REVIEW 3 major objections 7 minor 29 references
NuSTAR Observations of the Galaxy Cluster Abell 3667
T0 review · 3 major / 7 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read A 208 ks NuSTAR observation of Abell 3667's central region finds that the hard X-ray excess is primarily thermal, best fit by a two-temperature plasma with a hot component at kT = 25.8 keV, rather than non-thermal inverse Compton emission.
desk verdict A solid first NuSTAR look at Abell 3667 that finds the hard excess prefers a two-temperature thermal model over 1T+IC; the statistical case is decent but needs a clearer statement of how often the background realizations actually favor 2T. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central objects are the two competing spectral models: a two-temperature (2T) thermal plasma model built from APEC collisional-plasma components, and a single-temperature plus power-law (1T+IC) model representing inverse Compton emission from relativistic electrons. The discriminating evidence is the C-statistic (W-statistic) fit comparison over the full 4-80 keV band, combined with the NuSTAR background model nuskybgd and its systematic uncertainty realizations, which account for the background-dominated regime above 20 keV.
What would settle it
A reanalysis of the same NuSTAR spectra using an independently constructed background, such as blank-sky observations with a different background model, that shifts the C-stat difference in favor of the 1T+IC model would directly falsify the central claim; alternatively, a spatially resolved map showing that the 26 keV component originates from an unresolved point source rather than diffuse cluster gas would overturn the thermal interpretation.
Extended reading notes
Core claim
The central claim is that the hard X-ray excess in Abell 3667's core is thermal rather than non-thermal. Fitting the combined FPMA/FPMB NuSTAR spectra in the 4-80 keV band, the authors find that a two-temperature APEC model, with $kT_1 = 5.8^{+0.3}_{-0.3}$ keV and $kT_2 = 25.8^{+6.1}_{-2.1}$ keV, gives C-stat/dof $= 4211.2/3691$, while a single-temperature model with an additional power-law inverse Compton component gives C-stat/dof $= 4230.1/3691$. The lower C-stat for the 2T model indicates a statistically better fit, and the preference holds in nearly all of 1000 background realizations. The paper also rules out the brightest point source in the field as the origin of the excess and derives a lower limit on the magnetic field of about 0.2 microgauss.
Load-bearing premise
The conclusion rests on the accuracy of the NuSTAR background model above about 10 keV, where the spectra are background-dominated and where the 2T versus 1T+IC discrimination is decided; if the background or its systematic uncertainties are misestimated, the preference for the two-temperature model could be an artifact.
Editorial extensions
If this is right
- The central hard X-ray excess in Abell 3667 can be explained without a population of relativistic electrons in the core, removing the need for non-thermal inverse Compton emission there.
- A hot thermal component at roughly 26 keV implies that merger-induced heating can raise intracluster gas to unusually high temperatures, with consequences for the energy budget of merging clusters.
- The lower limit of about 0.2 microgauss on the central magnetic field is consistent with values in other dynamically active clusters and with equipartition and Faraday rotation estimates.
- Earlier non-thermal interpretations of the Abell 3667 hard excess from BeppoSAX, RXTE, and Suzaku are disfavored, at least for the cluster core.
- The outskirts near the radio relics may still host non-thermal emission, so the thermal-versus-IC question is not fully closed for the whole cluster.
Reading between the lines
- If the hot component is genuinely thermal, a spatially resolved spectral map could locate it at the merger shock front, a testable prediction with deeper NuSTAR observations or a future high-resolution X-ray calorimeter.
- The comparison uses a single power-law for the non-thermal alternative; a more flexible model, such as a broken power law or inverse Compton emission from a realistic electron energy distribution, could alter the C-stat balance.
- The magnetic field limit is computed from the full radio bridge flux, which extends beyond the NuSTAR field; a spatially matched radio measurement would likely lower the inferred field, so the 0.2 microgauss value may be an optimistic bound for the central region.
- The 2T model's hot 26 keV component could represent a blend of multiple temperature phases or non-equilibrium plasma rather than a single uniform hotspot, which future line-sensitive observations could distinguish.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript presents an analysis of deep NuSTAR observations (208 ks) of the central 4.5-arcmin region of the galaxy cluster Abell 3667. The authors fit the joint FPMA/FPMB spectra with a single-temperature (1T) APEC model, a two-temperature (2T) APEC model, and a single-temperature plus power-law (1T+IC) model. They report that the 2T model provides the best fit (C-stat 4211.2 for 3691 dof), with a hot component at kT = 25.8(+6.1/−2.1) keV, compared to the 1T+IC model (C-stat 4230.1 for the same dof). They interpret the hard X-ray excess as primarily thermal, likely from merger-induced heating, and derive a lower limit on the magnetic field strength of ~0.2 μG using the 1T+IC model's 20–80 keV flux and the radio flux of the bridge between the two radio relics.
Significance. If the thermal interpretation holds, this result strengthens the emerging NuSTAR-based view that many reported non-thermal hard X-ray excesses in clusters are instead thermal in origin, and it provides a new constraint on the hot gas phase in a well-known merging system. The analysis uses the largest NuSTAR exposure of Abell 3667 to date and includes a systematic test with 1000 background realizations, which is a commendable effort. The claimed preference for the 2T model over 1T+IC, however, is not accompanied by a formal significance test for non-nested models, and the magnetic field limit is conditional on the disfavored non-thermal model. These issues make the central claim moderately supported but in need of further quantitative justification. The authors also appropriately acknowledge the limitation that the observations cover only the central region, leaving the outer relics unexplored.
major comments (3)
- [Section 3, Figure 3] The central claim that the 2T model is preferred over the 1T+IC model rests on ΔC = 4230.1 − 4211.2 = 18.9 for equal dof (3691), but the two models are non-nested and no significance test (e.g., AIC, BIC, or a bootstrap p-value) is reported. The paper instead relies on Fig. 3, which shows a distribution of ΔC from 1000 background realizations, and states that 'in almost all cases' the 2T model is preferred, but does not quantify the fraction of realizations favoring each model or the width of the distribution. If, say, 5–10% of realizations favor the 1T+IC model, the observed ΔC would not be robust. Please report the number/percentage of realizations with ΔC>0, the median and 1σ spread of the distribution, and, ideally, a formal model-selection statistic calibrated for non-nested models.
- [Section 4.2] The lower limit on the magnetic field strength (B ∼ 0.2 μG) is derived from the 20–80 keV flux of the 1T+IC model (Table 1) using the radio-to-X-ray flux ratio. However, the paper's central conclusion is that the 2T thermal model is preferred over 1T+IC; under that interpretation, the non-thermal flux from the 1T+IC model is not a physical component. The B-field limit is therefore conditional on the disfavored model, and this tension should be acknowledged explicitly. The authors do note the condition 'under the assumption that this emission is produced by the same population of relativistic electrons...' but the text could be clearer that this assumption contradicts the preferred thermal interpretation. I recommend adding a sentence stating that the B-field limit is valid only if a non-thermal component exists, and is provided for comparison with previous work.
- [Section 3, background modeling] Above 20 keV the spectra are background-dominated, and the 2T vs 1T+IC discrimination depends on the spectral shape of the excess in that band. The background model (nuskybgd) is assigned systematic uncertainties of 3% (instrumental), 8% (aperture), 42% (focused cosmic), and 10% (solar) in the 1000-realization test, but the paper does not show how the fitted models and ΔC respond to plausible variations in the background normalization or shape, e.g., by fitting with the focused-cosmic or aperture components scaled to their systematic limits. A figure showing the background-subtracted spectra with systematic error envelopes (similar to the approach in Wik et al. 2014) would make the robustness of the ΔC=18.9 preference more transparent.
minor comments (7)
- [Throughout] The text contains numerous OCR-like artifacts, including 'NuST ARobservations', 'Datauction', 'NuST AR', and 'NUST ARDAS'; please proofread the manuscript carefully.
- [References] The reference list includes the placeholder line 'By default, this template uses biblatex and adopts the Chicago referencing style...' which is an artifact of the LaTeX template and should be removed.
- [Section 3] The sentence 'We initially allowed the redshift to vary freely, then fixed it to the value determined from the fitting...' is ambiguous; please report the adopted redshift value and clarify whether the redshift was fixed to the best-fit value or to the known cluster redshift after a calibration offset.
- [Section 3] The text cites 'Appendix B for Rojas Bolivar et al. 2023' but this manuscript has no appendix; either add the appendix or remove the citation.
- [Figure 3] The x-axis label is 'Cstat (C1T + IC C2T)' which appears incomplete; please label it as ΔC = C(1T+IC) − C(2T).
- [Table 1] The statistical and systematic uncertainties are presented as superscripts without explanation; please add a note in the table caption clarifying the notation (e.g., the first error bar is statistical, the second is systematic).
- [Section 4.2] The radio flux density used (44 ± 6 mJy at 2.3 GHz) is for the entire radio bridge, which extends beyond the NuSTAR field of view. The paper notes this caveat, but it would be helpful to also state the expected effect on B if a spatially matched radio flux were used (the text says it would reduce the field strength, but does not quantify the change).
Circularity Check
No significant circularity; the 2T-over-1T+IC preference is an empirical model comparison, not a self-derived result.
full rationale
The paper's central claim is that a two-temperature thermal model fits the NuSTAR spectra of Abell 3667 better than a single-temperature-plus-inverse-Compton model (C-stat 4211.2 vs 4230.1 for the same 3691 dof). This is a direct, data-driven model comparison using the observed spectra, the external nuskybgd background description, and standard XSPEC fitting; it does not reduce to any prior conclusion of the authors. The 1000-realization background systematic test is described in the text with the specific systematic fractions and is run on the present data, so the self-citation to Mirakhor et al. (2022) is for a methodological precedent only and is not load-bearing. The magnetic-field lower limit is a conditional calculation from the standard radio-to-X-ray flux ratio (Govoni and Feretti 2004), using the fitted 1T+IC flux and an external radio flux; it is not a prediction derived from the paper's own assumptions. The paper itself flags the caveat that the adopted radio flux covers the whole bridge, so a spatially matched lower radio flux would reduce the inferred field, which is a limitation on the constraint's interpretation but not a circularity. No equation or fitted parameter is renamed as an independent prediction, and no uniqueness claim is imported from the authors' prior work. The derivation chain is self-contained against the NuSTAR data and external references.
Assumptions & free parameters
free parameters (7)
- Redshift calibration offset =
not reported (fixed after free fit)
- T1 (cool APEC temperature) =
5.8 +0.3/-0.3 keV
- T2 (hot APEC temperature) =
25.8 +6.1/-2.1 keV
- Metal abundance Z =
0.24 +/- 0.02 Z_sun
- APEC normalizations (FPMA/FPMB) =
Norm1 ~2.09e-2, Norm2 ~6.2e-3 cm^-5 (Table 1)
- Power-law photon index Gamma =
1.6 +0.1/-0.2
- IC flux (20-80 keV) =
9.52 +0.46/-5.51 e-12 erg/s/cm2
assumptions (5)
- domain assumption APEC collisional plasma model accurately describes the thermal ICM in 4-30 keV.
- domain assumption nuskybgd background model and its systematic uncertainties are accurate.
- domain assumption Galactic absorption column is fixed at NH = 3.67e20 cm^-2.
- domain assumption Radio synchrotron and IC X-ray emissions arise from the same power-law electron population; the bridge radio flux approximates the NuSTAR region.
- domain assumption The brightest Chandra point source's spectrum can be extrapolated as a single power law to 20-80 keV.
Cite this review
Pith. "Pith review of NuSTAR Observations of the Galaxy Cluster Abell 3667." pith.science (2026). https://pith.science/paper/7TV5NYCJ
@misc{pith2026250520453,
author = {Pith},
title = {Pith review of: NuSTAR Observations of the Galaxy Cluster Abell 3667},
year = {2026},
howpublished = {\url{https://pith.science/paper/7TV5NYCJ}},
note = {Machine review of arXiv:2505.20453}
}
abstract
We present an analysis of the hard X-ray emission from the central region of Abell 3667 using deep NuSTAR observations. While previous studies on the nature of the hard X-ray excess have been controversial, our analysis of the central region suggests that the excess is primarily thermal, best described by a two-temperature (2T) model, with the high-temperature component likely arising from merger-induced heating. This interpretation contrasts with some earlier suggestions of non-thermal emission due to inverse Compton scattering of relativistic electrons. Additionally, we set a lower limit on the magnetic field strength of $\sim 0.2 \, \mu$G in the central region, consistent with values found in other dynamically active clusters and compatible with those inferred from equipartition and Faraday rotation measurements. Since our study is focused on the central region of the cluster, further high-resolution observations of the outer regions will be critical to fully disentangle the thermal and non-thermal contributions to the X-ray.
Figures
Reference graph
Works this paper leans on
-
[3]
Astronomy and Astrophysics 513 (April): A30
The Coma cluster magnetic field from Faraday rotation measures. Astronomy and Astrophysics 513 (April): A30. https://doi.org/10.1051/0004-6361/200913696. arXiv: 1002.0594 [astro-ph.CO]. Carilli, C. L., and G. B. Taylor
-
[7]
Rojas Bolivar, Randall A., Daniel R
arXiv: 2012.00236 [astro-ph.HE]. Rojas Bolivar, Randall A., Daniel R. Wik, Ayşegül Tümer, Fabio Gastaldello, Julie Hlavacek-Larrondo, Paul Nulsen, Valentina Vacca, et al
arXiv 2012
-
[16]
https://doi.org/10.1086/165529. Rephaeli, Y., J. Nevalainen, T. Ohashi, and A. M. Bykov
-
[19]
The Astrophysical Journal 606, no
Spectral Analysis of Rossi X-Ray Timing Explorer Observations of A3667. The Astrophysical Journal 606, no. 2 (May): 825–828. https://doi.org/10.1086/383123. arXiv: astro- ph/0401312 [astro-ph]. Rephaeli, Y oel, Duane Gruber, and Philip Blanco
-
[20]
The Astrophysical Journal 511, no
Rossi X-Ray Timing Explorer Observations of the Coma Cluster. The Astrophysical Journal 511, no. 1 (January): L21–L24. https://doi.org/10.1086/311828. Rojas Bolivar, Randall A., Daniel R. Wik, Simona Giacintucci, Fabio Gastaldello, Allan Hornstrup, Niels-Jorgen Westergaard, and Grzegorz Madejski
-
[24]
The Astrophysical Journall 556 (August): L91–L95
Collisional Plasma Models with APEC/APED: Emission-Line Diagnos- tics of Hydrogen-like and Helium-like Ions. The Astrophysical Journall 556 (August): L91–L95. https : / / doi . org / 10 . 1086 / 322992. eprint: arXiv:astro-ph/0106478. Tümer, Aysegul, Daniel R. Wik, Xiaoyuan Zhang, Duy N. Hoang, Mas- simo Gaspari, Reinout J. Weeren, Lawrence Rudnick, et al
-
[25]
The NuSTAR and Chandra view of CL 0217+70 and Its Tell-Tale Radio Halo
The NuSTAR and Chandra view of CL 0217+70 and Its Tell-Tale Ra- dio Halo. arXiv e-prints (June): arXiv:2206.09228. arXiv: 2206.09228 [astro-ph.HE]. van Weeren, R. J., F. de Gasperin, H. Akamatsu, M. Brüggen, L. Feretti, H. Kang, A. Stroe, and F. Zandanel
-
[26]
arXiv: 1901.04496 [astro-ph.HE]
https://doi.org/10.1007/s11214- 019- 0584- z. arXiv: 1901.04496 [astro-ph.HE]. Wik, Daniel R., A. Hornstrup, S. Molendi, G. Madejski, F. A. Harrison, A. Zoglauer, B. W. Grefenstette, et al
arXiv 1901
Show all 29 references
-
[29]
Madsen, Kristin K., Brian W
https://doi.org/10.1086/168737. Madsen, Kristin K., Brian W. Grefenstette, Sean Pike, Hiromasa Miyasaka, Murray Brightman, Karl Forster, and Fiona A. Harrison
- [30]
-
[48]
arXiv: 1403.2722 [astro-ph.HE]
https://doi.org/10.1088/ 0004-637X/792/1/48. arXiv: 1403.2722 [astro-ph.HE]. Wik, Daniel R., Craig L. Sarazin, Yu-Ying Zhang, Wayne H. Baumgartner, Richard F. Mushotzky, Jack Tueller, Takashi Okajima, and Tracy E. Clarke
-
[67]
arXiv: 1207.0506 [astro-ph.CO]
https://doi.org/10.1088/0004-637X/748/ 1/67. arXiv: 1207.0506 [astro-ph.CO]. ZuHone, J. A., M. Markevitch, and D. Lee
- [87]
-
[133]
8 Mohammad S
https://doi.org/10.1086/166729. 8 Mohammad S. Mirakhor et al. Rephaeli, Y oel, and Duane Gruber
-
[139]
arXiv: 1411.1573 [astro-ph.HE]
https://doi.org/10.1088/0004-637X/800/2/139. arXiv: 1411.1573 [astro-ph.HE]. Govoni, Federica, and Luigina Feretti
-
[1979]
https://doi.org/10.1086/156740
Relativistic electrons in the intracluster space of clusters of galaxies: the hard X-ray spectra and heating of the gas.The Astrophysical Journal 227 (January): 364–369. https://doi.org/10.1086/156740. Rephaeli, Y., D. E. Gruber, and R. E. Rothschild
-
[1999]
The Astrophysical Journall 513, no
Hard X-Ray Radiation in the Coma Cluster Spec- trum. The Astrophysical Journall 513, no. 1 (March): L21–L24. https: //doi.org/10.1086/311902. arXiv: astro-ph/9901018 [astro-ph]. Fusco-Femiano, Roberto, Daniele Dal Fiume, Mauro Orlandini, Gianfranco Brunetti, Luigina Feretti, a...
-
[2000]
The Astrophysical Journal 533, no
Nonthermal Bremsstrahlung and Hard X-Ray Emission from Clusters of Galaxies. The Astrophysical Journal 533, no. 1 (April): 73–83. https://doi.org/10.1086/308649. arXiv: astro-ph/9911335 [astro-ph]. Smith, R. K., N. S. Brickhouse, D. A. Liedahl, and J. C. Raymond
-
[2001]
The Astrophysical Journall 552, no
Hard X-Ray Emission from the Galaxy Cluster A3667. The Astrophysical Journall 552, no. 2 (May): L97–L100. https://doi.org/10.1086/320340. arXiv: astro-ph/0105049 [astro-ph]. Gastaldello, Fabio, Daniel R. Wik, S. Molendi, N. J. Westergaard, A. Horn- strup, G. Madejski, D. D. M....
-
[2002]
https://doi.org/ 10.1146/annurev.astro.40.060401.093852
Cluster Magnetic Fields.Annual Review of Astronomy and Astrophysics 40 (January): 319–348. https://doi.org/ 10.1146/annurev.astro.40.060401.093852. arXiv: astro-ph/0110655 [astro-ph]. Carretti, E., S. Brown, L. Staveley-Smith, J. M. Malarecki, G. Bernardi, B. M. Gaensler, M. H...
-
[2004]
International Journal of Modern Physics D 13, no
Magnetic Fields in Clusters of Galaxies. International Journal of Modern Physics D 13, no. 8 (January): 1549–1594. https : / / doi . org / 10 . 1142 / S0218271804005080. arXiv: astro-ph/0410182 [astro-ph]. Kim, K. -T., P. P. Kronberg, P. E. Dewdney, and T. L. Landecker
-
[2008]
Space Science Reviews 134, nos
Nonthermal Phenomena in Clusters of Galaxies. Space Science Reviews 134, nos. 1-4 (February): 71–92. https://doi.org/10.1007/s11214-008-9314-7. arXiv: 0801.0982 [astro-ph]. Rephaeli, Y oel, and D. E. Gruber
-
[2009]
The Astrophysical Journal 690, no
Galaxy Clusters in the Swift/Burst Alert Telescope Era: Hard X-rays in the Intracluster Medium. The Astrophysical Journal 690, no. 1 (January): 367–388. https://doi.org/ 10.1088/0004-637X/690/1/367. arXiv: 0809.0006 [astro-ph]. Ajello, M., P. Rebusco, N. Cappelluti, O. Reimer,...
-
[2010]
Galaxy Clusters in the Swift/BA T Era. II. 10 More Clusters Detected Above 15 keV. The Astrophysical Journal 725, no. 2 (December): 1688–1706. https://doi.org/10.1088/0004- 637X/725/2/1688. arXiv: 1009.4699 [astro-ph.CO]. Arnaud, K. A
-
[2013]
2 (April): 1414–1422
Detection of a radio bridge in Abell 3667.Monthly Notices of the Royal Astronomical Society 430, no. 2 (April): 1414–1422. https://doi.org/10.1093/mnras/ stt002. arXiv: 1205.1082 [astro-ph.CO]. Cova, F., F. Gastaldello, D. R. Wik, W. Boschin, A. Botteon, G. Brunetti, D. A. Buo...
-
[2014]
Astronomy and Astrophysics 562 (Febru- ary): A60
Investigating the hard X-ray emission from the hottest Abell cluster A2163 with Suzaku. Astronomy and Astrophysics 562 (Febru- ary): A60. https://doi.org/10.1051/0004- 6361/201322878. arXiv: 1312.4120 [astro-ph.HE]. Rephaeli, Y
-
[2019]
Astronomy and Astrophysics 628 (August): A83
A joint XMM-NuSTAR observation of the galaxy cluster Abell 523: Constraints on inverse Compton emission. Astronomy and Astrophysics 628 (August): A83. https://doi.org/10.1051/0004- 6361/201834644. arXiv: 1906.07730 [astro-ph.HE]. de Gasperin, F., L. Rudnick, A. Finoguenov, D. ...
1906 arXiv
-
[2020]
arXiv e-prints (May): arXiv:2005.00569
NuSTAR low energy effective area correction due to thermal blanket tear. arXiv e-prints (May): arXiv:2005.00569. arXiv: 2005.00569 [astro-ph.IM]. Mirakhor, MS, SA Walker, J Runge, and P Diwanji
2005 arXiv
-
[2022]
Astronomy and Astrophysics 659 (March): A146
MeerKA T view of the diffuse radio sources in Abell 3667 and their interactions with the thermal plasma. Astronomy and Astrophysics 659 (March): A146. https://doi.org/10.1051/ 0004-6361/202142658. arXiv: 2111.06940 [astro-ph.CO]. Edge, Alastair Clousten, GC Stewart, AC Fabian,...
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.