Thinning Trials

Abstract

The use is advocated of an experimental design, more particularly for thinning, in which a large number of lightly differentiated treatments are sited in serial order of intensity in order that negligible edge effects occur and surrounds are unnecessary. The large areas required for the more orthodox, highly discrete treatments, have been a major obstacle to satisfactory thinning trials. Evaluation is by regression using individual trees as units. An example is given of the diameter changes in a pilot experiment. Replication is dealt with and an elaborate design explained with a diagram. Selection of trees for removal involves priority for removal of trees of small diameter. Defective types are eliminated in the first thinning and subsequently only seriously damaged or diseased trees have priority for removal over small trees, Distribution is totally ignored. The diameter basis for thinning is supported by evidence of the high correlation between successive diameter measurements of individual trees. Some early trends of basal area and diameter in plots managed in accordance with the foregoing methods are given.