Mediterranean gravity field from the recent Satellite altimetry using Least Square collocation : analysis of analytical covariance functions
19/09/2016 | 17:45 | Session 2: Global gravity field modelling
Author(s): Adili Abulaitijiang, Ole Baltazar Andersen and Per Knudsen
Adili Abulaitijiang, Ole Baltazar Andersen and Per Knudsen
The CryoSat-2 satellite altimetry has improved along track resolution and range precision with its SAR mode. The 369 day repeat cycle also gives dense across track coverage making it possible to retrieve short wavelength components of the gravity signal. All the gravity field observables can be associated with the anomalous potential (T) by a linear functional. The covariance (kernel) functions of anomalous potential is then used to the derive covariance function of the observables through covariance propagation. Such covariance functions make up the base for the application of the least square collocation (LSC) in the modelling of gravity field, including the prediction of various quantities and its error estimate.
This paper is the first attempt on deriving the high resolution Mediterranean gravity field from 5 years of CryoSat-2 data testing various covariance functions. An initial investigation starts from the local covariance functions with local/spherical approximations fitted to the empirical covariance estimates. High resolution bathymetry data (e.g., GEBCO) is used to account for the strong correlation in the short wavelength gravity features with topography and bathymetry, which is necessary to comply with the assumption of homogeneity and isotropy in the LSC .The results will be compared to that produced from GEOCOL program which is using the global covariance model with a certain degree. The FFT method used to derive the DTU15 global marine gravity field will also be tested and results will be compared with that of LSC. The marine gravity is expected to resolve the ocean surface topography better closer to coasts with reasonable accuracy.