An improved measurement of the atmospheric neutrino flux in IceCube

Cosmic neutrinos in IceCube are the vogue these days, but atmospheric neutrinos are the popular ones if we look at the number of hits in the detector. Those neutrinos, created by the interaction of cosmic rays in the Earth’s atmosphere, are the main background in searches for astrophysical neutrinos.

The IceCube Collaboration has submitted a paper today to the European Physical Journal C describing a new analysis scheme for the measurement of the atmospheric neutrino spectrum with the IceCube detector. The analysis was performed using data from May 2009 to May 2010, when the detector was running with a configuration of 59 of the final 86 strings.

The spectrum was measured introducing a novel unfolding technique in the energy range from 100 GeV to 1 PeV, extending previous results of AMANDA by almost an order of magnitude. The new analysis also uses an improved selection, with results that showed a reduced atmospheric muon background contamination of 5 to 6 orders of magnitude and an 8% increase in the signal efficiency.

The unfolded atmospheric neutrino spectrum agrees with both previous experimental results and the current theoretical models. The new method reduces the impact of the systematic uncertainties on the measured flux, but at high energies they are still too large to allow for conclusive results about a prompt and/or an astrophysical component of the overall flux.

Fig 17 Ruhe
Comparison of the unfolding result obtained using IceCube in the 59-string configuration to previous experiments. Theoretical models are shown for comparison. Image: IceCube Collaboration.

The analysis scheme presented in this paper introduces a machine learning algorithm for the final event selection that uses 25 event variables to distinguish between atmospheric muon tracks and tracks produced by neutrino-induced muons.

“IceCube is a great detector for measuring atmospheric muon neutrinos. Those are, in fact, the vast majority of the neutrinos we detect. And by using tools and algorithms from data mining we can detect even more,” explains Tim Ruhe, a researcher at TU Dortmund University, in Germany.

For every neutrino detected by IceCube, about a million atmospheric muons are observed. A common way to look for neutrinos in this huge muon background consists of selecting only upgoing muon tracks, since muons created by the interaction of cosmic rays with the atmosphere will be absorbed by the Earth when approaching IceCube from below. Thus, if the event reconstruction and selection were perfect, the remaining muon tracks would have been created by the interaction of a neutrino with the ice in or around the IceCube detector.

However, previous to this analysis, the muon background rejection in IceCube was only 99.9% efficient because about 1,000 originally downgoing muons per every neutrino seen by IceCube were falsely reconstructed as upgoing tracks. With the new selection algorithm, IceCube researchers were able to reject 99.9999% of the incoming background events.

Info “Development of a General Analysis and Unfolding Scheme and its Application to Measure the Energy Spectrum of Atmospheric Neutrinos with IceCube,” IceCube Collaboration: M.G. Aartsen et al. European Physical Journal C75 (2015) 116,