Thursday, September 17, 2009

Arctic Sea Ice: Latest Figures

Regular readers of this blog will remember a series of posts (the three links are to three different posts) a few months back dealing with the question of whether NASA/JPL was lying when they claimed on a JPL web page that:

"The latest Arctic sea ice data from NASA and the National Snow and Ice Data Center show that the decade-long trend of shrinking sea ice cover is continuing."

I argued at the time that the latest data actually showed the shrinking to have reversed, although there was no way of knowing if that was more than a temporary deviation. The discussion got me into a correspondence first with someone at NASA who turned out to be a publicity person not a scientist, then with a scientist at NSDIC. It also resulted in a lot of comments on this blog.

Out of curiousity, I checked back today on the NSDIC web page, and found:

"The 2009 minimum is the third-lowest recorded since 1979, 580,000 square kilometers (220,000 square miles) above 2008 and 970,000 square kilometers (370,000 square miles) above the record low in 2007."

Or in other words, the extent of arctic sea ice has been increasing for the last two years, contrary to the claim I quoted above. The NSDIC puts the result in a way that emphasizes the fact that it is still below its long term level and obscures the fact that, for the past two years, arctic sea ice extent has been going up, not down. But at least they tell the truth about the facts.

To avoid irrelevant comments, I am not arguing for or against claims that the greater extent doesn't really count because it is thinner ice or that all the evidence taken together still supports long term shrinking of arctic sea ice. My claim is simply that the quote above, which is still up on the JPL web page, is false. When people lie to me about the evidence for their conclusions, offering other evidence that their conclusions are still true is not an adequate defense.

15 comments:

Tim Lambert said...

Since your earlier post, the Arctic sea ice extent anomaly has decreased by a considerable amount -- over one million square kilometres. Your earlier claim was that a decreasing trend had reversed itself. If you were consistent you would now be posting that that the trend had reversed itself again and that the anomaly trend had been decreasing since your post in May.

David Friedman said...

Tim accuses me of inconsistency. As should be clear from my post, I checked one source of information, and reported on what I found.

Checking the Cryosphere page, I observe that the anomaly has had a seasonal variation for quite a long time, presumably because summer and winter sea ice extent have been changing at different rates, and it continues to follow that pattern. The current anomaly looks, eyeballing the graph, higher than the anomaly for a year earlier, continuing the trend I initially pointed out.

In commenting on my first post of the series, Tim wrote:

"and because summer ice has been declining faster than winter ice there are now seasonal fluctuations in the anomaly."

What he described as seasonal anomalies then he now describes as a reversal of trend. I don't think I'm the one who is being inconsistent.

Since Tim didn't give the link to the Cryospher graph showing the anomaly, here it is; see for yourself:

http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.anom.jpg

Anonymous said...

I do minor work in aviation/land use consulting and I feel for the guy who published the "lie" because we're required to report certain "facts" that are ultimately too expensive to calculate (like the number and seating capacity of every "place of worship" within a such-and-such mile radius of the site). Sometimes, if we call up the "place of worship" and nobody answers, we leave it off the list, better for the client and faster for us--As long as we're careful not to miss the Crystal Cathedral or anything like that, the planning dept. usually doesn't care--this guy who posted the incorrect claim prolly had to summarize some minute data that he figured nobody would read (like I do a lot) and made a little mistake, then David Friedman comes down on him, sheesh.

David Friedman said...

I don't think the question of whether current data show the total extent of arctic sea ice increasing or decreasing is a little detail--the claim that it was decreasing was part of the point of the web page. It explicitly gave the ice and sea data center as one source--and their graphs show it increasing.

Mike Huben said...

The subject of the sentence David is criticizing is "decade long trend". For that to be false because of increases this year, David would have to misinterpret "decade" by a factor of 10.

I've seen David do this a few times: when he dislikes a statement, he declares it is wrong because it wasn't made the way he wanted it to be made. David wanted them to say shrinking has reversed somewhat this year. So he declares they're lying, trying to omit the fact that they're talking about a longer time frame. In the decade-long time frame, they are correct.

And of course it is misleading to insist on a short time frame: it's the difference between weather (short term) and climate (long term.) The subject is climate change.

David Friedman said...

Mike writes:

"The subject of the sentence David is criticizing is "decade long trend". "

And the verb is "is continuing."

Daniel [oeconomist.com] said...
This comment has been removed by the author.
Daniel [oeconomist.com] said...

I believe that all this illustrates a general observation, that most people will push aside what they otherwise acknowledge as protocols for arriving at a best shared approximation of the truth when they are convinced that they are in possession of some Greater Truth which is difficult or impossible to defend by those protocols.

At an extreme, people actively lie in defense of what they believe to be true. Or they may stop short of lying, but use misdirection to obscure inconvenient data or principles.

While some proponents and opponents of theories of anthropogenic global climate change are simply opportunists, I suspect that most of those who engaged in fudging of one sort or another have simply believe that they are employing "necessary evils" to protect Greater Truths.

Mike Huben said...

If the verb is "is continuing", the time period is then even more than a decade. It makes no sense in context to switch from climate to weather. Unless you want to make a hostile interpretation to claim somebody is a liar.

Daniel [oeconomist.com] said...

Mr Huber—

Let's look again at the original quote. It is not that the data can be reconciled with a trend of shrinking, it is that the data “show that the decade-long trend of shrinking sea ice cover is continuing.” (Emphasis mine.) One can say “X shows Y” and be lying even when Y is true. Dr Friedman has not said that the latest data show that the trend is not continuing; he has said that they very plainly do not show that it is.

Would you accept the same structure of argument (mutatis mutandis) on the part of your opponents that you are using here?

Anonymous said...

As one would expect from the fact that they are lying about the trend in area continuing, they are also lying about all the other stuff, such as thickness - the difference is that thickness is hard to estimate. We know that the trend of area diminishing is not continuing, but for all we know it might be well getting thinner - but what we do know is that their reasons for saying it is getting thinner are not true. Satellite measures of thickness suggest an increase, and ground level measures are only sampled in areas that the satellites say are thin.

Anonymous said...

Mike Huben's trend of idiotic commentary is continuing.

Anonymous said...

If someone has 999 facts correct and one wrong, do you immediately label them a "liar"? What's behind your obsession with this one web page?

Alex Perrone said...
This comment has been removed by the author.
Alex Perrone said...

I've liked the arctic sea issue. It is reflective of how we talk about climate, shows how the media can work (maybe even some scientists), and actually offers some hard numbers to cross-check.

Basically, as I interpret the English language, if a decade-long trend is continuing this year, then that means that the trend from this year alone (say from last year compared to this year, or maybe from the beginning of this year to the end of this year) is the same as the overall trend from the last ten years. That was not the case.

There is, however, an escape hatch, which is a perversion of this common sense approach. And that is: this year's data are such that it does not destroy the previous (negative) decade-long trend when you factor them in. In that case, the article was not lying, but it also has little meaning because it hardly narrows down what the data could be. It basically means this years' numbers were not extraordinarily positive, so positive that it would ruin the long-term negative trend.