Abstract

BOCA, T et al. Additive models for predicting biomass of Pinus elliottii var. elliottii x Pinus caribaea var. hondurensis in Misiones Argentina. Quebracho (Santiago del Estero) [online]. 2017, vol.25, n.1, pp.5-15. ISSN 0328-0543.

In recent years, forest management has changed the traditional concept for timber production to a global approach that includes the production of other goods such as dendro-energy and carbon dioxide fixation. In this context, it is necessary to improve the estimates of not only timber, but also other biomass fractions such as needles and branches. The predicting models of biomass components were fitted using Pinus elliottii var. elliottii x Pinus caribaea var. hondurensis hybrids data. To guarantee the additivity of the biomass components models, the simultaneous estimation method SUR (Seemingly Unrelated Regressions) was applied to the diameter at breast height and height as independent variables. The goodness of fit coefficient (R2) was 97 % for boles, 75 % for needles biomass and 68 % for branches mass. Considering that over 65 % of the aerial biomass corresponds to the bole, even in young trees, the results can be considered appropriate. The SUR methodology was compared to the traditional Ordinary Least Squares (OLS) method and showed a slight reduction in the coefficients confidence intervals. The statistical equations fitted are promising for future research of growth simulation models for hybrid pines.

Keywords : Seemingly unrelated equations; Tree biomass; Additive models.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )