Developing Individual Tree Crown Width Models For Larix Principis-rupprechtii

The Larix principis-rupprechtii forests are almost managed for the objectives of timber production,ecosystems or watershed protection,and habitats for specific animals.Larch forests in this region are usually characterized by a large amount of biomass and high primary productivity,which may be related to high adaptability of the tree species to the extreme low winter temperatures as well as to the efficient utilization of water from the melting zone of permafrost soil during the hot and dry summer season.This is why these larch species are widely used for timber production and afforestation in the temperature zone of China.Tree crown is an important place for photosynthesis and accumulation of energy.It reflects the grovwth vigor and competition ability of trees,and is an important health index of trees.The crown width is to describe the important variables to predict important index of canopy growth is also a harvest forest growth prediction model.This paper constructs a prediction model with a variety of crown diameter trees stand factors as covariates,and analyzes the influence of regional effect and group effect and nested in block in the kind of effect on the larch crown,on the basis of using the nonlinear mixed effect model constructed crown model,also consider the correlation the crown and the four corners of the world in four directions and crown diameter,and the use of four kinds of additive model method to build a model system,four models are solved compatibility the diameter of the crown and the four corners of the world four direction of the crown;finally through comprehensive comparison to determine the additivity of crown model system optimization.The main results are as follows:(1)Logistic model with three parameters:CW=a/l+exp[b+cln(D+1)]can better reflect the nonlinear relationship between the amplitude and the diameter of the natural secondary forest of larch,compared with other candidate models,the model has high fitting precision,and have certain biological significance of the parameters of the model object;wood crown length(CL),object height(H)and per hectare(M)has great influence on the crown,when the forecast accuracy is significantly improved model of these factors as predictors of the crown;the expressions of the improved model for:CW=1.4875-0.031IH+exp[(-3.6416-0.0002M)+(2.4346+0.0045CL)ln(D+1)]Through a large number of experimental data,the model has higher prediction accuracy.(2)The study found large random effects and nested block effects in block inside the plots of larch,when considering these random effects model prediction accuracy can be further increased significantly;the variance function index and predictive variables for breast height diameter can be effectively removed model heteroscedasticity,expression is:var(?ijk)= ?2 exp(2?xijk);the construction of the two level nested nonlinear mixed effect model to predict crown,using a random sample of 4 strains of samples calculated random effect parameter is better.When the regional effect in the fixed effect parameters and corresponding model,AIC(5425)and LogLik(-2697)nonlinear minimum maximum final construction of Larix natural forest crown models such as the regression effect shown in(5-19).(3)Model(4-3)as the based model,using nonlinear simultaneous equations(NSE),nonlinear likelihood regression(NSUR),the ratio adjustment method(AP)and least squares regression(OLSSR)method to construct independent crown additive model system,these methods can reflect the correlation between the total crown and the effective radius of the crown.Through the comprehensive comparison of these kinds of additive model for the total system,crown width,crown model of joint control system construction adjustment method and corresponding grading index,and were lower than NSUR,AP and OLSSR model system;therefore,the joint control of construction grading adjustment method the crown additive model system the highest prediction accuracy,the final model the expression as shown in(5-7).