Applied computing to study structural and enviromental properties of SDSS's galaxies / Computação aplicada ao estudo das propriedades estruturais e ambientais de galáxias do SDSS

Detalhes bibliográficos
Ano de defesa: 2017
Autor(a) principal: Diego Herbin Stalder Diaz
Orientador(a): Reinaldo Roberto Rosa, Reinaldo Ramos de Carvalho
Banca de defesa: Haroldo Fraga de Campos Velho, André Luís Batista Ribeiro, Irapuan Rodrigues de Oliveira Filho
Tipo de documento: Tese
Tipo de acesso: Acesso aberto
Idioma: eng
Instituição de defesa: Instituto Nacional de Pesquisas Espaciais (INPE)
Programa de Pós-Graduação: Programa de Pós-Graduação do INPE em Computação Aplicada
Departamento: Não Informado pela instituição
País: BR
Link de acesso: http://urlib.net/sid.inpe.br/mtc-m21b/2017/05.01.11.27
Resumo: The exponential growth of data from cosmological simulations and observational catalogs has motivated the development and application of new computational techniques for the study of galaxy properties. In this context, two topics are addressed in this thesis in applied computing: (i) The study of the galaxy structural properties using a Bayesian approach; (ii) The investigation of the gaussianity of the velocity distribution of groups and clusters. We study the use of a Bayesian approach for modeling images of elliptical galaxies using a tool called GALPHAT (GALaxy PHotometric ATtributes). This work has improved the accuracy of the numerical integration involved in this application, as well its capability to handle a large data sets. Thus, the present research proposes a new pipeline, written in python, for GALPHAT, called PyPiGALHAT, developed and tested, to analyze of a large set of galaxies in a high performance computing environment (HPC). PyPiGALPHAT has been validated considering several sets of synthetic galaxy images, generated using Sérsics law. This application allowed us to improve GALPHAT and measure its ability to recover the true galaxy parameters. The results indicate that the Bayesian approach provides more robust and reliable values, compared to frequentist approaches (GALFIT). Once the improvement was established via PyPiGALPHAT, it was applied to real images of bright elliptical galaxies observed by the Sloan Digital Sky Survey (SDSS). The results of SDSS data analysis indicate that the use of PyPiGALPHAT provides complementary informations and more reliable results than a frequentist approach (eg. GALFIT). The second part of this project is related to the study of a new systematics to characterize the galaxy environment. In general the environment is defined in terms of the local density of galaxies or the mass of the dark matter halo mas of the cluster / group. In this case, we classify the groups according to their galaxy velocity distribution. We study two particular techniques to measure how far the distributions are from a Gaussian, which indicates the state of equilibrium of the system. The first method, try to identify a mixture of gaussians (two) for justifying the velocity distribution while the second simply measures the distance between two distributions (Hellingers distance). We have shown that our measurements of gaussianity are robust and reliable, and that the environment is correlated with galaxy properties, suggesting that gaussian systems have a higher infall rate, assembling more galaxies which suffered a preprocessing before entering the groups. This technique, unprecedented in cosmological applications, has proved to be an excellent tool for analyzing large-scale structures in the Universe.