Abstract
Since 2011, massive strandings of holopelagic Sargassum have occurred on the coasts of the Caribbean and of West Africa. Although open ocean Sargassum mats are oases of biodiversity, their stranding has a number of negative ecological, economic and health consequences. To limit these impacts, Sargassum needs to be collected as quickly as possible to avoid its decomposition, which requires accurate predictions of the date, location and abundance of the strandings. Two complementary approaches have been developed for this purpose: satellite remote sensing technique, to detect Sargassum aggregations, and modeling, to forecast Sargassum displacement and growth. The objective of this review is to provide a synthesis of the current knowledge related to Sargassum monitoring in the tropical Atlantic Ocean. To better understand the issues surrounding Sargassum and its monitoring, the first two parts are devoted to an overview of the ecology of the two most prevailing holopelagic Sargassum species, to the current issues related to their strandings, to the causes of their occurrence in the tropical Atlantic Ocean and to their seasonal and interannual variabilities. The methods used to detect Sargassum from satellite images and their limitations are examined. The transport and biogeochemical models developed for seasonal forecast and stranding predictions are described along with their limitations. As both detection and modeling rely on validation data to assess their accuracy, previous works providing in situ characterization of Sargassum are also reviewed here. The last part provides recommendations to further increase knowledge on holopelagic Sargassum and improve the predictions of their strandings.