Reducing cross-track geoid gradient errors around TOPEX/Poseidon and Jason-1 nominal tracks: Application to calculation of sea level anomalies

Abstract

A new technique is developed and tested to correct for cross-track geoid gradients in altimeter data. The proposed method is based on direct estimations of geoid variations around nominal tracks and on knowledge of ocean signal variability. Apart from measurement errors, ocean variability is demonstrated to be the major source of error in cross-track geoid estimations using altimeter measurements. The method thus uses the outputs of multimission ocean signal mapping procedures to improve the estimation of geoid features. A detailed error analysis shows that such a technique allows reduction of the estimation error by a factor of 2. Therefore, the method is applied taking advantage of the unprecedented TOPEX/Poseidon mission length. It provides a gain of 50%, in terms of sea level anomaly (SLA) variance reduction, in the cross-track geoid gradient correction used in collocating the repeat-cycle data. It also improves the estimation of altimetric mean profiles. From this study, local mean sea surface estimates can be inferred and applied to present and future altimetric missions, since they can be easily updated using more data. New altimetric missions like Jason-I and Envisat, with the same ground track as the former TOPEX/Poseidon and European Remote Sensing Satellite (ERS) missions, make the method even more relevant.