Convergence of the UK OSGM05 GRACE-based geoid and the UK fundamental benchmark network

Abstract

The results of this study suggest that a traditional height network based on fundamental benchmarks can be used meaningfully to assess the quality of a satellite-derived geoid model. We compare the pre- Gravity Recovery and Climate Experiment (GRACE) OSGM02 and the recent OSGM05 geoid models with the datum surface inferred by the UK fundamental benchmark network combined with network monument ellipsoidal heights derived from GPS data. OSGM05 incorporates long-wavelength gravity field data from GRACE to degree/order 90. Switching from OSGM02 to OSGM05, the mean difference between the leveling datum and the geoid model reduces from -0.129 m to -0.004 m, and RMS differences reduce from 0.110 m to 0.048 m. The mean tilt between the height datum and the geoid model in the north-south direction reduces from 34 mm to 12 mm per degree of latitude. Hence we demonstrate a systematic convergence between the geoid models and the leveling datum surface in terms of mean offset, RMS scatter and tilt without any fitting of one surface to the other. The implication is that the leveling datum surface represented a more accurate model of the geoidal geometry compared to OSGM02 in terms of form (in both short- and long-wavelength features) and surprisingly also in terms of its offset from the GRS80 ellipsoid. While the latter point can be no more than a coincidence, the former indicates that the UK leveling network, based on its fundamental benchmark network, can be used to test the accuracy of geoid model improvements derived from missions such as GRACE, with implications for the forthcoming Gravity Field and Steady State Ocean Circulation Explorer (GOCE) mission. Copyright 2008 by the American Geophysical Union.