A study was conducted to extract relationships between terrain factors and soil properties by clustering terrain factors based on the fuzzy set theory.Similarity of each pixel to the typical ones was obtained from the classification results,with degrees of similarity referred to as fuzzy membership.Then,a continuous soil horizon thickness distribution map was drawn with the aid of ARC/in of software based on samples collected at the area with high value of fuzzy membership and population of the similarity model using the linear and additive weighting function.Comparison of the predicted value set from the map with the independently collected field sample set showed the derived soil map achieved 82% of accuracy.Further investigation showed that the model worked well in the area of low altitude,well-developed soil,thick soil horizon,stable surrounding for soil formation.The new approach was found to be higher in efficiency and accuracy of soil mapping,lower in mapping cost,and more capable of demonstrating microregional variance and accommodating higher information load for mapping.Hence,it is a feasible new method worth extrapolating in soil survey and mapping.