Ultra-short period (USP) planets, which orbit their host stars in less than a day, represent one of the most intriguing areas of development in exoplanetary science. These planets can offer unique insights into planetary formation, migration, and survival under extreme conditions, such as intense stellar radiation and strong tidal forces. With a growing catalogue of USP planets from missions like Kepler, CoRoT, and TESS, as well as spectroscopic data from high-resolution spectrographs such as HARPS, HARPS-N, and ESPRESSO, we are now capable of characterizing these planets with the indirectly obtained radii, masses, and bulk densities. However, usually only some system configurations are tested based on the Generalized Lomb-Scargle Periodogram (GLS) and the effect of stellar activity is not properly considered, which can lead to inaccurate results. This thesis aims to homogeneously re-analyse a sample of USP planets using radial velocity (RV) data by testing different models, finding more likely configurations and obtaining or improving the previously obtained planetary parameters. To this end, stellar parameters are uniformly determined through isochrone fitting, while RV and stellar activity data are gathered from literature and available catalogues. To mitigate the impact of stellar noise in the RV data, Gaussian processes (GP) are used within a Bayesian framework implemented in PyORBIT, a Python package that supports both Markov Chain Monte Carlo (MCMC) and Nested Sampling algorithms for robust parameter estimation and model selection. Multiple models are tested, including different methods for stellar activity mitigation using GPs and varying number of planets for each planetary system. We also verify the equivalence of two mathematical implementations of the same kernel for GPs. The statistical relevance of each model is calculated with the Bayesian evidence and information criteria. The final set of planetary parameters are obtained with MCMC algorithms. The derived properties of the USP sample are presented, together with the orbital architecture of their planetary systems. The sample is situated in a mass-radius plot to contextualize it within planetary composition theoretical models.
Ultra-short period (USP) planets, which orbit their host stars in less than a day, represent one of the most intriguing areas of development in exoplanetary science. These planets can offer unique insights into planetary formation, migration, and survival under extreme conditions, such as intense stellar radiation and strong tidal forces. With a growing catalogue of USP planets from missions like Kepler, CoRoT, and TESS, as well as spectroscopic data from high-resolution spectrographs such as HARPS, HARPS-N, and ESPRESSO, we are now capable of characterizing these planets with the indirectly obtained radii, masses, and bulk densities. However, usually only some system configurations are tested based on the Generalized Lomb-Scargle Periodogram (GLS) and the effect of stellar activity is not properly considered, which can lead to inaccurate results. This thesis aims to homogeneously re-analyse a sample of USP planets using radial velocity (RV) data by testing different models, finding more likely configurations and obtaining or improving the previously obtained planetary parameters. To this end, stellar parameters are uniformly determined through isochrone fitting, while RV and stellar activity data are gathered from literature and available catalogues. To mitigate the impact of stellar noise in the RV data, Gaussian processes (GP) are used within a Bayesian framework implemented in PyORBIT, a Python package that supports both Markov Chain Monte Carlo (MCMC) and Nested Sampling algorithms for robust parameter estimation and model selection. Multiple models are tested, including different methods for stellar activity mitigation using GPs and varying number of planets for each planetary system. We also verify the equivalence of two mathematical implementations of the same kernel for GPs. The statistical relevance of each model is calculated with the Bayesian evidence and information criteria. The final set of planetary parameters are obtained with MCMC algorithms. The derived properties of the USP sample are presented, together with the orbital architecture of their planetary systems. The sample is situated in a mass-radius plot to contextualize it within planetary composition theoretical models.
Homogeneous radial velocity analysis of ultra-short period planets
PACHECO DE MEDEIROS MESTRE, JÚLIA CRISTINA
2023/2024
Abstract
Ultra-short period (USP) planets, which orbit their host stars in less than a day, represent one of the most intriguing areas of development in exoplanetary science. These planets can offer unique insights into planetary formation, migration, and survival under extreme conditions, such as intense stellar radiation and strong tidal forces. With a growing catalogue of USP planets from missions like Kepler, CoRoT, and TESS, as well as spectroscopic data from high-resolution spectrographs such as HARPS, HARPS-N, and ESPRESSO, we are now capable of characterizing these planets with the indirectly obtained radii, masses, and bulk densities. However, usually only some system configurations are tested based on the Generalized Lomb-Scargle Periodogram (GLS) and the effect of stellar activity is not properly considered, which can lead to inaccurate results. This thesis aims to homogeneously re-analyse a sample of USP planets using radial velocity (RV) data by testing different models, finding more likely configurations and obtaining or improving the previously obtained planetary parameters. To this end, stellar parameters are uniformly determined through isochrone fitting, while RV and stellar activity data are gathered from literature and available catalogues. To mitigate the impact of stellar noise in the RV data, Gaussian processes (GP) are used within a Bayesian framework implemented in PyORBIT, a Python package that supports both Markov Chain Monte Carlo (MCMC) and Nested Sampling algorithms for robust parameter estimation and model selection. Multiple models are tested, including different methods for stellar activity mitigation using GPs and varying number of planets for each planetary system. We also verify the equivalence of two mathematical implementations of the same kernel for GPs. The statistical relevance of each model is calculated with the Bayesian evidence and information criteria. The final set of planetary parameters are obtained with MCMC algorithms. The derived properties of the USP sample are presented, together with the orbital architecture of their planetary systems. The sample is situated in a mass-radius plot to contextualize it within planetary composition theoretical models.File | Dimensione | Formato | |
---|---|---|---|
PachecodeMedeirosMestre_JuliaCristina.pdf
accesso aperto
Dimensione
7.75 MB
Formato
Adobe PDF
|
7.75 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/79649