pith. sign in

arxiv: 2606.29707 · v1 · pith:4EJXBDSNnew · submitted 2026-06-29 · 🌌 astro-ph.GA

Globular cluster formation with multiple stellar populations: A comprehensive overview of a star-cloud interaction scenario

Kenji Bekki , Madeleine McKenzie This is my paper

Pith reviewed 2026-06-30 05:42 UTC · model grok-4.3

classification 🌌 astro-ph.GA
keywords globular clustersmultiple stellar populationsAGB starschemical abundancesgiant molecular cloudsstar formationhelium abundance
0
0 comments X

The pith

A new globular cluster formation scenario using AGB pollution of giant molecular clouds reproduces observed multiple stellar populations and chemical patterns.

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

The paper proposes that both first and second generation stars in globular clusters form from giant molecular clouds polluted by asymptotic giant branch stars within and around the clouds. This approach addresses previous issues with mass budgets and timing of dilution in other models. It uses idealized analytic models to show that observed fractions of first population stars and helium spreads match data as a function of cluster mass. Various chemical abundance patterns like O-Na anticorrelations are reproduced, along with specific predictions for other observables such as carbon isotope ratios.

Core claim

In this scenario both the first and second populations of stars form from giant molecular clouds polluted by asymptotic giant branch stars within and around the clouds. Unlike previous scenarios with AGB stars as primary polluters, it alleviates the mass-budget and dilution-timing problems. The principal results based on idealized analytic models are that the observed fraction of 1P stars and the helium abundance spreads between the 1P and 2P as a function of GC masses can be well reproduced, the modelled GCs show O-Na, C-N, and Mg-Al anticorrelations and Si-Al, 25Mg-Al, 26Mg-Al correlations, and additional predictions hold for iron-complex GCs, young clusters, and specific abundance ratios.

What carries the argument

The star-cloud interaction scenario in which 1P and 2P stars form from GMCs polluted by AGB stars, implemented through idealized analytic models.

If this is right

  • The observed fraction of 1P stars and helium abundance spreads as a function of GC mass are reproduced.
  • Modelled GCs exhibit O-Na, C-N, and Mg-Al anticorrelations along with Si-Al, 25Mg-Al, and 26Mg-Al correlations.
  • Iron-complex Type-II GCs form through merging of two GCs from two GMCs at different epochs in a host dwarf galaxy.
  • Young massive clusters in environments with surface star formation rate densities below 1 solar mass per year per square kiloparsec are unlikely to evolve into GCs with multiple populations.
  • Specific predictions include low 12C/13C ratios around 5 for 2P stars, a [Na/Fe]-[F/Fe] anticorrelation, and P-rich stars with [P/Fe] > 0.5 and [N/Fe] > 0.5.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the models hold, full hydrodynamical simulations may not be required to capture the main chemical outcomes of AGB pollution in GMCs.
  • The scenario connects GC formation to chemical evolution processes in dwarf galaxies through the proposed merging mechanism for iron-complex clusters.
  • The requirement that about 20 percent of AGB stars produce Li-rich ejecta to match Li abundance observations could be tested by targeted abundance surveys in additional clusters.

Load-bearing premise

The idealized analytic models of star-cloud interactions accurately capture the timing and efficiency of AGB pollution within and around GMCs without requiring full hydrodynamical simulations or additional physics.

What would settle it

Checking whether young massive clusters formed in galaxy environments with surface star formation rate densities well below 1 solar mass per year per square kiloparsec exhibit multiple populations or chemical spreads would directly test one of the scenario's key predictions.

Figures

Figures reproduced from arXiv: 2606.29707 by Kenji Bekki, Madeleine McKenzie.

Figure 1
Figure 1. Figure 1: Observed correlations and anticorrelations between various chemical abundances of GC stars that this paper discusses in the context of the SCI scenario. The mean and 1σ dispersion of chemical abundances at different abundance bins are shown by filled circles and error bars, respectively. The details of these observational data sets are given in the main text. These are suggested to be the nine benchmark te… view at source ↗
Figure 2
Figure 2. Figure 2: An illustration of the SCI scenario. In this scenario, seed small molecular clouds (MCs) can grow through merging and accretion of other small MCs to become a GMC (Step 1). New stars (“1P”) can form during this growth, and the growing GMC can be also polluted by pre-existing field stars (“0P”) within and around GMCs (2). As star formation accelerates within the GMC, new stars (“2P”) start to form from the … view at source ↗
Figure 3
Figure 3. Figure 3: Threshold mass densities of AGB stars (ρagb,th) above which GCs with MPs can finally form from GMCs in gas-rich galaxies as a function of GMC lifetimes (tlife) for different model parameters: dependence on F2P in the upper left panel, on vgmc in the upper right, on ϵsf in the lower left, and on Rgmc in the lower right). 7 7 7 7 7  t   [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Threshold SFR densities (ΣSFR,th) above which GCs with MPs can finally form from GMCs in gas-rich galaxies as a function of tlife for F2P=0.1 (solid), 0.3 (dotted), and 0.5 (dashed). Clearly, ΣSFR,th should be higher for higher F2 for a given tlife. ρagb,th = F2PMgc πϵsffejR2 gmcvgmctlife(1 + Fdil)(1 − flost) . (6) Thus, ρagb,th depends both on GMC properties (e.g., Rgmc) and galactic dynamical parameters … view at source ↗
Figure 5
Figure 5. Figure 5: Time evolution of the total masses of pristine GMC gas (Mg, blue), AGB ejecta (Mej, purple), all new stars (Mns, orange), and 2P stars (M2P, green) in the fiducial model. not depend on fsf. We can estimate M1P from M1P,0 as done for M2P: M1P = (1 − flost)M1P,0, (26) where flost is fixed at 0.75. Our future simulations need to be run to find a reason￾able range of flost for 1P and 2P stars, because flost co… view at source ↗
Figure 6
Figure 6. Figure 6: Time evolution of [O/Fe] (blue), [Na/Fe] (purple), [Mg/Fe] (orange), [Al/Fe] (green) for gas in the fiducial model. observed abundance spread in various elements. We use the “one-zone” model in which gas and metals from AGB stars can be instantaneously mixed and converted into new stars. The mass and time units in the model is 106M⊙ and 108 yr, respectively, and all model parameters are given in these unit… view at source ↗
Figure 7
Figure 7. Figure 7: Mass fractions of 1P stars (F1P) as a function of the present-day GC masses (Mgc) for simulated 300 GCs (blue) and observations (gray). Observational data from M18 is used here. 3      MM⊙     [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Differences in helium abundances (δY ) between 1P and 2P stars as a function of Mgc. for simulated GCs (blue) and observations (gray). Observational data from M18 is used here. tic are provided in the corresponding subsections of Section 4 and 5 where the model predictions are compared directly with the data. © 2005 RAS, MNRAS 000, 1–?? [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Anticorrelations between [O/Fe] and [Na/Fe] in three models with α = 2.3 and magb,l = 4M⊙ (blue) α = 2.3 and magb,l = 6M⊙ (purple) and α = 1.7 and magb,l = 4M⊙ (orange) and observations from C09a and C09b (gray). [O/Fe] and [Na/Fe] at 1000 time steps are shown for each model in this figure. The AGB models from D10 are used to calculate the IMF-averaged AGB yields in these models. the convective envelope in… view at source ↗
Figure 11
Figure 11. Figure 11: Same as [PITH_FULL_IMAGE:figures/full_fig_p015_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Upper and lower four panels shows how [Al/Fe] dis￾tributions of stars depend on Fdil and [Fe/H] in GCs, respectively. the strength of this anticorrelation relative to O-Na reflects this higher temperature threshold [PITH_FULL_IMAGE:figures/full_fig_p015_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Same as [PITH_FULL_IMAGE:figures/full_fig_p016_13.png] view at source ↗
Figure 15
Figure 15. Figure 15: Same as [PITH_FULL_IMAGE:figures/full_fig_p017_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: A(Li) as a function of [Na/Fe] (left) and [Al/Fe] (right) in the five models with different fractions of Li-producing AGB stars (P(Li)): 0 (blue), 0.01 (purple), 0.03 (orange), 0.1 (green), and 0.18 (black). Observational results taken from Lind et al. (2009) and McKenzie et al. 2026 (MA26) are plotted in the left panel, and the mean and 1σ dispersion for five [Na/Fe] bins are also plotted with error bars… view at source ↗
Figure 17
Figure 17. Figure 17: 25Mg/Mg (left) and 26Mg/Mg ratios (right) as a function of [Al/Fe] for three models with different AGB ejecta compositions, denoted as R(25Mg/Mg) and R(26Mg/Mg), respectively. In the left panel, R(25Mg/Mg) is set to 0.2 (blue), 0.15 (purple), and 0.1 (orange). In the right panel, R(26Mg/Mg) is set to 0.6 (blue), 0.4 (purple), and 0.2 (orange). Reanalysis of archival data from Yong et al. (2003) for NGC 67… view at source ↗
Figure 18
Figure 18. Figure 18: 12C/13C ratios as a function of [Na/Fe] in the three models with magb = 4M⊙ (blue), magb = 5.5M⊙ (purple), and magb = 6M⊙ (orange). The observed data points from Carretta et al. (2015) are plotted by small gray circles, and the mean and 1σ dispersion for [Na/Fe] bins are indicated by large gray points and error bars, respectively. AGB yields from a selected mass (magb) are used in each model (i.e., no IMF… view at source ↗
Figure 19
Figure 19. Figure 19: Mass budget factors for s-rich stars (Fb,s−rich) as a function of [Ba/Fe] of polluting AGB stars for ϵsf = 0.4 (solid) and 0.1 (dotted). The observed total stellar masses and [Ba/Fe] in s-poor and s-rich stars in M22 are used to plot these. Fb,s−rich > 1 means that the total mass required to explain the observed [Ba/Fe] of s-rich stars is larger than the total-mass of s￾poor stars. The required large Fb,s… view at source ↗
Figure 20
Figure 20. Figure 20: Evolution of the four models with XF = XF(magb = 6M⊙) (blue), 0.5×XF(magb = 6M⊙) (purple), 0.25×XF(magb = 6M⊙) (orange), and 0.5 × XF(magb = 5M⊙) (green) on the [Na/Fe]-[F/Fe] plane. The observed abundances taken from Smith et al. (2005) for M4 are shown by gray circles with observational error bars. The observed Na-F anticorrelation can be better re￾produced by the models with reduced F yields. anticorre… view at source ↗
Figure 22
Figure 22. Figure 22: Total gaseous (blue) and stellar masses (purple) of a gas-rich dwarf galaxy as a function of [Fe/H] in the one-zone chemical evolution model of the galaxy for the first 1 Gyr evo￾lution. The blue and purple dotted lines indicate the threshold gaseous (Mg,th) and stellar masses (Mns,th) for GC formation with MPs, respectively. The crosspoint between the solid and dot￾ted lines can mark the threshold [Fe/H]… view at source ↗
Figure 23
Figure 23. Figure 23: MTO of intermediate-mass stars as a function of time (T) in six models with CSF = 0.02 (blue), 0.05 (purple), and 0.1 (orange) for the IMF slope α = 2.35 (solid lines) and 1.5 (dotted lines). The total mass of intermediate-mass stars (MTO(4 ⩽ mTO/M⊙ ⩽ 10)) that have just left the main-sequence at a give time step is calculated in these models. MTO normal￾ized by the total mass of in falling gas is plotted… view at source ↗
read the original abstract

We present a new scenario of globular cluster (GC) formation with multiple stellar populations (MPs) in which both the first and second populations (1P and 2P, respectively) of stars form from giant molecular clouds (GMCs) polluted by asymptotic giant branch (AGB) stars within and around the GMCs. Unlike previous GC formation scenarios with AGB stars being the primary polluters, the new scenario alleviates tensions with the mass-budget and dilution-timing problems The principal results based on idealized analytic models of the formation scenario are as follows. The observed fraction of 1P stars and the helium abundance spreads between the 1P and 2P as a function of GC masses can be well reproduced. The modelled GCs show O-Na, C-N, and Mg-Al anticorrelations and Si-Al, 25Mg-Al, 26Mg-Al correlations. The observed Mg-K anticorrelation can be reproduced, only if super AGB stars make a significant contribution to chemical enrichment within GMCs. The lack of correlations of Li abundances with [Na/Fe] and [Al/Fe] can be reproduced, only if about 20% of the polluting AGB stars produce Li-rich ejecta, which disfavours scenarios with polluters incapable of Li production. Iron-complex, Type-II GCs can be formed through merging of two GCs formed from two GMCs within a host dwarf galaxy at different epochs. The new scenario predicts young massive clusters formed in galaxy environments with surface star formation rate densities well below 1 M_sun/yr/kpc^2 are unlikely to evolve into GCs with MPs. It also predicts low 12C/13C ratios of 2P (~5), a [Na/Fe]-[F/Fe] anticorrelation, and P-rich star formation with [P/Fe]$>0.5$ and [N/Fe]>0.5. These predictions are tested against more than 30 observed properties of GCs with MPs, representing one of the most comprehensive observational benchmarks against a specific GC formation scenario to date.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper proposes a new globular cluster formation scenario with multiple stellar populations in which both 1P and 2P stars form from giant molecular clouds polluted by AGB stars via star-cloud interactions. Using idealized analytic models, it claims to reproduce the observed 1P star fractions and helium abundance spreads as a function of GC mass, O-Na/C-N/Mg-Al anticorrelations, Si-Al and Mg-Al correlations, and (with additional conditions) the Mg-K anticorrelation and lack of Li correlations. It further claims to explain iron-complex GCs via mergers, make several testable predictions (e.g., low 12C/13C in 2P stars, P-rich stars), and pass a benchmark of >30 observed GC properties while alleviating mass-budget and dilution-timing issues in prior AGB scenarios.

Significance. If the idealized analytic models prove robust and the parameter choices are shown to be non-circular, the work would offer a comprehensive, observationally benchmarked framework for GC formation with MPs that resolves longstanding tensions and generates falsifiable predictions across chemical patterns and cluster demographics.

major comments (2)
  1. [Abstract] Abstract: reproduction of the Mg-K anticorrelation is stated to require 'significant contribution' from super AGB stars, and reproduction of the lack of Li-[Na/Fe] and Li-[Al/Fe] correlations requires that 'about 20% of the polluting AGB stars produce Li-rich ejecta'; these are post-hoc conditions introduced to match specific observations rather than derived from the model, directly affecting the claimed success on multiple chemical patterns.
  2. [Abstract] Abstract: the central quantitative claims (1P fractions, He spreads vs. GC mass, multiple anticorrelations) rest on idealized analytic models of star-cloud interactions, yet the abstract provides no derivation details, error bars on the fits, or direct comparison to hydrodynamical simulations; any mismatch in assumed cloud structure, mixing efficiency, or feedback geometry would alter the predicted 1P/2P ratios and abundance spreads.
minor comments (1)
  1. [Abstract] The assertion that the scenario is tested against 'one of the most comprehensive observational benchmarks' is subjective and should be qualified or supported by an explicit enumeration of the 30+ properties.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful review and recommendation for major revision. We address each major comment on the abstract point by point below, providing clarifications on the model conditions and the scope of the idealized analytic approach while noting where revisions to the manuscript can improve clarity.

read point-by-point responses

  1. Referee: [Abstract] Abstract: reproduction of the Mg-K anticorrelation is stated to require 'significant contribution' from super AGB stars, and reproduction of the lack of Li-[Na/Fe] and Li-[Al/Fe] correlations requires that 'about 20% of the polluting AGB stars produce Li-rich ejecta'; these are post-hoc conditions introduced to match specific observations rather than derived from the model, directly affecting the claimed success on multiple chemical patterns.

    Authors: The abstract already qualifies these as necessary conditions ('only if') for reproducing the Mg-K anticorrelation and the absence of Li correlations. These requirements are not arbitrary post-hoc adjustments but follow directly from applying published AGB nucleosynthesis yields (including super-AGB models) to match the full set of observed abundance patterns while remaining consistent with independent constraints on AGB star populations. The manuscript uses these conditions to generate further predictions (e.g., disfavouring polluters unable to produce Li) and to benchmark against >30 GC properties. We agree the abstract could more explicitly note that the fractions are motivated by stellar evolution calculations rather than free parameters, and we will revise the wording accordingly for improved transparency. revision: partial

  2. Referee: [Abstract] Abstract: the central quantitative claims (1P fractions, He spreads vs. GC mass, multiple anticorrelations) rest on idealized analytic models of star-cloud interactions, yet the abstract provides no derivation details, error bars on the fits, or direct comparison to hydrodynamical simulations; any mismatch in assumed cloud structure, mixing efficiency, or feedback geometry would alter the predicted 1P/2P ratios and abundance spreads.

    Authors: The abstract explicitly describes the results as coming from 'idealized analytic models' and the main text supplies the full derivations, including the assumed GMC density profiles, star-cloud interaction geometry, mixing prescriptions, and feedback assumptions, together with direct comparisons to the observed 1P fractions, He spreads, and abundance anticorrelations. Because the models are deterministic analytic expressions rather than statistical fits, formal error bars are not applicable; parameter sensitivities are instead explored and discussed in the manuscript. We acknowledge that hydrodynamical simulations would provide valuable cross-checks on the mixing and geometry assumptions, but such comparisons lie beyond the present scope of this analytic framework. We will add a brief clarifying phrase in the abstract to emphasize the idealized character and refer readers to the detailed model sections. revision: partial

Circularity Check

2 steps flagged

Reproduction of Li correlations and other patterns requires tuned polluter fractions chosen to match observations

specific steps
  1. fitted input called prediction [Abstract]
    "The lack of correlations of Li abundances with [Na/Fe] and [Al/Fe] can be reproduced, only if about 20% of the polluting AGB stars produce Li-rich ejecta, which disfavours scenarios with polluters incapable of Li production."

    The 20% fraction is introduced as a condition required to reproduce the observed lack of Li correlations; the match is therefore obtained by setting the input parameter to the value that produces the target output rather than emerging as a model prediction.

  2. fitted input called prediction [Abstract]
    "The observed Mg-K anticorrelation can be reproduced, only if super AGB stars make a significant contribution to chemical enrichment within GMCs."

    The requirement for significant super-AGB contribution is stated as necessary to reproduce the Mg-K anticorrelation, indicating that the parameter is adjusted to fit the observation rather than predicted independently by the star-cloud interaction scenario.

full rationale

The paper's central results on reproducing observed GC properties (1P fractions, He spreads, Li lack of correlation, Mg-K anticorrelation) are achieved by selecting specific parameter values (20% Li-rich AGB fraction, significant super-AGB contribution) that are explicitly conditioned on matching those same observations. This reduces the claimed reproductions to fits by construction rather than independent predictions from the analytic models. No equations or self-citations are shown to create further circularity, and the models are presented as new, so the circularity is partial and limited to the fitted inputs.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The model depends on several fitted fractions and contributions to match observations, plus domain assumptions about AGB pollution efficiency and GMC dynamics. No new particles or forces are invented. Review is abstract-only so full parameter list is unknown.

free parameters (2)
  • fraction of polluting AGB stars producing Li-rich ejecta
    Set to ~20% to reproduce the lack of Li correlations with Na and Al.
  • contribution of super AGB stars to chemical enrichment
    Required to be significant to reproduce the Mg-K anticorrelation.
axioms (1)
  • domain assumption AGB stars are the primary polluters of GMCs both within and around the clouds
    Central premise of the new scenario stated in the abstract.

pith-pipeline@v0.9.1-grok · 7020 in / 1315 out tokens · 56516 ms · 2026-06-30T05:42:09.740738+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

186 extracted references · 3 canonical work pages · 2 internal anchors

  1. [1]

    Adamo, A., et al., 2020, MNRAS, 499, 3267

    2020

  2. [2]

    C., 2010, ARA&A, 48, 47 Alvarez Garay, D

    Adams, F. C., 2010, ARA&A, 48, 47 Alvarez Garay, D. A., et al., 2022, ApJ, 928 L11 Alvarez Garay, D. A., et al., 2025, A&A, 704, 114 Alvarez Garay, D. A., et al., 2026, A&A, 707, 280

    2010

  3. [3]

    Bailin, J., 2018, ApJ, 863, 99

    2018

  4. [4]

    Barbuy, B., et al., 2025, AJ, 170, 245

    2025

  5. [5]

    S., 2013, MNRAS, 436, 2852

    Bastian, N., Cabrera-Ziri, I., Davies, B., Larsen, S. S., 2013, MNRAS, 436, 2852

    2013

  6. [6]

    Bastian, N., Lamers, H. J. G. L. M., de Mink, S. E., et al. 2013, MNRAS, 436, 2398

    2013

  7. [7]

    2015, MNRAS, 449, 3333

    Bastian, N., Cabrera-Ziri, I., & Salaris, M. 2015, MNRAS, 449, 3333

    2015

  8. [8]

    Bastian, N., Lardo, C., 2018, ARA&A, 56, 83

    2018

  9. [9]

    Baumgardt, H., & Kroupa, P., 2007, MNRAS, 380, 1589

    2007

  10. [10]

    A., et al., 2019, MNRAS, 487, 1986

    Beasley, M. A., et al., 2019, MNRAS, 487, 1986

    2019

  11. [11]

    R., Piotto, G., Anderson, J., et al., 2004, ApJ, 605, L125

    Bedin, L. R., Piotto, G., Anderson, J., et al., 2004, ApJ, 605, L125

    2004

  12. [12]

    Bekki, K., 2010, ApJ, 724, L99

    2010

  13. [13]

    Bekki, K., 2011, MNRAS, 412, 2241

    2011

  14. [14]

    Bekki, K., 2015, MNRAS, 449, 1625

    2015

  15. [15]

    Bekki, K., 2019, A&A, 622, 53 (2019a)

    2019

  16. [16]

    Bekki, K., 2019, MNRAS, 486, 2570 (2019b)

    2019

  17. [17]

    Bekki, K., 2019, MNRAS, 490, 4007 (2019c)

    2019

  18. [18]

    Bekki, K., 2023, MNRAS, 518, 3274

    2023

  19. [19]

    E., 2006, ApJ, 637, L109

    Bekki, K., Norris, J. E., 2006, ApJ, 637, L109

    2006

  20. [20]

    Bekki, K., et al., 2007, MNRAS, 377, 335

    2007

  21. [21]

    D., 2009, MNRAS, 394, 124

    Bekki, K., Mackey, A. D., 2009, MNRAS, 394, 124

    2009

  22. [22]

    Bekki, K., Tsujimoto, T., 2012, ApJ, 761, 180 (BT12)

    2012

  23. [23]

    Bekki, K., Yong, D., 2012, MNRAS, 419, 2063 (BY12)

    2012

  24. [24]

    Bekki, K., Tsujimoto, T., 2016, ApJ, 831, 70 (BT16)

    2016

  25. [25]

    Bekki, K., Tsujimoto, T., 2017, ApJ, 844, 34

    2017

  26. [26]

    Bekki, K., Tsujimoto, T., 2024, ApJ, 937, L1

    2024

  27. [27]

    Bekki, K., Furuya, K., Shimonishi, T., 2025, MNRAS, 543, 2997

    2025

  28. [28]

    Bianchini, P., et al., 2026, prepint (arXiv:2603.26195)

    work page arXiv 2026

  29. [29]

    B., 2025, MNRAS, 544 3601

    Bobrick, A., Davies, M, B., Perets, H. B., 2025, MNRAS, 544 3601

    2025

  30. [30]

    et al., 2007 A&A, 470, 153

    Bonifacio, P. et al., 2007 A&A, 470, 153

    2007

  31. [31]

    Brauner, M., et al., 2023, A&A, 637, 123

    2023

  32. [32]

    Bragaglia, A., et al., 2017, A&A, 607, 44

    2017

  33. [33]

    Buder, S., et al., 2025, PASA, 42, 51

    2025

  34. [34]

    L., et al., 2025, A&A, 695, 124

    Buzzo, M. L., et al., 2025, A&A, 695, 124

    2025

  35. [35]

    Bruzual, G., Schweizer, F., 2014, MNRAS, 441, 2754

    2014

  36. [36]

    Cameron, A. G. W., Fowler, W. A., 1971, ApJ, 164, 111

    1971

  37. [37]

    Carretta, E., 2014, ApJ, 795, L28

    2014

  38. [38]

    Carretta, E., et al., 2005, A&A, 433, 597

    2005

  39. [39]

    Carretta, E., et al., 2009, A&A, 505, 117 (C09a)

    2009

  40. [40]

    G., Lucatello, S., 2009, A&A, 505, 117 (C09b)

    Carretta, E., Bragaglia, A., Gratton, R. G., Lucatello, S., 2009, A&A, 505, 117 (C09b)

    2009

  41. [41]

    Carretta, E., et al., 2010, ApJ, 722, L1

    2010

  42. [42]

    G., Lucatello, S., D’Orazi, V., 2012, ApJ, 750, L14

    Carretta, E., Bragaglia, A., Gratton, R. G., Lucatello, S., D’Orazi, V., 2012, ApJ, 750, L14

    2012

  43. [43]

    Carretta, E., Bragaglia, A., 2025, A&A, 696, 120

    2025

  44. [44]

    Chevance, M., et al., 2020, Space Science Reviews, 216, 50

    2020

  45. [45]

    V., 1979, ApJS, 41, 743

    Cohen, M., Kuhi, L. V., 1979, ApJS, 41, 743

    1979

  46. [46]

    & Spergel, D

    Conroy, C. & Spergel, D. N. 2011, ApJ, 726, 36

    2011

  47. [47]

    L.; Da Costa, G

    Cottrell, P. L.; Da Costa, G. S., 1981, ApJL, 245, 79

    1981

  48. [48]

    Cristallo, S., et al., 2011, ApJS, 197, 17

    2011

  49. [49]

    S., Norris, J

    Cunha, K., et al., 2003, AJ, 126, 1305 Da Costa, G. S., Norris, J. E., Yong, D., 2003, ApJ, 769, 8

    2003

  50. [50]

    2018, ApJ, 864, 33 D’Antona, F., & Caloi, V

    Dalessandro, E., Mucciarelli, A., Bellazzini, M., et al. 2018, ApJ, 864, 33 D’Antona, F., & Caloi, V. 2004, ApJ, 611, 871 D’Antona, F., et al. 2012, 426, 1710 D’Antona, F., Vesperini, E., D’Ercole, A., Ventura, P.,

    2018

  51. [51]

    P., Marino, A

    Milone, A. P., Marino, A. F., & Tailo, M. 2016, MNRAS, 458, 2122

    2016

  52. [52]

    & Ekstr¨ om, S

    Decressin, T., Meynet, G., Charbonnel, C., Prantzos, N. & Ekstr¨ om, S. 2007, A&A, 464, 1029 Dell’Agli, F., et al., 2018, 475, 3098 (D18) de Mink, S. E., Pols, O. R., Langer, N., 2009, A&A, 507, L1

    2007

  53. [53]

    A., Hartwick

    Denissenkov, P. A., Hartwick. F. D. A., 2014, MNRAS, 437, L21 D’Ercole, A., Vesperini, E., D’Antona, F., McMillan, S. L. W., & Recchi, S. 2008, MNRAS, 391, 825 (D08) D’Ercole, A., D’Antona, F., Ventura, P., Vesperini, E.,

    2014

  54. [54]

    McMillan, S. L. W., 2010, MNRAS, 407, 854 (D10) D’Orazi, V., Marino, A. F., 2010, ApJ, 716, L166 D’Orazi, V., et al., 2014, ApJ, 791, 39

    2010

  55. [55]

    Dobrovolskas, V., et al., 2014, A&A, 565, 121

    2014

  56. [56]

    L., et al., 2015, 446, 2599 ©2005 RAS, MNRAS000, 1–?? 36K

    Doherty, C. L., et al., 2015, 446, 2599 ©2005 RAS, MNRAS000, 1–?? 36K. Bekki, M. McKenzie,

    2015

  57. [57]

    L., et al., 2017, PASA, 34, 56 Elmegreen B

    Doherty, C. L., et al., 2017, PASA, 34, 56 Elmegreen B. G., & Efremov, Y. N., 1997, ApJ, 480, 235 Elmegreen B. G., 2017, ApJ, 836, 80

    2017

  58. [58]

    K., 2004, MNRAS, 353, 789 Fern´ andez-Trincado, J

    Gibson, B. K., 2004, MNRAS, 353, 789 Fern´ andez-Trincado, J. G., et al., 2019, A&A, 631, 97

    2004

  59. [59]

    R., 2009, Nat, 462, 483

    Ferraro, F. R., 2009, Nat, 462, 483

    2009

  60. [60]

    Forestini, M., Goriely, S., Jorissen, A., Arnould, M., 1992, A&A, 261, 157

    1992

  61. [61]

    2024, PhRvL, 132, 2701

    Fox, W., et al. 2024, PhRvL, 132, 2701

    2024

  62. [62]

    Grasha, K., et al., 2019, MNRAS, 483, 4707

    2019

  63. [63]

    Gavagnin, E., Mapelli, E., Lake, G., 2016, MNRAS, 461, 1276

    2016

  64. [64]

    P., Burkert, A., 2001, MNRAS, 323, 988

    Geyer, M. P., Burkert, A., 2001, MNRAS, 323, 988

    2001

  65. [65]

    Gieles, M., et al., 2025, MNRAS, 544, 483

    2025

  66. [66]

    G., Carretta, E., Bragaglia, A., 2012, A&ARv, 20, 50

    Gratton, R. G., Carretta, E., Bragaglia, A., 2012, A&ARv, 20, 50

    2012

  67. [67]

    G., et al., 2019, ARA&A, 27, 8

    Gratton, R. G., et al., 2019, ARA&A, 27, 8

    2019

  68. [68]

    Goudfrooij, P., 2018, ApJ, 857, 16

    2018

  69. [69]

    Goudfrooij, P., et al., 2026, accepted by MNRAS (arXiv.2606.07751)

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  70. [70]

    Harris, J., Zaritsky, D., 2009, AJ, 138, 1243

    2009

  71. [71]

    E., Pudritz, R

    Harris, W. E., Pudritz, R. E., 1994, ApJ, 429, 177

    1994

  72. [72]

    E., Gerasimov, R., Kirby, E

    Henderson, L. E., Gerasimov, R., Kirby, E. N., 2025, ApJ, 992, L14

    2025

  73. [73]

    G., 1980, 235, 986 K

    Hills, J. G., 1980, 235, 986 K. Hollyhead, K., et al., 2018, MNRAS, 476, 114

    1980

  74. [74]

    2016, ApJ, 818, 98

    Iliadis, C., et al. 2016, ApJ, 818, 98

    2016

  75. [75]

    G., et al., 2026, MNRAS, 546, 2272

    Kane, S. G., et al., 2026, MNRAS, 546, 2272

    2026

  76. [76]

    I., 2010, MNRAS, 403, 1413

    Karakas, A. I., 2010, MNRAS, 403, 1413

    2010

  77. [77]

    I., Lattanzio, J

    Karakas, A. I., Lattanzio, J. C., 2014, PASA, 31, 30

    2014

  78. [78]

    Kawamura, A., et al., 2009, ApJS, 181, 1

    2009

  79. [79]

    Kemp, A., et al., 2022, MNRAS, 509, 1175

    2022

  80. [80]

    N., Ji, A

    Kirby, E. N., Ji, A. P., Kovalev, M., 2023, ApJ, 958, 45

    2023

Showing first 80 references.