ABSTRACT
Geographic information systems (GIS) have been used successfully in many applications where digital representations of spatial resources and their management is a prime requirement. However, when using GIS for environmental management and decision support, many more considerations come into play. This awareness of the need to extend basic GIS functionality to encompass spatiotemporal data management as well as computational modelling capabilities has been documented (Abel 1993, Claramunt 1998). A number of current proprietary GIS are characterised by their flexibility and can be used for certain environmental applications but do not provide the specialised functionality required in the environmental GIS (EGIS) domain. Likewise, custom-built applications are good at solving specific problems but have very little generic flexibility. Both approaches often lack the facilities, extensibility and functionality that are required for environmental research which can be characterised by:
• large volumes of data, collected using a multiplicity of sensors which need to be processed before being used for analysis; the problem is often described as one of ‘too much data but not enough information’
• the need to merge raster, vector, scalar, hierarchical data with researchers’ inherent domain knowledge
• requirement to access, and make available, data over local and wide-area networks • the need for open software architecture that is extensible and flexible.
