REVIEW 2 major objections 5 minor 1 cited by
Absence of TeV halos around millisecond pulsars
T0 review · 2 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash
Pith's one-line read Millisecond pulsars show no TeV halos in a stacked search of 2565 days of HAWC data.
desk verdict Solid HAWC non-detection that convincingly refutes earlier tentative MSP-halo claims; main caveat is the Geminga-sized template, which is not stress-tested in the extended direction. 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 load-bearing machinery is a stacking maximum-likelihood search. Each MSP is modeled as a bidimensional Gaussian halo with extension $\sigma_{\rm source} = (d_{\rm Geminga}/d_{\rm source})\,2.0^\circ$, a power-law spectrum of index 2.7, and a common normalization $K$ shared by all sources; source-dependent weights $C_i$ set the relative flux either from spin-down flux or GeV flux. The log-likelihood profiles of all sources are summed to constrain the common efficiency $\eta$, and the resulting stacked test statistic is compared against a Monte Carlo background built from random sky positions that satisfy the same selection criteria. This machinery lets a population of individually undetectable halos be tested jointly, and converts a null result into upper limits on halo efficiency.
What would settle it
Observing one millisecond pulsar with a TeV halo whose efficiency matches Geminga's, or a stacked test statistic above the paper's $2\sigma$ background threshold of ${\rm TS}=3.99$ in a reanalysis with more exposure, would directly contradict the central claim.
Extended reading notes
Core claim
The central claim is that TeV halos around millisecond pulsars are absent, or at least far less efficient than those around isolated pulsars. The paper reaches this by fitting each of 57 MSPs with a Gaussian halo template whose angular size is scaled from Geminga's $2.0^\circ$ halo by distance, then adding the individual likelihood profiles with a common normalization for halo flux. In both weighting schemes (flux proportional to spin-down flux $L_{\rm sd}/(4\pi d^2)$ and flux proportional to Fermi 0.1-100 GeV flux $G_{100}$), the stacked test statistic matches background expectations in every energy bin from 0.32 to 100 TeV. The resulting 95% confidence upper limits on the efficiency $\eta = E^2\Phi/(L_{\rm sd}/4\pi d^2)$ or $\eta_{\rm GeV} = E^2\Phi/G_{100}$ lie below the efficiency band of detected TeV halos, particularly above 10 TeV. The paper also re-fits the four MSPs previously reported as tentative detections and finds all have $\sqrt{\rm TS}<2$ with the updated data, attributing the earlier excesses to stray Galactic plane emission.
Load-bearing premise
The analysis assumes every millisecond pulsar halo is a Gaussian of the same physical radius as Geminga's halo, with flux proportional either to spin-down power or to GeV brightness; if real MSP halos are smaller, larger, or differently shaped, the stacked search could miss them.
Editorial extensions
If this is right
- MSPs do not contribute significantly to TeV and higher-energy Galactic diffuse emission, leaving unresolved halos of isolated middle-aged pulsars as the viable leptonic explanation at a few TeV.
- The MSP interpretation of the Galactic Center GeV excess is strengthened, because MSPs can produce GeV emission without accompanying TeV halos, so previous TeV-halo constraints on that scenario are weakened.
- The four tentative MSP halo detections reported in earlier work are consistent with background fluctuations once more data and better event reconstruction are used.
- For individual bright MSPs, the current upper limits are approaching but still above the magnetospheric TeV emission models of PSR B1957+20 and PSR J2339-0533, implying that a 25-60% sensitivity improvement could lead to a detection.
Reading between the lines
- If MSPs genuinely lack the slow-diffusion regions that confine electrons around isolated pulsars, the confinement mechanism may require large-scale pulsar wind nebulae or long-lasting particle injection; a targeted search for small halos around the most energetic MSP binaries would test this.
- The null result sharpens the puzzle of what accelerates particles around middle-aged pulsars: MSPs have comparable GeV luminosity but apparently do not make TeV halos, so the efficiency difference may point to the wind nebula rather than the magnetosphere as the essential ingredient.
- A southern-sky survey with next-generation instruments could test whether the absence is intrinsic to MSPs or an artifact of the northern-sky HAWC sample, and could also probe MSP-rich globular clusters for faint TeV emission.
- The stacking technique of scaling a single halo template across many sources and summing likelihoods could be transferred to other faint source populations, such as radio-quiet pulsars or globular clusters, to constrain their TeV emission.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper presents a search for very-high-energy (TeV) gamma-ray halos around 57 millisecond pulsars (MSPs) using 2565 days of HAWC data. The authors find no significant emission from individual MSPs and, combining likelihood profiles under two weighting schemes (proportional to spin-down flux and to Fermi-LAT GeV flux), find that the stacked signal is consistent with background. They derive upper limits on a common halo efficiency and conclude that MSPs are less efficient than isolated pulsars in producing TeV halos, with implications for the Galactic Center GeV excess and the galactic diffuse emission. The analysis is careful in its treatment of background trials and systematic uncertainties, and it re-examines earlier tentative detections with more data.
Significance. If upheld, this is the first population-level constraint on TeV halos around MSPs and would weaken the case that unresolved MSPs explain the Galactic Center GeV excess via TeV counterparts. The paper is valuable for its null result, its explicit refutation of previous tentative detections, and its discussion of implications. The likelihood methodology is standard and the authors include Monte Carlo background distributions and systematic checks on spectral index, detector response, and source extension. However, the central comparative claim about halo efficiency rests on assumptions about the angular size of MSP halos that are not adequately tested.
major comments (2)
- [Methods, Eq. (1); Table IV] The upper limits on the halo efficiency in Table I and Figure 2 are derived from a Gaussian template whose angular size is fixed by Eq. (1) to the same physical radius as the Geminga halo. If MSP halos are physically more extended than Geminga's, the fitted normalization K is biased low by approximately (sigma_template/sigma_true)^2, so the reported limits underestimate the true total flux and efficiency. The 'Extension' systematic in Table IV replaces the per-source angular sizes with the sample mean (0.468 deg), which is not a coherent rescaling of the assumed physical radius and does not test the direction that matters; for nearby sources it actually shrinks the template. The Discussion explicitly suggests that MSPs may lack the slow-diffusion regions that confine isolated-pulsar halos, which would make MSP halos larger. The authors should present limits for a range of assumed physical radii (e.g., 0.5, 1, 2, and 4 times Geminga's radius) and assess whether the conclusion that MSPs are less efficient than isolated pulsars is robust, or alternatively qualify the abstract's unconditional claim.
- [Results, Table I; Eq. (3)] The stacking analysis assumes a single common efficiency eta for all MSPs, with source weights C_i fixed to the spin-down or GeV flux scaling. If only a subset of MSPs (e.g., those with favorable conditions) produce halos with efficiencies comparable to isolated pulsars, the stacked signal could be diluted and the upper limit on the common efficiency would not exclude such a population. The individual-source upper limits partially address this, but they are not combined into a statement about the fraction of MSPs that could be efficient. The authors should either quantify the sensitivity to a fraction f of MSPs contributing with efficiency eta, or restrict the conclusion accordingly.
minor comments (5)
- [Appendix: Distribution of the Stacked TS] The text says the background simulation is repeated 3093 times, while the caption of Figure 5 states the background distribution is composed of 11068 trials; these numbers are inconsistent and should be reconciled.
- [Methods: source selection] The source-count statement reports 57 sources selected from ATNF (53), 3PC (37, 34 overlap with ATNF), WVU (3), and LOTAAS (1); a direct sum gives 60 unique sources if the WVU and LOTAAS entries do not overlap with the others, so the overlapping membership of these catalogs should be clarified.
- [Methods: spectral index] The paper states the spectral index is set to p=2.7 and that its uncertainty is studied as a systematic, but it does not report the range of p values tested; please specify the tested values in the systematic uncertainty section.
- [Results: point-source stacking] The sentence beginning 'As the best fit TS of the two weight models over the full energy range is 2.43 and 5.04, which indicates that' is grammatically incomplete and should be rewritten.
- [Table II] The entry PSRJ1939+2134 lists RA=19.66 degrees, which is far from the pulsar's actual right ascension; this appears to be a typo and should be corrected.
Circularity Check
No significant circularity: the MSP halo efficiency limits are fitted outputs of the likelihood analysis, and the Geminga-scaled template and weighting schemes are stated input assumptions rather than results derived from the target conclusion.
full rationale
The derivation chain is: select MSPs; model each halo as a Gaussian with extension sigma_source = (d_Geminga/d_source) sigma_Geminga (Eq. 1); define the flux model Phi_i = K C_i (E/E_piv)^-p (Eq. 2); fit the common normalization K by stacking likelihoods; then convert K into efficiency limits via Eq. 3. The claimed result, that MSPs are not as efficient as isolated pulsars in producing TeV halos, is an output of that fit, not an input. K is a free parameter, and the upper limits in Tables I and III and Figures 2 and 6 are data-driven. The weighting schemes (C_i proportional to L_sd/d^2 or G100) use the same spin-down or GeV fluxes that appear in the efficiency denominator, but this is an explicitly stated scaling ansatz, not a definitional identity that forces the efficiency value; the fit could, in principle, have returned a detection or stronger signal. Equation (1) imports Geminga's measured angular size as a template scale; that is an external empirical input, and the paper treats the extension uncertainty in the systematic section by additionally fitting a common extension of 0.468 degrees. Whether real MSP halos are physically larger than Geminga's is a model-assumption or sensitivity question, not a circularity: changing the template changes the numerical limits, but even a systematically underestimated template would not make the derivation equivalent to its own conclusion. The prior tentative MSP halo claims [12,13] are not authored by the present authors, and the HAWC method papers cited are technical references rather than load-bearing self-citations. No step reduces by construction to its own inputs, and no fitted parameter is relabeled as a prediction.
Assumptions & free parameters
free parameters (3)
- Spectral index p =
2.7
- Halo angular extension sigma_source =
scaled from Geminga: 2.0 deg * d_Geminga/d_source
- Universal normalization K =
Best-fit consistent with 0; 95% C.L. upper limits in Table I
assumptions (5)
- domain assumption TeV halos have Gaussian spatial profile
- domain assumption MSP halo physical size equals Geminga's
- domain assumption Halo flux scales with spin-down power or GeV flux
- ad hoc to paper Source selection criteria are appropriate
- domain assumption HAWC detector response and background model are correct
Cite this review
Pith. "Pith review of Absence of TeV halos around millisecond pulsars." pith.science (2026). https://pith.science/paper/FLDC7L6E
@misc{pith2026250500184,
author = {Pith},
title = {Pith review of: Absence of TeV halos around millisecond pulsars},
year = {2026},
howpublished = {\url{https://pith.science/paper/FLDC7L6E}},
note = {Machine review of arXiv:2505.00184}
}
read the original abstract
TeV halos are extended very-high-energy (VHE; 0.1-100 TeV) gamma-ray emission around middle-aged pulsars. So far they have only been found around isolated pulsars, but it has been suggested that they may also be powered by millisecond pulsars (MSPs). We searched for VHE gamma-ray emission from MSPs reported by radio and GeV gamma-ray observatories in 2565 days of data from the High Altitude Water Cherenkov (HAWC) Observatory. We found no significant emission from individual pulsars. By combining the likelihood profiles of all MSPs accessible to HAWC, our analysis suggests that the excess emission around the MSP population is consistent with a background. Our result suggests that MSPs are not as efficient as isolated pulsars in producing TeV halos. This finding has strong implications on the physics interpretation of the Galactic Center GeV excess and high-latitude Galactic diffuse emission.
Figures
Figures from the paper (5 more)
Forward citations
Cited by 1 Pith paper
-
GeV-TeV Connections in Galaxies: Evolutionary Signatures from Pulsars in Globular Clusters
Globular cluster pulsar winds may make a significant contribution to GeV and TeV gamma-ray emission from massive quiescent galaxies, with strength tied to each galaxy's evolutionary history.
Reference graph
Works this paper leans on
-
[1]
A. U. Abeysekara et al. , Science 358, 911 (2017), arXiv:1711.06223 [astro-ph.HE]
arXiv 2017
-
[2]
A. Albert et al. , Astrophys. J. 905, 76 (2020), arXiv:2007.08582 [astro-ph.HE]
arXiv 2020
-
[3]
F. Aharonian et al., Phys. Rev. Lett. 126, 241103 (2021), arXiv:2106.09396 [astro-ph.HE]
arXiv 2021
-
[4]
at least 2 ◦ from any known sources in TeVCat [33] 3 FIG. 1. The HAWC significance map computed using Pass 5.1 data in the Galactic coordinate. The MSPs studied our analysis are marked as yellow circles. The color bar indicates the test statistic of a point source at a given pixel assuming a power-law spectrum with an index of 2.7. and the 3rd HAWC catalo...
- [5]
-
[6]
M. Di Mauro, S. Manconi, and F. Donato, Phys. Rev. D 101, 103035 (2020), arXiv:1908.03216 [astro-ph.HE]
arXiv 2020
-
[7]
A. Albert et al. (HAWC), Astrophys. J. Lett. 944, L29 (2023), arXiv:2301.04646 [astro-ph.HE]
arXiv 2023
-
[8]
A. K. Harding, Astrophys. Space Sci. Libr. 465, 57 (2021), arXiv:2101.05751 [astro-ph.HE]
arXiv 2021
Show all 50 references
-
[9]
D. A. Smith et al. , Astrophys. J. (2023), 2307.11132 [astro-ph]
2023 arXiv
-
[10]
L´ opez-Coto, E
R. L´ opez-Coto, E. de O˜ na Wilhelmi, F. Aharonian, E. Amato, and J. Hinton, Nature Astron. 6, 199 (2022), arXiv:2202.06899 [astro-ph.HE]
2022 arXiv
-
[11]
further to tens of TeV energies. Previous works identified excess emission around sev- eral MSPs using the public HAWC data and suggested that the population may also power TeV halos at a 4 σ significance level [12, 13]. The analyses concluded that MSPs produce very-high-energ...
-
[12]
Hooper and T
D. Hooper and T. Linden, Phys. Rev. D 98, 043005 (2018)
2018
-
[13]
A. K. Harding, C. Venter, and C. Kalapotharakos, As- trophys. J. 923, 194 (2021), arXiv:2110.09412 [astro- ph.HE]
2021 arXiv
-
[14]
C. J. T. van der Merwe, Z. Wadiasingh, C. Venter, A. K. Harding, and M. G. Baring, Astrophys. J. 904, 91 (2020), arXiv:2010.01125 [astro-ph.HE]
2020 arXiv
-
[15]
Linden and B
T. Linden and B. J. Buckman, Phys. Rev. Lett. 120, 121101 (2018)
2018
-
[16]
Hooper and T
D. Hooper and T. Linden, Phys. Rev. D 105, 103013 (2022)
2022
-
[17]
Atkins et al
R. Atkins et al. (Milagro), Phys. Rev. Lett. 95, 251103 (2005), arXiv:astro-ph/0502303
2005 arXiv
-
[18]
Vecchiotti, F
V. Vecchiotti, F. Zuccarini, F. L. Villante, and G. Pagliaroli, Astrophys. J. 928, 19 (2022), arXiv:2107.14584 [astro-ph.HE]
2022 arXiv
-
[19]
Cao et al
Z. Cao et al. , Phys. Rev. Lett. 131, 151001 (2023), arXiv:2305.05372 [astro-ph.HE]
2023
-
[20]
Alfaro et al
R. Alfaro et al. (HAWC), Astrophys. J. (2023), arXiv:2310.09117 [astro-ph.HE]
2023
-
[21]
Dekker, I
A. Dekker, I. Holst, D. Hooper, G. Leone, E. Simon, and H. Xiao, arXiv e-prints , arXiv:2306.00051 (2023), arXiv:2306.00051 [astro-ph.HE]
2023 arXiv
-
[22]
K. Fang, J. S. Gallagher, and F. Halzen, Na- ture Astronomy (2023), 10.1038/s41550-023-02128-0, arXiv:2306.17275 [astro-ph.HE]
2023 arXiv
-
[23]
Yan, R.-Y
K. Yan, R.-Y. Liu, R. Zhang, C.-M. Li, Q. Yuan, and X.-Y. Wang, arXiv e-prints , arXiv:2307.12363 (2023), arXiv:2307.12363 [astro-ph.HE]
2023 arXiv
-
[24]
Hooper and L
D. Hooper and L. Goodenough, Phys. Lett. B 697, 412 (2011), arXiv:1010.2752 [hep-ph]
2011 arXiv
-
[25]
Abbasi et al
R. Abbasi et al. (IceCube), Science 380, adc9818 (2023), arXiv:2307.04427 [astro-ph.HE]
2023
-
[26]
Fang and K
K. Fang and K. Murase, ApJ Letters 957, L6 (2023), arXiv:2307.02905 [astro-ph.HE]
2023 arXiv
-
[27]
or astrophysical reasons, specifically, a group of unre- solved MSPs [28, 29]. [12, 13] show that if MSPs generate TeV halos with Geminga-like efficiency, then the MSPs contributing to the GeV-scale excess should also produce TeV halo emission via inverse Compton radiation. Th...
-
[28]
Calore, I
F. Calore, I. Cholis, and C. Weniger, JCAP 03, 038 6 (2015), arXiv:1409.0042 [astro-ph.CO]
2015 arXiv
-
[29]
Ackermann et al
M. Ackermann et al. (Fermi-LAT), Astrophys. J.840, 43 (2017), arXiv:1704.03910 [astro-ph.HE]
2017 arXiv
- [30]
-
[31]
K. N. Abazajian and M. Kaplinghat, Phys. Rev. D 86, 083511 (2012), [Erratum: Phys.Rev.D 87, 129902 (2013)], arXiv:1207.6047 [astro-ph.HE]
2012 arXiv
-
[32]
Bartels, S
R. Bartels, S. Krishnamurthy, and C. Weniger, Phys. Rev. Lett. 116, 051102 (2016), arXiv:1506.05104 [astro- ph.HE]
2016 arXiv
-
[33]
A. U. Abeysekara et al. , Astrophys. J. 843, 39 (2017), arXiv:1701.01778 [astro-ph.HE]
2017 arXiv
-
[34]
Galactic millisecond pul- sars,
E. Ferrara and D. Lorimer, “Galactic millisecond pul- sars,” (2024)
2024
-
[35]
C. M. Tan, C. G. Bassa, S. Cooper, J. W. T. Hessels, V. I. Kondratiev, D. Michilli, S. Sanidas, B. W. Stappers, J. van Leeuwen, J. Y. Donner, J.- M. Grießmeier, M. Kramer, C. Tiburzi, P. Weltevrede, B. Ciardi, M. Hoeft, G. Mann, A. Miskolczi, D. J. Schwarz, C. Vocks, and O. Wu...
2020 arXiv
-
[36]
S. P. Wakely and D. Horan, Proceedings of the 30th Inter- national Cosmic Ray Conference (2008), aDS Bibcode: 2008ICRC....3.1341W
2008
-
[37]
Vianello, R
G. Vianello, R. J. Lauer, J. M. Burgess, H. Ay- ala, H. Fleischhack, P. Harding, M. Hui, S. Marinelli, V. Savchenko, and H. Zhou, in Proceedings of 7th Inter- national Fermi Symposium - PoS(IFS2017) (Sissa Medi- alab, Garmisch-Partenkirchen, Germany, 2017) p. 130
2017
-
[38]
A. U. Abeysekara et al., in Proceedings of 37th Interna- tional Cosmic Ray Conference — PoS(ICRC2021) , Vol. 395 (SISSA Medialab, 2022) p. 828
2022
-
[39]
A. U. Abeysekara et al., Astrophys. J. 881, 134 (2019)
2019
-
[40]
Linden, K
T. Linden, K. Auchettl, J. Bramante, I. Cholis, K. Fang, D. Hooper, T. Karwal, and S. W. Li, Phys. Rev. D 96, 103016 (2017)
2017
-
[41]
A. U. Abeysekara et al. , Astrophys. J. 843, 40 (2017), arXiv:1702.02992 [astro-ph.HE]
2017 arXiv
-
[42]
Brisbois, C
C. Brisbois, C. Riviere, H. Fleischhack, and A. Smith, The Astronomer’s Telegram 12013, 1 (2018), aDS Bib- code: 2018ATel12013....1B
2018
-
[43]
Riviere, H
C. Riviere, H. Fleischhack, and A. Sandoval, The As- tronomer’s Telegram 10941, 1 (2017), aDS Bibcode: 2017ATel10941....1R
2017
-
[44]
Aharonian et al., Physical Review Letters 126, 241103 (2021), aDS Bibcode: 2021PhRvL.126x1103A
F. Aharonian et al., Physical Review Letters 126, 241103 (2021), aDS Bibcode: 2021PhRvL.126x1103A
2021
-
[45]
Evoli, T
C. Evoli, T. Linden, and G. Morlino, Phys. Rev. D 98, 063017 (2018), arXiv:1807.09263 [astro-ph.HE]
2018 arXiv
-
[46]
Fang, X.-J
K. Fang, X.-J. Bi, and P.-F. Yin, MNRAS 488, 4074 (2019), arXiv:1903.06421 [astro-ph.HE]
2019 arXiv
-
[47]
Mukhopadhyay and T
P. Mukhopadhyay and T. Linden, Phys. Rev. D 105, 123008 (2022), arXiv:2111.01143 [astro-ph.HE]
2022 arXiv
- [48]
- [49]
-
[50]
C. J. T. van der Merwe, Z. Wadiasingh, C. Venter, A. K. Harding, and M. G. Baring, Astrophys. J. 904, 91 (2020). 7 Appendices Properties of the source sample Figure 3 presents the spin-down power and distances of the MSPs in our sample. The colored markers indi- cate the gamma...
2020
Reviewed August 16, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.