The recently completed NCEP Climate Forecast System Reanalysis (CFSR) from 1979 to 2009 used the NCEP coupled atmosphere-oceanland surface-sea ice system. This paper describes the sea ice concentration data used and how sea ice concentration is assimilated in the CFSR. The record minimum of Arctic sea ice in September 2007 is clearly shown in the CFSR output. Because of the realistic sea ice distribution and other improvements in the CFSR, it is expected that the coupled CFS reanalysis will improve in many aspects compared to the previous NCEP/NCAR Reanalysis-1 and NCEP-DOE Reanalysis-2. The improved surface air temperature in the fall is shown in this paper.
Sea ice is known to play a significant role in the global climate system. The accuracy of sea ice coverage is essential for the good performance of atmospheric and oceanic data assimilation models over the polar regions in the CFSR. Global climate model studies showed that sea ice concentration strongly affects the climate in the Antarctic regions (e.g., Simmonds and Budd, 1991; Simmonds and Wu, 1993). However, no sea ice concentration was presented in the previous NCEP reanalysis (NCEP/NCAR Reanalysis-1 (R1) and NCEP-DOE Reanalysis-2 (R2)) although sea ice concentration data from analysis were used to present the sea ice coverage in R1 and R2 with 55% cutoff (i.e. when sea ice concentration is greater than 55% it is considered as 100% sea ice coverage). The NCEP new coupled CFSR (Saha et al. 2010) allows us to add sea ice concentration from analysis into the reanalysis system which would provide a close sea ice-atmosphere link that should provide the main climate patterns and trends that include the polar regions. This paper discusses the sea ice data used in the CFSR and how sea ice concentration is assimilated.
