Updating forest road networks using single photon LiDAR in northern Forest environments

Abstract Knowledge about the condition and location of forest roads is important for management. Coupling accurate road information with planning conservation strategies supports resource In Canada, spatial data forestry networks are available provincially; however, they lack accuracy, up-to-date on key attributes such as width missing. this study, we apply a novel approach to update characterize conditions in Ontario’s Boreal Great Lakes—St. Lawrence (GLSL) Forest regions. We use airborne laser scanning (ALS), facilitate identification across densely forested landscapes. categorized into four classes based driveable width, edge vegetation, well surface degradation derived from high-density Single Photon LiDAR (SPL) data. Using extraction method, produced probability raster map centerlines. validated attribute using Global Navigation Satellite System (GNSS) ground truth two Ontario management units, boreal GLSL. Road segments some regions have been altered account land cover changes, flooding or fallen trees. other situations, path may deviate planned layout road, which not always followed field. Our results highlight inaccuracies existing networks, 30 per cent ‘Full access’ 29 ‘Partial being undriveable by standard vehicles 45 ‘Status unknown’ roads, make up 48 pre-existing network, vehicles. Results show that average positional accuracy updated centerlines 0.4 m, error 2 m. The production spatially characterizing large often minimal available.