A natural mixed broadleaf-conifer forest in the Lesser Khingan Mountains was selected as a sample,the main dominant tree species including Pinus koraiensis,Abies fabri,Picea asperata,Fraxinus mandshurica,Betula costata and Tilia amurensis were selected as the research objects,and correlation analysis of tree height vs.DBH,crown width vs. DBH were carried out respectively in this paper.By SPSS 20.0 software,the survey data was substituted into the eight classical mathematical models for regression analysis,the P value,the coefficient of determination R 2 and related parameters of each equation were obtained.The fitting accuracy was verified by using the total error,the average error,the average relative error and the root mean square error,the optimum tree growth model was finally selected.The results showed that the correlations between tree height vs.DBH,crown width vs.DBH for conifer trees were significant,and the average coefficients of determination in the tree height vs.DBH model and in the crown,width vs.DBH model were 0.809 and 0.498,respectively.Compared with the coniferous trees,the correlation coefficients between tree height vs.DBH,crown width vs.DBH for broadleaf trees were slightly worse,and the average coefficients of determination were 0.608 and 0.395,respectively.Different tree growth models should be used for different tree species,in the tree height vs.DBH correlation analysis,the power function and S function models were mostly used,while the cubic polynomial model was the most significant among the crown diameter vs.DBH models.
