Canadian Forest Service Publications
Within polygon grid based sampling for height estimation with LIDAR data. 2001. Wulder, M.A.; Seemann, D.; Bouchard, A. Pages 265-269 (Vol. 1) in M. Bernier and C. Duguay, editors. 23rd Canadian Symposium on Remote Sensing, Proceedings: Remote sensing in the third millennium: from global to local. August 21-24, 2001, Université Laval, Sainte-Foy, Québec. Canadian Aeronautics and Space Institute, Ottawa.
Available from: Pacific Forestry Centre
Catalog ID: 26717
The expected canopy top height of a forest stand polygon is an important forest inventory variable. The use of LIDAR, or laser altimetry, to develop a remote estimate of canopy top height is accordingly desirable. Forest stand polygons are delineated from air photographs by interpreters to indicate regions of homogeneous forest conditions. The perceived homogeneity of forest stand polygons is a complex amalgam of forest conditions grouped using visual clues and the experience of the interpreter. An averaging of all LIDAR hits within a polygon is expected to give a biased result. Placing a sampling grid over a polygon within which maximum height may be used as representative, will factor in for crown openings and differing strata.
In this research, we investigated the development of tools for extraction of the within polygon LIDAR values, such as for the processing of LIDAR values for separation of overstorey from the understorey values, and differing grid sizes and input values. The goal of the research was to develop a system for remote estimates that would result in a representative canopy height from LIDAR data. To do so, we developed a within polygon grid based sampling system. The within polygon sampling system allowed for setting of a series of parameters to investigate the sensitivity of changes in height estimates from alterations to the system parameterisation. Additionally, we compare the grid based within polygon estimates to a non-gridded control. Our results indicate understorey improved the results, the number of within grid hits to average has a nominal impact in height estimates, and that the size of the sampling grid has a large impact upon final height estimates.