Interannuale und dekadische Variabilität im CO₂ Austausch zwischen Ozean und Atmopshäre - eine Modellstudie

This study investigates the mechanisms of North Atlantic-European climate using atmosphere general circulation models (AGCMs). Experiments with the AGCM ECHAM4, in which the sea surface temperature (SST) forcing is restricted to either the Atlantic or the Indo-Pacific oceans, show that both oceanic regions have an influence on North Atlantic-European climate in winter. In the experiment with SST forcing restricted to the Indo-Pacific oceans the atmospheric response projects on the North Atlantic Oscillation (NAO), while in the experiment with SST forcing restricted to the Atlantic Ocean the atmospheric response projects on the East-Atlantic Pattern. A multi-model intercomparison study shows that the region with the dominant influence on North Atlantic-European winter climate varies between different AGCMs. The dominant forcing of the atmospheric variability in the North Atlantic-European region in ECHAM4 is from the tropical eastern Pacific. However, in three other AGCMs the dominant forcing of this mode is from the tropical North Atlantic region. The importance of North Atlantic SST for North Atlantic-European climate is shown in another multi-model intercomparison study. The idealized North Atlantic SST anomaly pattern for this experiment has the structure of a tripole, which is believed to have the strongest impact on North Atlantic climate. Agreements between the responses in the different AGCMs are found concerning the NAO, Eurasian temperatures, rainfall over America and Africa, and the Asian monsoon. The results suggest that the extratropical North Atlantic region response is associated with remote Caribbean and tropical Atlantic SST anomalies, and with local forcing. All of these results support the conclusion that the ocean has a significant influence on North Atlantic-European climate. In addition to the mechanisms, this study investigates the predictability of North Atlantic-European climate. A control integration and ensemble experiments with the coupled atmosphere-ocean general circulation model (AOGCM) ECHAM5/MPI-OM are analyzed to investigate the decadal climate predictability. The ensemble experiments are realized with slightly perturbed atmospheric but the same oceanic initial conditions. The results show that the North Atlantic thermohaline circulation (THC) and SST are potentially predictable on multidecadal timescales. Over the ocean the predictability of surface air temperature (SAT) is very similar to that of SST, and this signal proceeds into the lower troposphere. Over land there is little evidence of decadal predictability of SAT with classical predictability methods. However, the estimation of the potential predictability over the European continent with probabilistic methods, commonly used in seasonal and medium-range forecasting, exhibits some limited success on decadal timescales. A multi-model comparison study with five AOGCMs confirms the potential predictability of North Atlantic THC and SST, albeit with skill levels dependent on the AOGCM. In general, models with greater decadal THC variability have higher levels of potential predictability