A test of tau neutrino interactions with atmospheric neutrinos and K2K

Bottom line: this is a careful constraint paper, and the central claim survives — e–τ NSI at the weak scale are not excluded by atmospheric plus K2K data. The specific numbers in Table I, though, are more fragile than the abstract lets on, and the stress-test note is right about where the fragility sits. What is genuinely new: the full three-dimensional scan over (ε_ee, ε_eτ, ε_ττ) against the complete 1489-day SK phase I sample plus K2K, and the two analytical cutoff conditions, Eqs. (22) and (23), which come from requiring consistency between the low-energy atmospheric sample and the essentially matter-free K2K measurement. The two-state reduction in Sect. III A is derived cleanly and then checked against the full three-flavor numerics — that is a real consistency test, not a restatement. The flat-direction parabola, Eq. (17), is a prediction of the approximation that the fit then confirms. Where it is soft: the ε_ee window. Table I comes from marginalizing ε_ee over −1.6 to 1.6, a range the authors call “an example,” tied to a CHARM bound that only applies if the NSI sit entirely on the right-handed down quark. Footnote 74 concedes that a rigorous treatment needs a simultaneous CHARM reanalysis and a global fit, which they explicitly decline. They also note the contours practically coincide with ε_ee = 1.6, so Table I is essentially that single case. Widen the window and the intervals widen. The abstract's 0.2–0.7 and 1.5–4.0 are already slightly wider than the tabulated 3σ values (0.24–0.68, 1.7–3.9) — a small thing, but it is the number people will quote. The omission of detection effects (Sect. IV A.1) I would not hold against them: it is disclosed, argued, and partially quantified with a 20% cross-section estimate. Setting ε_μτ = 0 and taking ε_eτ real are stated and deferred. Who should read it: anyone setting targets for MINOS, OPERA, or long-baseline NSI sensitivity, and anyone quoting atmospheric NSI bounds. It deserves referees.

The paper's Table I intervals and abstract ranges are obtained by marginalizing over ε_ee restricted to -1.6 ≤ ε_ee ≤ 1.6, a window the authors describe as 'an example' corresponding to a CHARM bound 'if the NSI are present exclusively as flavor-preserving interactions of νe on the right-handed down quark' (Sect. IV B, footnote 74). Footnote 74 concedes that a rigorous treatment would require a simultaneous reanalysis of CHARM with all e-τ NSI couplings and a global fit, which is 'beyond the scope of this work.' The authors note the contours 'practically coincide with those obtained for normal hierarchy with ε_ee fixed at the upper limit, ε_ee = 1.6,' so Table I is essentially the ε_ee = 1.6 case. If the true allowed ε_ee window is wider, the marginalized intervals would widen, and the abstract's 0.2-0.7 and 1.5-4.0 ranges (already wider than the tabulated 3σ values 0.24-0.68 and 1.7-3.9) would shift further. This does not threaten the qualitative claim that O(1) couplings survive, but it means the specific numbers are window-dependent in a way the abstract does not disclose. The reader's detection-effects concern is real but is acknowledged and partially quantified (Sect. IV A.1 gives a 20% cross-section estimate for one point); the ε_ee window is more directly load-bearing for the quoted intervals.

The paper is a careful constraint study. The analytical two-state reduction is derived cleanly and checked against a full three-neutrino numerical fit to the complete SK phase I data set plus K2K. The parabolic flat direction and its cutoffs are genuine predictions of the approximation that the numerics confirms. The central claim — that e-τ NSI at the weak scale survive atmospheric and K2K data while widening the allowed θ23 and Δm² region — is supported by the analysis as stated. The reader's CONDITIONAL verdict is appropriate. I partially agree with the reader's weakest_assumption: the omission of detection effects is a real qualification, but the paper addresses it at length in Sect. IV A.1 and gives a quantitative estimate for one representative point, so it is not purely qualitative. The more directly load-bearing qualification for the headline numbers is the hand-imposed ε_ee marginalization window (footnote 74), which sets Table I and is not rigorously justified. A secondary concern is the absolute χ² of ~148 for ~55 data points, which is not discussed; if this indicates a poor fit, the Δχ² contours could be affected. A third is the abstract-vs-table discrepancy (0.2-0.7 vs 0.24-0.68; 1.5-4.0 vs 1.7-3.9), which should be reconciled. None of these breaks the central claim, which is a bound and is conservative in the sense that the omitted effects are argued to be uncorrelated. The result is an extension of the authors' own earlier work [21] with a fuller scan and updated data, not a new mechanism, and its significance is within neutrino phenomenology.

The paper is a careful and useful contribution. Its main strengths are: (i) a genuine analytic derivation of the flat direction and its cutoffs, obtained from the Hamiltonian independently of the fit and then checked against the full three-neutrino numerical scan — a real consistency test rather than a restatement; (ii) use of the complete SK phase I data set with 3D fluxes and an explicit K2K χ², so the bounds come from external data rather than from a fitted quantity relabeled as a prediction; (iii) an explicit, falsifiable forward look (Eq. (25) and Sect. V) to MINOS and long-baseline appearance/disappearance signals, which gives the result experimental traction; and (iv) the honest and physically important conclusion that O(1) NSI in the e–τ sector is not excluded. The paper is also transparent about what it does not do (detection effects, Sect. IV A.1; simultaneous CHARM reanalysis, footnote 74). The main reservation is that the specific numbers quoted in the abstract and Table I depend on a hand-imposed prior on ε_ee that is described only in the body and in a footnote.

Atmospheric neutrinos oscillate into tau neutrinos as they cross the Earth, so the tau component of the flux is a natural probe of how tau neutrinos scatter on matter. The paper asks whether Super-Kamiokande and K2K data allow nonstandard neutral-current interactions in the $nu_e$-$nu_tau$ sector, and if so, how large they can be. The authors combine a three-neutrino numerical fit with an analytic reduction that shows why the allowed region is a narrow parabola in coupling space rather than a small ball around zero. The payoff is a bound that improves on accelerator limits but still permits couplings of order the standard weak interaction, while shifting the preferred vacuum mixing and mass splitting.

What carries the argument. The load-bearing object is the three-neutrino matter Hamiltonian with NSI entries $epsilon_{ee}$, $epsilon_{e tau}$, and $epsilon_{tau tau}$. Diagonalizing the matter part gives eigenvalues $lambda_{e'}$, $lambda_{tau'}$ and a mixing angle $beta$, with $tan 2beta = 2|epsilon_{e tau}|/(1+epsilon_{ee}-epsilon_{tau tau})$. When one eigenvalue is hierarchically larger than the vacuum term, the electron state decouples and the problem reduces to effective two-neutrino $nu_mu$-$nu_{tau'}$ oscillations with matter potential $lambda_{tau'}$. That reduction explains both the flat parabolic direction and the cutoff: along the parabola the muon survival probability keeps vacuum-like $L/E$ behavior, while $beta$ controls how much the high-energy data force the vacuum parameters away from their low-energy values.

Core claim. The central claim is that atmospheric neutrino data, combined with K2K, do not exclude nonstandard interactions of tau neutrinos with matter at the scale of the standard weak interaction. For fixed $epsilon_{ee}$, the allowed region in $(epsilon_{e tau},epsilon_{tau tau})$ follows the parabolic flat direction $epsilon_{tau tau} sim |epsilon_{e tau}|^2/(1+epsilon_{ee})$, broadened by the condition that one matter eigenvalue stay below the vacuum term at $E sim 20$-$30$ GeV. The parabola is cut off where the matter mixing angle $beta$ becomes too large to reconcile the high-energy atmospheric sample with the low-energy sample and with the nearly matter-free K2K data. Marginalizing over these couplings widens the $3sigma$ allowed vacuum region to $sin^2 theta_{23}=0.24$-$0.68$ and $|Delta m^2_{31}|=(1.7$-$3.9)times 10^{-3}$ eV$^2$, compared with $0.32$-$0.66$ and $(1.7$-$3.6)times 10^{-3}$ eV$^2$ for standard interactions.

The premise it rests on. The fit includes only the propagation effect of the new interactions and deliberately omits any nonstandard contribution to the detection cross section, so the quoted bounds assume that the same new physics does not appreciably change the neutral-current event rate in Super-Kamiokande.

If the claim holds, tau-neutrino matter couplings comparable to the standard weak coupling remain compatible with all atmospheric and K2K data, so accelerator bounds are improved but not pushed to the percent level.

The allowed vacuum parameters expand to $sin^2 theta_{23}=0.24$-$0.68$ and $|Delta m^2_{31}|=(1.7$-$3.9)times 10^{-3}$ eV$^2$ at $3sigma$, with the lower mixing edge and larger splitting absent in the standard fit.

K2K is load-bearing because its 250 km baseline is nearly matter-free: it measures the vacuum parameters directly and therefore cuts the parabolic region by limiting how far $Delta m^2$ can be shifted.

Long-baseline beam experiments should see an irregular disappearance pattern and an unsuppressed $nu_mu to nu_e$ channel if the NSI direction is realized, in contrast with standard two-neutrino oscillations.

Because the flat direction is a parabola, a model can hide large NSI from atmospheric data while still leaving observable shifts in solar/KamLAND and supernova neutrino channels; the companion solar analysis is the natural next discriminant.

The omission of detection effects is not just a caveat: a UV completion that generates both the vector matter potential and a neutral-current rate would correlate two otherwise independent datasets, so the bounds should be read as propagation-only bounds.

A future long-baseline measurement that finds maximal $theta_{23}$ and standard $Delta m^2$ would not by itself kill the parabola; the decisive signature is the energy dependence of disappearance plus $nu_e$ appearance.

The sign/phase degeneracy of $epsilon_{e tau}$ at $theta_{13}=0$ means that a nonzero $theta_{13}$ measurement would turn the symmetric parabola into a one-sided constraint, which is a concrete target for reactor and long-baseline programs.

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review