| Stand quality is related to the function of wood production and ecological benefit.In the process of national reserve forest construction project,improving stand quality has become the main goal and task of forestry development.As one of the main timber species in Guizhou Province,Pinus massoniana has a wide planting area and ranks second in stock volume,but there is still a large space for improving stand productivity and ecological service function.In order to take reasonable management measures to promote the stand quality of P.massoniana.This study takes P.massoniana forest in Shiqian County and Yinjiang County as the research object,and the site index model was constructed based on the data of 1,286 subcompartments including 1,701 sample plots,and the suitability criteria of tree species were divided by site index value,then the tree species suitability was studied by using different machine learning algorithms.Further considering the strategic positioning and business objectives of the national reserve forest,the stand quality evaluation index system was constructed,and the index weight was determined by combination weighting method,so as to carry out the study of stand quality evaluation.The main conclusions are as follows:(1)Based on the six theoretical growth equations,the Richards equation was determined to be the optimal model for fitting the relationship between dominant height and stand age,and the site index model was derived by difference form.In the study area,the site index of subcompartment ranged from 11-19 m,with an average value of 15.8m.The site index value was used as the evaluation standard for the suitability of P.massoniana.There were 142 subcompartments that were most suitable for the growth of P.massoniana,914 subcompartments that were more suitable for the growth of P.massoniana,and 230 subcompartments that were unsuitable for the growth of P.massoniana.(2)Taking the suitability category of P.massoniana as the dependent variable and the site factor as the independent variable,naive Bayes,support vector machine and random forest were used to construct the suitability evaluation model of P.massoniana,and the importance of site factor was evaluated.The results show that: the fitting accuracy of the random model was86.21%,and the generalization accuracy was 89.42%,which has the best classification accuracy.Altitude,slope degree,soil thickness,gravel content and humus layer thickness are the key factors affecting the growth suitability of P.massoniana,while the influence of parent rock and soil type is relatively weak.The application of random forest model in tree species suitability study can be an important technical support for the construction of national reserve forest base and the improvement of stand quality.(3)In accordance with the principles of feasibility,hierarchy and dynamics,and considering the construction objectives of national reserve forest,the stand quality evaluation index system was established by selecting three factors of stand growth status,site conditions,stand structure and stability through frequency analysis method,expert consultation method and theoretical analysis.There were 15 evaluation indexes including: average diameter at breast height,average tree height,stock per unit area,annual average growth,soil thickness,humus layer thickness,gravel content,slope degree,slope aspect,altitude,stand community structure,large diameter material ratio,the degree of forest health,tree species composition coefficient,canopy density.(4)The analytic hierarchy process is combined with the entropy weight method,and the distance function discriminant method is introduced to determine the index weight.The results show that: in the criterion layer,the weight values were in the order of stand growth status(0.3810),stand structure and stability(0.3456),and site conditions(0.2734);in the index layer,the stock per unit area(0.1463),tree species composition coefficient(0.1425)and average tree height(0.1119)contributed the most to stand quality,while the gravel content(0.0401)and humus layer thickness(0.0381)contributed the least.(5)The matter-element analysis was used to evaluate the stand quality of P.massoniana in the study area.It was found that the stand quality grade was mainly "medium" and "good",and there were 862 subcompartments in total,accounting for 70.44% of the total area.The number of subcompartment belonging to "excellent","poor" and "bad" grades was relatively small,accounting for 8.25%,10.59% and 10.72% of the survey area,respectively.There was still a lot of room for improvement in stand quality.According to the statistical analysis of the evaluation results from different perspectives,with the improvement of age group and site index,the forest quality showed a rising trend,and the number of subcompartment with quality grade of "excellent" and "good" gradually increased.At the same time,corresponding management suggestions were put forward for stands of different quality levels to improve stand quality. |
PDF Full Text Request