Atmospheric circulation regimes and climate change

Sammanfattning: The Earth's atmosphere is expected to warm in response to increasing atmospheric concentrations of greenhouse gases (GHG). The response of the Earth's complex and chaotic climate system to the GHG emissions is, however, difficult to assess. In this thesis, two issues of importance for the assessment of this response are studied. The first concerns the magnitude of the natural and anthropogenic emissions of CO2. An atmospheric transport model is used, combined with inventories of anthropogenic CO2 emissions and estimates of natural emissions, to compare modelled and observed variations in the concentration of CO2 at an Arctic monitoring site. The anthropogenic and natural emissions are shown to exert approximately equal influence on Arctic CO2 variations during winter.The primary focus of this thesis is the response of the climate system to the enhanced GHG forcing. It has been proposed that this response may project onto the leading modes of variability. In the present thesis, this hypothesis is tested against the alternative that the spatial patterns of variability change in response to the enhanced forcing. The response of the atmospheric circulation to the enhanced GHG forcing as simulated by a specific coupled global climate model (CGCM) is studied. The response projects strongly onto the leading modes of present-day variability. The spatial patterns of the leading modes are however changed in response to the enhanced GHG forcing. These changes in the spatial patterns are associated with a strengthening of the waveguide for barotropic Rossby waves in the Southern Hemisphere. The Northern Hemisphere waveguide is however unchanged.The magnitude of the global mean responses to an enhanced GHG forcing as simulated by CGCMs vary. Moreover, the regional responses vary considerably among CGCMs. In this thesis, it is hypothesised that the inter-CGCM differences in the spatial patterns of the response to the enhanced GHG forcing are partially explained by inter-CGCM differences in zonal-mean properties of the atmospheric flow. In order to isolate the effect of these differences in the zonal-mean background state from the effects of other sensitivities, a simplified model with idealised forcing is employed. The model used is a global three-level quasi-geostrophic model. The sensitivity of the stationary wave pattern (SWP) to changes in the zonal-mean wind and tropopause height of similar magnitude as those found in response to the enhanced GHG forcing in CGCMs is investigated. The SWP in the simplified model shows a sensitivity of comparable magnitude to the analogous response in CGCMs. These results indicate that the CGCM-simulated response is sensitive to relatively small differences in the zonal-mean background state. To assess the uncertainties in the regional response to the enhanced forcing associated with this sensitivity, ensemble simulations of climate change are of great importance.