• Citer

Radiatively consistent retrieval of cloud, aerosol and surface properties from space-based observations in the solar spectral region

par Luffarelli, Marta
Président du jury Coheur, Pierre
Promoteur Clarisse, Lieven
Publication Non publié, 2022-06-03

Thèse de doctorat

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

Satellite instruments operating in the visible and near-infrared parts of the electromagnetic spectrum offer a unique opportunity to study the global distribution of aerosols and their optical properties. These can be obtained from the observed satellite radiances through specialised inversion algorithms. While significant progress has been achieved in the past decade, current algorithms still face three outstanding challenges. First, there is in general a strong radiative coupling among the different retrieval vari- ables. When this radiative coupling is not correctly accounted for or when different variables are retrieved with different radiometric assumptions, inconsistencies arise. Second, aerosol retrievals from satellites are normally performed in clear sky situations only, excluding a safety area around clouds to avoid potential cloud contamination in the aerosol measurements. However, recent studies suggest that excluding pixels in the vicinity of clouds from aerosol retrieval could potentially lead to aerosol load underestimation due to size swelling in more humid environments near clouds. Finally, the large availability of satellite data makes it desirable to be able to apply the same algorithm to different sensors.In this PhD project, these challenges are addressed with the development of the Combined Inversion of Surface and Aerosol (CISAR) algorithm for the joint retrieval of surface reflectance and aerosol single-scattering properties. It features a continuous variation of the state variables in the solution space, and by considering the coupling between the surface and atmosphere, delivers a radiatively consistent product. Thanks to the flexibility of its inversion algorithm, CISAR can be applied to different sensors operating in the visible and near-infrared onboard satellites flying at different orbits. The CISAR algorithm has also been adapted to include the retrieval of cloud properties, extending the aerosol retrieval to the vicinity of clouds to deepen the study of aerosol properties in such conditions. The first part of the thesis introduces aerosols and clouds and their retrieval from satellites. The second part consists of the three research papers detailing all the scientific results obtained during the PhD. The first paper (Govaerts and Luffarelli, 2018) presents the theoretical concepts of the CISAR algorithm. The second paper (Luffarelli and Govaerts, 2019) demonstrates CISAR’s applicability to both geostationary and polar-orbiting satellites and evaluates its performances. The third paper (Luffarelli et al., 2022) describes the extension of the algorithm to the retrieval of clouds and discusses the innovative combined aerosol-cloud product during a dust storm originating from Africa in June 2020.This research opens up new ways of approaching aerosol retrieval from satellites, al- lowing the retrieval in situations that are normally discarded by aerosol retrieval algorithms. This could allow a more realistic estimation of the direct effect of aerosols on the radiative forcing.