Characterization of the Three-Flavor Composition of Cosmic Neutrinos with IceCube

Abstract

Neutrinos oscillate over cosmic distances. Using 11.4 years of IceCube data, the flavor composition of the all-sky neutrino flux from 5\,TeV--10\,PeV is studied. We report the first measurement down to the $\mathcal{O}$(TeV) scale using events classified into three flavor-dependent morphologies. The best fit flavor ratio is $f_e:f_μ:f_τ\,=\,0.30:0.37:0.33$, consistent with the standard three-flavor neutrino oscillation model. Each fraction is constrained to be $>0$ at $>$ 90\% confidence level, assuming a broken power law for cosmic neutrinos. We infer the flavor composition of cosmic neutrinos at their sources, and find production via neutron decay lies outside the 99\% confidence interval.

Figures (5)

• Figure 1: Ternary diagram of the results of the flavor-composition fit: The axes show the fraction of $\nu_e$, $\nu_{\mu}$ and $\nu_{\tau}$ at Earth. 68% and 95% CL contours assuming the test statistic follows Wilks' theorem Wilks are shown as solid and dashed lines, respectively. Expected flavor composition at Earth, after standard oscillations, for benchmark production mechanisms (pion decay: circle, muon damping: square, neutron decay: triangle) and all possible flavor compositions after propagation (dash-dot line, from Song:2020nfh) are shown. The dotted line shows the 68% CL contour from IceCube's last measurement IceCube:2020fpi.
• Figure 2: Posterior distribution of flavor composition at source: Assuming $f_{\tau,\mathrm{S}}\,=0\,$ and global best-fit mixing parameters (normal ordering) with their $\chi^2$ distributions from Esteban:2024eli, and the likelihood profile obtained with the $\mathcal{L}_{\mathrm{max}}$ fit to the MESE dataset. The same with the likelihood profile from the IceCube (2015) analysis in icecube_collaboration_combined_2015, which reported more constraining rejection of neutron decay with the $\mathcal{L}_{\mathrm{max}}$ fit.
• Figure 3: Histograms of observables: The set of 7 observables used for performing the flavor measurement, with 11.4 years of data. We assume a broken power law model for the astrophysical flux. We fit a prompt contribution of zero, and hence it is not shown in the figure. No muon component survives the double cascade selection procedure.
• Figure 4: Ternary flavor composition for different spectral assumptions: The baseline measurement, assuming a broken power law, is compared to a single power law assumption. The contour size shrinks along the $\nu_{\tau}$ axis for an SPL, since it predicts more tau neutrinos due to its harder spectrum at higher energies. The best fits remains close to each other for both models. Also $f_e$ and $f_{\mu}$ CLs remain similar.
• Figure 5: Posterior at source, no prior on f$_{\tau,\,\rm{S}}$: Here we do not assume the additional prior on the fraction of $\nu_{\tau}$ at source. The remaining assumptions are the same as that in Fig. \ref{['fig:source_post']}. Contours describing the high probability density regions, constructed from the posterior distribution, are shown.