Presentation #102.354 in the session Poster Session.
Recent results indicate that Earth-sized planets orbiting within the habitable zone of Sun-like stars are extremely common. The next generation of space telescopes aims to characterize the atmospheric state of such planets, in search of potential habitable conditions suitable for the development of life.
Direct imaging is an essential technique for atmospheric characterization of planets with large orbital separation from their host star, yet our understanding of how such observations translate to constraints on atmospheric composition and habitability is still poorly developed. More precisely, it is unclear what data quality levels and precursor information are most critical for efficiently constraining their habitability.
At the stage of atmospheric and surface characterization, precursor information from radial velocity and/or astrometric observations may help place constraints on planetary orbital distance, star-planet-observer (phase) angle, and planet mass. Thus, such constrained parameters may help to significantly reduce the possible range of other inferred parameters. Yet, we do not know to what extent precursor information at different data quality levels will impact our ability to measure key planetary properties and habitability indicators.
To address this question, we use rfast, a newly developed inverse modeling framework tool to retrieve the range of atmospheric states that could explain a given set of noisy observations. We first simulate future telescope observations of an Earth twin by combining an exoplanet albedo model and an instrument noise model to generate an “observed” spectrum. Then, using a Markov Chain Monte Carlo sampler, we explore the parameters space and atmospheric states consistent with the observations.
Doing so at varying orbit/mass precursor information levels and data quality (wavelength coverage and signal-to-noise ratio) allows us to determine which prior observations and instrument design most efficiently recognize a potentially habitable world.
Here we present preliminary results obtained using our retrieval framework and discuss our ability to retrieve fiducial gas abundances, planetary bulk and orbital parameters, and cloud properties for different combinations of precursor information, spectral coverages and signal-to-noise ratios.