Objectives: Quantitative prediction of mineralization requires comprehensive study on threedimensional spatial information, and it is also confronted with the comprehensive analysis difficulties between qualitative and quantitative data, twodimensional and three-dimensional data.  Methods:A weighted shortest-distance field algorithm for the 3D heterogenetic space was developed, in which faults rupture zone can be treated as migration pathway of ore element and the shortest path can also be tracked back from any object voxel to source voxels. The algorithm was used to analysis the relation between the distance field value of the 3D ore-body model on the earth's surface and soil geochemical survey data in Hongtoushan copper mine area, China.  Results:Studies show the geochemical anomaly matches better with the joint distance filed of known and predicted ore-bodies than with the distance filed of only known bodies, which indirectly approved the reliability of predicted ore-bodies.  Conclusions: This method can interpret the most of geochemical survey anomaly which caused by orebodies, and can serves as a model for the comprehensive analysis and the relation extraction between geochemical survey, remote sensing data and threedimensional geological data model.