FSU Students Present Machine Learning Oceanography Research at Ocean Sciences Meeting 2026 in Glasgow

Students Jose Miranda and Jorge Velasco represented Florida State University's Department of Scientific Computing and the Center for Ocean-Atmospheric Prediction Studies (COAPS) at the 2026 Ocean Sciences Meeting held in Glasgow, Scotland this past February, presenting innovative research that applies machine learning to advance oceanographic observation and analysis.

NeSPReSO: Synthetic Ocean Profiles for the Gulf of Mexico

Jose Miranda presented "Neural Synthetic Profiles from Remote Sensing and Observations (NeSPReSO), an Open Web Interface and API for Synthetic Temperature and Salinity Profiles in the Gulf of Mexico." His work introduces a neural-network-based framework that integrates satellite observations and in-situ measurements to generate synthetic temperature (T) and salinity (S) profiles. The NeSPReSO system provides an accessible web interface and API, enabling researchers to easily retrieve high-resolution subsurface ocean estimates for the Gulf of Mexico and supporting broader data accessibility for oceanographic research.

Enhancing Sea Surface Height with Machine Learning

Jorge Velasco presented "Improving Sea Surface Height Resolution Using Satellite-Derived Chlorophyll-a and Temperature with Machine Learning." His research explores how biological and thermal satellite observations can be leveraged through machine learning models to enhance the spatial resolution of sea surface height estimates. By fusing multiple satellite-derived variables, the approach aims to improve the representation of ocean dynamics and mesoscale features.

Their participation highlights the department's continued contributions to interdisciplinary research combining advanced computational methods, machine learning, and oceanographic data to improve our understanding and monitoring of the marine environment.

Originally published by the FSU Department of Scientific Computing.