| Binary standing tree volume table is one of the commonly used tables in forest resources investigation.In the forest resource survey,the binary standing tree volume table of different regions and different species is often used to measure the volume of single tree and stand volume.Therefore,it is of great significance to compile the binary standing tree volume table with strong applicability and high accuracy.The existing binary standing tree volume table of Cunninghamia lanceolata in Hunan Province is prepared according to the standing tree volume formula and corresponding parameter values provided in the standing tree volume table(LY208-77)issued by the former Ministry of Forestry,and has been used up to now.With the passage of time,the change of ecological environment and the improvement of forest management level,the existing binary standing tree volume table of Cunninghamia lanceolata has a large deviation in the calculation of forest volume and stand volume,which directly affects the accurate measurement and quality evaluation of forest resources.Therefore,it is very necessary to test and adjust the existing binary standing tree volume model of Cunninghamia lanceolata in Hunan Province.The existing binary standing tree volume model of Cunninghamia lanceolata in Hunan Province(V=aDbHc)as the research object,and based on the measured data of 973 Cunninghamia lanceolata analytical trees in seven regions,including Changde,Chenzhou,Hengyang,Yongzhou,Shaoyang,Xiangxi and Yiyang,tests and analyzes the data in different regions,different diameter levels and different tree levels,drawing the binary distribution diagram of DBH and tree height for volume error,and looks for the error characteristics of the volume model,the paired T test was used to analyze the error characteristics of the model;Based on the Yamamoto volume formula,the variable parameter model of Cunninghamia lanceolata standing tree volume is further constructed by the method of re-parameterization,and the mixed effects model and dummy variable model of Cunninghamia lanceolata standing tree volume are respectively constructed with the region as the random variable and dummy variable,and the existing binary standing tree volume model of Cunninghamia lanceolata in Hunan Province is optimized and adjusted.(1)The existing binary standing tree volume model of Hunan Province was tested.In different regions,the average system deviation(MSB)of Chenzhou was4.358%,and the total relative deviation(TRB)was as high as-5.0898%,which were not controlled within the standard range.The existing binary standing tree volume model was not applicable to Chenzhou.Through the analysis of volume error by dividing different diameter steps,the existing binary standing tree volume model is not suitable for all diameter steps,especially for 24cm diameter steps.Both TRB and MSB are up to-7%,with large error.From the perspective of different tree levels,the error at the 18m tree level is large.By drawing the binary distribution diagram of DBH and tree height for the volume error,it is found that the error of analytic trees with large diameter order and high tree height is generally large.The paired T-test method was used to compare and analyze the existing volume model of Cunninghamia lanceolata in Hunan with the actual volume value,and the results showed that there were significant differences between the two.(2)Based on Yamamoto volume formula,a variable parameter model is constructed by reparameterization.Coefficient of determination(R2)of fixed parameter model and variable parameter model are above 0.95.Test the test data,and the TRB and MSB of the model are within ±3%,the method of diameter division was used to test.The total relative deviation and average systematic deviation of the fixed parameter model in the 10cm,22cm and 24cm diameter steps exceeded ± 3%.The total relative deviation and average systematic deviation in the 24cm diameter steps exceeded ± 7%,the variable parameter model has smaller deviation in each diameter order,and better fitting effect.(3)The mixed effects model and dummy variable model were constructed with regions as random variables and dummy variables.Coefficient of determination of the mixed effects model and dummy variable model were both above 0.95.On the whole,the TRB and MSB of the dummy variable model tended to zero,while the TRB of the mixed effects model in the 10cm,12cm and 24cm diameters exceeded ±3%,the Akchi information criterion and Bayesian criterion of dummy variables are smaller,which are obviously better than the mixed effects model.(4)The use of fixed parameter models,variable parameter models,mixed effects models,and dummy variable models to construct a binary standing tree volume model for Cunninghamia lanceolata in Hunan Province can improve the accuracy of Cunninghamia lanceolata to some extent,and their deviations are also controlled within the standard range.Taking into account the accuracy of quadrature,error characteristics and the practicality of compiling and applying the binary standing tree volume for Cunninghamia lanceolata in Hunan,it is recommended to use the variable parameter model V=0.00007595D1.6270.002764(D+H)H1.146+0.002483(D+H)as the binary standing tree volume model for Cunninghamia lanceolata in Hunan Province. |
PDF Full Text Request