Numerical simulation of the North Atlantic Oscillation and its impact on the South West Asia

The impact of the North Atlantic Oscillation (NAO) on meteorological parameters in the South West Asia (SWA), especially on temperature and precipitation, is certain according to recent studies. Due to differences in methods for analyzing the effects of the two phases of NAO, however, there are differences among researchers on the details of the impact. Perhaps one reason for this inconsistency is the weakness of the impact due to the fact that the NAO occurs far upstream of the SWA.  So the interaction with other stronger forcings makes it almost impossible to extract the net effects of NAO. Hence some of the previously published results are in doubt. More certain results can be obtained using numerical models when we are able to do control experiments by fixing certain desirable parameters of the atmosphere and changing the forcings of interest at the same time.
In this article the general climate model ECHAM5 is employed to simulate the NAO and its impact in the SWA. Two experiments named "control experiment" (COEX) and "climatological experiment" (CLEX) are designed to verify the results previously reported by the authors in 2008. In the COES, the actual sea surface temperature (SST) and in the CLEX, the long-term mean of SST is used as an input of the model. Artificially changing the NAO index is the aim of CLEX which is done by changing the lower boundary condition of the model. For both experiments, the NAO index is computed on the basis of the method suggested by Hurrell in 1995 using sea level pressure values in the northern and subtropical regions of the North Atlantic Ocean. Then on the basis of the COEX NAO index, the critical positive and negative months of NAO are determined and the ensemble mean of some important meteorological parameters in these two groups of months are computed and analyzed.
The simulation of the NAO index shows that the ECHAM5 is unable to simulate the real atmospheric values of the monthly NAO index. In particular, the positive trend of the NAO index from 1970 onwards is not predicted by ECHAM5. However this model is successful in simulation of the low-frequency variability of the NAO index. The latter findings show that the high-frequency variability of the NAO index is effectively controlled by forcings other than the SST. However it seems that the SST acts as a main forcing for low-frequency variability of the NAO index.
Analysis of the ensemble mean of meteorological parameters in the critical positive and negative months shows that the patterns obtained for COEX are similar to those reported by the authors in 2008.  Some small differences observed in location and magnitude of centers could be due to the differences in the type and resolution of the data. As mentioned above, the critical months are selected here on the basis of the NAO index computed for COEX. So the selected months are not critical in the CLEX. If we examine the NAO index for CLEX in the same critical months, the result is like a random distribution.  Hence, no significant difference is observed between the ensemble mean of meteorological parameters in CLEX in the critical positive and negative months. Therefore, it can be concluded that the results reported by the authors in 2008 are related to the real impact of NAO in the selected months.  Finally, since in the experiments carried out here only the boundary conditions and forcing functions related to the numerical representation of the atmosphere are present, it should be possible to explain the observed impact of NAO in the SWA by means of the dynamics represented by ECHAM5.