Posters

Poster Session 3, Wednesday,  October  5, 16:00–18:00

Poster 91

Operational sargassum detection in Lesser Antilles: the Meteo-France experience

The Sargassum algae heavy beaching has been affecting the Lesser Antilles since 2011, and disturbing local ecological and economic situation. To predict the time moment and the amount of the Sargassum arrival on islands, Meteo-France has developed a full-stack operational system that includes the remote sensing, the numerical surface drift model, and visualization components, and the human final expertise to help local management. We present this operational system of Sargassum detection and forecast, giving special attention to the remote sensing component. To detect the presence of Sargassum algae in the Tropical Atlantic, we use L1 NRT optical and infrared imagery (Terra & Aqua/MODIS, Sentinel-3/OLCI, NOAA-20 & Suomi-NPP/VIIRS). The pre-processing of swath images for the Sargassum detection includes masking all potentially misleading features, such as clouds, sun-glint, land/shallow waters, and some others, and applying the Rayleigh correction. Then MCI (Maximum Chlorophyll Index, for OLCI and VIIRS) or AFAI (Alternative Floating Algae Index, for MODIS) are calculated. The Sargassum retrieval is then processed as follows (adapting the method presented in (Wang & Hu, 2016), hereafter described for MCI): (1) a “background MCI” is calculated to obtain a (2) derived MCI (dMCI), compared to an empirical threshold to produce (3) the mask containing only Sargassum pixels. Finally, several swath products are combined into a daily product, where dMCI values are normalized and reprojected on a regular grid. The final sargassum product of Meteo-France operational system was compared to other existing products (CLS, France and USF, USA) and validated by a team of local experts in the Lesser Antilles, Meteo-France.

Anastasia Tarasenko, Météo-France, [email protected], 0000-0001-9749-1974

Stéphane Saux-Picart, Météo-France, [email protected]