Seismologist Probes Japan Quake

As part of the NOVA team filming Japan's Killer Quake, I arrived in Japan two days after the March 11 earthquake and tsunami. I flew by helicopter over coastlines damaged by the event, consulted with colleagues at Japan's Earthquake Research Institute, and generally tried to gain a preliminary understanding of the nature and impact of the devastating 9.0 earthquake.

My first reactions on arrival in Tokyo were of admiration for the seismologists and earthquake engineering community of Japan. There was almost zero damage to the buildings of Tokyo, and, in fact, little shaking damage throughout much of Honshu, the largest island of Japan. This surely must be considered a success story given the enormity of the earthquake. It was the tsunami that did most of the damage, which I'll touch on below.

Here, illustrated with diagrams I have put together, are some initial thoughts, together with a brief discussion of potential dangers awaiting Japan as well as the U.S.:

Foreshocks and aftershocks


As this graph shows, several foreshocks (in red) occurred near the epicenter of the mainshock, but, regrettably, these were not recognized at the time as precursory to the mainshock sequence. As if the monster mainshock of March 11 were not enough of a scare, the Japanese people have endured numerous aftershocks, some of them larger than the Haiti earthquake. They continue to be felt now, two weeks after the mainshock.
Not just quake--a megaquake

As the graph above shows, the March 11 Honshu earthquake, at magnitude 9.0, is one of five "megaquakes" in the world to have exceeded magnitude 8.4 since 2004. These recent five megaquakes were preceded by a four-decade gap that followed a cluster of other megaquakes between 1950 and 1964. No significance to the gap has been established.

On the graph, the largest of those recent megaquakes is marked as "Indonesia." This was the 2004 quake off Sumatra that triggered the Indian Ocean tsunami that killed over 200,000 people. The 2010 Haiti quake, which possibly killed even more people, is not on the graph because it was "only" a magnitude 7.0 (one-thousandth the energy of the 9.0 Japan quake). Its greater damage and loss of life occurred because the quake happened near the country's crowded capital, which sadly had few earthquake-proofed buildings.

Cross-section of the quake

The diagram above is a west-to-east section through the Honshu coastline showing the location of the main rupture, ending close to the shore at a depth of 18 to 25 miles. The quake occurred when slowly accumulating strain was released where the Pacific tectonic plate collides with a thin sliver of the North American plate, backed by the Eurasian plate.

The main rupture caused an undersea uplift of 10 to 15 feet, as inferred from GPS signals on land. This seafloor uplift pushed the sea overhead upwards, generating the tsunami. The quake also resulted in subsidence of nearby coastline of over three feet. This lowering of the pre-earthquake shoreline greatly exacerbated the effects of the tsunami, which raced ashore with heights in places of as high as 45 feet.

Nuclear crisis

Most significantly, the land's subsidence and the tsunami's size combined to overwhelm the defenses of the Fukushima Daiichi nuclear plant (above, photographed the day after the quake). The plant was protected by a 16-foot-high tsunami barrier, but the wave that came ashore there apparently exceeded 30 feet. The wave flooded the plant's generators and electrical wiring in the basement and lower levels, triggering the crisis that is still ongoing. Nuclear power plants are simply not designed to be immersed in seawater.

In hindsight it appears impossible to believe that nuclear power stations were located on a shoreline without recognizing the engineering difficulties attending prolonged immersion by a large tsunami. In 1896 a tsunami over 100 feet high drowned the Sanriku coastline 125 miles to the north of Fukushima. A 75-foot wave surged onto the same coast in 1933, and as recently as 1993 a 100-foot wave swept over Okushiri Island.

Global impact
The March 11 earthquake also brought about lateral shifts in coastal land. The city of Sendai, in one of the hardest-hit areas, initially moved up to 15 feet to the east before springing back to the west about a yard.

The quake actually shifted the entire island of Honshu, the east more than the west. Data from more than 1,000 GPS units that operated every five seconds during the quake revealed that, over a period of about three minutes, Honshu expanded to the west by up to 12 feet, adding new land equal to 150 soccer fields. More precisely, this "new" land area had been lost to the imperceptibly slow squeezing action of the west-moving Pacific plate over the previous few hundred years, and the earthquake enabled it to spring back to its circa 18th-century shape.

The March 11 quake even influenced the planet as a whole. Large subduction-zone earthquakes such as the Honshu tremblor alter the geometry of the Earth. The seafloor uplift during the March 11 quake raised global sea levels by 1/100th of an inch. In addition, as Richard Gross of NASA's Jet Propulsion Laboratory calculated, the quake shortened of the length of a day on Earth by 1.8 microseconds. The seismic waves from the earthquake set the Earth ringing like a bell at its fundamental period of around 60 minutes, and even now, two weeks after the earthquake, sensitive seismometers are recording the Earth's slow vibrations.

Multiple segments at once


The great earthquakes that accompany the incremental slip of plates into the Earth's mantle occur on segments that may rupture singly or in combination. In the past several hundred years, relatively small segments have slipped sequentially in damaging earthquakes along the Japan Trench (see map above).

In the March 11 quake, however, several segments slipped simultaneously, something that may not have occurred here for a thousand years. An earthquake on July 13, 869 A.D. appears to have resembled the 2011 event in that its tsunami also flooded the coastal plains near Sendai.

The figure above, adapted from the Japanese seismologist Kiyoo Mogi's 1985 book Earthquake Prediction, illustrates historical rupture zones prior to 1982 and, in violet, the 300-by-100-square-mile segment of the plate boundary that slipped between 6 and 60 feet on March 11 and in aftershocks.

What makes a big rupture area team up with a neighboring big rupture area to merge and become a megaquake? This is one of at least three missing pieces of information seismologists would ideally like to know before the next megaquake. The other two are: Where and to what degree is strain distributed across various segments of a fault zone like the Japan Trench? And what is the trigger for a big earthquake--a reduction in friction on the fault plane, or an increase in stress to overcome constant friction? Much research will need to be done before these questions can be answered.

Influence on possible Tokai earthquake
For several decades Japanese scientists have been expecting a large earthquake to shake Tokyo from the south. Such a quake would be the next iteration of the so-called Tokai earthquake, which recurred previously in 1498, 1605, 1707, and 1854. Seismologists have estimated it would be in the neighborhood of magnitude 8.4.

Aftershocks from the March 11 quake have migrated south, some within 30 miles of downtown Tokyo, and even to the south, close to the rupture zone of the anticipated Tokai earthquake. The possibility that the three Nankai Trough segments southwest of Tokyo could fail in a single earthquake hitherto has been considered unlikely. But after the March 11 quake, that view, too, will need reconsideration. Should a single earthquake prevail, its magnitude could match the recent megaquake, a scenario that would be accompanied not only by additional shaking for Tokyo and nearby cities, but also by a tsunami of large amplitude approaching from the south.

What about the U.S.?
Earthquake scientists have long known that a quake about the size of the March 11 quake will strike off the coast of the Pacific Northwest. Along the seafloor boundary where the Juan de Fuca plate plunges beneath the North American plate, magnitude-9 earthquakes occur about every 300 to 600 years. The last one was 300 years ago, so it may not happen for another century or two--or it could happen tomorrow.

Japan's devastating quake of March 11 serves as the most sobering of reminders about the effects Americans might anticipate from a magnitude-9 quake all at once shaking the western edge of northernmost California, Oregon, and Washington State--with a large tsunami slamming into the coast within a half an hour. We have a lot to learn from Japan's disaster to better prepare for our own.

Roger Bilham will appear in Japan's Killer Quake, which airs on most PBS stations on March 30 at 9 p.m. Check local listings for exact times in your area.

Image credits: (all graphs) Courtesy Roger Bilham; (photo of Fukushima plant) © STRINGER/epa/Corbis

User Comments:

Enjoyed this article. Lots of details and info not reported in the news. Thanks.

NOVA produces high quality material.

Darn, I missed your show on the Japanese Earthquake that was on PBS this evening. I caught the tail end of the show. I would have love to have seen the whole epidsole but I don't have money to buy your tape. I would like to say I did enjoy reading your post about the quakes. I found the post to be very, very interesting and informing. Thank you for sharing with us in Cyberspace.

The question I have is about the possibility of using bombs to blow up the tsunami's path-way. Why haven't we tried to re-create tsunami's and veiw the path's that the tsunami takes so that at least a little bomb could be used in the most (proposed based on evidence) devastating area's of the tsunami. Perhaps we could use force as the proper equation to verify what the evidence tells us.

(If you have some info on this already being done let me know! I am interested in hearing back either way!:))

Transcribed: Mrs. Brandon Allen

Hi Sherry,
Thank you for your comment! You can watch the show online by following this link:

Thank you nova that was an excellent show you
did on march 30th the japan killer quake.
I always like watching those kind of documenteries very fascinated and the "oh wooow" whisper. Or "holi cooow" whisper as well. Thank you very much Nova : )
PS; If I had the money to Save pbs from going 3 feet deep under water I would help.

As a expat living in Tokyo and present for this megaquake, I would like to know more about the possibility of another quake occuring in the Phillipine Sea Plate below Tokyo in the near future. You mentioned it very quickly towards the end of the program
but with very little explanation. Otherwise , it was an excellent and informative piece on why quakes and tsunami occurred.

Hi. As usual, another great show.

I heard it mentioned that the earth was knocked off its axis after the quake in Japan. Does anyone have more information on this?

Regards, Peter D

Great coverage... The Earth's magnetic field and seismic activity, is there a link? Did magnetic north move before, during or after the quake? Do sections of the Earth's crust move when magnetic north changes?

I know that certain rocks under pressure emit a magnetic charge. I also know that bore sample tests in Germany determined magnetic effects on rocks under pressure to be random and inconsistent. Despite that wouldn't you be able to tune a magnetic resonance device of some sort to monitor large scale deep magnetic activity for possible build ups or drop offs as a sign of an impending earthquake? Even if they data was inconclusive at first with a high enough resolution I'm positive a pattern could be distinguished.

blog comments powered by Disqus

Roger Bilham

NOVApbs Twitter Feed

    Other posts by this Contributor