| description abstract | Extratropical weather systems are an essential feature of the midlatitude climate and global circulation. At the last glacial maximum (LGM), the formation of regions of high transient activity, referred to as ?storm tracks,? is strongly affected by the presence of large ice sheets over northern America and Scandinavia and by differences in sea surface temperature (SST) distributions. In the framework of the Palaeoclimate Modelling Intercomparison Project, simulations of the LGM climate have been run with a wide range of atmospheric general circulation models (AGCMs) using the same set of boundary conditions, allowing a valuable comparison between simulations of a climate very different from the present one. In this study, the authors focus on the storm track representation in the models and its relationship with the surface temperatures, the mean flow, and the precipitation. Storm tracks are described using transient eddy diagnostics such as mean sea level pressure variance and three-dimensional E vectors, computed from daily output. It is found that the general response to the changes in boundary conditions from present day to LGM is consistent for all models: they nearly all give an eastward shift for both storm tracks, with a larger shift for the Atlantic one. This is intrinsically linked to changes in stationary waves, which is also studied using the E vector diagnostic. Differences between the models reside in the value of the shift of the storm tracks and the change in their amplitude, which the authors analyze in terms of differences in resolution and parameterizations in the models. The sensitivity of the storm tracks to the sea surface temperatures and sea-ice extent are also examined by comparing the differences between prescribed and computed SST simulations. All in all, it is the eastern part of the storm tracks that is found to be most model-dependent, which relates to differences in the simulated climates over America?s west coast and Europe, and has to be taken into account when analyzing GCM climate simulations. | |
