Heidelberg 2022 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
T: Fachverband Teilchenphysik
T 107: Data Analysis, Information Technology and Artificial Intelligence 5
T 107.5: Vortrag
Donnerstag, 24. März 2022, 17:15–17:30, T-H39
Deep-Learning-Based Reconstruction of Cosmic-Ray Masses from Extensive Air Shower Measurements with AugerPrime — Martin Erdmann, •Jonas Glombitza, Berenika Idaszek, Niklas Langner, and Dominik Steinberg — III. Physikalisches Institut A, RWTH Aachen University
Ultra-high-energy cosmic rays (UHECRs) that penetrate the Earth’s atmosphere induce extensive air showers. At the Pierre Auger Observatory, showers are measured from the ground using the fluorescence detector (FD) and the surface detector (SD) consisting of water Cherenkov detectors (WCDs). Currently, the SD is extended with scintillators (SSDs) as part of the AugerPrime upgrade.
Actual measurements of the UHECR mass composition are based on FD observations of the depth of shower maximum Xmax. Using deep learning, Xmax was successfully extracted using only the SD, exploiting the full statistics of the observatory. However, the precision of Xmax as a mass estimator at the event level is limited. The new SD upgrade offers the possibility to measure individual components of the shower, potentially improving the reconstruction of the mass composition.
We introduce our network to extract the properties of air showers by analyzing the signals of water Cherenkov detectors as well as the SSDs. We show that the mass-separation power when using only the observable Xmax is already fully exploited using only WCDs. Thus, we investigate additional observables, novel network architectures and new reconstruction strategies to increase the mass sensitivity with the combination of WCDs and SSD measurements.