Climate change is causing a long-term decline in Arctic sea ice, and scientists expect the Arctic Ocean to be largely ice-free in summer at some point this century.
But is that broad prediction too complacent? This week, the Guardian claimed scientists working for the US Navy believe summer sea ice could disappear as soon as 2016, based on the results of a sophisticated new computer model.
But having looked at the research, it turns out the 2016 prediction is from an older, simpler model, and isn’t the US Navy prediction of what’s going to happen in the Arctic. It’s also much sooner than most polar scientists would suggest.
An ice-free prospect
Arctic sea ice is declining by nearly four per cent per decade, according to the latest report from the Intergovernmental Panel on Climate Change (IPCC). The loss is particularly noticeable at the end of summer, when the ice reaches a seasonal low.
Average extent of Arctic sea ice in summer (colours represent different datasets). Source: IPCC 5th Assessment Report, Summary for Policymakers (p8)
The change has wide-reaching consequences, so when it might happen is an important question.
In a recent Guardian blog post, Dr Nafeez Ahmed reports on research suggesting the Arctic Ocean could be nearly sea ice-free in just a few years time.
Under the headline, ‘US Navy predicts summer ice free Arctic by 2016’, Ahmed writes:
“An ongoing Department of Energy-backed research project led by a US Navy scientist predicts that the Arctic could lose its summer sea ice cover as early as 2016 – 84 years ahead of conventional model projections.”
Where did the story come from?
The piece cites a research paper by Professor Wieslaw Maslowski from the US Naval Postgraduate School and colleagues. The research was published last year in the journal Annual Reviews: Earth & Planetary Sciences.
The paper assessed several different methods available at the time for predicting when Arctic summers will become sea ice-free.
The Guardian blog quotes from the paper:
“Given the estimated trend and the volume estimate for October [to] November of 2007 at less than 9,000 km3, one can project that at this rate it would take only nine more years or until 2016 Â± 3 years to reach a nearly ice-free Arctic Ocean in summer.”
The theme of the article is that “conventional” climate modelling is “out of pace with speed and abruptness of global warming”.
The piece also describes a new, state-of-the-art US Navy climate model, called a Regional Arctic System Model (RASM). The model “uses complex modelling techniques that make its projections more accurate than others”, the Guardian reports.
In describing the new, more sophisticated model, the blog gives the impression that it produced the 2016 figure, which would make it the most up to date prediction for sea ice free Arctic summers available.
But this isn’t the case. Maslowski tells us the 2016 prediction doesn’t come from the RASM model. The quote refers to another, simpler method the team used to examine sea ice loss.
The RASM model combines the latest satellite measurements and scientific understanding of how the atmosphere, ocean, rivers and vegetation interact in the Arctic to melt sea ice, Maslowski says.
But before this model came into existence, the team relied on looking at past trends in sea ice volume and extending them forwards to make their predictions. This method produced the rather concerning projection of 2016, based on assuming that the processes governing ice loss remain unchanged.
Maslowski tells us his team hasn’t yet used the brand new model yet to predict when Arctic summers will be sea ice free. In other words, while the RASM model might represent the team’s most up to date research, the 2016 prediction does not. He explains:
“We do expect to compare sea ice volume results [from the RASM model] with our earlier model for the same period … possibly next year or so â?¦ [so] the 2016 figure is not from RASM â?¦ our projection (from the 2012 paper) is not ‘the US Navy prediction’.”
The Guardian story may have been inspired by a piece in a US local newspaper, the Monterey County Weekly, which also attributes the 2016 date to the RASM model and carries the headline ‘NPS researchers predict summer Arctic ice might disappear by 2016, 84 years ahead of schedule.’
2016 is looking unlikely
It’s fair to say there is a range of scientific opinion on when the Arctic ice will largely melt away. While a handful of other scientists support the prospect that Arctic summers could be sea ice free within the next few years – notably Professor Peter Wadhams, also cited by the Guardian – it doesn’t appear to be a view that’s widely held in the scientific community.
Professor Mark Serreze from the US National Snow and Ice Data Centre in Colorado tells us he doesn’t see the remainder of Arctic sea ice being lost so soon. He says:
“Seeing an essentially ice free Arctic Ocean in summer by 2016 is extremely unlikely. While we have certainly seen a rapid loss of summer ice in recent decades (compared to the late 1970s, we’ve lost about 40 per cent of the summer ice cover), losing the remainder over the next few years would be very hard to do.”
Most climate models suggest sea ice free Arctic summers are still decades away.
Under a high emissions scenario, the models used by the IPCC suggest the Arctic could be up to 94 per cent ice free in September by 2100.
IPCC climate models (thin coloured lines) give a range of predictions about when the Arctic might be ice free (dashed horizontal line. The model average (in yellow) is around 2100. Overland & Wang, 2013
However, the Arctic has been changing faster than most IPCC models predicted. Using only models that come close to reproducing the rate of ice loss in the last few decades, the date for nearly sea ice-free summers jumps forward to about 2050.
Some scientists think it could happen faster. Serreze adds:
“I am on record stating that we may lose the summer ice cover as early as 2030, and I stand behind that statement. This is in itself much earlier than projections from nearly all climate model simulations. I would agree with Dr. Maslowski that the IPCC models have shortcomings.”
New satellite observations suggest this could be at least in part down to the fact that sea ice is thinning as well as shrinking in area, making it more vulnerable to breaking up.
This story is based on one model result from a single paper. Reporting an area of climate science based on a single set of results can increase the risk of giving a distorted impression of what’s going on, especially if those results are considered outliers by the scientific community or where research is ongoing, both of which are true here.
Ahmed took a similar approach with another recent paper. The research suggested thawing Arctic permafrost could trigger a 50 billion tonne pulse of methane – a powerful greenhouse gas – into the atmosphere over just ten years.
Scientists are clear that Arctic sea ice is in long term decline because of climate change. Most scientists appear to agree that the Arctic Ocean will be essentially ice-free in summer some time this century, but the most popular scientific view seems to be that it will be decades, rather than years, before summer sea ice is gone.
Get a Daily or Weekly round-up of all the important articles and papers selected by Carbon Brief by email.