Estimation of Forest Aboveground Biomass Using Backpack and UAV LiDAR Data
The application of Light Detection and Ranging(LiDAR)technology in forestry surveys is widespread,as it can accu-rately capture vertical forest structure information.This study verified the feasibility of using backpack LiDAR combined with field survey plots to replace traditional field surveys.Additionally,by integrating UAV-LiDAR data and employing multivariate stepwise regression(MSR),support vector machine(SVM),and random forest(RF)algorithms,models for estimating aboveground biomass(AGB)were established and compared.The research results showed that:(1)Under manual intervention,single tree parameters extracted using backpack LiDAR were highly correlated with measured values,with a coefficient of determination(R2)of 0.98 and a root mean square error(RMSE)of 0.35 cm for average diameter at breast height,and R2 of 0.96 and RMSE of 0.63 m for average tree height.(2)In the AGB estimation model built using bio-mass samples from backpack LiDAR and UAV-LiDAR data,the random forest model performed the best(R2=0.75,RMSE=23.58 t/hm2),followed by the support vector machine model(R2=0.63,RMSE=30.49t/hm2),with the multivariate step-wise regression model showing the lowest performance(R2=0.54,RMSE=35.60t/hm2).The high accuracy of individual tree diameter at breast height and tree height obtained from backpack LiDAR allows for replacing field surveys and expan-ding sample coverage,enabling rapid estimation of forest biomass based on airborne LiDAR data.This provides a feasible method for large-scale forest biomass estimation.
万方数据
维普
