American scientists from the University of Maryland have developed an artificial intelligence (AI) tool that predicts the formation of unpredictable and dangerous waves known as monster waves. The research was published in the scientific journal magazine Scientific Reports (SciRep).
Monster waves are single waves of 20-30 meters high that form for no apparent reason. They pose a great danger to ships and coastal infrastructure. Such waves have been recorded at least 16 times in the 21st century.
The authors trained a neural network using a dataset of 14 million 30-minute samples of sea surface elevation measurements taken from 172 buoys located off the coast of the United States and the Pacific Islands.
They used their AI to predict monster waves using a separate dataset of 40,000 sea surface height measurements taken from the same buoys.
The scientists found that their instruments could accurately predict the formation of 75% of monster waves one minute in advance, and the formation of 73% of waves five minutes in advance.
The neural network was also able to predict the appearance of giant waves near two buoys that were not included in the datasets used for training, with 75% accuracy one minute in advance.
The researchers suggest that the accuracy and lead time of development forecasts could be improved by incorporating water depth, wind speed and wave position data.
Earlier scientists I learnedwhere waves of devastating monsters are coming from.