A First Look at Inverse Problems for the Atmospheres of M and G Earth-like Planets

Abstract

This work addresses the main question regarding how much can we learn from spectroscopic observations of Earth-like exoplanets along with what can we identify in terms of atmospheric conditions and biosignatures that may be present on the planet and serve as evidence of life. Using a high resolution infrared radiative transfer model, we calculate the emission spectra of Earth-like exoplanets and their partial derivatives with respect to the unknowns of the inverse problem (i.e. the Jacobians). Sensitivity analysis based on Jacobian matrices revealed which parameters can be retrieved independently and which parameters have to be set a priori, depending on the type of observation and spectral domain. A set of few optimized state vectors were used for extensive retrieval simulations investigating impact of the initial guess and a priori data information content. The calculations provided information on data quality depending on the spectral range, resolution, and number of molecular species considered in the atmosphere. Larger correlations were found between certain parameters, what provided information about the difficulty of retrieving a certain parameter independently. A sensitivity study has also been conducted in order to assess the feasibility of retrieving CO2 and H2O molecular concentrations from an observed spectrum with a 1% noise level. Also, comparisons between analytical and numerical Jacobians were studied.