Effect of basic density, site conditions and genetic material on the allometry of Eucalyptus species

Abstract

The proportions of biomass across different components vary among species, developmental stages and growth environments, reflecting their allometric patterns. Allometric equations are widely used to estimate biomass, primarily based on diameter at breast height (DBH) and tree height. However, when developing generic equations for multiple species and environmental conditions, wood density becomes a key variable to consider. This study evaluated four hybrid clones of Eucalyptus species, aged between 97 and 110 months, in two regions of Minas Gerais, Brazil. Five trees per genetic material and region were sampled to determine aboveground biomass (stem, branches, bark and leaves) and average basic density. Biomass components and total biomass were estimated using different model fitting approaches: independent fitting, simultaneous fitting using non-linear seemingly unrelated regressions (NSUR) and weighted NSUR (WNSUR). Three modelling procedures were applied: (1) DBH and height; (2) DBH, height and basic density; and (3) DBH, height, basic density, and dummy variables for region and clone. The hypothesis tested was ‘Integrating basic density with clone and region effects through dummy variables improves predictive performance and estimator efficiency compared to systems based solely on allometric variables. The simultaneous fitting techniques, NSUR and WNSUR, proved to be efficient in developing accurate and high-quality equation systems. The inclusion of wood density enhanced the predictive performance of generic biomass equations and accounted for the effect of clone identity. The site effect was most pronounced in clone C1 .