Skip to Main content Skip to Navigation

Matter in the largest structures of the Universe : from galaxies to filaments, observations and data analysis

Abstract : Studying the evolution and the composition of the largest stuctures of the Universe, e.g., galaxy clusters and cosmic filaments, is one of the most challenging research topic in cosmology. During my phD thesis, I have analysed publicly available multi-wavelength surveys (namely SDSS, WISE, Planck), using new techniques in data analysis such as machine learning or Bayesian methods. The purpose of my work was to study the baryonic matter (hot gas and galaxies) in the intra-cluster medium, and in the cosmic web (bridges of matter between clusters of galaxies and cosmic filaments).During my thesis, I have developed an algorithm based on machine learning in order to estimate galaxy properties, such as star formation rate and stellar mass in the redshift range 0.050.5 were observed with VLT/FORS2. The analysis of the ~3000 obtained spectra is still ongoing and with the actual data, we expect the spectroscopic confirmation of about 10 galaxy clusters with about 10 members per clusters.The gas content of the cosmic web may account for about 40% of the baryons. Therefore, it is important to use optimised observational tracers of the gas. In this purpose, I have developed a deep learning algorithm using the full potential of the Planck data to detect the SZ effect tracing galaxy clusters and hot ionised gas. I have trained a convolutional neural network on the Planck frequency maps to recognize clusters selected in tSZ, X-rays, and optical. With this trained model I have generated a full-sky map of low noise SZ signal, in which I have detected about 10 times more candidates than the first Planck catalogue of SZ sources. I have shown that deep learning seems a promising approach to improve the y-map and detect smaller pressure halos and diffuse gas in the cosmic web.
Complete list of metadata

Cited literature [379 references]  Display  Hide  Download
Contributor : Abes Star :  Contact Connect in order to contact the contributor
Submitted on : Friday, June 26, 2020 - 6:34:13 PM
Last modification on : Thursday, September 9, 2021 - 5:33:14 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02882460, version 1


Victor Bonjean. Matter in the largest structures of the Universe : from galaxies to filaments, observations and data analysis. Astrophysics [astro-ph]. Université Paris sciences et lettres, 2019. English. ⟨NNT : 2019PSLEO010⟩. ⟨tel-02882460⟩



Record views


Files downloads