09 Novembre – Thesis defense - William Pluriel
14 h Room: Univers - LAB / building B18N (Pessac)
Effects of the three-dimensional structure of hot exoplanet atmospheres on observations and retrieval models.
Nowdays, we are able to discover more and more exoplanets, and even more important we can observe their atmospheres and we are starting to get information on their physical, dynamic and, chemical properties. With the new generation of space telescopes such as the JWST or Ariel, we will be able to observe spectral features that are now unobservable. However, during transit observations, the geometric structure of the atmospheres, and in particular the day to night dichotomy of hot and ultra hot Jupiters, affects the transmission spectrum and generates biases in the interpretations of retrieval outputs analysis.
The aim of my thesis is to study the effects of the three-dimensional structure of exoplanet atmospheres, and in particular the hottest ones, in order to determine the importance and the origin of these biases to allow a better analysis of the spectral data. I set up a numerical experiment simulating observations of hot Jupiters where I control the entire observational chain, from the atmosphere's simulation to the retrieval. In addition, I analyzed observations of Kelt-7 b, a hot Jupiter, from the Hubble Space Telescope, to link my numerical analysis to real data.
My work has shown that the particular structure of the hottest Jupiters significantly affects observations, implying significant biases in the results of 1D retrieval models. Although these models are valid over a large range of planets, I have demonstrated that for the hottest exoplanets, they are unable to find the abundances of species. I succeeded in quantifying these biases and in understanding their origins, hence an improvement in the interpretation of future results from 1D retrieval models. Furthermore, I conclude that it is necessary to develop 2D retrieval models to try to resolve these biases.