RADIAL ORBIT ERROR REDUCTION AND SEA SURFACE TOPOGRAPHY DETERMINATION USING SATELLITE ALTIMETRY

A method is presented in satellite altimetry that attempts to simultaneously determine the geoid and sea surface topography with minimum wavelengths of about 500 km and to reduce the radial orbit errors caused by geopotential errors.;To provide statistical estimates of the radial distances and the geoid, a covariance propagation is made based on the full geopotential covariance. Accuracy estimates for the Seasat orbits are given which agree quite well with already published results.;Observation equations are developed using sea surface heights and crossover discrepancies as observables. A minimum variance solution with prior information provides estimates of parameters representing the sea surface topography and corrections to the gravity field that is used to the orbit generation. The potential of the method is demonstrated in a solution where simulated geopotential errors and the Levitus sea surface topography are used to generate the observables for a three day Seasat arc. The simulation results show that the method can be used to effectively reduce the radial orbit error and recover the sea surface topography.;The modeling of the radial orbit error is made using the linearized Lagrangian perturbation theory. Secular and second order effects are also included. After a rather extensive validation of the linearized equations, alternative expressions of the radial orbit error are derived. A Fourier series formulation allows for easier computations, examination of the frequency content of the error and computation of different statistics. A geographic representation with respect to geocentric latitude and longitude gives a significantly better insight on the spatial variations of the error. Similar expressions are derived for the geoid undulation error and sea surface topography. Numerical estimates for the radial orbit error and geoid undulation error are computed using the differences of two geopotential models as potential coefficient errors, for a Seasat orbit.