Searches for neutrinos from fast radio bursts with IceCube

Alex Pizzuto; Ali Kheirandish; Justin Vandenbroucke (for the IceCube Collaboration)

REVIEW 1 major objections 5 minor 22 references

Searches for neutrinos from fast radio bursts with IceCube

T0 review · 1 major / 5 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 The paper reports two IceCube searches for neutrinos coincident in time and direction with fast radio bursts, found no excess, and consequently sets the first limits on MeV neutrino emission from FRBs and the tightest GeV–TeV limits to…

desk verdict A clean null result with a real first: IceCube reports the first MeV limits and improved GeV-TeV limits on neutrino emission from FRBs, though the MeV limit rests on a stated fiducial spectrum. read the letter →

arxiv 1909.00078 v1 pith:AJRYJ3V6 submitted 2019-08-30 astro-ph.HE

Ali Kheirandish , Alex Pizzuto , Justin Vandenbroucke (for the IceCube Collaboration) This is my paper · ORCID

classification astro-ph.HE

keywords fastradioburstsIceCubeneutrinoastronomymuonneutrinosMeVSNDAQmultimessengerupperlimits

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 attempts to establish whether fast radio bursts produce neutrinos, using two complementary IceCube searches: one for GeV–TeV muon-neutrino tracks and one for collective MeV hit-rate bursts. No statistically significant excess is found in either channel, so the authors report the first upper limits on MeV neutrino emission from FRBs and the tightest limits to date at GeV–TeV energies. If correct, the result means the observed FRBs are not, as a class, strong neutrino emitters, and it gives quantitative caps on the hadronic energy FRB progenitors can release as neutrinos. This matters because FRBs have been a candidate source of IceCube's diffuse astrophysical neutrino flux, and these limits test that possibility.

What carries the argument

The analysis rides on two likelihood engines. For GeV–TeV neutrinos, an unbinned maximum-likelihood test statistic (the paper's Eq. 2.1) compares a signal spatial PDF, a two-dimensional Gaussian around each FRB position, against a Poisson background PDF built from off-time data, and is maximized to find the best-fit number of signal events for a given time window $\Delta T$. For MeV neutrinos, a one-dimensional Gaussian likelihood over per-optical-module (DOM) hit rates (Eq. 2.2) measures a collective rate excess $\Delta\mu$ in the Supernova Data Acquisition (SNDAQ) system, which records photomultiplier hits in 1.6 ms bins; the resulting significance $\xi = \Delta\mu/\sigma_{\Delta\mu}$ is compared with seasonally corrected off-time significances. The new leverage is that FRBs' millisecond timescales allow the track analysis to loosen its event selection, increasing effective area, especially in the Southern sky, while keeping the background penalty small.

What would settle it

Observe a single coincident event: an IceCube muon track reconstructed within the directional uncertainty of a radio-detected FRB and inside a $\Delta T$ window such as 10 ms, or an SNDAQ significance above the $3\sigma$ threshold in any 10–1280 ms window around the 21 bursts, would contradict the reported null correlation. A purely archival cross-check would recompute the stacking test statistic for the 28 FRBs with the six-year high-purity track selection; if the per-burst limits for $\Delta T < 1$ s did not improve by roughly a factor of 50 over the six-year result, the claimed sensitivity gain would fail.

Watch

Extended reading notes

Core claim

The central claim is a null result with improved constraints. In a stacking search and a max-burst search over 28 FRBs using a loosened muon-track sample, the post-trial $p$-values are 0.35 and 0.33, consistent with atmospheric background. In the SNDAQ search over 21 FRBs, no significance exceeds the one-sided $3\sigma$ threshold, with the most significant burst at $2.55\sigma$. From these non-detections the paper derives 90% confidence upper limits on the time-integrated neutrino flux per FRB for time windows from milliseconds to seconds, finding that for $\Delta T < 1$ s the $E^{-3}$ stacking limits are about a factor of 50 stronger than the previous six-year analysis, and that the MeV limits are the first ever placed on FRBs.

Load-bearing premise

The MeV limits assume FRB neutrinos follow a core-collapse supernova spectrum with mean energy 15.6 MeV and pinching parameter $\alpha = 3$, chosen only because no FRB-specific neutrino spectrum model exists; the SNDAQ hit rate depends on the incident flux shape, so different spectra would change the quoted limits.

Editorial extensions

If this is right

The per-burst limits at short time windows constrain any model in which a homogeneous class of FRBs produces IceCube's diffuse flux: the stacking limits provide a benchmark obtained by dividing the measured diffuse $\nu_\mu$ flux among thousands of FRBs per day.
The first MeV limits rule out a supernova-like neutrino flash accompanying the observed bursts; any model that gives FRB progenitors a bright MeV-neutrino counterpart at these fluences is inconsistent with the data.
The same pipeline applied to the larger FRB catalogs expected from wide-field radio interferometers should reach the diffuse-flux benchmark for Northern-Sky sources once the source class grows by about an order of magnitude.
The loosened track selection shows that millisecond timing windows can compensate for higher atmospheric background, a strategy transferable to searches for other short-duration transients.

Reading between the lines

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

An unstated consequence is that the quoted MeV limits are tied to the specific 15.6 MeV supernova template; repeating the SNDAQ analysis with several spectral templates would yield spectrum-independent bounds, and the paper's noted dependence of hit rate on incident flux suggests results could shift materially for harder or softer spectra.
If only a subpopulation of FRBs, such as repeaters or bursts with persistent radio emission, produce neutrinos, the stacking test dilutes the signal; as the catalog grows, splitting FRBs by class and running the max-burst test separately may reveal a signal the current 28-source sample cannot resolve.
The SNDAQ burst technique could be applied to other millisecond astrophysical transients, such as magnetar giant flares or short gamma-ray burst precursors, where MeV neutrino emission is still essentially unconstrained.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Colleague,

This is a straightforward, competent null-result paper from IceCube. The one thing you should know: it finds no neutrino excess coincident with FRBs, and in doing so it publishes the first MeV limits from FRBs and tighter GeV-TeV limits than the earlier six-year analysis. The novelty is real but modest — established methods applied to a young source class.

What the paper does well: the high-energy search uses a loosened track selection than the six-year analysis, which buys effective area in the southern sky and at lower energies; the likelihood is standard, and they compute post-trial p-values from Monte Carlo (0.35 stacking, 0.33 max-burst). The MeV search is a genuinely new application of SNDAQ to FRBs via collective DOM hit-rate excess, searching 21 FRBs on timescales 10–1280 ms. The null result is robust, limits at 90% CL are clearly presented, and the sensitivity projections for future radio interferometers are a nice touch.

Soft spots: the MeV limit is quoted for a single fiducial CCSN spectrum (mean energy 15.6 MeV, alpha=3). The authors explicitly note the hit-rate dependence on incident flux, but they do not quantify it — a harder or softer spectrum could move the limit. That makes the 'first limit on MeV neutrino emission' claim somewhat model-dependent, though not misleading, since the assumption is stated. I would have liked one plot showing how the limit shifts for other plausible spectra. Also, the MeV limits have no systematic error band, and the 3-sigma detection threshold is not corrected for trials. The trial issue is minor because the largest observed significance is 2.55 sigma, so the null result stands regardless.

Overall, the citation pattern is clean — the self-citations are the prior IceCube FRB analyses that this work improves on. The paper is honest about its assumptions and the central null result is not in question. This is more of a conference-proceedings contribution than a breakthrough paper, but it is exactly the kind of reference that FRB modelers and neutrino advocates will want to cite.

If I were a journal editor, I would send this to a referee if it were submitted as a Letter or proceedings paper; it does not need major revision. The authors should be asked to add spectral-dependence quantification and a trials statement, but those are minor.

Referee Report

1 major / 5 minor

Summary. This manuscript, an IceCube Collaboration contribution to ICRC 2019, describes two complementary searches for neutrinos from fast radio bursts (FRBs). The first is an unbinned maximum-likelihood muon-track search applied to 28 FRBs, using a looser event selection than the prior six-year analysis and testing both a stacking hypothesis and a max-burst hypothesis over a range of time windows. No significant excess is found, with post-trial p-values of 0.35 (stacking) and 0.33 (max-burst), and 90% confidence upper limits are placed on the time-integrated neutrino flux per FRB for E^-2 and E^-3 spectra. The second search uses the SNDAQ hit-rate stream to look for collective MeV-scale neutrino bursts from 21 FRBs over eight time windows; again no significant signal is found, and the authors report 90% upper limits on anti-electron-neutrino fluence assuming a fiducial core-collapse supernova spectrum. The paper closes with projections for future searches using next-generation radio interferometers.

Significance. If the results stand, the paper provides the most stringent limits to date on high-energy neutrino emission from FRBs and, for the first time, places limits on MeV-scale neutrino emission from FRBs. The high-energy analysis is methodologically sound: the likelihood is standard, the background is parameterized from data, and the post-trial p-values are obtained from Monte Carlo simulations. The reported null result is robust and does not depend on the assumed spectral shape of any hypothetical FRB neutrino emission. The paper also gives credit to the improvement over the six-year analysis, noting factor-of-50 stronger limits for short timescales under an E^-3 spectrum. The MeV component is a novel use of IceCube's supernova trigger, and the explicit use of an external CCSN spectral model (rather than a fitted model) keeps the circularity burden low. The main weakness is the unquantified spectral dependence of the MeV fluence limits, which affects the generality of the abstract's 'first limit on MeV neutrino emission' claim.

major comments (1)

[Section 3, MeV search and Figure 6] The quoted upper limits on anti-electron-neutrino fluence from FRBs are derived for a single fiducial core-collapse supernova spectrum with mean energy 15.6 MeV and pinching parameter alpha=3. The manuscript itself notes that the signal hit rate depends on the incident flux, but the size of this spectral dependence is not quantified. Because the MeV detection channel combines the inverse-beta-decay cross section, Cherenkov yield, and DOM threshold, the conversion from a hit-rate excess to a fluence limit can vary strongly with the assumed neutrino spectrum. As written, the abstract's claim of 'the first limit on MeV neutrino emission from FRBs' overstates the generality of the result. Please either (i) state in the abstract and conclusions that the limits apply specifically to a CCSN-like spectrum and quantify, even roughly, the expected variation with spectral shape, or (ii) rephrase the claim as a limit on the fiducial model.

minor comments (5)

[Section 2.2] There is a typo in the text: 'backgrond' should be 'background'.
[Section 2.2, Eq. (2.2)] The quantity sigma_i is introduced as the standard deviation of the hit rate in DOM_i, but the likelihood uses <sigma_i> in the denominator; please clarify whether <sigma_i> is the time-averaged standard deviation and how it is estimated.
[Section 3, Figure 6] Unlike the high-energy limits in Figure 4, the MeV limits in Figure 6 are shown without any systematic uncertainty bands; a brief statement on the dominant systematics (e.g., DOM efficiencies, atmospheric muon rate modeling) would help the reader judge the robustness of the limits.
[Section 2.2, MeV search] The MeV search tests 21 FRBs across 8 time windows, but the reported significance is compared to a 3-sigma threshold from off-time windows without an explicit trials correction for the number of FRBs and time windows. The null result is unaffected because the largest observed significance is 2.55 sigma, but the procedure should be described more carefully to avoid an overclaim in future applications.
[Abstract and Section 3] The claim that the high-energy limits are 'the most constraining' is supported by comparison with the preceding IceCube six-year analysis and by the effective-area comparison with ANTARES, but a direct limit-to-limit comparison with ANTARES is not shown; please state the basis of the comparison more explicitly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the IceCube FRB-neutrino limits are derived from data using external, non-fitted inputs.

full rationale

The paper's central claims are a null search and upper limits on neutrino emission from FRBs. The high-energy muon-track analysis uses an unbinned likelihood whose signal and background PDFs are defined from detector properties and data-based background estimates; no astrophysical parameter is fitted to the FRB data and then renamed as a prediction. The quoted limits assume benchmark power-law spectra (E^-2, E^-3), which are conventional test spectra, not outputs of the analysis. The MeV search uses the SNDAQ collective hit-rate method, cited from established IceCube instrumentation literature [15], and converts an observed hit-rate significance into a fluence limit under a fiducial core-collapse supernova neutrino spectrum taken from Totani et al. [20]. That spectrum is external to the present data, is explicitly labeled as fiducial because no FRB neutrino spectrum model exists, and is not fitted to IceCube results. The paper even notes the dependence of the signal hit rate on varying incident fluxes, so the model dependence is disclosed rather than hidden. The self-citations to prior IceCube FRB analyses [13,14] are used for comparison and motivation, not as load-bearing justification for the new limits. No equation in the paper reduces by construction to its input, no fitted parameter is called a prediction, and no uniqueness claim is imported. Model dependence of the MeV limits is a caveat about interpretation, not circularity.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central result is an observational null result. No new particles, forces, or mechanisms are introduced. The main assumptions are statistical models and spectral inputs from prior literature. The MeV limit is the most assumption-dependent: it is quoted for a fiducial core-collapse supernova spectrum.

free parameters (2)

Signal normalization n_s = not reported (consistent with 0)
Maximized in the unbinned likelihood (Eq. 2.1) for each time window. It is a standard fit parameter in the search, not an ad hoc assumption; the null result has no significant excess.
MeV significance threshold = 3σ (one-sided)
Chosen by hand as the detection threshold from the background significance distribution (Sec. 2.2). It does not affect the reported limits but is not corrected for multiple tests.

assumptions (4)

domain assumption FRB positions are accurate enough for a 2D Gaussian spatial signal PDF
S(xi) in Eq. 2.1 models angular distance to each FRB; incorrect coordinates would degrade the spatial correlation test.
domain assumption Atmospheric background dominates and is stable outside each search window
B(xi) is parameterized from data excluding the search time window; seasonal muon variations are handled with the technique of [18].
domain assumption The fiducial supernova neutrino spectrum represents FRB MeV emission
The MeV limits are quoted for a core-collapse supernova spectrum (mean energy 15.6 MeV, pinching parameter α=3) because no FRB spectrum model exists; the paper notes the signal hit rate depends on incident flux.
standard math Standard statistical assumptions for the likelihoods and Monte Carlo trials
Unbinned Poisson likelihood (Eq. 2.1), Gaussian DOM rates (Eq. 2.2), and post-trial Monte Carlo for the track search (Sec. 3).

how reviews work

0 comments

Cite this review

Pith. "Pith review of Searches for neutrinos from fast radio bursts with IceCube." pith.science (2026). https://pith.science/paper/AJRYJ3V6

@misc{pith2026190900078,
  author       = {Pith},
  title        = {Pith review of: Searches for neutrinos from fast radio bursts with IceCube},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AJRYJ3V6}},
  note         = {Machine review of arXiv:1909.00078}
}

read the original abstract

Although IceCube has discovered a diffuse astrophysical neutrino flux, the underlying sources of these neutrinos remain unknown. Transient astrophysical objects, such as fast radio bursts (FRBs), could explain a large percentage of the measured flux. We present the analysis techniques of IceCube searches for MeV to TeV neutrinos from FRBs. As no significant correlation between IceCube neutrinos and FRBs has been found, we present the first limit on MeV neutrino emission from FRBs and the most constraining limits for neutrinos with GeV to TeV energies. We also describe the prospects for future IceCube neutrino searches coinciding with FRB detections from next generation radio interferometers.

Figures

Figures reproduced from arXiv: 1909.00078 by the authors.

**Figure 1.** Simulation of the various types of IceCube events analyzed in these analyses. Large fluxes of MeV neutrinos can lead to short tracks that interact with at most one DOM (left). At higher energies, individual charged-current interactions from incident muon-neutrinos lead to long tracks that can be reconstructed. The color scale indicates timing, with early hits in red and later hits in blue. The right shows an example… view at source ↗

**Figure 3.** The event rate for these data is significantly larger than that of the six-year analysis, due mostly to an increase in the number of muons produced in the atmosphere passing the event selection. The increase is largest in the Southern sky (cos(θ) > 0), and is nearly a factor of 105 larger at zenith. Features in the Southern sky are the result of selection methods. 2.2 Search for temporal MeV neutrino coincidence … view at source ↗

**Figure 4.** Upper limits on the time-integrated neutrino flux per FRB for a range of ∆T, assuming powerlaw spectra of E −3 (left). The blue line in the stacking limits is produced by dividing IceCube’s entire astrophysical νµ flux [19] among a homogeneous class of 3,000 FRBs per day [10]. We also constrain the maximum time-integrated neutrino flux among 28 FRBs, assuming the same spectrum (right). The error bands on these limi… view at source ↗

Figures from the paper (2 more)

**Figure 6.** Figure 6: Upper limits (90% C.L.) on the time-integrated flux of anti-electron neutrinos from FRBs. Each line represents a limit for an individual FRB of the 21 FRBs considered in the analysis, assuming the fiducial model for neutrino emission with a mean neutrino energies of 15…

**Figure 7.** Figure 7: Stacking sensitivity projections for an isotropically distributed class of FRBs, with number of radio detections indicated in the legend. For FRBs in the Northern Sky, sensitivity to a class of FRBs with an E −3 spectrum will surpass the limit set by the diffuse νµ flu…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 15 canonical work pages

[1]

IceCube, Fermi-LAT, MAGIC, AGILE, ASAS-SN, HA WC, H.E.S.S., INTEGRAL, Kanata, Kiso, Kapteyn, Liverpool Telescope, Subaru, Swift NuSTAR, VERITAS, VLA/17B-403 Collaboration, M. G. Aartsen et al., Science 361 (2018) eaat1378

work page 2018
[2]

IceCube Collaboration, M. G. Aartsen et al., Science 361 (2018) 147–151

work page 2018
[3]

Albert et al., Astrophys

ANTARES, IceCube, Pierre Auger, LIGO Scientiﬁc, VirgoCollaboration, A. Albert et al., Astrophys. J. 850 (2017) L35

work page 2017
[4]

IceCube Collaboration, M. G. Aartsen et al., Astrophys. J. 809 (2015) 98

work page 2015
[5]

IceCube Collaboration, M. G. Aartsen et al., Phys. Rev. Lett. 115 (2015) 081102

work page 2015
[6]

IceCube Collaboration, M. G. Aartsen et al., Eur . Phys. J.C79 (2019) 234

work page 2019
[7]

IceCube Collaboration, M. G. Aartsen et al., Submitted to: Phys. Rev. Lett. (2018)

work page 2018
[8]

D. R. Lorimer, arXiv:1811.00195

arXiv

Show all 22 references

[9]

E. F. Keane, arXiv:1811.00899

arXiv
[10]

Callister, J

T. Callister, J. Kanner, and A. Weinstein, Astrophys. J. 825 (2016) L12

2016
[11]

Platts, A

E. Platts, A. Weltman, A. Walters, S. P. Tendulkar, J. E. B. Gordin, and S. Kandhai, arXiv:1810.05836

arXiv
[12]

IceCube Collaboration, M. G. Aartsen et al., JINST 12 (2017) P03012

2017
[13]

Fahey, A

S. Fahey, A. Kheirandish, J. Vandenbroucke, and D. Xu, Astrophys. J. 845 (2017) 14

2017
[14]

IceCube Collaboration, M. G. Aartsen et al., Astrophys. J. 857 (2018) 117

2018
[15]

Abbasi et al., Astron

IceCube Collaboration, R. Abbasi et al., Astron. Astrophys. 535 (2011) A109. [Erratum: Astron. Astrophys.563,C1(2014)]

2011
[16]

Albert et al., Phys

ANTARES Collaboration, A. Albert et al., Phys. Rev. D96 (2017) 082001

2017
[17]

Albert et al., arXiv:1807.04045

ANTARES Collaboration, A. Albert et al., arXiv:1807.04045

arXiv
[18]

IceCube Collaboration, M. G. Aartsen et al., The IceCube Neutrino Observatory - Contributions to ICRC 2015 Part V: Neutrino Oscillations and Supernova Searches , 2015. arXiv:1510.05227

2015 arXiv
[19]

IceCube Collaboration, M. G. Aartsen et al., arXiv:1710.01191

arXiv
[20]

Totani, K

T. Totani, K. Sato, H. E. Dalhed, and J. R. Wilson, Astrophys. J. 496 (1998) 216–225

1998
[21]

Amiri et al., arXiv:1803.11235

CHIME/FRB Collaboration, M. Amiri et al., arXiv:1803.11235

arXiv
[22]

L. B. Newburgh et al., Proc. SPIE Int. Soc. Opt. Eng. 9906 (2016) 99065X. 8

2016

Pith tools

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

[1] [1]

IceCube, Fermi-LAT, MAGIC, AGILE, ASAS-SN, HA WC, H.E.S.S., INTEGRAL, Kanata, Kiso, Kapteyn, Liverpool Telescope, Subaru, Swift NuSTAR, VERITAS, VLA/17B-403 Collaboration, M. G. Aartsen et al., Science 361 (2018) eaat1378

work page 2018

[2] [2]

IceCube Collaboration, M. G. Aartsen et al., Science 361 (2018) 147–151

work page 2018

[3] [3]

Albert et al., Astrophys

ANTARES, IceCube, Pierre Auger, LIGO Scientiﬁc, VirgoCollaboration, A. Albert et al., Astrophys. J. 850 (2017) L35

work page 2017

[4] [4]

IceCube Collaboration, M. G. Aartsen et al., Astrophys. J. 809 (2015) 98

work page 2015

[5] [5]

IceCube Collaboration, M. G. Aartsen et al., Phys. Rev. Lett. 115 (2015) 081102

work page 2015

[6] [6]

IceCube Collaboration, M. G. Aartsen et al., Eur . Phys. J.C79 (2019) 234

work page 2019

[7] [7]

IceCube Collaboration, M. G. Aartsen et al., Submitted to: Phys. Rev. Lett. (2018)

work page 2018

[8] [8]

D. R. Lorimer, arXiv:1811.00195

arXiv