[Paper Review] Measurement of the Diffuse Astrophysical Muon-Neutrino Spectrum with Ten Years of IceCube Data

J. Stettner|arXiv (Cornell University)|Aug 26, 2019

Astrophysics and Cosmic Phenomena6 citations

TL;DR

This paper presents an updated measurement of the diffuse astrophysical muon-neutrino spectrum using ten years of IceCube data, with improved systematic uncertainty treatment and Pass-2 reprocessing. The best-fit power-law spectrum yields a normalization of $\Phi_{\text{astro}} = 1.44 \times 10^{-18}\ \text{GeV}^{-1}\text{cm}^{-2}\text{s}^{-1}\text{sr}^{-1}$ and a spectral index of $\gamma_{\text{astro}} = 2.28$, consistent with prior results but with enhanced precision and reduced systematics.

ABSTRACT

The IceCube Neutrino Observatory measured a flux of high-energy astrophysical neutrinos in several detection channels. The energy spectrum is fitted as unbroken power-law, but different best-fit parameters were obtained in the various analyses covering different energy ranges between few TeV to 10 PeV. Here, we present an update to the analysis of through-going muon-neutrinos from the Northern Hemisphere. It was extended to almost ten years of data and an improved treatment of systematic uncertainties on the atmospheric fluxes was implemented. The updated best-fit parameters for the astrophysical flux assuming a power-law energy spectrum are $\\Phi_{astro}=1.44$ and $\\gamma_{astro}=2.28$. We will present the results of the spectral fit and discuss how the measured flux compares to other IceCube results.

Motivation & Objective

• To measure the diffuse astrophysical muon-neutrino flux over a decade of IceCube data with improved systematic uncertainty control.
• To update the spectral fit of high-energy neutrinos using a consistent, reprocessed dataset (Pass-2) across all detector configurations.
• To assess the impact of atmospheric neutrino flux uncertainties—especially prompt and conventional components—on the astrophysical flux measurement.
• To compare the results with other IceCube analyses (e.g., starting events, cascades) and evaluate consistency across detection channels.

Proposed method

• The analysis uses through-going muon-neutrino events from the Northern Celestial Hemisphere, selected via a boosted decision tree for high-quality track reconstruction.
• Monte Carlo simulations model conventional and prompt atmospheric neutrino fluxes using MCEq with MSIS-00 and SIBYLL2.3c, while the astrophysical component is modeled as a single power-law.
• A Poisson likelihood is used to compare data and simulation binned in reconstructed muon energy and cosine of zenith angle, with signal and background parameters fit simultaneously.
• Systematic uncertainties are treated via nuisance parameters, including variations in the primary cosmic-ray flux model (e.g., H4a, GST-4gen, GSF-beta) and atmospheric models.
• The Pass-2 reprocessing campaign re-calibrated historical data (2010–2016), correcting a 4% charge-scale shift and ensuring consistency across IC59, IC79, and IC86 configurations.
• Profile likelihood scans are performed to determine confidence intervals and correlations between astrophysical normalization and spectral index.

Experimental results

Research questions

• RQ1 What is the best-fit power-law spectrum for the diffuse astrophysical muon-neutrino flux using ten years of IceCube data with improved systematics?
• RQ2 How do variations in the primary cosmic-ray flux model affect the measured astrophysical flux parameters?
• RQ3 What is the impact of prompt atmospheric neutrinos on the astrophysical flux measurement, and what upper limit can be placed on their normalization?
• RQ4 How do the updated results compare to previous IceCube measurements of the astrophysical flux in other channels (e.g., starting events, cascades)?
• RQ5 To what extent are the astrophysical flux parameters sensitive to uncertainties in atmospheric neutrino flux models?

Key findings

• The best-fit astrophysical flux normalization is $\Phi_{\text{astro}} = 1.44 \times 10^{-18}\ \text{GeV}^{-1}\text{cm}^{-2}\text{s}^{-1}\text{sr}^{-1}$ with a 68% confidence interval of $^{+0.25}_{-0.24}$.
• The best-fit spectral index is $\gamma_{\text{astro}} = 2.28$ with a 68% confidence interval of $^{+0.08}_{-0.09}$, indicating a softening compared to earlier results.
• The prompt atmospheric neutrino normalization is best-fit at zero, with a small impact on the astrophysical parameters: a $\Delta\gamma_{\text{astro}} = -0.05$ shift if fixed at the MCEq baseline.
• The GST-4gen primary cosmic-ray model provides the best fit to the data, with a $\Delta\text{LLH} = 4.4$ improvement over the H4a baseline.
• The results are consistent with previous IceCube measurements of the astrophysical flux in other channels, as shown in the comparison of profile likelihood contours.
• Systematic uncertainties from atmospheric flux models are largely absorbed by nuisance parameters, with minimal impact on the astrophysical parameters.