select search filters
roundups & rapid reactions
factsheets & briefing notes
before the headlines
Fiona fox's blog

expert reaction to Antarctic sea ice paper

A study published in Nature Geoscience suggests climate warming has led to Antarctic sea ice expansion, due to cool freshwater from melt beneath the Antarctic ice shelves insulating offshore sea ice from the warming ocean beneath.


Prof Martin Siegert, Professor of Geosciences at the University of Bristol, said:

“This work indicates that the ice ocean system is sensitive to several forcing factors, and that we’re still uncovering these and determining which are the most significant. 

“Past records show that long periods of warming or cooling are non-steady, and such small scale variability may be due to competing processes and their feedbacks.  The process proposed in Antarctica is unlikely to be significant, however, due to relative lack of ice shelves around the sea ice growth regions.”


Dr Emily Shuckburgh, Head of Open Oceans at British Antarctic Survey, said:

“It has long been known that ‘global warming’ is a vast over-simplification of the changes that might be expected from increased greenhouse gases.  Unusually cold weather in the UK and increasing sea ice in parts of Antarctica are both examples of disruption to our climate that may in fact be a symptom. 

“There are fascinating and subtle interactions across the globe between the air temperature and wind, the properties and circulation of the ocean, and ice in its different forms.  This paper demonstrates how these interactions can sometimes lead to unexpected results.”


Dr Paul Holland, Polar Oceanographer at British Antarctic Survey, said:

“It is already known that the existence of ice-shelf meltwater promotes the existence of sea ice.  Basically, a layer of fresh water from the ice shelves forms a ‘cap’ on the ocean that reduces the mixing upwards of ocean heat (fresh water is buoyant, so mixing it is hard).  Less ocean heat arriving at the sea surface means that the ocean can freeze more easily.  That concept was demonstrated quite clearly in 2004 by Hartmut Hellmer of the Alfred Wegener Institute in Germany.

“This paper claims that the observed increase in ice-shelf melting has made a significant contribution to the observed increase in sea-ice cover.  That result makes sense and might well be expected, but I don’t think this paper demonstrates it to be true.  The paper uses two strands of evidence:  a Met Office collection of ocean data, and the results of two models run with increased meltwater input.

“Regarding the data, I am sceptical that there are enough wintertime Antarctic data to reliably determine changes in the ocean for 1985-2010.  There are now lots of freely-floating sensors in the Southern Ocean (‘Argo floats’), so there will be a relatively large amount of data for recent years, but there is very little for older years, which makes the reliable calculation of changes very difficult.  Also, the paper shows trends for a ‘zonal mean’ – averaging around all longitudes of Antarctica.  That is complicated because the different regions of Antarctica have experienced very different changes.

“The paper also uses trends from a satellite SST dataset.  These cannot be trusted, because the SST is simply set to -1.9C, the freezing temperature of seawater, below sea ice.  So in the regions where sea ice has increased the ocean appears to cool simply because the SST is recently being set to -2C when previously it was not.

“Regarding the models, none of the results shown match the observed changes in sea ice (compare figures 2a and 4d, or figures 2a and Supplementary Figure 5), so I don’t find that at all convincing.  The authors are using a completely wrong distribution of freshwater increase from Antarctica, putting in increased ice-shelf melting where it is not occurring.  They are also using a rate that is vastly too high.  They are using 250 Gt/y of extra meltwater for 40 years, whereas the real melting increase has ramped up slowly since 1992, adding only 71 Gt/y on average in that period (and much less before that).  The real distribution and rates can be found in Shepherd et al 2012 [Science, 10.1126/science.1228102].  This lack of quantitative agreement is critical, because the authors are claiming that the freshwater effect is important.

“There are plausible alternative explanations for the Antarctic sea ice increase.  To use an example close to my heart, in a Nature Geoscience paper last year (attached), Ron Kwok and I showed that changes in the wind can explain changes in the ice cover.  Changes in the wind change the ice cover by both blowing the ice around directly, and by carrying colder or warmer air masses over the ocean, leading to more or less freezing.  The Bintanja paper claims that this is unlikely to be true because the observed change in sea ice does not correlate with various climate variables, such as the Southern Annular Mode (SAM).  However, the SAM and other variables describe only a fraction of the variability in the winds around Antarctica, so it is entirely possible that the winds are linked to the ice change but the SAM is not.

“The possibility remains that the real increase is the sum of wind-driven and meltwater-driven effects, of course.  That would be my best guess, with the meltwater effect being the smaller of the two.  But none of that is proven.

“In summary, the freshwater concept is plausible and its playing a role is not inconsistent with previous work.  However, this paper has not demonstrated that increased ice-shelf melting has made a significant contribution to the observed increase in sea-ice cover.  Alternative explanations remain entirely viable.”


Dr Andrew Russell, Lecturer in Climate Change at Brunel University, said:

“The Antarctic climate is complex and the increase in sea-ice is one aspect that’s been looked at a lot in the last few years.  It was shown recently that changing wind patterns around Antarctica played a role in driving the sea-ice growth but this latest study adds another important factor to that picture. 

“The new paper show that melting from the bottom of ice shelves – where the Antarctic glaciers flow out over the ocean – produces a layer of cold water that stops warmer water below reaching the surface and slowing sea-ice growth.  So overall we have two components of a warming climate – wind pattern changes and ice shelf melting – resulting in increased sea-ice, which perhaps isn’t what you’d expect but is consistent with our best description of climate change.”


‘Important role for ocean warming and increased ice-shelf melt in Antarctic sea-ice expansion’ by R. Bintanja et al. will be published in Nature Geoscience on Sunday 31st March.

in this section

filter RoundUps by year

search by tag