Abstract
A comparison is made of seven Antarctic sea ice data sets developed since 1980, on the basis of techniques of analysis and inferred temporal variations. Navy-NOAA Joint Ice Center sea ice charts are the basic data for all seven studies, but techniques used to derive ice areas vary significantly between studies. Sources of variation include the choice of a single week to represent a month, the characteristic measured (i.e., latitude of ice edge or actual ice area—with or without polynya), and the sea ice concentration used to determine the ice edge. The resulting data sets tend to indicate similar long term trends between 1973 and 1982. However, the estimates of mean annual and mean monthly ice areas vary distinctly between studies. This variability is often explainable in terms of the different techniques of analysis, but in some cases is not (e.g., Chiu's apparent overestimation of ice areas). The differences identified between these analyses suggest that caution should be taken in applying or extending these data sets.
