Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Hydrostatic and explosive $\alpha$-element chemical abundances of Milky Way globular clusters, halo substructures, and satellite galaxies

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read At fixed [Fe/H], Milky Way globular clusters and halo substructures have a hex ratio about 0.1 dex higher than dwarf satellite galaxies, implying the satellites formed with a more top-light initial mass function.

desk verdict A clean observational comparison with a suggestive ~0.1 dex hex-ratio offset, but the IMF interpretation needs chemical-evolution checking before it can be trusted. read the letter →

arxiv 2506.08079 v3 pith:TRH5OKWN submitted 2025-06-09 astro-ph.GA

Danny Horta , Melissa K. Ness This is my paper
classification astro-ph.GA
keywords alphaelementshexratioinitialmassfunctionglobularclustershalosubstructuresdwarfsatellitegalaxiesAPOGEEgalacticchemicalevolution
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

Stars forge two families of $\alpha$ elements in different places: oxygen and magnesium are made in the cores of the most massive stars, while silicon, calcium, and titanium are made in all core-collapse supernova explosions. The ratio of these two families, the hex ratio, therefore measures how many very massive stars contributed to a stellar population relative to the total number of supernovae. Using APOGEE spectra of red giants, this letter compares the hex ratio across Milky Way globular clusters, halo substructures, satellite galaxies, and the high- and low-$\alpha$ disc. It finds that globular clusters and halo substructures sit at a higher hex ratio than dwarf satellite galaxies of the same $[\mathrm{Fe/H}]$ — an offset of roughly 0.1 dex — and concludes that the satellites formed under a more top-light initial mass function, with fewer of the most massive stars. If this interpretation holds, the local universe preserves a record of IMF variation with star formation conditions, and the hex ratio becomes a practical way to read the high-mass end of the IMF from resolved stellar populations.

What carries the argument

The load-bearing object is the hex ratio, the difference between a star's hydrostatic and explosive $\alpha$ abundances: $\mathrm{hex} = ([\mathrm{Mg/Fe}]+[\mathrm{O/Fe}])/2 - ([\mathrm{Si/Fe}]+[\mathrm{Ca/Fe}]+[\mathrm{Ti/Fe}])/3$. Hydrostatic alphas (O, Mg) are taken to be produced only in the cores of the most massive stars, $M_\star \approx 15$–$30\,M_\odot$, while explosive alphas (Si, Ca, Ti) come from all core-collapse supernovae with $M_\star > 8\,M_\odot$, so the ratio counts the most massive-star contribution relative to the total core-collapse supernova count. The paper compares median hex ratios at fixed $[\mathrm{Fe/H}]$, uses a $K$-means split of globular cluster stars into first and second populations to remove light-element contamination, and reads the $[\mathrm{Fe/H}]$ drift of the ratio, attributed to Type Ia supernova enrichment, as the background against which the IMF offset must be seen.

What would settle it

Run a chemical evolution model with a fixed, invariant initial mass function and published nucleosynthetic yields through the same APOGEE abundances; if it reproduces the observed $\sim$0.1 dex lower hex ratio of dwarf satellites at fixed $[\mathrm{Fe/H}]$ without any change in the high-mass IMF, the IMF interpretation is falsified. A simpler observational check is whether the offset survives when the hydrostatic pair (O, Mg) and the explosive triplet (Si, Ca, Ti) are examined as separate single-element ratios such as $[\mathrm{O/Mg}]$ and $[\mathrm{Si/Ti}]$.

Watch

Extended reading notes

Core claim

The central claim is that, at fixed iron abundance, the hydrostatic-to-explosive $\alpha$-element ratio of first-population Milky Way globular clusters and of halo substructures is about 0.1 dex higher than that of dwarf satellite galaxies, and that this offset traces the high-mass endpoint of the initial mass function. The paper establishes the offset with median hex ratios, $\mathrm{hex} = ([\mathrm{Mg/Fe}]+[\mathrm{O/Fe}])/2 - ([\mathrm{Si/Fe}]+[\mathrm{Ca/Fe}]+[\mathrm{Ti/Fe}])/3$, for globular clusters, halo substructures, and nine satellite galaxies across $[\mathrm{Fe/H}]\approx -2.5$ to 0. The hydrostatic $\alpha$ abundance is about 0.3 dex higher in clusters and halo debris than in satellites, while the explosive average is also higher, leaving a net hex offset near 0.1 dex. The authors also report that the hex ratio decreases with increasing $[\mathrm{Fe/H}]$ in every population, with best-fit slopes near $-0.05$ and different intercepts for the cluster/halo relation versus the satellites, and they tentatively attribute that decline to delayed Type Ia supernova enrichment. They further report a weak increase of the hex ratio with globular cluster age, no dependence on globular cluster mass, and no difference between in situ and accreted clusters.

Load-bearing premise

The entire argument depends on the assumption that, at the same $[\mathrm{Fe/H}]$, a lower ratio of hydrostatic to explosive $\alpha$ elements means fewer very massive stars formed, rather than differences in Type Ia supernova enrichment, metallicity-dependent element yields, or abundance-measurement systematics; the paper does not test this assumption with a chemical evolution model.

Editorial extensions

If this is right

  • If the hex-ratio offset is an IMF signature, the Milky Way's dwarf satellite galaxies formed with fewer stars above roughly 15–30 solar masses per core-collapse supernova than did the progenitors of globular clusters and halo substructures.
  • Because the hex ratio declines with $[\mathrm{Fe/H}]$ in every population, IMF comparisons between systems must be made at matched iron abundance to avoid mistaking Type Ia enrichment for an IMF difference.
  • Globular cluster mass does not set the high-mass IMF endpoint: the hex ratio shows no correlation with initial or present-day cluster mass.
  • Accretion origin leaves no separate imprint in the hydrostatic/explosive alpha pattern of globular clusters; in situ and accreted clusters follow the same hex-to-$[\mathrm{Fe/H}]$ relation.
  • The weak positive hex-age trend among globular clusters, if real, implies the high-mass IMF endpoint drifted slightly toward a less top-heavy form over the epoch of cluster formation.

Reading between the lines

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

  • The paper's own logic implies a direct test it did not run: if the offset is truly the high-mass IMF, systems with independently constrained star formation histories, bursty dwarfs versus quiescent ones, should sort by hex ratio at fixed $[\mathrm{Fe/H}]$ according to their burstiness rather than their present-day mass.
  • I infer the same measurement could be extended to other Local Group dwarfs as a cheap probe of the high-mass IMF endpoint, provided the hydrostatic/explosive yield mapping holds.
  • Because the paper does not run a chemical evolution model, an alternative reading remains open: metallicity-dependent yields or an incomplete Type Ia correction could generate the 0.1 dex offset with no IMF change; comparing single-element ratios such as $[\mathrm{O/Mg}]$ and $[\mathrm{Si/Ti}]$ would separate those channels.
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.

Referee Report

3 major / 4 minor

Summary. This letter uses APOGEE DR17 data to compare the ratio of hydrostatic to explosive alpha-element abundances (the 'hex' ratio) among Milky Way globular clusters (GCs), halo substructures, dwarf satellite galaxies, and the high/low-alpha discs. The authors report that GCs and halo substructures have hex ratios about 0.1 dex higher than dwarf satellites at fixed [Fe/H], a decreasing hex ratio with [Fe/H] for all populations, a weak positive trend with GC age, and no dependence on GC mass or in situ/accreted origin. They interpret the satellite/GC offset as evidence for a more top-light IMF in the satellite galaxies.

Significance. If established, the result would provide an interesting observational constraint on IMF variations across Galactic environments, using a large homogeneous sample and explicit treatment of GC multiple populations. The strengths are the use of public APOGEE data, the separation of first-population GC stars, and the clearly presented figures. However, the central inference currently rests on an unquantified degeneracy between IMF variations and Type Ia SN enrichment histories, and the key offsets are presented without significance tests. The paper is a useful phenomenological contribution, but the headline conclusion is not yet established.

major comments (3)
  1. [§4 and Fig. 2] The interpretation of the ~0.1 dex hex-ratio offset between dwarf satellites and GCs/halo substructures as evidence for a top-light IMF is not uniquely supported. As the authors themselves argue, the decreasing hex ratio with [Fe/H] is attributed to delayed Type Ia SN contributions to the explosive alpha-elements (Section 4; Johnson et al. 2020). Dwarf galaxies are precisely the systems with the most extended star formation histories, so at fixed [Fe/H] they may have a larger cumulative SNIa-to-core-collapse ratio, which would lower the hex ratio without any change in the IMF. No chemical evolution model or yield convolution is presented to show that a ~0.1 dex offset requires a variation of the high-mass IMF endpoint rather than a difference in enrichment history. This is the central claim and must be tested quantitatively.
  2. [§3 (after Fig. 2)] The key quantitative results are reported without uncertainties or significance tests: the ≈0.1 dex offset, the slopes m=−0.047 and m_sat=−0.048, and the intercepts 0.089 and −0.050 are all quoted as single numbers. The figure caption states only that 'the error bars show the average standard error for each stellar population type,' but no confidence interval is given for the offset at fixed [Fe/H]. Given the significant scatter visible in Fig. A1, the claim that GCs and halo substructures are 'higher' than dwarf galaxies requires a stated significance level.
  3. [§1 and §3] The nucleosynthetic mapping that underpins the IMF inference—hydrostatic alpha (O, Mg) produced only in 15–30 Msun cores, explosive alpha (Si, Ca, Ti) from all >8 Msun core-collapse SNe—is adopted without sensitivity testing. Because the paper acknowledges that some explosive alpha also comes from Type Ia SNe (footnote 1), and because yields of O, Mg, Si, Ca, Ti are metallicity- and mass-dependent, the hex ratio may respond to enrichment channels other than the high-mass IMF endpoint. The authors should demonstrate robustness, for example by recomputing the hex ratio using subsets of elements (e.g., [O/Mg] versus [Si/Ca]) or by testing the effect of changing the treatment of Ti.
minor comments (4)
  1. [§2] The low-alpha disc selection criterion includes the interval '(−0.05<[Fe/H] <−0.6∧[Mg/Fe] < 0.12)'. This interval is empty as written because it requires [Fe/H] > −0.05 and [Fe/H] < −0.6 simultaneously; it likely intends the opposite inequality (e.g., −0.6 < [Fe/H] < −0.05). Please correct, as it affects the disc sample shown in Fig. 2.
  2. [§3.1] The quoted slope and intercept for the hex–age relation ('≈0.012 dex with increasing age and an intercept of ≈0.016 dex') lack units and uncertainties; presumably the slope is in dex/Gyr. Please clarify.
  3. [§3] The notation 'hexhigh−α≈ 0.15' and 'hexlow−α≈ 0.05' is ambiguous: it should state explicitly that these are median hex ratios, and the subscripts should be formatted clearly.
  4. [Appendix A] In Fig. A1, the legend entries for some substructures (e.g., 'Sequoia(Koppelman)', 'Sequoia(Myeong)', 'Sequoia(Naidu)') suggest multiple selection criteria, but the text in §2 mentions only that Sequoia was identified using three different criteria without specifying them. Please provide a reference or a sentence describing these criteria.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the hex-ratio comparison is a direct abundance measurement, and the IMF inference is a stated physical interpretation rather than a fitted or self-citational construct.

full rationale

The central observable, the hex ratio, is computed directly from APOGEE abundances as ([Mg/Fe]+[O/Fe])/2 - ([Si/Fe]+[Ca/Fe]+[Ti/Fe])/3; it is not a fitted parameter and is not constructed from the conclusions it is used to support. The dwarf-versus-GC offset is an empirical difference in these measured ratios, and the top-light IMF interpretation is stated by the authors as an inference from the standard nucleosynthetic assumption that hydrostatic alpha elements come from the most massive core-collapse supernovae, not as a quantity fitted to the same data. The paper explicitly flags the need for per-system star formation history characterisation to test the IMF hypothesis, which is the opposite of presenting the conclusion as forced by the construction of the hex ratio. The main self-references are to the authors' earlier halo substructure classifications (Horta et al. 2021, 2023) and the GES-debris hypothesis (Horta et al. 2023); these provide sample definitions and interpretive consistency but are not used as a uniqueness theorem or as a substitute for the abundance measurements, so they do not make the central claim circular. The high-/low-alpha disc classification does use [Mg/Fe] cuts, and [Mg/Fe] enters the hex ratio, so the disc hex difference is partly a consequence of the sample definition; however, this is an ancillary observation, explicitly disclosed, and it is not the basis of the paper's central GC/halo-versus-dwarf conclusion. No load-bearing step reduces by construction to its inputs.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new physical entities. The hex ratio is an observational quantity borrowed from prior literature. The descriptive best-fit lines are parameters fitted to the data, not independent inputs.

free parameters (3)
  • Best-fit slope and intercept for GC/halo/substructure hex ratio vs [Fe/H] = m = -0.047, b = 0.089
    Linear fit to median hex ratios across the combined GC/halo/substructure sample; reported without uncertainties.
  • Best-fit slope and intercept for satellite hex ratio vs [Fe/H] = m_sat = -0.048, b_sat = -0.050
    Linear fit to the dwarf satellite sample; reported without uncertainties.
  • Best-fit slope and intercept for GC hex ratio vs age = m_age = 0.012, b_age = 0.016
    Linear fit to median GC hex ratio against age from Forbes & Bridges (2010); scatter is noted as significant.
assumptions (3)
  • domain assumption Hydrostatic alpha elements (O, Mg) are produced only in the cores of the most massive stars (about 15-30 solar masses), while explosive alpha elements (Si, Ca, Ti) come from all core-collapse supernovae above about 8 solar masses.
    This nucleosynthetic mapping is the basis for interpreting the hex ratio as an IMF probe; it is stated in Section 1 and not tested here.
  • domain assumption APOGEE DR17 ASPCAP abundance ratios are on a common, unbiased scale across all stellar populations studied.
    The comparisons assume no population-dependent systematics in [O/Fe], [Mg/Fe], [Si/Fe], [Ca/Fe], and [Ti/Fe]; no cross-calibration or NLTE corrections are applied (Section 2).
  • domain assumption The K-means clustering on [Mg/Fe], [Al/Fe], [C/Fe], [N/Fe] reliably separates first-population from second-population globular cluster stars.
    Residual second-population stars with depleted Mg would lower the hydrostatic alpha and hex ratio measurements; the classification is described in Section 2.1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hydrostatic and explosive $\alpha$-element chemical abundances of Milky Way globular clusters, halo substructures, and satellite galaxies." pith.science (2026). https://pith.science/paper/TRH5OKWN

@misc{pith2026250608079,
  author       = {Pith},
  title        = {Pith review of: Hydrostatic and explosive $\alpha$-element chemical abundances of Milky Way globular clusters, halo substructures, and satellite galaxies},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TRH5OKWN}},
  note         = {Machine review of arXiv:2506.08079}
}
abstract

Stellar atmospheric element abundance ratios of stars retain information about their birth conditions, helping elucidate their origin and nature. In this letter, we analyse and contrast the hydrostatic and explosive $\alpha$-element abundance ratios, and the ratio of the two (the hex ratio), for a large sample of Galactic globular clusters (GCs), halo substructures, satellite galaxies, and the Milky Way high-/low-$\alpha$ discs using data from the $APOGEE$ survey. Our results show that: $i$) Milky Way GCs and halo substructures appear to have qualitatively similar hex ratios across a broad range of [Fe/H], that are higher than that of dwarf satellite galaxies of similar [Fe/H]; $ii$) for all stellar populations studied, there is a trend in the hex ratio with [Fe/H]; $iii$) there is a weak trend in the hex ratio with respect to age for Galactic GCs, but not with initial or final GC mass; $iv$) there are no differences in the hex ratio between GCs formed $in$ $situ$ versus those labelled as accreted.

Figures

Figures reproduced from arXiv: 2506.08079 by the authors.

Figure 1
Figure 1. First-/Second-Population classification results from our 𝐾-means clustering procedure for stars in the GC value-added-catalogue (Schiavon et al. 2024) in the [Mg/Fe]-[Al/Fe] plane ([C/Fe]-[N/Fe] plane not shown). First population GC stars (red circles) do not suffer from light-element variations, and are thus used for studying the 𝛼-element abundance ratios of Galactic GCs. element abundance ratio measurements for r… view at source ↗
Figure 2
Figure 2. Median values of the hydrostatic 𝛼-elements (namely, ([Mg/Fe]+[O/Fe])/2) on the left, the explosive 𝛼-elements (namely, ([Si/Fe]+[Ca/Fe]+[Ti/Fe])/3) in the middle, and the hex ratio ([𝛼hyd/Fe] − [𝛼exp/Fe]) on the right as a function of median [Fe/H] for (first population) globular cluster stars (red circles), halo substructures (purple squares), and Milky Way satellite galaxies (yellow upside-down triangles); the ru… view at source ↗
Figure 3
Figure 3. shows the median hex ratio now as a function of age for the Galactic GC sample. The ages are taken from the compilation of Forbes & Bridges (2010). Our results show that there appears to be a weak decreasing trend of hex with decreasing age2 . However, there is a significant scatter around this relation. We also investigated the relation between the hex ratio and GC mass —both present day and initial mass— using the… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4 [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Revealing the Origins of Galactic Globular Clusters via Their Mg-Al Abundances

    astro-ph.GA 2025-08 conditional novelty 6.0 of 10

    The authors take the lowest-aluminum third of stars in each globular cluster and find that metal-rich clusters separate into in-situ and accreted groups, reclassifying NGC 288, M4, and Terzan 9.

Reference graph

Works this paper leans on

42 extracted references · 2 canonical work pages · cited by 1 Pith paper

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.state := if if FUNCTION not #0 #1 if FUNCTION and 'skip pop #0 if FUNCTION or pop #1...

  2. [2]

    Abdurro'uf et al., 2022, @doi [ ] 10.3847/1538-4365/ac4414 , https://ui.adsabs.harvard.edu/abs/2022ApJS..259...35A 259, 35

    doi:10.3847/1538-4365/ac4414 2022

  3. [3]

    C., Wilhelm R., Newberg H

    Allende Prieto C., Beers T. C., Wilhelm R., Newberg H. J., Rockosi C. M., Yanny B., Lee Y. S., 2006, @doi [ ] 10.1086/498131 , https://ui.adsabs.harvard.edu/abs/2006ApJ...636..804A 636, 804

    doi:10.1086/498131 2006

  4. [4]

    Bastian N., Lardo C., 2018, @doi [ ] 10.1146/annurev-astro-081817-051839 , https://ui.adsabs.harvard.edu/abs/2018ARA&A..56...83B 56, 83

    doi:10.1146/annurev-astro-081817-051839 2018

  5. [5]

    Baumgardt H., H \'e nault-Brunet V., Dickson N., Sollima A., 2023, @doi [ ] 10.1093/mnras/stad631 , https://ui.adsabs.harvard.edu/abs/2023MNRAS.521.3991B 521, 3991

    doi:10.1093/mnras/stad631 2023

  6. [6]

    L., et al., 2021, arXiv e-prints, https://ui.adsabs.harvard.edu/abs/2021arXiv210811907B p

    Beaton R. L., et al., 2021, arXiv e-prints, https://ui.adsabs.harvard.edu/abs/2021arXiv210811907B p. arXiv:2108.11907

    arXiv 2021

  7. [7]

    V., Rybizki J., Bedell M., 2019, @doi [ ] 10.3847/1538-4357/ab39e5 , https://ui.adsabs.harvard.edu/abs/2019ApJ...883...34B 883, 34

    Blancato K., Ness M., Johnston K. V., Rybizki J., Bedell M., 2019, @doi [ ] 10.3847/1538-4357/ab39e5 , https://ui.adsabs.harvard.edu/abs/2019ApJ...883...34B 883, 34

    doi:10.3847/1538-4357/ab39e5 2019

  8. [8]

    S., Vaughan A

    Bowen I. S., Vaughan A. H. J., 1973, @doi [ ] 10.1364/AO.12.001430 , https://ui.adsabs.harvard.edu/abs/1973ApOpt..12.1430B 12, 1430

    doi:10.1364/ao.12.001430 1973

Show all 42 references
  1. [9]

    L., Sheffield A

    Carlin J. L., Sheffield A. A., Cunha K., Smith V. V., 2018, @doi [ ] 10.3847/2041-8213/aac3d8 , https://ui.adsabs.harvard.edu/abs/2018ApJ...859L..10C 859, L10

    2018 doi

  2. [10]

    Cunha K., et al., 2017, @doi [ ] 10.3847/1538-4357/aa7beb , https://ui.adsabs.harvard.edu/abs/2017ApJ...844..145C 844, 145

    2017 doi

  3. [11]

    Fernandes L., et al., 2023, @doi [ ] 10.1093/mnras/stac3543 , https://ui.adsabs.harvard.edu/abs/2023MNRAS.519.3611F 519, 3611

    2023 doi

  4. [12]

    Filion C., Platais I., Wyse R. F. G., Kozhurina-Platais V., 2022, @doi [The Astrophysical Journal] 10.3847/1538-4357/ac9383 , 939, 38

    2022 doi

  5. [13]

    A., Bridges T., 2010, @doi [Monthly Notices of the Royal Astronomical Society] 10.1111/j.1365-2966.2010.16373.x

    Forbes D. A., Bridges T., 2010, @doi [Monthly Notices of the Royal Astronomical Society] 10.1111/j.1365-2966.2010.16373.x

    2010

  6. [14]

    Freeman K., Bland-Hawthorn J., 2002, @doi [ ] 10.1146/annurev.astro.40.060401.093840 , https://ui.adsabs.harvard.edu/abs/2002ARA&A..40..487F 40, 487

    2002

  7. [15]

    E., et al., 2016, @doi [ ] 10.3847/0004-6256/151/6/144 , https://ui.adsabs.harvard.edu/abs/2016AJ....151..144G 151, 144

    Garc \' a P \'e rez A. E., et al., 2016, @doi [ ] 10.3847/0004-6256/151/6/144 , https://ui.adsabs.harvard.edu/abs/2016AJ....151..144G 151, 144

    2016 doi

  8. [16]

    Geisler D., et al., 2021, @doi [ ] 10.1051/0004-6361/202140436 , https://ui.adsabs.harvard.edu/abs/2021A&A...652A.157G 652, A157

    2021 doi

  9. [17]

    E., et al., 2006, @doi [ ] 10.1086/500975 , https://ui.adsabs.harvard.edu/abs/2006AJ....131.2332G 131, 2332

    Gunn J. E., et al., 2006, @doi [ ] 10.1086/500975 , https://ui.adsabs.harvard.edu/abs/2006AJ....131.2332G 131, 2332

    2006 doi

  10. [18]

    R., et al., 2020, @doi [Nature] 10.1038/s41586-020-2649-2 , 585, 357

    Harris C. R., et al., 2020, @doi [Nature] 10.1038/s41586-020-2649-2 , 585, 357

    2020 doi

  11. [19]

    Hasselquist S., et al., 2021, @doi [ ] 10.3847/1538-4357/ac25f9 , https://ui.adsabs.harvard.edu/abs/2021ApJ...923..172H 923, 172

    2021 doi

  12. [20]

    E., Kirby E

    Henderson L. E., Kirby E. N., de los Reyes M. A. C., Gerasimov R., Manwadkar V., 2025, @doi [ ] 10.3847/1538-4357/adbe7d , https://ui.adsabs.harvard.edu/abs/2025ApJ...983..117H 983, 117

    2025 doi

  13. [21]

    Horta D., et al., 2020, @doi [ ] 10.1093/mnras/staa478 , https://ui.adsabs.harvard.edu/abs/2020MNRAS.493.3363H 493, 3363

    2020 doi

  14. [22]

    Horta D., et al., 2021, @doi [ ] 10.1093/mnras/staa2987 , https://ui.adsabs.harvard.edu/abs/2021MNRAS.500.1385H 500, 1385

    2021 doi

  15. [23]

    K., Rybizki J., Schiavon R

    Horta D., Ness M. K., Rybizki J., Schiavon R. P., Buder S., 2022, @doi [ ] 10.1093/mnras/stac953 , https://ui.adsabs.harvard.edu/abs/2022MNRAS.513.5477H 513, 5477

    2022 doi

  16. [24]

    Horta D., et al., 2023, @doi [ ] 10.1093/mnras/stac3179 , https://ui.adsabs.harvard.edu/abs/2023MNRAS.520.5671H 520, 5671

    2023 doi

  17. [25]

    D., 2007, @doi [Computing in Science & Engineering] 10.1109/MCSE.2007.55 , 9, 90

    Hunter J. D., 2007, @doi [Computing in Science & Engineering] 10.1109/MCSE.2007.55 , 9, 90

    2007 doi

  18. [26]

    A., Fields B

    Johnson J. A., Fields B. D., Thompson T. A., 2020, @doi [Phil Trans R. Soc.] 10.1098/rsta.2019.0301 , 533

    2020

  19. [27]

    R., et al., 2017, @doi [ ] 10.3847/1538-3881/aa784d , https://ui.adsabs.harvard.edu/abs/2017AJ....154...94M 154, 94

    Majewski S. R., et al., 2017, @doi [ ] 10.3847/1538-3881/aa784d , https://ui.adsabs.harvard.edu/abs/2017AJ....154...94M 154, 94

    2017 doi

  20. [28]

    H., Helmi A., 2019, @doi [Astronomy &amp; Astrophysics] 10.1051/0004-6361/201936135 , 630, L4

    Massari D., Koppelman H. H., Helmi A., 2019, @doi [Astronomy &amp; Astrophysics] 10.1051/0004-6361/201936135 , 630, L4

    2019 doi

  21. [29]

    McWilliam A., Wallerstein G., Mottini M., 2013, @doi [ ] 10.1088/0004-637X/778/2/149 , https://ui.adsabs.harvard.edu/abs/2013ApJ...778..149M 778, 149

    2013 doi

  22. [30]

    J., Horta D., 2024, @doi [ ] 10.3847/1538-4357/ad6df6 , https://ui.adsabs.harvard.edu/abs/2024ApJ...974..186M 974, 186

    Mead J., Ness M., Andersson E., Griffith E. J., Horta D., 2024, @doi [ ] 10.3847/1538-4357/ad6df6 , https://ui.adsabs.harvard.edu/abs/2024ApJ...974..186M 974, 186

    2024 doi

  23. [31]

    M \'e sz \'a ros S., et al., 2020, @doi [ ] 10.1093/mnras/stz3496 , https://ui.adsabs.harvard.edu/abs/2020MNRAS.492.1641M 492, 1641

    2020 doi

  24. [32]

    Monty S., et al., 2024, @doi [ ] 10.1093/mnras/stae1895 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.533.2420M 533, 2420

    2024 doi

  25. [33]

    L., et al., 2015, @doi [ ] 10.1088/0004-6256/150/6/173 , https://ui.adsabs.harvard.edu/abs/2015AJ....150..173N 150, 173

    Nidever D. L., et al., 2015, @doi [ ] 10.1088/0004-6256/150/6/173 , https://ui.adsabs.harvard.edu/abs/2015AJ....150..173N 150, 173

    2015 doi

  26. [34]

    Pedregosa F., et al., 2011, Journal of Machine Learning Research, 12, 2825

    2011

  27. [35]

    A., et al., 2021, arXiv e-prints, https://ui.adsabs.harvard.edu/abs/2021arXiv210811908S p

    Santana F. A., et al., 2021, arXiv e-prints, https://ui.adsabs.harvard.edu/abs/2021arXiv210811908S p. arXiv:2108.11908

    2021

  28. [36]

    P., et al., 2017, @doi [ ] 10.1093/mnras/stw3093 , https://ui.adsabs.harvard.edu/abs/2017MNRAS.466.1010S 466, 1010

    Schiavon R. P., et al., 2017, @doi [ ] 10.1093/mnras/stw3093 , https://ui.adsabs.harvard.edu/abs/2017MNRAS.466.1010S 466, 1010

    2017 doi

  29. [37]

    P., et al., 2024, @doi [ ] 10.1093/mnras/stad3020 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.528.1393S 528, 1393

    Schiavon R. P., et al., 2024, @doi [ ] 10.1093/mnras/stad3020 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.528.1393S 528, 1393

    2024 doi

  30. [38]

    V., et al., 2021, @doi [ ] 10.3847/1538-3881/abefdc , https://ui.adsabs.harvard.edu/abs/2021AJ....161..254S 161, 254

    Smith V. V., et al., 2021, @doi [ ] 10.3847/1538-3881/abefdc , https://ui.adsabs.harvard.edu/abs/2021AJ....161..254S 161, 254

    2021 doi

  31. [39]

    M., 1979, @doi [ ] 10.1086/157039 , https://ui.adsabs.harvard.edu/abs/1979ApJ...229.1046T 229, 1046

    Tinsley B. M., 1979, @doi [ ] 10.1086/157039 , https://ui.adsabs.harvard.edu/abs/1979ApJ...229.1046T 229, 1046

    1979 doi

  32. [40]

    C., et al., 2019, @doi [ ] 10.1088/1538-3873/ab0075 , https://ui.adsabs.harvard.edu/abs/2019PASP..131e5001W 131, 055001

    Wilson J. C., et al., 2019, @doi [ ] 10.1088/1538-3873/ab0075 , https://ui.adsabs.harvard.edu/abs/2019PASP..131e5001W 131, 055001

    2019 doi

  33. [41]

    Zasowski G., et al., 2013, @doi [ ] 10.1088/0004-6256/146/4/81 , https://ui.adsabs.harvard.edu/abs/2013AJ....146...81Z 146, 81

    2013 doi

  34. [42]

    Zasowski G., et al., 2017, @doi [ ] 10.3847/1538-3881/aa8df9 , https://ui.adsabs.harvard.edu/abs/2017AJ....154..198Z 154, 198

    2017 doi

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.