Published online 19 May 2011 | Nature | doi:10.1038/news.2011.305


Quake risk in Japan remains high

Despite the massive release of seismic stress two months ago, another large earthquake could strike the region.

quake damageThe huge earthquake on 11 March did not release all of the seismic pressure built up along the fault near Japan.Photoshot

The tsunami-spawning, magnitude-9 earthquake that hit Japan on 11 March was the largest tremor to strike the region in hundreds of years, but immense amounts of seismic stress remain stored in the area's tectonic interfaces — and the next large quake could occur much closer to Tokyo, a troubling analysis suggests.

The 'Tohoku-Oki' earthquake in March occurred beneath the ocean east of Honshu, Japan's main island, where the Pacific tectonic plate subducts beneath, or slides under, Japan. Although magnitude-7 or larger quakes strike the same segment of the subduction zone every 30–40 years, the only previous event that might be comparable to that 10 weeks ago happened in July 869 AD, says Mark Simons, a geophysicist at the California Institute of Technology in Pasadena.

In the first of three studies published online today in Science, Simons and his colleagues used global positioning system (GPS) data recorded at more than 1,200 sites in Japan to reconstruct how the Tohoku-Oki quake unfolded1.

During the earthquake the northern half of Honshu shifted eastwards, with some onshore sites nearest to the epicentre moving horizontally more than 4.3 metres and dropping about 66 centimetres. These movements resulted from slippage and deformation of the tectonic plates, which prior to the quake had been flexed and loaded with unreleased stress.

The team's models suggest that the sea floor about 50 kilometres offshore dropped by about 2 metres, and about 100 kilometres offshore, near the epicentre of the quake, it rose by about 9 metres.

In the second study2, Mariko Sato, a geodesist with the Japan Coast Guard in Tokyo, and her colleagues tracked the long-term movements of transmitters installed on the sea floor in the region in 2000–04. Their results confirm the motions deduced by Simons. Measurements taken in the weeks following the quake indicate that one sea-floor site had moved about 24 metres to the east–southeast and risen about 3 metres since the previous measurement in February. Most of these motions were presumed to have occurred during the quake.

The tectonic plates slipped, on average, about 20 metres during the quake, but in some spots deep within Earth's crust they scraped more than 50 metres past each other, Simons and his colleagues calculate.

Deep vibrations

In the third study3, Satoshi Ide, a seismologist at the University of Tokyo, and his colleagues analysed seismic waves spreading from the quake. They suggest that for the first 40 seconds, rupture of the tectonic interface occurred deep beneath the sea floor. Only later did slippage migrate to shallower portions of the subduction zone. The largest amounts of slippage occurred at shallow depths, but most of the energy released in high-frequency vibrations — the type that are most easily felt and do the most damage — apparently radiated from deeper portions of the rupture.

Differences in behaviour between shallow and deep rupture aren't surprising, because the environmental conditions inside the rocks — including temperature, pressure and water content — vary widely according to depth, says Thorne Lay, a seismologist at the University of California, Santa Cruz. "But what's controlling that behaviour is what we don't quite know yet," he notes.

On average, the tectonic plates east of Japan move towards each other at a rate of between 8.0 and 8.5 centimetres per year. When plates are locked together, there's no relative movement along the tectonic interface, sometimes for centuries, but when stress builds to unresistable levels the stored energy is released all at once as a quake.


But even after slippage during the March megaquake has been tallied, approximately 90 metres of convergence since the quake in 869 AD remains unaccounted for. The stress from that amount of slip could have bled off during small quakes or through gradual slippage that didn't generate seismic vibrations, or it may be still stored in the plates, just ready to rip.

Likewise, the section of the subduction zone just south of the Tohoku-Oki rupture may hold large amounts of seismic stress, says Simons. In recorded history, that interface has experienced only one set of quakes larger than magnitude 8 — which means that that region, which lies much closer to Tokyo than the March rupture does, is at risk for its own Tohoku-sized earthquake.

Close monitoring of plate movements in the region should help scientists to determine seismic risk there, Simons and his colleagues suggest. "This area will warrant a lot of attention in the near future," he says. 


If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.

Comments on this thread are vetted after posting.

  • #62227

    I don't think I can ever watch one of those again! Too close to home now.

Commenting is now closed.