Summary: | The exchange between angular momentum of the solid portion of the Earth and of the atmosphere plays a vital role in exciting small but measurable changes in the rotation of our planet. Recent improvements in geodetic data and atmospheric research has allowed for new insights into fluctuations of the Earth's rotation. Atmospheric general circulation models (GCMs) are able to simulate mass movements and mass concentrations on a global scale in a realistic way. Due to enormous changes in the mass distribution and motions relative to the rotating Earth the atmosphere has an important impact on Earth rotation parameters (ERPs). For this thesis ensemble simulations with the ECHAM5 GCM for the period from 1880 to 2006 have been conducted - driven by sea surface temperatures, sea ice concentrations, green house gas concentrations, aerosols, and solar forcing. For the purpose of investigating possible future long term trends in ERPs, A1B scenario runs from 2000 to 2060 were carried out with the coupled atmosphere-ocean GCM ECOCTH. A detailed structure analysis compares time series of observed ERPs and those deduced from atmospheric reanalyses with the atmospheric ensemble simulations. In general ECHAM5 is able to explain a larger proportion of the variance in observed ERPs than its predecessors. Fluctuations in Earth rotation linked to the quasi-biennial oscillation are not present in the simulations, as the model fails to generate this stratospheric mode. Variations in the length-of-day (LOD) induced by ENSO and the annual cycle are well comprehended by the simulation. As for reanalyses the amplitude of the semiannual cycle is significantly underestimated and only reaches about 66 percent of the observed one. An analysis of variance of the ensemble simulations indicates that the semiannual variations can not be explained by external boundary conditions to a large extent. They are primarily driven by internal processes. Densities of local atmospheric angular momentum (AAM) variations tend to be mainly driven by boundary ...
|