Prediction of Forest Attributes with Field Plots, Landsat, and a Sample of Lidar Strips: A Case Study on the Kenai Peninsula, Alaska

To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The published article is copyrighted by the American Society for Photogrammetry and Remote Sensing and can be found at: http://www.asprs.org/Photogrammetric-Engineering-and-Remote-Sensing/PE-RS-Journals.html.In this study we demonstrate that sample strips of lidar in\ud combination with Landsat can be used to predict forest attributes\ud more precisely than from Landsat alone. While lidar\ud and Landsat can each be used alone in vegetation mapping,\ud the cost of wall to wall lidar may exceed users' financial\ud resources, and Landsat may not support the desired level of\ud prediction precision. We compare fitted linear models and\ud k nearest neighbors (kNN) methods to link field measurements,\ud lidar, and Landsat. We also compare 900 m² and 8,100\ud m² resolutions to link lidar to Landsat. An approach with\ud lidar and Landsat together reduced estimates of residual\ud variability for biomass by up to 36 percent relative to using\ud Landsat alone. Linear models generally performed better\ud than kNN approaches, and when linking lidar to Landsat,\ud using 8,100 m² resolution performed better than 900 m²