We develop and explore numerical codes rooted in the fundamental theories of fluid dynamics to aid geodetic and geophysical data analysis and interpretation for the benefit of an improved understanding of the System Earth. Central to our work are ocean general circulation models configured to cover global or regional domains that represent processes as ocean tides, self-attraction and loading feedbacks to the ocean dynamics, or surface pressure forcing in addition to the usually considered atmospheric fluxes. We further use numerical models of the terrestrial water cycle that consider vertical water and energy exchange with the atmosphere, and lateral water transport in the soil and through the river network.

In this topic we concentrate on joint analysis of gravity, seismic and other geophysical, geodetic and geological data aiming for construction of comprehensive models of the lithosphere and underlying mantle and finding a connection of these models with ongoing tectonic processes and geodynamics.

We use observations and numerical simulations to understand the dynamics of the large ice sheets of Greenland and Antarctica in the past and predict their dynamics in the future. A particular focus is on the mass balance of these ice sheets, since it has a direct impact on the global sea level. Interactions of the ice sheets with the ocean, the atmosphere and the solid earth play a key role in understanding their dynamics and for meaningful modeling.

In this topic, we consider the solid earth response to mass redistributions at its surface in relation to geodynamic and geophysical problems like climate, ocean and ice-sheet dynamics. The focus is numerical modelling of instantaneous to long-term processes of ten thousand years and more.

This topic systematically combines observations and numerical models of Earth's systems. Observations are interpolated in time and space under consideration of modelled physical laws, not directly observed quantities are estimated and numerical models are reinitialized to improve the forecasts. By using models, data assimilation can invert causal connections to find the sources of observed changes. Likewise we try to relate sensitivities of hardly observable quantities to observable quantities with the goal to derive the former by observing the latter. Closely related is the estimation and optimization of the expected gain of future observation campaigns.