Séminaire

To support the MEGHA-Tropiques space mission, a cloud mask and a cloud type classification are needed at high spatial and time resolutions over the tropical belt (30°S-30°N). For this purpose, visible and infrared radiance data from all geostationary satellites are used with the SAFNWC (Satellite Application Facility for Nowcasting) algorithm initially developed for Meteosat Second Generation. This algorithm is adapted to the spectral characteristics and field of view of each satellite here. The cloud cover characteristics (cloud mask, cloud classification and cloud top pressure) obtained by this method are evaluated over four months of summer 2009 against CALIOP lidar observations from the CALIPSO sun-synchronous satellite. To better identify atmospheric and instrumental issues, separate analysis is performed over land and ocean areas, for midday and midnight CALIPSO overpasses and for each geostationary satellite. It is found that each classification has specific features, which depend on observed cloud regimes and on the instrument capabilities. Most importantly, common behaviour depending on cloud types is found when comparisons from both statistical and pixel-to-pixel analyses are considered. The overall bias between CALIOP and geostationary satellite cloud occurrence is 10%, with the largest values over land during daytime and the smallest over ocean during daytime. A more detailed analysis reveals specific discrepancies in the retrieved cloud types. As expected, high clouds are detected more frequently by lidar, but it is shown that multi-spectral radiometry performs very similarly over the ocean when the optical depth of detected high clouds is limited to 0.1. However, the most significant difference is due to the non-detection of low clouds in some regions, which causes a reduction of up to 20% and even 30% in the low cloud fraction in the central Pacific region. Other differences are seen over land, where mid-level clouds are not detected or are miss-classified .

MP Lefebvre 0144272799