Heidelberg 2022 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 71: Neutrino Astronomy 3

T 71.6: Vortrag

Mittwoch, 23. März 2022, 17:30–17:45, T-H30

Data reduction for the Radio Neutrino Observatory Greenland — •Zachary Meyers for the RNO-G Collaboration collaboration — DESY, Platanenallee 6, 15738 Zeuthen, Germany — Erlangen Center for Astroparticle Physics (ECAP), Friedrich- Alexander-University Erlangen-Nuremberg, 91058 Erlangen, Germany

Continuing the search for utra-high energy neutrinos (> 10 PeV) beyond the range of optical detection methods, the Radio Neutrino Observatory Greenland (RNO-G) is now online after a successful first season of deployment. Total data taken during the shortened 2021 campaign from the three operational stations amounts to nearly ten million recorded events, requiring more than 330GB of storage. While this could be considered a manageable sum, next year another 7 stations are planned to come online, while the complete array will consist of 35 total. And for future experiments, requiring hundreds of similar stations, the data volumes rapidly increase to a level where it is no longer feasible to run direction and energy reconstruction algorithms on the entire dataset. Low level cuts must be made early in the data processing stages (or even onboard the detector itself in real time) in order to be computationally efficient. In an attempt to discriminate between thermal noise fluctuations, anthropogenic noise and neutrino-like signal, we show the potential effectiveness of deep learning approaches, specifically convolutional neural networks (CNNs), in both the time and frequency domains. When compared and combined with more traditional methods such as matched filtering, a comprehensive strategy for post trigger filtering can be established.