This paper presents several approaches to deal with the problem of identifying muons in a water Cherenkov detector with a reduced water volume and 4 PMTs. Different perspectives of information representation are used and new features are engineered using the specific domain knowledge. As results show, these new features, in combination with the convolutional layers, are able to achieve a good performance avoiding overfitting and being able to generalise properly for the test set. The results also prove that the combination of state-of-the-art Machine Learning analysis techniques and water Cherenkov detectors with low water depth can be used to efficiently identify muons, which may lead to huge investment savings due to the reduction of the amount of water needed at high altitudes. This achievement can be used in further research to be able to discriminate between gamma and hadron induced showers using muons as discriminant.

The concept of a small, single-layer water Cherenkov detector, with three photomultiplier tubes (PMT), placed at its bottom in a 120 • star configuration (Mercedes WCD) is presented. The PMTs are placed near the lateral walls of the stations with an adjustable inclination and may be installed inside or outside the water volume. To illustrate the technical viability of this concept and obtain a first-order estimation of its cost, an engineering design was elaborated. The sensitivity of these stations to low energy EAS electrons, photons and muons is discussed, both in compact and sparse array configurations. It is shown that the analysis of the intensity and time patterns of the PMT signals, using Machine Learning techniques, enables the tagging of muons, achieving an excellent gamma/hadron discrimination for TeV showers. This concept minimises the station production and maintenance costs, allowing for a highly flexible and fast installation. Mercedes WCDs are thus well-suited for use in high-altitude large gamma-ray observatories covering an extended energy range from the low energies, closing the gap between satellite and ground-based measurements, to very high energy regions, beyond the PeV scale.