Sign In

Communications of the ACM

ACM TechNews

Supercomputer-Developed AI Predicts Tsunami Flooding in Real Time on a PC

View as: Print Mobile App Share:
Wave front of a tsunami.

An artificial intelligence model developed with the help of the world's fastest supercomputer may offer disaster management teams a new tool in preparing for tsunamis.

Credit: ozaiachinn/Depositphotos

Researchers from Japan's Fujitsu, Tohoku University, and the University of Tokyo's Earthquake Research Institute have developed an artificial intelligence (AI) model that can predict tsunami flooding in real time with high spatial resolution.

The AI model was built using the world's fastest supercomputer, Fugaku, which was developed by Riken and Fujitsu.

The researchers leveraged Fugaku's computing power to create 20,000 potential tsunami scenarios to train the AI model.

They used a deep learning algorithm to corroborate the waveform data of offshore tsunamis at the time of an earthquake, with resulting flooding conditions on dry land.

After being trained on Fugaku, the model can be loaded onto a standard PC, fed waveform data, and calculate predictions within seconds.

The researchers' simulations of a large earthquake and tsunami impacting Tokyo Bay matched flood modeling data provided by the Cabinet Office of Japan.

From New Atlas
View Full Article


Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA


No entries found