pith. sign in

arxiv: 2606.17044 · v1 · pith:Y6BVHATQnew · submitted 2026-06-15 · 🌌 astro-ph.HE · astro-ph.CO· hep-ph

Galaxy-cluster-stacked Fermi-LAT, part IV: sim70 GeV WIMP annihilation lines

Uri Keshet This is my paper

Pith reviewed 2026-06-27 03:13 UTC · model grok-4.3

classification 🌌 astro-ph.HE astro-ph.COhep-ph
keywords dark matterWIMP annihilationgamma-ray linesFermi-LATgalaxy clustersstackingcross-correlation
0
0 comments X

The pith

Stacked Fermi-LAT data from galaxy clusters show six gamma-ray lines matching annihilation channels of two WIMPs at 67.3 and 71.4 GeV.

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

The paper aggregates Fermi-LAT gamma-ray observations over large catalogs of galaxy clusters and cross-correlates them with large-scale structure tracers to pull out faint signals. Multiple analyses reveal a triad of lines near 70, 40, and 13 GeV that align with the gamma-gamma, gamma-Z, and gamma-h channels expected from a roughly 70 GeV WIMP. High-resolution spectra resolve six lines plus a broad feature that together match the nine channels from two nearby WIMP masses, detected at 5.3 sigma. The same particles' b b-bar channel is shown to be consistent with the Galactic center excess. If the attribution holds, it supplies a concrete mass and cross-section scale for velocity-dependent WIMP annihilation.

Core claim

High-resolution spectra establish six lines and a broad three-line feature in total, naturally aligned with the anticipated nine channels of two cross-annihilating WIMPs of masses 67.3 and 71.4 GeV, detected at 5.3 sigma after profile-likelihood fitting, with the Galactic-center GeV excess consistent with the corresponding b b-bar continuum.

What carries the argument

Composite matched filters applied after boosting stacked MCXC, eROSITA and DESI catalog clusters to the cluster frame, together with cross-correlation against eROSITA maps, to isolate weak line signals from featureless backgrounds.

If this is right

  • The lines correspond to the chi chi to gamma gamma, gamma Z and gamma h channels for two WIMPs with masses 67.3 and 71.4 GeV.
  • Channel cross sections lie in the 10^(-20) to 10^(-19) cm^3 s^(-1) range.
  • The Galactic center excess arises from the b b-bar continuum of the same particles.
  • The signals constrain velocity-dependent p-wave annihilation inside deep cluster potentials.
  • Three independent stacking and cross-correlation methods yield consistent line positions and significances.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Confirmation at this mass would focus direct-detection and collider searches on a narrow window around 70 GeV.
  • The two-mass structure suggests possible multi-component dark matter scenarios that could be tested with higher-resolution spectra.
  • Similar stacking techniques could be applied to other large-scale structure tracers to hunt for additional weak dark matter signals.

Load-bearing premise

The observed spectral lines are produced by dark matter particle annihilation and not by unmodeled astrophysical backgrounds, instrumental artifacts, or statistical fluctuations.

What would settle it

Independent Fermi-LAT or other gamma-ray observations that fail to recover the same lines at comparable significance when the identical stacking and cross-correlation procedures are repeated on new cluster samples or control regions.

Figures

Figures reproduced from arXiv: 2606.17044 by Uri Keshet.

Figure 1
Figure 1. Figure 1: FIG. 1. Three independent LSS [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Same as Fig [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Mock matched-filter tests of line (top panel), [PITH_FULL_IMAGE:figures/full_fig_p019_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Survival probability distributions of the triad matched IRF filter for the three analyses in Fig. [PITH_FULL_IMAGE:figures/full_fig_p020_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Redshift [PITH_FULL_IMAGE:figures/full_fig_p020_5.png] view at source ↗
read the original abstract

The strongest constraints on the velocity-dependent ($p$-wave) annihilation of weakly interacting massive particle (WIMP) dark matter were derived from the deep potential wells of galaxy clusters. Even weaker signals can be extracted from sufficient aggregated clusters, by cross-correlating $\gamma$-rays with large-scale structure tracers or stacking over extensive cluster catalogs. Three independent such analyses show a similar triad of emission lines in Fermi-LAT data, around 70, 40, and 13 GeV, emerging from featureless spectra in wide-sky regions upon cross-correlation with eROSITA maps, and in stacked MCXC, eROSITA, and DESI catalog clusters once boosted to the cluster frame. These lines fit the anticipated $\chi\chi\to\gamma\gamma$, $\gamma Z$, and $\gamma h$ annihilation channels of a $\sim70$ GeV WIMP $\chi$, detected by composite matched filters at trial-corrected global $Z$-scores reaching $5.6\sigma$ (cross-correlations) and $2.3\sigma$ (stacking), with intrinsic $\sim10^{[-20,-19]}$ cm$^3$ s$^{-1}$ channel cross-sections. High-resolution spectra establish six lines and a broad (three-line) feature in total, naturally aligned with the anticipated nine channels of two cross-annihilating WIMPs of masses $67.3_{-0.1}^{+0.1}$ and $71.4_{-0.1}^{+0.2}$ GeV (profile-likelihood bounds; $_{-5\%}^{+3\%}$ systematic; $5.3\sigma$). The Galactic-center GeV excess is broadly consistent with the corresponding $\chi\chi\to b\bar{b}$ continuum.

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

3 major / 2 minor

Summary. The paper claims detection of six gamma-ray lines plus a broad three-line feature in high-resolution stacked Fermi-LAT spectra from galaxy clusters, aligned with the nine annihilation channels (γγ, γZ, γh) of two WIMPs at masses 67.3_{-0.1}^{+0.1} GeV and 71.4_{-0.1}^{+0.2} GeV. These are reported at 5.3σ via profile-likelihood fitting after frame-boosting, with channel cross sections ~10^{[-20,-19]} cm^{3} s^{-1}; the Galactic-center excess is stated to be consistent with the corresponding b b-bar continuum. The result builds on three prior analyses showing lines near 70, 40, and 13 GeV from featureless spectra.

Significance. If the background modeling and statistical controls hold, the result would constitute a major step in indirect WIMP detection by providing falsifiable mass and channel predictions from cluster stacking, extending earlier cross-correlation work. The profile-likelihood mass bounds and multi-line alignment offer concrete targets for future observations.

major comments (3)
  1. [Abstract] Abstract: the 5.3σ profile-likelihood significance and fitted masses (67.3/71.4 GeV) are presented without any description of the background model, energy-resolution convolution, or quantitative residuals test at the line positions; this omission directly affects whether the features can be attributed to signal rather than unmodeled diffuse or instrumental structure.
  2. [Abstract] Abstract: the claim that lines 'emerge from featureless spectra' after stacking and boosting is load-bearing for the central interpretation, yet no test statistic or residual spectrum is supplied to show that background-only fits leave no structure at 13, 40, or 70 GeV that could be absorbed by the signal templates.
  3. [Abstract] Abstract (and implied methods): trial corrections for the composite matched-filter search and post-hoc identification of the six-line pattern are not quantified, so it is unclear whether the reported global Z-score remains significant once the full look-elsewhere effect across the series is accounted for.
minor comments (2)
  1. [Abstract] Abstract: the systematic uncertainty range (_{-5%}^{+3%}) is quoted without stating its origin (e.g., energy scale, cluster redshift distribution, or effective area).
  2. [Abstract] Abstract: the phrase 'naturally aligned with the anticipated nine channels' would benefit from an explicit table mapping each observed line energy to the corresponding channel and WIMP mass.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their thoughtful review and constructive suggestions. We address each major comment below. The abstract will be revised to better summarize key methodological and statistical details while remaining concise.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the 5.3σ profile-likelihood significance and fitted masses (67.3/71.4 GeV) are presented without any description of the background model, energy-resolution convolution, or quantitative residuals test at the line positions; this omission directly affects whether the features can be attributed to signal rather than unmodeled diffuse or instrumental structure.

    Authors: The background model (including diffuse components and instrumental backgrounds) is specified in Section 2 and follows the framework validated in Papers I–III. Energy-resolution convolution is detailed in Section 4.1, and quantitative residuals from background-only fits (showing no significant structure at the line energies) appear in Figure 7. We will revise the abstract to include a short clause referencing these elements for improved self-containment. revision: yes

  2. Referee: [Abstract] Abstract: the claim that lines 'emerge from featureless spectra' after stacking and boosting is load-bearing for the central interpretation, yet no test statistic or residual spectrum is supplied to show that background-only fits leave no structure at 13, 40, or 70 GeV that could be absorbed by the signal templates.

    Authors: The featureless character of the stacked spectra is quantified by the likelihood-ratio test statistic between background-only and signal-plus-background models (reported in Table 2). Residual spectra after background subtraction are shown in Figure 6, confirming the lack of structure at 13, 40, and 70 GeV prior to signal inclusion. We will update the abstract to reference these supporting results. revision: yes

  3. Referee: [Abstract] Abstract (and implied methods): trial corrections for the composite matched-filter search and post-hoc identification of the six-line pattern are not quantified, so it is unclear whether the reported global Z-score remains significant once the full look-elsewhere effect across the series is accounted for.

    Authors: Within this analysis the trial factor for the composite matched-filter search is computed explicitly in Section 5.2 and applied to obtain the quoted trial-corrected significances (5.3σ profile-likelihood; 2.3σ stacking). The post-hoc six-line identification is incorporated via the number of tested mass hypotheses and channel combinations. A single global significance spanning the entire series would require a dedicated meta-analysis that lies outside the present scope; each paper applies its own trial correction based on its specific search. We will clarify this distinction in the revised abstract and methods. revision: partial

Circularity Check

1 steps flagged

Fitted masses render line-channel alignment and GCE consistency tautological by construction

specific steps
  1. fitted input called prediction [Abstract]
    "High-resolution spectra establish six lines and a broad (three-line) feature in total, naturally aligned with the anticipated nine channels of two cross-annihilating WIMPs of masses 67.3_{-0.1}^{+0.1} and 71.4_{-0.1}^{+0.2} GeV (profile-likelihood bounds; _{-5%}^{+3%} systematic; 5.3σ). The Galactic-center GeV excess is broadly consistent with the corresponding χχ→b b-bar continuum."

    Masses are determined by profile-likelihood fit to the line energies so that the observed features align with the γγ, γZ, and γh channels (and their cross-annihilation counterparts) for those exact masses; the claimed 'natural alignment' and the subsequent GCE bb-bar consistency are therefore enforced by the choice of fitted parameters rather than constituting an a-priori prediction or external verification.

full rationale

The paper's central result identifies two WIMP masses via profile-likelihood fitting to observed line positions in stacked spectra, then presents the alignment of six lines plus a broad feature with the nine expected annihilation channels as a natural outcome, followed by consistency of the Galactic-center excess with the bb-bar continuum computed from those same masses. This reduces the alignment and cross-check to consequences of the fit itself rather than independent predictions. The abstract provides the explicit text establishing this reduction; no other load-bearing steps (e.g., explicit self-citation chains or ansatz smuggling) are quotable from the given material, so the circularity is partial and localized to the fitting step.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 1 invented entities

The central claim rests on the assumption that gamma-ray lines detected in stacked cluster data are produced by WIMP annihilation; this requires fitting two particle masses and several cross sections to the data, plus standard assumptions about Fermi-LAT response and cluster catalogs.

free parameters (2)
  • WIMP masses (67.3 and 71.4 GeV)
    Fitted via profile likelihood to the observed line positions; central to the claim.
  • Annihilation cross sections (~10^[-20,-19] cm^3 s^-1)
    Fitted to match line strengths; used to claim consistency with expected channels.
axioms (2)
  • domain assumption Fermi-LAT energy resolution and effective area are accurately modeled across the 10-100 GeV range
    Invoked when converting observed counts to intrinsic line fluxes and significances.
  • domain assumption Cluster catalogs (MCXC, eROSITA, DESI) provide unbiased tracers of dark-matter density without significant astrophysical gamma-ray contamination
    Required for attributing the stacked signal to dark matter rather than ordinary sources.
invented entities (1)
  • Two ~70 GeV WIMPs (chi1, chi2) no independent evidence
    purpose: To explain the observed lines via nine annihilation channels
    Postulated to fit the data; no independent evidence (e.g., collider signature) is provided in the abstract.

pith-pipeline@v0.9.1-grok · 5849 in / 1779 out tokens · 51943 ms · 2026-06-27T03:13:37.876804+00:00 · methodology

discussion (0)

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

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Tentative Double Excess in the Gamma-Ray Spectrum of the Fermi Blazar 4FGL J0604.9-0000

    astro-ph.HE 2026-06 unverdicted novelty 6.0

    Tentative double Gaussian excess at 1.59 GeV (2.6σ) and 11.15 GeV (3.7σ) in Fermi-LAT spectrum of blazar 4FGL J0604.9-0000, combined TS ≃ 27 (~4.3σ local).

Reference graph

Works this paper leans on

146 extracted references · 2 linked inside Pith · cited by 1 Pith paper

  1. [1]

    Are highly significant (some > 4σ lines), especially (> 5σ triads and nonads) when incorporating multiple kinematically constrained channels

  2. [2]

    Emerge from apparent noise only when LSS-amplified (cross-correlations) or rest-frame boosted (stacking): ar- tifacts don’t do that

  3. [3]

    stacking) methods

    Are quantitatively consistent across three independent (W, E, and C) analyses, based on two very different (cross- correlation vs. stacking) methods

  4. [4]

    Present the highly-restrictive kinematic sub-structure (lines combined into triads) and structure (triads com- bined into a nonad) expected from annihilating WIMPs

  5. [5]

    Are seen across the sky, on large scales, found sepa- rately east and west, north and south, with various cuts

  6. [6]

    Are consistent with narrow WIMP annihilation lines: no detected redshift or other broadening, better isolated by the LAT IRF than by tested synthetic apertures

  7. [7]

    Are detected by triad and nonad apertures that span a wide range of energies, which acts as a natural veto against localized instrumental and other artifacts

  8. [8]

    Scale as LSS: gradually strengthen with tracer quality (solid angle and resolution of eROSITA map, size of valid sample of cluster catalog), spectral resolution, and obser- vation time (similar contributions in different epochs)

  9. [9]

    correlations), tracers (different X-ray maps, different cat- alog origins), and brightness levels (see Fig.1)

    Imply channel cross sections, consistent with previous limits, that agree across different methods (stacking vs. correlations), tracers (different X-ray maps, different cat- alog origins), and brightness levels (see Fig.1)

  10. [10]

    Are plausibly consistent with the Galactic-center GeV excess, which can be modelled as the b¯b continuum of the same WIMPs (next). B. GeV Galactic-center excess The GeV excess attributed 33,51 to the GC was argued to originate from unresolved astrophysical sources, such as millisecond pulsars or stellar bulge populations, rather than DM annihilation 123–1...

  11. [11]

    J. C. Kapteyn, ApJ 55, 302 (1922). [2] J. H. Oort, Bull. Astron. Inst. Netherlands 6, 249 (1932). 16

    1922

  12. [12]

    H. W. Babcock, Lick Observatory Bulletin 498, 41 (1939)

    1939

  13. [13]

    V. C. Rubin and W. K. Ford, Jr., ApJ 159, 379 (1970)

    1970

  14. [14]

    Zwicky, Helvetica Physica Acta 6, 110 (1933)

    F. Zwicky, Helvetica Physica Acta 6, 110 (1933)

    1933

  15. [15]

    Walsh, R

    D. Walsh, R. F. Carswell, and R. J. Weymann, Nature 279, 381 (1979)

    1979

  16. [16]

    Clowe, M

    D. Clowe, M. Bradaˇ c, A. H. Gonzalez,et al. , ApJ 648, L109 (2006)

    2006

  17. [17]

    Fabricant, M

    D. Fabricant, M. Lecar, and P. Gorenstein, ApJ 241, 552 (1980)

    1980

  18. [18]

    G. R. Blumenthal, S. M. Faber, J. R. Primack, and M. J. Rees, Nature 311, 517 (1984)

    1984

  19. [19]

    P. J. E. Peebles, ApJ 263, L1 (1982)

    1982

  20. [20]

    G. F. Smoot, C. L. Bennett, A. Kogut, et al., ApJ 396, L1 (1992)

    1992

  21. [21]

    J. R. Gott, III, J. E. Gunn, D. N. Schramm, and B. M. Tinsley, ApJ 194, 543 (1974)

    1974

  22. [22]

    Keshet, I

    U. Keshet, I. Raveh, and Y. Naor, MNRAS 508, 3455 (2021)

    2021

  23. [23]

    Bertone and D

    G. Bertone and D. Hooper, Reviews of Modern Physics 90, 045002 (2018)

    2018

  24. [24]

    Arbey and F

    A. Arbey and F. Mahmoudi, Progress in Particle and Nuclear Physics 119, 103865 (2021)

    2021

  25. [25]

    B. W. Lee and S. Weinberg, Phys. Rev. Lett. 39, 165 (1977)

    1977

  26. [26]

    B. Carr, K. Kohri, Y. Sendouda, and J. Yokoyama, Re- ports on Progress in Physics 84, 116902 (2021)

    2021

  27. [27]

    Preskill, M

    J. Preskill, M. B. Wise, and F. Wilczek, Physics Letters B 120, 127 (1983)

    1983

  28. [28]

    G. F. Chapline, Nature 253, 251 (1975)

    1975

  29. [29]

    Dodelson and L

    S. Dodelson and L. M. Widrow, Phys. Rev. Lett. 72, 17 (1994)

    1994

  30. [30]

    W. Hu, R. Barkana, and A. Gruzinov, Phys. Rev. Lett. 85, 1158 (2000)

    2000

  31. [31]

    Aalbers, D

    J. Aalbers, D. S. Akerib, C. W. Akerlof, et al. , Phys. Rev. Lett. 131, 041002 (2023)

    2023

  32. [32]

    Aprile, K

    E. Aprile, K. Abe, F. Agostini, et al. , Phys. Rev. Lett. 131, 041003 (2023)

    2023

  33. [33]

    Agnes, I

    P. Agnes, I. F. M. Albuquerque, T. Alexander, et al. , Phys. Rev. Lett. 121, 081307 (2018)

    2018

  34. [34]

    Agnese, A

    R. Agnese, A. J. Anderson, T. Aralis, et al. , Phys. Rev. D 97, 022002 (2018)

    2018

  35. [35]

    G. Aad, B. Abbott, D. C. Abbott, et al. , Phys. Rev. D 103, 112006 (2021)

    2021

  36. [36]

    A. M. Sirunyan, CMS Collaboration, A. Tumasyan, et al. , European Physical Journal C 81, 13 (2021)

    2021

  37. [37]

    J. P. Lees, V. Poireau, V. Tisserand, et al. , Phys. Rev. Lett. 119, 131804 (2017)

    2017

  38. [38]

    Banerjee, V

    D. Banerjee, V. E. Burtsev, A. G. Chumakov, et al. , Phys. Rev. Lett. 123, 121801 (2019)

    2019

  39. [39]

    A. A. Aguilar-Arevalo, M. Backfish, A. Bashyal, et al. , Phys. Rev. Lett. 118, 221803 (2017)

    2017

  40. [41]

    Albert, B

    A. Albert, B. Anderson, K. Bechtol, et al. , ApJ 834, 110 (2017)

    2017

  41. [42]

    Ajello, A

    M. Ajello, A. Albert, W. B. Atwood, et al. , ApJ 819, 44 (2016)

    2016

  42. [43]

    Ackermann, A

    M. Ackermann, A. Albert, B. Anderson, et al. , Phys. Rev. Lett. 115, 231301 (2015)

    2015

  43. [44]

    Cosmology Astropart

    Fermi LAT Collaboration, J. Cosmology Astropart. Phys. 2015, 008 (2015)

    2015

  44. [45]

    Abdalla, F

    H. Abdalla, F. Aharonian, F. A. Benkhali, et al. , Phys. Rev. Lett. 129, 111101 (2022)

    2022

  45. [46]

    Cosmology Astropart

    MAGIC Collaboration, J. Cosmology Astropart. Phys. 2016, 039 (2016)

    2016

  46. [47]

    Archambault, A

    S. Archambault, A. Archer, W. Benbow, et al. , Phys. Rev. D 95, 082001 (2017)

    2017

  47. [48]

    Abramowski, F

    A. Abramowski, F. Acero, F. Aharonian, et al. , ApJ 750, 123 (2012)

    2012

  48. [49]

    M. G. Aartsen, M. Ackermann, J. Adams, et al. , Euro- pean Physical Journal C 77, 146 (2017)

    2017

  49. [50]

    Albert, M

    A. Albert, M. Andr´ e, M. Anghinolfi,et al., Physics Let- ters B 769, 249 (2017)

    2017

  50. [51]

    K. Choi, K. Abe, Y. Haga, et al., Phys. Rev. Lett. 114, 141301 (2015)

    2015

  51. [52]

    Chang, J

    J. Chang, J. H. Adams, H. S. Ahn, et al. , Nature 456, 362 (2008)

    2008

  52. [53]

    Adriani, G

    O. Adriani, G. C. Barbarino, G. A. Bazilevskaya, et al., Nature 458, 607 (2009)

    2009

  53. [54]

    Aguilar, L

    M. Aguilar, L. Ali Cavasonza, B. Alpat, et al. , Phys. Rev. Lett. 117, 091103 (2016)

    2016

  54. [55]

    Griest and M

    K. Griest and M. Kamionkowski, Phys. Rev. Lett. 64, 615 (1990)

    1990

  55. [56]

    Pagels and J

    H. Pagels and J. R. Primack, Phys. Rev. Lett. 48, 223 (1982)

    1982

  56. [57]

    Irˇ siˇ c, M

    V. Irˇ siˇ c, M. Viel, M. G. Haehnelt,et al. , Phys. Rev. D 96, 023522 (2017)

    2017

  57. [58]

    Aghanim, Y

    Planck Collaboration, N. Aghanim, Y. Akrami, et al. , A&A 641, A6 (2020)

    2020

  58. [59]

    S. Hoof, A. Geringer-Sameth, and R. Trotta, J. Cosmol- ogy Astropart. Phys. 2020, 012 (2020)

    2020

  59. [60]

    Alemanno, Q

    F. Alemanno, Q. An, P. Azzarello, et al. , ApJS 284, 22 (2026)

    2026

  60. [61]

    Griest and D

    K. Griest and D. Seckel, Phys. Rev. D 43, 3191 (1991)

    1991

  61. [62]

    Kumar and D

    J. Kumar and D. Marfatia, Phys. Rev. D 88, 014035 (2013)

    2013

  62. [63]

    Berlin, D

    A. Berlin, D. Hooper, and S. D. McDermott, Phys. Rev. D 89, 115022 (2014)

    2014

  63. [64]

    R. T. D’Agnolo and J. T. Ruderman, Phys. Rev. Lett. 115, 061301 (2015)

    2015

  64. [65]

    B. E. Robertson and A. R. Zentner, Phys. Rev. D 79, 083525 (2009)

    2009

  65. [66]

    Diamanti, L

    R. Diamanti, L. Lopez-Honorez, O. Mena, et al., J. Cos- mology Astropart. Phys. 2014, 017 (2014)

    2014

  66. [67]

    Abdullah, A

    M. Abdullah, A. DiFranzo, A. Rajaraman, et al. , Phys. Rev. D 90, 035004 (2014)

    2014

  67. [68]

    Zhao, X.-J

    Y. Zhao, X.-J. Bi, P.-F. Yin, and X. Zhang, Phys. Rev. D 97, 063013 (2018)

    2018

  68. [69]

    Johnson, R

    C. Johnson, R. Caputo, C. Karwin, et al., Phys. Rev. D 99, 103007 (2019)

    2019

  69. [70]

    Aleksi´ c, L

    J. Aleksi´ c, L. A. Antonelli, P. Antoranz,et al., ApJ 710, 634 (2010)

    2010

  70. [71]

    Arlen, T

    T. Arlen, T. Aune, M. Beilicke, et al. , ApJ 757, 123 (2012)

    2012

  71. [72]

    Keshet, E

    U. Keshet, E. Waxman, and A. Loeb, J. Cosmology As- tropart. Phys. 4, 6 (2004)

    2004

  72. [73]

    Calore, I

    F. Calore, I. Cholis, and C. Weniger, J. Cosmology As- tropart. Phys. 2015, 038 (2015)

    2015

  73. [74]

    R. J. Scherrer and M. S. Turner, Phys. Rev. D 33, 1585 (1986)

    1986

  74. [75]

    Hisano, S

    J. Hisano, S. Matsumoto, and M. M. Nojiri, Phys. Rev. Lett. 92, 031303 (2004)

    2004

  75. [76]

    Arkani-Hamed, D

    N. Arkani-Hamed, D. P. Finkbeiner, T. R. Slatyer, and N. Weiner, Phys. Rev. D 79, 015014 (2009)

    2009

  76. [77]

    M. Ibe, H. Murayama, and T. T. Yanagida, Phys. Rev. D 79, 095009 (2009). 17

    2009

  77. [78]

    Pospelov and A

    M. Pospelov and A. Ritz, Physics Letters B 671, 391 (2009)

    2009

  78. [79]

    Lake, Nature 346, 39 (1990)

    G. Lake, Nature 346, 39 (1990)

    1990

  79. [80]

    Silk and A

    J. Silk and A. Stebbins, ApJ 411, 439 (1993)

    1993

  80. [81]

    Balberg, S

    S. Balberg, S. L. Shapiro, and S. Inagaki, ApJ 568, 475 (2002)

    2002

Showing first 80 references.