Using Isosurfaces To Explore Geo-Temporal Patterns Of Crime

Crime has been shown to exhibit marked variations over both space and time. Sherman (1989) reported that 3% of addresses are the subject of 50% of police call for service (Sherman, 1989 p. 3). This is also the case with the temporal dimension; Van Koppen (1999) in his study of commercial robberies reported daily and weekly peaks and distinct increases in winter rather than in summer months. From an operational policing perspective the job of resource allocation needs to collectively take into account both the spatial and temporal dimension.

Geographic visualisation is an area that has been the focus of much research (see for example, MacEachren and Kraak,(2001)). When applied to crime data geographic visualisation provides a powerful tool through which valuable insights in criminal dynamics can be derived. Given the large variability of crime in both space and time a technique capable of representing both in conjunction with one another is called for. The problem is that, to date, there is not any well established technique for achieving this with crime data. In this paper we present a technique that is capable of addressing this problem through the display of both space and time in a single visualisation.