Abstract

VIANA, Helder; MARQUES, Carlos Pacheco  and  FONSECA, Teresa. Prediction of Aboveground Biomass of Pinus pinaster Aiton Using a System of Integrated Additive Equations from the Open Source ModisPinaster Simulator. Silva Lus. [online]. 2013, vol.21, n.Especial, pp.77-86. ISSN 0870-6352.

This study presents a system of allometric equations to estimate the aboveground biomass (AGB) components (stem wood, stem bark, branches, top and needles) of maritime pine (Pinus pinaster Aiton.). In order to assure the additivity properties among the components of tree biomass and total tree biomass, a nonlinear regression equations system was developed by Seemingly Unrelated Regression (SUR) procedures. The data was collected, by destructive sampling, in 77 trees within 16 forest stands, located in the North-Centre of Portugal. The trees components were separated, individually weighed and, after determination of the moisture content, the biomass dry weight was determined. The biomass of individual components was modelled as a function of the diameter at breast height and total tree height. The goodness of fit (R²adj) and the percentage root mean square error (RMSE) indicates the good predictive ability to estimate the total AGB biomass (R²ad = 0986, RMSE = 13.7%). The lowest predictive ability was observed in top and branches components (R²ad = 0.71, RMSE = 22.3%; and R²ad = 0.79, RMSE = 51.1%, respectively). The comparison with previous equations, for the same species, showed that the new models achieve more accurate estimates in the stands studied. Finally, the system of equations was included in the open source simulator ModisPinaster, available in the Capsis platform (http://www.inra.fr/capsis/help_en/modispinaster), allowing the estimation of maritime pine AGB, under different management scenarios chosen by the user.

Keywords : Biomass; maritime pine; Seemingly Unrelated Regression; ModisPinaster.

        · abstract in Portuguese     · text in Portuguese     · Portuguese ( pdf )