Modeling stand-level forest attributes using lidar and Common Stand Exam data

Traditional forest inventories provide important information to forest managers regarding stand volume, structure, and species composition. While crucial for making informed decisions, forest inventories can be time intensive, costly, and acquisition can delay forest management actions. In some cases, publicly available and large-scale LiDAR datasets can serve as a means for assisting with or even substituting for pedestrian methods when collecting forest inventory data. This study focuses on the development of a new geospatial methodology and model development where LiDAR data was leveraged to recreate Common Stand Exam (CSE) results. CSE protocols are the U.S. Forest Service’s approach to conducting forest inventories, with Live Tree Stocking and Volume reports being major outputs following field data acquisition. Modelling efforts yielded statistically significant similarities in BA, TPA, board-feet volume, and tonnage volume when comparing traditionally acquired CSE data versus LiDAR-based analysis. While lidar-based approaches might not be appropriate for every forest management objective, these results demonstrate that they have the potential to be leveraged in scenarios where major forest metrics are required. This could represent significant time and cost efficiency for forest managers who are confronted with challenging deadlines, fiscal limitations, and harsh environmental conditions.