Consistency of Geoid models, radar altimetry, and hydrodynamic modelling in the North Sea

J. Schall*, A. Löcher, J. Kusche, R. Rietbroek, A. Sudau

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

6 Citations (Scopus)


Radar altimetry, when corrected for tides, atmospheric forcing of the sea surface, and the effects of density variations and mean and time-variable currents, provides an along-track realization of the marine geoid. In this study we investigate whether and how such an ‘altimetric-hydrodynamic’ geoid over the North Sea can serve for validating satellite-gravimetric geoids. Our results indicate that, using ERS-2 and ENVISAT along-track altimetry and water levels from the high-resolution operational circulation model BSHcmod, we do find distinct differences in RMS fits for various state-of-the art satellite-only models (beyond degree 145 for GRACE-only, and beyond degree 185 for GOCE models) and for combined geoid models, very similar as seen in GPS-levelling validations over land areas. We find that, at spectral resolution of up to about 200, an RMS fit as low as about 7 cm can be obtained for the most recent GOCE-derived models such as GOCO05S. This is slightly above what we expect from budgeting individual errors. Key to the validation is a proper treatment of the spectral mismatch between satellite-gravimetric and altimetric-hydrodynamic geoids. Comparison of data fits and error budget suggests that geoid truncation errors residual to EGM2008 (i.e. EGM2008 commission and omission error) may amount up to few cm.

Original languageEnglish
Pages (from-to)223-237
Number of pages15
JournalMarine Geodesy
Issue number3-4
Early online date18 Apr 2016
Publication statusPublished - 3 Jul 2016
Externally publishedYes


  • Geoid models
  • hydrodynamic modelling
  • North Sea
  • radar altimetry
  • ITC-CV


Dive into the research topics of 'Consistency of Geoid models, radar altimetry, and hydrodynamic modelling in the North Sea'. Together they form a unique fingerprint.

Cite this