Fundamentals of Spatial Data Accuracy and The Global Spatial Data Model (GSDM)

Three-dimensional (3-D) spatial data accuracy is described by the standard deviation of each component in the context of a global spatial data model (GSDM) that simultaneously accommodates both local and global perspectives, both high-level scientific and local “flat-earth” applications, and both activities that generate spatial data and activities that use spatial data. The goal in this paper is to identify and build on fundamental concepts of spatial data and error propagation to promote a better understanding of spatial data accuracy. Starting with a definition of the spatial data primitive and associated conventions, spatial data of various types are added to beginning control point values to build a 3-D database. Using the global spatial data model (GSDM), the unique location of each point is stored as geocentric X/Y/Z coordinates and the stochastic information for each point is stored in a covariance matrix. Conventional latitude/ longitude, UTM, and map projection coordinates (plus other geometrical elements such as heights, areas, and volumes) are derived from the stored values. The standard deviation of each derived quantity is readily available from the stored stochastic information using conventional error propagation techniques. Regardless of discipline, spatial data users worldwide can enjoy th