A data-driven approach to local gravity field modelling using spherical radial basis functions英文资料.pdf

J Geod (2008) 82:457–471 DOI 10.1007/s00190-007-0196-3 ORIGINAL ARTICLE A data-driven approach to local gravity field modelling using spherical radial basis functions R. Klees · R. Tenzer · I. Prutkin · T. Wittwer Received: 18 April 2007 / Accepted: 1 November 2007 / Published online: 2 February 2008 © The Author(s) 2008 Abstract We propose a methodology for local gravity field independent GPS-levelling control points, confirms that our modelling from gravity data using spherical radial basis func- solution is in no way inferior to NLGEO2004. tions. The methodology comprises two steps: in step 1, grav- ity data (gravity anomalies and/or gravity disturbances) are Keywords Local gravity field modelling · Spherical radial used to estimate the disturbing potential using least-squares basisfunctions · GPS-levelling data · Quasi-geoid techniques. The latter is represented as a linear combination of spherical radial basis functions (SRBFs). A data-adaptive strategy is used to select the optimal number, location, and depths of the SRBFs using generalized cross validation. 1 Introduction Variance component estimation is used to determine the opti- mal regularization parameter and to properly weight the dif- Spherical radial basis functions (SRBFs) are being used ferent data sets. In the second step, the gravimetric height extensively in gravity field modelling. Examples are the anomalies are combined with observed differences between point-mass kernel (e.g. Heikkinen 1981; Barthelmes 1986, global positioning system (GPS) ellipsoidal heights and nor- 1988; Vermeer 1982, 1983, 1984, 1995), the radial multipoles mal heights. The data combination is written as the solution of (e.g. Marchenko 1998; Marchenko et al. 2001), the Black- a Cauchy boundary-value problem for the Laplace equation.