Global vs. Regional Trends

A common problem in climate coverage

On January 1, the Daily Tech, an online magazine, published a somewhat misleading blog post about the “rapid recovery” of global sea ice, based on research released by the University of Illinois’s polar research group.

The research, based on satellite observations of the northern and southern hemisphere polar regions, didn’t get much pickup. And the Daily Tech story may be getting a bit stale now, but it bears mentioning because it illustrates a common mistake in science reporting: confusing the relevance of global versus regional climate trends.

The piece starts off accurately enough by reporting that:

Each year, millions of square kilometers of sea ice melt and refreeze. However, the mean ice anomaly — defined as the seasonally-adjusted difference between the current value and the average from 1979-2000, varies much more slowly. That anomaly now stands at just under zero, a value identical to one recorded at the end of 1979, the year satellite record-keeping began.

OK so far, but two paragraphs later the post suddenly flips to a confusing discussion of Arctic (northern hemisphere) ice levels. “Earlier this year,” wrote reporter Michael Asher, “predictions were rife that the North Pole could melt entirely in 2008. Instead, the Arctic ice saw a substantial recovery.”

That is patently misleading. The predictions Asher refers to were for a complete melt of summer sea ice, not of winter sea ice, to which the “substantial recovery” applies. And there was, indeed, a lot of melt last summer. According to the National Snow and Ice Date Center, “[2008] continued the negative trend in [Arctic] summer sea ice extent, with the second-lowest summer minimum since record-keeping began in 1979.” So far, it is not at all unusual for sea ice to make a full recovery during the winter. (The San Francisco Chronicle and The Associated Press succumbed to a similar, seasonal confusion two years ago.)

Worse still, Asher’s failure to include such key details leads the reader to believe that recent recovery has brought Arctic sea ice extent back to 1979 levels. That is simply not true. Asher is clearly mixing up global and regional trends. In response to his post, scientists at the University of Illinois’s polar research group issued a letter (pdf) clarifying the significance of their data:

Observed global sea ice area, defined here as a sum of N. Hemisphere and S. Hemisphere sea ice areas, is near or slightly lower than those observed in late 1979, as noted in the Daily Tech article. However, observed N. Hemisphere sea ice area is almost one million sq. km below values seen in late 1979 and S. Hemisphere sea ice area is about 0.5 million sq. km above that seen in late 1979, partly offsetting the N. Hemisphere reduction.

Although Asher does not mention climate change specifically in his post, his muddled coverage leaves the impression that the sea-ice data somehow refutes other evidence of global warming. Maybe that was not his intention, but scientists at the University of Illinois were clearly worried about that possibility. In their clarification letter, they noted:

One important detail about the article in the Daily Tech is that the author is comparing the GLOBAL sea ice area from December 31, 2008 to same variable for December 31, 1979. In the context of climate change, GLOBAL sea ice area may not be the most relevant indicator.

Much like droughts, floods, and other patterns of impact related to climate change, the extent of sea-ice melt varies by region. In other words, even if only one pole (the Arctic) is melting, we still have a problem. (And, in fact, referring only to the net ice gain in Antarctica belies the fact that its western half is warming and melting much more quickly that its eastern half.)

The relevance of global versus regional trends is confusing, however, because the applicability of either one depends on what the story is about. For example, last December, CJR and a number of other outlets blasted Politico for an article which reported that there is mounting scientific evidence for global cooling. The reporter cited data that shows only a meager rise in average temperature in the United States since 1930. Unlike the Daily Tech piece about sea ice, however, it was the global trend (i.e. in mean temperature) that Politico should have paid attention to. Unfortunately, the same kinds of mistakes are still far too common, as evidenced by a naïve column published last week in the Seattle Post-Intelligencer that suffered from all the same flaws as Politico’s piece. (It’s a mystery how this column ever made it past the P-I’s editors, who are usually more astute about environmental coverage.)

The only way for reporters (and columnists) to avoid making these mistakes in the future is to pay more attention to detail. When reviewing scientific data, it is imperative to note whether the data reflects trends that are global, regional, seasonal, annual, perennial, etc. Such nuance is as fundamental to good science journalism as an understanding of the difference between weather and climate.

Correction: This article was changed to reflect that the University of Illinois has no “Arctic Climate Research Center.” The researchers responsible for the sea ice data are informally known as the polar research group.

Has America ever needed a media watchdog more than now? Help us by joining CJR today.

Curtis Brainard writes on science and environment reporting. Follow him on Twitter @cbrainard.