Pith. sign in

REVIEW 2 major objections 4 minor 26 references

Detection of sub-100 GeV gamma-ray pulsations from PSR B1706-44 with H.E.S.S

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read PSR B1706−44 is detected as a pulsed gamma-ray source in 28.3 hours of H.E.S.S. II CT5 monoscopic observations at 4.74σ significance, making it the fourth pulsar seen from the ground.

desk verdict A credible fourth ground-based pulsar detection with a clean spectral match to Fermi-LAT, but the abstract's ~70 GeV 'significant signal' claim is unsupported by the paper's own 2.5 sigma bin. read the letter →

arxiv 1908.06464 v3 pith:ZVYNFYNC submitted 2019-08-18 astro-ph.HE

Marion Spir-Jacob , Arache Djannati-Ataï , Lars Mohrmann , Gianluca Giavitto , Bruno Khélifi , Bronis{l}aw Rudak , Christo Venter , Roberta Zanin (for the H.E.S.S. collaboration) This is my paper · ORCID
classification astro-ph.HE
keywords PSRB1706-44gamma-raypulsationsH.E.S.S.IICT5monoscopicCherenkovmodepulsartimingfoldingphase-resolvedspectrumsub-100GeVgammarayssteeppower-law
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 reports that the 102 ms pulsar PSR B1706−44 pulses in sub-100 GeV gamma rays, detected with the H.E.S.S. II array's CT5 telescope in monoscopic mode. Over 28.3 hours, 1,532,177 ON-phase events against 3,050,011 OFF-phase events give an excess of about 7,171 photons at 4.74σ significance. The pulsed spectrum above 10 GeV is very steep, with photon index about −3.8, consistent with the Fermi-LAT spectrum above 10 GeV. A significant signal of about 1,000 events is reported near 70 GeV, though the data cannot confirm or rule out a power-law shape at the highest energies. If correct, the source becomes the fourth pulsar detected from the ground, after the Crab, Vela and Geminga, bridging satellite and ground-based energy ranges.

What carries the argument

The analysis chain uses events imaged by CT5, the single 28 m telescope, in monoscopic mode, with a boosted decision tree trained on Monte Carlo gamma-ray signal and real data background to reject hadrons, plus a 0.3° spatial cut and ON/OFF phase windows defined from the Fermi-LAT phasogram (ON: phase 0.25–0.55; OFF: phase 0.6–0.2). Event timestamps are folded into pulsar phase with the Tempo2 timing package using an ephemeris built from LAT timing and radio data, and significance is computed with the Li&Ma test and a maximum-likelihood ratio test using a kernel-density estimate of the Fermi light curve. A forward-folding maximum-likelihood fit with Monte Carlo instrument response functions converts reconstructed energies into the spectral points.

What would settle it

Recompute the Li&Ma significance from the same 28.3 hours while shifting the OFF window to include phases 0.2–0.25 and 0.55–0.6, and repeat with a tightened or loosened boosted-decision-tree cut; a stable excess above 4σ across these choices would confirm the pulsation, while a drop below 3σ for a reasonable OFF definition would indicate the detection depends on the background interval.

Watch

Extended reading notes

Core claim

The central claim is that PSR B1706−44 exhibits pulsed gamma-ray emission in the sub-20 GeV to sub-100 GeV range when 28.3 hours of H.E.S.S. II CT5 monoscopic data are folded with a combined radio and LAT ephemeris. The phasogram matches Fermi-LAT above 15 GeV, and a maximum-likelihood ratio test using the Fermi phasogram as template gives 4.6σ significance, while the independent Li&Ma ON/OFF test gives 4.74σ. The spectrum from both instruments above 10 GeV is consistent with a steep power law of index about −3.8, but the statistics do not allow distinguishing a continued power law from a curvature or cutoff. This makes the source the fourth pulsar detected from the ground and extends its measured spectrum toward 100 GeV.

Load-bearing premise

The detection significance assumes that the residual background remaining after the boosted-decision-tree and spatial cuts is identical in the ON-phase interval (0.25–0.55) and the OFF-phase interval (0.6–0.2), so a phase-dependent cosmic-ray or diffuse background would not inflate the 4.74σ excess.

Editorial extensions

If this is right

  • PSR B1706−44 joins the short list of pulsars detected from the ground: the Crab, Vela and Geminga.
  • The pulsed spectrum above 10 GeV is very steep, with photon index about −3.8 to −3.9 in both H.E.S.S. II-CT5 and Fermi-LAT, extending the measured spectrum into the sub-100 GeV range.
  • The H.E.S.S. light curve, in which the second peak dominates, tracks the Fermi-LAT phasogram above 15 GeV, indicating that the pulsed mechanism seen by Fermi persists at ground-accessible energies.
  • Current statistics cannot distinguish a continued power law from a curvature or cutoff near 70 GeV; the paper states that further H.E.S.S. or CTA observations are needed to decide whether the tail behaves more like Vela's or Crab's.

Reading between the lines

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

  • If the steep power law continues past 100 GeV, exponential-cutoff models for this pulsar's spectrum would need a separate high-energy tail component; this is an inference beyond what the paper itself claims.
  • The same ON/OFF phase-definition strategy from Fermi phasograms could be applied to other imaging atmospheric Cherenkov telescopes to search for pulsed emission from Fermi pulsars not yet detected from the ground.
  • Because the significance is 4.74σ, roughly doubling the exposure would be a direct test of whether the excess remains stable and whether the spectrum extends to ~100 GeV.
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.

Referee Report

2 major / 4 minor

Summary. This paper reports on a search for pulsed gamma-ray emission from PSR B1706-44 using 28.3 hours of H.E.S.S. II CT5 monoscopic observations. After defining ON and OFF phase intervals from a Fermi-LAT phasogram above 15 GeV, the authors find 1,532,177 ON-phase events versus 3,050,011 OFF-phase events, corresponding to an excess of 7,171.5 ± 1,515 events and a Li&Ma significance of 4.74σ. An independent pipeline yields a compatible significance of 4.6σ. A forward-folding power-law fit to the CT5 data gives Γ_HESS = 3.76 ± 0.36 and a normalization of (4.3 ± 0.9)×10^-8 TeV^-1 cm^-2 s^-1 at 20 GeV, consistent with the Fermi-LAT spectrum above 10 GeV (Γ_LAT = 3.9 ± 0.1). The abstract claims a significant signal of ~1000 events at ~70 GeV, while the text reports a 2.5σ excess of 2,782 events in the 54–225 GeV bin. The paper concludes that this is the fourth ground-based detection of a pulsar, after the Crab, Vela and Geminga.

Significance. If the detection is confirmed, this would be the fourth pulsar detected from the ground and would extend the measured pulsed spectrum of PSR B1706-44 to the tens-of-GeV range, providing a new data point for pulsar emission models. The paper's main strengths are the use of an externally defined phase template from Fermi-LAT, an independent analysis pipeline cross-check, and the consistency of the measured spectral index with Fermi-LAT. The authors also transparently note the inability to constrain spectral curvature. However, the central advertised claim of a significant ~70 GeV signal is not supported by the reported 2.5σ excess in that energy bin.

major comments (2)
  1. [Abstract and §3.2] The abstract states that 'a significant signal of ~1000 events is detected at energies ~70 GeV', but §3.2 reports that the 54–225 GeV bin has an excess of 2,782 events at a significance of only 2.5σ, and the abstract itself says it is 'not possible to either confirm or rule out a power-law behaviour'. These statements are in conflict. Since the 'sub-100 GeV' claim in the title and abstract rests on this bin, the authors should either provide a trials-corrected significance for the 54–225 GeV bin or remove the word 'significant' and qualify the ~70 GeV detection as marginal. The number of excess events should also be reconciled: the reported 2,782 is not '~1000'.
  2. [§3.1] The ON and OFF phase intervals are chosen using the Fermi-LAT phasogram from the same pulsar, and the KDE template used in the maximum-likelihood test is also derived from Fermi-LAT data. Although these are external data, the paper should state explicitly whether any trials were performed in selecting the intervals [0.25–0.55] and [0.6–0.2] or the KDE variance (0.025). Without this statement, the quoted 4.74σ and 4.6σ significances do not account for any possible selection effects. The cross-check with the alternative pipeline mitigates this concern, but a sentence quantifying the trials would make the detection claim robust.
minor comments (4)
  1. [§3.1] The phase interval [0.6-0.2] is shorthand for the union [0.6,1.0] and [0.0,0.2]; please write this explicitly to avoid confusion.
  2. [§3.2] The exponent in the parabola model is ambiguous: 'dN(E)/dE = Φ0 (E/E0)−ΓLPB−β ln(E/E0)' should be written as Φ0 (E/E0)^{-Γ_LPB - β ln(E/E0)}.
  3. [Abstract and §3.2] The average energy of the highest bin is given as 62.7 GeV in §3.2, not ~70 GeV; harmonize the numbers between the abstract and the text.
  4. [§4] The summary contains a typographical spacing issue in 'index ∼ −3.8 to 3 .9'; it should read '−3.8 to −3.9'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the H.E.S.S. detection is a self-contained ON/OFF counting measurement with Fermi-LAT used as an external template, not as an input fitted to produce the result.

full rationale

The paper's central claim, a 4.74σ Li&Ma detection of pulsations, is obtained by counting 1,532,177 events in the ON-phase interval against 3,050,011 events in the OFF-phase interval, giving a 7171.5 ± 1515 excess. This significance does not depend on any parameter fitted from the H.E.S.S. data; the ON/OFF intervals are defined using the Fermi-LAT phasogram, but that is an external measurement of the same pulsar, not an output of the H.E.S.S. analysis. The maximum-likelihood ratio test using the Fermi-LAT KDE template is presented as a compatible cross-check (4.6σ), not as the primary detection. The spectral index ΓHESS = 3.76 ± 0.36 is a forward-folding fit to the H.E.S.S. data alone, and the comparison to the Fermi-LAT index 3.9 ± 0.1 is an external benchmark, not a fitted input called a prediction. The estimation of the 62.7 GeV mean energy of the highest bin uses a simulated spectrum with Fermi-LAT parameters, but this is an energy-bias correction with stated assumptions and does not determine the detection significance or the fitted spectral index. Self-citations to the Vela CT5 pipeline [12] and to the forward-folding method [20] are methodological references to previously validated tools, not load-bearing uniqueness claims. No step in the derivation reduces by definition to its input; the claimed detection and spectrum have independent content.

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

The paper is a measurement, not a derivation. Its central values come from maximum-likelihood fits: the H.E.S.S. spectral index and normalization, and the Fermi-LAT spectral parameters used for comparison. The detection claim relies on standard assumptions about background subtraction, ephemeris accuracy, and instrument response functions. No new theoretical entities are introduced.

free parameters (4)
  • H.E.S.S. pulsed spectrum power-law index = 3.76 ± 0.36 (stat)
    Central claim of a steep spectrum with index about -3.8 depends on this fit (Section 3.2).
  • H.E.S.S. spectrum normalization at 20 GeV = (4.3 ± 0.9) × 10^-8 TeV^-1 cm^-2 s^-1
    Normalization from the same power-law fit; needed for flux comparison with Fermi-LAT.
  • Fermi-LAT power-law index above 10 GeV = 3.9 ± 0.1
    Comparison spectrum; agreement with the H.E.S.S. index supports the steep-spectrum claim.
  • KDE variance for Fermi-LAT phase template = 0.025
    Used to build the phase PDF for the maximum-likelihood ratio test; a different variance would change that test, though the Li&Ma result is independent.
assumptions (4)
  • domain assumption The Fermi-LAT phasogram above 15 GeV correctly represents the pulse profile of PSR B1706-44 at H.E.S.S. energies.
    Used to define ON and OFF phase intervals for the H.E.S.S. analysis (Section 3.1). If the pulse shape changes with energy, the ON/OFF selection could be suboptimal.
  • domain assumption The OFF-phase interval contains only background, with no pulsed emission.
    Background subtraction in the Li&Ma test assumes the OFF interval (phase 0.6 to 0.2) is free of signal (Section 3.1).
  • domain assumption The instrument response functions from Monte Carlo simulations accurately model the CT5 monoscopic response at energies around 20 to 100 GeV.
    Spectral unfolding and energy scale corrections rely on MC-based IRFs, with energy scale uncertainty set to +5% (Section 3.2 and reference [12]).
  • domain assumption The Tempo2 ephemeris from LAT timing and Parkes radio data is accurate over the 2013-2015 observation period.
    Phase folding of H.E.S.S. events uses this ephemeris (Section 2.1); an inaccurate ephemeris would smear the pulsation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Detection of sub-100 GeV gamma-ray pulsations from PSR B1706-44 with H.E.S.S." pith.science (2026). https://pith.science/paper/ZVYNFYNC

@misc{pith2026190806464,
  author       = {Pith},
  title        = {Pith review of: Detection of sub-100 GeV gamma-ray pulsations from PSR B1706-44 with H.E.S.S},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZVYNFYNC}},
  note         = {Machine review of arXiv:1908.06464}
}
read the original abstract

We report on the detection of pulsations from PSR B1706-44 based on 28.3 hours of observations with the H.E.S.S. II array with CT5 in monoscopic mode. The lightcurve is similar to that obtained with the Fermi-LAT above 15 GeV and the pulsations exhibit a steep spectrum with index ~ -3.8 in the sub 20 GeV to sub-100 GeV energy range. While a significant signal of ~ 1000 events is detected at energies ~ 70 GeV, it is not possible to either confirm or rule out a power-law behaviour of PSR B1706-44 spectrum in this range.

Figures

Figures reproduced from arXiv: 1908.06464 by the authors.

Figure 1
Figure 1. Histogram of phases for two periods with Fermi-LAT data > 15 GeV (top) and H.E.S.S. II with CT5 in monoscopic mode data (bottom). The hashed box represents the OFF-pulse region and the grey box the ON-pulse region. The empty (or white) area between both isn’t considered as part of the ON nor OFF region. The dashed line shows the average level of background evaluated in the OFF-pulse region. From the Fermi-LAT data a… view at source ↗
Figure 2
Figure 2. PSR B1706-44 spectral energy distribution. The grey and blue flux points are obtained from 5 years of Fermi data. Two fits are plotted: the power law with a sub-exponential cutoff above 100 MeV (grey) and another power law for the tail of the emission above 10 GeV (blue). The green box is derived from 28.3 hours of H.E.S.S. II-CT5 data and includes systematic errors (see text). model and the power law yields a signi… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 22 canonical work pages

  1. [1]

    A., Ajello, M., Allafort, A., et al

    Abdo, A. A., Ajello, M., Allafort, A., et al. 2013, ApJS, 208, 17

    2013

  2. [2]

    A., Ajello, M., Antolini, E., et al

    Abdo, A. A., Ajello, M., Antolini, E., et al. 2010, ApJ, 720, 26

    work page 2010

  3. [3]

    2017, Phys

    Abdollahi, S., Ackermann, M., Ajello, M., et al. 2017, Phys. Rev. D, 95, 082007

    work page 2017

  4. [4]

    2015, ApJS, 218, 23

    Acero, F., Ackermann, M., Ajello, M., et al. 2015, ApJS, 218, 23

    2015

  5. [5]

    A., Antonelli, L

    Aleksi ´c, J., Alvarez, E. A., Antonelli, L. A., et al. 2011, ApJ, 742, 43

    work page 2011

  6. [6]

    A., Antoranz, P., et al

    Ansoldi, S., Antonelli, L. A., Antoranz, P., et al. 2016, A&A, 585, A133 5 Detection of PSR B1706 −44 with H.E.S.S. M. Spir-Jacob

    work page 2016

  7. [7]

    2017, in American Institute of Physics Conference Series, V ol

    Djannati-Ataï, A., Giavitto, G., Holler, M., et al. 2017, in American Institute of Physics Conference Series, V ol. 1792, 6th International Symposium on High Energy Gamma-Ray Astronomy, 040028

    work page 2017

  8. [8]

    P., Srinivasan, R., Saito, Y ., et al

    Finley, J. P., Srinivasan, R., Saito, Y ., et al. 1998, ApJ, 493, 884

    work page 1998

Show all 26 references
  1. [9]

    A., Goss, W

    Frail, D. A., Goss, W. M., & Whiteoak, J. B. Z. 1994, ApJ, 437, 781

    1994

  2. [10]

    B., Frail, D

    Giacani, E. B., Frail, D. A., Goss, W. M., & Vieytes, M. 2001, AJ, 121, 3133

    2001

  3. [11]

    V ., Halpern, J

    Gotthelf, E. V ., Halpern, J. P., & Dodson, R. 2002, ApJ, 567, L125

    2002

  4. [12]

    H. E. S. S. Collaboration, Abdalla, H., Aharonian, F., et al. 2018, A&A, 620, A66

    2018

  5. [13]

    H. E. S. S. Collaboration, Abramowski, A., Acero, F., et al. 2011, A&A, 528, A143

    2011

  6. [14]

    B., Edwards, R

    Hobbs, G. B., Edwards, R. T., & Manchester, R. N. 2006, MNRAS, 369, 655

    2006

  7. [15]

    G., Manchester, R

    Johnston, S., Lyne, A. G., Manchester, R. N., et al. 1992, MNRAS, 255, 401

    1992

  8. [16]

    2015, Astrophys

    Kerr, M., Ray, P., Johnston, S., Shannon, R., & Camilo, F. 2015, Astrophys. J., 814, 128

    2015

  9. [17]

    Li, T. P. & Ma, Y . Q. 1983, ApJ, 272, 317

    1983

  10. [18]

    2019, in International Cosmic Ray Conference, V ol

    López Moya, M. 2019, in International Cosmic Ray Conference, V ol. 36, 36th International Cosmic Ray Conference (ICRC2019), 728

    2019

  11. [19]

    2015, in Proc

    Murach, T., Gadjus, M., & Parsons, R. 2015, in Proc. 34th Intern. Cosmic Ray Conf. No. 1022

    2015

  12. [20]

    2001, A&A, 374, 895

    Piron, F., Djannati-Atai, A., Punch, M., et al. 2001, A&A, 374, 895

    2001

  13. [21]

    W., et al

    Pittori, C., Verrecchia, F., Chen, A. W., et al. 2009, A&A, 506, 1563

    2009

  14. [22]

    S., Kerr, M., Parent, D., et al

    Ray, P. S., Kerr, M., Parent, D., et al. 2011, The Astrophysical Journal Supplement Series, 194, 17

    2011

  15. [23]

    N., Bennett, K., Bignami, G

    Swanenburg, B. N., Bennett, K., Bignami, G. F., et al. 1981, ApJ, 243, L69

    1981

  16. [24]

    J., Arzoumanian, Z., Bertsch, D

    Thompson, D. J., Arzoumanian, Z., Bertsch, D. L., et al. 1993, Nature, 365, 188

    1993

  17. [25]

    2011, Science, 334, 69

    VERITAS Collaboration, Aliu, E., Arlen, T., et al. 2011, Science, 334, 69

    2011

  18. [26]

    N., et al

    Weltevrede, P., Johnston, S., Manchester, R. N., et al. 2010, PASA, 27, 64 6

    2010

Pith tools

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