Engineers Use Artificial Intelligence To Capture the Complexity of Breaking Waves

• Rating

• predicted waves
• ocean wave
• predict wave
• knowing waves
• surfer break
• wave behaviour

Once they reach a critical height, the waves break and crash into a shower of droplets and bubbles. These waves can be as large as the surfer's break point and as small as a gentle wave skimming over the shore. For decades, the dynamics of how and when a wave breaks has been too complicated for scientists to predict.

Now, MIT engineers have discovered a new method for modeling how waves break. The researchers transformed equations that had previously been used to predict wave behavior using machine learning data and wave tank tests. Engineers often use such equations to help design robust offshore platforms and structures. But so far, the equations have failed to capture the complexity of breaking waves.

The researchers found that the modified model predicted how and when the waves would break more accurately. For example, the model evaluated the stability of a wave shortly before it breaks, as well as its energy and frequency after it breaks, with greater precision than conventional wave equations.

Their results, recently published in the journal Nature Communications, will help scientists better understand how a breaking wave affects the water around it. Knowing how these waves interact can help improve the design of offshore structures. It could also improve predictions of how the ocean interacts with the atmosphere. Better estimates of how waves break could help scientists estimate how much carbon dioxide and other atmospheric gases the ocean can absorb.

"Wave breaking is what pushes air out into the ocean," says study author Themis Sapsis, an associate professor of mechanical and ocean engineering and an associate at MIT's Institute for Data, Systems, and Society. "It may sound like an understatement, but if you multiply its effect by the area of ​​the entire ocean, then wave breaking starts to become fundamentally important for climate prediction."

Study co-authors include lead authors and MIT postdocs Debbie Eultink, Hubert Branger and Christopher Luneau of the University of Aix-Marseilles, Amin Chabachub of Kyoto University, Jerome Kasparian of the University of Geneva, and T.S. Van den Bremer of Delft University of Technology.