5 6 7

• Main Author: Circulation Forcing
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Subjects: Pacific; Sea ice
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.384.2225; http://www.cgd.ucar.edu/cas/cdeser/Docs/Deser_GRL_SuppMat_Nov162007.pdf

Empirical orthogonal function analysis of sea ice concentration anomalies Are the Sea Ice Concentration (SIC) trend patterns shown in Fig. 1 of the manuscript preferred structures of variability, or are they simply a result of dividing the record in half? To address this question, we have applied Empirical Orthogonal Function (EOF) analysis to the covariance matrix of SIC anomalies over the full period of record, using a separate EOF analysis for winter and for summer. The two leading EOFs in each season and their associated principal component (PC) time series are shown in Fig. S1. In each season, the first and second EOFs account for approximately 30 % and 17 % of the variance, respectively, and are well separated according to the criterion of North et al. (1982). In winter, the leading EOF exhibits out-of-phase variations between the eastern and western Atlantic and between the eastern and western Pacific, strongly reminiscent of the trend pattern during the first half of the record (see Fig. 1 in the manuscript). This EOF is nearly identical to that in Ukita et al. (2007) based on February-March averages over the period 1979-2003, and consistent with results obtained using data sets beginning