Seminars at FGG
Exoplanet Atmospheric Characterization Using GUIBRUSHR: A Graphical User Interface for Bayesian Retrieval with High-Resolution Spectroscopy
Speaker: Francesco Amadori (INAF - Osservatorio Astrofisico di Torino (Italy))
Date and time: 2025-03-13 11:30
The field of exoplanetary research has advanced significantly in recent years, particularly in the characterization of exoplanetary atmospheres. Spectroscopic techniques reveal atmospheric composition and conditions by analyzing starlight during the planet transit (transmission) or emission from the planet’s dayside, with high-resolution spectroscopy providing detailed molecular insights. A tool currently in development for this purpose is GUIBRUSHR (Graphical User Interface for Bayesian Retrieval Using Spectroscopy at High-Resolution), a Python-based software designed to integrate data from both ground-based high-resolution (HR) instruments and space-based low-resolution (LR) telescopes. This synergy enables users to retrieve atmospheric properties using LR data from telescopes such as JWST and HST, alongside HR data from instruments like GIANO-B, HARPSN and the upcoming ANDES on the ELT. An advantage of combining HR and LR spectroscopy is the ability to inspect different atmospheric layers: HR is more sensitive to the line core, probing the upper atmosphere, while LR provides information on the lower layers. Additionally, HR can resolve individual spectral lines, allowing the identification of refractory elements and non dominant species in the bands. The software implements Differential Evolution Markov Chain Monte Carlo (DEMCMC) for Bayesian retrieval, optimizing parameter estimation. It supports radiative transfer codes such as PetitRADTRANS (Mollière et al., 2019) and Pyrat Bay (Cubillos et al., 2021) for both emission and transmission spectroscopy. Additionally, GUIBRUSHR enhances multi-instrument data integration, allowing for a more robust atmospheric retrieval process. GUIBRUSHR is still in development, but preliminary analyses have already been conducted to assess its scientific and technological performance. While some aspects still require improvements, others -such as cross-correlation- already have been validated with results from the literature.