Expanding stellar associations as Galactic accelerometers

(2) University of Barcelona); N\'uria Miret Roig (2) ((1) University of Helsinki; Till Sawala (1)

REVIEW 3 major objections 5 minor 62 references

Expanding stellar associations as Galactic accelerometers

T0 review · 3 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read Expanding young star clusters can be rewound to weigh the Milky Way's gravitational potential.

desk verdict A clean, honest proof-of-concept for a new local potential probe; read the abstract precision numbers as in-sample, not yet as Milky Way constraints. read the letter →

arxiv 2608.05297 v1 pith:2JFJXWOM submitted 2026-08-05 astro-ph.GA

Till Sawala (1) , Núria Miret Roig (2) ((1) University of Helsinki , (2) University of Barcelona) This is my paper · ORCID

classification astro-ph.GA

keywords GalacticpotentialyoungstellarassociationsdynamicaltracebackorbitreconstructionGaiaastrometrydarkmatterhaloMilkyWaymasstidaltensor

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

The paper claims that the Milky Way's gravitational potential can be measured by rewinding the orbits of stars in young, expanding stellar associations. Because all members were born in a compact region, the correct potential is the one in which the association shrinks to the smallest size when integrated backward in time. Using synthetic associations, the authors show that with Gaia DR4 astrometry and radial velocities below 0.2 km/s, a single association could measure the dark halo mass to about 0.6 x $10^{12}$ solar masses and its concentration to better than 0.8. They also show that the dynamical traceback age depends on the assumed potential, so independent age measurements can break degeneracies, but also that traceback ages are never free of the potential assumption.

What carries the argument

The central object is the position covariance matrix C of the association's member stars, summarized by the trace tr = Tr(C)^(1/2) and the determinant det = det(C)^(1/6), which measure the association's size as it evolves backward. The machinery is orbit integration in trial parametric potentials, namely a spherical NFW dark halo, a Miyamoto-Nagai disk, a bulge, and a nucleus, with the minimum of tr or det as the objective function. The trace behaves smoothly under noise; the determinant responds to contraction in any dimension but is numerically stiff and dominated by sample variance. The same covariance evolution defines the traceback ages tau_tr and tau_det, making the potential and the age jointly inferable.

What would settle it

Run the inference on synthetic associations generated from a simulated Milky-Way-like galaxy that includes a bar and spiral arms, fitting only the static axisymmetric family; if the recovered halo mass and concentration are offset from the true input values by more than the quoted uncertainties, the central premise fails in realistic potentials. A purely observational check is to take a real association with an independent isochronal age and near-future data, and test whether the potential that minimizes the covariance also makes the traceback age agree with the isochronal age to within the combined uncertainty.

Watch

Extended reading notes

Core claim

The central discovery is a new inference criterion: the true Galactic potential is the one that minimizes the trace or determinant of the position covariance matrix of an expanding young stellar association when its member orbits are integrated backward to birth. The same observed present-day positions and velocities are used in every trial potential; only the potential changes. In the true potential the members reconverge to the most compact configuration, and the time of that minimum is the dynamical traceback age. With near-future data the minimum-trace criterion can distinguish halo masses, concentrations, and disk masses, though with strong degeneracies; the determinant is sharper for true three-dimensional focusing but noisier. A second, complementary route infers the potential by requiring the dynamical traceback age to match an independently measured stellar age.

Load-bearing premise

The inference is only valid if the true Milky Way potential is well described by the assumed static, axisymmetric family, namely a spherical NFW dark halo plus a single smooth Miyamoto-Nagai disk, over the association's 20 to 60 million year lifetime, and the paper itself concedes that the single-disk treatment is 'certainly not accurate'.

Editorial extensions

If this is right

With 'next'-generation errors (astrometry at the 10 microarcsecond level and radial velocities at 0.2 km/s), a single association can constrain the halo mass with an uncertainty of about 5.4 x 10^11 solar masses, the halo concentration with an uncertainty of about 0.83, and the disk mass with an uncertainty of about 2.3 x 10^9 solar masses, when each parameter is fitted alone.
With further-improved 'future' errors (2.5 microarcsecond astrometry and 0.025 km/s radial velocities), these single-association uncertainties drop to about 6 x 10^10 solar masses, 0.14, and 4 x 10^8 solar masses respectively.
The inferred dynamical traceback age depends on the assumed potential; at 'next' precision the systematic age error from potential uncertainty can exceed the statistical error, so traceback-age estimates are not independent of the potential.
An independent age measurement, such as an isochronal age, can be used as a second likelihood: the correct potential is the one in which the traceback age best matches the independent age.
Degeneracies between halo mass, halo concentration, and disk mass persist, but they align with physically meaningful local quantities such as the circular velocity at the Sun, the radial tidal tensor, and the Oort constant, so combining multiple associations or independent priors can break them.

Reading between the lines

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

As an extension of the paper's logic, the covariance-minimum criterion is effectively a measurement of the local acceleration and tidal tensor, so any departure from the assumed parametric potential family, such as a bar, spiral arms, or disequilibrium disk motions, would bias the results regardless of observational precision.
The same criterion could be applied to older structures or to cluster families that share a common origin, allowing radial velocities to be averaged over many members and relaxing the per-star precision requirement.
The potential-dependence of traceback ages implies that published dynamical ages carry hidden model dependence; a systematic comparison with isochronal ages across many associations could map local variations in the Galactic potential rather than merely measuring a global halo.
If an independent age is available, the two inference routes of minimum size and age matching become a powerful consistency test, and a significant disagreement between them would flag missing physics such as non-axisymmetric forces or kinematic contamination.

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

You should know two things about arXiv:2608.05297. First, it does something genuinely new: it rewinds expanding young associations and looks for the trial potential that minimizes the trace or determinant of the position covariance matrix, using the smallest minimum as an objective function to infer the Galactic potential. That is not just traceback-age territory; the covariance-minimum criterion over trial potentials is a real novelty. Second, the eye-catching precision figures in the abstract—sigma(M_h) about 0.6e12 solar masses, sigma(c_NFW) below 0.8—are in-sample precision from mock data generated with the same static, axisymmetric potential family and the same Gala integrator used for inference. The paper is transparent about the major caveats in Section 5.3, but the abstract reads more optimistically than the full text supports.

The paper earns its keep as a well-executed proof-of-concept. The synthetic experiments are careful: three error budgets, Monte Carlo sampling, explicit treatment of sample variance, and sensible exploration of intrinsic parameters like the number of stars, initial size, and velocity dispersion. The code is available. The secondary point is also valuable: traceback ages depend on the assumed potential, so age and potential should be fitted jointly rather than treating one as known. That observation alone is worth keeping in mind for anyone doing orbit rewinding.

The soft spot is real and load-bearing for the headline claim: the demonstration is entirely in-sample. The forward and backward integrations use the same potential model and the same integrator, so the method's ability to recover the injected parameters is partly guaranteed by construction. If the real Milky Way tidal field near the Sun includes contributions from the bar, spiral arms, non-equilibrium disk motions, or a non-spherical halo, those act as unmodelled systematics that could shift best-fit parameters by more than the quoted statistical uncertainties. The authors concede the single-disk treatment is "certainly not accurate." For a proof-of-concept this is acceptable, but it means the precision numbers are not yet demonstrated for the real Galaxy. The obvious next step is a test with an independent, more realistic potential model or at least a second integrator, plus application to a real association. A minor point: the determinant metric is noisy and poorly behaved under sample variance; the trace is more robust, and the paper itself shows that clearly.

Who gets value from this: people working on Galactic potential inference, traceback ages, local dark matter density, and Gaia-era astrometry. It deserves a serious referee. The referee should push for out-of-sample validation and a more careful framing of the precision claims, but the core idea is sound and should go through review. I would send it to peer review rather than desk-reject it.

Referee Report

3 major / 5 minor

Summary. The paper proposes a novel method to constrain the Milky Way's gravitational potential using expanding young stellar associations. The method relies on the physical prior that association members share a compact birth region: when the observed stellar orbits are integrated backward in a trial potential, the true potential is argued to be the one that minimizes a measure of the association's concentration, specifically the trace or determinant of the position covariance matrix. The authors generate synthetic associations by forward integration in a parametric static, axisymmetric potential (NFW halo plus Miyamoto-Nagai disk, bulge, and nucleus) and add observational errors according to three budgets ('current', 'next', 'future'). They show that with 'next' errors (Gaia DR4-like astrometry plus radial velocities of 0.2 km/s) a single association can recover the halo mass and concentration to within roughly 0.5e12 Msun and 0.8, respectively, under one-parameter variations, while current errors are too large. They also demonstrate that the dynamical traceback age depends on the assumed potential and can itself be used to constrain the potential when an independent age is available.

Significance. If the forecast precision were robust to real-world model misspecification, the method would offer a genuinely complementary probe of the local gravitational acceleration and tidal field, with systematics very different from rotation curves, stellar streams, and satellite dynamics. The paper is notable for its careful synthetic experiments: it uses Monte Carlo sampling to propagate three explicit observational error budgets, publishes documented reproduction code, and is unusually transparent about the limitations of the adopted potential model in Section 5.3. The demonstration that traceback ages are not independent of the assumed potential is a valuable and transferable result. However, the headline precision estimates are currently in-sample quantities, derived from the same parametric potential family and integrator used to generate the mock data, and the paper does not test the method against an independent, more realistic Galactic potential. This gap prevents the quantitative claims from being taken at face value for the real Milky Way.

major comments (3)

[§4.1, Fig. 5, §5.3] The quoted precisions for the halo mass and concentration (σ(M_h)=5.4×10^11 M⊙, σ(c_NFW)=0.83 with 'next' errors) are derived from mock associations generated by forward integration in the same parametric potential model (Table 3) and then fitted with the same parametric family and the same Gala integration code. This is a closed-box test: the data-generation and inference procedures share the same model family and the same numerical machinery. As the authors note in §5.3, the assumed static, axisymmetric, single-disk model is 'certainly not accurate' for the Milky Way, and constraints on the potential are dependent on the assumed parametrisation. The covariance-minimum criterion responds to the local tidal tensor along the association's orbit, so any unmodelled contributions from the bar, spiral arms, non-axisymmetric halo, or non-equilibrium disk motions will shift the best-fit parameters, potentially by more than the quoted statistical uncertainties regardless of per-star measurement precision. The authors should validate the method on mock data generated from an independent, more realistic potential model (e.g., including a bar or spiral arms or a multi-component disk), or with an independent integrator, to quantify the systematic bias. At minimum, the abstract and §4.1 must clearly state that the quoted precisions are in-sample and conditional on the assumed parametric family. This is the central load-bearing issue for the paper's main claim.
[§4.1, Fig. 5] The histograms in Figure 5 are labelled 'posterior distributions', but no likelihood is defined for the size-based inference. The text describes drawing 1000 samples from a uniform distribution over each potential parameter and fitting Gaussian curves to the 'raw data', which is more consistent with an ensemble of best-fit parameter values (the argmin of trmin across sampled potentials for each Monte Carlo realization of the data) than with a Bayesian posterior. The quoted 1σ widths therefore do not have the usual posterior-probability interpretation. The authors should either derive a proper likelihood for the covariance-minimum statistic (for example via the sampling distribution of trmin as a function of the potential parameters) or explicitly relabel these as 'distributions of inferred parameters' and explain that the σ values represent the scatter of the point estimate over data realizations. Because the abstract's precision numbers are drawn from these σ values, the statistical interpretation needs to be corrected.
[Abstract, §4.1] The abstract states that with Gaia DR4 astrometry and radial velocities below 0.2 km/s, 'the halo mass can be constrained with a precision of <0.6×10^12 M⊙ and the concentration with a precision of <0.8 using a single association'. In §4.1, however, these are explicitly described as idealized one-parameter constraints, obtained with all other potential parameters fixed, and the text notes that these values 'should be interpreted as idealised one-parameter constraints'. The abstract should carry this qualification as well; without it, the reader may reasonably interpret the numbers as joint constraints on the full potential, which the paper itself shows are not achievable even with 'next' errors due to the degeneracies described in §4.2. Adding a short phrase such as 'in the idealized case where the other potential parameters are known' would resolve the discrepancy.

minor comments (5)

[Abstract] The abstract in the arXiv version states a precision of '0.6 trillion solar masses' while the full-text abstract says '<0.6×10^12 M⊙'; the two versions should be made consistent, and the '<' symbol should be used consistently with the reported σ values in §4.1.
[§3.2 / Conclusion] The 'next' error budget in Table 2 sets σ_RV=0.2 km/s, but the abstract says 'below 0.2 km/s' and the conclusion says 'to the level of ~0.2 km/s'; please unify the threshold language.
[§2.1] The symbols 'tr' and 'det' are used both for the linearized size measures (tr=Tr^{1/2}, det=Det^{1/6}) and for the time-dependent functions in Figures 1 and 2; a distinct notation would avoid confusion.
[Table 4] The two blocks of Table 4 share identical default rows, and the note explains this, but the layout makes it easy to misread the 'Effects of the potential' rows as independent from the 'Intrinsic properties' rows; consider merging or adding a separate default row with a clear label.
[Figure 3] The panels of Figure 3 are not labelled (a)-(f) etc.; adding panel labels and referring to them individually in the text would significantly improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the synthetic recovery experiment forward-generates associations in a known potential and then infers that potential by backward integration, so the headline precision figures are in-sample forecasts rather than fitted inputs renamed as predictions.

full rationale

The paper's inference chain is not circular. In Section 3, mock associations are generated by integrating the association centre backward in the assumed true potential to the birth time, sampling member positions and velocities with specified dispersions, then integrating forward again in the same potential, and finally convolving with assumed observational errors. The backwards inference then uses the same observed phase-space coordinates in different trial potentials and compares the minima of trace and determinant of the position covariance matrix. The 'true potential' is the one used in the forward generation, so its recovery is a genuine test of the method rather than a definitional tautology. No fitted parameter is renamed as a prediction: the quoted uncertainties, such as sigma(M_h)=5.4e11 Msun with 'next' errors, are derived from Monte Carlo realisations and Gaussian fits to posterior distributions obtained by varying trial potentials while holding the true potential fixed; they are not calibrated to force agreement. Self-citations (Sawala et al. 2016, 2023a,b, 2025; Miret-Roig et al. 2018, 2020, 2022, 2024) are contextual and not load-bearing. The acknowledged caveats in Section 5.3, notably that the static axisymmetric single-disk parametrisation is 'certainly not accurate' and that constraints are 'dependent on the assumed parametrisation', are robustness/correctness concerns about applying the method to the real Milky Way, not circularity in the demonstrated synthetic inference. The paper itself is transparent that this is a proof-of-concept with synthetic associations, and it does not claim external validation beyond its own forward model.

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

All inputs to the synthetic experiments (potential parameters, error budgets, intrinsic association properties) are chosen by hand rather than fitted to data. The central proof-of-concept rests on domain assumptions about the Milky Way potential and association birth configurations, which the authors explicitly enumerate in Section 5.3.

free parameters (4)

Default halo mass M_h = 1.0 x 10^12 Msun
Assumed true value in the synthetic forward model, not fitted to observations.
Default halo concentration c_NFW = 11
Assumed true value in the synthetic forward model, not fitted to observations.
Default disk mass M_d = 5.2 x 10^10 Msun
Assumed true value in the synthetic forward model, not fitted to observations.
Future observational error budget = sigma_RV = 0.025 km/s, sigma_astrometry = 2.5 uas
Chosen input scenario; the projected precision numbers depend on this assumed budget.

assumptions (4)

domain assumption The Galactic potential is static and axisymmetric over the association's lifetime
Section 5.3 states this assumption; if bars, spiral arms, or non-equilibrium structures perturb the orbits, the reconstructed minima could be biased.
domain assumption The halo is spherical and the disk is a single smooth Miyamoto-Nagai disk
Section 5.3 acknowledges this is not accurate, especially for the disk, and limits interpretation to local acceleration and tidal tensor combinations.
domain assumption Stars in an association share a common, compact origin much smaller than the present extent
Section 2.2 relies on this constraint to anchor the orbital reconstruction and break degeneracies.
domain assumption Observational errors are independent and Gaussian
Section 3.2 adds independent Gaussian uncertainties to all observables; real errors may be correlated or non-Gaussian.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Expanding stellar associations as Galactic accelerometers." pith.science (2026). https://pith.science/paper/2JFJXWOM

@misc{pith2026260805297,
  author       = {Pith},
  title        = {Pith review of: Expanding stellar associations as Galactic accelerometers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2JFJXWOM}},
  note         = {Machine review of arXiv:2608.05297}
}

read the original abstract

The gravitational potential of the Milky Way is fundamental for understanding the evolution of our Galaxy and the nature of dark matter. We introduce a new method to constrain the Galactic potential using expanding young stellar associations. We exploit the physical constraint that these stars share a common, compact origin to reconstruct their most likely orbits and infer the gravitational potential in which they have evolved. We define the size of an association using the trace and determinant of its position covariance matrix. By integrating synthetic associations backward in different trial potentials, we show how the true potential can be identified as the one that minimises these metrics. We demonstrate that, while current observational errors are still too large, upcoming observations will allow us to distinguish between different potentials. Our results suggest that with Gaia DR4 astrometry and radial velocity errors below 0.2 km/s, the halo mass can be constrained with a precision of 0.6 trillion solar masses and the concentration with a precision of < 0.8 using a single association, albeit with significant degeneracies. In addition, we show that the inferred dynamical traceback age is sensitive to the gravitational potential, suggesting that independent age information can help break existing degeneracies, but also that traceback-age estimates are not independent of the assumed potential. Expanding stellar associations carry information about the gravitational potential in which they have evolved. With the arrival of next-generation astrometry and high-precision radial velocities, they will provide a complementary tool for constraining the Galactic potential.

Figures

Figures reproduced from arXiv: 2608.05297 by the authors.

**Figure 2.** Effect of intrinsic properties on an association’s reconstructed evolution in the true potential. The top row uses the determinant of the covariance matrix. The bottom row uses its trace. From left to right, we vary the number of stars, N⋆, the initial dispersions in position, σx, and in velocity, σv, and the age of the association. All other properties are set to the default values. Solid lines denote medians, band… view at source ↗

**Figure 3.** Effect of individual observational errors on an association’s reconstructed evolution in the true potential. The top two rows use the determinant of the covariance matrix, the bottom two rows use its trace. Solid lines denote medians, bands indicate ±1σ uncertainties. Three error scenarios are considered, namely, ‘current’ (black), ‘next’ (blue), and ‘future’ (pink), with all other errors set to zero. We also show, … view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: Sensitivity of the association’s minimum size to components of the Galactic potential. We show the median and 1 [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

**Figure 5.** Figure 5: Posterior distributions for potential components inferred from single associations, assuming uniform priors over the range [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

**Figure 6.** Figure 6: Degeneracy between pairs of potential variables, disk mass and halo mass (top two rows), and concentration and halo mass [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]

**Figure 7.** Figure 7: Sensitivity of the traceback age to individual components of the potential. On each panel, we show the median and [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]

**Figure 8.** Figure 8: Posterior distributions for potential components inferred from single associations, assuming uniform priors over the range [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 35 canonical work pages

[1]

Ablimit, I., Zhao, G., Flynn, C., & Bird, S. A. 2020, ApJ, 895, L12

work page 2020
[2]

Ambartsumian, V . A. 1947, Stellar Evolution and Astrophysics (Armenian Academy of Sciences)

work page 1947
[3]

2018, Nature, 561, 360

Antoja, T., Helmi, A., Romero-Gómez, M., et al. 2018, Nature, 561, 360

2018
[4]

2023, A&A, 673, A115

Antoja, T., Ramos, P., García-Conde, B., et al. 2023, A&A, 673, A115

2023
[5]

W., et al

Benisty, D., Vasiliev, E., Evans, N. W., et al. 2022, ApJ, 928, L5

work page 2022
[6]

& Price-Whelan, A

Bonaca, A. & Price-Whelan, A. M. 2025, New A Rev., 100, 101713

2025
[7]

2018, ApJ, 859, L13

Bose, S., Ginsburg, I., & Loeb, A. 2018, ApJ, 859, L13

work page 2018
[8]

K., & Kallivayalil, N

Bovy, J., Bahmanyar, A., Fritz, T. K., & Kallivayalil, N. 2016, ApJ, 833, 31

2016

Show all 62 references

[9]

M., Cautun, M., Deason, A

Callingham, T. M., Cautun, M., Deason, A. J., et al. 2019, MNRAS, 484, 5453

2019
[10]

J., et al

Cantat-Gaudin, T., Jordi, C., Wright, N. J., et al. 2019, A&A, 626, A17

2019
[11]

J., et al

Cautun, M., Benítez-Llambay, A., Deason, A. J., et al. 2020, MNRAS, 494, 4291

2020
[12]

2023, ApJ, 946, 6

Couture, D., Gagné, J., & Doyon, R. 2023, ApJ, 946, 6

2023
[13]

2026, ApJ, 997, 129

Couture, D., Gagné, J., & Doyon, R. 2026, ApJ, 997, 129

2026
[14]

C., et al

Dalton, G., Trager, S., Abrams, D. C., et al. 2014, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, V ol. 9147, Ground- based and Airborne Instrumentation for Astronomy V , ed. S. K. Ramsay, I. S. McLean, & H. Takami, 91470L de Jong, R. S., Ager...

2014
[15]

K., Di Cintio, A., Battaglia, G., Brook, C., & Taibi, S

Fritz, T. K., Di Cintio, A., Battaglia, G., Brook, C., & Taibi, S. 2020, MNRAS, 494, 5178 Gagné, J., Moranta, L., Faherty, J. K., et al. 2026, arXiv e-prints, arXiv:2602.15695 Gagné, J., Mamajek, E. E., Malo, L., et al. 2018, ApJ, 856, 23

2020
[16]

Galli, P. A. B., Miret-Roig, N., Bouy, H., Olivares, J., & Barrado, D. 2023, MN- RAS, 520, 6245

2023
[17]

Galli, P. A. B., Miret-Roig, N., Bouy, H., Olivares, J., & Barrado, D. 2024, A&A, 689, A11

2024
[18]

Gibbons, S. L. J., Belokurov, V ., & Evans, N. W. 2014, MNRAS, 445, 3788

2014
[19]

& Portegies Zwart, S

Gieles, M. & Portegies Zwart, S. F. 2011, MNRAS: Letters, 410, L6 Grosbøl, P. & Carraro, G. 2018, A&A, 619, A50

2011
[20]

2010, MNRAS, 404, 1111

Guo, Q., White, S., Li, C., & Boylan-Kolchin, M. 2010, MNRAS, 404, 1111

2010
[21]

R., Millman, K

Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357

2020
[22]

2024, The Open Journal of Astrophysics, 7, 74

Hayati, E., Behroozi, P., & Patel, E. 2024, The Open Journal of Astrophysics, 7, 74

2024
[23]

2021, Experimental Astronomy, 51, 783

Hobbs, D., Brown, A., Høg, E., et al. 2021, Experimental Astronomy, 51, 783

2021
[24]

2016, arXiv e-prints, arXiv:1609.07325

Hobbs, D., Høg, E., Mora, A., et al. 2016, arXiv e-prints, arXiv:1609.07325

2016 arXiv
[25]

Hunt, E. L. & Reffert, S. 2023, A&A, 673, A114

2023
[26]

Hunt, J. A. S. & Vasiliev, E. 2025, New A Rev., 100, 101721

2025
[27]

Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90

2007
[28]

2025, arXiv e-prints, arXiv:2501.14868

Hussein, A., Necib, L., Kaplinghat, M., et al. 2025, arXiv e-prints, arXiv:2501.14868

2025 arXiv
[29]

A., Lewis, G

Ibata, R. A., Lewis, G. F., Thomas, G., Martin, N. F., & Chapman, S. 2017, ApJ, 842, 120

2017
[30]

V ., Zhao, H., Spergel, D

Johnston, K. V ., Zhao, H., Spergel, D. N., & Hernquist, L. 1999, ApJ, 512, L109

1999
[31]

V ., Benito, M., Iocco, F., Trotta, R., & Geringer-Sameth, A

Karukes, E. V ., Benito, M., Iocco, F., Trotta, R., & Geringer-Sameth, A. 2020, J. Cosmology Astropart. Phys., 2020, 033

2020
[32]

2014, MNRAS, 442, 2487

Kennedy, R., Frenk, C., Cole, S., & Benson, A. 2014, MNRAS, 442, 2487

2014
[33]

L., Murphy, S

Kerr, R., Kraus, A. L., Murphy, S. J., et al. 2022, ApJ, 941, 143

2022
[34]

F., Blanc, G

Kollmeier, J., Anderson, S. F., Blanc, G. A., et al. 2019, in Bulletin of the Amer- ican Astronomical Society, V ol. 51, 274

2019
[35]

Leane, R. K. 2020, arXiv e-prints, arXiv:2006.00513

2020 arXiv
[36]

2020, ApJ, 894, 10

Li, Z.-Z., Qian, Y .-Z., Han, J., et al. 2020, ApJ, 894, 10

2020
[37]

I., Brinchmann, J., et al

Mainieri, V ., Anderson, R. I., Brinchmann, J., et al. 2024, arXiv e-prints, arXiv:2403.05398

2024 arXiv
[38]

2023, MNRAS, 520, 5225

Mateu, C. 2023, MNRAS, 520, 5225

2023
[39]

2024, Nature Astronomy, 8, 216

Miret-Roig, N., Alves, J., Barrado, D., et al. 2024, Nature Astronomy, 8, 216

2024
[40]

2018, A&A, 615, A51

Miret-Roig, N., Antoja, T., Romero-Gómez, M., & Figueras, F. 2018, A&A, 615, A51

2018
[41]

Miret-Roig, N., Galli, P. A. B., Brandner, W., et al. 2020, A&A, 642, A179

2020
[42]

Miret-Roig, N., Galli, P. A. B., Olivares, J., et al. 2022, A&A, 667, A163

2022
[43]

& Nagai, R

Miyamoto, M. & Nagai, R. 1975, PASJ, 27, 533

1975
[44]

F., Frenk, C

Navarro, J. F., Frenk, C. S., & White, S. D. M. 1996, ApJ, 462, 563

1996
[45]

& Bonaca, A

Nibauer, J. & Bonaca, A. 2025, ApJ, 985, L22

2025
[46]

Olivares, J., Miret-Roig, N., Galli, P. A. B., & Bouy, H. 2025, A&A, 699, A122

2025
[47]

G., Wang, W., & Han, J

Palau, C. G., Wang, W., & Han, J. 2025, MNRAS, 539, 2718

2025
[48]

& Helmi, A

Posti, L. & Helmi, A. 2019, A&A, 621, A56

2019
[49]

Price-Whelan, A. M. 2017, The Journal of Open Source Software, 2, 388

2017
[50]

M., Hogg, D

Price-Whelan, A. M., Hogg, D. W., Johnston, K. V ., & Hendel, D. 2014, ApJ, 794, 4

2014
[51]

J., Lemasle, B., et al

Prudil, Z., Koch-Hansen, A. J., Lemasle, B., et al. 2022, A&A, 664, A148 Rodriguez Wimberly, M. K., Cooper, M. C., Baxter, D. C., et al. 2022, MNRAS, 513, 4968

2022
[52]

J., et al

Sawala, T., Delhomelle, J., Deason, A. J., et al. 2025, Nature Astronomy, 9, 1206

2025
[53]

S., Fattahi, A., et al

Sawala, T., Frenk, C. S., Fattahi, A., et al. 2016, MNRAS, 457, 1931

2016
[54]

Sellwood, J. A. 2014, Reviews of Modern Physics, 86, 1

2014
[55]

Sellwood, J. A. & Masters, K. L. 2022, ARA&A, 60, 73

2022
[56]

M., Murray, N., et al

Shen, J., Eadie, G. M., Murray, N., et al. 2022, ApJ, 925, 1

2022
[57]

Song, I., Zuckerman, B., & Bessell, M. S. 2003, ApJ, 599, 342

2003
[58]

2024, Nature, 631, 49 The MSE Science Team, Babusiaux, C., Bergemann, M., et al

Swiggum, C., Alves, J., Benjamin, R., et al. 2024, Nature, 631, 49 The MSE Science Team, Babusiaux, C., Bergemann, M., et al. 2019, arXiv e- prints, arXiv:1904.04907

2024 arXiv
[59]

2022, A&A, 659, A95

Tsantaki, M., Pancino, E., Marrese, P., et al. 2022, A&A, 659, A95

2022
[60]

E., et al

Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods, 17, 261

2020
[61]

Wang, W., Han, J., Cautun, M., Li, Z., & Ishigaki, M. N. 2020, Science China

2020
[62]

L., van der Marel, R

Watkins, L. L., van der Marel, R. P., Sohn, S. T., & Evans, N. W. 2019, ApJ, 873, 118 Page 13

2019

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [1]

Ablimit, I., Zhao, G., Flynn, C., & Bird, S. A. 2020, ApJ, 895, L12

work page 2020

[2] [2]

Ambartsumian, V . A. 1947, Stellar Evolution and Astrophysics (Armenian Academy of Sciences)

work page 1947

[3] [3]

2018, Nature, 561, 360

Antoja, T., Helmi, A., Romero-Gómez, M., et al. 2018, Nature, 561, 360

2018

[4] [4]

2023, A&A, 673, A115

Antoja, T., Ramos, P., García-Conde, B., et al. 2023, A&A, 673, A115

2023

[5] [5]

W., et al

Benisty, D., Vasiliev, E., Evans, N. W., et al. 2022, ApJ, 928, L5

work page 2022

[6] [6]

& Price-Whelan, A

Bonaca, A. & Price-Whelan, A. M. 2025, New A Rev., 100, 101713

2025

[7] [7]

2018, ApJ, 859, L13

Bose, S., Ginsburg, I., & Loeb, A. 2018, ApJ, 859, L13

work page 2018

[8] [8]

K., & Kallivayalil, N

Bovy, J., Bahmanyar, A., Fritz, T. K., & Kallivayalil, N. 2016, ApJ, 833, 31

2016