Error Characteristics of a Mean Dynamic Topography from Altimetry and GOCE Using full Covariance Information
Horvath, Alexander; Fecher, Thomas; Pail, Roland
Technical University Munich, GERMANY

Comprehensive error quantification and characterization of a Mean Dynamic Topography provides highly valuable information for a better understanding of the ocean and especially ocean currents.
This approach aims to employ covariance information not only from the gravity field (derived from GOCE gravity gradients) but also from the ocean surface height measurements from altimetry satellites in order to compute a Mean Dynamic Topography. The approach covers the propagation of the error from the two independent entities Sea Surface Height (SSH) and Geoid height (N) towards a Mean Dynamic Topography (MDT) following the basic equation MDT = SSH - N. Using this equation by subtracting SSH and N from each other requires the application of filters to ensure spectral consistency of the two entities SSH and N. This work embeds both, isotropic and anisotropic filters, and evaluates the advantages and disadvantages of the filter algorithms applied.
Subsequently the error propagation is carried further towards geostrophic velocities. The calculations are grid based for a regional test area (e.g. North Atlantic current, Gulf Stream). The initial covariance information for the geoid heights originates from a strict error propagation starting from the gravity gradients and is provided for all grid cells in the test area. For the SSH, the covariance information is provided as well for all grid cells in the test area and is derived from the level of single SSH measurements. In summary this work presents a stochastic model combining covariance information from gravity field and altimetry to derive a characterization of MDT and geostrophic velocity uncertainties.