These waves occur within the stratified ocean as water moves between layers of differing densities. They can exhibit high vertical velocities, reach amplitudes exceeding 100 meters, and impact anthropogenic structures, sediment resuspension, and ocean mixing, playing a crucial role in climate dynamics. Although they develop beneath the surface, they leave a distinct signature visible in satellite images, making them detectable from space.
In a pioneering effort, the AIR Centre independently developed a machine-learning-based classifier using Sentinel-1 WV mode imagettes, starting from scratch and manually labeling 5,000 images to train a supervised model for detecting internal waves. In addition, the AIR Centre built a comprehensive data processing pipeline and a web platform that enables experts to validate the detected data. This platform also allows the global consortium to share data and information, ensuring collaboration and transparency. After achieving promising results, the AIR Centre launched a
Kaggle competition that attracted international experts, who contributed high-performance classification algorithms. The system has since evolved into a near real-time global service, integrating automation with expert labelling to ensure efficiency and accuracy in internal wave detection.
This workshop will bring together the consortium for the first general meeting of the IWS, where discussions will focus on the existing data processing pipeline, the validation platform, the next development steps, and opportunities to expand the service's applications.
The AIR Centre has partnered with leading experts in internal waves and is assembling a global consortium that includes institutions such as the University of Porto (Portugal), the Interdisciplinary Centre of Marine and Environmental Research (CIIMAR - Portugal), the French Research Institute for Exploitation of the Sea (IFREMER - France), the Nansen Environmental and Remote Sensing Center (NERSC - Norway), the University of Southern Mississippi (USA), Woods Hole Oceanographic Institution (WHOI - USA), the Massachusetts Institute of Technology (MIT - USA), the University of Waterloo (Canada), the National Oceanography Centre (NOC - United Kingdom), Fisheries and Oceans Canada, the European Space Agency (ESA - Italy), NASA's Jet Propulsion Laboratory (NASA-JPL - USA), and NVIDIA RAPIDS (Brazil). Together, we aim to advance the understanding and monitoring of internal waves on a global scale, benefiting local communities, science, and industries.
Research
