Thesis of Raphael Peroni

Defense of thesis Raphael Peroni - laboratory LOA

Remote sensing of water vapour content above and around convective clouds

Abstracts :

Despite significant advances in atmospheric physics research over the past few decades, many uncertainties persist regarding our understanding of climate change. Current knowledge indicates that clouds play a major role in these uncertainties due to complex interactions involving aerosols, water vapor, clouds, global atmospheric circulation, convection, and precipitation. Water vapor plays a crucial role in clouds formation and development, especially those resulting from convective phenomena that redistribute water vapor in the atmosphere through exchanges between clouds and their immediate environment. Therefore, a better understanding of water vapor content above and around clouds is necessary to improve our comprehension of interactions between water vapor and clouds and to help the scientific community better constrain LES models and numerical weather forecasting models. Our research is part of the C³IEL space mission, which aims to enhance our knowledge of the 3D envelope of convective clouds, their horizontal and vertical development velocities, the water vapor content above and around clouds, and the electrical activity associated with these convective systems. The focus of this thesis concerns the retrieval of integrated water vapor content in the presence of clouds from satellite observations. This retrieval was achieved through a Bayesian probabilistic approach: the optimal estimation method. So far, few studies have explicitly demonstrated the feasibility of such inversion under cloudy atmosphere because of the complexity related to the penetration and scattering of radiation within the cloud. This increases the number of parameters involved in the relationship between radiance and water vapor content.

Radiative transfer simulations were conducted in the three SWIR spectral bands defined for the study of water vapor content in the context of the C³IEL mission. The atmosphere was assumed to be composed of homogeneous plan-parallel layers, and synthetic radiance datasets were generated for testing the retrieval algorithm developed in this thesis. The feasibility of retrieving integrated water vapor content above a cloud and over the ocean from SWIR radiances was shown with a precision of approximately 1 kg/m² for optically dense clouds. However, the precision of this retrieval decreases as the cloud optical thickness decreases. Tests were then realized with realistic water vapor and cloud extinction profiles that present non-homogeneous cloudy vertical profiles. This shows that integrated water vapor content above liquid water clouds could be retrieved with a positive bias related to cloud penetration of approximately 2.18 kg/m². This value is of the same order of magnitude than those obtained in previous work under clear-sky conditions. In the presence of convective clouds containing both liquid and ice water, characterized by a significant vertical extension and thus a high top altitude, very high optical thickness and very low water vapor content, the retrieval algorithm does not succeed to provide a valid retrieval. Suggestions are therefore proposed to improve water vapor content retrievals in realistic cases and define the retrievable limit for water vapor content.

Keywordss : Water vapor,Remote sensing,Optimal estimation,Clouds,C³IEL