Resolving Neutrino Mass Hierarchy and CP Degeneracy by Two Identical Detectors with Different Baselines

The core idea is worth your attention: two identical 0.27 Mton water Cherenkov modules on the same J-PARC off-axis beam, one at Kamioka (295 km) and one in Korea (~1050 km), exploiting the fact that the Korean module sits past the second oscillation maximum where the appearance probability's energy dependence is violent and hierarchy-dependent. The bi-probability plots (Figs. 1, 2, 4) make the mechanism clear, and the chi-squared in Eqs. (3)-(4) is explicit enough to check. The volume-ratio scan in Fig. 6 is useful work: 1:1 comes out near-optimal, and the flatness across 3:7 and 1:9 is itself a robustness statement. What is genuinely new is the quantitative claim, not the ingredients. Footnote 3 credits Bugey for identical-detector cancellation and Ref. [14] for the Korean site; the authors are upfront about that. The contribution is the demonstration that the Korean spectral shape, combined with that cancellation, buys sign discrimination down to sin² 2theta_13 ~ 0.03 at 2 sigma. Soft spots, in order of weight. First, the stress-test note is right and it is the one I would push on. Eq. (4) shares a single signal-efficiency nuisance parameter across all four detector/beam combinations. The 2/5/10% scan in Sec. IV.D varies the magnitude of a common systematic, not the effect of a detector-specific one. A 1-3% uncorrelated Korean efficiency offset from PMT coverage or water transparency would eat into the headline numbers in a way the scan cannot see. The abstract does say 'under reasonable assumptions,' so this is semi-acknowledged, but the correlation structure itself is never discussed. Second, and related: the 'identical beam spectrum' premise. Sec. I establishes that a 2-3 degree off-axis beam reaches Korea, but the geometry (Kamioka 1.3 degrees below horizon, Korea 4.5-5.6) does not obviously give the same off-axis angle at both sites. The error-cancellation argument needs equal spectra; the paper asserts this rather than demonstrating it. Worth a paragraph. Minor: constant density along the Korean path (they do vary 2.5/2.8/3.1, fine), five-bin binning not varied, no tables or code released. None of this sinks the qualitative conclusion, which is that a Korean module adds real sign-discrimination power. The numbers should be read as indicative. Who should read it: anyone working on long-baseline design or detector systematics for the J-PARC program. Send it out.

The central claim — mass hierarchy resolved for sin^2 2θ13 > 0.03 (2σ) / 0.055 (3σ) for any δ — depends on the two-detector comparison. In Eq. (4), a single signal-efficiency nuisance parameter ǫ_3 is shared between Kamioka and Korea (f_i3 = 1 for all i and all k). This encodes the assumption that the Korean detector's signal efficiency is perfectly correlated with Kamioka's. The robustness scan in Sec. IV.D varies σ̃_j (2%, 5%, 10%) but keeps the correlation perfect, so it tests only the magnitude of a common systematic, not the effect of a detector-specific one. Because the entire gain over a single 0.54 Mton Kamioka detector (Fig. 6, dashed blue) comes from the two-detector comparison, an uncorrelated efficiency difference — plausible at the 1-3% level from different PMT coverage, water transparency, or reconstruction tuning — would directly erode the headline numbers. The same applies to the assumed identical beam spectrum: the geometry discussion in Sec. I establishes only that the Korean off-axis angle is 'larger than 1 degree' and 'between 2 and 3 degrees,' not that it equals Kamioka's 2.5°; a 0.2° difference would make the signal-efficiency correlation spectrum-dependent and imperfect.

The paper is a careful sensitivity study with an explicit χ², a clear physical mechanism, and genuine robustness checks on systematic size, matter density, and θ23. The qualitative conclusion — that a Korean detector on the same beam adds sign-discrimination power — is well supported. The reader's weakest_assumption correctly identifies the load-bearing premise: the error-cancellation model of Eqs. (3)-(4). I agree, and sharpen it: the scan varies only the magnitude of a perfectly correlated nuisance parameter, so it does not test the correlation itself. The headline numbers are therefore indicative rather than demonstrated. This supports CONDITIONAL, not ACCEPT. It is not REJECT: the direction of the effect is robust, and the paper acknowledges the site and systematic-model caveats in Sec. V. The concrete test above would settle whether an uncorrelated detector-specific efficiency offset materially moves the thresholds.

If the premise holds, this is a significant result: it would elevate the J-PARC phase II program from a CP-violation search to a dual-purpose experiment that also determines the mass ordering. The paper's strengths are the explicit χ² definition in Eqs. (3)-(4), the genuine robustness scans in §IV.D, and the honest comparison of detector configurations in Fig. 6 and Fig. 7. The paper is a forward simulation study, not a proof, and its conclusions are explicitly conditional on the two detectors being identical and receiving the same beam spectrum. The key open question is whether those conditions can be met closely enough in practice.

The paper asks whether a single accelerator neutrino beam can be used to settle two open questions at the same time: whether the third neutrino is the heaviest or the lightest (the mass hierarchy), and whether leptons violate CP. The obstacle is a four-fold ambiguity: for a given appearance rate there are several combinations of $sin^2 2theta_{13}$ and $delta$ that fit, and the unknown sign of $Delta m^2_{31}$ duplicates them again. The authors' proposal is to take the planned megaton water Cherenkov detector and split it into two identical 0.27 Mton modules, one at Kamioka (295 km) and one in Korea (about 1050 km), both fed by the same 2.5-degree off-axis J-PARC beam at 4 MW. Because the two modules are identical and see the same beam spectrum when there were no oscillations, their systematic uncertainties cancel, which makes the shape of the observed energy spectrum trustworthy enough to break the ambiguity. The payoff claimed is a hierarchy resolution for $sin^2 2theta_{13} > 0.03$ at $2sigma$ and $> 0.055$ at $3sigma$ for any value of $delta$, together with CP-violation sensitivity from 4 years of neutrino plus 4 years of anti-neutrino running.

What carries the argument. The load-bearing device is the bi-probability plot: the pair $(P(nu_mu to nu_e), P(bar{nu}_mu to bar{nu}_e))$ traced as $delta$ runs from 0 to $2pi$ gives an ellipse for each $theta_{13}$, and the two hierarchies give two overlapping regions. The transformation $delta to pi - delta$ with $Delta m^2_{31} to -Delta m^2_{31}$ leaves the probability nearly invariant when matter effects are small, which is why Kamioka alone cannot tell the hierarchies apart. The Korean baseline breaks the symmetry because matter effects and the energy dependence of the ellipses differ there. The second piece of machinery is the $chi^2$ of Eqs. (3) and (4): five energy bins per channel, four detector-beam combinations, and three nuisance parameters (5 percent background normalization, 5 percent energy-dependent background shape, 5 percent signal efficiency) shared between the two detectors. Sharing them is what encodes the cancellation, and the paper checks that varying them between 2 and 10 percent barely moves the contours.

Core claim. The claim is that a two-detector complex with identical 0.27 Mton fiducial volumes, one in Kamioka and one in Korea, has the potential to resolve the neutrino mass hierarchy and to detect leptonic CP violation simultaneously. The hierarchy is resolved up to $sin^2 2theta_{13} > 0.03$ ($0.055$) at the $2sigma$ ($3sigma$) level for any value of $delta$, and CP violation is reachable with 4 years of neutrino and 4 years of anti-neutrino appearance running. The mechanism is not a larger event rate in Korea — the Korean module collects an order of magnitude fewer events — but the shape of its energy spectrum. At about 1050 km, slightly off the second oscillation maximum, the CP effect is two to three times larger than at Kamioka, the matter effect is more pronounced, and the bi-probability ellipses swing through more than $pi/2$ as the energy moves from 0.5 to 0.8 GeV, in opposite directions for the two hierarchies. The authors show that a single 0.54 Mton detector in Kamioka cannot separate the hierarchy from its mirror solution, whereas the pair can, and that the sensitivity depends only weakly on the assumed size of systematic uncertainties, on the matter density along the Korean path, and on deviations of $theta_{23}$ from maximal mixing.

The premise it rests on. The headline numbers rest on the two detectors being genuinely identical in response and on the three-parameter systematic model of Eqs. (3) and (4) capturing all the correlated uncertainty; the paper varies the size of those uncertainties but not their functional form or any detector-to-detector difference in acceptance, energy reconstruction, or background.

If the two-detector scheme works as modelled, a single J-PARC beam line delivers both the mass ordering and the CP phase, so the ordering does not have to be marginalized away when fitting for $delta$.

The Korean module resolves the hierarchy mainly through spectrum shape rather than statistics, so a modest fiducial volume there is enough; the paper finds the 1:1 split close to optimal and the sensitivity only weakly dependent on the volume ratio.

Because the two modules are identical and see the same unoscillated beam, the background rates are tied by the inverse-square of the baseline ratio, so a highly precise standalone background measurement is not needed.

The CP-violation reach of the current J-PARC phase II design is preserved or improved for $sin^2 2theta_{13} > 0.01$, turning a CP-search experiment into a dual-purpose one at roughly the same total cost.

The analysis suggests that even neutrino-beam-only running carries some CP information, because the two detectors' spectra constrain $delta$ directly.

The proposal is really a statement about correlated systematics: the physics gain comes from the correlation structure of the fit, not from the event count, so the same trick would apply to any pair of identical detectors on one beam line, e.g. a future long-baseline pair in the United States or Europe.

The Korean baseline sits deliberately off the second oscillation maximum; the paper's own reasoning implies an optimum, and a scan of baseline between roughly 900 and 1200 km would show how sharp that optimum is.

If the mass ordering turns out to be inverted, the same configuration should perform nearly symmetrically, since the paper presents normal and inverted true hierarchies side by side; a small asymmetry in reach would itself be a diagnostic.

The result is a statement about a possible future facility, so its practical weight depends on site availability in Korea and on whether two separated modules can be operated with genuinely matched response — neither of which the paper quantifies.

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