WASHINGTON Philip Liu of Cornell University in New York and several of his colleagues rushed to Sri Lanka after the Dec. 26 Indian Ocean tsunami. Liu's team had developed a computer model to predict how a quake might generate a tsunami and wanted to test it.
WASHINGTON Philip Liu of Cornell University in New York and several of his colleagues rushed to Sri Lanka after the Dec. 26 Indian Ocean tsunami.
Liu's team had developed a computer model to predict how a quake might generate a tsunami and wanted to test it.
Such a system might be plugged in to an early warning network to help determine which quakes are likely to cause a tsunami, Liu said.
The magnitude 9 earthquake off the coast of the Indonesian island of Sumatra lifted the sea floor 15 feet and displaced trillions of gallons of water, inundating coastlines thousands of miles away.
More than 300,000 people were killed or have disappeared. There were virtually no warnings, although in some instances it took hours for the giant waves to arrive.
Liu's team wants to develop a quick way to measure a quake and plug the data into the model to predict whether and where a tsunami may hit in time to warn people.
In the case of the Sumatra quake, the model showed it displacing a column of water both up and down.
It should, according to the model, have sent a trough first and then a wave in one direction, and a giant wave followed by a trough and then more waves in another.
And that is what happened, Liu told a news conference organized by the Smithsonian Institution's Smithsonian magazine.
While a "negative" wave went to Thailand, causing waters to recede before the wave hit, a "positive" wave hit parts of India and Sri Lanka, giving no warning before the waters rose.
Furthermore, the model showed the waves wrapping around the island of Sri Lanka, which also happened in December.
"You would think the (Colombo) region was protected," Liu said, pointing to Sri Lanka's capital on the west coast of the island, away from Indonesia. "But Colombo actually was affected."
Looking Into the Past
Tsunamis are rare events and while Liu looked for eyewitness verification of his prediction methods, other scientists said they are forced sometimes to literally dig for clues.
Jody Bourgeois and colleagues at the University of Washington examined sediments left in Chile by a tsunami generated by a magnitude 9.5 quake in May 1960.
They visited the village of Mehuin, at the mouth of Rio Lingue in southern Chile.
"People here knew to go to high ground," she said. Only 40 were killed, although the tsunami turned what had been a village and farmers' fields into a wide place in the river.
The villagers and farmers showed her where fields had been, and her team saw clear evidence of heavy deposits of dirt and sand washed over the area by the tsunami.
Back home, they saw similar patterns in Washington's Puget Sound and elsewhere. The Juan de Fuca plate off the Washington coast has a subduction zone, where one underwater plate slides under another, similar to the Sumatran zone.
As in Indonesia's Aceh province, the tsunami from such a quake would arrive in just minutes. Earthquakes are not the only threat.
James Luhr, a volcanologist at the Smithsonian's National Museum of Natural History, said Mount St. Helens, which erupted and collapsed in 1980, dumped 2.6 cubic km (92,000 square feet) of ash, mud and rock on the surrounding area.
California's Mt. Shasta deposited 46 cubic km (11 cubic miles) of material, changing the landscape in a way that can still be seen clearly 350,000 years later.
U.S. Geological Survey sonographic scans show extensive areas around old underwater volcanoes near Hawaii and the Canary Islands off Africa that make Mt. Shasta's deposits look tiny.
Some contain up to 500 cubic km (120 cubic miles) of debris. Luckily, Luhr said, they only seem to occur every 125,000 years or so.