A new model for handling vector data uncertainty in geographic information systems

The authors investigated the uncertainty associated with grid-cell data in geographic information systems (GIS), through the use of a model which permits a population of distorted (but equally probable) versions of the same map to be generated and analyzed. While this model is easily applied to data such as digital elevation models to help assess the uncertainty associated with the products derived from them, it was not directly applicable to vector data. The model has now been enhanced to allow for distortion of vector data through the creation of separate horizontal positional error fields in the x and y directions, which are then overlaid with the vector data to apply coordinate shifts by which new versions are created. The likely uncertainty of vector data in both position and attribute may be assessed in certain types of GIS outputs. This paper explains the background to the model and discusses its implementation and potential applications.