Résumé
Since 2011, the distribution extent of pelagic Sargassum algae has substantially increased and now covers the whole Tropical North Atlantic Ocean, with significant inter-annual variability. The ocean colour imagery has been used as the only alternative to monitor such a vast area. However, the detection is hampered by cloud masking, sunglint, coastal contamination and others phenomena. All together, they lead to false detections that cannot be discriminated with classic radiometric analysis, but may be overcome by considering the shape and the context of the detections. Here, we built a machine learning model based on spatial features to filter false detections. More specifically, Moderate-Resolution Imaging Spectroradiometer (MODIS, 1 km) data from Aqua and Terra satellites were used to generate daily map of Alternative Floating Algae Index (AFAI). Based on this radiometric index, Sargassum presence in the Tropical Atlantic North Ocean was inferred. For every Sargassum detections, five spatial indices were extracted for describing their shape and surrounding context and then used by a random forest binary classifier. Contextual features were most important in the classifier. Trained with a multi-annual (2016-2020) learning set, the classifier performs the filtering of daily false detections with an accuracy of 90%. This leads to a reduction of detected Sargassum pixels of 50% over the domain. The method provides reliable data while preserving high spatial and temporal resolutions (1 km, daily). The resulting distribution on 2016-2020 is consistent with the literature for seasonal and inter-annual fluctuations, with maximum coverage in 2018 and minimum in 2016. In particular, it retrieves the two areas of consolidation in the western and eastern part of the Tropical Atlantic Ocean associated with distinct temporal dynamics. At full resolution, the dataset allowed us to semi-automatically extract Sargassum aggregations trajectories from successive filtered images. Using those trajectories will help to better quantify the drift of aggregations with respect to the currents, the wind and sea state. Overall, this new dataset will be useful for understanding the drivers of Sargassum dynamics at fine and large scale and validate future models.