Probing the effects of magnetic fields on ultra-high energy cosmic ray arrival directions at the Pierre Auger Observatory

When ultra-high-energy cosmic rays (UHECRs) travel from sources to Earth, they are deflected by extragalactic and galactic magnetic fields. Since the deflection depends on the charge of the nuclei, UHECRs with very high magnetic rigidity propagate almost ballistically. Consequently, when detected on Earth, the arrival directions point near their origin. Hence, backtracking the high-rigidity UHECRs could set a limit on different source classes.

However, indirect detection of UHECRs poses challenges in obtaining information about their energy and mass simultaneously. Machine learning-based mass estimators show the potential to improve the reconstruction of mass-sensitive variables, such as the depth of the shower maximum.

In this contribution, we investigate the effect of the galactic magnetic field on the propagation of the UHECRs using neural network-based mass estimators. We test different scenarios of source distributions. In particular, we present the developed methodology to test the hypothesis that sources follow the distribution of matter in space, hence the Supergalactic plane. We use data from the Pierre Auger Observatory.