Classical determination of the earth's gravity field from the viewpoint of modern geomathematics (Q734084)

The topic of the present work is the investigation of the canonical connection between classical global gravity field modeling according to the concepts of Neumann and Stokes and modern local multi-scale calculation using local-compact adaptive wavelets. Especially the zoom-in method to find geoid heights by local gravity anomalies and disturbances is explained. First, the basic equations of the klassical theory of the gravity field are derived. The phenomena of gravity anomaly and gravity disturbance (becoming more interesting in connection with new GPS-methods) are discussed. The Neumann boundary condition problem at known gravity disturbances is explained. As data of gravity acceleration of sufficiently high resolution are available only locally, the related potential of gravity disturbances cannot be found with high enough accuracy. The authors show that using a multiscale approximation using local-compact Neumann wavelet kernels, locally a high accuracy with respect to the description of density and distribution of disturbances is obtained. Further, the Stokes boundary problem for known gravity anomalies is solved. A regularized global presentation of the potential of disturbances is obtained. It is shown that an integral under which Stokes wavelet functions occur has -- as in the case of the Neumann wavelet functions -- to be integrated not over the whole sphere, but, in dependence on the scale, only over a cap of the sphere. Further, Green vector wavelets are introduced. Finally a local multiscale reconstruction of the potential of the disturbances is performed in the Neumann case. The Stokes case follows analogously, substituting the Neumann kernel by the Stokes kernel, and the gravity disturbances by the gravity anomalies. Applying the presented theory, global and local zooming-in results to determine the potential of disturbances are presented. It is concluded that also using a multi-scale technique, as done in the present work, the global potential of disturbances cannot be found with very high resolution solving a classical boundary value problem. But the resolution may be increased adaptively according to the volume and precision of available data. More global data might be observed using satellite technique (CHAMP, GRACE, GOGE), but there other methods of data evaluation have to be introduced.