Scientific geovisualization of the dynamics of Sargassum dispersion and landfall in the Caribbean, based on satellite imagery and numerical forecasts.

Abstract

Listen

Translate article

This study focuses on the spatial and temporal representation of Sargassum dispersal and landfall dynamics. An automated prototype is developed incorporating the following components: 1) Detection of Sargassum Rafts: Individual sargassum rafts are identified using Sentinel-2 images with a revisiting period of five days. 2) Forecasting/Hindcasting Vector Fields: One-week forecasts (or hindcasts) are obtained at hourly intervals for the primary forces affecting raft movement—currents, tides, waves, and wind—using supercomputing services (Copernicus Marine Service) 3) Lagrangian Simulation: The movement of detected rafts in step 1 is simulated using the vector fields obtained in step 2. For statistical purposes, rafts that land or drift outside the simulation range are logged with details of location, date, and time. 4) Animation Generation: Four animations are produced to visualize: a) Rafts movement, b) Rafts trajectories, c) The dynamics of surface forcings (currents, tides, and waves), and d) The dynamics of above-surface factors (i.e. wind drag, modeled as a percentage of wind speed). 5) Interactive 3D Visualization: All elements are integrated into an interactive globe featuring 3D bathymetry, allowing users to explore sargassum dispersion and landfall predictions (or hindcasts) for specific satellite observation dates. While the prototype shown takes into account all elements of a monitoring system, it should not be considered as an operational early warning system. Further steps, beyond the scope of this study, would be required, including optimization of the remote sensing technique, improvement of the transport simulation methods, and an experimental framework for accuracy assessment. In this study, we present the interactive geovisualization and discuss its potential for improving the scientific understanding of Sargassum dispersal and landfall patterns, as well as its potential for the implementation of coastal management policies. A comparison is made with existing systems, highlighting the limitations and advantages of our approach while discussing its potential for developing a robust Sargassum monitoring and early warning system for the Caribbean Sea.