The space-born geodetic temporal Mean Dynamic Topography (MDT) is obtained from the difference of altimetric Mean Sea Surface (MSS) $h$ and the geoid height $N$. With the geostrophic surface currents obtained from its gradient the MDT is an essential parameter when discribing the ocean dynamics. Spectral consistency of $h$ and $N$ is crucial to minimize MDT errors. Usually, $h$ is globalized to allows for a Spherical Harmonic (SH) analysis and small scales beyond maximum degree and order (d/o) resolved in the geoid are cut-off. However, the usual globalization causes ocean-land steps in $h-N$ and spectral inconsistencies of $N$ and $h$ over land. To overcome both issues a new methodology is proposed based on globalization of the MDT. A Laplacian smoother with the coastal MDT values as boundary condition is applied resulting in a flat surface over land and a continuous ocean-land transition. The new methodology strongly reduces Gibbs effects and the need to work with high resolution MDTs to minimize them. Reduction of resolution is tested to reduce MDT uncertainties caused by the commission error expected to increase whith decreasing scale. Applying drifter data and a high resolution hydrodynamic ocean model it is shown, that for the Gulf Stream and the Kuroshio geodetic MDTs applying recent combined geoid models contain physical information up to at least d/o 420 (48km spatial scale). Since for oceanic regions with strong gradients in $N$ still inconsistencies between the geoid and the MSS exist, it depends on application/region if a higher resolution MDT is needed.

The effects of temporal changes in the marine geoid on estimates of the ocean dynamic topography is being investigated. Influences from changes of land hydrology, ice sheets, Post-Glacial Rebound (PGR), and ocean and atmospheric dynamics are considered and the associated crustal deformation is included. The strongest signals are associated with the seasonal cycle caused by changes in terrestrial water content and ice sheets as well as the redistribution of atmospheric mass. Second to this is the importance of an overall trend caused by PGR and decreasing ice sheets over Greenland and Antarctica. On long spatial scales, the amplitude of regional trends estimated for the geoid height have a sizable fraction of those from Sea Level Anomaly (SLA) for the period 1993–2016, also after subtraction of steric height of the upper 1000m to analyze trends in deep ocean geostrophic currents. The estimated strong negative geoid height trend south of Greenland for the period 1993–2016 opposes changes in dynamic sea level for the same period thereby affecting past studies on changes of both the strength of the Subpolar Gyre based on SLA and the meridional overturning circulation on a section between Cape Farewell and Portugal applying ocean dynamic topography and hydrography. We conclude that temporal geoid height trends should be considered in studies of (multi-)decadal trends in sea level and circulation on large spatial scales based on altimetry data referenced to a geoid field.