LONDON -- Japan, one of the world's most active seismic hot spots, spends $100 million a year trying to predict earthquakes, but that will come as little consolation to the people of Kobe.


Tuesday's quake, which hit the port city hardest, is the fourth to strike Japan since late December, all without warning, and recently the transport minister, who oversees the Meteorological Agency, demanded an explanation for the lack of success.


"It's a frontier science still," said Dr. David Booth of the global seismology research group at the British Geological Survey in Edinburgh, Scotland. "We're only at a very early stage despite all the work; we know more about the solar system than we do about the earth."


"We can't predict earthquakes," added Dr. Manfred Baer of the Swiss Seismological Service in Zurich, Switzerland. "Scientists are very much involved in trying, but no real method exists to do this accurately.


"Scientists can give you a statistical value of a 10 to 20 percent chance of an earthquake occurring at a certain time, but it's not like a weather forecast which can say you ought to take your umbrella with you today."


Scientists place monitoring devices close to known faults in the earth's crust, and measure the minutest movements in a bid to establish the preconditions for the buildup to an earthquake.


But scientists say few of the faults have been mapped and it was likely that local seismologists were unaware Kobe was at particular risk from a quake.


"It is possible to get closer to predicting earthquakes if instruments are close to faults, but we don't know where all the faults are, even in areas of constant earth tremor activity like Japan or California," said Dr. Sarada Sarma, reader in engineering seismology at London's Imperial College. The British Geological Survey's Booth said that only two earthquakes have ever been predicted, in Haicheng, China in 1975, and Mexico in 1978, but added the evidence was tenuous.


Booth said other indicators of unusual seismic activity include changes in the level of water tables, wells and spring outflows.


Other evidence included the emission of radioactive radon gas from rocks, which was linked to a pressure buildup. Unusual behavior by animals sensing the escape of gases was a possible clue.


Reuters


LONDON -- Japan, one of the world's most active seismic hot spots, spends $100 million a year trying to predict earthquakes, but that will come as little consolation to the people of Kobe.


Tuesday's quake, which hit the port city hardest, is the fourth to strike Japan since late December, all without warning, and recently the transport minister, who oversees the Meteorological Agency, demanded an explanation for the lack of success.


"It's a frontier science still," said Dr. David Booth of the global seismology research group at the British Geological Survey in Edinburgh, Scotland. "We're only at a very early stage despite all the work; we know more about the solar system than we do about the earth."


"We can't predict earthquakes," added Dr. Manfred Baer of the Swiss Seismological Service in Zurich, Switzerland. "Scientists are very much involved in trying, but no real method exists to do this accurately.


"Scientists can give you a statistical value of a 10 to 20 percent chance of an earthquake occurring at a certain time, but it's not like a weather forecast which can say you ought to take your umbrella with you today."


Scientists place monitoring devices close to known faults in the earth's crust, and measure the minutest movements in a bid to establish the preconditions for the buildup to an earthquake.


But scientists say few of the faults have been mapped and it was likely that local seismologists were unaware Kobe was at particular risk from a quake.


"It is possible to get closer to predicting earthquakes if instruments are close to faults, but we don't know where all the faults are, even in areas of constant earth tremor activity like Japan or California," said Dr. Sarada Sarma, reader in engineering seismology at London's Imperial College. The British Geological Survey's Booth said that only two earthquakes have ever been predicted, in Haicheng, China in 1975, and Mexico in 1978, but added the evidence was tenuous.


Booth said other indicators of unusual seismic activity include changes in the level of water tables, wells and spring outflows.


Other evidence included the emission of radioactive radon gas from rocks, which was linked to a pressure buildup. Unusual behavior by animals sensing the escape of gases was a possible clue.