By Tan Ee Lyn
HONG KONG (Reuters) - Japan's battered northeastern coast suffered many large tsunamis in the past and nuclear power stations there should have been built to withstand these huge walls of water, a scientist said on Thursday.
In a commentary in the journal Nature, geophysics professor Robert Geller singled out two tsunamis -- the 38-meter Sanriku tsunami of 1896 that killed 22,000, and the Jogan tsunami of 869 that was comparable in size to the March 11 disaster -- which pummeled the very same Tohoku region in the northeast.
"There were very many documented large tsunamis in that area but the point is ... even one would have been enough to warrant precaution in designing nuclear power plants," Geller, of the Graduate School of Science at the University of Tokyo, said in a telephone interview.
Tokyo Electric Power's troubled Fukushima Daiichi nuclear plant, which has been spewing radioactive substances for more than a month, was designed to handle water of only up to 6 meters (20 feet) -- way below not only the 14-meter tsunami of March 11, but these other documented giants of the past.
"It's known to have happened before and it's well documented and so when they built a nuclear power plant they should have provided for a tsunami of the same size," Geller said.
"But they planned for a tsunami to be 5.6 meters and the actual tsunami was about 15 meters, so it completely overwhelmed the defenses."
Geller said all nuclear plants in Japan should be reviewed to ensure they are tsunami-resistant.
"They definitely should, they are almost all on the oceanfront. You need a lot of water to cool them," he said.
IMPOSSIBLE TO PREDICT
In his commentary, Geller said the equipment and the science that seismologists now have at their disposal were insufficient to make reliable short-term quake predictions.
"Theoretically we try to make a prediction ... that an earthquake is going to happen in a day or two. In my opinion that system is not scientifically sound and should be terminated," Geller said.
"It's a futile effort. Why should we be pretending to do something that we cannot do?"
In his article, Geller refers to Japan's elaborate plan to predict the next major quake in the central Tokai region, which suffered major earthquakes in 1498, 1605, 1707 and 1854.
The country fears that such a disaster will kill thousands of people, damage millions of buildings and concerns have also arisen over the Hamaoka nuclear power plant in the area.
But just as the March 11 earthquake could not have been predicted, neither could the next Tokai quake, Geller said.
"None are predictable ... it's like you bend a pencil and you bend it a little more and at some point it snaps. But you can't say when that will happen even though it's right there in our hands," he said.
"It is time to tell the public frankly that earthquakes cannot be predicted ... all of Japan is at risk from earthquakes and the present state of seismological science does not allow us to reliably differentiate the risk level in particular geographic areas," he wrote.
"We should instead tell the public and the government to 'prepare for the unexpected' and do our best to communicate both what we know and what we do not know."
(Reporting by Tan Ee Lyn; Editing by Alex Richardson)