Biomass Estimation for Individual Trees using Waveform LiDAR

Abstract:

Vegetation biomass information is important for many ecological models that include terrestrial vegetation in their simulations. Biomass has strong influences on carbon, water, and nutrient cycles. Traditionally biomass estimation requires intensive, and often destructive, field measurements. However, with advances in technology, airborne LiDAR has become a convenient tool for acquiring such information on a large scale. In this study, we use infrared full waveform LiDAR to estimate biomass information for individual trees in the Sangamon River basin in Illinois, USA. During this process, we also develop automated geolocation calibration algorithms for raw waveform LiDAR data. In the summer of 2014, discrete and waveform LiDAR data were collected over the Sangamon River basin. Field measurements commonly used in biomass equations such as diameter at breast height and total tree height were also taken for four sites across the basin. Using discrete LiDAR data, individual trees are delineated. For each tree, a voxelization methods is applied to all waveforms associated with the tree to result in a pseudo-waveform. By relating biomass extrapolated using field measurements from a training set of trees to waveform metrics for each corresponding tree, we are able to estimate biomass on an individual tree basis. The results can be especially useful as current models increase in resolution.