Skip to Content

Reaching Safe Ground

Some researchers are combining geophysical data with findings from social science in an effort to save more lives during a natural disaster.
July 26, 2005

If a major earthquake happened in the Cascadia Subduction Zone, a tectonic plate margin similar to the one where the devastating earthquake and tsunami occurred last December in the Indian Ocean, people in low-lying towns along the Pacific Northwest coast would have an estimated 30 minutes to evacuate.

Given that dire, but realistic scenario, would it be more effective to broadcast evacuation instructions to residents of, say, Seattle via a fleet of cars that sound an alarm, announcements over the radio, or reverse-911 calling with cell-phone paging?

As geoscientists and engineers throughout the world study the physics of “killer” waves and how to design structures that might withstand them, other researchers are focusing on an even slipperier area: human behavior.

“Social science has to play a major role in hazard preparedness,” says George Crawford, earthquake program manager for the state of Washington’s emergency management division.

Leading this effort at incorporating human factors into the field of disaster mitigation is Toshitaka Katada, a professor of social engineering at Gunma University in Japan. About six years ago, he realized that the flood modeling programs he was working on might also be valuable for reducing the losses from tsunamis. Today, in association with Harry Yeh, a professor of ocean engineering at Oregon State University (OSU), Katada is developing a model called the Tsunami Scenario Simulation.

At its core, this simulator applies classic physical factors involved in a tsunami: hydrodynamics, wave propagation, inundation, impact, and destruction and debris within a geographic area. 

“We have studied tsunamis theoretically and experimentally for a long time,” says another researcher in the field, civil and environmental engineer Tomoyuki Takahashi of Akita University. “We know the characteristics and have the governing equations.”

This complex physical understanding, in part, is due to investigations at places like OSU’s wave laboratory, where researchers run waves through a giant, concrete channel large enough for someone to surf in (see Sources in Notebook). Scientists also carry out mock seismic events in another, equally huge tank that sends waves hurtling toward miniaturized coastlines.

The Tsunami Scenario Simulation takes this traditional physical-world model – and then overlays it with social phenomena that simulate the movement of both information and people through time and space. Modeling these social dynamics allows researchers to test the efficacy of different types of warning transmissions, different evacuation routes, the role of word-of-mouth, demographic details about age and culture, and local information such as the location of schools and nursing homes.

In one such simulation, a series of blue waves sweep toward land from the side of the screen, while blinking lines indicate evacuation routes and color-coded symbols identify people who have drowned, major debris hazards, traffic bottlenecks, and more.

Although the tsunami simulator emphasizes communications, there’s no limit to the layers of data it can simulate. It could include architectural designs that predict if structures will stand or be damaged, what bridges could survive a wave impact, the importance of the time of day, or even the likelihood that people will hear a warning, but not heed it. Furthermore, the model incorporates not only vehicular movements but also foot traffic.

“This is something they’re really focusing on in Japan: getting beyond dependence on automobiles for evacuation,” says Crawford. In geographically similar areas, such as the narrow, low-lying peninsula of Long Beach, Washington, this approach would be critical, as well, since a tsunami there would likely take out major roads and bridges.

In fact, Crawford says officials in Washington, Oregon, and northern California stand to benefit greatly from such a social science model. For instance, it has already helped Crawford and other planners figure out where to locate safe “assembly areas” that evacuees could quickly reach by foot.

“What this computer simulation does is allow us to take inundation maps and models showing wave arrivals, combine them with information about infrastructure, and then take into account the social science as well,” says Crawford. “Running these models gives a fairly close prediction of whether we can evacuate our people in time and what the collateral damage would be.”

The simulations are not perfect, admits co-researcher Yeh: “We still can’t simulate flow of details like the flow of water between just a few houses or by a certain road. We’re also trying to make a casualties model more accurate.”

Further, modeling human behavior – especially during a natural disaster – requires some assumptions that can limit the accuracy of any model. “We hear from the social scientists that on 9/11, there was spontaneous organization of evacuation across the Hudson [River],” says Yeh. “This is remarkable because the assumption is that people would be acting chaotically. Yet it looked as if they had done practice drills.”

In the case of a tsunami, it’s possible some people will still head to the beach out of curiosity, despite knowing about the danger. “How can we predict this stuff? It’s not simple,” says Yeh. “At least these simulations can give some quantifiable ideas about what may happen.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.