by Robert Henson

As Andrew Freedman noted in last week's Undercast, the Arctic Ocean lost a staggering amount of sea ice this past summer. During the 1980s and 1990s, the typical summertime melt brought the Arctic's sea ice extent, as measured by the National Snow and Ice Data Center, down to about 6 million square kilometers. This summer the extent dropped to 4.13M sq km-about 1M below the previous record (set in 2005) and 2M below the 1979-2000 average.


Even as this saga of a soggy Arctic astounded both scientists and laypeople, a far different story was unfolding on the other side of the globe, where winter was in progress. In July, Buenos Aires, Argentina, got its first snowfall of significance since 1918, and in August Johannesburg, South Africa, saw its first snow since 1981. The cold intrusions feeding these rare snowfalls swept north from Antarctica, which was ringed by the largest amount of sea ice area measured since satellite records began in 1979: about 16.2M sq km, as noted by the University of Illinois' Crysophere Today website. (An important point to note: Cryosphere Today reports sea ice *area*, whereas NSIDC tracks sea ice *extent*. In a nutshell, the difference is that "area" omits large regions of open water within sea ice, counting only the ice itself, whereas NSIDC's "extent" includes the entire zone that has at least some significant ice, including the open water between ice floes. This difference isn't important for the comparisons presented below.)

The Antarctic freeze-up didn't get much attention from the mainstream press, but climate contrarians picked up on it quickly. The story got major play on such websites as climateskeptic.com and the Heartland Institute. Heartland claimed, "The record Antarctic sea ice and the revelation that recently shrinking Arctic sea ice is unrelated to global warming should have deflated sensationalist media reports, but global warming mania proved difficult to stop."

At first blush, it may look as though this year's trends in sea ice balanced each other out: a record low to the north, a record high to the south. But looking at things this way brings you perilously close to the "balance trap" discussed in my CapitalWeather post of July 29. In this case, the "balance" disappears upon closer inspection, as the departures from average were far greater in the north than in the south.

The Cryosphere Today site depicts the full 28-year record of sea-ice area for both north and south. Conveniently, it also tracks the combined total for both hemispheres, and that's where the "balance" argument falls apart. As you can see on the red trace at bottom, the anomaly in *total* global sea ice-the departure from the average for a given time of year-hit an all-time low around August 2007, dropping to about 3 million sq km below average. No previous negative anomaly had ever gone much below 2 million sq km. This record was driven almost entirely by the northern melt, which had reached hugely anomalous levels by August, whereas the southern freeze didn't reach exceptional levels till late September.

When you look at the separate 28-year records for Arctic and Antarctic ice, you'll see huge swings from summer to winter, as one would expect. Typically, the north loses about 60% of its ice area each summer, while the south loses a bigger fraction: almost 90%. That's largely because Antarctica sits atop the South Pole and spreads far around it. The ice that form around the continent's edge is thus a bit closer to the equator than in the north, extending into the Southern Ocean, and so it's thinner and more prone to summer melt. (See Wikipedia's entry on polar ice caps for a great explanation and graphics.) Since winters over much of Antarctica could remain as cold or colder than they are now for some years to come-in part because of circulation changes induced by ozone depletion-we can expect the relatively thin sheets of winter sea ice around Antarctica to remain impressive large, perhaps even setting more records.

Here's how the last few seasons have stacked up on average, in millions of square kilometers of sea ice *area* (a smaller value than *extent*), with the recent anomalies called out:

Sea ice area	Arctic	Antarctic
Winter	13-14	15-16 (2007: about 16.3)
Summer	4.0-5.0 (2007: about 3.0)	1.5-2.5

Note how much more summertime sea ice area there is in the Arctic than the Antarctic. That's a sign of multi-year or perennial ice: thick slabs, typically 3 to 4 meters deep and sometimes up to 20 meters deep, that persist across the Arctic Ocean over years or even decades. It's the loss of this perennial ice that is so disturbing, because the first-year ice that replaces it in winter is thinner and less resistant to melting, and the vicious cycle of Arctic melt will thus continue. (Open water absorbs more than 90% of the sunlight that hits it, whereas ice reflects up to 90% or more, so the more melt there is, the more heat the ocean gathers, which feeds even more melt.)

What made this particular summer so exceptional in the Arctic? There's no doubt that unusual weather played a big part. A huge, persistent dome of high pressure sat over the western Arctic, shunting multi-year ice toward Greenland and helping keep the western areas sunny and warm for weeks on end. Eventually, open water extended from Alaska and Siberia nearly to the North Pole for the first time on record. As noted by Julia Slingo and Rowan Sutton (UK National Centre for Atmospheric Science) in a letter to the journal Nature, the pattern was part of a set of global anomalies that included the La Nina event now unfolding. An October news release from NASA stated "the rapid decline in [Arctic] winter perennial ice the past two years was caused by unusual winds." This appears to be what prompted the Heartland Institute to claim that this summer's ice loss was "unrelated to global warming."

In fact, this NASA study wasn't debunking climate change: it was simply explaining the mechanism that moved the shrinking perennial ice to one side, like a meteorological broom, and opened the other side to more dramatic melting. There will always be proximate causes for the extremes we encounter as the climate warms. When Washington hits 109°F or 110°F in some future heat wave, it'll no doubt be blamed on a huge dome of high pressure. Will we pin that dome's formation on natural variability? Will we stick the blame on the heat-island effect, or will we point to fossil fuels as the main culprit? Or will we acknowledge that weather is always with us, even as it unfolds in a changing climate?

Meteorologist Robert Henson is a writer and editor at the University Corporation for Atmospheric Research, which operates the National Center for Atmospheric Research. He is the author of "The Rough Guide to Weather" (second edition, 2007) and "The Rough Guide to Climate Change" (2006; second edition to be released in February 2008). Henson is filling in for Andrew Freedman this week.