STAND-LEVEL PROGNOSIS OF EUCALYPTUS CLONES USING ARTIFICIAL NEURAL NETWORKS

Main Article Content

Mayra Luiza Marques da Silva Binoti Helio Garcia Leite Daniel Henrique Breda Binoti José Marinaldo Gleriani

Abstract

The objective of this study was to train, implement and evaluate the efficiency of artificial neural networks (ANN) to perform production prognosis of even-aged stands of eucalyptus clones. The data used were from plantations located in southern Bahia, totaling about 2,000 acres of forest. Numeric variables, such as age, basal area, volume and categorical variables, such as soil class texture, spacing, land relief, project and clone were used. The data were randomly divided into two groups: training (80%) and generalization (20%). Three types of networks were trained: perceptron, multilayer perceptron networks and radial basis function. The RNA that showed the best performance in training and generalization were selected to perform the prognosis with data from the first forest inventory. We conclude that the RNA had satisfactory results, showing the potential and applicability of the technique in solving measurement and forest management problems.

Article Details

How to Cite
BINOTI, Mayra Luiza Marques da Silva et al. STAND-LEVEL PROGNOSIS OF EUCALYPTUS CLONES USING ARTIFICIAL NEURAL NETWORKS. CERNE, [S.l.], v. 21, n. 1, p. 97-105, apr. 2016. ISSN 2317-6342. Available at: <http://cerne.ufla.br/site/index.php/CERNE/article/view/1044>. Date accessed: 22 sep. 2019.
Keywords
Modeling forest growth and yield, approximation of functions, unthinned stands.
Section
Article