The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Download
Citation
Articles | Volume XLI-B8
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLI-B8, 625–628, 2016
https://doi.org/10.5194/isprs-archives-XLI-B8-625-2016
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLI-B8, 625–628, 2016
https://doi.org/10.5194/isprs-archives-XLI-B8-625-2016

  23 Jun 2016

23 Jun 2016

IMPACTS OF TREE HEIGHT-DBH ALLOMETRY ON LIDAR-BASED TREE ABOVEGROUND BIOMASS MODELING

R. Fang R. Fang
  • College of Forestry, Oregon State University, 97331 Corvallis, Oregon, USA

Keywords: tree aboveground biomass, lidar, allometric equations, simulation, regression

Abstract. Lidar has been widely used in tree aboveground biomass (AGB) estimation at plot or stand levels. Lidar-based AGB models are usually constructed with the ground AGB reference as the response variable and lidar canopy indices as predictor variables. Tree diameter at breast height (dbh) is the major variable of most allometric models for estimating reference AGB. However, lidar measurements are mainly related to tree vertical structure. Therefore, tree height-dbh allometric model residuals are expected to have a large impact on lidar-based AGB model performance. This study attempts to investigate sensitivity of lidar-based AGB model to the decreasing strength of height-dbh relationship using a Monte Carlo simulation approach. Striking decrease in R2 and increase in relative RMSE were found in lidar-based AGB model, as the variance of height-dbh model residuals grew. I, therefore, concluded that individual tree height-dbh model residuals fundamentally introduce errors to lidar-AGB models.