A project to wire up an entire plate of the Earth's crust with a grid of seabed observatories is set to usher in a new era in understanding the inner workings of our planet, says Geoff Watts.
Something big is stirring off the west coast of North America.
An underwater observatory larger than England and Scotland combined is being devised. An entire tectonic plate of the Earth's crust is to be wired up to scrutinise the workings of our planet as never before. A vast variety of diverse projects ranging from earthquake prediction and climate-change studies to fisheries monitoring will be integrated for the first time. The Neptune project, with its network of 30 seabed observatories linked by 3,000km of cable and serviced by robots, is expected to do for earth sciences what the Human Genome Project has done for genetics and the Hubble Space Telescope for astronomy.
The man in charge of what amounts to an expedition across a scientific frontier is John Delaney, professor of oceanography at the School of Oceanography at the University of Washington in Seattle. He says that when Neptune, at a cost of hundreds of millions of dollars, begins monitoring the small Juan de Fuca tectonic plate, possibly as early as 2006, it will usher in a new era.
Whether it's the chemistry of the sea or the migration of whales, most researchers in Delaney's discipline operate in small, separate groups. "But once you begin fully instrumenting a volume of the ocean that is 500km by 1,000km, the way opens for collaborative research," he says. "Think of what happens when an earthquake takes place. Fluids migrate out of the rocks into the ocean, carrying nutrients that microbes can dine on. You may get a bloom of bacteria." Such events show how seemingly unrelated fields such as seismology and bacteriology are linked.
It is a vision that excites many of the world's leading scientists. Alan Chave, senior scientist in applied ocean physics and engineering at the Woods Hole Oceanographic Institution in Massachusetts, explains that most ocean data have to be collected episodically: you mount an expedition, take measurements, then come home. Neptune will allow continuous studies over its 30-year life span. "And the monitoring will be interactive," Chave says. "You'll be able to change what the instruments do depending on what you find down there. You'll be able to reprogram the science as you go."
Picture just one of these 30 instrument stations that will border the Juan de Fuca plate. Linked by two thick power cables to its neighbouring stations 100km distant, it sits on the sea floor. Mounted on its framework is an array of measuring equipment. Connected to it by shorter cables are sensors monitoring events in the surrounding water or beneath the seabed.
Suddenly there's movement. The sand is disturbed by the arrival of an unmanned mini-submarine. This remotely operated vehicle unplugs a sensor and plugs in another it has brought from a surface ship. Using the station as an underwater garage, it recharges its batteries. Minutes later, the area is flooded with light. A shoal of fish has triggered the station's video camera. Pictures of the migrants join the data travelling along the fibre-optic component of the cables back to land-based scientists.
At Neptune's heart is the theory of plate tectonics. The surface of the Earth appears to be rigid, but appearances are deceptive. The outermost layer of our planet, the crust, is thin - as thin as 5km under the oceans.
It forms the top of the lithosphere, a shell some 80km thick that is mostly made up of the upper reaches of the hot, semi-liquid rock of the mantle.
Early in the Earth's history, this shell broke into a dozen or so slabs or plates that vary in size from a few hundred to several thousand kilometres across. These plates move around on the liquid depths of the mantle, propelled by underlying convection currents at speeds of a few centimetres a year. In some places, they grind together; sometimes, one slides under another; elsewhere, new crust is formed by material rising to the surface.
Earthquakes, mountains, volcanoes and ocean trenches all find an explanation in plate tectonics.
The Juan de Fuca plate is one remnant of the much larger Farallon oceanic plate that is sliding beneath North America. One of Neptune's principle research targets will be earthquakes, which are often generated at plate margins. According to Chris Barnes, professor of palaeobiology at the University of Victoria and head of the Canadian part of the project, there is a big movement of the plates along the Northwest Pacific coast every 3,000-5,000 years, most recently in 1700. "Cities such as Vancouver, Seattle, Victoria and Portland have all been built since the last event of that proportion," he says. "So it behoves us to try to understand the processes and see if there's a way in which we can get a bit more warning."
Even a little warning time might be enough to shut key systems to prevent fire damage.
Barnes is also confident that Neptune will provide a powerful lesson in biology - and not just about fish. "Where new crust is formed," he says, "great quantities of organic matter get pumped out, because bacterial activity penetrates at least a kilometre into the ocean crust." Geophysics will likewise benefit. As the Juan de Fuca plate slides beneath North America, most of the organic-rich sediments on the seabed are scraped off and pile up. In some places these sediments include "gas hydrates", a slushy, waxy material made of methane molecules surrounded by what amount to tiny cages of water molecules. "Gas hydrates contain more hydrocarbons than all the world's oil reserves put together," says Adam Schultz, professor of geophysics at Cardiff University. "If someone could find a way of extracting the methane commercially, the beneficiaries would include countries such as Japan, which have no oil of their own." Neptune could provide the vital clues.
Ambition on this scale is quite new to the earth sciences, Schultz says.
"As a community, we've always been rather hesitant to make the leap to big science. We tend to think about relatively small amounts of money for specialist groups that aren't globally coordinated."
Indeed, Neptune will not be cheap: it will cost $250 million (£158.6 million) to install and run for the first five years; $10 million-$15 million a year to operate thereafter. But compare that with an ocean-going research vessel - the US ship Healy cost $380 million to build and equip, and it has annual operating costs of $17 million. The Canadians hope to get government approval soon for their 30 per cent share of the bill, while the US money was earmarked in the president's recently announced budget plans.
Chave is confident. "The tea leaves are all looking right," he reckons.
Two pilot programmes dubbed Mars and Venus have already been funded. Mars - the Monterey Accelerated Research System - will see a fibre-optic cable laid 60km out in California's Monterey Bay by 2005. "One of its functions will be to assess instruments before they go to Neptune," says project manager Keith Raybould, chief operations officer project coordinator at the Monterey Bay Aquarium Research Institute. "Juan de Fuca is a hostile environment in winter. To get a ship out, you might have to wait months. In Monterey Bay, we can locate instruments, test them, bring them back if there are problems, and get them out to sea again the next day." Venus - the Victoria Experimental Network under the Sea - is, like Mars, a near-shore test bed for Neptune equipment and experiments.
Long term, Delaney and his colleagues have an even more ambitious vision.
The names Neptune, Mars and Venus offer an unintentional hint. Delaney makes it explicit. "Many of us view this as a lead effort for exploring other planets," he says. "The project could serve as a test bed for sensors and robots designed to search oceans elsewhere in the solar system."