The project “Multi-source modelling of the ocean: Coupling Earth observations with acoustic waves” led by Leonardo Azevedo has been recommended for funding. These projects are in partnership with 2 PI's: one in PT (Leonardo Azevedo) and on in UTAustin (Prof. Tan Bui-thanh).
Creating ocean models that describe the spatial distribution of temperature and salinity is difficult and requires resource-intensive computers and very specialized knowledge of model parameterization. Despite efforts, these models are usually of low resolution, they can only predict phenomena at a kilometer scale. The main idea of this project is to create a computational tool based on machine learning algorithms that dispenses with these models using indirect measurement data (geophysics and satellite images) that exist in most regions of the globe.
'What we propose is a new paradigm in the study of the oceans. An easier-to-use tool that allows you to proxy real physical models. This allows research groups with fewer resources to obtain data and simulate scenarios that would otherwise be impossible to consider.'
Image: NASA at Unsplash