Geoscientist Online

Geoscientist Online 23 May 2007

Amarendra Swarup reports on a new understanding of how ice sheets melt. It's not as simple as you might think…

Global warming and the dwindling glaciers around the world are an all too common refrain in today's news. But behind the headlines and the endless guides to reducing your carbon footprint lies an increasing worry that our understanding of how and what changes may occur is still very much an inexact science. Now, evidence is building up that our theories about how great ice sheets respond to changing temperatures – a vital factor in predicting the future behaviour of global sea-levels – are indeed far from complete, and scientists are rushing to try and plug the gap in understanding.

“Recent observations suggest that major changes in parts of the ice sheets are taking place over timescales of a few years to decades, not thousands of years as traditionally believed” explains Jonathan Bamber at the University of Bristol, who has recently published a paper with colleagues examining the growing evidence that there is still much we don’t understand about ice sheets. “These changes were not predicted by numerical models and the underlying causes are uncertain.”

The finding is unsettling. The great ice sheets covering Antarctica and Greenland contain about 80% of the Earth's fresh water and cover 10% of the world's land surface. If they were to melt completely, global sea levels would rise some 70 metres, making understanding their behaviour critically important. Previous simulations of future climate, such as those carried out by the United Nations Intergovernmental Panel on Climate Change (IPCC), have assumed that the dominant changes in the size of continental-scale ice sheets take place over thousands of years. According to the IPCC’s third assessment, the primary impact of increasing temperatures on the ice sheets will be increased surface melting along coastal Greenland combined with increases in snowfall in Antarctica and central Greenland. The result is probably ice-sheet growth over the next few decades and then perhaps some shrinkage, depending on the extent and speed of warming.

However, observations in the last decade are starting to paint a different picture and point the finger at an overlooked factor – ice sheet flow (see below). Important marginal regions of the Greenland and Antarctic ice sheets have exhibited increased ice flow, contributing to sea-level rises in recent years, with warming – atmospheric, oceanic or both – the likely cause. In the case of Greenland, a recent estimate suggests this has resulted in an almost threefold increase in its contribution to sea-level rises in less than a decade. Other results from three independent satellite-based approaches indicate that the West Antarctic Ice Sheet is also losing significant mass.

“Snowfall and surface-melting for ice sheets is not perfect, but is pretty good” explains Richard Alley at Pennsylvania State University, who worked with Bamber on the paper. “Too warm, and ice melts; but the IPCC summary gave at least decades before notable sea-level rise resulted from this. Ice flow is much more problematic because of great observational and modelling difficulties.”

Though there is uncertainty about the absolute magnitude of losses, there is certainly a growing body of observational evidence that mass loss has increased in the last decade – a trend likely to be maintained if the amplified warming in the polar regions continues. The implication is that profound changes in ice sheet flow (and hence mass) are possible over much shorter timescales, complicating the already difficult task faced by scientists of projecting the future of the ice sheets. A question mark still hangs over whether these speed-ups are minor perturbations that will stabilise soon, or harbingers of larger future changes.

According to Alley, it is hard to find mechanisms that would greatly slow the flow and contribute to any stabilisation or drop in sea levels. In contrast; mechanisms to speed flow and contribute to sea-level rises are now known.

Waffle

“Flow will go faster if the bed becomes more slippery” says Alley. “Think of pancake batter spreading on a griddle – it goes faster if you grease the griddle, and if it’s on a smooth griddle than on a waffle iron.”

Observations from Greenland indicate that global warming could grease continental griddles in the future. They point to surface meltwater reaching the bed and lubricating ice motion near the edge of the Greenland ice sheet. While the increased speed from the additional water is not huge, it does appear real. In a warming world, this meltwater could also move inland to parts of the bed that are currently frozen and cause thawing, speeding the ice motion even further.

Another potential problem is that oceanic or atmospheric warming can thin and remove ice shelves – those flange-like edges of the ice sheet where the ice spreads and floats over the ocean waters, running aground on islands or scraping past the rocky sides of fjords. Such shelves provide a valuable restraint on ice flows due to the friction with the islands or fjord sides. Removing them can speed the process, raising sea levels.

“The processes responsible are not new but glaciologists were not aware of the existence of some, and their importance in influencing ice sheet dynamics” says Bamber. “The magnitude of the effect is far larger than we had expected and it is likely that these large fluctuations in ice dynamics have existed in the past but our observational capability has only recently provided us with hard evidence in the last decade or so.”

As oceans and the atmosphere warm, Bamber and his colleagues expect to see more examples of rapid and large changes in ice sheet volume. The race is now on to understand how quickly and to what degree these ice sheets may impact sea levels worldwide, as well as determining the relative influences of regional oceanic and atmospheric warming in the process.

“The real difficulties exist in a full understanding” says Alley. “There is a wealth of fascinating and important science to be done.”

Bamber agrees. “The observations provide a compelling picture of how the ice sheets are behaving” he says. “We need to better understand the physics underlying the mechanisms responsible and incorporate this into models so that we can better predict their future behaviour.”