Engineering: Tool to predict giant waves up to 5 minutes in advance | Scientific Reports |

A research paper reports on a new tool that can predict the appearance of abnormally large, unpredictable waves called rogue waves at sea up to five minutes in advance.Scientific reportsPublished in The authors believe that this tool could be used to provide advance warning to ships and offshore platforms so that operators can evacuate, conduct emergency shutdowns, maneuver, or minimize the impact of approaching large waves.

The tool, developed by Thomas Brunung and Balakumar Balachandran, consists of a neural network trained to distinguish between ocean waves that follow giant waves and those that do not. The researchers trained their neural network on a dataset of 14 million samples of 30-minute sea-surface height measurements taken from 172 buoys located off the coasts of the Americas and Pacific Islands. They also used another dataset of 40,000 sea-surface height measurements taken from the same buoy to predict the appearance of giant waves.

As a result, the tool was able to accurately predict 75% of superwaves one minute ahead, and 73% of superwaves five minutes ahead. The tool was also able to accurately predict 75% of superwaves one minute ahead near two buoys that were not included in the training dataset. The results show that the tool has the potential to predict large waves in new locations.

The authors suggest that the tool's predictive accuracy and advance warning time could be further improved by incorporating data on water depth, wind speed, and wave location. Further research could help predict the height and timing of giant waves in the future, the authors added.

Breuning, T., Balachandran, B. Predicting strange waves from buoy measurements. science representative 1416048 (2024).

doi:10.1038/s41598-024-66315-3

“Highlights from Nature Magazines” is a translation of a press release prepared by Nature's Public Relations Department. If you require more accurate and detailed information, please be sure to refer to the original paper.

Return to the list of articles “Highlights”.