红松人工林叶面积分布规律及回归模型研究.

The size of the leaf area can have a profound impact on the growth and life activities of the tree, and then it is related to the change of the biomass of the tree and the development trend of the forest ecology. The research on leaf area can play a key role in the correlation analysis of trees. Based on the analytical wood data of 48 Korean pine plantations in Heilongjiang Province and the leaf area data measured by the drainage method, a basic linear regression model was established, and then random effect parameters were introduced to construct a mixed effect regression model to improve the fitting effect and prediction accuracy. When random effects at the tree level were added, the optimal mixed effect model obtained contained three random effect parameters, namely logarithm of branch base diameter(lnBD), logarithm of branch length (lnSL), and tree height logarithm(lnHT). The R ² a was 0. 86, the MAE was 0. 342 8, and the RMSE was 0. 484 1. The statistical indicators were better than the basic linear model that only included fixed effects. The comprehensive analysis showed that the model can better describe the leaf area of Korean pine branches. Then, based on the model, the crown leaf area of the Korean pine tree was calculated and a mixed effect prediction model was obtained. The mixed model R ² a was 0. 83 and the RMSE was 0. 442 9. Through the test results, it can be seen that the model can predict the crown and leaf area of Korean pine trees better, and provide a better guiding direction for the future management of artificial Korean pine in this area. [ABSTRACT FROM AUTHOR]