The latest announcement of ice expansion in the Antarctic may have been a mistake, triggered by an error in processing satellite data. The earlier conclusion stated that ice in the Antarctic was expanding in the 1990's and first years of the 2000's as much as one-third as quickly as glaciers in the Arctic were retreating. Since that time, expansion of glaciers in the Antarctic has slowed, although coverage is slowly increasing.
Data collected by climatologists from 1991 to the mid 2000's were reprocessed in 2007. A team of researchers now believe a change in the way temperatures were measured could have introduced an error, leading investigators to believe ice levels grew during those years.
Another intriguing possibility is that the recalibration seven years ago eliminated an error, rather than cause one. If this were the case, then all the recent data - as well as the studies derived from it - would be called into question.
"This implies that the Antarctic sea ice trends reported in the... [2007 and 2013 assessment reports] can't both be correct: our findings show that the data used in one of the reports contains a significant error. But we have not yet been able to identify which one contains the error," Ian Eisenman from the Scripps Institution of Oceanography at University of California San Diego, and co-author of the study, said.
The Intergovernmental Panel on Climate Change (IPCC) released two reports, the Fourth Assessment Report (AR-4) in 2007 and its successor, the AR-5, in 2013. In the earlier reports, the international group announced ice levels between 1979 and 2005 were largely unchanged. The fifth report in the series reported annual increases in the levels of Antarctic ice during those 26 years, up to 6,350 square miles.
Data on ice coverage is not collected by a single instrument, but is patched together from several satellites and high-altitude weather balloons. Changes in some of these instruments may have led to a step-by-step change in readings, starting in 1991. All of this information has to be compiled using a formula, the most popular of which is called the Bootstrap algorithm. Eisenman and his team compared the AR-4 findings using one version of the algorithm to the AR-5 data, compiled with an updated formula for Bootstrap.
Researchers are unable to determine exactly where the error could be located due to the great natural variability in temperatures. There is too much "noise" in the data to pinpoint where an error may be found.
A report detailing how an error in the collection of satellite data could have led to the erroneous conclusion was published in the journal The Cryosphere.
