Forma de tronco e sortimentos de madeira de Eucalyptus grandis Hill ex Maiden., na região sudeste do estado do Rio Grande do Sul
This study had the objective of studing the stem shape of Eucalyptus grandis Hill ex Maiden., based on 261 trees, with diameters at breast height (DBH) ranging from 8 cm to 74 cm, originated from Aracruz Celulose S. A. forests, located at Depressão Central and Escudo Sul-Riograndense, in Rio Grand...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | Portuguese |
Published: |
Universidade Federal de Santa Maria
2005-01-01
|
Series: | Ciência Florestal |
Online Access: | http://www.redalyc.org/articulo.oa?id=53415309 |
Summary: | This study had the objective of studing the stem shape of Eucalyptus grandis Hill ex Maiden.,
based on 261 trees, with diameters at breast height (DBH) ranging from 8 cm to 74 cm, originated from
Aracruz Celulose S. A. forests, located at Depressão Central and Escudo Sul-Riograndense, in Rio Grande
do Sul state, to determine the volume and assortment of wood in two classes: timber (logs with 25 cm of
minimum diameter, with bark, and lengths of 5,6; 4,2 and 2,8 m, prioritizing the higher length assortments
formation) and industry (stems or portions of stem with diameters between 8 and 25 cm with bark). The
fiftieth degree polynomial, having as dependent variable the relative diameters (di/d) and independent
variable the relative heights (hi/h) was selected to describe the shape, with data stratification in three
diameter classes (DBH): diameter until 20 cm; between 20 and 30 cm and larger than 30 cm, with coefficient
of determination higher than 0,97 and standard error of the estimation lower than 6,2 %, allowing to estimate
the relative and absolute volume of the assortments through integration of form function. The model
adjustment for the trees complex covering all the diameters breadth generated a coefficient of determination
equal to 0,98 and a standard error of the estimation equal to 6,54%, also allowing to select the model to
describe the shape and assortment prediction. |
---|---|
ISSN: | 0103-9954 1980-5098 |