Friday, February 10, 2012

Arctic sea ice melt the second largest on record


Starting in 1979, NASA started using satellites to monitor sea ice cover in Arctic Ocean. Each summer and fall climatologists and other interested people (like me) look forward with interest, and in some cases anxiety, to see how much of the sea ice melts during a given calendar year.  I, for one, have been following this annual cycle of sea ice production and melt for many years.

Scientists at the National Snow and Ice Data Center at the University of Colorado, Boulder, collect and analyze the NASA data and provide near real time updates on the status of Arctic sea ice, among lots of other things, throughout the year.  You can access their analysis as well as many fine sources of raw data at their website: http://nsidc.org/index.html

Sea ice extent is defined by the NSIDC as the total area that has at least 15% of sea ice cover.  Arctic sea ice floats.  This means that wind, wave, and current action together with other physical factors cause the sea ice to be constantly on the move.  It forms pressure ridges, and is constantly shifting.  In the Spring, temperatures increase and ice starts to melt, crack, and move.  If wind is particularly strong it will drive the melting ice together and expose larger areas of open water, especially between the shore and the ice pack.  When wind action is less powerful the ice remains more spread out and the 15% ice cover extends over a larger area.

The pink lines on the map below shows the 1979-2000 average sea ice extent (minimum of 15% ice cover).  The white area represents the area actually covered by at least 15% sea ice for the minimum sea ice extent for 2011.  The 2011minimum sea ice extent is significantly lower than the historical average.  (All images in this posting are courtesy of the NSIDC.)

This year the minimum sea ice extent is the second lowest ever recorded. 


The graph below compares the 2011 sea ice minimum extent to the 1979-2000 minimum sea ice extent (dark gray line) for the months of June-September.  The 1979-2000 line presents a baseline value that is useful for comparing annual sea ice extents to the baseline as well as to each other.  The lighter gray area surrounding the 1979-2000 average indicates two standard deviations of the historical data.  This means that observations that fall outside of that light gray area are considered to be statistically different than the baseline value for that date.  Sea ice extents for the years 2007, 2008, 2010, and 2011, are also included in this graph.

The 2011 sea ice minimum reached its minimum in early September, and only 2007 had a smaller sea ice minimum.


The graph below shows the minimum sea ice extend for the month of September from 1979-2011.  You will note that there is considerable variation around the blue line that represents the overall trend.  Sometimes people who do not understand how natural systems work will look at a single month's observation, such as 1992 or 1996, and jump to the conclusion that Earth's climate is not warming after all.  This is a faulty way to think about this kind of data.  A single observation does not always indicate the overall trend of the system.  It is important to realize that Earth's climate, like every natural system, contains variability around the prevailing trend.

OK, here's an example of what I'm talking about with respect to variation around a trend line.   One natural system that everyone is familiar with is heart rate.  Just about everyone knows how to take their pulse.  If you were to take your pulse a doze random times throughout the day, every day for a year those data would reveal two things: 1) your average heart rate; and 2) the amount of variability that exists in your heart rate.  Heart rate data can also reveal longer-range trends if monitoring continues and lifestyle changes.  For example, let's say that you have not been all that active for many years, but you decide to start exercising.  For lack of another option you start to jog, and then run on a regular basis, say 3-4x/week.  After doing this for several months you are consistently running 12-20 miles a week.  If you have been monitoring your heart rate all this time you will probably have seen a drop in your average heart rate as you have gotten in increasingly better shape.  That improvement would be reflected in a lowering average heart rate.  At the same time, you would still see lots of variability in heart rate each day, depending on what you are doing.  That's characteristic of any natural system.

Because there tends to be a lot of variability (also called noise) in the data collected on natural systems, such as Arctic sea ice extent, climatologists recommend that when considering climate trends a data set of at least 30 years, and longer when possible, is needed to identify overall trends.  Sadly, the overall trend of sea ice extent reveals a pattern of increasing sea ice melt.


When we consider differences in prevailing conditions in 2007, the year with the smallest measured sea ice extent, and 2011, the year with the second lowest extent, it's notable that the sea ice extent got as small as it did in 2011.  Why was 2011 a surprise?  The figure below shows the prevailing direction and rate of sea ice movement during the 2007 and 2011.  In 2007 there was a combination of strong prevailing winds and surface currents pushing the ice toward the Canadian/Greenland margins of the Arctic Ocean.  The size of the arrows represent the rate of ice movement (larger and longer arrows mean faster movement).  Remember that sea ice extent is measured in terms of 15% sea ice cover.  This means that there can be significant amounts of water between ice floes and still be included in the area of sea ice cover.  In 2007 forces jammed the ice together, minimizing the amount of water between ice floes, and produced a smaller sea ice extent than ever seen before or since (so far).  By comparison, if you look at the map of sea ice movement in 2011, there was sea ice movement, of course, but the rate and direction of movement was nothing like what was observed in 2007.  Wind and current action did not tend to compact ice in the central Arctic Ocean in 2011 the way they did in 2007.  This means that the low sea ice extent in 2011 is due to a higher amount of sea ice melt rather than due to sea ice compaction.  


The chart below shows the sea ice extents for September for the years 2007-2011 along with the 1979-2000 average.  While the annual sea ice extent data are interesting, showing the 2011 sea ice extent being only 300,000 km2 larger than that of the record minimum in 2007, the most significant result of the analysis to me is that the overall rate of decline in sea ice extent is decreasing significantly.  The overall rate of decline, based on a rolling 10-year average, is now at 12% per decade.  That's a lot!  These data support the explanation that climate change is happening, and that the Earth is warming.


Before I quit, I want to mention one last thing about scientists and what they do.  They do not want the climate to change, sea ice to melt, polar bears to be at risk, or other consequences of climate change to occur.  What they do want is to try to understand patterns and processes that explain what is happening.  This holds true for all scientists in all fields.  Though, like anyone, they may have pre-existing notions of what they think is happening, scientists are not content to stop there.  They are driven to make observations, analyze data, and then find out what is actually happening.  They then work to find the best explanations for the observations.  Lastly, once scientists have developed their best explanation about what is happening and why, they present that idea to the larger scientific community for critical review.  Only the best ideas, those supported by data and appropriate methods of analysis, survive that review.  

So, what's the bottom line here?  The Arctic polar region is warming, and quickly.  As for the Antarctic, that's a topic for another time.

(Originally posted 10-6-2011)

No comments:

Post a Comment