MODELLING OF ALLOMETRIC EQUATIONS FOR BIOMASS ESTIMATE IN DECIDUOUS FOREST

Resumo

This paper aimed at testing and adjusting the allometric models for the biomass estimate in deciduous forest. The data result from seven 12 x 12 m plots, where 91 trees were cut down. Their diameter at breast height (DBH) was greater than 5 cm, and the measurement was made based on the DBH, total height (H) and total dry biomass (DAB) for each individual. The adjusted equations, without stratification, present an adjusted determination coefficient (R²_aj) between 0.726 and 0.972, and estimate standard error in percentage (S_yx%) ranging from 33.5 to 119.6. The best-adjusted model for the non-stratified data set was obtained through the Stepwise procedure, which resulted in the following equation: dab = β₀ + β₁.(DBH³) + β₂.H + β₃.(DBH³.H), with R²_aj of 0.954 and S_yx% of 44.0. For the stratified form, only the diameter class above 15 cm presented acceptable parameters, with R²_aj of 0.968 and S_yx% of 26.5.  That way, the current database has shown good results in the adjustment of stochastic models to estimate the DAB of each tree.