Estimation of yotal biomass and carbon for caixeta State, Brazil trees in Parana

Abstract

The objective of this work was to adjust models to estimate biomass and total carbon of individuals of Caixeta (Tabebuia cassinoides (Lam.) DC.), located in Guaratuba, coast of Parana State, Brazil. Thirteen models were tested, and the choice of the best was based on the statistical indicators R²_aj, S_yx (%) and graphical analysis of residuals. It was also an additional choice criteria , the evaluation of residual from the constrained regression for the best models. No adjustments were made for biomass and carbon by compartments due to the low correlation between the dependent variables (biomass and carbon) and independent, diameter at breast height (dbh) and total height (h). For total biomass, the model Y = β0 + β1*dbh presented better results, with R²_aj = 0.96 and S_yx(%) = 7.94. For carbon, the best adjustment was given by model Y = β0 + β1*dbh + β2*dbh² + β3*dbh³ + β4*dbh⁴, with R²_aj= 0.97 and S_yx (%) = 8.09. It was noticed the low variation of residues for both models. The variable total height, used isolated has proved to be inadequate to explain the variables total biomass and total carbon.