Thématique : Cosmologie observationnelle
The various observations of the Universe have been indicating for twenty years now that the
expansion of the Universe is accelerating. The standard model of cosmology, known as the
CDM model, describes the Universe as composed of 27% dark matter and 68% dark energy.
Understanding the nature of these two energy components remains one of the greatest
challenges in contemporary physics.
The future Euclid space mission is dedicated to the study of dark energy and dark matter in
the Universe and to test gravity on cosmological scales. Euclid was selected by the European
Space Agency (ESA) in 2011 and will be launched in 2022 to probe the Universe over a 6 yearperiod.
These data will revolutionize our ability to map the Universe and better understand
the nature of dark energy or put Einstein's General Relativity (GR) in default.
Two instruments will be embarked on board Euclid, the Near Infrared Spectrometer and
Photometer (NISP) and the visible imager (VIS). The spectroscopic survey with the NISP
instrument will target fifty millions of galaxies in the redshift range 0.9 < z < 1.8. The
photometric survey will get the image and photometric redshift of two billions of galaxies
down to a magnitude of 24.5 AB covering the redshift range 0 < z < 2.5.
The subject of the thesis is to measure the galaxy clustering from the Euclid photometric
catalog using a tomographic approach. The tomographic approach (2D) has several
advantages compared to the standard 3D mapping. On the one hand, the number of observed
objects is two orders of magnitude larger, since 2 billion galaxies are expected in imaging
compared to the 50 million galaxies expected in spectroscopy. The division into several tens
of slices in redshift makes it possible to keep a competitive statistic in each slice. On the other
hand, the angular approach does not require fiducial cosmology for the conversion of
distances, which makes it possible to constrain the cosmological models as cleanly as possible.
The interest of applying this method to Euclid data is to take advantage of the double
photometric/spectroscopic observation on the same fields to calibrate the photo-z.
The first step of the thesis work will consist in calibrating the photometric redshifts (photo-z)
of Euclid from the spectroscopic sample. This step will be done first from the Euclid Flagship
simulation which contains more than 3 million objects. A machine learning approach will be
considered. Then the angular clustering tools will be developed in order to extract the
clustering signal from the photometric samples. This analysis will require the optimization of
the processing chain in order to be able to process more than a thousand mocks for the
calculation of covariance matrices. The PhD student will be member of the Euclid Consortium,
with full access to Euclid data.
Cosmology, Dark Energy, Dark Matter, General Relativity, Galaxy Clusturing, BAO, Data Analyses, Tomograpgy, Big Data, Deep Learning, EUCLID, NISP, Photometry, Spectroscopy
<strong>Applicant profile :</strong>
Master in fundamental physics or astrophysics. Interest for cosmology and machine learning
tools. Programming skills (python, C++), strong motivation, ability to work in teams and in
large collaborations are highly recommanded.