Reconstrução da altura por meio da anatro completa do tronco de Pinus taeda L.

Abstract

Complete stem analysis is a technique used to rebuild the past growth of trees dendrometric variables that have visible growth rings. In this way, the present work aimed to evaluate different methods for height reconstruction using the Pinus taeda L. complete stem analysis. It was selected dominant trees (h100) of circular plots with an area of 400 m2. For complete stem analysis data were used 10 trees felled, dried and sanded in every 1m slices from diameter to full height. In order to measure growth rings between slices, the Image-Pro Plus program was used. For height growth reconstruction, six methods were used: i) Graves; ii) Lenhart; iii) Carmean; iv) Modified Carmean; v) ISSA; and vi) Targ. For each method the growth was evaluated from different positions from the diameter at 1,3m, being them variable, 2, 3 and 4 meters between slices. In the comparison between methods, the t-test, the analysis of variance and the Tukey test were used to evaluate differences or not at each age between the trees. In order to describe height growth (h), the Chapman-Richards model was tested. The performance of the adjusted models was determined by the coefficient of determination (R2), the root mean square error (RMSE), the mean absolute percentage error (MAPE), the corrected Akaike information criterion (AICc), the Akaike weights information criterion - wi (AIC) and graphical analysis of the residues. From the analysis of the results it was possible to conclude that the Targ method presented higher predictive capacity than the others, which is possibly associated to the use of the internal rings on the disks in the height calculation. The 4 m position between slices presented the best results and lowest error in the evaluations, this can be justified by the smaller number of slices evaluated.