Super-Kamiokande atmospheric neutrino data, zenith distributions, and three-flavor oscillations - art. no. 033001

Citation
Gl. Fogli et al., Super-Kamiokande atmospheric neutrino data, zenith distributions, and three-flavor oscillations - art. no. 033001, PHYS REV D, 5903(3), 1999, pp. 3001
Citations number
103
Categorie Soggetti
Physics
Journal title
PHYSICAL REVIEW D
ISSN journal
05562821 → ACNP
Volume
5903
Issue
3
Year of publication
1999
Database
ISI
SICI code
0556-2821(19990201)5903:3<3001:SANDZD>2.0.ZU;2-Y
Abstract
We present a detailed analysis of the zenith angle distributions of atmosph eric neutrino events observed in the Super-Kamiokande (SK) underground expe riment, assuming two-flavor and three-flavor oscillations (with one dominan t mass scale) among active neutrinos. In particular, we calculate the five angular distributions associated with sub-GeV and multi-GeV mu-like and e-l ike events and to upward through-going muons, for a total of 30 accurately computed observables (zenith bins). First we study how such observables var y with the oscillation parameters, and then we perform a fit to the experim ental data as measured in SK for an exposure of 33 KTy (535 days). In the t wo-flavor mixing case, we confirm the results of the SK Collaboration analy sis, namely, that nu(mu)<->nu(tau) oscillations are preferred over nu(mu)<- >nu(e), and that the no oscillation case is excluded with high confidence. In the three-flavor mixing case, we perform our analysis with and without t he additional constraints imposed by the CHOOZ reactor experiment. In both cases, the analysis favors a dominance of the nu(mu)<->nu(tau) channel. Wit hout the CHOOZ constraints, the amplitudes of the subdominant nu(mu)<->nu(e ) and nu(e)<->nu(tau) transitions can also be relatively large, indicating that, at present, current SK data do not exclude sizable nu(e) mixing by th emselves. After combining the CHOOZ and SK data, the amplitudes of the subd ominant transitions are constrained to be smaller, but they can still play a non-negligible role both in atmospheric and other neutrino oscillation se arches. In particular, we find that the nu(e) appearance probability expect ed in long baseline experiments can reach the testable level of similar to 15%. We also discuss Earth matter effects, theoretical uncertainties, and v arious aspects of the statistical analysis. [S0556-2821(99)06401-2].