Scientists at the University of Washington have been analyzing data since the Feb. 28 Nisqually earthquake that shook both structures and nerves in the Puget Sound region, and have some conclusions to present this week at a national conference. They also have a mystery or two.
Derek Booth, research associate professor in the UW Department of Civil and Environmental Engineering, tracked the temblor’s effects by surveying damaged chimneys and found a clear pattern of damage along the Seattle Fault. The fault didn’t move during the quake but, according to Booth, appears to have focused energy – and shaking intensity – along the fault line.
However, he also found pockets of damage outside the fault line, an effect that so far remains unexplained. Meanwhile, colleague Kathy Troost, a research scientist in the university’s Department of Earth and Space Sciences, discovered patterns of ground failure that correspond with spots that have been heavily altered by human activities.
Both will present their findings this week at the national convention of the Seismological Society of America in San Francisco. Booth said his interest in the chimneys was piqued by the apparent concentration of damage in West Seattle – something the 6.8-magnitude Nisqually earthquake had in common with a 1965 temblor that shook the region. Booth and a team of collaborators from the U.S. Geological Survey meticulously catalogued and categorized 60,000 chimneys in a broad west-to-east swath from neighborhoods in Bremerton to Mercer Island – all well north of the deep earthquake’s epicenter. What chimneys lack in precision as earthquake monitors they make up for in sheer numbers.
“It’s a broadly distributed seismological array,” Booth said. One clear zone of damage lay along the Seattle fault, Booth said, indicating that the fault somehow affected the seismic waves. “The energy didn’t just pass through as if nothing happened,” he said. “When vibrations come into a zone where material properties change they tend to be both refracted and reflected.” In other words, scientists suspect that the fault may have focused and amplified the shaking, resulting in greater damage on the land above. If this pattern is borne out, it suggests the location of a zone of chronically greater earthquake damage for this, and any subsequent, deep earthquakes.
The mystery is what caused patterns of damage in several other spots that don’t lie along the fault. “That’s something we just don’t know yet,” Booth said. Immediately after the February quake, Troost went out looking for signs of liquefaction. Liquefaction occurs when shaking from an earthquake turns loose, wet sand into a slurry, causing the mixture to behave like a liquid. Liquefied soil may then escape from the ground in the form of “sand blows.” The condition often causes structures to fail during earthquakes due to the loss of soil strength.
To her surprise, Troost found multiple sites of ground failure induced by liquefaction in the Seattle area. The major zones were on Harbor Island, in the SODO or South Downtown district of Seattle, the north end of Boeing Field and near Green Lake. “The surprise was finding liquefaction at all,” Troost said, given the distance and depth of the earthquake and the fact that Tacoma had very little liquefaction.
As she analyzed the data, she realized that the liquefaction sites on Harbor Island and in the SODO district occurred on tidal flats that had been filled, then built upon. The area around Green Lake also includes some fill over sandy soil. And, in the early 1900s, workers altered the course of the Duwamish River to straighten it out. Boeing Field was built along the old river valley, which had been filled.
“The places we’re seeing the failures are places where we’ve done a lot of alteration,” Troost said. “It’s a given that the geologic conditions make the area vulnerable to start with. But it looks like we’ve enhanced the susceptibility by extensive filling with uniform sand.”
Materials provided by University Of Washington. Note: Content may be edited for style and length.
Cite This Page: