Estimation of Geopotential Differences Over Intercontinental Locations Using Satellite and Terrestrial Measurements
Loading...
Date
1991-08
Journal Title
Journal ISSN
Volume Title
Publisher
Ohio State University. Division of Geodetic Science
Abstract
An error analysis study was conducted in order to assess the currently achievable accuracies and the future anticipated improvements in the estimation of geopotential differences over intercontinental locations. Extending the ideas put forward by Colombo (1980), an observation/estimation scheme was proposed and studied, whereby gravity disturbance measurements on the Earth's surface, in caps surrounding the estimation points, are combined with corresponding data in caps directly over these points at the altitude of a low orbiting satellite, for the estimation of the geopotential difference between the terrestrial stations. The gravity disturbance data at altitude are inferred from GPS measurements made from the low orbiter to the high-altitude GPS satellites, in a multiple-high- single-low Satellite-to-Satellite Tracking (SST) configuration. The mathematical modeling required to relate the primary observables to the parameters to be estimated, was studied both for the terrestrial data and the data at altitude. Emphasis was placed on the examination of systematic effects and on the corresponding reductions that need to be applied to the measurements to avoid systematic errors. For the gravitational accelerations inferred from SST data, a mismodeling related to a centrifugal acceleration term was identified and corrected. Alternative formulations related to the sampling (or discretion) and the propagated errors arising in the truncation theory considerations were derived. Recurrence relations for the altitude generalized truncation coefficients implied by Hotine's kernel, and for the degree variances implied by a first-order Gauss-Markov covariance model were originally developed in this study. The error estimation for the geopotential differences was performed using both truncation theory and least-squares collocation with ring-averages, in case observations on the Earth's surface only are used. The error analysis indicated that with the currently available global geopotential model OSU89B and with gravity disturbance data in 2° caps surrounding the estimation points, the error of the geopotential difference arising from errors in the reference model and the cap data is about 23 kgal cm, for 30° station separation. This error is expected to reduce to about 12 kgal cm, when the lower-degree harmonics of the reference model are improved by the incorporation of the global GPS tracking data on Gravity Probe-B. The incorporation of gravity disturbance data at altitude was studied using least-squares collocation with ring-averages. It was found that for a low-degree (Nmax = 45) reference model, the data at the altitude of GP-B (600 km) can improve the geopotential difference accuracy by about 7%, as compared to the use of terrestrial data only. However, additional high-frequency observables at lower altitude are needed to achieve the results obtainable when a high-degree reference model is used, and to this end the gradiometer data from ARISTOTELES will provide a significant contribution.