There is an increasing recognition of the value of forests for carbon storage, and hence interest in estimating Above Ground Biomass (AGB) for individual trees and across forests. Field-based inventories of AGB for hundreds of trees based on variables such as height and Diameter at Breast Height (DBH) can be costly and labor intensive, and hence there is increasing interest in evaluating remote sensing methods as an alternative. In this study, we assessed the suitability of Light Detection and Ranging (LiDAR) derived Canopy Height Model (CHM) based on drone data to estimate tree heights and hence AGB in a mixed species (56 species) even aged (∼10 years old) arboretum in subtropical Queensland, Australia. First, field-based estimates of AGB were obtained for a stratified random sample of 287 trees across the arboretum based on height, DBH and species wood density. Then CHM values were obtained based on LiDAR data for the whole arboretum and compared with the field data. The CHM values were correlated with field data for individual trees including height (r = 0.85), DBH (0.52), and AGB (0.52), using Pearson’s correlation coefficients. Using a linear regression, CHM was used to predict the height (R2 = 0.73, height = 1.9436 + 0.8471 *CHM), and AGB (R2 = 0.43, ln(AGB) = 0.4904 + 0.1335 *CHM) for individual trees. Incorporating other remote sensing data, such as Sentinel-2 derived vegetation indices (vegetation index, enhanced vegetation index and leaf area index) did not significantly improve the accuracy of the AGB estimates per tree. The outcome of this work confirms that LiDAR derived CHM is a good predictor of individual tree heights and a moderate predictor of AGB and could therefore be used as a proxy for biomass estimation for individual trees and over urban forests.