On the flavor composition of the high-energy neutrinos in IceCube

TL;DR

This paper analyzes IceCube's high-energy neutrino data, revealing a potential deviation from the expected flavor ratio of 1:1:1, which could imply background misestimation, misclassification, or new physics.

Contribution

It provides the first detailed statistical analysis of the flavor composition of high-energy neutrinos in IceCube data, highlighting possible tensions with standard expectations.

Findings

The canonical flavor ratio (1:1:1) is disfavored at 92% confidence level.

Best-fit flavor ratio at Earth is (1:0:0), inconsistent with standard oscillation predictions.

Results suggest potential background issues, misclassification, or new physics.

Abstract

The IceCube experiment has recently released 3 years of data of the first ever detected high-energy (>30 TeV) neutrinos, which are consistent with an extraterrestrial origin. In this talk, we compute the compatibility of the observed track-to-shower ratio with possible combinations of neutrino flavors with relative proportion (alpha_e:alpha_mu:alpha_tau). Although this observation is naively favored for the canonical (1:1:1) at Earth, once we consider the IceCube expectations for the atmospheric muon and neutrino backgrounds, this flavor combination presents some tension with data. We find that, for an astrophysical neutrino E_nu^{-2} energy spectrum, (1:1:1) at Earth is currently disfavored at 92% C.L. We discuss the trend of this result by comparing the results with the 2-year and 3-year data. We obtain the best-fit for (1:0:0) at Earth, which cannot be achieved from any flavor ratio at sources with averaged oscillations during propagation. Although it is not statistically significant at present, if confirmed, this result would suggest either a misunderstanding of the expected background events, or a misidentification of tracks as showers, or even more compellingly, some exotic physics which deviates from the standard scenario.