A Washington University seismologist has identified the structure of the hypocenter of a January 1995 earthquake in Kobe, Japan, and has discovered that high fluid content and high pressure initiated the massive earthquake that killed more than 6,000 people.
Dapeng Zhao, Ph.D., senior research scientist in earth and planetary sciences in Arts and Sciences, used his own method of seismic tomography -- a computer-aided technique that makes pictures of the Earth from seismic waves -- to show the first high-resolution, three-dimensional images of the earthquake's hypocenter, the point of major impact. This is the first time a seismologist has been able to make such a clear picture of the "heart" of an earthquake to understand the cause.
Zhao's research indicates a high likelihood that an anomalous body of over-pressurized, fluid-filled, fractured rock material at the hypocenter was a major contributor to the Kobe earthquake. Measuring 7.2 on the Richter scale, the earthquake was the most destructive in Japan in more than 70 years.
"This result has important implications," said Zhao, whose research was funded by the National Science Foundation. "Earthquakes like the Kobe one do not occur everywhere, just in anomalous areas. We could probably detect the anomaly before the earthquake by using geophysical methods."
Zhao, in fact, previously had shown anomalies in the Kobe region that weaken the crust and make the region seismically dangerous, compared with the rest of southwestern Japan. He published an article about the seismic tomography of all of Japan in November 1994, two months before the Kobe earthquake.
Zhao published his most recent findings last December in Science with co-authors Hiroo Kanamori, Ph.D., of the California Institute of Technology; Hiroaki Negishi, Ph.D., of Kyoto University in Japan; and Douglas A. Wiens, Ph.D., a Washington University professor of earth and planetary sciences.
The results of Zhao's Kobe analysis offer hope for pinpointing troublesome structures in earthquake-generating faults, giving scientists better clues about earthquake possibilities in a region.
Similar anomalies have been reported in northeastern Japan, where many active volcanoes exist. In earlier work, Zhao found that nearly all of the big earthquakes in this region occurred in areas with slow seismic velocities similar to those found in the Kobe seismic images. He discovered that volcanoes and underground magma chambers caused the weakening of the crust, and seismic velocity becomes slow in such areas, facilitating the occurrence of big earthquakes. But there are no volcanoes in the Kobe area.
In his most recent work, Zhao analyzed the seismic waves of more than 3,200 Kobe aftershocks and 431 local "micro" earthquakes that generated more than 64,000 P- and 49,200 S-wave arrival times. From this database, the seismic tomography revealed a low velocity of seismic waves and a high ratio of the transverse strain and longitudinal strain at the hypocenter. This ratio is called Poisson's ratio, which is the major indicator of high fluid content in the fault. The body of low velocity and high Poisson's ratio is located at the Kobe earthquake's hypocenter, at a depth of 16 to 23 kilometers, extending about 25 kilometers laterally.
Zhao is the first seismologist to obtain high-resolution images of the three physical parameters (P- and S-wave velocities and the Poisson's ratio) at a big earthquake's hypocenter to infer what initiated the earthquake.
Zhao also used his seismic imaging technique in analyses of the 1992 Landers and 1994 Northridge earthquakes in California. He detected anomalies at these earthquakes' hypocenters and found that those anomalies were closely related to the aftershock locations and mechanisms. He recently received a grant from the U.S. Geological Survey for more detailed work on the Landers and Northridge earthquakes.
However, Zhao could use only P-wave velocity structures in the Landers and Northridge areas because only P-wave data were recorded by the California seismic stations. Most of the California stations are equipped with only one-component seismometers that measure only the vertical motion of the ground. The seismic stations in the Kobe region are equipped with three-component seismometers that measure vertical, north-south and east-west motions of the ground, so both P- and S-wave arrival times can be recorded.
Zhao's findings also explain other seismological and geophysical observations in the Kobe region, among them the gaps in aftershock activity, a change in the local stress field, and swarms of "micro" earthquakes that occurred months before the major Kobe earthquake.
"Many people think prediction of an earthquake is impossible, and certainly it is difficult," Zhao said. "However, with more, continuous funding from government and general support from the public, we can certainly do better at reducing earthquake hazards, and we also will make progress toward earthquake prediction."
-- Tony Fitzpatrick
Please send comments and suggestions to: