Abstract

The paper is an account of some statistical analyses carried out in conjuction with a silvicultural research project in the department of Operational Efficiency, College of Forestry, Garpenberg. Sweden (Eriksson, L & Eriksson, O in preparation 1981). In Section 2, a statistic due to Moran (1950) is used to detect spatial interaction amongst counts of the numbers of young trees in sample plots laid out in a systematic grid arrangement. Section 3 discusses the construction of bivariate distributions for the numbers of wild and planted trees in a sample plot. Section 4 considers the relationship between the distributions of the number of trees in a plot and their total basal area. Section 5 is a review of statistical methods for ute in connection with preliminary surveys of large areas of forest. In particular, Section 5 discusses tests of spatial randomness for a single species, tests of independence between two species, and estimation of the number of trees per unit area.