Constraints on flavor-dependent long-range interactions of high-energy astrophysical neutrinos

Ashish Narang; Mauricio Bustamante; Sanjib Kumar Agarwalla; Sudipta Das

REVIEW 3 major objections 6 minor 18 references

Constraints on flavor-dependent long-range interactions of high-energy astrophysical neutrinos

T0 review · 3 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Flavor-composition measurements by IceCube place the strongest current limits on long-range, flavor-dependent neutrino interactions mediated by ultra-light $Z'$ bosons in four anomaly-free $U(1)'$ models, and IceCube-Gen2 is projected to…

desk verdict A clean, useful Gen2 projection paper, but the headline 'stringent constraints' outruns what is actually computed: fixed pion-decay source composition only. read the letter →

arxiv 2501.12662 v1 pith:YFMFQZWA submitted 2025-01-22 hep-ph astro-ph.HEhep-ex

Sanjib Kumar Agarwalla , Mauricio Bustamante , Sudipta Das , Ashish Narang This is my paper · ORCID

classification hep-phastro-ph.HEhep-ex

keywords astrophysicalneutrinosflavorcompositionlong-rangeinteractionU(1)'gaugemodelsZ'bosonIceCubeCube-Gen2neutrinooscillations

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

This paper claims that the flavor composition of high-energy astrophysical neutrinos can serve as a long-baseline probe of new, extremely feeble long-range interactions between neutrinos and ordinary matter. It considers four anomaly-free $U(1)'$ extensions of the Standard Model in which an ultra-light $Z'$ boson mediates a flavor-dependent potential, and it adds that potential to the standard neutrino oscillation Hamiltonian. Using IceCube's 8-year flavor-composition estimates, the paper derives 95% C.L. upper limits on the interaction potential between $1.79$ and $4.41 \times 10^{-19}$ eV depending on the model, which it says improve on existing limits. A projected 15 years of IceCube plus 10 years of IceCube-Gen2 would tighten the limits by roughly a factor of two. A sympathetic reader should care because flavor ratios measured after gigaparsec journeys turn a neutrino telescope into a laboratory for ultra-light bosons that are difficult to reach any other way.

What carries the argument

The central object is the long-range interaction potential matrix $V^f_{\alpha\beta} = \delta_{\alpha\beta} a_f b_\alpha V_0$, a diagonal matrix in flavor space whose entries are fixed by the $U(1)'$ charges of the matter fermions and neutrinos. The overall scale $V_0$ is built from the coupling $g'$, the mediator mass $m'_Z$, and the density of fermions in the Earth, Moon, Sun, Milky Way, and cosmological matter, with interaction range set by $1/m'_Z$. Adding $V_{\rm LRI}$ to the vacuum and standard matter Hamiltonians and diagonalizing the sum gives the averaged oscillation probabilities $\bar P_{\alpha\beta} = \sum_i |U^m_{\alpha i}|^2 |U^m_{\beta i}|^2$, which convert the assumed source flavor ratio into the predicted Earth flavor fractions that are compared with IceCube's measurements. For example, the $L-3L_\mu$ model gives $V_{\rm LRI}^{(L-3L_\mu)} = \mathrm{diag}(0,3V,0)$, while $L_e-L_\mu$ gives $\mathrm{diag}(V,-V,0)$.

What would settle it

Recompute the 95% C.L. limits using the latest public IceCube flavor-composition contours, for example from the 10-year high-energy starting event sample, instead of the 8-year estimates used in the paper; if those contours are materially wider, the claimed improvement over existing limits would vanish. Conversely, a single well-measured source with pion-dominant production whose Earth flavor ratio deviates from the standard oscillation prediction by more than the quoted uncertainty would contradict the limits.

Watch

Extended reading notes

Core claim

The paper's central claim is that IceCube's flavor-composition data are precise enough to set the tightest current bounds on flavor-dependent long-range neutrino interactions in the four anomaly-free lepton-number models $L-3L_\mu$, $L-3L_\tau$, $L_e-L_\mu$, and $L_e-L_\tau$. With a pion-decay source composition of $(1/3:2/3:0)$, standard oscillations give an almost democratic $(1:1:1)$ flavor ratio at Earth; a long-range interaction potential comparable to or larger than $\Delta m^2/(2E)$ shifts this ratio and, in the strong-potential limit, pushes it back toward the source ratio. The paper computes the resulting flavor compositions, convolves them with the IceCube and IceCube-Gen2 flavor-measurement contours, and integrates over oscillation-parameter priors to obtain 95% C.L. upper limits on the potential. Those limits, listed in Table II, are the quantitative content of the paper, and the corresponding exclusion contours in the coupling-mass plane are the figures a reader should look at.

Load-bearing premise

The limits assume every astrophysical source produces neutrinos by pion decay, giving a source flavor ratio of one part electron neutrino, two parts muon neutrino, and no tau neutrino; if real sources use other production channels, the baseline Earth flavor composition changes and the inferred limits shift.

Editorial extensions

If this is right

If the limits are correct, any ultra-light $Z'$ boson with one of these charge assignments must have a coupling and mass combination that lies below the exclusion curves in Fig. 2, or IceCube would have seen an anomalous flavor ratio.
The projected combination of 15 years of IceCube and 10 years of IceCube-Gen2 is claimed to improve the limits by about a factor of two, reaching $0.731$ to $1.69 \times 10^{-19}$ eV.
Flavor-composition measurements become a complementary, cosmological-baseline tool for bounding new neutrino-matter forces, alongside terrestrial oscillation, atmospheric, solar, and reactor experiments.
The bounds apply over interaction ranges that extend, depending on $m'_Z$, from Earth-scale distances up to the causal horizon, with step-like features from the Sun, the Galactic center, and horizon-scale matter.
Among the four models, the $L_e-L_\tau$ symmetry receives the tightest current limit ($1.79 \times 10^{-19}$ eV), showing that the constraint strength is model-dependent.

Reading between the lines

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

The paper fixes the source flavor ratio to pion decay, so the numbers in Table II are conditional on that production mechanism; a muon-damped or neutron-decay source would change the baseline Earth ratio and likely shift the limits, and the paper does not quantify by how much.
Because the LRI effect competes with $\Delta m^2/(2E)$, the distortion of the flavor ratio is energy dependent; a detector that reconstructs flavor as a function of energy could search for a tilt in the ratio rather than a single average, which the paper does not exploit.
The same flavor-composition data could also be used to constrain other beyond-standard-model propagation effects with similar signatures, such as invisible neutrino decay or quantum decoherence, although the paper does not discuss them.
If a future source with an independently known production mechanism is observed, a measurement of its Earth flavor ratio could separate the LRI contribution from standard oscillations more cleanly than the diffuse all-sky average used here.

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

Dear colleague,

What you should know first: this proceedings paper gives IceCube-Gen2 projections for constraints on flavor-dependent U(1)' long-range interactions. The genuinely new numbers are in Table II and the right panel of Fig. 2, and they are not in the earlier JHEP work. If you build BSM models with ultra-light Z' bosons, those projected limits are useful benchmarks to have on hand. The calculation is standard: diagonalize the propagation Hamiltonian, average over the fast oscillations, build a Bayesian posterior over the potential V0. I do not see a technical error in that core machinery.

What the paper does well: it is candid about being an extension of ref. [4], it spells out the four anomaly-free models and the Hamiltonian clearly, and it separates current IceCube 8-year limits from the 15-year + Gen2 projection. The choice to fold in future oscillation-parameter improvements from DUNE, T2HK, and JUNO is a sensible addition. For a proceedings article, the level of detail is appropriate.

The soft spots are real but not fatal. The main one is the fixed source composition. The analysis assumes pion decay, (1/3:2/3:0), from start to finish. The stress-test note is right: neutron-decay and muon-damped sources remain viable given current flavor-composition contours, and because the LRI potential is flavor-dependent, those different baselines would map the same measured contour onto different V0 bounds. The paper does not quantify that shift, so the Section VI statement that IceCube 'put stringent constraints ... improving over the existing limits' is broader than what was actually computed. What we have are limits for a single production scenario, not model-level U(1)' constraints. Second, the IceCube 8-year limits rest on ref. [10], a private reanalysis with overlapping authorship, and no code or data are shipped. That is common in this area, but it does limit independent verification. Third, the comparison with existing limits in Fig. 2 draws on very different observables (atmospheric, solar, reactor), so the 'improving' language is suggestive, not rigorous.

None of this undermines the core calculation. The paper is a competent extension that deserves to be read and cited for its Gen2 projections, but those projections should be treated as pion-decay benchmarks. The authors could strengthen this substantially by scanning over a few representative source compositions and showing how the limits move.

Recommendation: yes, send to peer review. It is a legitimate calculation with new projected numbers, and a referee can ask for the source-composition scan without sending the authors back to square one. I would flag that missing scan in the report, but this is not a desk-reject.

Best, [Your name]

Referee Report

3 major / 6 minor

Summary. The manuscript studies flavor-dependent long-range interactions (LRI) mediated by an ultra-light Z' boson in anomaly-free U(1)' models. It computes the flavor composition at Earth for astrophysical neutrinos under a constant LRI potential V0, taking a fixed pion-decay source composition, and compares the resulting flavor ratios with IceCube 8-year flavor-composition estimates and with projected IceCube 15-year plus IceCube-Gen2 10-year measurements. Using a Bayesian posterior with marginalization over oscillation parameters, it derives 95% C.L. upper limits on the LRI potential for four U(1)' models (Table II), translates these into g'-m_Z' planes (Figure 2), and claims that IceCube already improves on existing limits and that IceCube-Gen2 improves by about a factor of two. The paper is a concise proceedings-style presentation and draws heavily on the authors' previous work in ref. [4].

Significance. The result is potentially useful for interpreting current and future neutrino-telescope flavor measurements as probes of ultra-light new physics. Its main strengths are the transparent one-parameter framework for LRI potentials, the use of public IceCube flavor-composition contours, and the explicit projection to IceCube-Gen2, which offers a falsifiable target for future searches. If the central claim is correct, the derived limits would be among the most stringent constraints on flavor-dependent U(1)' long-range interactions in the considered mass range. However, the limits are conditional on a single assumed source production scenario and on several simplified detector assumptions, so the model-level claim in Section VI is stronger than the analysis currently supports.

major comments (3)

[Section III and Eq. (7)] The analysis fixes the source flavor composition to the pion-decay case, fS=(1/3:2/3:0), and Eq. (7) marginalizes only over the oscillation parameters, not over the source composition. Other production scenarios such as neutron decay or muon-damped sources remain viable at the current level of IceCube flavor precision, and because the LRI potential acts as a flavor-dependent Hamiltonian term, the same measured flavor contour maps to different V0 upper limits for different fS. The Section VI claim that IceCube puts stringent constraints on the U(1)' models therefore needs either a quantitative study of the source-composition dependence or a softer formulation that explicitly limits the claim to pion-decay sources.
[Section II, Table I and Eq. (6)] In Eq. (6), the LRI matrices for L-3L_mu and L-3L_tau are written with +3V on the muon and tau diagonal entries, respectively, but Table I gives b_mu=-2 and b_tau=-2, and Eq. (2) then yields -2V0 for the same models. Please define V relative to V0 in Eq. (6) or correct the signs and coefficients. As written, the equations are internally inconsistent, and the limits for these two models in Table II and Figure 2 are not reproducible without additional assumptions.
[Section IV] Section IV states that the computed flavor composition is averaged over neutrino energies in the range 25 TeV to 2.8 PeV, but it does not specify the energy weighting or how the energy-dependent vacuum Hamiltonian is handled after the LRI contribution is included. Since the oscillation probability is not linear in E after diagonalizing the full Hamiltonian, the averaged flavor composition is not uniquely defined without this information. Please state the assumed spectrum and, ideally, test the sensitivity of the final limits to the weighting choice.

minor comments (6)

[Section III] The sentence 'flavor composition at the Earth calculated using eq. (1)' should refer to Eq. (3), not Eq. (1).
[Section III] There is a typo in 'thr neutrino mixing matrix'; it should read 'the neutrino mixing matrix'.
[Figure 2] The label 'Work in progress' appears in both panels of Figure 2 but is not explained in the text or caption.
[Section IV] The text should make explicit that the IceCube flavor-composition estimates are not an official IceCube collaboration result but the reanalysis of ref. [10] by Song et al.
[Section V] The sentence 'the addition of the projected measurements from IceCube-Gen2 with 8 years of IceCube estimates leads to a significant improvement' is grammatically unclear; it should be rephrased to separate the 8-year IceCube estimate from the projected additions.
[Eq. (7)] The notation V_alpha_beta in Eq. (7) should be defined more carefully, since Eq. (2) uses the double-index V^f_alpha_beta and Eq. (6) uses a single V; the relationship between these symbols is not stated.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the LRI potential is the fitted target, not an input, and the comparison uses external IceCube flavor-composition data.

full rationale

The paper's derivation chain is self-contained. The long-range-interaction potential enters the propagation Hamiltonian in Eq. (5), the flavor composition at Earth is computed by diagonalizing that Hamiltonian via Eq. (3), and the result is compared to external IceCube flavor-composition contours using the posterior in Eq. (7). The constrained quantity V0 is the target parameter, not an input: no flavor-composition data point is used to define the model, and the 95% limits in Table II are obtained by fitting V0 to the external contour. The pion-decay source composition (1/3:2/3:0) is a stated assumption, and the limits are conditional on it, but this is a robustness limitation rather than circular reasoning because the assumption is not inferred from the data being predicted. The self-citations to refs. [4] and [7] for the celestial-body potential are not load-bearing in a circular sense: the potential formula is given explicitly in Eq. (4), and the cited prior work is not invoked as an unexamined uniqueness theorem. The paper also discloses that part of the results already appear in ref. [4], which is transparency rather than a hidden input. Oscillation parameters come from external global fits, and the IceCube contour is a published data analysis; overlap in authorship on ref. [10] does not make the comparison circular because the contour is not constructed from the LRI model being tested. No step reduces by construction to a fitted parameter renamed as a prediction.

Assumptions & free parameters 1 free parameters · 6 assumptions · 1 invented entities

The analysis rests on standard neutrino oscillation physics plus several domain choices: the U(1)' model set, a fixed pion-decay source composition, a particular treatment of celestial-body potentials from refs. [4,7], normal mass ordering, and energy/antineutrino averaging. The only fitted parameter is the LRI potential V0, which is the target of the constraint. No new entity is invented; the Z' mediator is inherited from existing BSM models.

free parameters (1)

LRI potential strength V0 = 95% C.L. upper limits: 1.79-4.41e-19 eV (IceCube 8 yr); 0.731-1.69e-19 eV (IceCube 15 yr + Gen2 10 yr)
This is the target parameter of the constraint analysis. It is varied and marginalized over; the paper reports upper limits, not a preferred value. It is not a derived constant.

assumptions (6)

standard math Standard three-flavor neutrino oscillations with PMNS mixing and averaged probabilities P = sum_i |U_alpha_i|^2 |U_beta_i|^2
Used in Section III to compute flavor composition; assumes decoherence over astrophysical baselines.
domain assumption U(1)' anomaly-free models with tree-level Z' mediation and no Z-Z' mixing
Section II, Eq. (1) and Table I list charge assignments; the analysis ignores sub-leading Z-Z' mixing, as stated in Section II.
domain assumption Pion-decay source composition (1/3:2/3:0)
Section III: 'This work considers the pion decay scenario as the neutrino production mechanism at the source.' Not tested against other production mechanisms.
domain assumption Matter distribution of Earth, Sun, Milky Way, and cosmological average potential
Section III: VLRI = V_Earth + V_Sun + V_MW + <V_cos>; the calculation follows refs. [4,7] and is assumed correct.
domain assumption Normal mass ordering (NMO) and Gaussian priors on oscillation parameters from NuFit 5.1
Section IV: 'assuming normal mass ordering (NMO) of neutrino'; inverted ordering and non-Gaussian systematics are not included.
domain assumption Neutrino/antineutrino and energy averaging over 25 TeV to 2.8 PeV with an unbroken power law
Section IV: the computed flavor composition is averaged over neutrino and antineutrino modes and over energies in this range; a spectral break would affect the limits.

invented entities (1)

Ultra-light Z' gauge boson independent evidence
purpose: Mediates the long-range flavor-dependent neutrino-matter interaction; inherited from U(1)' BSM models, not introduced by this paper.
Figure 2 maps limits onto the coupling-mass plane, giving a falsifiable handle for other experiments. The paper does not invent new entities beyond adopting the existing Z' framework.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Constraints on flavor-dependent long-range interactions of high-energy astrophysical neutrinos." pith.science (2026). https://pith.science/paper/YFMFQZWA

@misc{pith2026250112662,
  author       = {Pith},
  title        = {Pith review of: Constraints on flavor-dependent long-range interactions of high-energy astrophysical neutrinos},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YFMFQZWA}},
  note         = {Machine review of arXiv:2501.12662}
}

read the original abstract

Astrophysical neutrinos with energy in the TeV-PeV range traverse megaparsecs (Mpc) to gigaparsecs (Gpc) scale distances before they reach the Earth. Tiny physics effects that get accumulated over these large propagation paths during their journey may become observable at the detector. If there is some new interaction between neutrinos and the background matter, that can potentially affect the propagation of the astrophysical neutrinos. One such possible case is the flavor-dependent long-range interaction of neutrinos, which can affect the standard neutrino flavor transition, modifying the flavor composition of the astrophysical neutrinos at Earth. Using the present-day and future projection of the flavor-composition measurements of IceCube and IceCube-Gen2 along with the present and future measurement of the oscillation parameters, we explore the sensitivity of these experiments to probe long-range neutrino interaction with matter.

Figures

Figures reproduced from arXiv: 2501.12662 by the authors.

**Figure 1.** FIG. 1. Neutrino flavor composition at Earth as a function of LRI potential induced by all the anomaly-free models considered in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

**Figure 2.** FIG. 2. 95% C.L. limits on the coupling strength of the LRI arising from gauged [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 15 linked inside Pith

[4]

S. K. Agarwalla, M. Bustamante, S. Das, and A. Narang, JHEP 08, 113 (2023), arXiv:2305.03675 [hep-ph]. 5

arXiv 2023
[10]

N. Song, S. W. Li, C. A. Arg¨ uelles, M. Bustamante, and A. C. Vincent, JCAP 04, 054 (2021), arXiv:2012.12893 [hep-ph]

arXiv 2021
[1]

A. S. Joshipura and S. Mohanty, Phys. Lett. B 584, 103 (2004), arXiv:hep-ph/0310210

arXiv 2004
[2]

Bandyopadhyay, A

A. Bandyopadhyay, A. Dighe, and A. S. Joshipura, Phys. Rev. D 75, 093005 (2007), arXiv:hep-ph/0610263

arXiv 2007
[3]

M. C. Gonzalez-Garcia, P. C. de Holanda, E. Masso, and R. Zukanovich Funchal, JCAP01, 005 (2007), arXiv:hep- ph/0609094

arXiv 2007
[5]

Singh, M

M. Singh, M. Bustamante, and S. K. Agarwalla, JHEP 08, 101 (2023), arXiv:2305.05184 [hep-ph]

arXiv 2023
[6]

S. K. Agarwalla, M. Bustamante, M. Singh, and P. Swain, JHEP 09, 055 (2024), arXiv:2404.02775 [hep- ph]

arXiv 2024
[7]

Bustamante and S

M. Bustamante and S. K. Agarwalla, Phys. Rev. Lett. 122, 061103 (2019), arXiv:1808.02042 [astro-ph.HE]

arXiv 2019

Show all 18 references

[8]

Esteban, M

I. Esteban, M. C. Gonz´ alez-Garc ´ ıa, M. Maltoni, T. Schwetz, and A. Zhou, JHEP 09, 178 (2020), arXiv:2007.14792 [hep-ph]

2020 arXiv
[9]

NuFIT 5.1 (2021), http://www.nu-fit.org/

2021
[11]

M. G. Aartsen et al. (IceCube-Gen2), J. Phys. G 48, 060501 (2021), arXiv:2008.04323 [astro-ph.HE]

2021
[12]

M. G. Aartsen et al. (IceCube), Astrophys. J. 809, 98 (2015), arXiv:1507.03991 [astro-ph.HE]

2015 arXiv
[13]

Coloma, M

P. Coloma, M. C. Gonz´ alez-Garc ´ ıa, and M. Maltoni, JHEP 01, 114 (2021), arXiv:2009.14220 [hep-ph]

2021 arXiv
[14]

Mitsuka et al.(Super-Kamiokande), Phys

G. Mitsuka et al.(Super-Kamiokande), Phys. Rev. D 84, 113008 (2011), arXiv:1109.1889 [hep-ex]

2011 arXiv
[15]

Ohlsson, Rept

T. Ohlsson, Rept. Prog. Phys. 76, 044201 (2013), arXiv:1209.2710 [hep-ph]

2013 arXiv
[16]

M. C. Gonz´ alez-Garc ´ ıa and M. Maltoni, JHEP09, 152 (2013), arXiv:1307.3092 [hep-ph]

2013 arXiv
[17]

Baryakhtar, R

M. Baryakhtar, R. Lasenby, and M. Teo, Phys. Rev. D 96, 035019 (2017), arXiv:1704.05081 [hep-ph]

2017 arXiv
[18]

Arkani-Hamed, L

N. Arkani-Hamed, L. Motl, A. Nicolis, and C. Vafa, JHEP 06, 060 (2007), arXiv:hep-th/0601001

2007 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [4]

S. K. Agarwalla, M. Bustamante, S. Das, and A. Narang, JHEP 08, 113 (2023), arXiv:2305.03675 [hep-ph]. 5

arXiv 2023

[2] [10]

N. Song, S. W. Li, C. A. Arg¨ uelles, M. Bustamante, and A. C. Vincent, JCAP 04, 054 (2021), arXiv:2012.12893 [hep-ph]

arXiv 2021

[3] [1]

A. S. Joshipura and S. Mohanty, Phys. Lett. B 584, 103 (2004), arXiv:hep-ph/0310210

arXiv 2004

[4] [2]

Bandyopadhyay, A

A. Bandyopadhyay, A. Dighe, and A. S. Joshipura, Phys. Rev. D 75, 093005 (2007), arXiv:hep-ph/0610263

arXiv 2007

[5] [3]

M. C. Gonzalez-Garcia, P. C. de Holanda, E. Masso, and R. Zukanovich Funchal, JCAP01, 005 (2007), arXiv:hep- ph/0609094

arXiv 2007

[6] [5]

Singh, M

M. Singh, M. Bustamante, and S. K. Agarwalla, JHEP 08, 101 (2023), arXiv:2305.05184 [hep-ph]

arXiv 2023

[7] [6]

S. K. Agarwalla, M. Bustamante, M. Singh, and P. Swain, JHEP 09, 055 (2024), arXiv:2404.02775 [hep- ph]

arXiv 2024

[8] [7]

Bustamante and S

M. Bustamante and S. K. Agarwalla, Phys. Rev. Lett. 122, 061103 (2019), arXiv:1808.02042 [astro-ph.HE]

arXiv 2019