DOI: 10.1016/j.jag.2014.09.016
Scopus记录号: 2-s2.0-84919356322
论文题名: Sub-footprint analysis to uncover tree height variation using ICESat/GLAS
作者: Tian J ; , Wang L ; , Li X
刊名: International Journal of Applied Earth Observation and Geoinformation
ISSN: 15698432
出版年: 2015
卷: 35, 期: PB 起始页码: 284
结束页码: 293
语种: 英语
英文关键词: Gaussian decomposition
; GLAS
; Levenberg-Marquardt
; Sub-footprint analysis
; Tree height
Scopus关键词: altimeter
; forest cover
; forestry
; Gaussian method
; height
; ICESat
; tree
; waveform analysis
; China
; Ejina
; Nei Monggol
; Populus euphratica
英文摘要: Detailed forest height data are an indispensable prerequisite for many forestry and earth science applications. Existing research of using Geoscience Laser Altimeter System (GLAS) data mainly focuses on deriving average or maximum tree heights within a GLAS footprint, i.e. an ellipse with a diameter of 65 m. However, in most forests, it is likely that the tree heights within such ellipse are heterogeneous. Therefore, it is desired to uncover detailed tree height variation within a GLAS footprint. To the best of our knowledge, no such methods have been reported as of now. In this study, we aim to characterize tree heights' variation within a GLAS footprint as different layers, each of which corresponds to trees with similar heights. As such, we developed a new method that embraces two steps: first, a refined Levenberg-Marquardt (LM) algorithm is proposed to decompose raw GLAS waveform into multiple Gaussian signals, within which it is hypothesized that each vegetation signal corresponds to a particular tree height layer. Second, for each layer, three parameters were first defined: Canopy Top Height (CTH), Crown Length (CL), and Cover Proportion (CP). Then we extracted the three parameters from each Gaussian signal through a defined model. In order to test our developed method, we set up a study site in Ejina, China where the dominant specie is Populus euphratica. Both simulated and field tree height data were adopted. With regard to the simulation data, results presented a very high agreement for the three predefined parameters between our results and simulation data. When our methods were applied to the field data, the respective R2 become 0.78 (CTH), CL (R2 = 0.76), CP (R2 = 0.74). Overall, our studies revealed that large footprint GLAS waveform data have the potentials for obtaining detailed forest height variation. © 2014 Elsevier B.V. Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/79594
Appears in Collections: 气候变化事实与影响
There are no files associated with this item.
作者单位: Capital Normal University, Key Lab of 3D Information Acquisition and Application MOE, Beijing, China; Department of Geography, State University of New York at Buffalo, Buffalo, NY, United States
Recommended Citation:
Tian J,, Wang L,, Li X. Sub-footprint analysis to uncover tree height variation using ICESat/GLAS[J]. International Journal of Applied Earth Observation and Geoinformation,2015-01-01,35(PB)