I spent four days in Providence RI for the Netroots Nation 2012 convention. I'm sure I would have had a better time if I wasn't attacked by some tainted tuna I had at a restaurant nearby on Friday evening (not affiliated with the hotel, but it was VERY ugly Friday night and Saturday morning, I can assure you). Whether I attend NN13 in San Jose, CA or not I'm not sure yet, but one thing I can tell you for sure is that the National Snow and Ice data center came out with its May 2012 report on Arctic sea ice extent, and since that report was issued there have been some interesting happenings as well.
But first, here's the picture I usually post near-real time from Barrow, AK the North Pole webcam launched near 90°N every April. It's not a victim of spending cuts (yet), probably because anything Arctic gives Republicans visions of oil wells dancing in their pointy little heads. Bless their hearts. No sign of open leads in the ice as yet. And the snow looks to be mostly pristine. But that sunshine and the clear sky portends change.
More below the orange squiggle.
Arctic Sea Ice Conditions to 17 June 2012
After a slow start in which mean April 2012 sea ice extent actually approached the normal 1979-2000 average (and nearly met it, late in the month), Arctic sea ice has been melting at a very rapid pace. This started at the end of April through about 2 May, then slowed to near normal ice melt for the rest of May.
This can be seen in the graphic below, which shows the mean and ±2 (90% of all years) standard deviations from the mean, for daily Arctic sea ice extent from 1979-2000. The record minimum sea ice year of 2007 is included for reference. Dates run from 1 March to 31 July, except for 2012 which ends at 17 June. Note the dramatic plunge since early June. More on that will follow.
How we got here:
Higher than normal sea ice coverage in the Alaska sector was a large part of the reason for Arctic sea ice extent higher than recent years during April and May 2012. This was the result of the coldest November 2011 and January 2012 in many years in this area (though the rest of the planet certainly was anything but cold for the most part). During much of this past spring, winds have been from a northerly direction in the Bering Strait (between Siberia and Alaska), which has done two things: slowed the melting of ice because of below-normal temperatures, and pushed the ice edge southward.
May 2012 overall had below normal sea ice extent, but not as low as some recent years; only the 12th lowest since 1979 and higher than 8 of the last 10 years. The May average sea ice extent from 1979 to 2012 can be seen below.
A big "sea-change" and a rapid drop in sea ice extent
The wind flow reversed at about mid-May and began moving ice from the Bering Sea into the Arctic Ocean basin. There's evidence it has also started pushing it away from the east Siberian coast soon. On the other side of the Arctic basin north of Eurasia, ice has been getting pushed out into the north Atlantic and melts in warmer waters of the North Atlantic drift current. The most recent wind regime has been responsible for rapid decreases in warm-season sea ice in recent years, especially during 2007, the year of record sea ice minimum. This abnormal wind pattern over the 30 days ending 15 June 2012 is shown next. The red arrows highlight the abnormal winds.
Some open water has appeared in the Arctic Ocean in other areas as well, as can be seen below in the spatial extent graphics from 12 June (right) and 17 June (left). The orange contour in the graphics shows the normal sea ice edge from 1979-2000 for the respective dates.
The most recent losses in Arctic sea ice have been most obvious on the North American side of the Arctic in Hudson Bay and the Canadian Archipelago. The seas north of western Siberia also continued to lose significant ice, after already having less ice than normal all last winter.
Other climate factors:
Temperatures have been above normal in the Arctic (except for Alaska) during Spring 2012. The temperature anomalies for Spring 2012 in the lower atmosphere (about 2-3,000 feet above ground) in the Northern Hemisphere poleward of 30° are shown in the next graphic. Note also the warm weather in central North America, which led to a record early spring greenup and damage to crops when there was (what used to be) a normal April frost and freeze.
The same general pattern in temperature has been present over the 30 days ending 15 June 2012 as well, but now with Alaska's north slope warming to above normal as well.
High pressure and fine weather has been found on average over the North American to European side of the Arctic for the last 30 days, which has increased the direct sunlight available to melt snow and ice. You can see the anomalies in sea level pressure at this link at the NOAA Earth Systems Research Lab (ESRL).
What does the near-future hold?
Weather and short-term climate
Since snow disappeared early in northern Siberia, and has long-vanished from AK all the way to the Arctic coast (in late May), much of the Arctic land has warmed earlier than normal. And once fully exposed, the open water at the Arctic coasts of AK and Siberia will also have an impact, as sunlight is absorbed by that water, rather than reflected by ice. These both are part of a positive feedback loop that is strongly affecting Arctic climate known as the albedo feedback, where the reflective ice and snow are replaced by much less reflective water and land surface.
NOTE: A special report was issued by the National Snow and Ice Data Center (NSIDC) on 19 June 2012 (this past Tuesday), making note that the current Arctic sea ice extent is the lowest in the satellite record since 1979. Temperature, sunlight/cloud cover, and winds will all matter for the next several months, and cannot be predicted with sufficient accuracy to determine how they will ultimately impact the September 2012 Arctic sea ice minimum. The NSIDC was careful to reiterate this:
While these patterns and conditions have looked similar to 2007, over the last couple days the high pressure pattern over the Beaufort Sea has broken down. And while the extent is at a record low for the date, it is still early in the melt season. Changing weather patterns throughout the summer will affect the exact trajectory of the sea ice extent through the rest of the melt season.
Arctic sea ice volume, not area, tells us the quantity of sea ice
To better assess the near-future for the Arctic sea ice, we need to know its area and thickness (i.e. the volume of sea ice to melt). We have a good bit of information on the thickness of the sea ice, both in surrogate form (age of ice; the older the thicker, generally) and in actual measurements (derived from satellite, and now plane overflight, measurements).
Below is the measurement of sea ice age for early May 2011, late September 2011, and early May 2012. The legend at the lower right hand side of the figure gives the age in years represented by each color.
There seems to be more 4th year ice this early May than last year, particularly on the Greenland/North American coastline. Note the 5+ year-old ice covers a somewhat narrower band (white color) along the coast this year, though it extends further east this year. There is also more 4 year-old ice (yellow) than last year at this time.
Another measurement was taken of the area coverage of multiyear sea ice within the Arctic basin at January 1 from 2000 through 2012. That measure has been steadily decreasing since the beginning of the plot as we can see below.
What about those overflight measurements? Arctic sea ice extent peaked last March and held steady at about 15 million km2 through the first 10 days of April. The graphics below show, from top to bottom, the thickness of sea ice determined from plane overflights (Operation "IceBridge") during most of this period (14 March to 2 April 2012). "IceBridge" is being done to bridge the gap between the loss of ice thickness remote sensing capability and the launch of a new satellite in a few years.
Snow depth is important because the snow acts to insulate the ice beneath from both solar radiation and warm air near the surface.
Arctic sea ice model
These kind of data is used to update models of Arctic sea ice volume like the University of Washington's Applied Physics Lab/Polar Science Center, using the Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS, Zhang and Rothrock, 2003). Other data used to determine the sea ice volume includes high resolution sea surface temperatures, observed and analyzed weather conditions, and so on. For the climate scientists out there, lots more details can be found at the link above.
The first graphic below from PIOMAS shows the linear trend in Arctic sea ice volume during the satellite record commencing in 1979. This trend shows a loss of about 3,000 cubic km of ice per decade, plus or minus 1,000. The dark and light gray shading about the trendline represents ±1 and ±2 standard deviations away from the trend, respectively. I've eyeballed in a curved red line representing what appears to be an acceleration in the loss of sea ice volume since about 1990.
The second graphic from PIOMAS shows actual Arctic sea ice volume: the 1979 through 2011 mean seasonal cycle and ±1 and ±2 standard deviations in black circles/line, dark gray shading, and light gray shading, respectively; and the 2007, 2010, 2011, and 2012 (through 30 May only) seasonal cycles.
While the mean minimum volume is about 13,000 cubic km, the last 3 years have had a volume of about 4,000 cubic km, about 30% of the mean. If we were to follow the non-linear trend in the anomaly graphic, September Arctic sea ice would essentially be gone by about 2015; that seems unlikely given that the last couple of years have seen some stabilization around what seems to be a "new normal", or at least a return to a more linear trend.
Arctic sea ice minimum forecasts for 2012
The Arctic Research Consortium of the U.S. (ARCUS) gathers sea ice minimum forecasts from scientists and other interested parties every year beginning 1 June. This year's June forecasts for the September 2012 minimum have been reported in the the Study of Environmental Arctic Change (SEARCH) website. The forecast summary is shown below. More details can be found at the link above. Methods used by those submitting include the use of dynamical ice models, statistical models based on present and past ice conditions, heuristic (i.e. "trial-and-error") models, and a mix of heuristic and dynamical models.
The range of forecasts is smaller than it has been in previous years, ranging from 4.1 million km2 to 4.9 million km2. And (in)famous global warming "skeptic" Anthony Watts from "wattsupwiththat.com" has come more into line with the others, unlike June 2011 (see below) when he missed the final value by about 900,000 km2.
I'll keep doing these types of diaries over the course of the summer until the sea ice minimum is reached. Stay tuned.