An algorithm for the reconstruction of high-energy neutrino-induced particle showers and its application to the ANTARES neutrino telescope

Abstract

A novel algorithm to reconstruct neutrino-induced particle showers within the ANTARES neutrino telescope is presented. The method achieves a median angular resolution of [Formula: see text] for shower energies below 100 TeV. Applying this algorithm to 6 years of data taken with the ANTARES detector, 8 events with reconstructed shower energies above 10 TeV are observed. This is consistent with the expectation of about 5 events from atmospheric backgrounds, but also compatible with diffuse astrophysical flux measurements by the IceCube collaboration, from which 2-4 additional events are expected. A [Formula: see text] C.L. upper limit on the diffuse astrophysical neutrino flux with a value per neutrino flavour of [Formula: see text] is set, applicable to the energy range from 23 TeV to 7.8 PeV, assuming an unbroken [Formula: see text] spectrum and neutrino flavour equipartition at Earth.

Fig. 1

Left angular error of the direction reconstruction for shower-like neutrino events as a function of the MC shower energy. Right the ratio of the MC and the reconstructed shower energy, as a function of the MC shower energy. Blue squares denote the median of the distributions. The lower and upper end of the vertical bars in both figures show the 10 and 90% quantiles of the distributions, respectively

Fig. 2

Left the neutrino effective area after applying the vertex-quality cut to triggered events, and integrated over all directions, as a function of simulated neutrino energy for ${\overline{\mathrm{\nu }}}_{\text{e}}$ (black full squares) and ${\mathrm{\nu }}_{\text{e}}$ (red open squares) CC events, and $\overline{\mathrm{\nu }}$ (black triangles) and for $\mathrm{\nu }$ (red open triangles) NC events. Right reconstruction efficiency for all triggered shower-like events (black squares) and including the vertex-quality cut (red triangles) as a function of MC shower energy

Fig. 3

Reconstructed zenith-angle distribution for 1247 days of data taking, with events selected as described in Sects. 4 and 5. Data points and their statistical errors are depicted with black markers and compared to simulated distributions of atmospheric muons (blue), atmospheric neutrinos (red) and the astrophysical flux reported in Ref. [6] (green). The coloured bands indicate the uncertainties on the simulated and measured flux normalisations

Fig. 4

Distribution of the reconstructed shower energy for 1247 days of data taking, selected as described in Sect. 4 and with a cut on the reconstructed zenith angle applied at ${\mathrm{\Theta }}_{\text{rec}}\ge {94}^{\circ }$ (black markers, statistical errors only). Simulated contributions from atmospheric muons (blue), atmospheric neutrinos (red) and an astrophysical flux  [6] (green) have been overlaid for comparison. Coloured bands indicate the uncertainties on the simulated and measured flux normalisations. The atmospheric muon contribution beyond 10 TeV has been extrapolated as described in Sect. 6

Fig. 5

The 90% C.L. upper limit on the diffuse all-flavour astrophysical neutrino flux obtained in this work (solid red line) in comparison to previously set upper limits (dotted lines, AMANDA-II  [59], Baikal NT-200 [60], and ANTARES ${\mathrm{\nu }}_{\mathrm{\mu }}$ [16]) and 2 different measurements of a diffuse astrophysical neutrino flux reported by IceCube (solid blue lines, IC ${\mathrm{\nu }}_{\text{x}}/3$  [5], and IC ${\mathrm{\nu }}_{\mathrm{\mu }}$ [6])