Waveform- and Terrestrial Lidar Assessment of the Usual (Structural) Suspects in a Forest Canopy

Abstract:

Forest inventory has evolved from standard stem diameter-height relationships, to coarse canopy metrics, to more involved ecologically-meaningful variables, such as leaf area index (LAI) and even canopy radiative transfer as a function of canopy gaps, leaf clumping, and leaf angle distributions. Accurate and precise measurement of the latter set of variables presents a challenge to the ecological and modeling communities; however, relatively novel remote sensing modalities, e.g., waveform lidar (wlidar) and terrestrial lidar systems (TLS), have the potential to adress this challenge. Research teams at Rochester Institute of Technology (RIT) and the Virginia Commonwealth University (VCU) have been collaborating with the National Ecological Observation Network (NEON) to assess vegetation canopy structure and variation at the University of Michigan Biological Research Station and the NEON Northeast domain (Harvard Forest, MA). Airborne small-footprint wlidar data, in-situ TLS data, and first-principles, physics-based simulation tools are being used to study (i) the impact of vegetation canopy geometric elements on wlidar signals (twigs and petioles have been deemed negligible), (ii) the analysis of airborne wlidar data for top-down assessment of canopy metrics such as LAI, and (iii) our ability to extract “bottom-up” canopy structure from TLS using scans registered to each other using a novel marker-free registration approach (e.g., basal area: R²=0.82, RMSE=7.43 m²/ha). Such studies indicate that we can potentially assess radiative transfer through vegetation canopies remotely using a vertically-stratified approach with wlidar, and augment such an approach via rapid-scan TLS technology to gain a better understanding of fine-scale variation in canopy structure. This in turn is key to quantifying and modeling radiative transfer based on understanding of forest canopy structural change as a function of ecosystem development, climate, and anthropogenic drivers.