A Japanese scientist has discovered a way to forecast earthquakes, but despite the accuracy of some of his predictions, the scientific world still remains skeptical, writes David McNeill
Yoshio Kushida may have made an earth-shaking discovery. But the experts are not listening. Two months ago, a message was posted on the Internet bulletin board of the Tokyo Foreign Correspondents' Club warning of rumours that a devastating earthquake would hit the city on Tuesday, September 16th or the day after. The writer had heard the story from a wealthy friend who had fled to Hawaii with his family for two weeks. "What should I do?" he wrote.
If he is like most of us who have fallen victim to dozens of similar rumours over the years, he probably stocked a couple of extra bottles of mineral water and carried on as before. By the middle of that week, 60 per cent of the city's population had heard that the quake was on the way, according to a Tokyo television station, yet there was no sudden run on plane tickets, no noticeable last-minute buying and certainly no hysteria.
Pushing the panic button in a city that has cried wolf as often as Tokyo is not easy, and when Thursday passed off in seismic peace, the joke was on the few lily-livered toffs who had headed for the airport.
Then came Saturday, September 20th. At 12.55pm, nearly 30 million people in the world's largest metropolis felt the familiar shaking of a quake measuring 5.5 on the Richter scale, and you can be sure that millions of them, for a few queasy seconds, wondered if the prediction had been right.
It was a big one, not the Big One, but six days later, the world's biggest quake in over two years struck Hokkaido in northern Japan, injuring hundreds and causing widespread power cuts. Both earthquakes came slightly later than predicted, and the Hokkaido trembler was a few hundred kilometres from Tokyo, but they have still left many people here asking: Where did the rumour come from? The answer lies in an observatory in the mountains of Yamanashi, about 90 minutes west of Tokyo, where Yoshio Kushida, his wife Reiki and a handful of young researchers work with a bank of machines that monitor and record VHF radio waves. Since stumbling on the possibility that these waves may well hold the key to accurate earthquake forecasting a decade ago, the Kushidas have been sending their warnings to a small group of supporters.
"Every time I get data that says there will be a big quake, I struggle with what to do," says Kushida. "If I keep it to myself, thousands may die. If I release it and I'm wrong, people will lose faith in me. It's so difficult."
The 46-year-old self-taught scientist says his grandfather taught him about the horrific 1923 earthquake that killed more than 100,000 people in Tokyo and Yokohama and reduced much of the area to rubble and ash. "He told me you could see the fires glowing for miles. I was so scared growing up there. It's the most dangerous city in the world. They should never have made it the capital."
The Kushidas both started out as astronomers in Tokyo before deciding to abandon the city and build the Yamanashi observatory. One night in August 1993, while recording VHF radio echoes to monitor the passing of meteors, Kushida's machines recorded what he calls "an extraordinary baseline fluctuation." Convinced his equipment was faulty, he ignored the data but remembered the anomaly a few days later when he heard about a large quake in Hokkaido.
"I assumed that the correlation between VHF waves and earthquakes was known, so I thought little of it." The same thing happened in mid-January 1995 and again he ignored the data. Three days later on the morning of January 17th, he sat horrified, watching television footage of fires raging after the Kobe earthquake, which eventually took 6,400 souls and left close to half a million people homeless. "It shocked me to my core," says Kushida. "I just sat watching for days thinking, if only I'd taken my work seriously maybe those people would still be alive."
His first reaction was to call a press conference to announce his findings, "but the journalists talked to earthquake scientists who dismissed my work out of hand." The next time he called the press about a quake in Sakhalin in May 1995, he says he was "laughed at". It was Kushida's baptism of fire into the world of the seismologists who make their living studying the earth's fault zones and only reluctantly share the research stage with thousands of others they consider amateurs and cranks.
Can earthquakes be predicted? The mere question invites reams of well-researched and argued derision from the scientists. Thousands of quakes rattle around the world each year, including about six to 18 with a magnitude of 7.0 or larger, some of them in the wealthiest, most technologically advanced societies on the planet. Uncountable sums of money have been spent over the last century, studying rocks, ground temperature, ground water levels, sunspots, the moon, the tides, the behaviour of dogs, catfish, the minds of psychicists, anything that might give some warning, all without success. The definitive in-house judgment was delivered in a 1997 paper for Science magazine by the superstars in their field, Robert J. Geller, David D. Jackson, Yan Kagan and Francesco Mulargia, entitled, "Earthquakes Cannot Be Predicted." "I don't predict," says Kushida. "I forecast. My research is new and incomplete and the mechanism is not clearly understood, but the data clearly shows that earthquakes give warnings before they happen, and I have to let people know."
The "mechanism" is based on the premise that the electron density in the upper atmosphere changes before a quake, a phenomenon that can be picked up by FM radio waves. "Most of the scientists who criticise my work have never come here or read anything I've written. They're hostile to new research. We operate in parallel universes." Professor Keiji Doi of the University of Tokyo's Earthquake Research Institute is happy to agree.
"There is some relationship between earthquakes and VHF waves, but they only happen simultaneously. There's no proof that radio waves can be used to forecast quakes so I'm quite suspicious of Kushida's methods." Kushida's reaction to the hostility of the scientists and the derision of the media was to abandon his work on comets (he had discovered two, and his wife was Japan's leading supernova researcher) and borrow 10 million yen to buy new equipment. "I had to collate a lot more data to prove this worked. At first I only looked for support from wealthy people, but then I thought it wasn't fair to only help the rich, so I asked for small donations of 5000 yen a month. We started with four people and now we have 1,000".
Then the two astronomers, who had "fallen in love because we both loved to look at the stars," says Kushida, shifted their scientific gaze earthwards.
It was a momentous decision and one that has cost them dearly in time and health.
Kushida says he has worked 16-hour days without a break since he took up this research. The stress of the job is evidenced by the graying shown in photos of him on the wall of his office.
Over the years he has forecast 36 major quakes. In August this year, his stress levels soared when his machines started telling him a big earthquake was imminent in Tokyo, likely to result in 160,000 injuries or deaths, according to government estimates. "I’d never seen anything like it," he says, pointing to his data. "The baseline was very thick, indicating a strong quake, but in mid-August it suddenly disappeared, which has never happened before. I thought it could mean a silent earthquake, or possibly volcanic activity or maybe even a double earthquake but I couldn’t be sure."
He decided to put all the data on his Web page (http://www.yatsugatake-eorc.org/ – Japanese only) and write an article for one of Japan's top-selling magazines. The earthquake forecaster had really stuck his neck out. "I said Tuesday or Wednesday 16th or 17th plus or minus two days either way, so I was pretty close, but I was wrong about the magnitude," admits Kushida. "Still, I've learned a lot from this."
The latest forecast has given him a bigger audience and he now gets 100 e-mails from abroad a day, but the skeptics remain unconvinced. "It’s coincidence," says Tokyo University’s Professor Doi. "There are quakes all the time in and around Tokyo." Such sentiment saddens Kushida. "Science should be for everyone, it should be a useful living thing, not buried away in academic journals. I’d like us all to cooperate, but even if they don’t it doesn’t matter. This is the future. I’m very positive." And his advice to people who live in Tokyo? "I wouldn’t live there, that’s for sure. It’s only a matter of time before something terrible happens."
