Japan to predict tsunamis with AI

By Medha Basu

Kawasaki city will predict tsunamis and their impact to improve disaster preparation.

The coastal city of Kawasaki in Japan is looking to predict tsunamis and their impact using artificial intelligence.
It will study ways to predict the “profile” of tsunamis, such as the waves’ height and their arrival times, and simulate how tsunamis could flood coastal areas.

It will combine this data with a simulation of residents’ evacuation routes and behaviour, to find ways to better prepare for disasters and mitigate their impact.

Kawasaki city will also use simulations of different possible types of tsunamis to better understand their complex behaviour.



In 2011, a 9.0 magnitude earthquake created massive tsunami waves reaching 40.5 metres and travelling 10 kilometres inland. The disaster left 15,894 dead, displaced hundred of thousands and caused a nuclear meltdown. The World Bank estimated the damage costs to be US$235 billion.

Kawasaki sits next to a major fault line underneath the Nankai Trough, where historically earthquakes have produced damaging tsunamis. Another such tsunami could reach certain parts of Japan within minutes, and could take over an hour to hit Kawasaki and parts of nearby Tokyo Bay.

The city is working with The Earthquake Research Institute at the University of Tokyo, the International Research Institute of Disaster Science (IRIDeS) at Tohoku University, and Fujitsu to study how AI can be used to prepare for such devastating tsunamis.

After the study in Kawasaki is completed this year, it will be expanded to the nearby Nankai Trough coastal region.