Modern geographic information systems and model linking

Abstract

Geographical analysis involves the use of a wide range of techniques and models to apply formal, usually quantitative structures to systems in which the prime variables of interest vary significantly across space (Longley and Batty 1996). Such techniques of spatial analysis form a well-defined subset of the larger set of analytic methods, defined by an invariance property: the results of spatial analysis are not invariant under changes in the locations of the objects being analyzed. The tradition has its historical roots in locational analysis (e.g. Haggett 1965), geostatistics and the theory of regionalized variables (Matheron 1971; Cressie 1991) and spatial statistics (e.g. Cliff and Ord 1973), yet there are many reasons why modern geographic information systems (GIS) are causing the field to undergo profound change. Today’s GIS (broadly defined) provides a seamless computing environment for every stage of spatial analysis. In this chapter we assess the implications of this state-of-affairs, and its possible implications for the future development of spatial analysis in a GIS environment.