Download (PDF 1.65 MB): https://research.fs.usda.gov/download/treesearch/44093.pdf
Abstract
Remote sensing has many applications in forestry. Light detection and ranging (LiDAR) and high resolution aerial photography have been investigated as means to extract forest data, such as biomass, timber volume, stand dynamics, and gap characteristics. LiDAR return intensity data are often overlooked as a source of input raster data for thematic map creation. We utilized LiDAR intensity and elevation difference models from a recent Morgantown, WV, collection to extract land cover data in an urban setting. The LiDAR-derived data were used as an input in user-assisted feature extraction to classify forest cover. The results were compared against land cover extracted from high resolution, recent, true color, leaf-off imagery. We compared thematic map results against ground sample points collected using realtime kinematic (RTK) global positioning system (GPS) surveys and to manual photograph interpretation. Th is research supports the conclusion that imagery is a superior input for user-assisted feature extraction of land cover data within the software tool utilized; however, there is merit in including LiDAR-derived variables in the analysis.
Parent Publication
Citation
Maxwell, Aaron E.; Riley, Adam C.; Kinder, Paul. 2013. Comparison of LiDAR-derived data and high resolution true color imagery for extracting urban forest cover. In: Miller, Gary W.; Schuler, Thomas M.; Gottschalk, Kurt W.; Brooks, John R.; Grushecky, Shawn T.; Spong, Ben D.; Rentch, James S., eds. Proceedings, 18th Central Hardwood Forest Conference; 2012 March 26-28; Morgantown, WV; Gen. Tech. Rep. NRS-P-117. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station: 261-275.
