Abstract

Future direct imaging observations will start probing the actual peak of the gas giant planet occurrence rate at _{3 au. This will enable the study of the most common giant planet type, instead of making do with hot Jupiters, or today’s population of directly imaged companions, found at separations >} 10 au. The atmospheric composition of gas giant planets can be extracted from their emission spectra, and is widely regarded as the key for placing constraints on their formation. Because gas giant planets dynamically dominate the process of planet formation in the systems they reside in, understanding their formation is crucial to inform planet formation in general. In my talk I will discuss our work that strives to derive the atmospheric composition of directly imaged planets using so-called free retrievals. Compared to previous analysis methods, the strength of retrievals is their model flexibility, allowing to let the observational data constrain uncertain physical processes, most importantly the formation of atmospheric clouds. I will show how current observations of cloudy planets often lead to degenerate or biased solutions when inferring atmospheric properties, and will highlight how observations of the just-launched James Webb Space Telescope (JWST) will break such degeneracies, and enable training our methods for situations where JWST data is not available. In doing so, the atmospheric composition of cloudy directly imaged planets may finally be decipherable. I will end my presentation with a summary of how atmospheric compositions may actually allow constraining planet formation, and what obstacles need to be overcome.