SMuK 2023 – wissenschaftliches Programm

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

T: Fachverband Teilchenphysik

T 65: Gamma Astronomy III

T 65.3: Vortrag

Mittwoch, 22. März 2023, 16:20–16:35, POT/0151

Machine Learning Methods for an Increased Understanding of AGN Flares^* — •Yannick Hartych^{1, 2}, Julia Becker Tjus^{1, 2}, Wolfgang Rhode^{2, 3}, and Marcel Schroller^{1, 2} — ¹Theoretische Physik IV, Ruhr Universität Bochum, Bochum, Germany — ²RAPP-Center at Ruhr Universität Bochum, Bochum, Germany — ³Experimentelle Physik 5, Technische Universität Dortmund, Dortmund, Germany

Blazars are some of the brightest known sources in the Universe and are considered possible sources of the highest energy cosmic rays (CRs). Hence they are of high interest to astronomers to understand the processes accelerating those CR. One of those blazars is TXS 0506+056, from which a gamma-ray flare arrived in temporal and spatial coincidence with a high-energy neutrino of high probability to be of astrophysical origin. For this reason, the source was brought into focus for further investigation to understand the underlying processes leading to this observation. It is crucial to physically model blazars thoroughly. In order to find the related parameters responsible for this behaviour, we set up simulations in CRPropa3 and develop theoretical flare templates that can be compared to observational signatures. With those templates, the next step would be to train a machine learner to search the galactic catalogues for other blazars with a high probability of showing behaviour similar to TXS 0506+056. In this talk, we will present first preliminary results of such simulations and evaluate their significance in the context of the parameter study.
^* Financial support by the DFG (SFB 1491) is gratefully acknowledged