AI-based particle dispersion model

The particle dispersion model used is Gnome. To drive the dispersion simulations, a neural network is first employed to model the sea current for the next six hours, using the wind and current data from the previous six hours as input. The current signal forecasted by the neural network is then used as a forcing input for Gnome to model the particle dispersion. Full metadata: https://erddap.s4raise.it/erddap/info/unige-dicca_dispersion_forecast_ai/index.htmlThis dataset is part of the RAISE Spoke 3 project outcomes. RAISE is an innovation ecosystem funded by the Ministry of University and Research under the National Recovery and Resilience Plan (NRRP, Mission 4, Component 2, Investment 1.5).