Algorithm for muon electromagnetic shower reconstruction

https://doi.org/10.1016/j.nima.2008.01.013Get rights and content

Abstract

The ANTARES neutrino telescope is presently being built in the Mediterranean Sea at a depth of 2500 m. The primary aim of the experiment is the detection of high energy cosmic muon neutrinos, which are identified by the muons that are produced in charged current interactions. These muons are detected by measuring the Cherenkov light which they emit traversing the detector. Sometimes a high momentum muon produces electromagnetic showers. The subject of this paper is a method to reconstruct these showers which includes several steps: an algorithm for the fit of the muon track parameters, preselection of detected photons belonging to a shower, and a final fit with the preselected detected photons to calculate the electromagnetic shower position. Finally a comparison between data obtained with that part of the detector that is currently in operation and simulations is presented.

Introduction

The ANTARES [1] detection principle relies on the observation of Cherenkov (CK) light emitted by relativistic charged particles in water. The emitted CK photons are detected by a three-dimensional grid of 900 photomultiplier tubes (PMTs) arranged in 12 detection lines at a depth of 2500 m in the Mediterranean Sea. The detector will be completed in early 2008. Since the end of January 2007, the detector has five operating detection lines. Amongst others first data of atmospheric downward going muons have been measured, which are up to 500 m long tracks with energies larger than 100 GeV. This gives the unique opportunity to study the average number of showers per track length for high energy muons passing through sea water.

The muon track is reconstructed from the arrival times of the CK light detected by the PMTs, whose positions are known. The muon energy loss mechanism in water is ionization, e+e- pair production, bremsstrahlung, and photonuclear interactions. Above several hundred GeV the muon energy loss is dominated by the latter three effects [2], which are discrete and are characterized by large energy fluctuations. Furthermore, the average muon energy loss due to these effects is proportional to the energy of the muon.

In the following a method to reconstruct discrete electromagnetic (EM) showers is presented, which gives additional information about a muon track (e.g. the shower multiplicity) as compared to existing algorithms only measuring the direction and position of the muon track.

Section snippets

Identification of muon EM showers

The challenge of EM shower reconstruction is to distinguish CK photons from EM shower photons. The EM shower light has two key characteristics: it is produced in one point on the muon path and it arrives, in general, delayed compared to the CK light.

The first characteristic distinguishing CK light from EM shower light is illustrated in Fig. 1. The CK photons are emitted continuously along the muon path, whereas EM showers are emitted stochastically and discretely. Furthermore, the number of

Algorithm for shower reconstruction

The algorithm consist of three steps using digitized PMT signals which are referred to as hits.1 The first step selects muon candidates. Then, hits are preselected which originate from one distinct shower. The third step is a χ2-minimization yielding the space–time position of the shower within the preselected hits.

Conclusion

The performance of the reconstruction algorithm for EM showers induced by muons has been validated in a sample of simulated multiple atmospheric muon events with background noise. Next, the algorithm was applied to data obtained with ANTARES. The results of the shower multiplicity for down going muon events are shown in Fig. 4. The agreement between data and simulation is satisfactory. For the used set of selection cuts, the one shower muon events constitute about 5% of all muon events.

In

References (4)

  • E. Aslanides, et al.,...
  • W.-M. Yaom

    J. Phys. G Nucl. Part. Phys.

    (2006)
There are more references available in the full text version of this article.

Cited by (11)

View all citing articles on Scopus
View full text