文章基本信息

标题：A Method of Directly Estimating Stemwood Volume from Glas Waveform Parameters
本地全文：下载
作者：J. Rosette ; P. North ; J. Suárez 等
期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2007
卷号：XXXVI-3/W52
页码：344-349
出版社：Copernicus Publications
摘要：Methods of estimating stemwood volume from ICESat/ GLAS lidar waveforms are explored for a mixed temperate forest, the Forest of Dean, Gloucestershire, UK. Previous methods have used maximum canopy height estimations incorporating a digital terrain model (DTM) and requiring calibration using a sample of within-footprint tree heights. This study focuses on deriving methods which do not require such supplementary data. Maximum canopy height is estimated as the distance between Signal Begin and the ground peak within the waveform. The ground peak is determined using the centroid of either Gaussian Peak 1 or 2, identified by whichever has the greatest amplitude. This canopy height estimation was used to isolate the region of the waveform returned from the vegetation, from which heights of cumulative energy percentiles were calculated. For the tallest species within footprints, stemwood volume estimates for conifers produced R 2 of 0.59, RMSE 98.3 m 3 /ha and for broadleaf species, R 2 of 0.75, RMSE 59.1 m 3 /ha were found. Stemwood volume estimates taking account of the mixed species composition within stands were also calculated. For mixed stand estimates, R 2 of 0.66, RMSE 82.5 m 3 /ha was found for stands dominated by conifers whilst stands with greatest percentage cover provided by broadleaf species produced R 2 of 0.47, RMSE 75.6 m 3 /ha. Potential is shown for satellite lidar stemwood volume estimates to be derived directly from waveforms and therefore suggests that similar techniques could be applied where a suitable DTM or field measurements are not available
关键词：Lidar; Satellite; Vegetation; Forestry; Inventory