Intelligent Geographic Information Systems ( IGIS ) and surface water modeling

Geographic Information Systems (GIS) are emerging as major spatial data handling tools for natural resources inventory and management. For simulation model applications to surface water related problems, a GIS can serve as a powerful, interactive, and graphics oriented data platform and user environment. However, prior to utilizing available data and models, analysis of the problem's information requirements, the data base contents, and the given boundary conditions is necessary to configure models and data in a most effective and economical way. Problem diagnosis and preprocessing steps such as domain discretization, model parameter selection and calibration require problem-, methodand tool-specific expertise. This paper describes a conceptual approach towards the development of an "intelligent" Geographic Information System (IGIS), which incorporates knowledge processing capabilities for the surface water problem domain. Special attention is given to object oriented methodology, interfaces between numeric data and symbolic knowledge representation, and to dealing with uncertainty.