Japan's 32-year-old earthquake-prediction research program has failed to meet its goal of warning the public of impending earthquakes and has overstated the chances of developing accurate forecasts. So says a draft report from a government panel in what is the sharpest official criticism to date of the centerpiece of the country's $147 million a year earthquake research effort.
Summarizing the still-confidential report, panel member Masayuki Kikuchi, a seismologist at the University of Tokyo's Earthquake Research Institute, says the main message is that "trying to predict earthquakes is unreasonable." Instead, the report says, the government should publicly acknowledge that earthquake forecasting is not currently possible and shift the program's focus in a direction that should be set by another committee. The report is not scheduled to be released until after it is reviewed this summer, but a summary of its findings appeared over the weekend in the newspaper Yomiuri Shimbun.
Japan has been in the vanguard of efforts to predict the time, location, and magnitude of impending earthquakes. Although most of these programs, including one in the United States, gradually lost support by the 1980s when researchers could find little or no link between presumed precursors and earthquakes, Japan's prediction program is now midway through its seventh 5-year plan. But the devastating 1995 earthquake that struck Kobe, a lightly monitored area that had not suffered a major quake in almost a century, prompted officials to take a closer look at the overall research program. The report's harsh tone also reflects the rotating membership of the panel, with seismologists more skeptical of prediction replacing several firm supporters.
Despite its strong words, the report must clear several hurdles before it alters the direction of earthquake research. "It's a step in the right direction," says longtime prediction opponent Robert Geller, associate professor of geophysics at the University of Tokyo. "But it remains to be seen how much it really affects actual research."