Modelos sigmoidais na estimativa da altura de fustes em um sítio florestal amazônico, em Rondônia

Abstract

Accurate estimation of tree height is essential for forest inventories, timber stock assessment, and sustainable management planning, especially in complex tropical ecosystems such as the Amazon. This study evaluated the performance of three sigmoidal models—Morgan-MercerFlodin (MMF), Gompertz, and Logistic—to estimate the total height of tree trunks in na Amazonian domain fic in the municipality of Porto Velho, Rondônia, Brazil. The inventory comprised 26 clusters with four rectangular subunits of 2,000 m², arranged in a Maltese cross, totaling 4,715 tree trunks with diameters at breast height (DBH) ranging from 10 cm to 191 cm and ficien from 1.5 m to 48 m. The evaluation of the models considered the Root Mean Square Error (RMSE), Pearson’s correlation ficiente e, and Akaike Information Criterion (AIC). The MMF model showed less dispersion ficie the observed ficien and adequately reflected the asymptotic biological growth of the stems, even for the largest diameters. It is concluded that the MMF model is ficiente and recommended for hypsometric estimates in complex Amazonian forests.