NEW evidence suggests that the Pacific Northwest may be more susceptible to a major earthquake than once thought.
For years, geologists have known that two colliding plates off the coast could do more than just rattle the cappuccino cups of Seattle residents.
But now new research indicates that the seismic zone could generate a so-called ''great quake'' - one of magnitude 8 or higher. By comparison, the earthquake that hit Kobe, Japan, last January was magnitude 7.2.
The sobering evidence presented in two recent studies could hold implications for building codes and other earthquake readiness efforts in the region.
The focus of concern is two major plates, the Juan de Fuca and the North American, running from northern California to British Colombia.
Because no great earthquake has taken place there since pioneers pushed West and began to settle the area 150 years ago, it has been difficult to pin down how big these quakes could be.
This has lead to a debate among geologists over whether the Cascadia subduction zone, where the Juan de Fuca Plate plunges under the North American Plate, is capable of magnitude 9 earthquakes or whether the plate releases pent up energy in a series of smaller quakes.
The answer is of more than academic interest. The results have a direct bearing on estimating earthquake hazards in the region and designing building codes and emergency preparedness plans, says Robert McCaffery, a geologist at Rensselaer Polytechnic Institute in Troy, N.Y. Based on his own work studying how Cascadia was deformed, and comparing the patterns with other subduction zones around the world, Dr. McCaffery holds that the Cascadia zone is incapable of generating quakes greater than magnitude 8.
''Thirty years ago, when plate tectonics was still in its infancy, there was no basis at all - other than volcanoes - for inferring that major quakes could take place here,'' says Brian Atwater, a geologist with the US Geological Survey in Seattle and a coauthor of one of the two recent studies. But once it became clear that the Cascadia zone was similar to others known to generate big quakes, ''geologists began looking for evidence of big earthquakes.''
They found it in coastal areas where land abruptly dropped and became part of the intertidal zone and where coastal muds were covered with sand though to be deposited by quake-generated tidal waves.
''By the early 1990s, it was pretty clear there had been big subduction-zone earthquakes,'' he explains. ''People began asking, 'When's the next one and how big will it be?'''
Dr. Atwater and colleagues used radiocarbon dating of plants killed when the shore sank to determine the rupture length of the latest quake, though to have occurred about 300 years ago. Because a quake's size increases with the length of the rupture, it then might be possible to use the length to infer a magnitude.
Initial work in 1991 suggested a rupture length of about 55 kilometers (34 miles). Late last month, a team of a dozen researchers, including Atwater, reported in the journal Nature that additional plant samples from a longer stretch of the coast suggest the quake occurred around 1707. The data strongly suggest that at least 900 kilometers between northern California and southern British Columbia ruptured in a quake that affected up to 50,000 square kilometers.
''The whole business went at once,'' he says. ''We gave magnitude 9 a chance to disprove itself, and it didn't.''
Researchers from the United States and Chile, which sits alongside a subduction zone that yielded a magnitude 9.5 earthquake in 1960, recently reported they had come up with an explanation for why some subduction zones generate large quakes while others do not.
The key lies in the angle at which the subducting plate slides under another plate. This is affected by the direction the two plates are moving relative to each other, and the resistance the subducting plate meets from the molten material below the Earth's crust. Their analysis places Cascadia in the same quake-potential class as the subduction zone off Chile.