On the spectral combination of satellite gravity model, terrestrial and airborne gravity data for local gravimetric geoid computation

One of the challenges for geoid determination is the combination of heterogeneous gravity data. Because of the distinctive spectral content of different data sets, spectral combination is a suitable candidate for its solution. The key to have a successful combination is to determine the proper spectral weights, or the error degree variances of each data set. In this paper, the error degree variances of terrestrial and airborne gravity data at low degrees are estimated by the aid of a satellite gravity model using harmonic analysis. For higher degrees, the error covariances are estimated from local gravity data first, and then used to compute the error degree variances. The white and colored noise models are also used to estimate the error degree variances of local gravity data for comparisons. Based on the error degree variances, the spectral weights of satellite gravity models, terrestrial and airborne gravity data are determined and applied for geoid computation in Texas area. The computed gravimetric geoid models are tested against an independent, highly accurate geoid profile of the Geoid Slope Validation Survey 2011 (GSVS11). The geoid computed by combining satellite gravity model GOCO03S and terrestrial (land and DTU13 altimetric) gravity data agrees with GSVS11 to ±1.1 cm in terms of standard deviation along a line of 325 km. After incorporating the airborne gravity data collected at 11 km altitude, the standard deviation is reduced to ±0.8 cm. Numerical tests demonstrate the feasibility of spectral combination in geoid computation and the contribution of airborne gravity in an area of high quality terrestrial gravity data. Using the GSVS11 data and the spectral combination, the degree of correctness of the error spectra and the quality of satellite gravity models can also be revealed.