This site uses cookies. By continuing to use this site you agree to our use of cookies. To find out more, see our Privacy and Cookies policy.
Skip to the content

IOP A community website from IOP Publishing

Powered by Movable Type 4.34-en

In from the cold: June 2009 Archives

Superposition means putting one thing on top of another. Nature does it all the time, but it is only in the past thousand years or so that we have worked out how to exploit it. Ibn Sina, an 11th-century Persian better known in the West as Avicenna, understood how Nature piles younger sediment on top of older, and Leonardo came very close. The first person to articulate the Principle of Superposition clearly, though, was a 17th-century Dane, Steno: In any pile of sediment, the youngest is on top and the oldest is on the bottom.

It is an idea so blindingly obvious as to sound stupid, but for a long time the obviousness blinded us to its potential: depth in the pile is equivalent to time before the date of the top. With patience and hard work, there is a historical record waiting there for us to decode.

There are exceptions that prove the rule. For example folding can overturn the layers. Little beasts that live just under the sea floor can blur the layers by burrowing. In glaciers, where the accumulating snow is a sediment just as much as the mud on the sea floor, the main problems are flow, which stretches and squeezes the layers, and refreezing of meltwater, which mixes this year's accumulation with that of earlier years. A satisfactory solution is to drill at the summit of an ice sheet, where there is no melting and the flow rate is negligible.

The payoff has been invaluable. Ice cores give us our most detailed picture of the Earth's history over the past million years. We have barely begun to unravel the story. The wealth of incident in the story is so rich that it is hard to know how to pick and choose, but a recent technical advance by Elizabeth Thomas and colleagues makes a good start. They cut slices just 2 millimetres thick from a 4.5-metre section of a core from the interior of the Greenland Ice Sheet. This section, 2070 metres beneath the surface, is estimated to represent the years from 36,401 to 36,169 BC - at a rate of 7 to 11 samples per year. The assignment of calendar years is a bit dodgy. The dates could be out by more than 1400 years. But the relative error, from bottom to top of the section, is only about three years, and the march of the seasons all those years ago can be seen distinctly in the varying concentrations of dissolved ions. We also learn interesting facts such as that 36,263 BC was a rather dry year, while 36,262 BC was so-so and 36,261 BC rather snowy.

There is more to this work than minute detail. It tells the story of the transition from a full glacial state to the warm climatic stage DO-8. The last ice age is peppered with these DO or Dansgaard-Oeschger events, warmer episodes that lasted 1000-1500 years and began abruptly.

The authors are properly cautious about interpretation. Their aim was more to show what attention to detail can uncover than to write the last word about the transition to DO-8. But they do suggest that the transition lasted just 21 years, during which snowfall increased by a half and temperature rose by 11.4 °C. This last number calls for particular caution. It needs to be seen in context, because it probably represents a local rather than a global change, and there are some technical complications to be sorted out. But at face value it implies warming at 0.5°C per year, a hundred times faster than the global warming of the 20th century and ten times faster than some extreme predictions for the 21st century.

Dansgaard-Oeschger transitions are not like the warming that is about to happen this century. For one thing, they are almost certainly not due to increases in greenhouse-gas concentrations, at least not primarily. They are more probably related to abrupt changes in the circulation of the north Atlantic Ocean. But they do share the attribute of abruptness with our near future, and that makes them intensely interesting. Avicenna and Leonardo would have understood why.

One of the truths about field work on glaciers is that most of the time the weather is rotten, even in summer, when most field work is done. But every so often the clouds lift and even disappear altogether. Whether or not the temperature goes up on one of those infrequent sunny days, the view makes up for all the sleet, wind and fog that represent the norm, and the field workers get out their cameras. The remote-sensing specialists are also grateful for these cloud-free days, because they make air photography and satellite imaging possible.

The outcome of all this fair-weather photographic activity is pretty spectacular, and much of the best work has found its way onto the internet. Here are a few of my favourite places in cyberspace for pictures of glaciers.

Glaciers Online is a web site maintained by Jürg Alean and Michael Hambrey. Jürg Alean is a Swiss teacher who studied Baby Glacier on Axel Heiberg Island, northern Canada, as an M.Sc. student. Baby Glacier is a glacier in which my university, Trent University, has a special interest, and we were fortunate to be able to arrange a return visit to Axel Heiberg Island for Alean in summer 2008. You can see the results at Glaciers Online.

Alean has translated his prowess with the camera into a distinguished career showing the world what glaciers look like. Together, he and Hambrey, a structural glaciologist (among other things) at Aberystwyth University, have published the magnificently illustrated Glaciers (2004, Cambridge University Press). Many of the illustrations are posted at Glaciers Online.

Much of the photoglaciology on the web has a flavour about it of Last Chance to See, Douglas Adams and Mark Carwardine's 1990 book about filming animals that are on the verge of extinction. All of the glaciers, almost without exception, are getting smaller, and if you return to a place from which somebody photographed a glacier several decades ago there is an increasing chance that there won't be any ice left to see. Al Gore exploited this plain fact in the award-winning An Inconvenient Truth . At OceanAlaska Kenai Fjords, Bruce Molnia, of the United States Geological Survey, uses the technique of morphing - animating a transition between before and after images - to impressive effect to show what has been happening to the glaciers of the Kenai Mountains in southern Alaska. Most are much smaller now than they used to be.

Molnia is also the author of Glaciers of Alaska, chapter 1386-K in the Satellite Image Atlas of Glaciers of the World, which has been appearing since the 1970s as U.S. Geological Survey Professional Paper 1386. Chapter K, like the Satellite Image Atlas as a whole, is a tour de force in the patient assembly of scattered information about some well-known and a great many almost unknown glaciers.

I have to say, though, that the U.S. Geological Survey's idea of a "chapter" is not well aligned with mine. I don't know how much chapter 1386-K weighs because I only have it as a 90-MByte PDF file, but chapter 1386-J, Glaciers of North America (excluding Alaska), weighs 1.6 kg according to the scales in our kitchen. Most of the chapters that have appeared so far are gorgeous.

Finally, back briefly to Gutenberg space. Three visually stunning books about glaciers, not available electronically as far as I know, are Glacier Ice by Austin Post and Ed LaChapelle (revised edition, 2000, University of Washington Press, Seattle); The Opening of a New Landscape: Columbia Glacier at Mid-retreat by Tad Pfeffer (American Geophysical Union, 2007); and Glaciologi by Per Holmlund and Peter Jansson (Stockholm University, 2002). If you are looking for visual delight, then like me you will not mind if you are unable to read the Swedish text of the latter.

On the morning of 12 October 2007 I sat down at my computer and learned that the IPCC, the Intergovernmental Panel on Climate Change, had been awarded a share of the Nobel Peace Prize. I was a contributor to the report of IPCC Working Group I, and I recall the exhilaration vividly. I also remember thrilling, or at least impressing, the students in my climatology lecture later that day by telling them the news and suggesting that a little bit of the magic Nobel dust might settle on them if they listened to me carefully.

Contributing authors are the lowest form of life in the IPCC pond, but the facts that there were 800 of us, and that we were repeating an exercise carried out three times before, has a lot to do with the success of the IPCC. Plans are afoot now for the next, fifth IPCC assessment, developing concurrently with a post-mortem on the fourth. Already there are signs that the fifth assessment may have to be more disturbing than the fourth.

Some of the reasons are laid out in a set of commentaries, The road to Copenhagen, in a recent issue of Nature. For example, earlier estimates of climatic sensitivity, the amount of warming for a given amount of extra greenhouse gas, were too low. The same target for maximum carbon dioxide concentration now means a higher maximum temperature. Second, new modelling efforts show persuasively that recovery will be much slower than we thought - many centuries, not just a couple. The biosphere and the ocean cannot soak up greenhouse gas fast enough to draw down the atmospheric concentrations at rates that previously seemed probable.

In part this is just the natural evolution of understanding. The accumulating facts yield a clearer picture as they are subjected to more and more study. But this works for past events as well as for things that haven't happened yet.

With three or four colleagues, my contribution to the IPCC's fourth assessment was to show that glaciers have been losing mass more and more rapidly over the past three or four decades. We had assembled as many of the relevant facts as we could, but a leading problem was that there aren't enough of such facts: too few measurements, too unevenly distributed.

Now, further study is showing that the IPCC numbers for glacier mass balance need revising in the pessimistic direction. First, in a paper in Annals of Glaciology I brought in a large quantity of previously unused facts by working out a way to handle measurements made by so-called geodetic methods (based on repeated mapping, as opposed to direct measurements on the glacier). These newly-accessed facts make the mass balance appreciably more negative.

Second, a study led by Regine Hock shows that the IPCC work probably didn't allow properly for the glaciers around Antarctica. We had to handle these by guesswork, because there are practically no measurements down there. The new study uses alternative but credible information, modelling the mass balance from a knowledge of temperature and precipitation, to find that the IPCC guesses were too optimistic. Call their work educated guesswork if you like, but their guesses are very likely to be better than the IPCC guesses.

It now seems probable that the glaciers were contributing about 1.3 mm/yr to sea-level rise in recent years, rather than the IPCC estimate of about 1.0 mm/yr.

We IPCC contributors did our best. If you can't find any facts, you have to think of a substitute. And the facts that you do have will keep evolving. It takes time to find, process and test them, and therefore the picture will keep changing in detail. Sometimes it will look more and sometimes less rosy. But it hasn't changed in broad outline for a long time. Indeed, it hasn't changed much since Arrhenius calculated that doubling the atmospheric concentration of carbon dioxide would increase the temperature by 5 to 6 ºC. That was in 1896. The very latest estimates of this number, higher than that of the IPCC, are about the same.

Like all specialists, glaciologists are fond of acronyms and need to be reminded not to use them if they want to be understood by normal people. I don't know why acronyms exert such power over the modern mind. It has to be more than laziness. Perhaps it is the way the string of letters squeezes such a lot of concept into such a small space.

If you are a normal person who would like some clues to the decoding of glaciobabble, two acronyms stand out. One is ELA, short for equilibrium line altitude. The one I would like to focus on here, though, is AAR, which is short for accumulation-area ratio and is a close relative of ELA.

Two recent studies by Dave Bahr, Mark Dyurgerov and Mark Meier illustrate nicely why the AAR is a powerful concept. The first, with Bahr as first author, appeared in Geophysical Research Letters, and the other, with Dyurgerov as first author, will appear shortly in Journal of Glaciology.

Glaciers are moving bodies of ice on the Earth's surface. Ice, being near or at its melting point, is a soft solid, and flows from where there is net accumulation, usually at higher altitude, to where there is net loss due to melting or calving or both, usually at lower altitude. (Calving glaciers, however, are special cases, not considered here.)

The ELA is the altitude separating net gain above from net loss below. The AAR is the extent of the upper or accumulation area, above the ELA, divided by the total extent of the glacier.

There are some important ideas squeezed into these acronyms. First, the glacier is a whole. You must consider the accumulation area and the ablation area (the area of net loss) together. Next, the size of the whole depends on the shape. For the glacier to be in balance, neither growing nor shrinking in an unchanging climate, there has to be a balance between the accumulation area and the ablation area.

If nothing happens to alter the balance between gain and loss, the ice flow adjusts matters until a particular, equilibrium AAR is attained. At typical speeds of flow, a glacier somewhat out of balance will need from a few years to a few centuries to reach equilibrium.

The authors' best estimate of the equilibrium AAR is about 58 percent (give or take 1). Earlier estimates were larger. This finding comes from well-studied glaciers for which the authors plot annual averages of measured AAR against measured mass balance. They get a cloud of dots that defines a clear straight-line relationship. The equilibrium AAR is the AAR at which this line crosses the line representing zero mass balance.

The Dyurgerov study makes it clear that the AARs actually measured over recent decades are well below 58. The average for 1997 to 2006 is 44, give or take 2. Today's accumulation areas are too small, and today's glaciers too big, for today's climate, consistent with today's mass balances nearly always being negative.

The argument which follows from this finding rests mainly on an earlier demonstration that glacier volume is related to glacier area. Given current total area and AAR, we can estimate the change in volume required to reach the equilibrium total area and its AAR of 58. The authors find that to get to equilibrium with today's climate the glaciers will have to shed meltwater equivalent to 184±33 millimetres of sea-level rise. That would require somehow stopping climatic change in its tracks. But if we carry on with business as usual, the loss over the next hundred years comes out at 373±21 mm, or 3.7 mm/yr. The recent rate of mass loss is near to 1.3 mm/yr.

The upshot of all this is a new twist on the notion of committed change. Just as we will have to live with the greenhouse gases we have already added to the atmosphere, so we will have to watch our glaciers melt away - no matter what we do.

For an object that is supposed to be in the frozen state, your typical glacier has a surprising amount of liquid water in it. Even in the coldest parts of the Antarctic Ice Sheet, you can find water in the spaces between the ice crystals. It is liquid partly because it is under pressure but mainly because it is very salty, and in truth the amounts are tiny.

At the glacier bed, though, there is often a great deal of water. This makes sense when you think about the energy balance down there. The ice may, as in Antarctica, begin its sojourn on the Earth's surface at a temperature tens of degrees below the freezing point, but as more of it accumulates the ice at the bottom can only get warmer. Some of the energy comes from friction as the ice moves over the bed. More comes from geothermal heat, the slow leakage of energy from the Earth's interior. The ice has to warm up to its melting point eventually.

The basal meltwater has to go somewhere. A lot of it ends up in subglacial lakes, such as Subglacial Lake Vostok in East Antarctica. Subglacial Lake Vostok is about the same size, and coincidentally about the same shape, as Lake Ontario, but the resemblance is superficial. This is hardly an appropriate use of "superficial": Subglacial Lake Vostok, apart from being hundreds of metres deep and millions of years old, is buried under 3.5 kilometres of ice. By comparison Lake Ontario is an ephemeral puddle.

Vostok is the biggest one we know of, but a couple of hundred other subglacial lakes are also known and more are being found all the time. One was newly identified last year only 15 km from Amundsen-Scott, the American base at the South Pole. Now, in a paper in Journal of Glaciology, Helen Fricker and Ted Scambos have described in more detail than before a complex system of connected subglacial lakes in West Antarctica.

Two things fascinate me about this report. First, although you can find subglacial lakes by hauling a radio echo-sounder across the surface of the ice sheet, these were found and their behaviour tracked by repeated observation from 600-700 km away. MODIS is an imaging satellite, and ICESat is a satellite travelling in a precisely repeating orbit from which it points a laser altimeter at the Earth's surface. Subglacial lakes show up in ICESat track records as flat patches. The flatness means that the ice must be afloat. When Fricker and Scambos compared earlier with later observations, they saw these flat patches rising and falling through a few tenths of a metre, sometimes more. The only reasonable explanation is filling and draining of the subglacial meltwater. Filling pushes the surface up, draining allows it to subside.

The second fascinating thing is the connection of the subglacial lakes one with another. When one drains, one or more others fill a short time later. This kind of thing has been reported before (with an interesting very recent update by Sasha Carter and others), but the accumulation of evidence is showing that Antarctic subglacial hydrology must be a lively and a complicated business.

Fricker and Scambos show a block diagram of the subglacial topography, in which the lakes are arranged in the sequence you would expect. The uphill ones, when they are full (whatever that means), feed water to downhill ones. But "downhill" doesn't mean what you would expect. This topography is a fiction, consisting of the real topography of the bed of the ice sheet plus the effect of the varying thickness of ice. The water is pressurized by this overburden, so to understand the flow of the water you have to redefine downhill so that it means "direction of the force of gravity plus the force due to the ice-thickness gradient".

There is no suggestion that this work is revealing environmental change. Antarctic meltwater has presumably been behaving like this for ages. But there are two outstanding questions, both with intriguing implications. How, as it surely must, does subglacial hydrology affect the behaviour of the ice sheet? And how does the meltwater get out?