Individual Tree Biomass Estimation Of Larix Gmelinii Combining Airborne LiDAR And High Resolution Images

Forest biomass estimation is the foundation of carbon cycle and carbon dynamic analysis of terrestrial ecosystem.Remote sensing technology makes up for the lack of the traditional forest biomass estimation methods;it can estimate the forest biomass rapidly and non-destructively in large temporal and spatial scales in a certain extent.Chang Chun Moon Lake National Forest Park is taken as the study area,and airborne LiDAR data and Remote sensing image with 0.2m resolution were used as the basic data in this paper.Estimate individual Larix gmelinii tree parameters and biomass of larch from the remote sensing date.In this paper,the main research results are as follows:(1)3-D geometry model of larch canopy was established.The model was established by using the individual tree's parameters that extracted from laser radar and remote sensing image,and canopy volume was calculated.The canopy volume was used to replace the original parameter as a biomass parameter in model,and the problem of low utilization rate of the biomass model of crown was solved.At the same time,the existing biomass model was reformed and the estimation precision of biomass was improved.(2)Individual Larix gmelinii compatible biomass model was set up.In order to get larch compatible biomass model,the method of second step nonlinear measurement model was used to jointly estimate the whole larch wood and biomass model by fixing the coefficient of the parameters in the trunk in model.After verification,total and canopy model test of root mean square error were 2.4736 and 0.6958 respectively,the total relative error were-0.3317 and-1.2638 respectively,the average relative error were-0.9701 and-4.4679 respectively,the average relative error absolute value were 7.1602 and 25.3964 respectively,the fitting precision were 96.12%and 91.41%respectively,thus the model fitting effect of the entire compatibility,canopy and trunk biomass was better.(3)Individual biomass remote sensing inversion.Individual tree height,crown diameter,crown high,diameter at breast height and other parameters extracted from laser radar and remote sensing image was substituted into the compatibility of biomass model,getting the estimated Individual biomass data.After analysis,the model of estimate accuracy was 4.25%.RMSE(root mean square error)was 4.0628kg.The Individual crown breadth and height estimation accuracy were 82.13%and 89.49%respectively.