Abstract:

Based on neighboring relationship of features, forestry database acquired multi-scale data through spatial data generalization in order to meet the research of forest, landscape and natural geographic region. Taking the 1:10000 forest inventory spatial data as the data source, this article analyzed the topological neighboring, distance neighboring and semantic neighboring, and completed the aggregation and integration operations of polygon in the sub-compartments and constructs the spatial database of 1:25000 by graphics generalization of ArcGIS. Taking Guniushan as experimental area,the results showed that the number of the sub-compartments changed from 3520 to 2416 by means of the integrated operation which merged them according to the order of soil factors, slope position, slope, aspect, topography factor, canopy density, forest age and area. And it meeted the law of data integration of the open square root. The proportion of each property factor had not changed so much after the integrated operation, and the biggest change was the mature forest's age which changed from 10.27% to 12.19%.The mainly reason was, after the integrated operation, the nearly mature forest, the mature forest and the over mature forest were regarded as the mature forest. The proportion among species changed slightly whenever before or after the integration,and it basically meete the area balance between species.

Key words: multi-scale, neighboring relationship, data generalization, GIS