At the bottom of the Arctic Ocean is an ecosystem that has been isolated for the past 65 million years, since geological activity separated it from the Mid-Atlantic Ridge. Like the teeming ecosystems that line volcanic ridges in the Atlantic and Pacific, its denizens, which are likely to be unique, subsist on heat and nutrients coughed up from the Earth鈥檚 interior by hydrothermal vents.
Apart from one mapping mission and an expedition that returned a few snapshots and bits of rock in 2001, for all we know about this intriguing marine life, it might as well exist on Mars. 鈥淲ith the possible exception of Lake Vostok in Antarctica, this is the closest you can come on this planet to an alien life form,鈥� says David Akin of the University of Maryland, College Park, who plans to plumb these unknown depths with robots later this year.
The lack of knowledge is partly due to difficulties inherent in exploring the ocean under thick ice-sheets that even ice-breaking ships have problems getting through, but it also stems from the fact that until recently, it was far from clear that there was life there anyway. That changed in 2001 when an underwater Arctic mission discovered eight hydrothermal vents along the Gakkel Ridge, which lies between Greenland and Siberia. These could well harbour exotic life forms because of the heat and nutrients they provide. The submarine was able to map their locations to within 5 kilometres and since then scientists have been itching to return to these vents, some of which lie more than 5 kilometres below sea level.
Advertisement
Now that looks as if it will happen. In August 2007 two intrepid robots that Akin helped to design and build will venture into the Gakkel Ridge to bring back the first ever biological samples from under the Arctic ice.
Much like the Martian rovers Spirit and Opportunity, the two Autonomous Underwater Vehicles (AUVs), Puma and Jaguar, will explore unaided. Almost entirely cut off from the world above the ice, they will sniff out active hydrothermal vents, scan for life, sample what they find and return to the surface, using cutting-edge sensing and navigational technologies.
The journey will be fraught with difficulty, not least because communicating with underwater vehicles is even harder than talking to distant spaceships.
Since electromagnetic waves barely propagate through water, the best way to send data is via acoustic signals, but these weaken rapidly with distance. 鈥淲ithin 5 kilometres, you might have some idea of where it is,鈥� says Ken Collins of the National Oceanography Centre in Southampton, UK. 鈥淎fter that, it鈥檚 easier to talk to a probe on Mars.鈥�
This won鈥檛 hamper AUVs operating in open water, because they can carry out pre-programmed missions while maintaining only minimal contact with the mother ship, and surface at a predetermined location. But pulling off this rendezvous with an ice sheet in the way is a different matter.
The bots will have a relatively easy start to their mission 鈥� being simply lowered through a hole punched in the ice by an ice-breaking ship 鈥� but retrieving them will be no mean feat. Although the bots record the location of the initial hole and are programmed to return to that spot, the hole can freeze over or the ice sheet can drift by several kilometres.
鈥淭here鈥檚 always the potential that you鈥檒l drop the thing in a hole in the ice, wave goodbye and it鈥檒l never come back,鈥� Akin frets.
鈥淚f we were going to be in the open ocean, my state of stress and panic would be way lower,鈥� adds Hanumant Singh of the Woods Hole Oceanographic Institute in Massachusetts, leader of the team that designed and built the two AUVs.
鈥淚f we were going to be in the open ocean, my state of stress and panic would be lower鈥�
Indeed, in 2005, Collins鈥檚 team, lost a 拢5.5-million AUV called Autosub beneath the Fimbulisen ice shelf in Antarctica. They believe it smashed into the ice while trying to get back to the surface. Now the sub remains stuck until global warming sees fit to extricate it.
To avoid this occurring again, Singh鈥檚 team has equipped Puma and Jaguar with a navigation system that can communicate with the mother ship via sonar as the bots near their original hole, rather than relying purely on the coordinates recorded when they first dropped through the ice. 鈥淲e can talk to the vehicle and say 鈥極kay, the hole has moved, here鈥檚 the new location. Go here鈥�,鈥� says Singh. But there is still nothing to be done about the ice when a bot has to surface due to an emergency, such as a loss of power or bearings. In these cases the bot is simply programmed to surface as quickly as possible, with the team hoping for the best.
While retrieving the submersibles may be the most hair-raising part of any mission beneath the ice, Puma and Jaguar鈥檚 real task lies in finding Gakkel鈥檚 hidden life forms. The two car-sized bots will work in tandem (see Denizens of the Gakkel Ridge). Puma will set off first to carry out the initial rough location work, reducing the likelihood of losing Jaguar, which is more expensive and delicate as it is equipped with the robotic arm, dubbed SAMURAI. 鈥淵ou look at what happened to Autosub and the odds are fairly high that that could happen to us. They are not sacrificial but, well, that鈥檚 why we built two.鈥�
Puma will sniff out active hydrothermal vents using specialised chemical and temperature sensors. The vents constantly spew into the ocean a plume of chemicals that over time cool and change in chemical composition. By constantly monitoring the concentrations of dissolved iron and gases, as well as the temperature of the water, Puma can calculate how old the plume is and use this to zero in on the vent.
When it has narrowed the source of the plume to within a few tens of metres, it records its location and returns to the mother ship. There, researchers load these co-ordinates onto Jaguar and dispatch it for more accurate mapping and imaging.
Jaguar has a pair of cameras to see in three dimensions, and shape and colour-recognition software for homing in on objects that are likely to be alive. When it finds something it would like to sample, it settles on the ocean floor and directs its robotic arm to scoop it up.
The robots are designed to perform the entire mission without a human looking over their shoulders. However, like nervous parents their creators will initially watch them closely, sending them on short, incremental missions. 鈥淲e want to push the envelope every time we go out, but won鈥檛 bet the house on them,鈥� says Singh. 鈥淲e鈥檒l say 鈥楧o your best,鈥� and if they perform well the first few times, we can let them make more decisions on the fly.鈥�
This is especially true for the biological samples. Although Jaguar is supposedly capable of identifying and retrieving samples in one trip, Akin says that after its computer completes visual scans of the vent field for possible life, Jaguar will return to the ship and biologists will decide what should be sampled. 鈥淲e can look at an object and say, 鈥楾his is probably a tube worm or a shrimp,鈥� and designate it as a target,鈥� he says. Then Jaguar will be sent back down to search out specific targets.
One snag is that the team may be no better than the robots at spotting life because the form it takes is likely to be unique. 鈥淕akkel Ridge has been separated from the Mid-Atlantic Ridge for tens of millions of years. No one knows what to expect there,鈥� says Richard Camilli, a chemist at Woods Hole.
But if the mission is successful, it will be the first time an autonomous seagoing robot has managed to grab a biological sample. 鈥淭his represents a whole new level of autonomy,鈥� says Akin, who works on similar devices for space. 鈥淭hey don鈥檛 even give this level of autonomy to astronauts.鈥�
The mission could even inspire an expedition to the frozen oceans of Jupiter鈥檚 moon Europa in search of truly alien life.
