The weight function in see more this case is equal to: w(i,j)=max0,R2−di,j2R2+di,j2,In addition, the parameter E used in the successive correction method was introduced. E2 is an estimate of the ratio of the observation error to the first guess field error. E was set to 0.5 (E2 = 0.25), which means that the satellite data are treated

as more accurate than the model data. However, they never have a weight equal to one. In the absence of this parameter (E2 = 0), the satellite data, if present at a particular location, would be given a weight of one. This means that the model data at this point would be omitted. The presence of E2 ensures that the model data are taken into account everywhere and ensures smoothing of the analysis product, which prevents possible instabilities. The product of assimilation is then used as the new initial state of the model from which

the new forecast is calculated. The current version of the 3D CEMBS (3D Coupled Ecosystem Model of the Baltic Sea) is based on the CESM (Community Earth System Model) developed at the National Center for Atmospheric Research. It was adapted for the Baltic Sea region as a coupled sea-ice model consisting of POP (The Parallel Ocean Program) and CICE (The Los Alamos Sea Ice Model). Atmospheric fields from the ICM (Interdisciplinary Centre for Mathematical and Computational Modelling) of Warsaw University are used to force the model together with historical data of river inflows. 71 main rivers are taken into account. All these components high throughput screening are coupled by a CPL7 (Coupler, version 7), which controls time and data exchange between these components. The model is configured in a horizontal resolution of 1/48 degrees and it is divided into 21 vertical levels. In the first half of 2013 the Cressman

Rho analysis scheme was used to implement satellite SST data assimilation. The data gathered in the SatBaltyk project were used as the source of satellite data. The aim of this implementation was to improve the model’s accuracy. The model and satellite data are complementary to each other as in the case of high cloud coverage over the Baltic Sea the model is the main source of data. The 3D CEMBS_A model is currently running in operational mode. This mode is split into two separate sub-modes. The regular mode produces 48-h forecasts using new weather forecasts from the ICM as forcing fields. The forecasts are produced on a regular basis every 6 h. The hydrodynamic part of the model produces sea temperature, salinity, current speed and direction, sea surface height, ice area cover and ice thickness (Dzierzbicka-Głowacka et al., 2013a). It also provides several biological, chemical and ecological parameters (Dzierzbicka-Głowacka et al., 2013b). Results are then stored in the local archive and posted to a model website. The parameters from the surface are interpolated to 1 km resolution, uploaded onto the SatBaltyk server and are available from the project’s website.