For the first time: Artificial intelligence maps ocean currents from high-resolution satellite images

The GOFLOW system, developed under the leadership of Tel Aviv University, allows for the identification of subtle flow movements in the ocean and may improve models of the climate crisis, storms, and heat and gas exchange between the sea and the atmosphere.

Ocean currents. Illustration: depositphotos.com
ocean currents. Illustration: depositphotos.com

A new study led by Tel Aviv University presents a breakthrough in ocean research: An innovative system called GOFLOW makes it possible for the first time to map ocean flow movements at extremely high resolution, directly from satellite images. The researchers estimate that the new capability will help in a deeper understanding of processes that affect weather, the climate crisis, and the interactions between the sea and the atmosphere.

The study was conducted by Professor Roy Barkan, from the Department of Geophysics at Tel Aviv University, in collaboration with M. Luke Lennon from the Scripps Institution of Oceanography, Kaushik Srinivasan from the University of California, Los Angeles (UCLA), and Nicholas Piso from the University of Rhode Island (URI). The study was published in the journal Nature Geosciences.

The oceans cover more than 70% of the Earth's surface and are a major driver of global climate regulation. Ocean currents, including the Gulf Stream in the Atlantic Ocean, are responsible for transporting heat, carbon, and energy over vast distances. However, to date, many of the small, rapid processes that occur in them—on scales of tens of kilometers or less—have remained beyond the direct measurement capabilities of satellites.

This is where GOFLOW comes in. The system was developed using artificial intelligence and trained on extremely advanced ocean simulations. It analyzes series of infrared photographs of sea surface temperature and deduces the horizontal flow fields of the water from them. Unlike previous methods, which were based on restrictive physical assumptions, the new tool is also able to identify more complex and subtle movements.

The analysis reveals that in regions such as the Gulf Stream, powerful dynamic processes occur on scales smaller than 30 kilometers – processes accompanied by sharp changes in temperature and flow convergence. These phenomena are related to vertical mixing of water, the rise or subsidence of water masses, and the influence on heat and gas exchange between the ocean and the atmosphere.

The researchers note that the significance of the findings has a clear climatic aspect: mixing and energy transfer processes in the oceans affect the intensity of storms, the development of marine heat waves, the dispersion of pollutants, and the ocean's ability to absorb carbon dioxide from the atmosphere. A better understanding of these mechanisms could improve climate prediction models and provide more accurate predictions of extreme events.

Vertical movements of water

Prof. Roy Barkan. Photo: Tel Aviv University
Prof. Roy Barkan. Photo: Tel Aviv University

In addition, the researchers emphasize that the system provides for the first time a direct satellite representation of the horizontal divergence field in the ocean – a key measure for understanding vertical water movements. This is information that has until now been available mainly from computer simulations or point-based measurements at sea.

Professor Roy Barkan concludes: "The ocean is one of the central components of the Earth's climate system, but a large part of the processes that drive it occur on small and rapid scales that are very difficult to measure directly. Through GOFLOW, we are succeeding, for the first time, in extracting dynamic information from satellites that was previously inaccessible - and revealing mechanisms that affect mixing, heat transfer, and gas exchange between the sea and the atmosphere. In an era of accelerating climate change, the ability to see these small details is critical to better understanding the big picture."

More of the topic in Hayadan:

One response

  1. Looks like a blood circulation, so fitting for the Gia hypothesis… mega living organism

Leave a Reply

Email will not be published. Required fields are marked *

This site uses Akismet to filter spam comments. More details about how the information from your response will be processed.