro-ph.SR

keywords asteroseismologyconvectiveovershootingstellarevolutionmodelsMESAexponentialovershootcorenumericalresolutionfrequencyinterpolation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper is about a small numerical cutoff in stellar models and the large effect it can have on the predicted pulsation frequencies used to read stellar interiors. Exponential overshooting, a standard way to add mixing just beyond a convective zone, decays forever, so a model must decide when to stop tracking it. Comparing a 1.35 solar-mass model with cutoff values $10^{2}$, 1, and $10^{-2}$ $cm^{2}$/s, the paper finds that the highest value, the MESA default, leaves a sharp step in the hydrogen abundance profile at the end of the overshoot region, and that step jumps between successive models and makes the radial-mode frequency evolution noisy. Interpolation errors exceed 1 µHz, well above the roughly 0.1 µHz precision typical of Kepler asteroseismic observations. The paper concludes that overshoot_D_min should be set to $10^{-2}$ $cm^{2}$/s or lower for main-sequence stars with growing convective cores.

What carries the argument

The central object is the exponential overshoot prescription $D_{\rm ov}(z)=D_0\exp(-2z/(f H_p))$, in which the mixing coefficient outside a convective boundary decays exponentially with distance $z$, set by the scale height $H_p$ and a free parameter $f$. Because the exponential never reaches zero, the prescription introduces a numerical cutoff $D_{\rm ov}^{\rm limit}$ (the MESA inlist parameter overshoot_D_min) below which mixing is set to zero. The paper's diagnostic machinery is the cubic-spline interpolation error $\epsilon_{\rm interp} = \nu_{\rm GYRE} - \nu_{\rm interp}$ for radial-mode frequencies, which converts mesh-dependent composition jumps into a number comparable to observed Kepler frequency uncertainties.

What would settle it

Run the same comparison at another stellar mass (for example, 2 $M_\odot$), with a finer MESA mesh, or with a non-radial mode; if interpolation errors above roughly 0.1 µHz or discontinuous abundance steps persist when overshoot_D_min is $10^{-2}$ $cm^{2}$/s, the recommendation is falsified. Conversely, if mesh spacing alone moves the apparent cutoff, the threshold is a numerical artifact rather than a universal physical constant.

Watch

Extended reading notes

Core claim

The central claim is that the threshold at which exponential overshooting is truncated is not a harmless numerical detail but a physical choice that changes the smoothness of stellar structure and the reliability of calculated frequencies. At a cutoff of $10^{2}$ $cm^{2}$/s, the diffusion coefficient is still large enough at the cutoff edge to mix material fully, so the hydrogen profile ends in a sharp chemical step; the step's location depends on MESA's mesh placement and therefore moves discontinuously from model to model. As the convective core grows, this discontinuous composition evolution produces a noisy frequency derivative and interpolation errors above 1 µHz. Setting the cutoff to $10^{-2}$ $cm^{2}$/s, the value originally proposed for the prescription, removes the composition step and yields smooth frequency evolution. The author's recommendation is therefore to set overshoot_D_min to $10^{-2}$ $cm^{2}$/s or lower for these stars, and to consider changing the MESA default.

Load-bearing premise

The load-bearing premise is that one 1.35 solar-mass model, one radial mode, and one fixed input-physics setup can stand in for all main-sequence stars with growing convective cores; if the safe cutoff changes with mass, composition, numerical resolution, or mode, the paper's recommended global threshold does not transfer.

Editorial extensions

If this is right

Asteroseismic models of main-sequence stars with growing convective cores should adopt overshoot_D_min of 10^-2 cm^2/s or lower to keep abundance profiles smooth from model to model.
With the current MESA default of 10^2 cm^2/s, radial-mode frequency interpolation errors exceed 1 µHz during core growth, larger than the roughly 0.1 µHz uncertainty of Kepler observations, so model grids built at the default may not be reliable for frequency interpolation.
The smooth mixing profiles produced by the lower cutoff should also make the stellar models more numerically robust, beyond the specific frequency-based test.
The same abrupt-truncation problem may affect other masses and evolutionary stages that use exponential overshooting, such as massive main-sequence stars and shell-burning stars, though those cases are not tested here.
If wider tests confirm the result, the MESA default value itself may need to be lowered to the original 10^-2 cm^2/s proposal.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A direct test of the threshold's generality would be to repeat the comparison for several stellar masses, metallicities, and grid resolutions; the paper's single 1.35 $M_\odot$ track cannot rule out that the required cutoff is resolution- or mass-dependent, and the recommended value may need to be lower for finer meshes.
The tell-tale frequency noise at high cutoffs resembles real asteroseismic signals such as structural glitches or mode trapping, so grid-based fitting pipelines that use default MESA models could mistake numerical truncation for stellar physics.
It is worth checking whether the interpolation-error criterion generalizes to non-radial modes; if dipole or quadrupole modes probe different depths of the overshoot region, the safe cutoff could differ from the one calibrated on a single radial mode.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: the paper makes a modest but useful point. MESA's default overshoot_D_min of 100 cm^2/s truncates exponential overshooting with a sharp composition step, and that step shows up as noisy GYRE frequencies and interpolation errors above the typical Kepler uncertainty. Lowering the cutoff to Herwig's 1e-2 cm^2/s smooths the composition profile and the frequency evolution. The demonstration uses three tracks of a 1.35 solar-mass model, one radial mode, and a cubic spline interpolation test. It's a clean numerical experiment, not deep physics.

What's genuinely new is the direct test of this numerical parameter on asteroseismic frequencies for a growing convective core. I also give credit for reproducibility: the paper states the MESA version, provides a Zenodo link for all files, and is explicit about the mechanism. The author notes that the abrupt step's location depends on where MESA places mesh points, which is the right caveat to flag.

The stress-test concern is fair and lands. The paper varies only overshoot_D_min. It never varies spatial resolution, mesh control, or timestep control, and it tests nothing below 1e-2. So we cannot tell whether the improvement at 1e-2 reflects numerical convergence or an accidental match to this particular mesh. The abstract's recommendation of 1e-2 'or lower' is stronger than the evidence, which shows only that 1e-2 is better than 100 and 1 for one star and one mode. The generalization to 'these stars' is plausible but not established.

Still, the core observation is likely correct. The mechanism is physically sensible: when the cutoff is high, Dov at the boundary is still large enough to mix, and the sharp edge moves discretely with mesh points. The interpolation error exceeding 0.1 uHz is a real practical problem for grid-based asteroseismology. The paper is honest about its limitations and does not overclaim the broader applicability beyond noting it 'may affect' other cases.

Who is this for: anyone running MESA with exponential overshooting for asteroseismology, especially those building model grids. It deserves a serious referee. The right request would be a convergence test across resolution and timestep, plus at least one other mass or metallicity, and a check below 1e-2 to see if the effect plateaus. I'd encourage engaging with it as a useful numerical caution, with the recommendation treated as provisional until the robustness check is done.

Referee Report

3 major / 6 minor

Summary. The paper examines the role of the numerical cutoff parameter overshoot_D_min in MESA's exponential overshoot prescription. Using a 1.35 solar-mass main-sequence model, the author compares three values of overshoot_D_min (100, 1, and 1e-2 cm^2/s) while keeping other parameters fixed, computes radial-mode frequencies with GYRE, and analyzes the frequency evolution, its numerical derivative, and a cubic-spline interpolation error. The paper finds that the highest cutoff produces a sharp truncation of the overshoot region, mesh-dependent abundance steps, and interpolation errors above 1 uHz, whereas the lowest tested value yields smooth abundance profiles and frequency evolution. The central conclusion is that overshoot_D_min should be set to 1e-2 cm^2/s or lower for main-sequence stars with growing convective cores, and the author suggests this value may be worth adopting as a MESA default.

Significance. If the conclusion is robust, the paper provides a simple, practical recommendation for MESA users who model main-sequence stars with exponential overshooting, and it identifies a numerical artifact that can compromise asteroseismic frequency predictions. The paper has clear strengths: it uses a reproducible setup (Zenodo files are provided), a direct three-track comparison, and a plausible physical mechanism linking the cutoff value to discontinuous composition profiles and frequency noise. However, the significance is currently limited by the narrow empirical basis: one stellar mass, one radial mode, one set of input physics, and no quantitative convergence tests. The claim is not circular, as the recommendation is based on independent MESA/GYRE calculations rather than on fitting to observed data, but the generalizability and robustness of the absolute threshold remain open.

major comments (3)

[§2, right column of Fig. 1] The paper itself states that the sharp composition step's location 'depends on where MESA places mesh points, which can change between successive models.' This admission makes the observed discontinuity at overshoot_D_min=100 cm^2/s at least partly a mesh-sampling artifact. However, no experiment varies spatial resolution, mesh control, or time-step control while holding overshoot_D_min fixed. Without such a convergence test, the claim that 1e-2 cm^2/s is sufficient (and that 100 cm^2/s is always problematic) is not established even for the tested 1.35 Msun model. Please add a resolution study (e.g., varying mesh_delta_coeff or a similar mesh control) and show that the interpolation error at each overshoot_D_min value is robust to resolution.
[§2 and §3] The recommendation is based on a single stellar mass (1.35 Msun), a single radial mode (radial order 19), and one fixed set of input physics. Section 3 extends the conclusion to 'these stars' and suggests broader applicability to other masses and evolutionary stages, but no evidence is provided that the artifact or the recommended threshold is mass-, metallicity-, or mode-independent. This is a load-bearing generalization: the mechanism involving a growing convective core and mesh placement could plausibly vary with stellar parameters. Either restrict the conclusion to the tested configuration or add additional tracks and modes to support the broader statement.
[Eq. (2) and Fig. 1] The interpolation-error metric uses a single holdout scheme (excluding every fifth model) and a cubic spline, with no sensitivity test to the holdout fraction, spline order, or sampling density. The comparison to the 0.1 uHz Kepler uncertainty is introduced in the text, but no uncertainty or convergence criterion is attached to epsilon_interp itself. Since the central recommendation is justified by the smoothness of the frequency evolution as quantified by this metric, the paper should demonstrate that the reported differences in epsilon_interp are stable with respect to the interpolation setup and are not accidental products of the chosen sampling.

minor comments (6)

[References] The reference list contains two instances of 'therin' (in the Anders & Pedersen and Jermyn et al. entries) that should be 'therein.'
[§2] The phrase 'radial mode with radial order 19' should be made more precise; please state the quantum numbers (e.g., n=19, l=0) and clarify whether this is a p mode, so that readers can reproduce the mode identification.
[Fig. 1, right column] The horizontal axis label 'm/M' should be defined in the caption or text as the mass coordinate in units of the total stellar mass.
[Fig. 1, bottom-left panel] The '1 sigma range typical of frequencies of stars observed by Kepler' is shown as dark gray shading; please specify in the caption whether this corresponds to 0.1 uHz or a range, and state the source of this uncertainty estimate.
[Eq. (2)] Define nu_GYRE and nu_interp as mathematical symbols in the text before or immediately after the equation, rather than only in parentheses, to avoid ambiguity.
[§2] Please state the exact MESA and GYRE version identifiers and the full inlist settings used, or confirm that the Zenodo deposit contains all of these, so that the three-track comparison is fully reproducible.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the recommendation is an empirical comparison of three MESA cutoff values against GYRE frequencies; no fitted quantity is renamed a prediction.

full rationale

The paper's central claim is that overshoot_D_min should be set to 10^-2 cm^2/s or lower for main-sequence stars with growing convective cores. This is supported by a direct parameter study: three values (100, 1, and 10^-2 cm^2/s) are run in MESA with all other parameters fixed, and the resulting abundance profiles and GYRE radial-mode frequencies are compared. No parameter is fitted to the asteroseismic frequencies, and the recommended threshold is not an input to the calculation of the very quantity used to justify it. The interpolation error is defined as epsilon_interp = nu_GYRE - nu_interp, where nu_interp comes from a cubic spline fit; this is a diagnostic of smoothness, not a fitted quantity masquerading as a prediction. Herwig (2000) is cited historically for proposing the same value, but the paper independently demonstrates smoother abundance and frequency evolution at that value relative to higher cutoffs; no uniqueness theorem or load-bearing self-citation is invoked. The manuscript does contain a limitation relevant to robustness: it states that 'The location of this sharp change depends on where MESA places mesh points, which can change between successive models,' and only three D_min values are tested without varying mesh resolution or time-step controls. This is a real concern about whether the absolute threshold generalizes across numerical configurations, but it is not circularity: the conclusion is not defined in terms of itself, and the empirical observation at the tested settings stands. The single 1.35 solar-mass model is likewise a scope limitation, not a circular step. No circular chain is present, so the score is 0.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The work rests on standard domain assumptions that MESA and GYRE correctly implement the stellar physics, and on an implicit representativeness assumption for the single 1.35 M_sun test case. No new physical entities are introduced.

free parameters (2)

overshoot_D_min = tested values: 100, 1, and 0.01 cm^2/s
The numerical cutoff below which exponential overshoot mixing is truncated; this is the central parameter varied in the study and the subject of the recommendation.
f (exponential overshoot extent)
Determines the overshoot length scale in Herwig (2000) exponential profile; fixed across models but its value is not reported in the text, which is needed for exact reproduction.

assumptions (4)

domain assumption Exponential overshoot prescription (Herwig 2000) is the correct model for additional mixing beyond convective boundaries.
Used in Eq. 1 to define Dov(z); standard in stellar evolution modeling, not independently tested in this paper.
domain assumption MESA v24.08.1 correctly implements stellar evolution physics and the overshoot cutoff.
The central computational tool; the paper does not verify against independent stellar evolution codes.
domain assumption GYRE frequencies are accurate enough for the interpolation-error analysis.
Used to compute the test mode frequencies; no convergence test or comparison with other oscillation codes is reported.
ad hoc to paper The 1.35 M_sun track is representative of all main-sequence stars with growing convective cores.
Assumed when generalizing beyond the single tested mass in Section 3; not demonstrated with additional masses or evolutionary stages.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Exploring Mixing Thresholds in Asteroseismic Stellar Evolution Models." pith.science (2026). https://pith.science/paper/T7SKOWNH

@misc{pith2026250723275,
  author       = {Pith},
  title        = {Pith review of: Exploring Mixing Thresholds in Asteroseismic Stellar Evolution Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T7SKOWNH}},
  note         = {Machine review of arXiv:2507.23275}
}

abstract

Inferences from observations clearly show that mixing in stars extends beyond the convective boundaries defined by mixing length theory. This triggered the proposal of a variety of prescriptions to include additional mixing in stellar models. These prescriptions typically introduce free parameters to set the extent of the additional mixing and may also introduce numerical parameters. In the case of exponential overshooting, one must decide the threshold at which the exponential decay of the mixing coefficient can be treated as zero. Using the MESA stellar evolution code, I explore the effect of varying this parameter on asteroseismic models of main-sequence stars with growing convective cores. From this, I conclude that overshoot_D_min should be set to $10^{-2}$ cm$^2$/s or lower for these stars. The default value in MESA is four orders of magnitude higher than this recommendation, which results in discontinuous evolution.

Figures

Figures reproduced from arXiv: 2507.23275 by the authors.

**Figure 1.** Left: The top panel shows the test mode frequency over time. The inset zooms in to show some differences in the frequency evolution. The middle panel shows the numerical time derivative of the test frequency. The bottom panel shows the interpolation error as defined in Equation 2. The dark gray horizontal shading shows the 1σ range typical of frequencies of stars observed by Kepler. In all three panels, the vertical… view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

6 extracted references · 1 canonical work pages

[1]

- [1] #1 = = ^ ^ ^ .\!\!^ d .\!\!^ h .\!\!^ m .\!\!^ s .\!\!^ @mss

thebibliography [1] 20pt to REFERENCES 6pt =0pt \@twocolumntrue 12pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key o...

work page 2017
[2]

H., & Pedersen , M

Anders , E. H., & Pedersen , M. G. 2023, title Convective Boundary Mixing in Main-Sequence Stars: Theory and Empirical Constraints , Galaxies, 11, 56, 10.3390/galaxies11020056

doi:10.3390/galaxies11020056 2023
[3]

2000, title The evolution of AGB stars with convective overshoot , , 360, 952, 10.48550/arXiv.astro-ph/0007139

Herwig , F. 2000, title The evolution of AGB stars with convective overshoot , , 360, 952, 10.48550/arXiv.astro-ph/0007139

arXiv doi:10.48550/arxiv.astro-ph/0007139 2000
[4]

S., Bauer , E

Jermyn , A. S., Bauer , E. B., Schwab , J., et al. 2023, title Modules for Experiments in Stellar Astrophysics (MESA): Time-dependent Convection, Energy Conservation, Automatic Differentiation, and Infrastructure , , 265, 15, 10.3847/1538-4365/acae8d

doi:10.3847/1538-4365/acae8d 2023
[5]

2023, title A Review of the Mixing Length Theory of Convection in 1D Stellar Modeling , Galaxies, 11, 75, 10.3390/galaxies11030075

Joyce , M., & Tayar , J. 2023, title A Review of the Mixing Length Theory of Convection in 1D Stellar Modeling , Galaxies, 11, 75, 10.3390/galaxies11030075

doi:10.3390/galaxies11030075 2023
[6]

Townsend , R. H. D., & Teitler , S. A. 2013, title GYRE: an open-source stellar oscillation code based on a new Magnus Multiple Shooting scheme , , 435, 3406, 10.1093/mnras/stt1533

doi:10.1093/mnras/stt1533 2013

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

- [1] #1 = = ^ ^ ^ .\!\!^ d .\!\!^ h .\!\!^ m .\!\!^ s .\!\!^ @mss

thebibliography [1] 20pt to REFERENCES 6pt =0pt \@twocolumntrue 12pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key o...

work page 2017

[2] [2]

H., & Pedersen , M

Anders , E. H., & Pedersen , M. G. 2023, title Convective Boundary Mixing in Main-Sequence Stars: Theory and Empirical Constraints , Galaxies, 11, 56, 10.3390/galaxies11020056

doi:10.3390/galaxies11020056 2023

[3] [3]

2000, title The evolution of AGB stars with convective overshoot , , 360, 952, 10.48550/arXiv.astro-ph/0007139

Herwig , F. 2000, title The evolution of AGB stars with convective overshoot , , 360, 952, 10.48550/arXiv.astro-ph/0007139

arXiv doi:10.48550/arxiv.astro-ph/0007139 2000

[4] [4]

S., Bauer , E

Jermyn , A. S., Bauer , E. B., Schwab , J., et al. 2023, title Modules for Experiments in Stellar Astrophysics (MESA): Time-dependent Convection, Energy Conservation, Automatic Differentiation, and Infrastructure , , 265, 15, 10.3847/1538-4365/acae8d

doi:10.3847/1538-4365/acae8d 2023

[5] [5]

2023, title A Review of the Mixing Length Theory of Convection in 1D Stellar Modeling , Galaxies, 11, 75, 10.3390/galaxies11030075

Joyce , M., & Tayar , J. 2023, title A Review of the Mixing Length Theory of Convection in 1D Stellar Modeling , Galaxies, 11, 75, 10.3390/galaxies11030075

doi:10.3390/galaxies11030075 2023

[6] [6]

Townsend , R. H. D., & Teitler , S. A. 2013, title GYRE: an open-source stellar oscillation code based on a new Magnus Multiple Shooting scheme , , 435, 3406, 10.1093/mnras/stt1533

doi:10.1093/mnras/stt1533 2013