Saturday, August 13, 2011

Assessing the Arctic-toc of the countdown clock for northern sea ice

I stumbled upon another media mangling the reporting of the results of a study involving climate science. It happens regularly and is not surprising, but it still earns an eye-rolling sigh.

In this case the issue is the decline in Arctic sea ice and how the decline will not necessarily be continual and steady. But in some corners of the interwebs we get left with only the idea of "more arctic sea ice in the next decade", which is merely cited as a possibility, while the firm conclusion that "there’s no escaping the loss of ice in the summer" slips out of sight.

Let us start with the paper itself, an entry in Geophysical Research Letters, Kay, J. E., M. M. Holland, and A. Jahn (2011), Inter-annual to multi-decadal Arctic sea ice extent trends in a warming world, Geophys. Res. Lett., 38, L15708, doi:10.1029/2011GL048008.. Right there in the abstract we find the conclusions (1) that the Arctic melt of recent decades (through 2005, thus not inlcuding subsequent low years like current extrema leader 2007) is about half internal variability and (2) that in their modelling results the natural variability even "[i]n a warming world" can "until the middle of the 21st century" produce periods of 2-20 years with increases in Arctic sea ice.

Now let us jump to the NSF (National Science Foundation is the sponsor of the scientists employer the National Center for Atmospheric Research) press release, which includes quotes from the authors. A press release can easily be a first step in muddling the message of what research concludes. The release is significant because it may serve as basically the only source for media reports.

Here is the first paragraph and a chunk in the middle (emphasis mine):
Despite the rapid retreat of Arctic sea ice in recent years, the ice may temporarily stabilize or somewhat expand at times over the next few decades, new research indicates.
...
the NCAR research team found that Arctic ice under current climate conditions is as likely to expand as it is to contract for periods of up to about a decade.

"One of the results that surprised us all was the number of computer simulations that indicated a temporary halt to the loss of the ice," says NCAR scientist Jennifer Kay, the lead researcher.

"The computer simulations suggest that we could see a 10-year period of stable ice or even a slight increase in the extent of the ice.

"Even though the observed ice loss has accelerated over the last decade, the fate of sea ice over the next decade depends not only on human activity but also on climate variability that cannot be predicted."

Kay explains that variations in atmospheric conditions such as wind patterns could, for example, temporarily halt the sea ice loss. Still, the ultimate fate of the ice in a warming world is clear, she says.

"When you start looking at longer-term trends, 50 or 60 years, there's no escaping the loss of ice in the summer."

Again, the very start and very end of the above summarizes the results - shorter periods may show Arctic sea ice not declining, but over the long term the trend is clearing downward. In general that is absolutely unsurprising, but the unexpected result encountered by the authors is the quantity of instances in their simulations with temporary halts in the ice decrease. There is note of a generic 10-year such halt, but the bolded bit that says "next decade" invites the misinterpretation that the authors are particularly emphasizing that ice loss may well halt over literally the next decade from now rather than the following decade from any arbitrary time in coming years.

Next let us move on to some items found by the Google and its news aggregator. I will post the source and headline (with link). Remember with headlines that they may not come from the article/post writer and that whoever writes them the idea is to draw eyes and not necessarily provide a fully accurate summary.


For the most part when you actually read them these articles do a good job of matching what was in the press release, often using much content directly. But look at the range of impressions provided going down the list, from the idea of a slowdown or pause in Arctic melting to a stop of melting or even that more ice is imminent. The press release page itself would probably fall in the middle of that range with its headline, "Arctic Ice Melt Could Pause in Coming Decades".

And this is all from people (presumably) trying to get the story right! It is quite easy to find denialist commentors using this result to push their same empty nonsense, like that this shows climate is only driven by natural cycles, that climate scientists are grant whores trying to have it both ways so they cannot be proven wrong, etc. But then those people will use anything or even nothing to spout that.

To conclude, let us summarize the situation. The Kay et al. paper (this research, not all, as some contends we may not go much longer before seeing a summer-ice-free Arctic) says that any particular stretch of 10 years or so through the middle of this century, like say 2024-2034, has a good chance of not having a negative trend in Arctic sea ice due to natural variability in the system. There remains no question that as the world continues to warm, the trend is for less Arctic sea ice across the whole of this century.

No comments:

Post a Comment