PATH ANALYSIS APPLIED TO EVALUATION OF BIOMASS ESTIMATES IN SUBTROPICAL FORESTS AT BRAZIL

Correlation analysis among different variables do not always fully express the real direct effect of one variable on another, especially in forestry. The objective of the present study was to evaluate the correlations among direct/indirect effects of tree diameter, height, and biomass components on total aboveground tree biomass and consequences when using path analysis. We also aimed to define which of the variables should be included in biomass modeling. The tree data were collected from eight sites located in Parana and Rio Grande do Sul, where the diameter 1.30 m aboveground ( dbh ), total height ( h ), biomass of stems, branches, leaves and total aboveground biomass were measured. Spearman's correlation showed that dbh  (0.93), stems (0.94), and branches (0.90) had the highest association with total aboveground biomass when the direct and indirect effects of these variables were deconstructed by path analysis. The stem (0.44) and branch (0.35) components provided greater direct effects when compared with dbh  (0.17). For the component stem, branches, and leaves the total aboveground biomass presented the greatest direct effects of 0.59, 0.74, and 0.36, respectively. Thus, these results indicate the convenience of including at least one biomass component in the biomass regression model, along with dbh  and height.