Astronomer Yoshio Kushida believes he will receive forewarning should a major earthquake hit.

“I am absolutely confident,” he said when asked about predicting a major quake in the Tokyo metropolitan area.

“(The signs of) all major earthquakes that have occurred could have been observed,” Kushida said in a recent interview at his private observatory in Yamanashi Prefecture.

On the night of Jan. 14, 1995, an FM radio receiver began recording an extraordinary baseline fluctuation at Kushida’s Yatsugatake South Base Observatory.

Kushida didn’t realize it was the day that would totally change his life.

He was recording radio echoes in the very-high-frequency band to observe the passage of meteors through the atmosphere.

Kushida was at a loss that night, believing the machine was malfunctioning.

The thickness of the baseline, usually about 1 mm, was more than 2 mm on the nights of Jan. 15, 16 and 17, leaving Kushida puzzled. But in the early morning of Jan. 17, he realized the hidden message in the signal.

He turned pale when he switched on the TV. Breaking news told him that a major earthquake, measuring a magnitude of 7.2, struck Kobe and southern Hyogo Prefecture at 5:46 a.m. The earthquake led to the loss of more than 6,400 lives, destroyed 248,410 structures and left 446,485 households homeless.

“I was overwhelmed by a sense of self-reproach. If only if I had studied more closely when I found (the baseline phenomenon) two years earlier,” Kushida wrote in the book “Jishin Yoho-ni Idomu” (“The Challenges of Earthquake Prediction”), published Sept. 1.

In 1993, after roughly examining the correlation between abnormal electric waves and earthquakes, Kushida was convinced that some fluctuating patterns in the VHF band appear several days before an earthquake.

But Kushida, who had no expertise on earthquakes at the time, didn’t pay much attention to the data. He thought seismologists probably already knew about the phenomenon, but in the end, that was not the case.

Most seismologists think that accurate prediction of earthquakes is almost impossible, let alone early warning. But Kushida, through carefully observing the phenomena over the past five years, has continued his studies to challenge the common notion about earthquake prediction.

Analyzing radio echoes from a number of FM stations across the country, Kushida believes he has found five basic wave patterns that appear several days before a major earthquake.

From January 1997 to September 1999, using these patterns, Kushida predicted specific dates, strengths and locations of the focus of 36 major earthquakes measuring a magnitude of 5 or stronger.

The average margin of error in the dates predicted was 1.97 days.

As for location, Kushida now claims he can specify the focus of most earthquakes within a radius of 50 km.

“I think (the accuracy) is practical enough,” he said.

Some seismologists have argued that the results could be a coincidence because earthquakes occur very often in the Japanese archipelago.

But the Institute of Physical and Chemical Research, a major quasi-governmental think tank, independently examined the correlation between earthquakes and Kushida’s predictions and concluded the results were not random.

“(The correlation) has much significance,” said Toshiyasu Nagao, director at the institute’s Earthquake Prediction Research Center, which now supports Kushida’s project.

But what is the mechanism that causes this presaging of an earthquake in VHF radio echoes?

Kushida’s system was originally designed to observe meteors by catching radio echoes from a commercial FM radio station.

According to Kushida’s hypothesis, before an earthquake, electric charges accumulate on the Earth’s surface due to the generation of numerous microcracks in magma.

The charge and discharge process of a capacitor formed with the Earth’s surface and the ionosphere changes density of electric plasma in the ionosphere, and the phenomenon is observed by the FM receiver.

Indeed, it has been long known to scientists that some electromagnetic phenomena appear before an earthquake on the Earth’s surface.

In Greece, scientists have conducted studies on the prediction of earthquakes for more than 10 years based on the theory that solid matter emits an electric current just before it breaks down.

The reliability of the method, however, is still a focus of debate by seismologists, although the scientists conducting the test claim the success rate is about 60 percent.

Kushida recalled that seismologists’ response to his method was not good when he first held a press conference and contributed to an article in a spring 1995 physics magazine.

After five years of studies, many seismologists — many of whom have little knowledge of the ionosphere or electromagnetism — remain skeptical, or simply ignore Kushida’s achievements.

Kushida now only publicizes his analysis and predictions to people who have subscribed to his fax service, believing open publication of his predictions would only cause confusion or panic.

“What would you do about nuclear power plants, or railway service if a major earthquake is forecast to hit? You may want to stop them, but there is no legal basis (to support such actions). There is nothing I can do,” Kushida said.

Kushida said he cannot take responsibility for possible results of his predictions, as they still contain a margin of error.

Much more public understanding, legislation for early warning systems and more efforts to improve accuracy will be necessary before advance publication of earthquake information will be possible, he said.

But interested parties can subscribe to Kushida’s fax service if they sign an oath not to leak the information to other people or use it for secondary purposes.

In a time of both misinformation and too much information, quality journalism is more crucial than ever.
By subscribing, you can help us get the story right.