Prediction skill in the Subpolar North Atlantic and Nordic Seas in the Norwegian Climate Prediction Model (NorCPM)

Temperature and salinity anomalies in the subpolar North Atlantic (SPNA) and Nordic Seas are considered important sources for predictability of Arctic ocean heat content, sea ice, and air-sea interaction on interannual to decadal time scales. In this work, the ability to predict the ocean state in the SPNA and Nordic Seas is investigated using different versions of the Norwegian Climate Prediction Model (NorCPM). We specifically compare one version with data assimilation of only sea surface temperature with another version with assimilation of both sea surface temperature and sub-surface hydrographic profiles. Preliminary results show an improvement of prediction skill when the data assimilation is applied both at the surface and in the subsurface ocean. For this version, the ensemble mean anomaly correlation coefficient for winter SST is 0.7 up to 3 years lead time over the time period 1983-2009 in the eastern part of the Nordic Seas. In the SPNA, the correlation is 0.7 in large areas up to 7 years lead time. Additional analysis will shed further light on prediction skill in the SPNA and the Nordic Seas – in particular on the inflow region to the Nordic Seas – and the role of different initialization methods.