Disentangling atmospheric cascades started by gamma rays from cosmic rays with CORSIKA

• Autor: Rengifo Gonzáles, Javier
• Formato: tesis de maestría
• Fecha de Publicación: 2017
• Institución: Consejo Nacional de Ciencia Tecnología e Innovación
• Repositorio: CONCYTEC-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.concytec.gob.pe:20.500.12390/1840
• Enlace del recurso: https://hdl.handle.net/20.500.12390/1840
• Nivel de acceso: acceso abierto
• Materia: Rayos Gamma; Rayos cósmicos; Partículas (Física nuclear); https://purl.org/pe-repo/ocde/ford#1.03.00

In this work we search for a method to differentiate between particle showers produced by cosmic rays and by gamma rays at TeV energies, using CORSIKA simulations. This method tries to solve the dominant hadron flux background problem when looking for gamma-ray signals measured by different experiments. The results of this work can be applied to the study of Gamma-Ray Bursts (GRBs). GRBs emit very energetic photons, which after interacting in the Earth's atmosphere, produce a large detectable electromagnetic cascade of secondary particles. The procedure will be to simulate events produced by photons, the signal, and protons, the most abundant cosmic-ray background. We extract several parameters from fitting particle air-shower longitudinal profiles, characterizing the simulated showers. Some of the most important _t parameters are the shower maximum (Xmax), the width of the shower FWHM, the asymmetry parameter, the maximum number of particles Nmax and the shower start Xstart. There are different experiments using water Cherenkov tanks and fluorescence detectors which can measure these shower parameters. We tested two methods. The first relies on simple cuts, while the second is based on a multivariate analysis using the TMVA package, which improves individual cuts. The first method was applied to single simulated energies of 102, 103, 104 and 105 GeV to find adequate cuts. We found that Xmax, FWHM, Xstart and Nmax depend on the energy. Later we applied these energy-dependent cuts and other fixed cuts to a realistic sample, which consists of 104 signal events (photons) and 106 background events (protons) covering an energy range from 102 to 105 GeV with different spectra. Moreover, we introduced an energy smearing to simulate a detector energy reconstruction efficiency. The obtained result leaves 54% signal events and 12% background events. Applying the multivariate analysis TMVA, we found that the Boosted Decision Trees (BDT) method was the best for distinguishing signal from background. The result for a similar signal efficiency was 0:7% of background events. Finally using tighter cuts on the BDT to improve the significance results in 1 photon event for every 1000 protons. Given the initial ux proportion, it means a 103 background rejection capability. Thus the feasibility of gamma/hadron separation requires further improvement.