Life in some of the deepest ocean trenches in the world is being revealed by a UK-built video system designed to withstand pressures that would cause any other device to implode.
The autonomous camera, developed at Aberdeen University's Oceanlab, recently filmed fish at 7,700m beneath the surface of the Pacific Ocean — the deepest any fish have been captured on video — and it has the potential to explore even greater depths.
In laboratory tests the camera withstood pressure of 1,400bar, which would be found at sea depths of 14,000m. The Mariana Trench in the eastern Philippine Sea, the deepest part of the world's oceans, has a maximum depth of 11,000m.
The camera, which uses a light-sensitive digital image sensor called a charge-coupled device to form images, is housed inside a stainless steel case and attached to a PC and recorder card encased in a stainless steel tube. All this is plugged into a 12V lead acid car battery that is pressure compensated with oil so it can be dropped to any depth.
The equipment is installed in an aluminium tripod and the whole package is known as the Hadal-Lander.
The lead designer of the kit, Alan Jamieson of the Oceanlab, said his biggest challenge was finding a transparent material for the system's window that would not deform under deep-sea pressure.
He said his group looked at the materials used in past deep-sea exploration projects for inspiration.
The first they studied was the US Navy's Trieste, a deep diving manned submersible that dived nearly to the bottom of the Mariana Trench and back up again in 1960. It is now out of service.
The viewports of the Trieste were made of pressure-resistant acrylic plastic, which were reported to be effective at depths of 10,000m.
Jamieson said his group's test on that material ruled it out for use on Hadal-Lander. 'We did a whole series of tests with acrylic and basically wrote that off,' he said. 'When the acrylic was put under pressure mimicking 8,000m, it became fluid and deformed.' He claimed that the crew would have seen little outside the Trieste window at 10,000m.
Following that, Jamieson said his group considered using a glass hemisphere, but they quickly determined that it would be too expensive to make and difficult to mount.
After reviewing plastic and glass, the only strong transparent material left was ceramic. 'The only ceramic you can properly see through is sapphire,' he said.
The team bought sapphire that was grown in Russia and sold through a Cambridge company. With this material, Jamieson designed a 80mm diameter disc for the window.
The camera only requires a 30mm diameter viewing area, but he said the extra space was required to accommodate the seals that stop the water getting into the camera.
The sapphire window, which is 15mm thick, is thinner than the stainless steel housing it is in. 'It's that strong,' he said. 'It's also optically much better than any of the other materials that we studied because it is flat and polished.'
While one might think sapphire would be the most expensive option for the group to choose, Jamieson said this is not the case. 'The cost of machining a glass hemisphere becomes more expensive than a simple, plain sapphire disc,' he said.
The sapphire disc, like all the materials, was put through rigorous testing. Jamieson said Oceanlab has a 75cm diameter, 1.8m deep, pressure vessel capable of simulating pressure equivalent to 7,000m below the surface, about 700bar.
For their camera they needed an operational pressure test to 1,000bar but given the 'riskiness' of the project, they over-designed it slightly and tested to 1,400bar for 24 hours.
In order to do this, Jamieson went to an offshore industry facility in Peterhead, Aberdeenshire, where there were large, heavy-duty sub-sea valves.
The group sealed up a valve, placed the sapphire test piece inside and ramped up the internal pressure to 1,400bar.
'So it wasn't a real pressure vessel but rather a subsea valve that had been recently manufactured and just happened to be able to take enormous internal pressures,' he said.
Over the past year the camera has been used dozens of times. It was dropped to a depth of 5,500m eight times in the Mariana region. It explored depths between 6,000m and 8,000m in the Kermadec Trench, located near Samoa and New Zealand, and it was used to search for life at even greater depths between 9,000m and 10,000m in the nearby Tonga Trench.
Jamieson said it was their expedition at the Japan Trench between September and October this year that gained the most attention. They filmed a large group of highly active snailfish fighting over bait in front of their camera at 7,700m. The camera was programmed to record for one minute every five minutes until it ran out of power, which lasted about 10 hours.
Jamieson said his team was surprised to catch so many fish on camera. 'The populations are so low at that depth it's surprising we saw so many,' he said.
Even though they have not recorded fish at a greater depth yet, Jamieson said there is no ecological reason why there would not be fish at 10,000m. 'It seems very odd that fish would only go halfway down these trenches and not to the very bottom,' he said. 'The more we find out with this camera the more we are coming to the opinion that they probably do. It's just that no one has found them yet. It's just a case of getting this camera in the water as much as we can.'
Next March Jamieson and his team plan to go back to Japan to explore a trench near the Japan Trench called the Izu Bonin. They hope to use the camera at a depth of 9,500m.
'Fingers crossed I think we'll find something probably more interesting than the last one,' he said. 'If there are that many fish at 7,700m, the chances are there are more much deeper.'