Abstract
This paper presents a review of the spatial pattern analysis of individual trees in natural forest stands by the quadrat and distance methods. These methods have been used for three purposes (1) to identify spatial pattern, (i.e. uniform, random, or aggregated), (2) to estimate density and (3) to analyze spatial pattern structure. The separate use for each of the above purposes causes bias in density estimation. The concept of "intensity and grain" developed by PIELOU (1977) was applied to overview spatial pattern analysis. The intensity expressed by an index of aggregation has three roles; (1) for identifying spatial pattern, (2) for correcting bias in density estimation and (3) for analyzing grain of spatial pattern, i.e. mosaics, or sizes of clumps etc. Spatial pattern analysis methods from the point of the three roles are reviewed and spatial patterns of individual trees in natural forests and the problems in natural forest inventories discussed.
