The recent developments of infrared astronomy, and in particular space-based observatories, have revealed the ubiquitous presence of water in the Universe, including the unexpected and puzzling presence of water in warm giant and supergiant stars. In order to properly interpret these observations and prepare programs with future facilities, The new method I develop allows to solve radiative transfer for molecules in the conditions prevailing in these stars, which are out of equilibrium. This will have a strong impact on our understanding of these stars, their mass loss and their role in the chemical evolution of our Galaxy.Potential applications of the code in other research areas (e.g. planetary atmospheres, greenhouse effect) will also be examined.Researchers at Lund Observatory are strongly involved in the preparation of future international astronomical facilities, either ground-based (Extremely Large Telescope) and space missions (SOFIA, GAIA), as they will strongly impact our knowledge in stellar and Galactic physics. The exhaustive analysis of the forthcoming data will require precise analysis tools such as the code I developed. This will also reinforce the collaboration with the spectroscopy group.

Water is an important constituent of the atmosphere of red supergiant stars (RSG), which influence remains however poorly understood. The water spectrum of these stars can apparently be only reproduced through the addition of a detached shell of cool molecular gas, the so-called MOLspheres. However, this hypothesis is still cautious. In order to better interpret observed spectra, non local thermodynamic equilibrium (NLTE) spectrum synthesis may be potentially important. NLTE effects being potentially important, they may alleviate the problems in the interpretation of water spectra, and affect the atmosphere dynamics. We thus developed an original method to solve the radiative transfer equation, adapted to the numerous water transitions and without the LTE approximation. This method has been implemented in an original parallel code. Preliminary results show that NLTE effects in RSG atmospheres and their impact on observables such as the emergent spectrum are very important.

Development of a new NLTE radiative transfer code in red supergiantes atmospheres context.