Published online by Cambridge University Press: 04 April 2001
The use of the SSM/I sea ice concentration data together with a sea surface temperature (SST) analysis in a numerical weather prediction (NWP) model is described. The ice concentration data provide a useful representation of the SST freezing-point isotherm while the SST field allows a quality control of the sea ice data. A land-sea mask, an orography field and a climatological SST field are also used to improve the analysis of surface temperature in some inland seas and lakes. A monthly mean sea ice concentration climatology has been created from 13 years of the ECMWF re-analysis ice concentration dataset which was also derived from satellite multi-channel microwave radiometer data. Data assimilation experiments have shown a small positive impact in the southern hemisphere model forecast scores when this new surface field is used in the NWP model.