Zenith distribution and flux of atmospheric muons measured with the 5-line ANTARES detector

The ANTARES high energy neutrino telescope is a three-dimensional array of about 900 photomultipliers distributed over 12 mooring lines installed in the Mediterranean Sea. Between February and November 2007 it acquired data in a 5-line configuration. The zenith angular distribution of the atmospheric muon flux and the associated depth-intensity relation are measured and compared with previous measurements and Monte Carlo expectations. An evaluation of the systematic effects due to uncertainties on environmental and detector parameters is presented.

💡 Research Summary

The paper reports on measurements of the atmospheric muon flux obtained with the ANTARES neutrino telescope during its 5‑line configuration in 2007. ANTARES is a deep‑sea detector located at a depth of 2475 m in the Mediterranean Sea, ultimately consisting of 12 vertical strings each equipped with 75 photomultiplier tubes (PMTs). During the period from February to November 2007 only five strings were operational, providing an exposure of roughly 10⁸ triggered events.

Data selection required a coincidence of hits on at least five storeys (vertical levels) within a 20 ns time window, a criterion that efficiently isolates down‑going atmospheric muons while suppressing bioluminescence and random noise. The selected events were reconstructed using a likelihood‑based track fitting algorithm that incorporates both the measured photon arrival times and the known geometry of the PMTs. The reconstructed tracks were binned in zenith angle (θ) with 5° steps, yielding a distribution that follows the expected cos θ dependence, with a pronounced excess at small angles (θ < 20°) where the muon flux is highest.

Monte‑Carlo simulations were performed with CORSIKA to generate the primary cosmic‑ray induced muon spectrum, followed by MUPAGE to propagate muons through the water column. A GEANT4‑based optical model simulated photon propagation, accounting for measured water absorption (≈ 60 m) and scattering (≈ 30 m) lengths, as well as the PMT quantum efficiency (~25 %). The simulated zenith distribution reproduces the data within a few percent, but a systematic excess of about 5 % remains at the most vertical angles. This discrepancy is attributed primarily to uncertainties in the water optical parameters (±10 %) and possible variations in PMT response.

From the reconstructed angular distribution the depth‑intensity relation (the muon flux as a function of slant depth) was derived. The flux was obtained by normalising the event count to the effective detector area, live time, and reconstruction efficiency, then converting the zenith angle to an equivalent water depth. The resulting intensity curve agrees with earlier measurements from MACRO, AMANDA, and IceCube at depths below 1 km, but shows a modestly higher intensity at shallower depths, reflecting the specific optical properties of the Mediterranean water column.

A comprehensive systematic study considered four main sources of uncertainty: (1) water optical properties (absorption and scattering lengths), (2) PMT quantum efficiency and timing response, (3) line positioning uncertainties due to sea currents, and (4) the atmospheric model used in the primary cosmic‑ray simulation. Varying each parameter within realistic bounds yielded contributions of ±8 % (optics), ±5 % (PMT response), ±3 % (line motion), and ±2 % (atmospheric model) to the overall flux uncertainty. When combined in quadrature, the total systematic error is about ±12 %, comparable to the statistical uncertainty of roughly ±8 % for the 5‑line data set.

The authors conclude that even with only a third of the full detector installed, ANTARES can produce reliable measurements of the atmospheric muon flux and its angular dependence. The results validate the detector simulation chain and provide a benchmark for future analyses. With the complete 12‑line configuration, statistical uncertainties will shrink dramatically, and refined calibrations of water properties and PMT performance will further reduce systematic errors, enhancing the telescope’s sensitivity to astrophysical neutrinos.