Epsilon Archive for Student Projects

Abstract

Management based modifications of forest structure require a precise description of
the forest state both before and after a silvicultural interference. While the average
values of variables such as diameter, height and basal area provide useful descriptions
of the forest state, these variables do not give information about the structural variety
of the forest on a tree to tree basis.
The aim of this study is to introduce and discuss a simple index that can be used to
measure the dissimilarity in sizes between trees. The index, called the tree size
diversity index di, is based on the coefficient of variation of the diameter sizes of two
neighbouring trees. For a theoretical all-sized forest, di has an expected value of0.5.
This value is considered to be the standard or 'yardstick' for measuring tree size
diversity. di always ranges from 0 to 1. It is close to zero when the trees are of similar
size, and tends to one as the difference between the trees gets larger.
The tree size diversity index can also be used to test hypotheses concerning size
segregation, whereby large trees may tend to occur in the neighbourhood of other
large trees and small trees in the neighbourhood of other small trees. This can be
accomplished by carrying out randomisation tests as discussed in the study.
The tree size diversity index therefore provides a variable that can be used to describe
a particular forest state, that is a practical measure of control for assessing silvicultural
interference, and is a nearest neighbour statistic for testing hypotheses concerning tree
size segregation.