Quantifying the individual contributions of GOCE gradients for regional quasi-geoid modeling in a relatively rough topography

Abstract

Gravity field and steady-state Ocean Circulation Explorer (GOCE) is the first satellite mission that measures the gravity gradients (GGs) in space with primary objective of determining the gravity field of the Earth with maximum spatial resolution and accuracy. GOCE GGs have been used for the computation of global gravity field models (GGMs) represented by the surface spherical harmonics for about 10 years. However, they may potentially improve the regional quasi-geoid solutions computed from terrestrial gravity and a Global Geopotential Model (GGM) particularly in mountainous areas with rough gravity field. This study exploits the full signal content in the GOCE Level-2 GGs given in Local North-Oriented Frame for improved regional quasi-geoid modeling in Auvergne test area located in France. The least squares collocation method is used in single-input and single-output (SISO) mode where the input is the residual terrestrial gravity data corrected for GGM contribution and residual terrain modeling effects, and also in multiple-input and single-output mode (MISO) where the terrestrial data input is augmented with the four precise GOCE GGs or their combinations. The single outputs for both modes are the quasi-geoid heights at 75 GPS collocated leveling benchmarks distributed in the central area and at 7857 Digital Elevation Model (DEM) points covering the whole area. The combination of the terrestrial gravity with the GOCE along-track gradient, having a lower noise level than the others, gives the best results when a single gradient is used. The further incorporation of the off-diagonal gradient to the along-track gradient results in the best MISO solution showing improvements with respect to the SISO solution only at GPS/leveling benchmarks near a large data gap. Intercomparison of the best MISO and SISO solutions at the DEM points suggests substantial quasi-geoid differences up to 13.6 cm in areas of sparse or no gravity data near the Alps where these results could not be assessed due to the lack of GPS-leveling data.