Geodetic and oceanographic results from the analysis of 1 year of Geosat data

The first year of the Geosat Exact Repeat Mission has been analyzed to obtain improved orbits, gravity model, and representation of sea surface topography. The starting point for this study were the original geophysical data records merged with orbits by Haines et. al. (1990) computed with the GEM-T1 gravity model. A radial orbit error model following the development of Engelis (1987a) was modified to account for resonant effects above degree 36. Twenty-four 17-day arcs were analyzed. Separate 17-day and a single 1-year solution for a gravity model and a spherical harmonic representation of the sea surface topography to degree 10 were found. The root-mean-square crossover discrepancy with the a priori (i.e., GEM-T1) model was ±1.1 m, which was reduced to ±0.22 m with the 1-year solution. Comparison of overlaps in the arcs showed discrepancies of ±1.3 m with the a priori field which were reduced to ±0.07 m with the 1-year solution. The determination of the degree (1, 0) term in the sea surface topography was not possible in this analysis because of the high correlation of this term with an orbit error term. In addition, the (1, 1) terms showed serious incompatibility with that implied by historical oceanographic information. The final solution used here incorporated the degree 1 coefficients from oceanographic data. The problem with the (1, 1) coefficients may imply a small center of mass offset in the tracking station positions used in the orbit calculations or may result from the poor distribution of the available tracking stations. Error analysis indicated a ±10 cm sea surface topography commission error to degree 10. The geoid commission error to degree 36 was ±9 cm in the ocean areas. The new potential coefficient model was compared with terrestrial data, in a manner to that done by Engelis and Knudsen (1989) to show the improvement of the new model.