Pith. sign in

REVIEW 2 major objections 5 minor 55 references

Intrinsic iron abundance spreads in globular clusters

T0 review · 2 major / 5 minor · reviewed 2026-07-14 · grok-4.5

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

Pith's one-line read Most Galactic globular clusters show no intrinsic iron spread; only three of thirteen do, and the data reject widespread iron variation.

desk verdict Clean differential Fe analysis of 92 sibling stars in 13 GCs that largely refutes ubiquitous iron spreads claimed from photometry; three real exceptions, small-N limits, solid enough to engage. read the letter →

arxiv 2607.09966 v1 pith:XZYHYLJQ submitted 2026-07-10 astro-ph.SR astro-ph.GA

J. Schiappacasse-Ulloa , S. Lucatello , L. Magrini , A. Bragaglia , E. Carretta This is my paper · ORCID
classification astro-ph.SRastro-ph.GA
keywords globularclustersmultiplestellarpopulationsironabundancedifferentialspectroscopychemicalenrichmentfirst-generationstars
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

Globular clusters are famous for multiple stellar populations that differ in light elements such as sodium and oxygen, but iron was long thought to be the same in almost every star. Recent photometric work claimed large iron spreads even among first-generation stars, challenging the standard picture that only light-element polluters operated. This paper re-measures iron in 92 carefully matched “sibling” stars across 13 clusters using a strictly differential spectroscopic method that cancels most systematic errors. Monte Carlo tests then ask whether the residual scatter exceeds pure measurement noise. In ten clusters the answer is no; only NGC 1851, NGC 3201 and NGC 5634 show statistically significant iron spreads, and even those are modest. The authors therefore conclude that intrinsic iron inhomogeneity is rare, not a common feature of Galactic globular clusters.

What carries the argument

Sibling-star differential analysis with the Q2 code: stars whose effective temperatures differ by at most 100 K are compared line-by-line so that non-LTE effects, continuum errors and model-atmosphere biases largely cancel, yielding iron abundances precise enough for Monte Carlo significance tests of the residual spreads.

What would settle it

A homogeneous differential re-analysis of a larger sample of first-generation stars in the same clusters (or in the clusters claimed by photometry to have the largest spreads) that recovers iron dispersions clearly larger than the measurement uncertainties would falsify the claim that widespread iron variation is absent.

Watch

Extended reading notes

Core claim

Differential iron abundances of 92 sibling stars in 13 Galactic globular clusters reveal no statistically significant internal iron spreads in the majority of the sample. Only NGC 1851, NGC 3201 and NGC 5634 display highly significant spreads; the data as a whole do not support widespread iron variation among globular-cluster stars.

Load-bearing premise

Stars whose photometric temperatures differ by no more than 100 K are similar enough that residual systematic errors cancel to well below the 0.02–0.05 dex level claimed for the non-detections.

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

2 major / 5 minor

Summary. The paper presents a strictly differential iron-abundance analysis of 92 sibling RGB stars (groups with photometric Teff differing by ≤100 K and N≥4) in 13 Galactic globular clusters, using UVES spectra previously analysed by Carretta and collaborators. Equivalent widths are measured independently with ARES and REvIEW, averaged after cross-calibration, and fed to Q2/MOOG to obtain line-by-line differential stellar parameters and ΔFe relative to a reference star in each group. Monte-Carlo realisations (10 000 iterations) that inject only the reported measurement uncertainties are used to assign p-values to the observed group dispersions. Most clusters yield σ Fe ≈ 0.02–0.05 dex that are statistically consistent with pure error; only NGC 1851, NGC 3201 and NGC 5634 show p ≤ 0.05. An additional injection test demonstrates that the large FG iron spreads reported by Legnardi et al. (2022) are difficult to reconcile with the present data. The authors conclude that widespread intrinsic iron variations are not supported.

Significance. If the non-detections hold, the work supplies the largest homogeneous differential Fe data set yet assembled for GCs and places a quantitative upper limit on the prevalence of iron inhomogeneities. The dual-EW pipeline, the tight sibling criterion, the external validation against McKenzie et al. (2022) (recovering differential [Fe/H] to ~0.01 dex), and the Monte-Carlo null tests constitute a clean, falsifiable methodology that can be extended to larger samples and additional elements. The result directly addresses the tension between photometric claims of ~0.1–0.3 dex Fe spreads and traditional spectroscopic upper limits, and therefore has clear impact on models of GC formation and chemical enrichment.

major comments (2)
  1. Section 3.1 and Table 3: several clusters that drive the global non-detection claim (e.g. NGC 104, NGC 1904, NGC 6121) contain only a single FG star, so the generation-resolved test has essentially zero power for those systems. While the authors correctly note the limitation, the abstract and final conclusion still speak of “no widespread iron variation” without a quantitative statement of the minimum detectable spread given the actual N and error budget. A short power analysis (or an explicit upper-limit column in Table 3) would make the statistical reach of the sample transparent and would strengthen the claim against the Legnardi et al. results.
  2. Section 4.2 (comparison with Lardo et al. 2023 on NGC 2808): the paper notes the absence of common stars and the different radial coverage but does not quantify whether the discrepancy could be produced by the different spatial sampling alone. Because the central claim is that iron spreads are rare, a brief estimate of the expected dispersion difference under a radially varying Fe model (or an explicit statement that such a model cannot be tested with the present data) would close this residual tension.
minor comments (5)
  1. Figure 5: the p-value labels are useful, but the observed-spread vertical lines would be clearer if accompanied by a short legend or caption note that “p ≤ 0.05 is adopted as significant”.
  2. Table 1 is said to be available only at the CDS; a short excerpt (or at least the column definitions) in the main text or an appendix would help readers evaluate the differential parameters without leaving the paper.
  3. Section 2.2: the 20 % EW-difference rejection criterion is stated but not justified; a one-sentence note on how many lines were discarded and whether the final results are sensitive to that cut would be useful.
  4. Abstract and Section 5: “highly significant iron spread” for NGC 1851, 3201 and 5634 is slightly stronger language than the p-values alone (0.02–0.04) warrant; “statistically significant at the p ≤ 0.05 level” would be more precise.
  5. Figure 9 caption: the colour coding is described as green/orange/grey, but the text of Section 4.4 refers to blue/orange/grey; a single consistent scheme would avoid confusion.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: differential Fe abundances and Monte-Carlo null tests are independent of the photometric spreads and generation labels being assessed.

full rationale

The derivation chain is self-contained and non-circular. Sibling groups are defined solely by photometric Teff proximity (≤100 K) from the input catalogue; EWs are remeasured independently with two codes and averaged; Q2 then solves for differential parameters and line-by-line ΔFe relative to an internal reference star. Observed group dispersions are compared to a pure measurement-error null via 10 000 Monte-Carlo draws that use only the Q2 abundance uncertainties; no free parameters are fitted to the Fe data themselves. Generation labels are taken from prior [Na/Fe] (Carretta et al. 2009a) that are chemically orthogonal to Fe and are used only for a secondary split, not to force the primary non-detection. External validation on three M22 stars recovers literature differential [Fe/H] to ~0.01 dex. The additional injection test of Legnardi et al. (2022) spreads is a proper falsification exercise, not a re-statement of those spreads. Minor self-citations supply the input spectra and Na classifications but do not underwrite the Fe-spread statistics or the Monte-Carlo significance thresholds. The central claim (most clusters consistent with zero intrinsic Fe spread) therefore stands on new measurements rather than on definitional or fitted tautologies.

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

The analysis rests on standard 1D LTE model atmospheres, the assumption that sibling stars cancel systematics, and the Carretta Na-based population labels. No new free parameters are fitted to the Fe data themselves; the only numerical thresholds (100 K, 4 stars, p≤0.05) are analysis choices, not fitted constants.

free parameters (3)
  • sibling ΔTeff threshold = 100 K
    Maximum temperature difference of 100 K used to define sibling groups; chosen by the authors as more stringent than literature values of 200–250 K.
  • minimum group size = 4 stars
    Groups retained only if they contain at least four stars; an analysis cut that affects which clusters enter the final sample.
  • p-value significance threshold = 0.05
    Dispersion declared significant when p≤0.05; conventional but still a free choice that partitions the sample into ‘significant’ and ‘not’.
assumptions (3)
  • domain assumption 1D LTE radiative transfer with α-enhanced Kurucz atmospheres is adequate for differential Fe I/II analysis of RGB stars.
    Invoked throughout Section 2.2–2.3 via Q2/MOOG; standard in the field but known to leave residual non-LTE effects that the differential method is assumed to cancel.
  • ad hoc to paper Stars whose photometric Teff differ by ≤100 K share essentially identical non-LTE corrections, continuum errors and model-atmosphere biases.
    Core premise of the sibling-star method (Section 2.2); if residual systematics remain at the 0.03 dex level the non-detections become inconclusive.
  • domain assumption Carretta et al. (2009a) [Na/Fe] classification correctly separates first- and second-generation stars.
    Used in Section 3.2 to split the sample by generation; any misclassification would scramble the FG/SG spread comparison.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Intrinsic iron abundance spreads in globular clusters." pith.science (2026). https://pith.science/paper/XZYHYLJQ

@misc{pith2026260709966,
  author       = {Pith},
  title        = {Pith review of: Intrinsic iron abundance spreads in globular clusters},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XZYHYLJQ}},
  note         = {Machine review of arXiv:2607.09966}
}
read the original abstract

Globular Clusters (GCs) host multiple stellar populations, likely composed of a subset of FG stars that enriched the intracluster medium and gave rise to SG stars, each characterised by distinctive chemical patterns. These patterns typically include enhancements in elements such as N, Na, and Al, coupled with depletions in C, O, and Mg in SG stars. Traditionally, heavier elements such as those in the Fe-peak were considered unaffected in most GCs. However, recent studies have reported significant internal spreads in these elements, suggesting a more complex picture of chemical enrichment within GCs. This study seeks to derive precise and homogeneous differential iron abundances in a large sample of GCs. By doing so, our aim is to investigate the presence of intrinsic iron spreads within them and to assess the existence of differences in Fe between their stellar populations. We used the Python-based tool Q2 to determine both differential stellar parameters and iron abundances for 92 sibling stars, defined by the similarities in their stellar parameters, across 13 Galactic GCs. This differential approach reduces the influence of non-LTE effects, and minimises observational errors, and systematic biases linked to stellar parameters. We performed Monte Carlo simulations to evaluate the statistical significance of the measured spreads. Most of the GCs in our sample do not show evidence of statistically significant Fe spreads. Only a few exceptions emerge, namely NGC1851, NGC3201, and NGC5634, which display a highly significant iron spread. In particular, NGC3201 shows a particularly pronounced spread in its FG population, reflecting a potential inhomogeneous iron abundance in its pristine material. Finally, through statistical tests, we conclude that our data do not support the presence of a widespread iron variation in GCs.

Figures

Figures reproduced from arXiv: 2607.09966 by the authors.

Figure 1
Figure 1. Spectral region containing Fe I lines (shaded areas). The di [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Comparison of EW measurements obtained with [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Colour–magnitude diagrams of the 13 GCs analysed using Gaia DR3 photometry. Small black symbols indicate cluster [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Comparison between the EWs measured by McKenzie et al. (2022) and those obtained with our method for the M 22 stars [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Histogram of 10,000 Monte Carlo simulations of Fe spread. The dashed purple line indicates the observed Fe spread in our [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Measured abundance spreads for the full GC sample. [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: Histogram of 10,000 Monte simulations of Fe spread by stellar generation. The pink and grey histograms display the [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: [Upper panel:] Histogram of the number of clusters that [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: Differential abundance spreads for the full sample (upper panels), FG stars (middle panels), and SG stars (lower panels), shown as a function of cluster mass (left column), stel￾lar density within the half-mass radius (middle column), and distance from the Sun (right c…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 1 linked inside Pith

  1. [1]

    2020, ApJ, 892, 20

    Alfaro-Cuello, M., Kacharov, N., Neumayer, N., et al. 2020, ApJ, 892, 20

    2020

  2. [2]

    1999, A&AS, 140, 261

    Alonso, A., Arribas, S., & Martínez-Roger, C. 1999, A&AS, 140, 261

    1999

  3. [3]

    & Lardo, C

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

    2018

  4. [4]

    & Hilker, M

    Baumgardt, H. & Hilker, M. 2018, MNRAS, 478, 1520

    2018

  5. [5]

    L., et al

    Bedell, M., Meléndez, J., Bean, J. L., et al. 2014, ApJ, 795, 23

    2014

  6. [6]

    R., & Ibata, R

    Bellazzini, M., Ferraro, F. R., & Ibata, R. 2002, AJ, 124, 915

    2002

  7. [7]

    M., Cautun, M., Deason, A

    Callingham, T. M., Cautun, M., Deason, A. J., et al. 2022, MNRAS, 513, 4107

    2022

  8. [8]

    2015, ApJ, 810, 148

    Carretta, E. 2015, ApJ, 810, 148

    2015

Show all 55 references
  1. [9]

    & Bragaglia, A

    Carretta, E. & Bragaglia, A. 2022, A&A, 659, A122

    2022

  2. [10]

    G., et al

    Carretta, E., Bragaglia, A., Gratton, R. G., et al. 2013, A&A, 557, A138

    2013

  3. [11]

    2017, A&A, 600, A118

    Carretta, E., Bragaglia, A., Lucatello, S., et al. 2017, A&A, 600, A118

    2017

  4. [12]

    G., Bragaglia, A., & D’Orazi, V

    Carretta, E., Lucatello, S., Gratton, R. G., Bragaglia, A., & D’Orazi, V . 2011, A&A, 533, A69

    2011

  5. [13]

    & Kurucz, R

    Castelli, F. & Kurucz, R. L. 2003, in IAU Symposium, V ol. 210, Modelling of Stellar Atmospheres, ed. N. Piskunov, W. W. Weiss, & D. F. Gray, A20

    2003

  6. [14]

    2000, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Se- ries, V ol

    Dekker, H., D’Odorico, S., Kaufer, A., Delabre, B., & Kotzlowski, H. 2000, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Se- ries, V ol. 4008, Optical and IR Telescope Instrumentation and Detectors, ed. M. Iye & A. F. Moorwood, 534–545

    2000

  7. [15]

    F., Milone, A

    Dondoglio, E., Marino, A. F., Milone, A. P., et al. 2025, A&A, 697, A135

    2025

  8. [16]

    2008, ApJS, 178, 89 Gaia Collaboration, Vallenari, A., Brown, A

    Dotter, A., Chaboyer, B., Jevremovi´c, D., et al. 2008, ApJS, 178, 89 Gaia Collaboration, Vallenari, A., Brown, A. G. A., et al. 2023, A&A, 674, A1

    2008

  9. [17]

    S., & Ramírez-Galeano, L

    Gieles, M., Padoan, P., Charbonnel, C., Vink, J. S., & Ramírez-Galeano, L. 2025, MNRAS, 544, 483

    2025

  10. [18]

    2019, A&A Rev., 27, 8

    Gratton, R., Bragaglia, A., Carretta, E., et al. 2019, A&A Rev., 27, 8

    2019

  11. [19]

    2004, Annual Review of Astronomy and Astrophysics, 42, 385

    Gratton, R., Sneden, C., & Carretta, E. 2004, Annual Review of Astronomy and Astrophysics, 42, 385

    2004

  12. [20]

    G., Carretta, E., & Bragaglia, A

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

    2012

  13. [21]

    I., Kraft, R

    Johnson, C. I., Kraft, R. P., Pilachowski, C. A., et al. 2005, PASP, 117, 1308

    2005

  14. [22]

    Johnson, C. I. & Pilachowski, C. A. 2010, ApJ, 722, 1373

    2010

  15. [23]

    I., Rich, R

    Johnson, C. I., Rich, R. M., Pilachowski, C. A., et al. 2015, AJ, 150, 63

    2015

  16. [24]

    2022, ApJ, 939, 118

    Kacharov, N., Alfaro-Cuello, M., Neumayer, N., et al. 2022, ApJ, 939, 118

    2022

  17. [25]

    2022, A&A, 662, A117

    Lardo, C., Salaris, M., Cassisi, S., & Bastian, N. 2022, A&A, 662, A117

    2022

  18. [26]

    2023, A&A, 669, A19

    Lardo, C., Salaris, M., Cassisi, S., et al. 2023, A&A, 669, A19

    2023

  19. [27]

    2025, A&A, 694, A248

    Latour, M., Kamann, S., Martocchia, S., et al. 2025, A&A, 694, A248

    2025

  20. [28]

    J., Johnson, C

    Lee, Y .-W., Kim, J. J., Johnson, C. I., et al. 2019, ApJ, 878, L2

    2019

  21. [29]

    V ., Milone, A

    Legnardi, M. V ., Milone, A. P., Armillotta, L., et al. 2022, MNRAS, 513, 735

    2022

  22. [30]

    2012, MNRAS, 427, 50

    Lind, K., Bergemann, M., & Asplund, M. 2012, MNRAS, 427, 50

    2012

  23. [31]

    2023, The Messenger, 190, 13

    Lucatello, S., Bragaglia, A., Vallenari, A., et al. 2023, The Messenger, 190, 13

    2023

  24. [32]

    2023, arXiv e-prints, arXiv:2312.08270

    Magrini, L., Bensby, T., Brucalassi, A., et al. 2023, arXiv e-prints, arXiv:2312.08270

    2023 arXiv

  25. [33]

    F., Milone, A

    Marino, A. F., Milone, A. P., Dondoglio, E., et al. 2023, ApJ, 958, 31

    2023

  26. [34]

    F., Milone, A

    Marino, A. F., Milone, A. P., Sills, A., et al. 2019, ApJ, 887, 91 Article number, page 11 of 12 A&A proofs:manuscript no. aa60891-26 Fig. 9.Differential abundance spreads for the full sample (upper panels), FG stars (middle panels), and SG stars (lower panels), shown as a fun...

    2019

  27. [35]

    L., Smolinski, J

    Martell, S. L., Smolinski, J. P., Beers, T. C., & Grebel, E. K. 2011, A&A, 534, A136

    2011

  28. [36]

    H., & Helmi, A

    Massari, D., Koppelman, H. H., & Helmi, A. 2019, A&A, 630, L4

    2019

  29. [37]

    F., et al

    McKenzie, M., Yong, D., Marino, A. F., et al. 2022, MNRAS, 516, 3515 Mészáros, S., Martell, S. L., Shetrone, M., et al. 2015, AJ, 149, 153 Mészáros, S., Masseron, T., García-Hernández, D. A., et al. 2020, MNRAS, 492, 1641

    2022

  30. [38]

    P., Bedin, L

    Milone, A. P., Bedin, L. R., Piotto, G., et al. 2008, ApJ, 673, 241

    2008

  31. [39]

    P., Piotto, G., Renzini, A., et al

    Milone, A. P., Piotto, G., Renzini, A., et al. 2017, MNRAS, 464, 3636

    2017

  32. [40]

    F., et al

    Monty, S., Yong, D., Marino, A. F., et al. 2023, MNRAS, 518, 965

    2023

  33. [41]

    R., & Lanzoni, B

    Mucciarelli, A., Lapenna, E., Massari, D., Ferraro, F. R., & Lanzoni, B. 2015, ApJ, 801, 69

    2015

  34. [42]

    2017, A&A, 601, A112 Ramírez, I., Meléndez, J., & Asplund, M

    Pancino, E., Romano, D., Tang, B., et al. 2017, A&A, 601, A112 Ramírez, I., Meléndez, J., & Asplund, M. 2009, A&A, 508, L17 Ramírez, I., Meléndez, J., Bean, J., et al. 2014, A&A, 572, A48

    2017

  35. [43]

    2015, A&A, 579, A104

    Sbordone, L., Monaco, L., Moni Bidin, C., et al. 2015, A&A, 579, A104

    2015

  36. [44]

    & Lucatello, S

    Schiappacasse-Ulloa, J. & Lucatello, S. 2023, MNRAS, 520, 5938

    2023

  37. [45]

    2024, A&A, 685, A10

    Schiappacasse-Ulloa, J., Lucatello, S., Cescutti, G., & Carretta, E. 2024, A&A, 685, A10

    2024

  38. [46]

    J., & Pietrinferni, A

    Schiappacasse-Ulloa, J., Lucatello, S., Rain, M. J., & Pietrinferni, A. 2022, MN- RAS, 511, 231

    2022

  39. [47]

    2025, A&A, 699, A41

    Schiappacasse-Ulloa, J., Magrini, L., Lucatello, S., et al. 2025, A&A, 699, A41

    2025

  40. [48]

    P., Phillips, S

    Schiavon, R. P., Phillips, S. G., Myers, N., et al. 2024, MNRAS, 528, 1393

    2024

  41. [49]

    & Zinn, R

    Searle, L. & Zinn, R. 1978, ApJ, 225, 357

    1978

  42. [50]

    I., Filler, D., et al

    Simmerer, J., Ivans, I. I., Filler, D., et al. 2013, ApJ, 764, L7

    2013

  43. [51]

    1973, ApJ, 184, 839

    Sneden, C. 1973, ApJ, 184, 839

    1973

  44. [52]

    G., Santos, N

    Sousa, S. G., Santos, N. C., Israelian, G., Mayor, M., & Monteiro, M. J. P. F. G. 2007, A&A, 469, 783 Tautvaišien˙e, G., Drazdauskas, A., Bragaglia, A., et al. 2022, A&A, 658, A80

    2007

  45. [53]

    & Baumgardt, H

    Vasiliev, E. & Baumgardt, H. 2021, MNRAS, 505, 5978

    2021

  46. [54]

    2013, MNRAS, 434, 3542

    Yong, D., Meléndez, J., Grundahl, F., et al. 2013, MNRAS, 434, 3542

    2013

  47. [55]

    1985, ApJ, 293, 424 Article number, page 12 of 12

    Zinn, R. 1985, ApJ, 293, 424 Article number, page 12 of 12

    1985

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.