关键词: 白桦, 形率, 单木材积, 生物量, 交叉检验

Abstract:

Objective: Taking Betula platyphylla in the Daxing'an Mountains as the research object, the volume model with form quotient was constructed and compared with the two-variable B. platyphylla volume model being used in the forestry production in northeast China and the traditional basic volume model. Combined with the biomass expansion factor, the effect of the form quotient on biomass was studied to provide a scientific basis for accurate prediction of individual tree volume and biomass. Method: 15 form quotients on the trunk were introduced into the traditional volume model to establish the two-variable and three-variable volume models, respectively. Specifically, the GNLS module of R software was used to fit the models, and variance functions were introduced to eliminate the heterosedasticity in the fitting process of each volume model. The mean percentage of bias(MPB), mean absolute bias(MAB), root mean square error(RMSE) and determination coefficient(R²) were used as evaluation indexes to compare and analyze each model, and the cross-validation method was used to validate the models. Result: 1) Introducing form quotient could significantly improve the fitting effect of volume model. When the relative tree height was 40%, the fitting effects of two-variable volume model and three-variable volume model with form quotient q_0.4 were the best. 2) The cross-validation results showed that the accuracy of volume prediction could be significantly improved by introducing form quotient into the volume model. Compared with the traditional two-variable model(2) and one-variable model(1), the RMSE of the three-variable model(15) and two-variable model(14) with form quotient were reduced by 38.9% and 45.8%, respectively. 3) Compared with the two-variable volume model(5) currently used in northeast China, the RMSE, MAB and MPB of the two-variable volume model(14) with form quotient was reduced by 46.7%, 40.2% and 47.7%, respectively; RMSE, MAB and MPB of the three-variable model(15) with form quotient was reduced by 59.1%, 59.2% and 64.1%, respectively. 4) Form quotient could significantly improve the prediction accuracy of biomass, The RMSE, MAB and MPB of the individual tree biomass model based on the three-variable model(15) with the biomass conversion factor(BEF) was reduced by 46.9%, 46.5% and 46.3%, respectively, compared with that of the basic model without the introduction of form quotient. Conclusion: The introduction of form quotient into the volume model could not only improve the prediction accuracy of individual volume, but also be helpful for the accurate prediction of individual tree biomass. The three-variable volume model(15) with form quotient might be more suitable for predicting Betula platyphylla individual volume and biomass in this area.

Key words: Betula platyphylla, form quotient, individual tree volume, biomass, cross validation