Development of a General Analysis and Unfolding Scheme and its Application to Measure the Energy Spectrum of Atmospheric Neutrinos with IceCube

TL;DR

This paper introduces a new generic analysis scheme for measuring atmospheric neutrino spectra with IceCube, utilizing advanced event selection and unfolding techniques to extend the energy range of previous measurements.

Contribution

A novel, general analysis framework combining machine learning and regularized unfolding for neutrino spectrum measurement with IceCube.

Findings

Successfully measured neutrino spectrum from 100 GeV to 1 PeV.

Achieved 99.9999% rejection of atmospheric muon background.

Extended the energy range of atmospheric neutrino measurements beyond previous limits.

Abstract

We present the development and application of a generic analysis scheme for the measurement of neutrino spectra with the IceCube detector. This scheme is based on regularized unfolding, preceded by an event selection which uses a Minimum Redundancy Maximum Relevance algorithm to select the relevant variables and a Random Forest for the classification of events. The analysis has been developed using IceCube data from the 59-string configuration of the detector. 27,771 neutrino candidates were detected in 346 days of livetime. A rejection of 99.9999% of the atmospheric muon background is achieved. The energy spectrum of the atmospheric neutrino flux is obtained using the TRUEE unfolding program. The unfolded spectrum of atmospheric muon neutrinos covers an energy range from 100 GeV to 1 PeV. Compared to the previous measurement using the detector in the 40-string configuration, the analysis presented here, extends the upper end of the atmospheric neutrino spectrum by more than a factor of two, reaching an energy region that has not been previously accessed by spectral measurements.