Uso de redes neurais na estimativa volumétrica de clones de Eucalyptus spp no Pólo Gesseiro do Araripe, Pernambuco

Abstract

The Araripe Gypsum Pole in Pernambuco is responsible of 97% of national production of plaster. The main source of energy for the gypsum calcination process, raw material for plaster production is the wood from the natural vegetation of Caatinga. Due to the high costs of other energy sources, increasing the gypsum production implies more deforestation of the Caatinga. An economic and environmental solution for that problem is the implementation and the sustainable management of native species or the reforestation with fast growing forest species. Among the fast growing forest the genues Eucalyptus stands out for it productivity and adaptation of the Northeast semi-arid region. The objective of this study was to estimate the volume of the Eucalyptus spp clones in Gypsum Araripe Pole employing the methodology of Artificial Neural Networks (ANN) comparing it with the volumetric models of Schumacher and Hall and Spurr. Data came from an experiment implanted in the Experimental Station of the Agronomic Institute of Pernambuco, where was tested 15 clones of Eucalyptus spp planted in 2002, with final cut in 2009.The function of interest estimated was the volume of the tree in function of the diameter at the breast height (DBH), total height (Ht) and the clone type. It was also valued the adjustment of the best models for sample size. The results were evaluated with the adjusted coefficient of determination (R2aj), square root of the percentual mean error (RMSE%), standard error estimate (Syx%) and an analysis graphic of the residues. The obtained results confirmed the expectation showing efficiency of adjustments independent of the sample size.