Why we need to build our own tsunami

Nobody likes tsunamis, but earthquake engineer Tiziana Rossetto at University College London (TEDx Talk: Engineering against tsunami) is hard at work building one of her own. Don’t worry — it’s not going to destroy your coastline. It’s a scale model designed to help Rossetto and other engineers better understand the precise sequence of events that take place during and after a tsunami. That could help them build better coastal defenses and more resilient buildings — and perhaps even tame the terrible toll of the next big one. Here’s why Rossetto’s ideas matter:

A tsunami is more than just a big wave. These monsters can be tens of meters tall and over a thousand kilometers wide, and can reach dozens of kilometers inland. The Tōhoku tsunami in Japan in 2011 killed around 19,000 people and cost an estimated $275 billion; the 2004 Indian Ocean tsunami cost around $14 billion and killed around 250,000 people. It’s hard to imagine, but the energy of a tsunami can move through the ocean at up to 800 kilometers per hour. Unlike wind- or storm-driven waves, tsunamis are formed when an earthquake below the sea pumps the seafloor — and the sea itself — up and down. As a result, the wave has a different shape from the waves on a nice beach: it leads with the trough and follows with the crest, has a much longer wavelength, and carries much more energy.

Although they pack a punch, it’s hard to study tsunamis. Field measurements are still rare and incomplete. As with the earthquakes that cause them, tsunamis happen almost without warning; to get good field data would mean having an ocean-wide network of sensors standing by at all times. So far, there is only a very widely scattered, recently built network. That means we don’t really know how tsunamis work. “There’s uncertainty in the wave profiles,” Rossetto says. “How do they impact?” In other words: Do the waves hit land all at once, over multiple surges — or do they build up to a climax or some even more exotic combination? We just don’t know.

And we don’t really know how tsunamis behave on land. Rossetto visited Sri Lanka after the 2004 Indian Ocean tsunami to learn how buildings there had failed. “I realized that tsunamis are a sort of horizontal mode that pushes buildings sideways, sort of like earthquakes,” she says. “The problem being that we couldn’t identify what the forces were from the water on the buildings.” Unlike with earthquakes, the force of a tsunami on buildings changes as the tsunami moves inland. Rossetto has found that there are at least two modes in which tsunamis behave: in one, the water rushes around buildings and most of the force on structures comes from friction with the water. In the other mode, water piles up against one side of the building faster than it can flow away from the other side of the building, creating a pressure differential that can eventually knock over the building. Tsunamis also carry debris inland from the coast, which can cause more damage. Water can also scour the soil around a building’s foundation, further complicating engineering calculations of a building’s structural integrity.

Actually, the most damaging tsunamis are rare. This is good news, but it also makes them insidious: coastal dwellers may never encounter a tsunami, despite living in the grown-over debris field of the last big one. The US National Oceanographic and Atmospheric Administration (NOAA) has records of 2,500 tsunamis of different sizes over the last two millennia. Yet only a handful of those have occurred since the advent of modern instruments and computers with the ability to model them (here’s a list of the ten biggest tsunamis). Rossetto has had to look to geoscientists, she says. “It’s only because of their work that we know how frequent tsunamis are and how devastating they have been in the past,” even before human record-keeping began. Other factors, such as growing coastal populations and rising sea levels could mean the impact of future tsunamis could be even higher.

How to build your own tsunami. Starting in 2007, Rossetto recruited colleagues from earthquake, civil, and mechanical engineering, and together they built a simulated tsunami chamber. Made from concrete, it was 45 meters long and used air pressure to suck water up and down, much like an earthquake pushes water up and down to create a tsunami. The first such generator to create trough-led waves, it replicated some of the features of the 2004 Indian Ocean tsunami. The results were immediately intriguing: Rossetto found that the distance the scaled-down tsunami reached on the simulated land was different from that predicted by previous non-tsunami-specific wave generators. But there was still a problem, Rossetto says: “The flume was too short.” The first waves they generated were bouncing back from the wall and interfering with the shapes of subsequent waves.

We’re going to need a bigger flume. Now, Rossetto is building the next-generation tsunami generator, in Wallingford, Oxfordshire. This one will be 100 meters long and the team will be able to adjust how it generates its experimental waves. The waves should behave even more like real tsunamis while the extra length gives them room to include multiple sensor-laden models. That means they can examine what happens when the water passes over, not just around, their model buildings. “We can test out different structural arrangements and different coastal defense types,” Rossetto says. So-called benchmark data useful for civil engineering design will be open-sourced so that NOAA and engineers around the world can use the simulations in their own computer models for planning new inundation-zone infrastructure. Rossetto and her colleagues also plan to publish their additional data and analyses in peer-reviewed journals.

How the simple informs the complex. Rossetto is collaborating with social scientists to understand how to get people to better prepare for and conduct evacuations. They are studying and comparing how residents in various tsunami zones communicate risk to one another. Knowing what works best could help them generate recommendations for policymakers and disaster planners around the world. That will matter more as coastal populations grow and baseline sea levels rise. But, says Rossetto, it all goes back to experiments. “It’s really interesting how far very small-scale experiments in labs here at UCL, very simple stuff, can actually go to explain very complex phenomena that we see in real life.”

First published by ideas.ted.com: [html] [pdf].