The spatial pattern of trees in lodgepole pine sites in the SBSdw3 and SBSmc2 biogeoclimatic variants were identified and modelled. The K(t) statistic was used to detect the patterns and provide a basis for determining an appropriate spatial process model. Most sites had an aggregated pattern, which was modelled effectively by a Poisson cluster process. The clusters range in size and intensity from many small clusters to a few dense tightly packed clusters, to large wide-spread clusters. An equation describing the relationship between the parameters of the Poison cluster process model was developed. Random and regular patterns occurred equally often. The random patterns were modelled by a Poisson process. The regular patterns were modelled with a Markov point process and a lattice-based process; the former model resulted in better simulations.
FRDA Research Report 244 (1922 KB)
To view this document you need the current version of
Adobe Acrobat Reader, available free from the
Adobe Web Site.
Updated July 24, 2015