This chapter assesses to what extent the factors causing global warming affect the Baltic Sea area. Summertime near-surface warming in northern Europe exceeds natural internal variability of the climate system, and the observed warming cannot be explained without human influence. Regional changes in extreme temperatures, growing-season length and timing of the onset of spring are consistent with the large-scale signal of a human influence (mainly greenhouse gases). Shifts in large-scale circulation in the Northern Hemisphere and precipitation changes in northern Europe and the Arctic have been detected to exceed natural internal variability, but the climate models used to assess these quantities seem to underestimate the observed changes. To what extent this discrepancy between simulated and observed changes also affects the attribution of regional warming to human influence is still a matter of debate. Other aspects of regional climate change including changes in storminess, snow properties, run-off and the changing physical properties of the Baltic Sea have not been formally attributed to human influence yet.
KeywordsRegional climate change Á Detection and attribution
IntroductionThis chapter assesses how the factors causing global warming (mainly anthropogenic greenhouse gases) affect climate in the Baltic Sea area. In contrast to the following chapters on the effect of anthropogenic aerosols (Chap. 24) and land-use and land-cover changes (Chap. 25), this chapter focuses on globally uniform or at least large-scale forcing including changes in greenhouse gases, solar irradiance and stratospheric volcanic aerosols.To demonstrate an external influence on the observed climate change, the concept of detection and attribution is often used. Formal detection and attribution imply (i) the demonstration that the recent observed change is different from natural internal variability-the detection-and (ii) the comparison of different combinations of external forcing and assessment of their relative contribution in explaining the detected change-the attribution (see also Annex 1). Such a framework has been successfully applied at the global and continental scale to detect and attribute anthropogenic nearsurface and upper-level warming, as well as large-scale changes in other climatic parameters (Hegerl et al. 2007a). At the regional scale, however, there are only very few formal detection and attribution studies available (see Stott et al. 2010 for a review of recent advances).Climate change detection and attribution at the regional scale is complicated by various factors. First, variability increases with decreasing area of aggregation, that is the influence of small-scale phenomena does not average out. This generally leads to a decrease in the signal-to-noise ratio of externally forced changes and thus reduces the detectability of regional climate change (Stott 2003;Zwiers and Zhang 2003). Second, model biases play a more important