Planetary Atmospheres

Atmospheric retrieval frameworks allow us to provide estimates on the main atmospheric and planetary parameters from an observed exoplanetary spectrum. Using a Bayesian approach, the space of input parameters is explored iteratively to assess which combination of values explains the data best.

While being the standard approach for the characterisation of Hot Jupiters, this technique has yet to be applied systematically to terrestrial exoplanets. Such planets are supposed to be very common objects, but they are still very challenging to be detected because of the extremely low contrast in signal compared to their host star. Nonetheless, Earth twins and Super Earths represent very good candidates for hosting life in the universe and a statistical quantitative approach concerning the habitability of such exoplanets or the existence of biosignatures in their atmospheres is needed. First steps in this direction were presented in the context of reflected light observations with missions like NASA’s HabEx or LUVOIR.

In our group we are developing and utilizing a Bayesian atmospheric retrieval approach by combining the 1D forward model petitRADTRANS (Mollière et al., 2019) and the Nested Sampling algorithm MultiNest (Feroz et al., 2009) for Bayesian parameter inference. As one of the first applications we are investigating how planetary/atmospheric properties of an Earth-twin exoplanet can be derived from the MIR thermal emission spectrum. This work is done in the context of the LIFE initiative and hence we include a corresponding noise model in our simulated data. We investigate how the derived planetary/atmospheric parameter depend on the wavelength range, the SNR and the spectral resolution R of the data (Konrad et al., in prep., see Figure 1 for an example). Cloud coverage, host star type and atmospheric/surface composition also play a role in shaping the emission spectrum and will be taken into account in the near future (Alei et al., in prep).

Overall, this work will deliver a key piece for a statistical framework assessing habitability (see Catling et al. 2017 for a recent proposal) and will also inform us about the knowledge to be gained from future observations with upcoming facilities.