Spatio-temporal data model based on dynamic correlation

Geographic information systems (GISs) have become important tools in public planning activities, such as environmental and utility management. This type of activity requires the creation and management of different period geodata, as well as analysis of spatio-temporal data evolution. Existing GISs usually provide limited support for these operations. However, changes in the application domain are tracked in the system by performing updates and erasing information in the past. On the other hand, numerous spatiotemporal applications demand a general spatio-temporal model for storing and managing mass geo-data, analyzing changes over time and visualizing these changes. The purpose of this paper is to explore a general spatio-temporal model that may be useful in new type of applications, including urbanization analysis system and environmental monitoring system. Firstly, this paper reviews the spatio-temporal conceptual models that have been developed and points out their characteristics. Based on the research above, an integrated spatio-temporal data model based on dynamic correlation is proposed. This model uses the terms version and difference to represent spatio-temporal data at an instance and changes over time. A default version, which focuses on and always represents the present data state, is introduced in this model. And four version sets are introduced: current sets, historical sets, process tracing sets and version sets. Through the verification based on case studies, this model can reduce data reduction and accelerate data access based on the dynamic correlation among four sets.