pith. machine review for the scientific record. sign in

arxiv: 2604.02814 · v1 · submitted 2026-04-03 · 🌌 astro-ph.CO · astro-ph.GA

Recognition: 1 theorem link

· Lean Theorem

Inferring population III star properties from the 21-cm global signal

Sho Ukai , Hayato Shimabukuro , Kenji Hasegawa , Kiyotomo Ichiki
Authors on Pith no claims yet

Pith reviewed 2026-05-13 18:36 UTC · model grok-4.3

classification 🌌 astro-ph.CO astro-ph.GA
keywords population III stars21-cm signalreionizationFisher analysisstar formation efficiencyescape fractioncosmic dawnradio astronomy
0
0 comments X

The pith

The 21-cm global signal can constrain the typical mass and star formation efficiency of Population III stars.

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

The paper explores whether measurements of the 21-cm global signal during cosmic reionization can reveal properties of the first stars. It uses semi-numerical simulations that model how these stars ionize and heat surrounding gas, with the escape fraction of photons depending on stellar and halo mass, and heating profiles taken from radiation hydrodynamics runs. Applying Fisher analysis to future observations while including foreground contamination, the work shows that instruments like REACH could place meaningful limits on the average mass and efficiency of star formation in these early halos. This matters because direct detection of Population III stars is extremely difficult, so indirect probes like the 21-cm line offer one of the few ways to test models of the first stellar generation.

Core claim

By incorporating a mass-dependent escape fraction of ionizing photons and heating structures derived from radiation-hydrodynamics simulations into semi-numerical models of the 21-cm signal, Fisher-matrix forecasts indicate that observations with the Radio Experiment for the Analysis of Cosmic Hydrogen (REACH) can provide meaningful constraints on the typical mass and star-formation efficiency of Population III stars even after accounting for foreground emissions.

What carries the argument

Fisher analysis of the 21-cm global signal from semi-numerical simulations that include mass-dependent ionizing-photon escape fractions and RHD-derived heating structures around halos.

Load-bearing premise

The heating structure around each halo is taken directly from separate radiation-hydrodynamics simulations without re-validation inside the semi-numerical code, and the escape fraction follows a fixed dependence on stellar and halo mass.

What would settle it

If REACH or similar 21-cm experiments measure a global signal whose shape and amplitude cannot be reproduced by varying the Population III mass and efficiency within the allowed ranges after foreground removal, the claim of meaningful constraints would be falsified.

Figures

Figures reproduced from arXiv: 2604.02814 by Hayato Shimabukuro, Kenji Hasegawa, Kiyotomo Ichiki, Sho Ukai.

Figure 1
Figure 1. Figure 1: FIG. 1. Escape fraction–halo mass relation. Relationship [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. The global 21-cm brightness temperatures as func [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. The star formation rate density as a function of red [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. The minimum halo mass, [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. The ionized fraction as a function of redshift. The [PITH_FULL_IMAGE:figures/full_fig_p007_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. The expected 1 [PITH_FULL_IMAGE:figures/full_fig_p008_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10. The parameter derivatives of the global 21-cm [PITH_FULL_IMAGE:figures/full_fig_p009_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: FIG. 11. Same as Figure 9, but for the case [PITH_FULL_IMAGE:figures/full_fig_p010_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: FIG. 12. Estimated noise for the global 21-cm measurement [PITH_FULL_IMAGE:figures/full_fig_p011_12.png] view at source ↗
read the original abstract

Investigating the properties of the first stars in the universe is essential, yet it remains an open question. One way to explore these stars is by examining their effects on the surrounding gas during the epoch of reionization. In this study, we investigate whether the 21-cm global signal can constrain the typical mass and star formation efficiency of first-generation stars. We perform semi-numerical simulations that include the escape fraction of ionizing photons, which depends on stellar and halo masses, as well as the heating structure surrounding a halo that hosts the first star, determined by radiation hydrodynamics (RHD) simulations. By applying Fisher analysis, while accounting for foreground emissions, we demonstrate that future observations with instruments such as the Radio Experiment for the Analysis of Cosmic Hydrogen (REACH) could provide meaningful constraints on these properties.

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 / 2 minor

Summary. The manuscript presents semi-numerical simulations of the 21-cm global signal that incorporate a stellar- and halo-mass-dependent escape fraction for ionizing photons together with heating profiles imported from radiation-hydrodynamics (RHD) runs. Fisher-matrix forecasting that includes foreground emission is then used to argue that future global-signal observations with REACH can deliver meaningful constraints on the typical mass and star-formation efficiency of Population III stars.

Significance. If the underlying thermal and ionization modeling is shown to be internally consistent, the work supplies a concrete forecast that links an observable (the global 21-cm signal) to two key Pop-III parameters. The explicit inclusion of foregrounds and the use of Fisher analysis are standard and useful for experiment planning; the paper therefore has moderate significance for the 21-cm cosmology community.

major comments (2)
  1. [Section 3] Section 3 (semi-numerical simulation description): the heating structure around each halo is taken directly from separate RHD simulations and inserted into the semi-numerical code without any re-validation or residual comparison against the code’s own ionization, recombination, and halo-mass-dependent escape treatment. Because the global-signal amplitude and timing are exponentially sensitive to IGM temperature evolution, any mismatch scales the Fisher-matrix elements and the resulting error bars on Pop-III mass and star-formation efficiency.
  2. [Section 4] Section 4 (Fisher analysis): no quantitative error budget is provided for foreground-subtraction uncertainties, nor is an explicit validation of the semi-numerical code against full RHD simulations reported. These omissions are load-bearing for the central claim that REACH observations “could provide meaningful constraints.”
minor comments (2)
  1. [Abstract] Abstract: the claim of “meaningful constraints” would be strengthened by quoting the forecasted 1-sigma uncertainties on stellar mass and efficiency that emerge from the Fisher matrix.
  2. [Section 2] Notation: the functional form and numerical coefficients of the escape-fraction prescription should be written explicitly (with reference to the RHD source) rather than described only qualitatively.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and constructive feedback. The comments highlight important aspects of our hybrid modeling approach and the robustness of the Fisher forecasts. We address each major comment below and have revised the manuscript accordingly to improve clarity and transparency.

read point-by-point responses

  1. Referee: [Section 3] Section 3 (semi-numerical simulation description): the heating structure around each halo is taken directly from separate RHD simulations and inserted into the semi-numerical code without any re-validation or residual comparison against the code’s own ionization, recombination, and halo-mass-dependent escape treatment. Because the global-signal amplitude and timing are exponentially sensitive to IGM temperature evolution, any mismatch scales the Fisher-matrix elements and the resulting error bars on Pop-III mass and star-formation efficiency.

    Authors: We agree that ensuring consistency between the imported RHD heating profiles and the semi-numerical ionization/recombination treatment is essential given the sensitivity of the global signal. The RHD simulations supply high-resolution local thermal and ionization structures around individual halos that are difficult to capture self-consistently in large-volume semi-numerical codes, while our code applies the same stellar- and halo-mass-dependent escape fraction for the global evolution. We have added a new paragraph in Section 3 explicitly discussing this hybrid methodology, its assumptions, and the expected level of residual mismatch based on comparisons in the literature. A full re-simulation of the RHD profiles within our exact semi-numerical framework is not feasible due to computational cost, but we note that the localized nature of the heating limits propagation of small inconsistencies to the global signal. We have also included a brief sensitivity test in the revised manuscript showing that moderate variations in the heating profile do not qualitatively alter the Fisher constraints. revision: partial

  2. Referee: [Section 4] Section 4 (Fisher analysis): no quantitative error budget is provided for foreground-subtraction uncertainties, nor is an explicit validation of the semi-numerical code against full RHD simulations reported. These omissions are load-bearing for the central claim that REACH observations “could provide meaningful constraints.”

    Authors: We acknowledge that a more explicit quantitative error budget for foreground residuals would strengthen the presentation. In the revised manuscript we have added a dedicated subsection in Section 4 (and a new appendix) that quantifies the impact of residual foregrounds using conservative estimates drawn from REACH instrument papers and standard 21-cm foreground modeling (e.g., 10–20% residual power after subtraction). This is folded into the Fisher matrix as an additional systematic covariance term, and we show that the resulting degradation in constraints on Pop-III mass and star-formation efficiency remains modest. Regarding explicit validation against full RHD, we have clarified in the text that our approach follows established hybrid methods in the field; a complete cosmological-volume RHD validation is currently intractable. We have revised the abstract and conclusions to state that the reported constraints are meaningful within the adopted modeling framework and to emphasize the model-dependent nature of the forecasts. revision: yes

Circularity Check

0 steps flagged

No circularity: forward-model forecasting with external inputs

full rationale

The paper constructs a semi-numerical 21-cm model that imports heating profiles from separate RHD simulations and imposes a mass-dependent escape-fraction prescription; it then runs Fisher-matrix forecasts for REACH-like observations. No equation or step defines a target quantity in terms of itself, renames a fitted parameter as a prediction, or reduces the central claim to a self-citation chain. The derivation chain remains independent of the output constraints, satisfying the criteria for a non-circular forecasting study.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The forecast rests on the validity of the semi-numerical code, the accuracy of the imported RHD heating profiles, and the assumed functional form of the escape fraction; none of these are independently verified inside the paper.

axioms (2)
  • domain assumption The escape fraction of ionizing photons depends on stellar and halo masses in a manner that can be parameterized for semi-numerical use.
    Invoked to link stellar properties to reionization and heating.
  • domain assumption Heating structure around each halo is accurately captured by separate radiation-hydrodynamics simulations.
    Used to set the thermal state of the IGM in the semi-numerical runs.

pith-pipeline@v0.9.0 · 5444 in / 1415 out tokens · 40427 ms · 2026-05-13T18:36:45.524199+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

  • IndisputableMonolith/Foundation/RealityFromDistinction reality_from_one_distinction unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    We perform semi-numerical simulations that include the escape fraction of ionizing photons, which depends on stellar and halo masses, as well as the heating structure surrounding a halo that hosts the first star, determined by radiation hydrodynamics (RHD) simulations.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

52 extracted references · 52 canonical work pages

  1. [1]

    J. D. Bowman et al.Nature, 555(7694):67–70, Mar. 2018. doi:10.1038/nature25792

    work page doi:10.1038/nature25792 2018

  2. [2]

    2011, Annual Review of Astronomy and Astrophysics, 49, 373, doi:10.1146/annurev-astro-081710-102608

    V. Bromm and N. Yoshida.ARA&A, 49(1):373–407, 13 Sept. 2011. doi:10.1146/annurev-astro-081710-102608

    work page doi:10.1146/annurev-astro-081710-102608 2011

  3. [3]

    J. O. Burns et al.Advances in Space Research, 49(3): 433–450, Feb. 2012. doi:10.1016/j.asr.2011.10.014

    work page doi:10.1016/j.asr.2011.10.014 2012

  4. [4]

    Cullen et al.MNRAS, 540(3):2176–2194, July 2025

    F. Cullen et al.MNRAS, 540(3):2176–2194, July 2025. doi:10.1093/mnras/staf838

    work page doi:10.1093/mnras/staf838 2025

  5. [5]

    de Lera Acedo et al.Nature Astronomy, 6:984–998, July 2022

    E. de Lera Acedo et al.Nature Astronomy, 6:984–998, July 2022. doi:10.1038/s41550-022-01709-9

    work page doi:10.1038/s41550-022-01709-9 2022

  6. [6]

    G. B. Field.ApJ, 129:536, May 1959. doi: 10.1086/146653

    work page doi:10.1086/146653 1959

  7. [7]

    S. R. Furlanetto, S. P. Oh, and F. H. Briggs. Phys. Rep., 433(4-6):181–301, Oct. 2006. doi: 10.1016/j.physrep.2006.08.002

    work page doi:10.1016/j.physrep.2006.08.002 2006

  8. [8]

    Gessey-Jones et al.MNRAS, 516(1):841–860, Oct

    T. Gessey-Jones et al.MNRAS, 516(1):841–860, Oct

    work page

  9. [9]

    doi:10.1093/mnras/stac2049

    work page doi:10.1093/mnras/stac2049

  10. [10]

    Gessey-Jones et al.Nature Astronomy, 9:1268–1279, Aug

    T. Gessey-Jones et al.Nature Astronomy, 9:1268–1279, Aug. 2025. doi:10.1038/s41550-025-02575-x

    work page doi:10.1038/s41550-025-02575-x 2025

  11. [11]

    Hartwig et al.ApJ, 936(1):45, Sept

    T. Hartwig et al.ApJ, 936(1):45, Sept. 2022. doi: 10.3847/1538-4357/ac7150

    work page doi:10.3847/1538-4357/ac7150 2022

  12. [12]

    Hasegawa et al.arXiv e-prints, art

    K. Hasegawa et al.arXiv e-prints, art. arXiv:1603.01961, Mar. 2016. doi:10.48550/arXiv.1603.01961

    work page doi:10.48550/arxiv.1603.01961 2016

  13. [13]

    Hills, G

    R. Hills, G. Kulkarni, P. D. Meerburg, and E. Puch- wein.Nature, 564(7736):E32–E34, Dec. 2018. doi: 10.1038/s41586-018-0796-5

    work page doi:10.1038/s41586-018-0796-5 2018

  14. [14]

    Hirano et al.Monthly Notices of the Royal Astronom- ical Society, 448(1):568–587, 02 2015

    S. Hirano et al.Monthly Notices of the Royal Astronom- ical Society, 448(1):568–587, 02 2015. ISSN 0035-8711. doi:10.1093/mnras/stv044. URLhttps://doi.org/10. 1093/mnras/stv044

    work page doi:10.1093/mnras/stv044 2015

  15. [15]

    Hosokawa et al.The Astrophysical Journal, 824(2): 119, jun 2016

    T. Hosokawa et al.The Astrophysical Journal, 824(2): 119, jun 2016. doi:10.3847/0004-637X/824/2/119. URL https://dx.doi.org/10.3847/0004-637X/824/2/119

    work page doi:10.3847/0004-637x/824/2/119 2016

  16. [16]

    Hui and N

    L. Hui and N. Y. Gnedin.MNRAS, 292(1):27–42, Nov

    work page

  17. [17]

    doi:10.1093/mnras/292.1.27

    work page doi:10.1093/mnras/292.1.27

  18. [18]

    Ishiyama and S

    T. Ishiyama and S. Hirano.arXiv e- prints, art. arXiv:2501.17540, Jan. 2025. doi: 10.48550/arXiv.2501.17540

    work page doi:10.48550/arxiv.2501.17540 2025

  19. [19]

    Jaacks, R

    J. Jaacks, R. Thompson, S. L. Finkelstein, and V. Bromm.MNRAS, 475(4):4396–4410, Apr. 2018. doi: 10.1093/mnras/sty062

    work page doi:10.1093/mnras/sty062 2018

  20. [20]

    Jester and H

    S. Jester and H. Falcke.New A Rev., 53(1-2):1–26, May

    work page

  21. [21]

    doi:10.1016/j.newar.2009.02.001

    work page doi:10.1016/j.newar.2009.02.001 2009

  22. [22]

    W., Navarro J

    X. Jin et al.ApJ, 942(2):59, Jan. 2023. doi:10.3847/1538- 4357/aca678

    work page doi:10.3847/1538- 2023

  23. [23]

    2013, Reviews of Modern Physics, 85, 809, doi:10.1103/RevModPhys.85.809

    T. Karlsson, V. Bromm, and J. Bland-Hawthorn. Rev. Mod. Phys., 85:809–848, May 2013. doi: 10.1103/RevModPhys.85.809. URLhttps://link.aps. org/doi/10.1103/RevModPhys.85.809

    work page doi:10.1103/revmodphys.85.809 2013

  24. [24]

    R. S. Klessen and S. C. O. Glover.ARA&A, 61:65–130, Aug. 2023. doi:10.1146/annurev-astro-071221-053453

    work page doi:10.1146/annurev-astro-071221-053453 2023

  25. [25]

    Konno et al.ApJ, 797(1):16, Dec

    A. Konno et al.ApJ, 797(1):16, Dec. 2014. doi: 10.1088/0004-637X/797/1/16

    work page doi:10.1088/0004-637x/797/1/16 2014

  26. [26]

    Konno et al.PASJ, 70:S16, Jan

    A. Konno et al.PASJ, 70:S16, Jan. 2018. doi: 10.1093/pasj/psx131

    work page doi:10.1093/pasj/psx131 2018

  27. [27]

    Koopmans et al

    L. Koopmans et al. InAdvancing Astrophysics with the Square Kilometre Array (AASKA14), page 1, Apr. 2015. doi:10.22323/1.215.0001

    work page doi:10.22323/1.215.0001 2015

  28. [28]

    A. Liu, J. Pritchard, M. Tegmark, and A. Loeb. InAmerican Astronomical Society Meeting Abstracts #221, volume 221 ofAmerican Astronomical Soci- ety Meeting Abstracts, page 229.05, Jan. 2013. doi: 10.48550/arXiv.1211.3743

    work page doi:10.48550/arxiv.1211.3743 2013

  29. [29]

    X.-W. Liu, C. Heneka, and L. Amendola.J. Cos- mology Astropart. Phys., 2020(5):038, May 2020. doi: 10.1088/1475-7516/2020/05/038

    work page doi:10.1088/1475-7516/2020/05/038 2020

  30. [30]

    2024, A&A, 687, A67, doi: 10.1051/0004-6361/202347087

    R. Maiolino et al.A&A, 687:A67, July 2024. doi: 10.1051/0004-6361/202347087

    work page doi:10.1051/0004-6361/202347087 2024

  31. [31]

    I. D. McGreer, A. Mesinger, and V. D’Odorico.MNRAS, 447(1):499–505, Feb. 2015. doi:10.1093/mnras/stu2449

    work page doi:10.1093/mnras/stu2449 2015

  32. [32]

    Mirocha et al.MNRAS, 478(4):5591–5606, Aug

    J. Mirocha et al.MNRAS, 478(4):5591–5606, Aug. 2018. doi:10.1093/mnras/sty1388

    work page doi:10.1093/mnras/sty1388 2018

  33. [33]

    Ouchi et al.ApJ, 723(1):869–894, Nov

    M. Ouchi et al.ApJ, 723(1):869–894, Nov. 2010. doi: 10.1088/0004-637X/723/1/869

    work page doi:10.1088/0004-637x/723/1/869 2010

  34. [34]

    Planck Collaboration et al.A&A, 641:A6, Sept. 2020. doi:10.1051/0004-6361/201833910

    work page doi:10.1051/0004-6361/201833910 2020

  35. [35]

    J. R. Pritchard and A. Loeb.Phys. Rev. D, 82(2):023006, July 2010. doi:10.1103/PhysRevD.82.023006

    work page doi:10.1103/physrevd.82.023006 2010

  36. [36]

    J. R. Pritchard and A. Loeb.Reports on Progress in Physics, 75(8):086901, Aug. 2012. doi:10.1088/0034- 4885/75/8/086901

    work page doi:10.1088/0034- 2012

  37. [37]

    Sarmento, E

    R. Sarmento, E. Scannapieco, and B. Cˆ ot´ e.ApJ, 871(2): 206, Feb. 2019. doi:10.3847/1538-4357/aafa1a

    work page doi:10.3847/1538-4357/aafa1a 2019

  38. [38]

    R. K. Sheth, H. J. Mo, and G. Tormen.MNRAS, 323(1): 1–12, May 2001. doi:10.1046/j.1365-8711.2001.04006.x

    work page doi:10.1046/j.1365-8711.2001.04006.x 2001

  39. [39]

    Shimabukuro et al.PASJ, 75:S1–S32, Feb

    H. Shimabukuro et al.PASJ, 75:S1–S32, Feb. 2023. doi: 10.1093/pasj/psac042

    work page doi:10.1093/pasj/psac042 2023

  40. [40]

    Singh et al.Nature Astronomy, 6:607–617, Feb

    S. Singh et al.Nature Astronomy, 6:607–617, Feb. 2022. doi:10.1038/s41550-022-01610-5

    work page doi:10.1038/s41550-022-01610-5 2022

  41. [41]

    Sobacchi and A

    E. Sobacchi and A. Mesinger.MNRAS, 440(2):1662– 1673, May 2014. doi:10.1093/mnras/stu377

    work page doi:10.1093/mnras/stu377 2014

  42. [42]

    Spinelli, G

    M. Spinelli, G. Bernardi, and M. G. Santos.Monthly Notices of the Royal Astronomical Society, 489(3): 4007–4015, 09 2019. ISSN 0035-8711. doi: 10.1093/mnras/stz2425. URLhttps://doi.org/10. 1093/mnras/stz2425

    work page doi:10.1093/mnras/stz2425 2019

  43. [43]

    Tanaka and K

    T. Tanaka and K. Hasegawa.MNRAS, 502(1):463–471, Mar. 2021. doi:10.1093/mnras/stab072

    work page doi:10.1093/mnras/stab072 2021

  44. [44]

    Tanaka et al.MNRAS, 480(2):1925–1937, Oct

    T. Tanaka et al.MNRAS, 480(2):1925–1937, Oct. 2018. doi:10.1093/mnras/sty1967

    work page doi:10.1093/mnras/sty1967 1925

  45. [45]

    2023, A&A, 678, A173, doi: 10.1051/0004-6361/202346981

    E. Vanzella et al.A&A, 678:A173, Oct. 2023. doi: 10.1051/0004-6361/202346981

    work page doi:10.1051/0004-6361/202346981 2023

  46. [46]

    E. M. Ventura, Y. Qin, S. Balu, and J. S. B. Wyithe.MNRAS, 540(1):483–497, June 2025. doi: 10.1093/mnras/staf699

    work page doi:10.1093/mnras/staf699 2025

  47. [47]

    Visbal et al.MNRAS, 445(1):107–114, Nov

    E. Visbal et al.MNRAS, 445(1):107–114, Nov. 2014. doi: 10.1093/mnras/stu1710

    work page doi:10.1093/mnras/stu1710 2014

  48. [48]

    Visbal, G

    E. Visbal, G. L. Bryan, and Z. Haiman.ApJ, 897(1):95, July 2020. doi:10.3847/1538-4357/ab994e

    work page doi:10.3847/1538-4357/ab994e 2020

  49. [49]

    Wang et al.ApJ, 967(2):L42, June 2024

    X. Wang et al.ApJ, 967(2):L42, June 2024. doi: 10.3847/2041-8213/ad4ced

    work page doi:10.3847/2041-8213/ad4ced 2024

  50. [50]

    S. A. Wouthuysen.AJ, 57:31–32, Jan. 1952. doi: 10.1086/106661

    work page doi:10.1086/106661 1952

  51. [51]

    Yajima and Y

    H. Yajima and Y. Li.MNRAS, 445(4):3674–3684, Dec

    work page

  52. [52]

    doi:10.1093/mnras/stu1982

    work page doi:10.1093/mnras/stu1982