Abstract: Taking Yunnan Province as the study area, we proposed a method of dividing forest spatial homogeneous zones integrating the characteristics of forest resources, natural geographic features and the differences in socio-economic development of the counties aimed at revealing the spatial heterogeneity of Yunnan's forest resources and laying the foundation for assets inventory work of the national forest resources . Based on the spatial differences in 17 factors associated with forest resource quality such as forest origin, volume per unit area, elevation, slope, topographic relief, climatic conditions and socio-economics among counties in Yunnan Province, the degree of similarity of forest resources was expressed in spatial distance between county units and Yunnan Province was classified into spatially more consistent homogeneous units by K-mean clustering and hierarchical clustering respectively. The results show that K-mean clustering is more effective, dividing Yunnan Province into six homogeneous zones, showing more conductive, aggregated and coherent in practice. In addition to considering the attributes of forest resources, the introducing of economic and anthropogenic factors in the cluster analysis can reflect the regional difference in forest resources more objectively.The study can provide reference for constructing the price system of national forest resource asset inventory and asset accounting, and has important theoretical and practical value for supporting the national forest resource inventory work.