When it comes to natural disasters, the Japanese government is good with numbers. It expects, for instance, a magnitude-8.1 quake to strike in the next 30 years with an epicentre in the Nankai trough — a depression in the sea-floor 100 kilometres off the country's east coast. And when it hits, it is likely to kill between 12,000 and 18,000 people.

Credit: JAMSTEC

The Nankai trough lies in a subduction zone, a perilous region in which one tectonic plate dives under another, building up the sort of rock strain that can unleash the world's most powerful earthquakes. All earthquakes with a magnitude of greater than 9 have occurred in these zones. And although the next earthquake at Nankai is not expected to be quite this big, the region could prove key in understanding why earthquakes in subduction zones release such vast amounts of energy.

On 21 September, a brand new research ship is due to depart from the city of Shingu in Japan on the first leg of a five-year project. The iniative, called the Nankai Trough Seismogenic Zone Experiment (NanTroSEIZE), is the latest and most ambitious of a series of deep-drilling research projects that stretch back decades (see 'Staying afloat'). Everything about the ship, named Chikyu for 'Earth', is large: its 210-metre length, its 10 kilometres of drill string and its ¥60-billion (US$526-million) price tag. Chikyu is the first research ship to use a massive pipe known as a 'riser' to encompass the main drill pipe — making the rig more stable and enabling it to drill more than three times deeper than any other scientific drill ship1.

Having an exposed system to observe is like being able to examine a live squid rather than a dried one. Asahiko Taira

But once Chikyu gets down to its ultimate goal — an earthquake-generating zone some six kilometres below the sea-floor — its work will become very small-scale. Scientists onboard the vessel will be looking at minute changes in the porosity and other characteristics of the rocks drilled from the depths. Eventually, they will place instruments in a deep borehole that will gather data over several years. The goal is to monitor the build up of strain in the rocks — to see an earthquake in the making.

Down under

Previous attempts have been made to monitor earthquake zones — for instance, at the Parkfield site in California atop the infamous San Andreas fault — but Chikyu will be the first to take such a precise look in a subduction zone. “It will be the first chance to see how such an earthquake is being prepared,” says Asahiko Taira, director-general of the Center for Deep Earth Exploration in Yokohama, Kanagawa, which manages Chikyu and its attempts to unearth the very origins of earthquakes. “Having an exposed system to observe is like being able to examine a live squid rather than a dried one to understand its biology,” he says.

Japan, which footed the bill for building Chikyu, has a good reason to focus on the Nankai trough. Here, the Philippine tectonic plate dives beneath the Eurasian plate, on which Japan sits, at a rate of about four centimetres per year. But at some points along the boundary, the plates 'stick' together and pressure builds. Two of these sticky patches, both roughly 100 kilometres wide, were responsible for earthquakes in 1944 and 1946 (ref. 2) that each killed around 1,300 people. And it is these patches that are thought to be where pressure is building for the next big quake. It's a good bet. For the past 1,300 years the Nankai trough has unleashed a large earthquake, of magnitude 8 or greater, every 90 to 210 years.

This regularity offers scientists an opportunity for a before-and-after look at an earthquake in a subduction zone. “There's no place in the world like it,” says Taira.

Going deep is the best way to study the trough. Using the riser system, Chikyu scientists intend to dig the 'ultimate' borehole. (The record for the deepest scientific hole in the ocean is held by the JOIDES Resolution vessel, which drilled to a depth of 2,111 metres in 1993.) Chikyu will also drill at least 5 other boreholes along a 70-kilometre line, spanning a range of depths above the plate boundary (see graphic above). “We can look at temperature, pressure, material composition, as well as the degree of dehydration, and see how these properties change from the shallow part of the plate boundary into the deep,” says Masataka Kinoshita, a researcher at the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) in Yokosuka and a chief project scientist for NanTroSEIZE.

Chikyu will drill boreholes through sediments washing off Japan's coast (upper left), and into the boundary between the Philippine and Eurasian plates (dark wavy lines in sandy-coloured cross-section). Credit: JAMSTEC

We can record information about the rock types before we've disturbed them. Harold Tobin

But those comparisons will have to wait until the project is completed, which won't be before 2012. The leg starting on 21 September is an 8-week-long rapid survey of the six planned borehole sites. The 16 scientists on board will use sensors attached above the drill bit to pick up signals such as γ-radiation, electric currents, and sound waves transmitted from the drill to obtain information about the porosity and density of the surrounding rock. “We can record information about the rock types before we've disturbed them,” says Harold Tobin of the University of Wisconsin in Madison, the project's other chief scientist. “This is about as close to pristine conditions as you can get.”

To keep the project moving at a fast pace, no cores will be taken. That will be the task of the next leg — a 4–5 week mission scheduled for late November. One of the cores will go down to 1,000 metres, and they will all be analysed with Chikyu's plush scientific facilities, which include a computed-tomography (CT) scanner that can reveal the internal structure of the core without destroying it. In late December, a third leg will involve penetrating to a depth of 1,000 metres at two other sites.

After that, Chikyu will be temporarily side-lined, partly because of an agreement with Japan's fisheries, partly to save money and partly to do maintenance work. Drilling is slated to resume again in October 2008, when the massive riser system will come into play. A riser, common on oil-drilling vessels but rare for scientific missions, surrounds the drill pipe all the way from the ship down to the sea-floor and below. Heavy mud is circulated at high pressure between the riser and the drill pipe to stabilize the rocks and stop them from collapsing.

The ship will recover cores in continuous nine-metre stretches, which will provide valuable information about the geological history of the region. The layers within the cores “are like tree rings”, says Tobin. “You don't want to miss any dates.” But coring is a slow process — every nine-metre section must be hoisted up into the ship before drilling can continue. Raising a core can take as little as 15 minutes, to more than an hour, depending on the depth of the water at the drill site.

Because of the premium on ship time, the crew will work around the clock. A helicopter will make runs every two weeks from the shore to exchange scientists and drilling crews. In the end, Tobin says, it will take six or more legs of concentrated eight-week drill stints to reach all the way down to six kilometres.

Chikyu's cores will provide unprecedented information about earthquakes around Japan. Credit: IODP/JAMSTEC

Half of the cores will be analysed on board; the other half will be stored in a facility at Kochi University, on the island of Shikoku, for permanent archiving. Onboard researchers get first dibs at studying them, but a year after being extracted anyone can apply to study them.

Detectors in the depths

But many scientists are more excited about the possibilities once the drilling has finished. Project scientists plan to place long-term observatories down some of the boreholes to measure rock tilt, seismic activity, strain, pore pressure and temperature — key variables for understanding how the rocks behave. The sensors must be designed to withstand very high temperatures, and will cost around ¥1.5 billion over the next five years to develop. It's not yet clear whether they will be ready before Chikyu finishes its drilling. The researchers hope to operate the sensors for at least five years after they have been installed, perhaps uploading their data to remotely operated submersibles or sending them back to shore via cables on the bottom of the ocean.

The observatories will measure changes that are surprisingly small given what can be felt on the ground during earthquakes. But “these measurements will be the key to getting a quantitative description of how the earthquake is building its energy,” says Kinoshita. Over the long term, Taira adds, the observatories might even be able to provide a way to identify the very start of earthquakes, and perhaps even to warn areas that have not yet been hit.

The measurements should also shed light on some important research questions. What happens to the rocks, and the water they carry, in the subducting plate? How does the strain released during an earthquake propagate to the surface? Under what conditions do earthquakes trigger tsunamis? Previous drilling has helped to answer other questions about the behaviour of plate boundaries. At the Parkfield site, for instance, scientists have drilled three kilometres into the San Andreas fault and found that rocks there contain talc, which could explain the ease with which the plates slide along each other at those points3.

At Nankai, excitement has grown in the past five years, after seismologists there discovered earthquakes that generate very-low-frequency waves. Before that, it was generally thought that a subduction system had 'creeping' regions that slid past other, and 'locked' zones in which pressure builds up. But using broad-band seismometers, Japanese scientists found seismic events at lower frequencies than had previously been detected, and in places thought to be devoid of seismic activity. Taira speculates that these low-frequency earthquakes, which are typically of magnitude 3 or 4, relieve strain over the long term. “It is clear they have something to do with the earthquake cycle,” he says.

The new-found earthquakes could be partly caused by water carried by the subducting tectonic plate, says Kazushige Obara, at the National Research Institute for Earth Science and Disaster Prevention in Tsukuba. The water creates a clay-like formation that “acts like a cushion” to slow the action of the earthquakes, he says.

When Obara and his colleague Yoshihiro Ito discovered these low-frequency earthquakes in the shallower region of Chikyu's drilling area, it gave the mission a whole new target to study4.

In 2001, when Chikyu's drilling was being planned, “no one had heard of these earthquakes”, says Greg Moore, also of the JAMSTEC. “We now know there is a lot of seismic activity.”

By 2012, Chikyu may have spent as much time as it needs to study these and other details of the Nankai trough. After that, project managers expect that the ship will be in high demand for other missions. Two vying to be next are a palaeoclimate study in the Indian Ocean, and a seismogenic study in the Middle America trench off Costa Rica. Both require such deep drilling that only Chikyu can do it.

A riser system will hopefully allow Chikyu to drill into Earth's mantle. Credit: JAMSTEC

And eventually, Chikyu could achieve one of scientific ocean-drilling's greatest dreams. In the 1960s, scientists envisioned an ocean drilling project that could pierce Earth's mantle. The project, called 'Mohole', never got deeper than 200 metres beneath the sea-floor — let alone to 10 kilometres, where the crust borders the mantle. But one day Chikyu might try its own version of Mohole. To do so, it would need a costly extension of the riser from its current 2.5 kilometres to 4 kilometres so that it could operate in the deep waters where the mantle is closest. But it would be worth it, says Kinoshita. “This is a long-held dream of all mankind, or at least of all Earth scientists.”