Evidence for Oscillation of Atmospheric Neutrinos

The analysis

Atmospheric neutrinos arise from cosmic-ray interactions in the upper atmosphere and reach Super-Kamiokande from all directions: downward-going neutrinos have traversed only the ~15 km atmospheric depth, while upward-going neutrinos have traversed up to the Earth’s full diameter of ~12,800 km. In the absence of oscillation — and after correction for geomagnetic and other small effects — the two fluxes should be equal.

The Super-K analysis used 535 days of data from the SK-I phase, with fiducial mass 22.5 kt and electron/muon discrimination based on Cherenkov-ring pattern fuzziness. The event sample comprised 988 “fully contained” e-like events and 1,231 μ-like events, plus 301 partially contained μ-like events.

The result

The observed zenith-angle distributions showed:

• Electron-like events: up-down ratio consistent with unity, showing no oscillation signal in the explored parameter range
• Muon-like events: significant deficit for upward-going events compared to downward-going, with the deficit growing monotonically with zenith angle — consistent with the expected $L/E$ dependence of oscillation

The double-ratio $R=(\mu /e{)}_{\text{data}}/(\mu /e{)}_{\text{MC}}$ was measured at $0.63\pm 0.03\pm 0.05$, a 6σ deviation from unity.

A two-flavor oscillation fit to the sub-GeV and multi-GeV samples, separately for e-like and μ-like events, gave a best fit of ${\sin }^{2}2\theta \approx 1.0,\Delta m^2\approx 2\times 10^{-3}{\text{ eV}}^2$ with ${\nu }_{\mu }\to {\nu }_{\tau }$ as the most consistent interpretation. ${\nu }_{\mu }\to {\nu }_e$ was disfavored by the absence of an e-like excess and by the CHOOZ reactor limit on ${\theta }_{13}$.

The Takayama talk

The result was presented by Takaaki Kajita at the Neutrino ‘98 conference in Takayama on 5 June 1998. The combination of zenith-angle dependence, energy dependence, and flavor specificity made the oscillation interpretation inescapable. By the end of the conference week, the community had accepted that neutrinos have mass.

The paper was submitted to Physical Review Letters shortly after and published in August 1998.

Reception and follow-up

The K2K and MINOS long-baseline accelerator experiments subsequently confirmed the $\Delta m_{atm}^2$ measurement at higher precision using controlled beams. T2K and NOvA have since refined it further. The initial “near-maximal mixing” interpretation has been tightened: ${\sin }^2{\theta }_{23}$ is now known to be between about 0.45 and 0.57, close to but not exactly maximal, with residual octant ambiguity.

The 2015 Nobel Prize in Physics was awarded jointly to Takaaki Kajita (for this atmospheric-oscillation result) and Arthur McDonald (for the complementary SNO solar-oscillation result), “for the discovery of neutrino oscillations, which shows that neutrinos have mass.”

Significance in the evidence chain

The Super-K result is the single most-cited demonstration that the minimal massless-neutrino Standard Model is incomplete. Every subsequent discussion of neutrino mass, neutrino mass ordering, CP violation, the seesaw mechanism, and leptogenesis begins from this experimental fact.