Allometric models for estimating aboveground biomass in tree sprouts of three dendroenergetic crops

Recurrent problems have been observed for biomass measurement in tree sprouts, mainly dueto differences in patterns of biomass distribution and the bias generated by using models for mature trees.The objective of this research was to evaluate models to estimate aboveground biomass in sprouts from two-year-old stumps of Short Rotation Woody Crops (SRWC) used for energetic purposes. The study was conducted in Central Chile (Bío-Bío region) under a complete block design and four replications.Three plant species were included: Eucalyptus globulus, E. denticulata, and Acacia dealbata, established in three population densities (5,000, 10,000 and 15,000 trees per ha). Two groups of nonlinear models wereused on three independent variables: root collar diameter (squared mean of two dominant stems), height (mean of two dominant stems) and number of epicormic stems. In the allometric model, the total aboveground biomass and the aboveground biomass by component were explained as the product of the mean square of the root collar diameter, mean height of the two dominant new sprouts and total number of epicormic sprouts as the predictive variable. Values of coefficient of determination (R2) ranged from 0.78 to 0.95 and RMSE ranged from 168 to 913 g. The species of E. globulus showed higher RMSE for biomass estimation in the all components except the branch component. The inclusion of dummy variables to identify the differences in each parameter of the models explained the seasonalityof the biomass accumulation in tree sprouts at 10, 21, and 31 months of age, improving the model goodness of fit RMSE by 27%. Accurate prediction of sprouts aboveground biomass was obtained by the models only evaluating the two dominant epicormic sprouts in the stump, avoiding the need of individual sprouts evaluation or using destructive methods for biomass measurement.