Estimation Model Of Poplar Plantation Volume Based On Airborne LiDAR Data

Airborne LiDAR(light detection and ranging,LiDAR)is an active remote sensing technology with many advantages such as high accuracy,low interference and strong penetration.It can obtain the three-dimensional spatial structure information of stand in a non-destructive way,and effectively improve the estimation accuracy of stand volume.Taking Chenwei forest farm,Sihong County,Suqian City,Jiangsu Province as the research area,taking 36 sample plots of poplar artificial pure forest with a forest age of 15 years as the research object,this study processed and analyzed the airborne LiDAR data collected in March 2019(T1)and April 2021(T2),extracted height variables and intensity variables,screened the variables by Pearson correlation analysis,and developed the optimal estimation model of stand volume.The best data source is determined by analyzing the influence of different aerial cameras and flight altitude on the modeling accuracy;According to the afforestation density,99%cumulative height percentile(AIH99)and coefficient of variation of tree height,establish models in groups to determine the optimal stock estimation model based on single period airborne LiDAR data;The dynamic change of stand volume is analyzed according to the two periods of airborne LiDAR data.The influence of afforestation density on the growth change of stand volume of poplar plantation is discussed.The results are as follows:(1)The results show that the airborne LiDAR data with flight altitude of 100 m collected by Riegl-miniVUX aerial camera has the best modeling effect.Comparing the differences between linear model and nonlinear model in stand volume estimation model,the results show that the estimation accuracy of nonlinear model is higher.(2)An improved stand average height extraction method is proposed,which can improve the effect of stand volume estimation model.Taking the sampling average height of digital vertical angle gauge points(Vth)as the characteristic variable to participate in the research and construction of the stand volume estimation model,it is found that the accuracy of the model is improved.The R2 of the stand volume estimation model is 0.664,and the RMSE and RRMSE of the test data are 12.40 and 7.40%respectively.(3)There are three grouping methods to establish the estimation model of stand volume.The first method is to build an estimation model according to the grouping of afforestation density.The result is that the estimation accuracy of high-density sample plot(R2=0.689)is better than that of non grouping modeling,and better than that of low-density sample plot(R2=0.627).The second method is to extract 99%cumulative height percentile(AIH99)for grouping modeling.lt is found that the estimation effect of the model lower than AIH99<AIH99(R2=0.675)is better than that of non grouping modeling(R2=0.664),and better than that of AIH99 ≥ AIH 99(R2=0.650);The third method is based on the grouping modeling of the coefficient of variation of tree height(Hcv).it is found that the estimation accuracy of a group of models with Hcv<Hcv(R2=0.753)is higher than the modeling effect with Hcv≥Hcv(R2=0.698),and is higher than that of non grouping modeling.This grouping method is the best among the three grouping methods.(4)Based on the two periods of airborne LiDAR data,the dynamic estimation analysis of stand volume is carried out.The results show that the effect of dynamic estimation using 2021 data is better when the 2019 airborne LiDAR data are grouped according to the coefficient of variation of tree height(Hcv).The estimation model built in groups has high accuracy,and can accurately estimate the change of stand volume during the study period.R2 of the measured change of stand volume and the estimated change reaches 0.7071.(5)The effects of afforestation density on the growth changes of average height,average DBH and volume of Poplar Plantation in the study area were discussed.It was found that the growth of average height,average diameter at breast height and volume of each stand in the sample plot of 6 m × 6 m is the largest.The variation of stand parameters in the this sample plot is the largest,the average growth of stand height is 0.93 m,the average growth of DBH is 2.12 cm,the average growth of volume is 31.50 m3/ha.The amount of change was ordered as:6 m × 6 m>4.5m ×8m>5m ×5m>3m ×8m.