Comparing logistic function and clutter prognosis models

Abstract

The aim this study was to compare the performance of two non-linear models and the classic Clutter´s system on volume prognosis in fully stocking Eucalyptus stands from Espírito Santo, Brazil. Mean standard error percentual, Akaike and Bayesian criteria, supplemented by graphical analysis were used to evaluate models fit. Models performance presented better results up to four times when covariables were included in the adjustment process. It was observed that logistic function with covariables was superior than it´s original form and more appropriate then Clutter´s system. The results showed better results of non-linear system to modeling forest growth and yield indicating forest rotation near five years. These results are consistent with Eucalyptus plantation for biomass in Brazil.