PS 3-44 - CANCELLED - Using terrestrial LiDAR to quantify forest structure changes post-wildfire in a northern California mixed hardwood forest

Abstract: Using remote sensing approaches, such as terrestrial laser scanning (TLS), to understand forest structure before and after disturbance can provide critical insight related to management, fire prediction and wildlife assessment. Since TLS accurately captures spatially explicit stand measurements, it can be used to quantify fuel loads and estimate 3D forest texture metrics that describe heterogeneity (i.e., lacunarity). Here, we collected field data using traditional forest inventory approaches (diameter at breast height (DBH), species identification) on all trees >10 cm DBH in plots (400 m2) in an oak-woodland mixed hardwood forest in northern California before natural wildfire (n=24). We also scanned each plot using TLS to quantify changes in fuel loads at multiple heights (via occupied voxels) and texture metrics (via lacunarity). TLS scans were completed before disturbance, and then two times post-wildfire (1 month, 13 months). In a subset of plots (n=5), there was a significant increase in quadratic mean diameter 1 month post-wildfire (29.6 cm2 to 43.64 cm2; p = 0.04). Trees per plot significantly decreased from 29 trees to 15 trees (p = 0.06), as did lowest live crown height (5.46 m to 3.45 m; p = 0.04). Basal area did not significantly change 1 month post-wildfire (p = 0.62). TLS-derived lacunarity significantly increased following disturbance and explained 73-79% of the variance in other forest structural variables. Voxel analyses of surface fuels ( < 1 m in height) revealed regrowth of live vegetation one-year post-fire. These results imply that forest structure changes quickly following wildfire and may have important implications for forest health and future wildfire risk. Further work will examine these relationships across other ecosystem types and disturbance, including disease, and explore the use of TLS datasets to extract other relevant forest structure and texture variables.