Spatial Interpolation Using Fuzzy Reasoning

Spatial interpolation is an important feature of a Geographic Information System, which is the procedure used to estimate values at unknown locations within the area covered by existing observations. In this paper, we describe a conservative spatial interpolation technique that incorporates the advantages of local interpolation, Euclidean interpolation, and conservative fuzzy reasoning, and a dynamic fuzzy–reasoning–based function estimator with parameters optimised by a genetic algorithm. The main objective of this paper is to formulate a computationally efficient spatial interpolation technique similar to the IDWA technique that can be used in real time application. The main feature of our spatial interpolation technique is a capability for spatial interpolation and extrapolation in a higher–dimensional space. Examples from a rainfall spatial interpolation problem are used to illustrate the applicability of the proposed technique.