In summary, in spite of some delay in filling one Postdoc position (FUB) due to a lack of suitable candidates, good progress has been made in SHARP-STC in most of the work packages. No serious problems were faced that would have led to major deviations from the proposed work schedule. All (transient and time-slice) simulations with the DLR and FUB CCMs and the MPI-M AOGCM were carried out and model data were provided to other SHARP projects as arranged. Analyses of stratosphere-troposphere coupling were carried out using observations and the time-slice and transient model simulations for the past and future. Mechanisms of STC were investigated. The coupling of the EMAC-CCM with the MPIOM ocean model has been successfully carried out; current test runs will provide a balanced atmosphere-ocean model system that will be applicable by the start of SHARP-II. The study of STC mechanisms using a SGCM (collaboration with E. Gerber) was not yet carried out following the recommendation of the reviewers. The analysis of STC in specific climate states will be finished by the end of Phase I. Similarly, the comparison of STC in different model configurations, started at MPI-M, will be extended to include the CCMs of DLR and FUB during the remaining Phase I period. This inter-comparison will be very important and exciting as the individual model analyses show a different behaviour of stratospheric warmings in the future, the reasons of will be investigated.

FUB studied the tropospheric driving mechanisms of the stratosphere by analysing the observed planetary wave forcing prior to the two major stratospheric warmings (MSWs) in January 2009 and 2010 (Ayarzagüena et al., 2011). Further, the downward influence of stratospheric dynamical anomalies after strong or weak vortex events was analysed in the EMAC REF-B1 and SCN-B2d simulations for the past and compared with NCEP/NCAR and ERA-40 observations. The different EMAC CCM versions showed a realistic occurrence frequency of MSWs, while the coupled atmosphere-ocean GCM EGMAM underestimates MSW occurrence. In a future climate, the model does not see an increase in the occurrence of MSWs, however a significant shift of MSWs towards the end of the winter. More-over, future weak vortex events seem to be more efficiently ‘blocked’ at tropopause level, while the influence of strong vortices at the surface is going to be more evident. The downward influence of the stratosphere was further highlighted in a study of Scaife et al. (2011) who show that future stratospheric circulation changes have the potential to modify tropospheric regional weather patterns, a result that could not be found in models with poor stratospheric resolution. Another focus at FUB was the development of the coupled atmosphere-ocean CCM EMAC-MPIOM as a preparation for studying the role of atmosphere-ocean feedback for stratosphere-troposphere coupling in Phase II of SHARP.

DLR was primarily concentrating on the dynamical coupling of the stratosphere and troposphere (STC), particularly the mechanisms of STC, based on transient REF-B1, SCN-B2c and SCN-B2d simulations with the CCM E39C-A. It was found that on average, the signal is propagating downward to the troposphere during all situations, but there is a high case-to-case variability. Comparison of the strength of the signal in the stratosphere to the strength of the signal in the troposphere suggests that the strength of downward propagation is independent from the strength of the signal in the stratosphere. In addition, the duration of propagation of the signal from the stratosphere to the troposphere seems to be independent from the strength of the stratospheric perturbation as well. Applying a new method showed that significant differences between events with enhanced and weakened downward propagation are found in the troposphere, whereas the stratosphere is not distinguishable. It was shown for the first time that in case of the events with enhanced downward propagation (weak and strong events) significant changes of the tropospheric jet in the mid-latitudes, of tropospheric wave generation and dissipation were found, already two to three months before the geopotential height anomaly signal reaches the troposphere. It is shown, that the precondition for an enhanced impact of stratospheric perturbations on tropospheric weather and climate is already given around two to three months before the impact on the troposphere occurs.

MPI-M addressed the impact of the vertical resolution and the vertical extent of the model and performed the SEN simulations in three different model configurations: T63L47, T63L95 (both high-top) and T63L31 (low-top). It was found that the frequency of SSW events increases in all model configurations in the future. In the distribution of major SSW events over the northern hemispheric winter months, no systematic change was found. No significant differences in the downward propagation of the NAM index between different model configurations and climate states were found. A separation of the weak vortex events into downward propagating and not downward propagating events revealed that the majority of the events (70-80%) propagates down into the troposphere.