Arctic sea ice at a record low again; a warmer February for the U.S. coming

Arctic sea ice extent for January 2011 was the lowest on record for the month, and marked the second consecutive month a record low has been set, according to the National Snow and Ice Data Center. Most of the missing ice was concentrated along the shores of Northeast Canada and Western Greenland. Relative to the 1979 - 2000 average, the missing ice area was about twice the size of Texas, or about 60% of the size of the Mediterranean Sea. Hudson Bay in Canada did not freeze over until mid-January, the latest freeze-up date on record, and at least a month later than average. The late freeze-up contributed to record warm winter temperatures across much of the Canadian Arctic in December and January. Bob Henson of the National Center for Atmospheric Research has a very interesting post on this, noting that Coral Harbor on the shores of Hudson Bay had a low temperature on January 6 that was 30°C (54°F) above average--a pretty ridiculous temperature anomaly. He quotes David Phillips, a senior climatologist with Environment Canada, who discussed the lack of ice near Canada's Baffin Island: "The Meteorological Service of Canada was still writing marine forecasts as of 7 January, well beyond anything we have ever done." Henson also writes:

"The extremes have been just as impressive when you look high in the atmosphere above these areas. Typically the midpoint of the atmosphere's mass--the 500-millibar (500 hPa) level--rests around 5 kilometers (3 miles) above sea level during the Arctic midwinter. In mid-December, a vast bubble of high pressure formed in the vicinity of Greenland. At the center of this high, the 500-mb surface rose to more than 5.8 kilometers, a sign of remarkably mild air below. Stu Ostro (The Weather Channel) found that this was the most extreme 500-mb anomaly anywhere on the planet in weather analyses dating back to 1948.

Farther west, a separate monster high developed over Alaska in January. According to Richard Thoman (National Weather Service, Fairbanks), the 500-mb height over both Nome and Kotzebue rose to 582 decameters (5.82 km). That's not only a January record: those are the highest values ever observed at those points outside of June, July, and August."


Figure 1. Monthly January sea ice extent for 1979 to 2011 shows a decline of 3.3% per decade. Image credit: National Snow and Ice Data Center.

The warm temperatures in Canada and record sea ice loss in the Arctic were also due, in part, to a strong negative phase of the Arctic Oscillation (AO). The Arctic Oscillation and its close cousin, the North Atlantic Oscillation (NAO) are naturally occurring pressure patterns in the Arctic and mid-latitudes. A negative AO and NAO results when we have weaker than normal low pressure over the Arctic, and weaker than normal high pressure over the Azores Islands. This fosters an easterly flow of air off the warm Atlantic Ocean into the Canadian Arctic, and also weakens the winds of the polar vortex, the ring of counter-clockwise spinning winds that encircles the Arctic. A weaker polar vortex allows cold air to spill southwards out of the Arctic into eastern North America and Western Europe. Thus, the strongly negative AO and NAO the past two winters have been largely responsible for the cold and snowy winters in these regions, and exceptionally warm conditions in the Arctic. I described this pattern in more detail in my December post titled, Florida shivers; Hot Arctic-Cold Continents pattern is back. It is possible that Arctic sea ice loss is largely responsible for the unusual Arctic Oscillation pattern we've observed during the past two winters, as well as for the record-strength ridges of high pressure observed over Greenland and Alaska this winter. It should not surprise us that Arctic sea ice loss would be capable of causing major perturbations to Earth's weather, since it is well known that changes from average in sea surface temperatures over large regions of the ocean modify the jet stream, storm tracks, and precipitation patterns. The El Niño and La Niña patterns are prime examples of this (though the area of oceans affected by these phenomena are much larger than what we're talking about in the Arctic.) Another example: Feudale and Shukla (2010) found that during the summer of 2003, exceptionally high sea surface temperatures of 4°C (7°F) above average over the Mediterranean Sea, combined with unusually warm SSTs in the northern portion of the North Atlantic Ocean near the Arctic, combined to shift the jet stream to the north over Western Europe and create the heat wave of 2003, the deadliest heat wave in history with 30,000 - 50,000 deaths in Europe.

References
Feudale, L., and J. Shukla (2010), "Influence of sea surface temperature on the European heat wave of 2003 summer. Part I: an observational study", Climate Dynamics DOI: 10.1007/s00382-010-0788-0


Figure 2. The 6-10 temperature forecast issued by NOAA's Climate Prediction Center calls for an above-average chance of warm temperatures across most of the U.S. by mid-February.

A warmer forecast for February
Over the past two weeks, the Arctic Oscillation has undergone a major transition, changing from negative to positive. This means that low pressure over the Arctic has intensified, which will act to speed up the counter-clockwise spinning winds (the polar vortex.) This spin-up of the polar vortex will tend to keep cold air bottled up the Arctic, leading to more Arctic sea ice formation and warmer winter conditions over the U.S. This warm-up is reflected in the latest 6 - 10 day temperature outlook from NOAA's Climate Prediction Center (Figure 2.) Could it be the groundhog was right, and we have only three more weeks of winter left? Time will tell--we have little skill predicting what may happen to the Arctic Oscillation more than about two weeks in advance.

Jeff Masters