the climate brief Accurate Reporting of Climate Science

Three-quarters of Arctic sea ice lost in 30 years? We check.

• 07 Jun 2012, 14:45
• Verity Payne

Over the weekend an Observer Magazine article featuring a Greenpeace stat - that 75 per cent of Arctic sea ice has been lost over the last 30 years - prompted outrage amongst climate skeptic bloggers, who suggest that even "[f]ive minutes or less of checking would have prevented this blunder."

We took up the suggestion and spent rather more than five minutes checking both the 75 per cent figure, and the responses to it. The outcome? A salient lesson that when quoting facts and figures, particularly about Arctic sea ice, being clear and quoting sources is vital in order to avoid misunderstandings.

First though, the story so far:

Sunday's Observer magazine featured an article about bottom-baring Britpop icon Jarvis Cocker and his new role as frontman for the Greenpeace campaign to bring about a moratorium on exploiting the Arctic resources exposed by melting sea ice. Cocker, we are told, has become concerned about the fate of the Arctic since visiting it in 2008:

"Not that I'm a massive expert, but when I heard that they wanted to dig it up, I thought: hold on a minute - that's not good."

Right.

Beneath the Observer article is a Greenpeace-sourced list of '[c]hilling facts about the Arctic', begining with:

"Of the Arctic sea ice, 75% has been lost over the past 30 years. Last year saw sea-ice levels plummet to the second-lowest since records began."

This short statement caused a fair amount of discord in the blogosphere over the weekend, with climate skeptic bloggers WUWT and Haunting the Library disputing the figure and branding it a "total departure from both reality and common sense". On the other hand, climate modeller and blogger William Connelly seems to back the 75 per cent figure wholeheartedly.

So where has this figure come from? And is it right?

For a clue as to where it comes from, the 75 per cent figure, which was attributed to Greenpeace, also appeared in a Greenpeace blogpost back in February:

"If there's one fact to remember which underlines the urgency in protecting the Arctic it's this: in 30 years we've lost 75 per cent of the Arctic sea ice."

According to the blogpost, the 75 per cent figure comes from research into sea ice volume, and not the more commonly discussed 'sea ice extent' (which is roughly the same as 'area').

It's obviously more difficult to measure volume, where a lot of change happens below the surface, than extent, where satellite measurements of the Arctic can be used.

But sea ice volume is an important measurement, as we have written in the past, and as Greenpeace highlights in its blogpost:

"Some people compare Arctic sea ice amount by looking at the surface area of the ice and calculating how much it contracts by. But if you want to know how much of the ice is actually left, you've got to look at the volume - which is both the area and the thickness of the ice."

The Greenpeace blogpost links to the University of Washington's Polar Science Center website.

Faced with the difficulties of directly measuring sea ice volume, The Polar Science Center produces an assessment based on computer modelling - specifically something called the Panarctic Ice Ocean Modeling and Assimilation System (PIOMAS). The 75 per cent figure appears to be taken from the PIOMAS site, which reads:

"Monthly averaged ice volume for September 2011 was 4,200 [cubic kilometres]. This value is 66% lower than the mean over this period, 75% lower than the maximum in 1979, and 2.0 standard deviations below the 1979-2011 trend."

It's worth noting that this statement is somewhat ambiguous. Arctic sea ice goes through a yearly seasonal cycle, growing to a maximum volume in the cold winter months and shrinking to a minimum volume in the warm summer months; according to PIOMAS, the average sea ice volume changes from 28,700 cubic kilometres in April to 12,300 cubic kilometres in September.

Hence the PIOMAS website, which describes ice volume in September 2011 (a summer minimum) as 75 per cent lower than 'the maximum in 1979', could be read as comparing a minimum with a maximum, which would mean the 75 per cent figure was a misleading comparison that compares summer to winter.

On the other hand, it could mean a comparison between the biggest summer minimum volume, which occurred in September 1979, and the smallest summer minimum volume, in September 2011. In that case, noting that the second was 75 per cent smaller would be a reasonable comparison to make when discussing PIOMAS results.

To check, we contacted PIOMAS scientist Dr. Axel Schweiger. He explained that the 75 per cent figure was actually comparing minimum to minimum - it refers to a comparison between the September 1979 minimum ice volume and the September 2011 minimum ice volume.

So, as Dr. Schweiger told us:

"[W]hen referring to one particular measure of "Arctic sea ice loss" : PIOMAS ice volume, then that statement [the 75% figure from the PIOMAS website] is correct."

This is confirmed in the graph below, where the red line shows the PIOMAS average monthly arctic sea ice volume for September, and the blue line for April. Note that the September 2011 volume is only about a quarter of the September 1979 volume:

PIOMAS mean monthly arctic sea ice volume for April and September. Dashed lines parallel to linear fits represent one and two standard deviations from the trend. Error bars are estimated based on comparison with thickness observations and model sensitivity studies ( Schweiger et al. 2011) Source:  PIOMAS/Real Climate.

Picking two numbers exaggerates the noise

It seems to us that while the claim that there's been 75 per cent Arctic sea ice loss isn't out and out incorrect, as some bloggers have argued, that doesn't mean it's the most appropriate measure to highlight in a soundbite - for a few reasons.

First, as Schweiger told us, comparing just two single points in time rather than a trend 'exaggerates the noise'. Looking at the decadal trends in sea ice volume - i.e. an average of what the ice has been doing over each decade, rather than just a comparison of the start and end points - produces more confidence. According to PIOMAS, decadal trends suggest that over the past few decades sea ice volume has diminished by around 3,100 cubic kilometres per decade.

However, there's a significant degree of uncertainty in this measurement, of around +/- 1,300 cubic kilometres per decade. It's this kind of uncertainty that might suggest caution with presenting the PIOMAS figures, which are only one bit of scientific evidence, as fact.

PIOMAS might be bang on the money, but as WUWT and Haunting the Library point out, other scientists appear to disagree - Met Office Chief Scientist Julia Slingo is quoted in a Guardian article from earlier this year disputing the 75 per cent figure:

"[Slingo] said that suggestions the volume of sea ice had already declined by 75% already were not credible. 'We know there is something [happening on the thinning of sea ice] but it's not as dramatic as those numbers suggest.'
"The problem, [Slingo] explained, was that researchers did not know the thickness of Arctic sea ice with any confidence. [Slingo] hoped a new ice-monitoring satellite launched in 2010, Cryosat2, would help with more accurate measurements."

Sea ice: extent, area or volume?

Sea ice volume is a useful measurement of how the Arctic is changing, and it's important to remember that the sea ice is changing beneath the surface when assessing its state of health. In his email to us Schweiger highlighted:

"It is important to distinguish between ice-extent, area, and volume. Because the ice has shrunk in extent and thinned, the loss in volume is greater than in extent alone."

There can be very little doubt that Arctic sea ice is in decline. The Arctic is losing its older sea ice, leaving younger, thinner ice in its place. So a measure like sea ice volume, which includes this thinning, is probably more representative of the changes taking place in the Arctic than sea ice extent, which doesn't include this thinning.

And, as Schweiger told us:

"No matter how you slice it, there is very solid evidence from both observations and modeling that shows that the Arctic Sea ice has not only shrunk in extent but also thinned, thus yielding a loss in ice volume."

But ultimately, this example indicates that complicated scientific conclusions don't survive the process of being reduced to soundbites particularly well.