Quantifying vertical and horizontal stand structure using terrestrial LIDAR in the Pacific Northwest forests

Background/Question/Methods

Stand level spatial distribution is a fundamental part of forest structure that influences many ecological processes and ecosystem functions. Vertical and horizontal spatial structure provides key information for forest management. Although horizontal stand complexity can be measured through stem mapping and spatial analysis, vertical complexity within the stand remains a mostly visual and highly subjective process. Tools and techniques in remote sensing, specifically LiDAR, provide three-dimensional datasets that can help get at three-dimensional forest stand structure. Although aerial LiDAR (ALS) is the most widespread form of remote sensing for measuring forest structure, it has a high omission rate in dense and structurally complex forests. In this study we used terrestrial LiDAR (TLS) to obtain high resolution three dimensional point clouds of plots from stands that vary by density and composition. We used point cloud slicing techniques and object-oriented image analysis (OBIA) to produce canopy profiles at multiple points of vertical gradient. At each height point we produced segments that represented canopies or parts of canopies for each tree within the dataset. The resulting canopy segments were further analyzed using landscape metrics to quantify vertical canopy complexity within the stand.

Results/Conclusions

Based on the developed method, we have successfully created a tool that utilizes three dimensional spatial information to accurately quantify the vertical structure of forest stands. Preliminary results show significant differences in the number and the total area of the canopy segments and gap fraction between each vertical slice within and between individual plots. We find a positive relationship between the stand density and composition and the vertical canopy complexity. The methods described in this research make it possible to create horizontal stand profiles at any point along the vertical gradient of the stand with high frequency, therefore providing ecologists with invaluable dataset that measures horizontal and vertical stand structure.