Skip to content

Study Finds Antarctic Sea Ice Increases When It Gets Colder

August 17, 2013

By Paul Homewood




Some people have been tying themselves into knots to try to prove that the increasing extent of Antarctic sea ice is due to increasing temperatures.

Fortunately, some scientists still do proper work. In 2011, Shu et al analysed the relationship between surface air temperature and sea ice trends. They found,


Antarctic SIC trends agree well with the local SAT trends in most Antarctic regions. That is, Antarctic SIC and SAT show an inverse relationship: a cooling (warming) SAT trend is associated with an upward (downward) SIC trend.



Surface air temperature (SAT) from four reanalysis/analysis datasets are analyzed and compared with the observed SAT from 11 stations in the Antarctic. It is found that the SAT variation from Goddard Institute for Space Studies (GISS) is the best to represent the observed SAT. Then we use the sea ice concentration (SIC) data from satellite measurements, the SAT data from the GISS dataset and station observations to examine the trends and variations of sea ice and SAT in the Antarctic during 1979–2009. The Antarctic sea ice extent (SIE) shows an increased trend during 1979–2009, with a trend rate of 1.36 ± 0.43% per decade. Ensemble empirical mode decomposition analysis shows that the rate of the increased trend has been accelerating in the past decade. Antarctic SIE trend depends on the season, with the maximum increase occurring in autumn. If the relationship between SIC and GISS SAT trends is examined regionally, Antarctic SIC trends agree well with the local SAT trends in the most Antarctic regions. That is, Antarctic SIC and SAT show an inverse relationship: a cooling (warming) SAT trend is associated with an upward (downward) SIC trend.


Of course, this work not only debunks some of the fanciful arguments of warmists, it also suggests that temperatures in Antarctica have been falling since satellites started to monitor sea ice levels in 1979.



The study refers to both sea ice extent and concentration, so let’s have a look at concentration as well.

Concentration is well above normal in most areas, except for the Weddell Sea, which is, of course, the area where extent is also below normal.

  1. Brian H permalink
    August 17, 2013 9:03 pm

    Huda thunk it?

  2. David permalink
    August 18, 2013 4:01 pm


    The paper you’re referring to only considers *autumn* temperature and sea ice concentration trends. It makes it clear in the introduction that, overall, surface air temperatures in Antarctica have increased in recent decades.

    “If we take the Antarctic as a whole, we can find that SIE has a positive trend, and SAT also has a positive trend as noticed by Zhang (2007).”

    Zhang, et al (2007) clearly states in its abstract:

    “Estimates of sea ice extent based on satellite observations show an increasing Antarctic sea ice cover from 1979 to 2004 even though in situ observations show a prevailing warming trend in both the atmosphere and the ocean.”

    • August 18, 2013 6:08 pm

      And both extent and concentration were well above normal in the autumn.

      Are you suggesting temperatures were colder than normal this autumn, but warmer the rest of the year?

      They also say:-
      But if we examine SIC trends and SAT trends regionally and seasonally, we find that most of the Antarctic SIC and local SAT trends have an inverse relationship.

      In other words, although the correlation is strongest in Autumn, it is there most of the time.

      • David permalink
        August 18, 2013 6:33 pm

        You seem to have missed the point Paul.

        As I made clear (I hope) above, I was merely pointing to the fact that Shu et al. were specifically addressing *autumn* temperature trends, not overall temperature trends.

        While they point out that 6/11 temperature stations show falling autumn trends, they also clearly state in their introduction that the overall surface air temperature trend in Antarctica is upward (warming).

        They cite Zhang (2007) as their source for this, and Zhang is unambiguous: overall Antarctic surface air temperatures are rising over the longer term.

      • August 18, 2013 6:46 pm

        Wrong David.

        They say there is correlation for most regions/seasons, albeit not as strong as autumn.

        Interestingly they state

        But there are few studies investigating the Antarctic sea ice and SAT trends. One possible reason may be that some SAT datasets have large error in the Antarctic and thus it is difficult to study the relationship between sea ice trends and SAT trends.

        Perhaps we should pay more attention to what the ice is telling us, and less to a small number of temperature stations, that could be open to biases.

  3. dannyboy permalink
    January 3, 2014 11:43 pm

    Air temps having a relatively smaller heat capacity than water or ice would seem to have little effect on the overall temp in the antarctic. Would not the thermal storage oh heat be a more interesting subject to investigate?

    • January 4, 2014 12:01 pm

      Good point.

      There is so much “cold” stored in the ice, that it controls the atmospheric temperature to a large degree.


  1. Antarctic Sea Ice Increases When It Gets Colder! | sunshine hours

Comments are closed.

%d bloggers like this: