This study investigated the changes in the spatial distribution of lodgepole pine (Pinus contorta var. latifolia Dougl.) stands as regeneration proceeds. Twenty-nine plots were established in regenerating lodgepole pine stands and were remeasured two years later. Nine of these plots had enough ingrowth to warrant an analysis of their spatial distribution dynamics. Ripley's K(t) statistic was used to identify the spatial pattern of the trees at the initial measurement, the ingrowth trees, and the combined initial and ingrowth trees. The initial trees were mainly aggregated, but some plots had random or regularly distributed trees. The ingrowth trees were all aggregated, and the initial and ingrowth trees, when analyzed together, were aggregated or random. The bivariate K(t) statistic was also used to detect correlation between the locations of the initial trees and the ingrowth trees. Two of the nine plots showed correlation between the ingrowth and the initial trees; the initial and ingrowth tree locations were independent in the remaining seven plots. The spatial patterns were modelled by a Poisson cluster process, a Poisson process, or a Markov point process when the trees were aggregated, random, or regularly distributed, respectively. These models resulted in satisfactory fits to the data. Correlation between the initial and ingrowth trees occurred only when both types of trees were clustered; the correlation was modelled by using the same cluster centres to generate tree locations. When the initial and ingrowth trees were uncorrelated, then tree locations were generated independently.
Download RR12 PDF file (2255 KB)
To view this document you need the current version of
Adobe Acrobat Reader, available free from the
Adobe Web Site.
Updated October 24, 2008