Prediction Effect of Seedling Biomass on Different Indexes of Pinus yunnanensis

In order to improve the estimation accuracy of biomass of Pinus ynnanensis seedling, a regression model was used to estimate the biomass for two-year-old P. yunnanensis seedlings. Growth indexes including ground diameter(D), seedling height(H), product of ground diameter and seedling height(DH), product of ground diameter square and seedling height(D²H), and aboveground biomass(fresh weight) indexes including aboveground biomass fresh weight(W_Af), stem fresh weight(W_Sf), leaf fresh weight(W_Lf) were taken as independent variables. The underground part(total root, major root, lateral root) and individual plant biomass were predicted respectively. The determination coefficient(R²), the standard error of the estimated value(SEE) and the significance level of regression test were used as the selection criteria of the optimal model, and the accuracy of the optimal model was tested. The results show that the biomass optimal models constructed were all power functions. The product of ground diameter square and seedling height(D²H) in growth index and the fresh biomass of aboveground part(W_Af) in biomass index were the best models fitted with independent variables respectively. D²H could better predict the total fresh root weight and major root fresh weight of underground part. W_Af could better predict the total root dry weight, major root dry weight, lateral root dry weight, individual plant dry weight, lateral root fresh weight and individual plant fresh weight, and the goodness of fit of lateral root biomass was significantly higher than that of underground parts. There was no significant difference between the predicted fresh weight and dry weight of the same variable in the optimal model. Therefore, in actual production, for different variables, the most appropriate prediction index should be selected to fit the variables so as to improve the accuracy of estimation.