Assessing components of a competition index to predict growth in an even-aged Pinus nigra stand

The relationship between tree growth and competition may depend on some subjective choices that are commonly left to the researcher. Among these are the neighborhood radius, the number of years of growth that are integrated, and tree age. We have evaluated the importance of these factors when relating growth and competition in a forest stand with contrasted densities of the dominant tree species (Pinus nigra) and understory shrub species (Adenocarpus decorticans). Previous to this evaluation we performed a randomization test to assess the relationship between tree growth and neighbors. By using Daniels index of competition we found that the use of a fixed neighborhood radius underestimated the importance of tree competition. The coefficient of determination (r(2)) of the relationship between tree growth and Daniels index increased asymptotically with the number of years considered. Five years of growth gave high r(2) independently of the density of trees and shrubs. The intensity of competition was weakly affected by the characteristics of the plot (tree and shrub,densities), and did not change with time. In contrast, the potential growth at equal competition - as represented by constant "a" in the allometric model - changed with time suggesting a gradual decrease in potential tree growth in the plots with higher tree density, and a gradual increase in those plots with high density of shrubs. These results may reflect tree canopy closure and the senescence of Adenocarpus decorticans. A method is suggested to select optimum neighborhood radius and growing period for the calculation of competition indices. By applying this method we were able to explain as much as 79-84% of the variability in tree growth of this stand.