Improving the space-domain methods of forward modeling for geodynamic studies by GRACE

The present and future satellite missions CHAMP, GRACE and GOCE make it possible to study from space the local and the regional geophysical signals. This implies the revision of the methods of forward modelling of the gravitational attraction generated by known sources; e.g. the topography of the Earth. The standard space-domain method of prism integration is inadequate for two reasons. Firstly, even though the gravitational field for two neighbouring field points in space is almost identical, the computations of the gravitational attraction by prism integration start from scratch for each field point. Secondly, it is difficult to include intra-source mass density variations except by increasing the number of prisms. In the proposed method, the gravitational field generated by the known source is pre-computed and stored in few fixed field points. These points are regularly distributed in space and cover (for some chosen area) the satellite orbit; e.g. the altitudes between 300 km and 500 km for GOCE. The stored field quantities are the coefficients of the power series expansion of the gravitational field in the vicinity of the fixed field point. The information about the particular generating source has been integrated once-for-all into these coefficients. There are many merits of the proposed method. For example, the flexible integration of the effects of arbitrarily complex intra-source mass density variations; the precise and continuous field description based on power series expansion; the explicit method of choosing the size of the field domain for a chosen accuracy of the computed field; the straight-forward description of the modelled field in global spherical or ellipsoidal geometry. The method is illustrated by the example of computing the gravitational field generated by the topography in the Nordic and Baltic countries for studying the post-glacial rebound in Scandinavia by the GRACE satellite.