Orateur
Description
Implementation of the ultra-high energy photon candidate data stream to AMON using AugerPrime data will benefit from our design of such a data stream, which has been approved by the Pierre Auger Collaboration for the implementation. The design is based on the analysis developed in a PhD thesis by Lukas Zehrer (2021) and supplied by an internal document (GAP2021_012) describing the data stream handling workflow. Using the photon search optimization results, I will develop and apply, in coordination with the Pierre Auger Collaboration, a photon candidate data stream including the Auger Phase II (WCD/SSD/RD) data. The current surface detector (SD) array data file at the Auger data acquisition center in Malargue contains the latest (small amount of) reconstructed SD events and is updated every few minutes. This latency, similar to the one of the existing Auger data stream to AMON (see e.g. "MM Physics with Auger" paper, by Karl-Heinz et al.), is mostly due to the fully calibrated event reconstruction and analysis steps. I will use this SD data file as an input for the ROOT TMVA toolkit for the MVA analysis of the event parameters. Event selection is applied at this step. Since the goal of the proposed photon candidate data stream is to select sub-threshold events, rather loose cuts will be applied. While the analysis training will be done at a computer of the University of Nova Gorica, the classification of the Auger events will take place on the computer in Malargue. The data file, after applying the cuts, will be combined with the MVA file resulting from the training. The MVA creates a new file containing all the results. Events that have been classified in the MVA as photons (for which the MVA variable value > MVA-cut-value) will be selected with a script and written into a photon candidate event list. The format of this output event list (ASCII text, root, FITS, etc.) to be sent to AMON can be changed easily in the future if needed.