A 30 year multi-sensor satellite SST product for the Arctic Ocean
Hoeyer, Jacob; Howe, eva; Tonboe, Rasmus; Dybkjaer, Gorm
Danish Meteorological Institute, DENMARK

This presentation will focus upon a 30 year data set of Arctic satellite sea surface temperature. The data set consists of daily Optimally Interpolated (OI) level 4 SST fields without gaps in a 0.05 degree resolution for the entire Arctic Ocean from 1982 to 2012. The OI method uses the A(ATSR) Re-analysis for Climate (ARC), the Pathfinder satellite data set and in situ observations from the ICOADS 2.5 data set. A dedicated high latitude bias correction method has been developed for adjust the Pathfinder AVHRR observations to the ARC data, and the effect of this correction will be assessed. In situ observations not included in the data set are used to give an estimate of the errors on the data, and indicate very stable performance with a overall bias of less than 0.1 degC and a standard deviation of 0.6 DegC. The error statistics in the Marginal Ice Zone, in data sparse regions and close to the coast will be presented. Initial analysis of trends and variability in the data set will be presented, as well as an improved climatology, based upon the entire data set. Finally, the 30 years of SST variability will be investigated for 3 selected regions: the Disko Bay, The Nuuk Fjord System and in the Labrador Sea.