Fitting alometric equations for Araucaria angustifolia (Bert.) O. Ktze. using principal components analysis and linear regression

Abstract

The main objective of the present research was to select variables for fitting alometric equations in order to estimate stem fresh weight (sfw), needles fresh weight (nfw) and branches fresh weight (BFW), through Principal Components Analysis, for native and planted Araucaria angustifolia trees, in the southern regions of Paraná State, Brazil. Data from 15 trees of several ages (29 to 33 years) being evaluated the following variables: dbh (diameter at breast height), h (total height) hc (commercial height), sfw, nfw and bfw. As independent variables (x) they were used dbh, dbh², log(dbh), ht, dbh²ht, log(h), hc, dbh²hc and log(hc) and as dependent variables sfw, nfw, bfw. Once data were computed for all variable combinations, a correlation matrix was generated by Excel 2000 and Principal Components Analysis was made by Statgraphics 5.0. The study leaded to the conclusion that the use of Principal Components Analysis together with Regression Analysis showed a great potential for fitting alometric equations for Araucaria angustifolia, as well as for other wood species. A greater number of tree measurements for biomass variables is advisable, to improve the fitting of biomass equations for Araucaria angustifolia.