Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Polar Research

Buried treasure

Air bubbles trapped in the Antarctic ice sheet could yield precious information about Earth's climate more than a million years ago. But to access this record, scientists first have to climb one of the coldest peaks on Earth. Nicola Jones reports.

Dome Argus in Antarctica is a silent and lonely place. Snow stretches to the horizon in all directions, unbroken by any sign of life. The mountain is one of the least accessible parts of the frozen continent, and quite likely the coldest place on Earth — although no one has been there long enough to catch a record-breaking low. People first stepped onto the summit just two years ago, after a Chinese team made an arduous month-long, 1,200-kilometre trek from the coastal station of Zhongshan.

A Chinese team hopes that Dome A will provide the oldest ice core in the world. Credit: S. HOU

If you think that buried treasure should be hidden somewhere as remote as possible, then Dome Argus, or 'Dome A', would be an ideal spot. But if you think that it should be gold and shiny, you will be disappointed. Dome A's bounty, if it exists, is tiny pockets of gas trapped in its depths.

Antarctica's great ice cap is crowned by several flat snowy summits known as Domes A, C and F (see map). Reaching the top of Dome A, the tallest and the last to be conquered, was hailed as a triumph of exploration. The scientific importance, though, is not the height itself (although astronomers are excited by the clear air that comes with it) but the ice below. From more than 3,000 metres below Dome A's 4,093-metre peak, researchers hope to extract the oldest ice core in the world, and with it a treasury of climate information.

As part of the International Polar Year, which launches this month, a Chinese team plans to return to Dome A this austral summer to set up a camp. Next year, a larger, international group intends to storm the dome in aircraft to map the ice below. And the Chinese will return too, this time lugging a giant drill. Dome A might not be so lonely or so quiet for long.

An ice core from Dome A would join a formidable list of samples taken from previous Antarctic expeditions. A core from the Russian station Vostok eventually reached 3,600 metres deep and yielded measurable ice some 420,000 years old1. A core from Dome C reached only 3,200 metres down2 but, thanks to a better preserved bottom section, pushed 800,000 years back in time. And a 3,000-metre core being drilled at Dome F caused fleeting excitement last year when the Japanese team spearheading the project said that the ice might be even older. But more recent tests have shown that the usable ice from close to the bottom of the dome was some 720,000 years old, and the team now suspects that the last few chunks will hold only a few thousand years more.

A wealth of information

The reason for this old-ice treasure hunt is simple. Earth used to experience periods of glaciation once every 41,000 years or so. Today, glacial cycles come every 100,000 years. Evidence from sediment cores suggest that the key transition between these states took place over a period of several hundred thousand years, about a million years ago. No one knows why it happened. One idea is that levels of carbon dioxide in the atmosphere plummeted and cooled Earth enough for a substantial extra layer of ice to form. This massive burden of ice would have made it hard for the planet to respond so nimbly to the orbital drivers of climate change, thus shifting it into a more stately pace of glaciations.

The only way to confirm the idea is to find air bubbles that date back 1.5 million years or so, and track how the carbon dioxide levels changed over time. Ice is the only place to find such bubbles. And for old ice, Antarctica is the place to go. Ice cores from Greenland, where the ice flows more dynamically, reach back only 100,000 years or so. But Antarctica has been covered in ice for an estimated 30 million years, and models of glacial flow suggest that there could be an uninterrupted record of ice that stretches back a few million years — probably beneath Dome A. The International Partnership in Ice Core Science (IPICS), a 19-nation group co-chaired by Eric Wolff, an ice-core specialist with the British Antarctic Survey in Cambridge, UK, and Ed Brook of Oregon State University in Corvallis, has made finding that ice one of its main goals.

Dome A not only has plenty of ice to drill, it also gets very little snow — just 1.25 to 1.5 centimetres of its equivalent in water per year, compared with the 3 centimetres at Domes C and F or the 50 centimetres that dumps on the coastal station of Halley. That means that the ice on Dome A contains snow from a very long time period.

But there are complications. The Antarctic ice cap acts as an insulator, lying like a blanket across the continent and trapping geothermal heat below it. The thicker the ice, the greater the insulation, and so when the ice gets really thick its base will frequently become warm enough to melt, shortening the record.

Also, the topography of the rock beneath the ice is complex, to say the least. A mountain range lies down there (see 'The hidden mountains'). If the ice at the bottom of the ice sheet has been forced to move up and over rocky ridges, it will be folded, muddled and mixed, making it impossible to date it or to extract clean information from it. The bottom 70 metres of the Dome C core were like this, making its oldest ice unusable.

Models suggest that the flow of ice away from the base of Dome A is small, so older ice should still be preserved at the bottom. A map of ice ages by modeller Philippe Huybrechts of the Dutch-speaking Free University Brussels in Belgium (see map), confirms that an area near the peak of Dome A — a vast swath about the size of Britain — is suitable for an old-ice hunt.

Figure 1: Antarctica's oldest ice


Apart from Dome A, other candidate sites for the oldest ice do exist. The Aurora basin, near Dome C but closer to the coast, for example, is about 4,500 metres deep and could potentially hold very old ice. Australia plans to drill a 400-metre test core there in the 2008–09 season, says Vin Morgan of the Australian Government Antarctic Division near Hobart, Tasmania. But this area is lower and warmer than Dome A, increasing the chances that its bottom ice has melted substantially.

The nearby Astrolabe basin has an even deeper 4,700 metres of ice, but covers a small area in which the surrounding rock may have distorted the ice at the bottom. Farther afield, there are other areas in Antarctica likely to hold very old ice (see map), but they are trapped in mountain ranges where the ice record is much thinner and more squashed.

Disturbing the peace

Dome A thus remains the prime candidate for drilling. But too little is known for researchers to draw an X in the snow and plant their drill. So starting next year, if funding comes through, the sky over Dome A will be filled with the rare noise of low-flying planes, burdened with radar and equipment to measure gravity and magnetic fields.

Particularly helpful will be the radar surveys, which can pick up changes in density, crystal structure or dust content in the ice. By flying from Dome C to Dome A, the IPICS team hopes to be able to track ancient layers along the 1,000-kilometre flight path, thus revealing the depth of correlated layers in the ice at Dome A.

Given the remoteness of Dome A, the planes will need local bases from which to refuel. Ideally, these will be placed around the dome at slightly lower altitudes, as the height of the dome makes the air so thin that propeller planes have trouble taking off and pilots' functioning can be impaired. “Technically, the pilots should be using supplementary oxygen,” says Wolff. And in similar circumstances, loaded Twin Otter planes have had to use jets strapped to their wings to gain enough lift.

That same season, the Chinese researchers plan to return, this time carrying French drilling equipment capable of bagging a 500-metre core — an upgrade on the 110 metres they pulled from the ice in 2005. Their work will be dangerous; last time, a team member fell into a crevasse while working at the summit camp. “He was lucky — it wasn't so deep,” says expedition member Shugui Hou, of the Chinese Academy of Sciences' Cold and Arid Regions Environmental and Engineering Institute in Lanzhou. The good news is that summer temperatures are only −35 °C. “When the weather was good we were wearing only one sweater, perhaps because it's so dry,” says Hou. “It's quite comfy.”

Pooling resources

Funding a serious drilling effort will be a bigger project than any country can tackle alone. “No one nation has the resources. We're going to have to do it internationally,” says Alan Rodger, head of science programmes at the British Antarctic Survey. With the help of many countries, Wolff estimates, a site could be selected and a drilling operation under way by 2012.

“We want to drill two cores, to give ourselves two chances,” says Wolff. One idea would be to blast through the top part of the first hole with a heat drill, not bothering to pull up usable cores until reaching a predetermined depth. Meanwhile, a second site, tens or hundreds of kilometres away, could be drilled in detail from the surface. This strategy of extracting two cores would be a new one; most other sites poured all their resources into a single drill hole (although jammed drills have often meant that teams had to start a new core).

The technology, at least, exists. Drilling deep into the Antarctic is tricky but doable. The difficulties lie in such things as getting the drill fluid right. The cold, dense liquid that drillers insert into the hole to stop it from closing up on itself must not dissolve the snow around it or get clogged with ice chips that can form a sludge; kerosene with chemical additives is often used. And when the drill gets near the bottom, where the ice can be near the melting point, the problem is to stop ice that melts and refreezes from jamming the drill bit, by using some antifreeze.

Once the core has been extracted, researchers will use all the techniques they can to date the ice and pull information from it. Old layers of ash from known volcanic eruptions act as date markers, and the top part of the core can be matched up to previous ones, already dated, to pin down the age. Another dating trick for the older ice will be to search for the higher amounts of beryllium produced by the increased flux of cosmic rays when Earth's magnetic field reverses itself3 — as happened 780,000 and 900,000 years ago. Analysing the amount of nitrogen and oxygen in the ice can also help; the ratio of the two changes in concert with a 23,000-year cycle of alterations in the amount of solar radiation reaching the Antarctic4.

Technicians will then crush the ice in a vacuum to release the air, and measure the amounts of greenhouse gases such as carbon dioxide and methane. The ratio of oxygen isotopes can also be used to estimate past temperature. But the big prize will be the carbon dioxide — the amount of the gas in each bubble could confirm the idea that a drop in carbon dioxide caused the change in glaciation cycles a million years ago.

As always, there is still the chance that the ice won't hold the expected treasure. If so, the answer to this million-year-old puzzle may lie elsewhere: perhaps, some speculate, the vast plateau of the Canadian shield was at some point scrubbed clean of lubricating mud by all the glaciers, and it was this, rather than a drop in carbon dioxide, that allowed ice sheets to build up enough to slow the planet's cycle of glaciations. Finding the oldest ice could go a long way to answering these questions.

If not the bounty of plummeting carbon dioxide, then a core from Dome A would still guarantee the gem of an ancient climate record. Xiao Cunde, another member of the Chinese expedition, puts the chances of finding million-year-old ice under Dome A at 95%, and 80% for ice older than 1.5 million years. With odds like that, most would be happy to go on a treasure hunt.


  1. 1

    Petit, J. R. et al. Nature 387, 359–360 (1997).

    ADS  CAS  Article  Google Scholar 

  2. 2

    EPICA community members Nature 429, 623–628 (2004).

  3. 3

    Raisbeck, G. M., Yiou F., Bourles D. & Kent D. V. Nature 315, 315–317 (1985).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Bender, M. Earth Planet. Sci. Lett. 204, 275–289 (2002).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Additional information

See Editorial, page 110 .

Related links

Related links

Related links in Nature Research

Climate warming 'seesaws' between the poles

Ice core shows its age

Greenhouse-gas levels highest for 650,000 years

Palaeoclimate: Frozen time

Climate change in focus

Polar research in focus

Ice cores Web focus

EPICA adventure

Related external links

IPICS white papers

IPY Chinese proposal for exploration of Dome A

IPY Chinese proposal for traverse to Dome A

IPY Australian proposal for astronomy at Dome A

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Jones, N. Buried treasure. Nature 446, 126–127 (2007).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing