Warming Up to Scott:
A First Glance at Controls on Glaciers
Richard B. Alley
We know that warming melts ice. And, we’ll be visiting many places where the geology tells us that ice is melting. Can we blame it on warming? With high confidence, yes.
Scott had it backward in 1905, Ewing and Donn were similarly confused at the other pole in 1956, and a few people today continue to stand on their heads to read their thermometers. But, anyone who has watched the leftovers from the latest “snowstorm of the century” melting in a Walmart parking lot knows better. And knowing better shows that most glaciers worldwide are shrinking because of warming.
On the British National Antarctic “Discovery” Expedition of 1901-04, geologist Hartley Ferrar and Captain Robert Scott observed the unmistakable evidence “everywhere…of the vastly greater ancient extent of the ice-sheet” (Scott, 1905, p. 364). Reporting later, Scott argued that “…a great glacial epoch is the result of a comparatively mild climate…the chief argument is, of course, that it is physically impossible for cold air to contain much moisture…living in a severe climate, it was impossible not to realize that greater severity would have meant more sterile ice-conditions”—he thought cold brought less ice (p. 366). Scott went on to note that the greatest snowfall experienced by the expedition occurred in the warmer summer. However, Scott failed to make the additional observation that even a big summer snow melts quickly on the exposed rocks of Ross Island, with mud and dust causing grittier camps as the summer progresses—warming brings less snow overall even on the coast of Antarctica.
At the other pole, the great Maurice Ewing, then director of Lamont-Doherty Geological Observatory, and William Donn followed Scott even if they did not cite him. They proposed (1956) that ice ages are triggered by Arctic warmth removing sea ice, to feed huge snowfall on land. However, Ewing and Donn had to arm-wave away the data, which showed reduced Arctic sea ice when land ice was also reduced during mid-Holocene warmth; they instead suggested based on no evidence that the field workers had the ages wrong, with open water during the cold part of the ice age.
In reality, the Arctic field scientists were correct that reduced sea ice and land ice occurred together. Scott and Ferrar followed glacier tracks from the colder ice age when “greater severity” did what Scott thought impossible by bringing “more fertile” ice conditions. What was going on?
Here, I’ll start with the influence of changing climate on “mountain” glaciers, those that flow down from the heights and end on land. The ice-age Antarctic ice sheet probably was responding more to changing ocean temperature than anything else, which will require a bit more consideration later, but the interested reader can get some of the lowdown in Alley et al. (2007) and Joughin et al. (2012).
For a mountain glacier, snowfall feeds a high-altitude accumulation zone, which may also gain snow by wind drift or avalanche. The snow is packed to ice under the weight of more snowfall, and spreads downhill under its own weight until reaching an ablation zone where mass loss, primarily by melting, exceeds snow supply. In many cases, the accumulation and ablation zones meet near the 0oC isotherm, with almost 2/3 of the glacier area above. If climatic cooling reduces melting, the glacier extends down the mountain looking for somewhere hot to melt; subsequent warming removes the lower-elevation areas, and the glacier shrinks towards its high, cold origin. By straddling the freeze-thaw boundary, the ice is “trying” to be as sensitive to climate as possible.
If the temperature increases for the clouds that are raining or snowing, they have more water vapor to condense—by slightly less than 10% per degree Celsius. Some of the snow may switch to rain, so a glacier can expect a few percent more snow per degree warming if nothing else changes.
But, warmer air also speeds melting. Even the snowiest places on Earth give only about 10 m of ice per year, but any decent mid-latitude parking lot can melt more than that by June. The physical controls on melting of natural ice are somewhat involved, but across a great range of glaciers, mass loss increases by roughly 30-40% per degree Celsius (e.g., Oerlemans, 2001, p. 128; Alley, 2003), with some glaciers going as low as 20% or so and others above 50%. Typically, a small warming causes a much larger increase in melting than in snowfall, and the glacier retreats.
Because the extra melting from a small warming is only a few times bigger than the extra snowfall, you might imagine situations in which a big change in storminess could cause a glacier to grow in a warming climate. But, growing a glacier with storms generally involves “robbing Peter to pay Paul”, as the storm track shifts snow from one place to another, so if you look across a wide area and see similar behavior, you can have considerable confidence that temperature is in control.
With all the fun complexities of glaciers—debris cover, and surges, and so much more—serious people have not yet suggested getting rid of glaciologists (including me!) and just assuming that glaciers are simple thermometers. But, unless there is a really good reason to expect something else, you are wise to start out assuming that the glacier toe rises as the temperature rises.
As we head off for the Antarctic, we’ll see widespread evidence of ice loss. And, we have a pretty good idea why.
Alley, R.B. 2003. Comment on “When Earth’s Freezer Door Is Left Ajar”. Eos 84(33), p. 315, 319.
Alley, R.B., S. Anandakrishnan, T.K. Dupont, B.R. Parizek and D. Pollard. 2007. Effect of sedimentation on ice-sheet grounding-line stability. Science 315(5820), 1838-1841.
Ewing, M. and W.L. Donn. 1956. A Theory of Ice Ages. Science 123, 1061-1065.
Joughin, I., R.B. Alley and D.M. Holland. 2012. Ice-sheet response to oceanic forcing, Science 338, 1172-1176.
Oerlemans, J. 2001. Glaciers and Climate Change. Balkema, Lisse, Netherlands.
Scott, R.F. 1905. Results of the National Antarctic Expedition.—I. Geographical. The Geographical Journal, 25(4), 353-370.