GPS for Geophysics: Glacial Isostatic Adjustment and Tests of Ionospheric Models

Doctoral thesis, 2006

This dissertation is divided into two parts that separately utilize high precision Global Positioning System (GPS) satellite observations for inferences of solid earth and ionospheric parameters. In the first part of the dissertation, solid earth movements in Sweden and Finland are observed with GPS at a precision of the order 0.1 mm/yr. GPS is the first technique that is able to resolve the horizontal movements that are related to the post-glacial rebound or---more generally---glacial isostatic adjustment in the region and the longest baselines are found to be expanding at a rate of the order 3 mm/yr. These results are used for inferences of the viscosity of the mantle of the earth as well as the thickness of the elastically behaving lithosphere without the trade-off between these two parameters that affects observations of the vertical components. The upper mantle viscosity was found to be $ 5\times10^{20} $ Pa~s and the lithospheric thickness larger than 90 km. The position uncertainties for northern stations in the observing network have earlier been found to be larger than in the south due to winter-time snow coverage of the receiving antennas. A fractal model was therefore used in an evaluation of the underlying time series in order to derive realistic uncertainties of the individual velocity estimates. The spectral power of the time series were estimated directly from a smoothed power spectrum using uncorrelated bins and a new transform which normalizes the power spectrum variance. In the second part of the dissertation, the impact of the proposed ionospheric model for the coming European global navigation satellite system Galileo is investigated. A non-physical equivalent sunspot number was used as an input parameter to optimize the global response of the model. Simulations indicate an ionospheric model misfit on single-frequency receiver position estimates generally less than 3 m, with better fits for mid-year months and intermediate to high latitude geomagnetic regions and worse fits for regions associated with the ionospheric equatorial anomaly. An investigation that exploited the multi-layered formulation of the model indicates an improvement with a model change for stand-alone users as well as for surveyors with network real time kinematic abilities.