2011), emphasizing the urgent need to incorporate climate change into available decision support systems (DSSs). The DSS Nest (http://nest.su.se/nest) developed in the MARE program (http://www.mare.su.se) is today the only scientifically-based tool available to support the development Trichostatin A of cost-effective measures against eutrophication for the entire Baltic Sea (Wulff et al. 2001, Savchuk & Wulff 2007, 2009). The Nest has been used to set the targets of the Baltic Sea Action Plan (BSAP, http://www.helcom.fi/stc/files/BSAP/BSAP_Final.pdf); however, the Nest

does not take the effect of climate change (e.g. changing hydrography) into account. In this study the first steps towards a DSS are described, which considers the combined effects of changing climate and changing nutrient loads on the Baltic Sea ecosystem. For this purpose a hierarchy of existing state-of-the-art, regional sub-models of the Earth system is applied (Figure 1). The atmospheric forcing for these regional sub-models is provided by an RCM, the Rossby Centre Atmosphere Ocean model (RCAO; Döscher et al. 2002), driven with boundary data from scenario simulations for the 21st century of Global Climate Models (GCMs). In these downscaling experiments, GCMs provide lateral boundary data and sea surface temperature (SST) and sea ice data for all sea

areas of the model domain except for the Baltic Sea region, Proteasome inhibitor where atmosphere and ocean sub-models are interactively coupled. Compared to earlier scenario simulations for the Baltic Sea, summarized by the BACC (2008), the downscaling approach Interleukin-3 receptor is novel because 1. time-dependent (transient) scenario simulations from the present climate until 2100 are performed instead of selected time slices for present and future climates (e.g. Räisänen et al. 2004), Results from GCM scenario simulations described in the fourth Assessment Report of the Intergovernmental Panel on

Climate Change (Solomon et al. 2007) are used as lateral forcing for RCAO. The DSS is built on the confidence of the models’ capacity to simulate changing climate in the Baltic Sea region. By comparing the observed and simulated present climate, the predictive skills of the models are assessed and model uncertainties are quantified. We investigate the quality of atmospheric surface fields over the Baltic Sea from an ensemble of 16 RCM simulations recently performed at the Swedish Meteorological and Hydrological Institute, SMHI (Kjellström et al. 2011). Our approach is to select two out of eight available GCMs and two greenhouse gas emission scenarios to minimize the computational burden of the DSS simulations based upon the following criteria: 1. The downscaled atmospheric surface fields should have sufficiently high quality during the present climate to force coupled physical-environmental Baltic Sea models.