A Practical Global Spatial Data Model (GSDM) for the 21st Century

Earl F. Burkholder

Abstract:	We often work with two dimensions, but spatial data are 3 dimensional(3-D). Modern measurement systems collect 3-D spatial data. Computer databases store digital spatial data in 3-D. Human perception of spatial relationships is primarily visual. Mathematical models provide a conceptual connection between digital spatial data and its graphic representation - data visualization. Digital spatial data are also used to make analog products such as maps and other hardcopy diagrams. Numerical representation of spatial data includes listings of coordinates for individual points and relative point-pair relationships (such as bearings and distances) between points. Traditional models for spatial data include a simple flat earth model (used extensively in plane surveying) and the more complex ellipsoidal earth model (used in geodetic surveying). In each case, a natural distinction is made between horizontal and vertical due to the local perception of up. Given traditional use of horizontal and vertical datums, it follows that 3-D measurements are handled in terms of those models. However, except for local flat-earth problems, traditional equations for manipulating 3-D spatial data are more complex than equations for rectangular coordinates due to 1) using the ellipsoidal earth model, 2) using mixed units (angular units for latitude/longitude and meters for vertical), and 3) extensive use of 2 dimensional map projections to flatten the earth. A simple practical 3 dimensional global spatial data model (GSDM), which includes both functional and stochastic components, is presented. The GSDM 1) accommodates all modes of spatial data measurement, 2) utilizes a digital spatial data base containing earth-centered earth-fixed (ECEF) X/Y/Z coordinates, 3) does not distort physical measurements as does a 2 dimensional map projection, 4) uses one set of solid geometry equations world-wide, 5) portrays an accurate view of spatial data based upon any origin selected by the user, 6) preserves geometrical integrity by using local coordinate differences, 7) optionally stores stochastic model information in the X/Y/Z covariance matrix of each point, and 8) where covariances are stored, provides the standard deviation of each spatial data component and subsequently derived quantities such as distance, direction, area or volume.
Published in:	Proceedings of the 1998 National Technical Meeting of The Institute of Navigation January 21 - 23, 1998 Westin Long Beach Hotel Long Beach, CA
Pages:	791 - 798
Cite this article:	Burkholder, Earl F., "A Practical Global Spatial Data Model (GSDM) for the 21st Century," Proceedings of the 1998 National Technical Meeting of The Institute of Navigation, Long Beach, CA, January 1998, pp. 791-798.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In