GOFLOW Turns Satellite Data into Hourly Ocean Currents
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- GOfLOW turns thermal imagery from geostationary weather satellites into hourly ocean surface current maps using deep learning, without needing new hardware.
- Luc Lenain co‑led the study with Kaushik Srinivasan, publishing the findings in Nature Geoscience and receiving funding from the Office of Naval Research, NASA, and the European Research Council.
- GOfLOW was trained on a high‑resolution computer simulation of ocean circulation to recognize temperature‑pattern deformations caused by currents, then applied to GOES‑East images captured every five minutes.
- GOfLOW's output was validated against shipboard velocity measurements in the Gulf Stream and satellite altimetry, showing agreement and revealing finer detail of small, fast‑moving eddies and boundary layers.
- Cloud cover blocks the thermal imagery GOFLOW relies on, limiting its coverage; the researchers intend to incorporate other satellite data to fill gaps.
- Data and code from the study are being made publicly available to enable further research and applications.
Why it matters: Oceanographers, climate modelers, and emergency responders gain real‑time, high‑resolution current data that can refine heat‑and‑carbon transport estimates and improve search‑and‑rescue or spill‑tracking operations, while the method’s cloud‑cover limitation means regions with frequent cloudiness remain under‑observed, and integrating GOFLOW into forecasts could boost weather and climate prediction accuracy.
