Authors: L. Monika Moskal*, University of Washington, Travis Axe, University of Washington
Topics: Environment, Remote Sensing, Natural Resources
Keywords: lidar, structure from motion, leaf area index
Session Type: Paper
Start / End Time: 1:10 PM / 2:50 PM
Room: Harding, Marriott, Mezzanine Level
Presentation File: No File Uploaded
We demonstrate the estimation of effective leaf area index (eLAI) using two remote-sensing techniques: discrete-return Airborne Laser Scanning (ALS) and airborne Structure-from-Motion (SfM). The study examines riparian forest-buffers in a watershed in Washington State which was chosen for both its hydrologically complex landscape and its range of riparian forest types. Hemispherical photos were taken at each of the 113 plots and were the source of eLAI reference-values. These reference data were compared to the output of ALS analysis, which replicated several models used in similar studies. These results showed that the penetration rate of ALS first returns was strongly related to eLAI even when keeping the elevation threshold of penetration consistent with the actual height of the field camera. Models that tested light attenuation variations in accordance with of the Beer-Lambert law saw similar results. However, these were more limited due to the complexity of leaf angle distribution, canopy structure, and terrain. The reference data was then compared to SfM output, which utilized the point cloud of a digital surface model, rendered from airborne photography. A multivariate linear regression utilized the distribution elevation values of upper-canopy point returns and the elevation values representing mid and max stand-level totals for each observation's respective point cloud. A spectral analysis yielded a second-order statistics grey-level co-occurrence matrix entropy: adding this variable further improved the regression results. SfM performed less well in the Beer-Lambert approach because of the relative lack of appropriate ground points, even when altering the elevation of the penetration threshold.