Changes in forest structure after wildfires as measured by simulated spaceborne lidar waveforms

Student: Evangeline Viray

Faculty Mentor: Matthew Clark

Geography, Environment & Planning

College of Humanities, Social Sciences, and the Arts

Multispectral satellite burn indices like difference Normalized Burn Ratio (dNBR) from optical satellites have traditionally been used to assess fire severity and guide research, but it has limitations. Landsat only captures information from the top of vegetation canopies, presenting limited information of forest structure, or fuels, below the canopy. This study investigates simulated Global Ecosystem Dynamics Investigation (GEDI) spaceborne lidar as an alternative approach to mapping forest fuels. Simulated GEDI provides 3D forest structure measurements utilizing airborne laser scanning (ALS), or lidar, offering better insight into full canopy structure. Simulated GEDI does not have the geolocational, ground-finding, and noise issues found in real, on-orbit GEDI data acquired from the International Space Station. The main objectives of this research are to: 1) determine how well bi-temporal simulated GEDI metrics reflect the change in fuel loads after three recent fires in the Sonoma County area; and 2) assess significant changes in the pre- and post-fire footprints based on traditional dNBR severity. For this study, we utilized lidar data from before (2013) and after (2022) the Wallbridge, Glass, and Kincade wildfires. Using R packages, rGEDI and rGEDIsimulator, the ALS point cloud data were used to simulate GEDI. Mixed-effects and spatial regression models were used to assess the data and show the resulting impacts of the fires.