Researchers have developed a new deep learning-based method to measure ocean surface currents using data from existing weather satellites, offering higher-resolution insights into interactions between the ocean and atmosphere.
The approach, known as GOFlow (Geostationary Ocean Flow), analyzes thermal imagery from geostationary weather satellites to track how ocean surface temperature patterns evolve over time. By applying artificial intelligence, the system can infer the movement of underlying currents at a much finer temporal and spatial scale than traditional methods.
The study, published in Nature Geoscience, was led by scientists from the UC San Diego Scripps Institution of Oceanography and collaborating institutions.
Ocean currents play a critical role in global weather and climate systems, influencing heat distribution, carbon exchange and atmospheric processes. However, capturing their behavior at small scales and high frequency has remained a challenge. Conventional satellite methods typically revisit the same location every 10 days or so, while ship-based measurements and coastal radar systems are limited in coverage.
GOFlow addresses this gap by using weather satellite imagery captured as frequently as every five minutes. A neural network is trained to recognize how temperature patterns shift in response to ocean currents, using high-resolution simulations as a reference. The system then applies this learning to real satellite data to generate hourly current maps.
“Weather satellites have been observing the ocean surface for years,” said Luc Lenain, oceanographer at Scripps Institution of Oceanography. “The breakthrough was learning how to turn that time-lapse into hourly maps of currents by tracking how temperature patterns bend, stretch and move from one hour to the next.”
The method was validated against ship-based measurements and conventional satellite observations in the Gulf Stream region. Researchers said the results matched existing data while providing significantly more detail on smaller, rapidly changing features such as eddies and boundary layers.
These small-scale processes are important for vertical mixing in the ocean, which had previously only been documented in computer simulations.
“This opens a range of exciting possibilities in physical oceanography that, until now, were largely accessible only through simulations,” Lenain said. “Using GOFlow, we can now measure key signatures of these small, intense currents using real observations rather than relying almost entirely on simulations. This opens the door to testing long-standing ideas about how the ocean takes up heat and carbon.”
Because the method works with existing geostationary satellites it does not require new instruments to be launched into space. Over time, GOFlow could be incorporated directly into weather forecasts and climate models, and may ultimately help improve forecasts by resolving rapidly evolving currents that influence air-sea exchange, marine debris transportation and ocean ecosystems.
Cloud cover remains a limitation, as it can obscure thermal imagery. The team is working to combine additional satellite data sources to improve coverage and expand the system for global use.
The project was supported by grants from the Office of Naval Research, NASA and the European Research Council.
Related news, AAC Clyde Space wins €76.3m contract for EPS-Sterna satellite instruments
