May 3, 2012

expert reaction to new research on Greenland glacier

According to a study in Science, a decade-long record of nearly all of Greenland’s major glaciers suggests that that Earth’s sea level may not rise 2 full meters over the next century as previously predicted.

Prof W. Tad Pfeffer, Dept. of Civil, Environmental, Architectural Engineering at the University of Colorado, said:

“The analysis presented here by Moon et al looks good to me, and their conclusions are not surprising at this point. But I should point out that the 2 metre sea level rise from Greenland by 2100 is not an appropriate estimate to compare this analysis to. In actual fact the 2 metre estimate was my own extreme value, which I arrived at by letting every variable go to the highest conceivable value immediately and see what the result was. The idea was that 2 metres by 2100 isn’t physically impossible, but is at the very outer limit of “plausibility”, and estimates beyond this really make no sense whatever.

“We went still further in that paper and chose a set of variables we considered to be much more reasonable, and using those came up with an estimate of 0.8 metres from Greenland by 2100, which is in fact still large but closer to other subsequent estimates, including those in this paper.

“Overall, Moon et al’s analysis is working toward a very important aspect of ice dynamics: the accelerations observed on outlet glaciers, most notably in Greenland but elsewhere as well, aren’t permanent features, and outlet glaciers won’t just keep melting faster and faster forever. Next, we need a good way to model this. Price et al (PNAS, 2011) have a good start on that”

Dr Jeff Kargel, Glaciologist at the University of Arizona, said:

“The authors’ excellent track records warrant high confidence in this latest work. Moon et al. have done much to extend knowledge of Greenland glacier flow speed oscillations and trends. As the authors note, the temporal baseline of observations is brief; so short, in fact, that it verges toward being the glaciological equivalent of weather. However, the authors have analyzed their data to peer beyond the interannual and inter-glacier “noise” and to discern significant trends and patterns.

“This work does not directly address ice sheet mass balance. They examined accelerations and decelerations in surface flow speeds, which can relate to but is different from changes in mass balance. The observed acceleration of mean flow speeds of marine-terminating glaciers presumably signals accelerating loss of ice into the ocean from the Greenland ice sheet. However, instead of finding as rapid acceleration as some other researchers have found for small samples of glaciers, the bigger sample of glaciers assessed by Moon et al. shows a more gentle acceleration of mean flow speeds and a lot of seeming random variation in speed. Their conclusion is that Greenland is not undergoing a rapid or catastrophic acceleration of ice into the sea.

“A couple important issues are not fully discussed in the paper.

“First, we know from microwave and other satellite observations over an equivalently short temporal baseline, there has been a tendency for surface melting to become more intense, to last longer in the summer, and occur across more of the Greenland ice sheet surface. Surface water ponding has become important, and in places surface meltwater is penetrating to the bed of the ice sheet and affecting glacier dynamics, such as increasing the sliding speed and enhancing glacier mass loss into the ocean. This process is likely contributing to the variability of flow speeds seen by the authors. Eventually, there must be a close coupling between surface meltwater formation and basal sliding. However, nobody knows the timescale over which it operates in the aggregate of the Greenland ice sheet; there have been instances where meltwater descends quickly to the bed and affects basal sliding almost instantly. However, in areas where the ice sheet may be frozen to the bed, or otherwise is resistant to sliding, there may be a long delayed effect of surface melting on basal sliding. In other words, the considerable warming of recent decades might not be felt in ice sheet dynamics, except in certain restricted locales, for several decades. The dramatic cases of rapid variation of Greenland outlet glacier speeds may be restricted to places where sliding is now favored. We can reasonably suspect that sliding will become more widespread as the ice sheet surface warms, as melting increases, and as meltwater penetration to the bed increases. Greenland may be in the barest beginnings of this expected response. The full dynamical impact of warming–which we know is happening across most of Greenland and the surroundings seas– may be highly nonlinear; this expected nonlinearity has not been picked up in this study, which necessarily was geared more to linear analysis.

“Second, some of the most dramatic dynamics observed for Greenland glaciers involve wild oscillations in flow speeds akin to the surge/waste cycle of temperate alpine glaciers (on a smaller scale than Greenland’s outlet glaciers) and the “binge-purge” oscillations of the Laurentide ice sheet of the past (on a larger scale). It appears that the authors have excluded from their trend analysis some of the more wildly behaving glaciers. These highly consequential glaciers are thus thrown out of the aggregate analysis, and this may have biased the results to the glaciers exhibiting tamer dynamics and less influence on the ice sheet mass balance changes. I do not disagree with the methodology, since it is hard to deal with high variability systems that can skew results one way or another way for any short baseline of observations, but this approach may affect the aggregate analysis.

“All said, I think this is a valuable analysis, and it probably does mean that Greenland is undergoing a gentler acceleration of mass loss than hitherto considered likely.”

’21st-Century Evolution of Greenland Outlet Glacier Velocities’ by T. Moon et al, published in Science on Thursday 3 May.