Representing and processing uncertain soil information in a fuzzy constraint framework

This paper presents a combined GIS (Geographical Information System) and co nstraint satisfaction approach to map soil hydrological properties from inf ormation contained in a small scale (1:250,000) pedological map. The region al soil databases available at this scale are rather imprecise. Because of the lack of reliable statistical data the imprecision is represented in a p ossibility theory framework. For every point in the covered area,an identif ication of the more or less possible soil typological units is computed. Fo r each candidate soil typological unit the water storage capacity is comput ed by using constraint satisfaction techniques that implement fuzzy interva l calculus on interactive variables. The method returns a map of the water storage capacity that accounts for the imprecision of the processed informa tion. Copyright (C) 1998 IFAC.