Coupled regional Earth system modeling in the Baltic Sea region

Abstract. Nonlinear responses to externally forced climate change are known dampen or amplify the local impact due complex cross-compartmental feedback loops in Earth system. These feedbacks less well represented traditional stand-alone atmosphere and ocean models on which many of today's regional assessments rely (e.g., EURO-CORDEX, NOSCCA BACC II). This has promoted development for Baltic Sea region by coupling different compartments system into more comprehensive models. Coupled realistically represent than information imposed prescribed boundary conditions and, thus, permit degrees freedom. In past, several coupled model systems have been developed Europe region. article reviews recent progress that allow two-way communication between models; land surface, including terrestrial biosphere; wave at air–sea interface hydrology water cycle closure. However, processes mostly realized one-way date, such as marine biogeochemistry, nutrient cycling atmospheric chemistry aerosols), not considered here. contrast uncoupled models, can modify mean near-surface air temperatures locally up compared with their counterparts using surface conditions. The representation small-scale oceanic processes, vertical mixing sea-ice dynamics, appears essential accurately resolve heat exchange over Sea, these parameters only be provided online high-resolution addition, ocean–atmosphere allows a explicit formulation microphysical temperature large-scale upwelling. Over land, important arise from dynamical vegetation changes implementation land-use scenarios afforestation/deforestation further alter albedo, roughness length evapotranspiration. Furthermore, good open sea areas is critical climatic extremes heavy precipitation, storms, tropical nights (defined where daily minimum does fall below 20 ∘C), appear sensitive coupling. For present-day climate, atmosphere–ocean atmosphere–land demonstrated added value single variables, particular when low-quality data were used respective model. makes prospective tool downscaling global because often large biases scale. close remains problematic, accuracy precipitation is, most cases, insufficient simulate runoff without bias adjustments. Many tuned suitably climatologies rather change. Therefore, research required how sensitivity models' response given temperature) affected spread altered multi-model multi-scenario ensembles ones.