Reconstruction of events from the surface detector of the Pierre Auger Observatory using air shower Universality

The chemical composition of ultra-high energy cosmic rays (UHECRs) is a key element to solve several open questions about the origin and propagation of these rare particles. When reaching the Earth, UHECRs interact with the atmosphere generating showers of secondary particles. The atmospheric depth of the maximum in the electromagnetic part of the shower (Xmax) is a quantity commonly employed to infer information about the nature of primary cosmic rays. It is measured with good accuracy through fluorescence detection, but the related duty cycle is insufficient to extend our knowledge about composition at the highest energies, namely above 1019.5 eV with the current experimental setups. Therefore many efforts have been put in the development of a composition analysis which makes use only of data recorded with ground arrays of surface detectors, operational nearly 100% of the time. With this aim, in the framework of the Pierre Auger Observatory a new reconstruction procedure has been developed, based on the concept of air shower Universality. According to the Universality paradigm, extensive air showers created by ultra-high energy cosmic rays can be described, to a remarkable degree of precision, in terms of a reduced set of macroscopic parameters: shower geometry, energy, Xmax and muon content. The socalled Universality reconstruction consist in a best-fit of the values of these macroscopic parameters to describe the data collected by the water Cherenkov detectors of the ground array, comparing measurements and expectations from specific parameterizations of the integrated and time-dependent signals. The work I carried out during my PhD led to two main outcomes. The first one is an improvement of the fitting procedure already implemented in the Offline software of the Collaboration, to achieve a higher accuracy in the reconstructed quantities, in particular of the Xmax determination. The adjustments were guided by the comparison between the results of the Universality and the standard reconstruction for a selected set of high quality hybrid events, i.e. events measured simultaneously by the surface and fluorescence detectors. In this way an improved likelihood function and a better description of uncertainties in the time-dependent signal are achieved, allowing the definition of an optimized fitting procedure. The second outcome is the reconstruction of the dataset collected by the surface detector between 2004 and 2016, using the optimized version of the Universality fit. The resulting values of the shower maximum are used to calculate the first two moments of the Xmax distributions above 1019 eV, from which the detector independent mean and variance of the logarithmic mass, ln A, of the primary cosmic rays reaching the Earth are obtained. Finally, the Xmax distributions in their entirety are used to estimate the contribution of different groups of masses (light, intermediate, heavy) and to attempt a study of the energy spectrum separated in a lighter and a heavier component.