Comparison of potential coefficient determinations with 5° and 1° anomalies
Loading...
Date
1980-04
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Ohio State University. Division of Geodetic Science
Abstract
Potential coefficients can be derived from mean gravity anomalies, satellite data analysis, or a combination of both data types. In using gravity data, a choice of the size of the mean anomaly block to use must be made. Usually this has been 5° equal area blocks as opposed to smaller size blocks such as 1° x 1°. Tests are described in this paper that have been made using 1° x 1° anomalies and 5° anomalies that are consistent with the 1° data. Potential coefficients have been determined with just the anomaly data, and the anomaly data in combination with the GEM 9 potential coefficients. In the combination solution the average percentage difference between the solutions using the two anomaly sizes was 29%. The root mean square undulation difference was ± 1.1 m and the root mean square anomaly difference was ± 3.8 mgals. These differences are caused by the perturbation of the low frequency information by the high frequency information in the mean anomaly blocks. These differences suggest that for the highest accuracy, even if coefficients just to degree 36 are sought, 1° x 1° anomaly blocks should be used. We have also examined an approximate technique for the combination solution using 1° x 1° data that requires a significantly less amount of computer time than the rigorous solution. In comparing two 180 x 180 fields from a rigorous and approximate adjustment we found an average percentage difference of 9%; a RMS undulation difference of 0.8 m; and a RMS anomaly difference of 2 mgals.