Massive star clusters in the gamma-ray sky: the role of HII regions

Elena Amato; Giada Peron; Giovanni Morlino; Stefano Gabici

REVIEW 3 major objections 5 minor 11 references

Massive star clusters in the gamma-ray sky: the role of HII regions

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Young massive star clusters embedded in HII regions can be detected in gamma rays before any supernova explodes, and Fermi-LAT data reveal four such regions emitting up to about 1 TeV.

desk verdict A compact proceedings paper that mostly repackages already-published results; the four HII-region follow-ups and the strong correlation are real but need the cited ApJ papers for full support. read the letter →

arxiv 2501.12767 v1 pith:GS7EGV6Y submitted 2025-01-22 astro-ph.HE

Giada Peron , Giovanni Morlino , Stefano Gabici , Elena Amato This is my paper · ORCID

classification astro-ph.HE

keywords starclustersHIIregionsgamma-rayastronomyFermi-LATcosmic-rayaccelerationstellarwindsunidentifiedsourcesWISEcatalog

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 sets out to show that very young massive star clusters, still wrapped in the gas clouds where they formed, are gamma-ray emitters in their own right. Because the clusters are younger than about 3 million years, no supernova has yet exploded in them, so any gamma-ray emission must come from particles accelerated by the collective winds of the massive stars themselves. Analyzing Fermi-LAT data, the authors find a highly significant positional correlation between unidentified gamma-ray sources and infrared-selected HII regions, with 127 matches versus about 50 expected by chance, and targeted analysis of four regions reveals emission up to roughly 1 TeV whose shape follows the infrared emission. If this reading is right, stellar winds alone can accelerate cosmic rays in dense environments, and young clusters contribute to the Galactic cosmic-ray budget before they ever produce supernova remnants.

What carries the argument

The analysis rests on two linked tools. The first is a catalog cross-correlation: unidentified sources from the Fermi-LAT 4FGL catalog are matched to the WISE catalog of HII regions, an infrared 22-micron sample that traces hot dust around massive young stars, and the significance of the number of matches is assessed against 1000 randomized catalogs through the statistic Sigma = (N_real - <N_sim>)/sigma_sim. The second is a Fermi-LAT residual analysis: the background model includes Galactic and extragalactic diffuse emission and 4FGL point sources; after optimization, the sources overlapping each candidate HII region are removed, and the leftover test-statistic maps are compared with WISE infrared contours. The physical mechanism carrying the argument is hadronic gamma-ray production, in which protons accelerated by stellar winds collide with the dense gas of the HII region, at densities of $10^{2}$ to $10^{3}$ $cm^{-3}$, and produce gamma rays.

What would settle it

Recomputing the residual maps with an updated Fermi source catalog or a revised interstellar emission model and finding that the four excesses disappear, or assigning any of them to a known pulsar or supernova remnant, would overturn the claimed association.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the four WISE HII regions G018.426+01.922, G028.746+03.458, G040.554+02.443, and G051.978+00.542 are genuine GeV gamma-ray sources: after the standard Fermi-LAT background, made of Galactic and extragalactic diffuse emission plus 4FGL point sources, is optimized and the catalog sources overlapping each region are removed, residual test-statistic maps show emission up to about 1 TeV that is spatially matched to the 22-micron infrared morphology of the HII regions. Because these clusters are younger than the supernova timescale, the authors attribute the emission to hadronic interactions of protons accelerated by stellar winds in gas with density of order $10^{2}$ to $10^{3}$ $cm^{-3}$. The accompanying catalog-wide correlation, with significance Sigma = 12 overall and greater than 35 in the inner Galaxy, is used to argue that the four targeted objects are not isolated curiosities but representatives of a population of pre-supernova cosmic-ray accelerators.

Load-bearing premise

The paper assumes that the 4FGL catalog plus the standard Galactic and extragalactic diffuse models completely describe everything in each field except the HII region, so the leftover gamma-ray signal is real emission rather than an artifact of an incomplete background model.

Editorial extensions

If this is right

If the four detections hold, young massive clusters are confirmed as pre-supernova particle accelerators, and the observed emission traces wind-accelerated protons interacting with dense gas.
The statistical correlation, with Sigma about 12 overall and above 35 in the inner Galaxy, implies that many unidentified Fermi sources coincident with HII regions are likely the same kind of stellar-wind accelerators, not chance alignments.
The measured GeV fluxes fall within the sensitivity of the planned ASTRI and CTAO arrays, so the next generation of Cherenkov telescopes can test whether the emission extends to TeV energies and probe the parent proton spectrum.
With acceleration efficiency around one percent, as found in the earlier Vela pilot study, the total power from such clusters may be large enough to contribute a meaningful fraction of the Galactic cosmic rays, supplementing supernova remnants at the high-energy end.

Reading between the lines

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

If the correlation generalizes, a sizable share of the currently unidentified Fermi-LAT sources in the inner Galaxy could be young stellar clusters, which would shift the inferred source population of the 4FGL catalog.
A natural test is to rank all matched HII regions by 22-micron flux and gas density and predict which should be the next brightest gamma-ray sources; the present paper selects by infrared brightness and size, so this ranking is not yet tested.
The same age-based selection could be applied to older clusters above about 3 million years to isolate when supernova remnants start to dominate; if emission persists or brightens past that age, the pre-supernova attribution would need revision.
Because the HII-region gas provides a target for hadronic collisions, neutrino production at comparable flux is expected from the same proton population, so a neutrino observation from one of the four directions would independently test the hadronic origin.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

Giada,

Quick read: this is a conference proceedings, not a full paper. Most of the content is a summary of Peron et al. 2024 (Nature Astronomy and ApJ 972, L22), which the authors cite openly. The genuinely new bit is the follow-up analysis of four WISE HII regions with Fermi-LAT, presented as residual TS maps and SEDs up to 1 TeV. The statistical correlation between WISE HII regions and unidentified 4FGL sources, with Sigma ~ 12 and >35 in the inner Galaxy, is a strong result and is well explained here.

What the paper does well: the selection criterion (young clusters, <3 Myr, so no SN contamination) is physically motivated, and the correlation method using Monte Carlo against random catalogs is sound. The residual maps do show emission that spatially matches the infrared tracers, which is suggestive. For a proceedings, the description is clear and the references point to the detailed analyses.

Soft spots, in proportion: the 1 TeV claim is load-bearing but rests on residual TS maps after removing 4FGL sources and the diffuse model. The text does not report a likelihood test that adds an extended source with the WISE template, nor per-bin significances for the highest-energy points. All four regions sit at |b| <= 3.5 deg in the inner Galaxy, where diffuse mismodeling and source confusion are known issues. So the emission could in principle be an artifact of the leftover background. That said, this is a proceedings—the full details are in the referenced papers, and the strain does not necessarily mean the result is wrong. It means this text alone cannot independently confirm the 1 TeV part of the claim. The selection function for the follow-up targets is also not quantified, but again, the proceedings format is compressed.

Who this is for: someone wanting a quick overview of the star-cluster/HII-region gamma-ray connection, or a citable pointer to the team's broader work. The paper itself does not deserve a serious referee as a standalone research article—its value is as a summary—but the underlying science does merit attention. If the authors submit this as a full paper, it needs the missing likelihood analysis and a fuller background treatment.

My call: if this crossed my desk as a proceedings contribution, I'd let it through without fuss. If it were submitted as a regular paper, I'd send it to review but with the expectation that the 1 TeV claim be backed by an extended-template test. For your own work, cite the ApJ paper, not this proceedings.

Best, [Your name]

Referee Report

3 major / 5 minor

Summary. The paper reports a Fermi-LAT analysis of four WISE HII regions associated with young massive star clusters, claiming gamma-ray emission up to 1 TeV from G018.426+01.922, G028.746+03.458, G040.554+02.443, and G051.978+00.542. The detection is based on residual TS maps obtained after removing nearby 4FGL sources from the background model, and the spectral energy distributions are extracted with the source extension fixed to the WISE size. The paper also summarizes a statistical correlation between unidentified Fermi-LAT sources and WISE HII regions, reporting a global significance of Σ=12 and Σ>35 in the inner Galaxy. The authors conclude that these young systems are pre-supernova cosmic-ray accelerators and that future CTAO/ASTRI observations can test the emission at TeV energies.

Significance. If the claimed 1 TeV emission is confirmed, it would provide strong evidence that collective stellar winds in very young massive star clusters accelerate particles before any supernova has occurred, with direct implications for the contribution of star clusters to Galactic cosmic rays. The statistical correlation between HII regions and unidentified Fermi sources is a useful quantitative step toward resolving source confusion. The paper also benefits from explicitly stating the young ages (< 3 Myr) of the targets, which cleanly separates the stellar-wind component from supernova remnants, and the Monte Carlo approach to evaluate the correlation is appropriate. However, the current manuscript does not yet present the statistical evidence needed to fully support the central 1 TeV claim, as detailed in the major comments.

major comments (3)

[Section 3, Fig. 2] The claim of emission up to 1 TeV rests entirely on residual TS maps after subtracting 4FGL sources and a diffuse model. The manuscript never reports the test statistic for adding an extended source with the WISE template to the full background model, nor the per-energy-bin significance of the highest-energy spectral points. In the crowded inner Galaxy (|b| ≤ 3.5°, l ~ 18–52°), the interstellar emission model is known to have systematics and 4FGL is incomplete or confused, so the residual TS peaks could be artifacts of mismodeled diffuse emission or unremoved point sources. The paper itself hedges in Section 4 ('Further study would be needed to confirm whether this type of sources could be seen at higher energies'), which underscores that the present evidence is not conclusive. A likelihood-ratio test comparing the full model with and without the WISE-extended source, together with energy-resolved TS maps, is required to support the detection claim.
[Section 3, Fig. 3] The spectral energy distribution is extracted with the source extension fixed to the WISE size, so the SED cannot independently validate the morphological association. The claimed morphological match between gamma-ray residuals and infrared emission is only qualitative (visual comparison in Fig. 2). The paper should provide a quantitative morphology test, e.g., TS_ext comparing a point-like source against a source extended with the WISE shape, and report the best-fit position and extension with uncertainties. Without this, the statement that 'the morphology of the Fermi-LAT emission matches well with the infrared emission' is not tested against alternative hypotheses.
[Section 2, Fig. 1] The Monte Carlo significance calculation for the correlation between unidentified Fermi sources and WISE HII regions is described too briefly to assess the validity of the reported Σ values, especially for the longitude-sliced analysis. It is unclear whether the 1000 simulated catalogs with 'random extraction in l and b' preserve the actual source density or the longitude-dependent distribution when slices of 60° are considered. If the simulation draws uniformly over the whole sky, the high Σ > 35 in the inner Galaxy could reflect the higher real source density there rather than a physical association. The manuscript should specify the simulation procedure for the slice test, or explicitly refer to the detailed derivation in Peron et al. (2024) with the necessary information.

minor comments (5)

[Section 3, Fig. 2 caption] The caption describes 'light-blue circles' for the HII region extensions, but the text in Section 3 refers to 'red circles' for the 4FGL source localization uncertainties; the color coding in the figure and text should be made consistent.
[Table 1] The column header contains a typo: '4FGL souces' should be '4FGL sources'. Also, the table does not give uncertainties on the extension and 22 μm flux, which would be helpful for judging the precision of the association.
[Abstract] The abstract uses 'we will present' for the statistical investigation, but the results are already presented in Section 2; the tense should be adjusted to reflect the actual content.
[References] There are two 'Peron et al. 2024' entries with different coauthor lists (Nature Astronomy and ApJ 972, L22); the in-text citations do not distinguish them. The authors should label them 2024a and 2024b or use the author abbreviations consistently.
[Section 3] The Fermi-LAT analysis details (version of the fermitools, event selection, energy range, and the specific diffuse models used) are not stated. For a proceedings paper this may be acceptable, but a reference to the detailed analysis in Peron et al. (2024) would improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the gamma-ray detection is a new Fermi-LAT residual analysis anchored to external catalogs and simulations, not a repackaging of its inputs.

full rationale

The paper's derivation chain is: (1) statistically correlate WISE HII regions with unidentified 4FGL sources using 1000 randomized catalogs; (2) select bright, resolved HII regions with overlapping Fermi sources; (3) perform a standard Fermi-LAT likelihood fit with diffuse templates and 4FGL point sources, remove the overlapping catalog sources, and produce residual TS maps; (4) use the WISE extension only as a spatial template for SED extraction; (5) claim emission up to 1 TeV in the residuals. The central detection does not reduce to its inputs by construction: the residual maps in Fig. 2 are generated without any WISE-template source, so the morphological coincidence is an output, not an input. The SED extraction uses the WISE extension as an assumed spatial model, but the normalization is free and the detection is against a no-source null hypothesis, so the measured excess is not equivalent to the assumed template. Reliance on prior work by the same authors (Peron et al. 2024; Celli & Peron 2024) concerns the HII-region catalog correlation, acceleration efficiency, and future-telescope sensitivities; these are cited as published independent results and are not the source of the new residual detection. The manuscript's own caveat that further study is needed to confirm emission at higher energies reinforces that the TeV claim is a data-driven prospect rather than a tautology. No self-definitional, fitted-input-as-prediction, or unique-import-by-self-citation pattern is present.

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

The paper introduces no new free parameters or entities. It relies on the assumption that the HII regions are younger than about three million years and free of supernovae, that the WISE catalog traces these systems, that the Fermi-LAT background models are accurate, and that the gamma-ray emission is hadronic. These are domain assumptions adopted from prior literature rather than inputs fitted here.

assumptions (4)

domain assumption Young massive clusters younger than about 3 Myr have not yet hosted any supernova explosion, so the observed gamma-ray emission is entirely due to stellar winds.
Section 1 and 4: this premise excludes SNRs and pulsars as contributors, but the stellar population may include older field stars or runaway stars.
domain assumption The WISE catalog of HII regions (Anderson et al. 2014) is complete and traces the same structures that emit gamma rays.
Section 2: the correlation and target selection rely on WISE 22 micron detections as positional and morphological tracers.
domain assumption The Fermi-LAT Galactic and extragalactic diffuse emission models and the 4FGL catalog correctly describe the sky in the analyzed regions.
Section 3: residual TS maps after model subtraction are interpreted as real sources; any catalog or background error creates spurious residuals.
domain assumption The gamma-ray emission is produced by hadronic interactions of accelerated protons with the dense gas in the HII regions.
Section 1 and 4: the hadronic origin is used to infer particle acceleration, but the spectral and morphological evidence is not shown in this paper.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Massive star clusters in the gamma-ray sky: the role of HII regions." pith.science (2026). https://pith.science/paper/GS7EGV6Y

@misc{pith2026250112767,
  author       = {Pith},
  title        = {Pith review of: Massive star clusters in the gamma-ray sky: the role of HII regions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GS7EGV6Y}},
  note         = {Machine review of arXiv:2501.12767}
}

read the original abstract

Massive Star Clusters (SCs) have been proposed as important CR sources, with the potential of explaining the high-energy end of the Galactic cosmic-ray (CR) spectrum, that Supernova Remnants (SNRs) seem unable to account for. Thanks to fast mass losses due to the collective stellar winds, the environment around SCs is potentially suitable for particle acceleration up to PeV energies and the energetics is enough to account for a large fraction of the Galactic CRs, if the system is efficient enough. A handful of star clusters have been detected in gamma-rays confirming the idea that particle acceleration is taking place in this environment. However, contamination by other sources often makes it difficult to constrain the contribution arising from SCs only. Here we present a new analysis of Fermi-LAT data collected towards a few massive young star clusters. The young age (< 3 Myr) of the clusters guarantees that no SN has exploded in the region, allowing us to determine the power contributed by the stellar component alone, and to quantify the contribution of this type of sources to the bulk of CRs. Moreover, we will present a recent statistical investigation that quantifies the degree of correlation between gamma-ray sources and these astrophysical objects and briefly discuss the observational prospect for ASTRI and CTAO.

Figures

Figures reproduced from arXiv: 2501.12767 by the authors.

**Figure 1.** Statistical significance of associations of Hii regions with Fermi-LAT unidentified sources, as a function of their galactic longitude (red errorbars). The histogram underneath shows the distribution in longitude of the considered Hii regions (Anderson et al. 2014) and Fermi-LAT unidentified sources (Abdollahi et al. 2022). Both distributions are normalized [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 2.** Test statistic maps obtained from Fermi-LAT data after the optimization of the background sources and the removal of the sources potentially associated to HII regions. The light-blue circles represent the extension of the HII regions in the field, while the contours represent their 22-µm emission as traced by WISE [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Spectral energy distribution of the analyzed Hii regions compared to the sensitivity towards extended sources of the future ASTRI-mini array and CTAO facilities. The sensitivity curves are calculated as in Celli & Peron (2024) [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

11 extracted references · 7 canonical work pages

[1]

, " * write output.state after.block = add.period write newline

ENTRY address archiveprefix author booktitle chapter edition editor howpublished institution eprint journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.all := #1 ...
[2]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in " " * FUNCTION format....
[3]

2022, The Astrophysical Journal Supplement Series, 260

Abdollahi, S., Acero, F., Baldini, L., et al. 2022, The Astrophysical Journal Supplement Series, 260

work page 2022
[4]

D., Bania, T

Anderson, L. D., Bania, T. M., Balser, D. S., et al. 2014, Astrophysical Journal, Supplement Series, 212, 1

work page 2014
[5]

& Peron, G

Celli, S. & Peron, G. 2024

work page 2024
[6]

2024, , 686, A118

Celli , S., Specovius , A., Menchiari , S., Mitchell , A., & Morlino , G. 2024, , 686, A118

2024
[7]

2023, arXiv e-prints, arXiv:2301.06505

Gabici , S. 2023, arXiv e-prints, arXiv:2301.06505

arXiv 2023
[8]

2021, Monthly Notices of the Royal Astronomical Society, 504, 6096

Morlino, G., Blasi, P., Peretti, E., & Cristofari, P. 2021, Monthly Notices of the Royal Astronomical Society, 504, 6096

work page 2021

Show all 11 references

[9]

2024, Nature Astronomy

Peron, G., Casanova, S., Gabici, S., Baghmanyan, V., & Aharonian, F. 2024, Nature Astronomy

2024
[10]

2024, , 972, L22

Peron , G., Morlino , G., Gabici , S., et al. 2024, , 972, L22

2024
[11]

2018, Journal of the Korean Astronomical Society, 51, 37

Seo, J., Kang, H., & Ryu, D. 2018, Journal of the Korean Astronomical Society, 51, 37

2018

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

, " * write output.state after.block = add.period write newline

ENTRY address archiveprefix author booktitle chapter edition editor howpublished institution eprint journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.all := #1 ...

[2] [2]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in " " * FUNCTION format....

[3] [3]

2022, The Astrophysical Journal Supplement Series, 260

Abdollahi, S., Acero, F., Baldini, L., et al. 2022, The Astrophysical Journal Supplement Series, 260

work page 2022

[4] [4]

D., Bania, T

Anderson, L. D., Bania, T. M., Balser, D. S., et al. 2014, Astrophysical Journal, Supplement Series, 212, 1

work page 2014

[5] [5]

& Peron, G

Celli, S. & Peron, G. 2024

work page 2024

[6] [6]

2024, , 686, A118

Celli , S., Specovius , A., Menchiari , S., Mitchell , A., & Morlino , G. 2024, , 686, A118

2024

[7] [7]

2023, arXiv e-prints, arXiv:2301.06505

Gabici , S. 2023, arXiv e-prints, arXiv:2301.06505

arXiv 2023

[8] [8]

2021, Monthly Notices of the Royal Astronomical Society, 504, 6096

Morlino, G., Blasi, P., Peretti, E., & Cristofari, P. 2021, Monthly Notices of the Royal Astronomical Society, 504, 6096

work page 2021