Printer friendly version

Lab studies offer insight into tsunamis

Before the tsunami disaster in the Indian Ocean, Galen Gisler of Thermonuclear Applications (X-2) and his research team of Bob Weaver and Mike Gittings, also of X-2, and former Lab employee Charles Mader studied potential sources of destructive tsunamis. The team's initial studies focused on three causes of tsunamis: 1) landslides, 2) asteroid impacts and 3) submarine volcanic eruptions.

Spawned by the extensive study by Mader of the Lituya Bay Alaska landslide of 1958, the team first evaluated the potential tsunami hazards caused by landslides. This event produced the highest run-up, or peak height of the wave on shore, of any measured tsunami and was successfully modeled in a laboratory by Hermann Fritz in Switzerland. Mader's work with the SAGE code, developed jointly by the Laboratory and Science Applications International, accurately reproduced both the laboratory results and the real event at Lituya Bay.

"This study gave insight into the dangers of landslide-generated tsunamis and can help inform communities as to how to specifically prepare for such events," said Gisler.

Image from a 3-D calculation of an asteroid impact model in 2002, in which the jet from the transient crater is collapsing and producing a wave of a couple kilometers initial amplitude.

The team then conducted a comparative study of tsunamis generated by asteroids of different sizes, velocities and compositions. According to its asteroid impact paper published in the Science of Tsunami Hazards Journal, the team performed a series of two-dimensional and three-dimensional simulations of asteroid impacts into an ocean, using realistic equations of state for the atmosphere, seawater, the oceanic crust and the mantle. "These asteroid-caused tsunamis can have large amplitudes at the source, but will decline faster with distance than landslide-caused tsunamis," Gisler said.

The team also published a paper in Computers in Science and Engineering giving an overview of its asteroid impact work, including the application to the meteor impact that killed off the dinosaurs at the end of the Cretaceous Period 65 million years ago. In addition, it produced a paper for the Caribbean Tsunami Symposium in March 2003 in Puerto Rico, which studied the potential tsunami-generating ability of the Kick-em Jenny submarine volcano in the eastern Caribbean.

A conclusion of this work is that tsunamis generated by point events, such as asteroid impacts and explosive submarine volcanoes, propagate less effectively across great distances than do tsunamis generated by extended events, such as landslides and earthquakes. This was true with the Tsunami from the Indian Ocean event; the rupture was at least 500 kilometers long and it propagated effectively across the Indian Ocean and in to the other oceans of the world.

The team's research is part of Department of Energy's program in Advanced Simulation and Computing, and the code, SAGE, written by Gittings, is part of the Lab's Crestone Project and a joint development between the Lab and Science Applications International Corporation (SAIC).

Regarding his current research activities Gisler said, "We're undertaking two new studies now, one of the potential landslide threat off the side of the Cumbre Vieja volcano in the Canary Islands that is perceived as a danger to the Atlantic, and the other of the Sunda Trench earthquake event that unfolded tragically [in the Indian Ocean in December]."

In a preliminary report given to the International Tsunami Information Center, which is part of National Oceanic and Atmospheric Administration (NOAA), for its press conference today, the team finds that potential landslides from Cumbre Vieja would produce waves that are considerably shorter in wavelength and period than the destructive tsunamis observed in the Indian Ocean and previously in the Pacific. As such, tsunamis from Cumbre Vieja would likely be very dangerous to the other islands in the Canaries as well as to the coasts of Africa and Spain.

The analysis of the potential threat to more distant shores depends on completion of a high-resolution three-dimensional run to be completed when the team obtains high-resolution bathymetry of the Canary Islands region, which has been requested from the Spanish government.

"Naturally, the results of these calculations will be shared with the Spanish government and with all other regions that might be affected," said Gisler.

"The ultimate goal is to protect peoples' lives. The more we understand, the better off we are," he concluded.

-- Hildi T. Kelsey

A 2-D calculation done in 2004 of a very large underwater landslide that produced a wave of 0.5 km amplitude. Images courtesy of Galen Gisler, X-2

Other Headlines