Monte Carlo simulations are a most useful tool to study the response of a detector and to monitor its performance. In a deep-sea neutrino telescope as ANTARES, the variability of the environmental conditions that can change the behaviour of the data acquisition system must be considered in order to provide a reliable description of the active parts of the detector. Similarly, the effects of particles passing through the apparatus and their signals on the active parts of the detector must be simulated in order to realistically reproduce the detector response to the possible neutrino signals that are the targets of the ANTARES analyses.
Neutrino interactions and cosmic ray muons passing through the detector, and the particles produced by their interactions as the move in matter, must be correctly modeled. The neutrino interactions are simulated with the GENHEN software, developed in the ANTARES collaboration specifically for this scope. Similarly, cosmic ray muons are simulated using the MUPAGE code, which has also been developed by ANTARES collaborators [Computed Physics Communications 179(12):915-923].
A sensitive volume around the apparatus is defined and all charged particles in this volume are followed. The light emitted along their path is the propagated to the active elements of the detector, its PMTs. This is done using the KM3 software, a GEANT-based application, specifically developed by the ANTARES Collaboration in which the probability of obtaining a photon at a certain distance from a particle is tabulated. KM3 allows for a simulation of the water properties that are based on the measured optical properties of the medium.
Figure: scheme of the simulation procedure of the ANTARES telescope. The sensitive volume (the can) is defined around the instrumented volume of the apparatus, in which detailed particle tracking is performed and light is tracked to the sensitive parts of the detector.
Finally, the PMT electronics – with the digitisation of the PMT signals – and the data acquisition of the detector, reproducing the same filtering algorithms that are applied on the real data at the shore station. At this stage, also the PMT behaviour is reproduced, also with specific actions to reproduce the time-dependent behaviour of the environmental conditions in terms of optical noise as well as the PMT efficiency modification along time. For the former, real detector data is used to sample the measured rate on the individual PMTs and reproduce this in the simulations. For the latter, in situ measurements based on the detection of 40K decays in sea-water are used to estimate the PMT-by-PMT efficiency [The European Physics Journal, 78: 669 (2018)]. This constitutes the run-by-run approach.
After this stage, simulated data are equivalent to real data as they come out of the shore station and they can be treated with the same analysis tools.