Brown Bag Seminar: Atticus Stovall

University of Virginia Environmental Sciences Dept.

Abstract:

Accurately capturing forest biomass and carbon storage is imperative for addressing ecological questions and management decisions. However, everything we know about where and how much carbon exists in terrestrial plants across the globe is built on the assumption that regional- or national-scale allometry – scaling relationships common among all organisms –accurately captures growth form across the wide spectrum of plant size. Allometry is painstaking to create: trees must be cut, dried, and weighed over the span of months. This bottleneck has left most equations low in sample size and without the largest trees, which can contain around 40% of forest carbon. While airborne LiDAR has become the silver bullet for improving biomass carbon mapping, LiDAR still relies on plot-level training data inferred from this potentially unrepresentative allometry – essentially fitting a model to another model. It is clear that improvements in biomass estimation must come from “the ground up” with novel methods that are more accurate and precise than conventional approaches; terrestrial laser scanning (TLS) can facilitate this goal. In this talk I explore TLS as a calibration and validation approach in LiDAR-based biomass mapping though virtual plot-level tree volume reconstruction and biomass estimation at the scale of a single tree, a forest stand, and the landscape. My findings suggest TLS is not only appropriate for plot level biomass estimation, but may be the most effective method of significantly improving forest biomass measurement, whether it be in the form of non-destructive local allometry or full plot-level 3D reconstruction. From management to modeling, adaptation of this framework has the potential to revolutionize our understanding of forested ecosystems.

Bio:

Atticus Stovall is a forest ecologist with an emphasis on LiDAR remote sensing for precision measurements of forest biomass and carbon. During his PhD at the University of Virginia he developed methods of improving biomass mapping through direct three-dimensional modeling of trees with terrestrial LiDAR. He used this novel approach of estimating biomass to quantify the uncertainty from commonly used allometric equations and how they can potentially impact airborne and spaceborne LiDAR-based biomass mapping. He also led and collaborated on projects investigating the influence of microtopography on forest productivity in bog-lands, the relationship between fine-scale 3D forest complexity and ecosystem productivity, and landscape scale drought mortality mapping of individual trees with LiDAR. He is currently working with Lola Fatoyinbo after the recent AfriSAR field campaign to improve mangrove biomass mapping in Gabon with terrestrial LiDAR allometry.