Prediction Of Growth And Establishment Of Stand Harvest Tables For Oak Secondary Forests In Hunan

Forest not only provides material basis for human survival and development,but also plays ecological functions such as protecting biodiversity,preventing wind and fixing sand,conserving water and soil,conserving water and adjusting climate.Sustainable utilization of forest resources is the goal of modern human beings in the use of forest resources,and the realization of this goal is inseparable from the scientific management and management of forest resources.Stand growth and harvest model can accurately monitor and predict stand growth dynamics,and it is also an important basis for evaluating silvicultural effect and formulating silvicultural measures.The forest coverage rate of Hunan Province is 59.82%,which is one of the important forest areas in China.The Oak secondary forest is widely distributed in Hunan Province,with large area and many species.In this study,the Oak secondary forest in 6 issues(1989-2014)of national forest resources continuous inventory data of Hunan Province was taken as the object,using 125 plots with Oak as the dominant tree species or main tree species,a total of 195 groups of 346 sample data,comprehensively consider the three factors of stand age,site class index and stand density,a parameterized and mixed effect based basal area growth rate model and basal area growth model of Oak in Hunan Province were established;meanwhile,a variable density model of Oak in Hunan Province was established and a variable density harvest table was compiled.The results provide a scientific theoretical basis for monitoring and predicting the growth dynamics of Oak secondary forest in Hunan Province and formulating reasonable forest management measures.The main conclusions are as follows:(1)Two theoretical growth equations and four empirical equations were used to simulate the relationship between basal area growth rate and basal area of Oak secondary forest in Hunan Province.The Weibull model had the best fitting effect(R2=0.817,P=91.27%,RMSE=1.399).A parameterized model including stand age a,site class index SI and stand density(stand numbern)was constructed(SI and N were continuous)The test index and residual diagram showed that the prediction accuracy of mixed effect model(ME=1.111,RMSE=1.326,R2=0.836,P=96.77%)was the highest.F-test(F=1.055<F0.05=3.183)was used to test the mixed effect model of stand basal area growth rate based on the measured value of stand basal area at the end of the period.It also showed that there was no significant difference between the measured value and the predicted value of stand basal area at the end of the period.When the site index and stand age are the same,the basal area growth rate of different density stands has the same starting point,the growth rate of low density stands is significantly higher than that of high density stands in a certain period of time,and tends to be close in the process of growth rate decline;the growth rate speed is directly proportional to stand density and site index,and in the 9(12)site index level,the decline speed of density I is the maximum In 30-35(20-25)years,density II in 25-30(15-20)years,density III in 20-25(15-20)years.(2)Richards growth model was the best(R2=0.314,P=92.89%,RMSE=6.211).The mixed effect model was established by adding the site class index SI to the parameter a representing the maximum basal area and the number of trees N to the parameter b representing the growth rate of the basal area Variable and parametric model(N is continuous variable).Compared with the basic model,the prediction accuracy is significantly improved(Pvariable=98.36%,Pparametric=97.71%,Pbasic=92.84%).When any two factors in a,N and SI are the same,the total growth of stand basal area(5-year growth of stand basal area)increases with the increase of the other factor,and the larger the density is,the smaller the difference of total growth of two adjacent densities is;when the growth reaches the maximum in 5 years,the larger the density is,the faster the growth decreases;when other factors are the same,the more trees there are The earlier the maximum value of growth appeared(20 years),the greater the site index and the greater the growth of basal area in 5 years.(3)The growth rate obtained by the stand basal area growth model decreases from fast to steady with the growth of age,while the growth rate obtained by the stand basal area growth rate model decelerates slowly before 20 years of age,and then decreases steadily after 20-25 years of age;when the stand age is 25-70 years,the difference of stand basal area growth rate between the two models is very small.(4)In the stand variable density harvesting model,the stand density index model and stand average diameter at breast height model are based on Korf equation(RMSE=192.672,R2=0.1837,P=93.41%;RMSE=1.104,R2=0.498,P=97.77%),and the relationship between N and parameter a,SI and parameter b in the stand density index model is a power function.The fitting accuracy of stand basal area model was still high after N was replaced by SDI(R2=0.945,P=97.32%).Schumacher model was used as stand volume prediction model with high fitting accuracy.The results of F test(FSDI=0.998,FG=1.019,FD=0.974,FM=1.074,all less than the critical value,F0.05=3.140)were both actual values and theoretical values,and the difference was not significant.Therefore,the forest model system established in this paper has high fitting accuracy,which can scientifically predict the growth dynamics of Oak secondary forest in Hunan Province,and provide scientific basis for the management activities of Oak secondary forest in Hunan Province.