2006: warmest year on record in the U.S.

The United States recorded its warmest year ever in 2006, according to today's report issued by the National Climatic Data Center (NCDC). The 2006 annual average temperature was 55�F, 2.2�F (1.2�C) above the 20th Century mean and 0.07�F (0.04�C) warmer than the previous warmest year, 1998. The NCDC had estimated that 2006 would be the 3rd warmest year in U.S. history last month, but an unusually warm December pushed 2006 to the top. It was the warmest December on record in the Northeast U.S., and the 4th warmest December for the country as a whole. Only 1939, 1957, and 1933 had warmer Decembers. However, the statistics partially hide the extraordinary warmth that began on December 10 and continued until January 6, when New York City tied their all-time record January high temperature of 72�. During the month ending January 6, the Northeast was 14 �F above average, and the U.S. as a whole was 7� above average.



No cause for alarm?
"No cause for alarm. Enjoy it while you have it," said Mike Halpert, head of forecast operations at the National Oceanic & Atmospheric Administration's Climate Prediction Center, in a story run by CNN just before New York City's record warmth. The story continued, "The weather is prone to short-term fluctuations, and forecasters said the mild winter does not necessarily mean global warming is upon us. In fact, the Plains have been hit by back-to-back blizzards in the past two weeks." True, the weather across most of the U.S. has finally cooled off this week, and the rest of January should have near average temperatures. And I agree that one warm month of winter in one country in its warmest year in 112 years of record keeping is not evidence of global warming, particularly when there is a moderate El Nino episode going on. An El Nino can lead to significantly warmer winters in the U.S.--exceptional December warmth has also occurred in 1877, 1939, and 1957, all of which were moderate or strong El Nino years. I've plotted up a comparison of temperatures in December of 1957 vs 2006 (Figure 1), and one can see that the unusual warmth of December 2006 does have historical precedent. Taking a look at average U.S. December temperatures for all years in the historical record (Figure 2), we see that these temperatures do show quite a bit of noise, and there is no evidence of dramatic warming in the past 30 years.


Figure 1. Comparison of the departure of average temperature from normal for December 1957 (the the second warmest December on record in the U.S.) and December 2006. Image credit: NOAA.

Figure 2. Average December temperatures for the U.S. from 1895 to 2006. Image credit: National Climatic Data Center.

Rolling thirteens with the weather dice
Take a look at the trend December temperatures in Figure 2. It shows that the average temperature has warmed a little more than 1� F in the past century. It may not seem like much, but that is enough to significantly load the dice in favor of warmer winters. Six of the ten warmest U.S. winters on record have occurred in the past 15 years. Month long spells where winter is seemingly absent--as also occurred in January 2006, the warmest January in U.S. history--have become more common. Keep in mind that the weather of January of 2006--which blew away the previous record for warmest January by a huge margin (2� F)--occurred during a La Nina year, not an El Nino. What concerns me most is that the warming trend is not isolated to the U.S. The 1� F rise in temperatures the past century has occurred world-wide, thanks to global warming, and the temperature increase has been much higher in the Arctic--something the climate models have predicted would occur as a telltale sign of the human-caused addition of greenhouse gases to the atmosphere. In the past, an exceptionally warm winter month in the U.S., like December 1957 (Figure 3), was offset by much cooler weather elsewhere, such as we see in Alaska, Greenland, and northern Siberia. However, December 2006 had no such offsetting cool temperatures--it was more than 1� C above average over almost all the land areas of the Northern Hemisphere north of 40� north latitude (Figure 4). Colorado, whose three blizzards have been widely cited as evidence that winter has been severe elsewhere, still recorded temperatures about 1� C above normal in December 2006.


Figure 3. Global departure of temperature from average for December 1957, the second warmest December on record in the U.S. Note that the exceptionally warm temperatures over the U.S. are offset by much cooler weather elsewhere, such as in Alaska, Greenland, and northern Siberia.

Figure 4. Global departure of temperature from average for December 2006. Note that the almost the entire globe north of 40� north latitude was more than 1� C above average, with large areas more than 6� C (11� F) above average.

All this unusual heat in the northern high latitudes is going to significantly slow down the formation of ice over the Arctic Ocean this winter. Furthermore, the lack of the usual snows across the Arctic may allow the snowpack to melt much earlier than normal in spring, resulting in more record warmth in the Arctic this summer. Arctic sea ice coverage, already down 20% in the past 20 years, is likely to continue to shrink in 2007. As sea ice melts in response to rising temperatures, it creates a positive feedback loop: melting ice means more of the dark ocean is exposed, allowing it to absorb more of the sun's energy, further increasing air temperatures, ocean temperatures, and ice melt. The observed changes in the ice cover (Figure 5) indicate that this feedback is now starting to take hold, and the weather dice will continue to get more loaded towards rolling higher numbers in 2007. I do think we're due for a cold winter next year--part of the warmth of the past two winters is probably due to the normal random fluctuations in the weather, and Nature has been rolling twelves more often than snake eyes of late. However, we're not going to see snake eyes too much more. December's weather in the Northeast U.S. may have been a case of the weather dice coming up thirteen--weather not seen on the planet since before the Ice Age began, 118,000 years ago. The weather dice will start rolling an increasing number of thirteens in coming years, and an ice-free Arctic Ocean in summertime by 2040 is a very real possibility, as indicated by computer modelling studies published in the Journal of Geophysical Research last month. This possibility is cause for alarm, and I, for one, had a lot of trouble enjoying the phenomenally warm weather of the past month here in Michigan.


Figure 5. Percent change in coverage of Arctic sea ice in Decembers from 1979-2006, compared to the 1979-2000 average. The Polar Ice Cap has shrunk by about 15% in December, and 20% in summer, over the past 20 years. Image credit: National Snow and Ice Data Center.

Check out the realclimate.org post on this winter's anomalous warmth.

I'll be back Thursday afternoon or Friday with a look at the status of El Nino. Will it still be around during hurricane season?

Jeff Masters