This article is reproduced with the permission of New Scientist for exclusive use by Nova users.

Deep secrets
20 April 2002
From New Scientist Print Edition.
Kate Ravilious

Layer by layer the rock grew. For 200,000 years it felt temperatures swing from high to low and then back again. It watched the sea level rise and fall. The world hurried by, two ice ages passed, Neanderthals came and went, civilisations developed and crumbled. All the while, the stalagmite sat in its quiet cave watching the sea roll in and out. Then, one day, it met a violent end as Fabrizio Antonioli hacked it from the cave floor.

Antonioli took it back to his lab at the Italian Institute for Alternative Energy in Rome and sliced it in half down its length. He and his collaborator, Edouard Bard of the University of Aix-Marseille III in southern France, took a look inside. They were amazed at what they saw. It contained a perfect record of the long-term changes in sea level dating back 200,000 years—enough to cover the two most recent ice ages.

The researchers now believe this stalagmite—and others like it—could be the Rosetta stone for climate change. Earth has swung in and out of ice ages for at least the past two million years, and alongside these variations the oceans have gone up and down. Some scientists predict that sea levels may rise by 2 metres or more in the next few hundred years, inundating low-lying land and cities occupied by hundreds of millions of people (New Scientist, 30 October 1999, p 5). But our only real chance of knowing what dangers we face in the future is to find out what has happened in the past, and that's where the stalagmite comes in.

Antonioli's discovery was an enormous piece of good luck. Aside from his job studying how the Earth's past climate is reflected in its geology, he is also a keen scuba diver. In 1991 he was diving near a small island called Argentarola, just off the west coast of Italy about 100 kilometres north of Rome. He chanced upon an uncharted labyrinth of caves nearly 30 metres below the surface.

As soon as he saw a host of knobbly grey stalagmites on the cave floor, he knew it must once have been above sea level. That's because stalagmites form as drips of water strike the floor of a cave and precipitate tiny amounts of dissolved calcite. Antonioli thought the rocks might have something to say about past sea levels, but didn't have the means to find out what.

Eight years later, though, he mentioned the cave to Bard, who is one of the world's experts in dating rocks. Bard was convinced he could decipher the rocks' secrets, so they set out to collect a sample.

Cutting off a stalagmite and carrying it such a long way up to the surface is no easy matter. Antonioli went back to the cave equipped with a hacksaw and set to work. "The biggest problem I had is that the cave walls are covered in a blanket of thick mud," he says. "After one minute of sawing the base of the stalagmite I had stirred up so much mud that I couldn't even see my hand in front of my face." But he persevered, and eventually managed to detach a 30-centimetre long stalagmite and haul it to the surface.

"The first step was to cut the stalagmite in half lengthways so that we could see its internal structure," said Bard. When they did, they were thrilled by what it revealed: yellow-brown rock layers alternating with white deposits. "It told us the stalagmite had observed the sea going up and down," Bard says. In the past the sea must have fallen far enough to expose the stalagmite to fresh air on more than one occasion. During these periods it collected the yellowish layers as calcite-laden water dripped from the roof of the cave. Then, when the sea level rose again, marine worms colonised the surface and left the white deposit. "It is at the perfect altitude to record the sea as it yo-yos up and down over the ice ages," says Antonioli.

The pair realised that if they could accurately date the changes between the white marine deposit and the yellowish rock layers, they would have an excellent record of when the sea rose and fell. And it would be far more precise than other methods, which are based on corals and tiny organisms called foraminifers. These are often unreliable, not least because the seabed rises and falls due to movements of the tectonic plates it rests on (see "Maritime monuments"). "This section of the Italian coast has been very stable tectonically for at least the last 200 to 300 thousand years," says Antonioli.

They dated the yellowish rock layers using a technique Bard has developed. It involves measuring minute amounts of the isotopes of uranium and thorium trapped in the rock. Water always contains trace amounts of uranium, a highly soluble element. Whenever rock precipitates out of water—as happens when a stalagmite forms—it retains a small amount of uranium-238. This uranium isotope radioactively decays over time, first into uranium-234, and then into thorium-230, which is very insoluble and stays put in the rock. As time goes by, increasing amounts of thorium become trapped in the mineral lattice. The older the rock, the greater the ratio of thorium-230 to uranium-238. "We know the half-life of each isotope so this allows us to calculate the time elapsed since the rock was formed," Bard says. "By measuring the uranium and thorium isotopes we can date a rock very accurately."

The researchers dated the stalagmite along its whole length and found that it had started growing a staggering 206,000 years ago. By dating the boundaries between white deposits and yellowy rock they could identify two periods when the sea level had reached a "highstand", or maximum, and times in between when it had fallen to a "lowstand" (see Graphic, p 41). They found that the first white marine layer began growing 202,000 years ago and stopped 12,000 years later. This tells us the sea level was at least high enough to submerge the stalagmite for that period. After that, an ice age began, reducing the sea level and exposing the rock to further deposits from water dripping off the cave's ceiling. Then, 145,000 years ago, the ice melted, the sea levels rose, and the rock has remained submerged. Until, that is, Antonioli brought it to the surface.

Antonioli and Bard believe their estimates are accurate to around 2000 years, and their work is to be published in Earth and Planetary Science Letters (vol 196, p 135). An estimate for the same highstand made by looking at corals and foraminifers is much more uncertain. Errors are about 4000 years either way and results seem to vary wildly. Gideon Henderson, a geochemist at Oxford University, thinks the stalagmite is a great find, since researchers have long sought something that is tectonically stable and can be dated so accurately. It's particularly welcome because the dates it covers are important ones for understanding climate change—and ones that we previously had little information about. "These results provide robust constraints on the timing of sea level change in a crucial period," Henderson says.

Having established the rock as an accurate record of sea levels Bard and Antonioli are now investigating whether it says anything about why the ice ages that cause the fluctuations come and go.

The root cause of this cycle is thought to be the natural variation in the shape and attitude of the Earth's orbit around the Sun. Over many thousands of years, the orbit changes from an almost circular shape to more of an oval, and then back again. The tilt of the Earth's axis also varies over time, and these two effects cause periodic variations in the distribution of heat that the Earth receives from the Sun.

In the 1920s a Serbian geophysicist called Milutin Milankovitch suggested that the Earth dips in and out of ice ages depending on where we are in the orbital cycle. Since then astronomers have calculated extremely precisely how the Earth's orbit has changed over the past million years, providing accurate dates of when we would expect the Earth to be in glacial and interglacial stages. The periodic variations are called Milankovitch cycles in honour of his idea.

When Bard and Antonioli checked their dates from the stalagmite with the dates predicted by the Milankovitch cycles, they found precise agreement for a period of high sea level just before the penultimate ice age. This lasted from 202,000 years ago until 190,000 years ago when the penultimate glaciation started. Both the orbital data and the stalagmite indicate that the sea was at a similar height to today around 195,000 years ago. "No coral or foram has ever given such good agreement and this gives us real confidence in the stalagmite," Bard says.

But that doesn't mean the stalagmite—or the Milankovitch cycle—has all the answers. The relationship between climate change and the Milankovitch cycle is not straightforward because there are actually three subtle, superimposed cycles. And their effects are complicated by feedback mechanisms that kick in here on Earth. The cycles could set off a chain of events—mostly to do with atmospheric carbon dioxide and ice-sheet reflectivity—that would skew the apparent correlation between the cycles and the climate.

This complexity is borne out by correlating information from the stalagmite with data from other sources. The date it gives for the rise in sea level after the penultimate ice age—140 000 years ago—is several thousand years before the date predicted by the Milankovitch cycle. To find out why, Bard and Antonioli compared the stalagmite dates for sea level shifts with records of atmospheric carbon dioxide preserved in the ice cores removed from Vostok, Antarctica. The sea level rise 140,000 years ago was associated with a jump in atmospheric carbon dioxide of 100 parts per million. But the sea level rise 202,000 years ago had a tiny increase of around 20 parts per million. Although both of the highstands indicated by the stalagmite were influenced by Milankovitch cycles, the most recent one was accelerated by a carbon dioxide feedback reaction, offsetting it from the astronomical cycles.Mark Maslin of University College London thinks that stalagmite data could contribute a great deal to the debate over the mechanisms behind the ice ages. "The precise dating feeds into the discussion on whether ice-sheet feedback mechanisms or carbon dioxide are important in stopping and starting ice ages," he says.

Finding out could give us a much better idea of what the results of our carbon emissions might be. However, the Argentarola stalagmite won't tell us enough because it doesn't date back far enough to reveal the full variation in Milankovitch cycles. To see the long-term pattern in the cycles you'd need data going back 440,000 years.

And so Antonioli is about to go diving again. He's heading back to his cave labyrinth to look for an older rock. After years of expensive Arctic exploration and satellite imaging, who'd have thought some diving gear and a hacksaw would be the best tools for monitoring climate change?

Kate Ravilious is a palaeoclimatologist and science writer based in Oxford

From issue 2339 of New Scientist magazine, 20 April 2002, page 38

For the latest from New Scientiist visit www.newscientist.com