Simultaneous Estimation of the Gravity Field and Sea Surface Topography From Satellite Altimeter Data By Least‐Squares Collocation

Abstract

The signal content of Seasat altimeter data is evaluated with respect to magnitude and wavelength. Based on this information covariance functions associated with the different signals/errors have been determined. the covariance function associated with the sea surface heights contains covariances associated with the geoid, the stationary sea surface topography, and the time‐varying sea surface topography. These three components have been estimated from Seasat altimeter data in the Faeroe Islands region using least‐squares collocation. In the solution error covariances associated with errors in the orbits, corrections for instrumental, atmospherical, and tidal effects, and sea state related bias are taken into account. the accuracies of the solutions are evaluated by computing error estimates and error correlations. A special effort has been made to evaluate the correlations between the estimated geoid and the other signals/errors. the results show that geoid undulations may be estimated with an accuracy of 0.13–0.16m. Approximately 50 per cent of the error variance has a long‐wavelength signature, which is mainly correlated with errors in the estimation of the permanent sea surface topography. the short‐wavelength part of the error is mainly correlated with errors due to liquid water in rain and clouds. Free air gravity anomalies may be estimated with an accuracy of 8.8–11.7 times 10‐5 ms‐2. This error has a short‐wavelength signature which is mainly correlated with errors due to liquid water in rain and clouds. However, a comparison with gravity observations resulted in an agreement of 7.8 times 10‐5 ms‐2. Both the permanent and the time‐varying sea surface topography may be estimated with accuracies of about 0.21 m. Copyright © 1991, Wiley Blackwell. All rights reserved