Study On Inversion Model Of Picea Schrenkiana Var.tianschanica Stand Volume

Forest stock volume is one of the important indicators reflecting the quality of forest resources,and it is also a key parameter for estimating forest biomass.Picea schrenkiana var.tianschanica is the main single-quality tree species in Xinjiang mountain natural forest,and its stock volume survey has always been the focus of mountain forest resources survey.Then,the volume table method is used for measurement,which has a large workload and a long cycle.It is particularly urgent to extract the technical methods of forest accumulation information conveniently and quickly.In view of this,this study took the Tianshan spruce forest in the internship forest farm of Xinjiang Agricultural University in the middle of the Tianshan Mountains as the research object,based on airborne lidar images,the terrain factors(slope,aspect,altitude)and stand factors(tree height,number of plants per hectare)were extracted,and the multiple regression relationship between the stock volume and the stand age,site quality,and stand density was constructed to construct Picea schrenkiana var.tianschanica.Stand volume inversion model.The main findings are as follows:For the estimation model of Picea schrenkiana var.tianschanica forest volume based on systematic sampling technology,referring to the Schumacher harvest estimation model,using the systematic sampling method to set up the quadrats,with the relevant factors of site quality and stand density as independent variables,the inversion model of the Picea schrenkiana var.tianschanica stand volume was constructed.The average accuracy of the model test reached 91.49%,and the fitting effect was good,indicating that the stand volume was closely related to factors such as slope aspect,slope,altitude,age,tree height,and the number of plants per hectare.In the inversion model of Picea schrenkiana var.tianschanica forest volume based on airborne lidar,the airborne lidar data were used to extract factors such as the average tree height,number of trees per hectare,slope,slope aspect and altitude,and the extraction accuracy was relatively high.reached 89.44%,84.76%,88.12%,87.25% and 97.20% respectively;The extracted remote sensing feature parameters were substituted into the Picea schrenkiana var.tianschanica forest volume estimation model based on systematic sampling technology,and an airborne lidar-based Picea schrenkiana var.tianschanica forest volume inversion model was constructed.The average accuracy of the model test reached 90.89%.The sample test shows that the model has a good inversion effect on the volume of Picea schrenkiana var.tianschanica forest.In terms of the inversion measurement error model of Picea schrenkiana var.tianschanica forest volume,taking the inversion model of Picea schrenkiana var.tianschanica forest volume based on airborne lidar as a mathematical function relationship,the measurement error method is used to construct the inversion measurement error of Picea schrenkiana var.tianschanica forest volume.The average test accuracy of the model is 93.41% at the quadratic level and 86.54% at the small-class level.The model accuracy has been significantly improved compared with the original regression model,and the small-class survey accuracy is greater than 85%.According to the technical regulations for forest resource planning and design surveys,the small-class survey accuracy belongs to the grade "A".