Reference
Belchansky, G.I., Douglas, D.C., Alpatsky, I.V. and Platonov, N.G. 2004. Spatial and temporal multiyear sea ice distributions in the Arctic: A neural network analysis of SSM/I data, 1988-2001. Journal of Geophysical Research 109: 10.1029/2004JC002388.
What was done
The authors "present a comparative evaluation of multiyear sea ice inversions of SSM/I brightness temperature data using different multiple layer perceptron neural networks that were constructed with learning data extracted from ERS synthetic aperture radar and Okean multiyear ice map products," after which they discuss the results they obtained and their several implications.
What was learned
Belchansky et al. report that from 1988 to 2001, total January multiyear ice area declined at a mean rate of 1.4% per year. In the autumn of 1996, however, they note that "a large multiyear ice recruitment of over 106 km2 fully replenished [our italics] the previous 8-year decline in total area," but they add that the monumental replenishment "was followed by an accelerated and compensatory decline during the subsequent 4 years." In addition, they learned that 75% of the interannual variation in January multiyear sea area "was explained by linear regression on two atmospheric parameters: the previous winter's Arctic Oscillation index as a proxy to melt duration and the previous year's average sea level pressure gradient across the Fram Strait as a proxy to annual ice export."
What it means
Belchansky et al. conclude that their 14-year analysis of multiyear ice dynamics is "insufficient to project long-term trends." Hence, they also conclude it is insufficient to reveal "whether recent declines in multiyear ice area and thickness are indicators of anthropogenic exacerbations to positive feedbacks that will lead the Arctic to an unprecedented future of reduced ice cover [the climate-alarmist position], or whether they are simply ephemeral expressions of natural low frequency oscillations [our position]." It should be noted in this regard, however, that low frequency oscillations are what the data actually reveal; and such behavior is not what one would predict from a gradually increasing atmospheric CO2 concentration. Hence, we continue to believe that the world's climate alarmists are truly "skating on thin ice" when they continue to claim what they do about this subject.