Welcome to the 2026 GIS in Action Annual Conference hosted by the Oregon & SW Washington Chapter of the Geospatial Professional Network & Cascadia ASPRS.
LiDAR-based forest inventory has advanced over the past decade, with studies reporting cm-level DBH accuracy and sub-meter height estimates. However, most of this work was done in urban forests, plantations, or younger forests with simple structure. In Pacific Northwest natural forests, undergrowth, tall trees, and closed canopy limit high-precision GNSS, making spatially explicit inventory difficult. We tried to register Handheld Laser Scans (HLS) with RTK-based Aerial Laser Scanning (ALS) as a reference. ICP registration failed, likely due to density differences and lack of common features. Therefore, alternatively, we used a two-step process: rough alignment with 'Register-Clouds' and manual refinement using canopy gaps and crown patterns. Despite good crown matching quite well, stem positions from the two scans were far apart, and stem sizes differed from under the canopy. These uncertainties propagate through DBH, height, volume, and biomass estimates. Quantifying their impact on operational forest inventory is essential, and developing matching algorithms robust to complex environments and dense tree clusters is encouraged
