BALTSEM - a marine model for decision support within the Baltic Sea Region

A longstanding recognition of eutrophication as the most common threat to the entire Baltic Sea has led to the international agreement on nutrient load reductions within the Baltic Sea Action Plan. The nutrient load reductions were based on quantitative estimates of the “maximum allowed nutrient inputs” evaluated with a help of the decision support system (DSS) Baltic Nest developed within the MARE (Marine Research on Eutrophication) program. As demonstrated by a comparison to available data derived from observations, the marine biogeochemical model SANBALTS (Simple As Necessary Baltic Long-Term large-Scale) used in this evaluationis capable to realistically simulate both contemporary and pre-industrial trophic states of the Baltic Sea. A key to successful performance of SANBALTS lays in accounting for major sources and sinks that determine the size of internal nutrient pools and, thus, govern the large scale Baltic Sea eutrophication. Particularly, the most important phenomena that have to be reproduced by eutrophication models are a) spatial gradients of environmental conditions and limiting nutrients, b) interconnectivity of the major Baltic Sea basins, c) sporadic ventilation of the hypoxia prone deep-water layers with saltwater inflows, d) redox alterations of the coupled nitrogen and phosphorusbiogeochemical cycles, and d) nitrogen fixationby cyanobacteria.At the same time, both a ratherhigh aggregation of ecosystem variables (organic and inorganic forms of nutrients without explicit description of biota) and their correspondent spatial-temporal averaging (annual within homogeneous basins) implemented in SANBALTS make this model not appropriate enough for further revision and elaboration of the BSAP. Because such revision has also to take into consideration indicators required by the Marine Strategy Framework Directive (MSFD) of the European Union and characterized by higher spatial and temporal resolution, e.g.basin-wise winter surface nutrient concentrations and summer phytoplankton biomasses, and because of a necessity to factor in the possible effects of climate fluctuations, the appropriate model must simulate changes in ecosystem seasonal dynamics occurring over tens of years in response to both nutrient load reductions and climate changes. In principle, the continuing development of computing resources has made it feasible to implement for such purposes three-dimensional coupled physical-biogeochemical models with a relatively high resolution. However, with such models a simulation of the entire Baltic Sea over several decades still requires many days of computation even at super computer centers, which greatly hinders numerical experimentation needed for both model calibration and sensitivity analysis, including scenario responses. Therefore, there is a need for the model that is both reliable and convenient enough to be used for the revision of BSAP and implementation of MSFD,as well as for similar managerial tasks within an ecosystem approach. To serve this need, the model should be computationally fast for allowing multiple numerical runs necessary for finding and testing suitable distributions of the water-protection measures. Furthermore, for a building of credibility necessary in the national deliberations and international negotiations it should also be publicly accessible through the decision support system Nest allowing to any interested party running hindcast and scenario experiments as well as visualize its results. For these purposes, we present here the latest developments of the BAltic sea Long-Term large Scale Eutrophication Model (BALTSEM), which captures the main features of the Baltic Sea eutrophication, and now servesas a next generation marine model in the Baltic Nest system. These results as well as hindcast for 1850-2006 and future scenarios can be reproduced and analyzed on-line. Since BALTSEM performance at long-term scales has already been presented by Eilola et al. (2011a) and Gustafsson et al. (2012), this paper is especially focused on a seasonal scale