Dynamic GIS

Most GIS are an epitome of static, which is why to many people the word ‘dynamic GIS’ is an oxymoron. GIS usually is about data and to a much lesser degree about what can be done with the data. As such, GIS can be compared to early 20 century geography – an Aristotelian description of ‘where is what’ and ‘what is where’ (Barnes 1984). Process, and the notion of change, are acknowledged on the sidelines of GIScience research but have not yet become main stream in everyday GIS use. This is in spite of some 15 years since Tomlin’s ‘cartographic modeling’ (1990), the success of the GIS and Environmental Modeling conference series (Goodchild et al. 1996, Clarke and Parks 2001), and well-established pockets of spatial process modeling research such as Peter Burroughs group in Utrecht, NL. This chapter will reframe some of the work usually subsumed under the header of geocomputation by focusing on a single but new dimension time and uncovering low-level structural problems (Burrough 1992) that make it so hard to merge the spatial and temporal aspects of GIS data models.