| This thesis deals with the use of spectral methods for the post processing of airborne vector gravity data. In general, two aspects are discussed: (1) filtering and combining airborne measurements for the estimation of gravity disturbance vector components with higher accuracy; (2) the downward continuation of airborne gravity data to a level surface and to an arbitrary surface as well. Two spectral methods, namely input-output system theory and frequency domain least-squares adjustment, are discussed and examined for filtering and combination purposes. Numerical results show that, by filtering and combining all three components of airborne gravity vector rather than filtering one of the components alone, the estimation accuracy will be improved approximately by 10 percent for the vertical component and by 60 percent for the horizontal components within the wavelength range from 10 km to 100 km. In the downward continuation, more stable and accurate results for every component can be expected when it is combined with other components. |
