Canadian Forest Service Publications
Area-Based Mapping of Defoliation of Scots Pine Stands Using Airborne Scanning LiDAR. 2013. Vastaranta, M.; Kantola, T.; Lyytikäinen-Saarenmaa, P.; Holopainen, M.; Kankare, V.; Wulder, M.A.; Hyyppä, J.; Hyyppä, H. Remote Sensing 5: 1220-1234.
Year: 2013
Issued by: Pacific Forestry Centre
Catalog ID: 34432
Language: English
Availability: PDF (download)
Available from the Journal's Web site. †
DOI: 10.3390/rs5031220
† This site may require a fee
Abstract
The mapping of changes in the distribution of insect-caused forest damage remains an important forest monitoring application and challenge. Efficient and accurate methods are required for mapping and monitoring changes in insect defoliation to inform forest management and reporting activities. In this research, we develop and evaluate a LiDAR-driven (Light Detection And Ranging) approach for mapping defoliation caused by the Common pine sawfly (Diprion pini L.). Our method requires plot-level training data and airborne scanning LiDAR data. The approach is predicated on a forest canopy mask created by detecting forest canopy cover using LiDAR. The LiDAR returns that are reflected from the canopy (that is, returns > half of maximum plot tree height) are used in the prediction of the defoliation. Predictions of defoliation are made at plot-level, which enables a direct integration of the method to operational forest management planning while also providing additional value-added from inventory-focused LiDAR datasets. In addition to the method development, we evaluated the prediction accuracy and investigated the required pulse density for operational LiDAR-based mapping of defoliation. Our method proved to be suitable for the mapping of defoliated stands, resulting in an overall mapping accuracy of 84.3% and a Cohen’s kappa coefficient of 0.68.
Plain Language Summary
This study demonstrates how LiDAR can be used to detect and characterize forest defoliation. This illustrates the potential for LiDAR to provide more than just the standard forest inventory variables, at little or no additional cost. The study provides support for inventory-based LiDAR collection and demonstrates the potential for LiDAR to address an increasingly broad range of information needs.