REVIEW 4 major objections 6 minor 5 references
The Baikal-GVD neutrino telescope: First results of multi-messenger studies
T0 review · 4 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read The Baikal-GVD telescope, with only two clusters operational in 2017, reports no neutrinos coincident with GW170817 and sets 90% C.L. upper limits on the neutrino fluence.
desk verdict Baikal-GVD's first multi-messenger search: a credible non-detection, but the fluence limits rest on a referenced calculation that this paper itself does not make checkable. 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 mechanism is the cascade detection mode of Baikal-GVD: Cherenkov light from the particle cascades produced by high-energy neutrino interactions in water is reconstructed into direction and energy. The median angular resolution is about 4.5 degrees and the energy resolution averaged over an $E^{-2}$ spectrum of electron neutrinos is about 30%, with roughly 90% of reconstructed cascade events falling between 5 TeV and 10 PeV. Around this reconstruction sits an alert-search strategy that fixes a direction and time window for each external trigger, applies cascade-quality cuts, and searches both prompt and delayed windows. That procedure is what turns the recorded hits into a null result, and the null result is what is then converted into the quoted fluence limits.
What would settle it
One concrete check is to rerun the cascade reconstruction on the archived 2017 Baikal-GVD data in the $\pm 500$ second and 14-day windows around GW170817 with the same cuts and see whether any event with reconstructed energy above 100 TeV falls inside the spatial selection; any such event would contradict the non-detection claim. A second check is to recompute the expected atmospheric background in the 74 to 150 degree zenith band used for the delayed search with the detector simulation: if the background is substantially larger than the assumed value, the 90% C.L. fluence limits would be understated.
Extended reading notes
Core claim
The central claim is that Baikal-GVD, with only two of its planned clusters operational during 2017, detected no neutrinos coincident with GW170817/GRB170817A and therefore established 90% confidence upper limits on the high-energy neutrino fluence from that neutron-star merger. In cascade mode the source lay slightly below the horizon at the detector (zenith angle 93.3 degrees), and no events were found either in the prompt $\pm 500$ second window or in a delayed window of 14 days, restricted to the zenith range from 74 to 150 degrees. The same analysis applied to the blazar TXS 0506+056 in the IC170922 follow-up found no coincident neutrinos in $\pm 1$ hour and $\pm 1$ day windows. The fluence limits assume an $E^{-2}$ neutrino spectrum and equal fluence in all three neutrino flavors, with the details of the limit calculation deferred to reference [5].
Load-bearing premise
The upper limits assume the neutrino flux falls as $E^{-2}$, that all three neutrino flavors contribute equally, and that the detector-response model from reference [5] is correct; if any of these assumptions fails, the quoted 90% C.L. limits would change.
Editorial extensions
If this is right
- With two clusters operating in 2017, the cascade channel is already sensitive enough to place 90% C.L. upper limits on the neutrino fluence of a gravitational-wave source.
- The lack of coincident neutrinos in the 14-day delayed window around GW170817 extends the constraint beyond the prompt emission phase.
- The same cascade search found no neutrinos from the blazar TXS 0506+056 in either a one-hour or a one-day window, adding an independent non-detection to the IC170922 picture.
- Following external neutrino alerts, no time-direction correlations were found in six months of observations, with an estimated atmospheric background toward a given direction of about 1.75 events per day inside a 10-degree cone and 0.5 events per day inside a 5-degree cone for three clusters.
- With the array growing and an online alert system under development, the same selection logic is intended to run in real time and issue alerts to other observatories.
Reading between the lines
- The published fluence limits are benchmark bounds rather than model-independent exclusions: if the true neutrino spectrum is softer than $E^{-2}$ or the flavors are not equally populated, the 90% C.L. limits would shift outside the quoted bands.
- The paper's own background estimates suggest that at the three-cluster scale a handful of cascade events near an alert direction remains consistent with atmospheric muons, so individual coincidences should be treated as candidate associations until the array grows and the live time lengthens.
- A combined analysis of null results from neutrino telescopes with complementary sky coverage, Baikal-GVD among them, would tighten the joint upper limits on high-energy neutrino emission from neutron-star mergers without waiting for a single detection.
- If a future binary neutron-star merger is observed while the source is above the Baikal-GVD horizon, the same cascade search would have substantially better sensitivity, because the GW170817 search was handicapped by a source slightly below the horizon.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper, a proceedings contribution from the 36th ICRC, reports the first multi-messenger follow-up searches performed with the Baikal-GVD neutrino telescope while it was under construction. Using two operating clusters in 2017 and the cascade detection mode, the authors searched for high-energy neutrinos in coincidence with GW170817/GRB170817A and with the blazar TXS 0506+056 (IC170922). No neutrino candidates were found that are spatially and temporally coincident with either source. Assuming an E^-2 neutrino spectrum and equal flavor partition, the paper presents 90% confidence-level upper limits on the neutrino fluence from GW170817 for both the ±500 s prompt window and a 14-day delayed window. The paper also briefly describes the status of the Baikal-GVD alert system and a follow-up of ANTARES neutrino alerts, again with no observed coincidences.
Significance. If the reported limits are correct, this is a valuable early demonstration that a partial neutrino detector under construction can already contribute to multi-messenger astronomy, placing meaningful constraints on high-energy neutrino emission from a binary neutron star merger. The non-detection itself is credible, and the paper explicitly states the spectral and flavor assumptions used in the interpretation. The paper is also transparent about its preliminary nature and about the fact that many details are deferred to a companion paper. However, the central quantitative result — the 90% C.L. fluence limits — is not derived in this manuscript; the reader is referred to Ref. [5] without being given the exposure, effective volume, search cone, or energy threshold needed to verify the calculation. In addition, the paper's own statement that the event selection cuts vary with the time window introduces a potential inconsistency in how the prompt and delayed limits are computed. These gaps make the paper more a status report than a self-contained measurement, but the underlying approach is sound and the issues are addressable in a revision.
major comments (4)
- [Section 2] The 90% C.L. fluence limits shown in Fig. 2 (left) are asserted without derivation. The text only states that "more details of analysis are presented in Ref. [5]" and that "assuming an E^-2 spectral behavior and equal fluence in neutrino flavors, upper limits at 90% c.l. have been derived." Since these limits are the central quantitative claim of the paper, the manuscript should at least specify the exposure (number of clusters, livetime), the search cone radius, the energy threshold, and the effective volume or acceptance used, or it should reproduce the calculation in an appendix. Without this information, the reader cannot verify that the limits correspond to the analysis described in the text.
- [Sections 1 and 2] Section 1 states that "the values of cuts are slightly varied in dependence on time windows in the follow up analysis, so that they are weaker for shorter times." Section 2 then presents separate limits for the prompt (±500 s) and delayed (14-day) emission windows. Because the effective volume of the detector depends on the event selection cuts, the two limits must have been computed with different selection efficiencies. The paper does not state how these time-dependent cuts enter the limit calculation, nor whether Ref. [5] used the same varying cuts. Without this information, the two limit curves in Fig. 2 (left) cannot be interpreted consistently, and the normalization of both could be wrong if a single effective volume was used.
- [Section 2] The paper reports no systematic uncertainties on the fluence limits. The source NGC 4993 was at a zenith angle of 93.3°, i.e., slightly below the horizon for Baikal-GVD, so the cascade effective volume is affected by energy-dependent Earth absorption and by the assumed neutrino cross-section. In addition, uncertainties in the energy scale, angular resolution, and event selection efficiency could shift the limits by an amount that is not quantified. Since the limits are intended for comparison with theoretical models of neutron-star merger neutrino emission, the paper should include at least a discussion of the dominant systematic errors and their effect on the 90% C.L. limits.
- [Section 2] The search for neutrino events "associated with these sources" over a fixed direction is not fully specified: the paper does not state the angular acceptance radius used to define spatial coincidence, nor the energy range over which the limits apply. Given the quoted median cascade angular resolution of about 4.5°, the meaning of "no events spatially coincident" is ambiguous. The paper should state the search cone (for example the 90% angular containment radius) and the energy range for the GW170817 search so that the non-detection can be properly interpreted.
minor comments (6)
- [Section 1] The text reads "TMV A implementation"; this should be "TMVA" (Toolkit for Multivariate Analysis).
- [Section 1] The phrase "an E −2.46 spectrum" should be typeset as a power law, e.g., E^{-2.46}, to avoid ambiguity.
- [Section 2] The naming of the blazar is inconsistent: the text uses both "IC170922" and "IC170922A", as well as "TX-0506+056" and "TXS 0506+056"; a single consistent nomenclature should be adopted.
- [Section 2] Figure 2 (right) is described only as a "distribution of number of events in a time window of ±1 hour"; the caption should define the quantity on the horizontal axis and the meaning of the plotted histogram.
- [Section 3] The phrase "half-open cone of 10°" and "half-open cone 5°" is not defined; the authors should specify that the cone is defined by an angular radius around the alert direction and state the exact convention.
- [Section 3] The sentence "In best case of happened coincidence within time window of ±1 day..." is awkwardly worded and should be rewritten for clarity.
Circularity Check
No circularity: the paper reports an observational non-detection and derives upper limits from stated physical assumptions and referenced detector-response calculations.
full rationale
The paper's derivation chain is observational, not self-referential. It searches for neutrino cascades in time and direction coincidence with GW170817/GRB170817A, finds zero events after cuts, and then states that upper limits on the neutrino fluence were derived assuming an E^-2 spectrum and equal flavor partition. These assumptions are external physical inputs, not quantities fitted to the data being used to produce the claimed result. The non-detection is the raw observational outcome; the conversion to fluence limits is deferred to Ref. [5], which is cited for analysis details rather than as a substitute for the present measurement. The paper does not define its observable in terms of the claimed result, does not rename a fitted parameter as a prediction, and does not invoke a uniqueness theorem by the same authors to forbid alternatives. The self-citations to Refs. [3] and [5] provide reconstruction methods and the limit-calculation procedure, and nothing in the quoted text indicates that those references assume the GW170817 fluence limits or the zero-event outcome. The statement that cuts are slightly varied with time windows is a methodological caveat about efficiency changes, not a circular step. Any concern that Ref. [5]'s effective volumes or cuts may not exactly match the reported limits is a correctness or reproducibility risk, not evidence that the result reduces by construction to its inputs.
Assumptions & free parameters
assumptions (4)
- domain assumption The detector simulation correctly models the Cherenkov light from cascade events in Baikal water.
- domain assumption An E^-2 neutrino spectrum and equal flavor partition are assumed to set upper limits.
- domain assumption The background estimate for the ANTARES follow-up is reliable.
- domain assumption The zenith angle restriction 74 degrees < theta < 150 degrees captures the relevant field of view for the delayed GW search.
Cite this review
Pith. "Pith review of The Baikal-GVD neutrino telescope: First results of multi-messenger studies." pith.science (2026). https://pith.science/paper/ZFKGBS7W
@misc{pith2026190805450,
author = {Pith},
title = {Pith review of: The Baikal-GVD neutrino telescope: First results of multi-messenger studies},
year = {2026},
howpublished = {\url{https://pith.science/paper/ZFKGBS7W}},
note = {Machine review of arXiv:1908.05450}
}
read the original abstract
Multi-messenger astronomy is a powerful tool to study the physical processes driving the non-thermal Universe. A combination of observations in cosmic rays, neutrinos, photons of all wavelengths and gravitational waves is expected. The alert system of the Baikal-GVD detector under construction will allow for a fast, on-line reconstruction of neutrino events recorded by the Baikal-GVD telescope and - if predefined conditions are satisfied - for the formation of an alert message to other communities. The preliminary results of searches for high-energy neutrinos in coincidence with GW170817/GRB170817A using the cascade mode of neutrino detection are discussed. Two Baikal-GVD clusters were operating during 2017. The zenith angle of NGC 4993 at the detection time of the GW170817 was 93.3 degrees. No events spatially coincident with GRB170817A were found. Given the non-detection of neutrino events associated with GW170817, upper limits on the neutrino fluence were established.
Figures
Reference graph
Works this paper leans on
-
[5]
A.D. Avrorin et al., JETP Lett. 108, no.12, 787-790 (2018). 3
work page 2018
- [1]
-
[2]
M.G.Aartsen et al., IceCube Coll., Science, 361, eaat1378 (2018)
work page 2018
-
[3]
Dvornický et al., Baikal Coll., these proceedings, Search for cascade events with Baikal-GVD
R. Dvornický et al., Baikal Coll., these proceedings, Search for cascade events with Baikal-GVD
-
[4]
The Baikal-GVD neutrino telescope: muon track events reconstruction
G.B. Safronov et al., "The Baikal-GVD neutrino telescope: muon track events reconstruction", in preparation
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.