Estimation of high density wetland biomass: combining regression model with vegetation index developed from Worldview-2 imagery

被引:9
作者
Adam, Elhadi M. I. [1 ]
Mutanga, Onisimo [1 ]
机构
[1] Univ KwaZulu Natal, Discipline Geog, ZA-3209 Pietermaritzburg, South Africa
来源
REMOTE SENSING FOR AGRICULTURE, ECOSYSTEMS, AND HYDROLOGY XIV | 2012年 / 8531卷
关键词
wetland biomass; red edge bands; random forest algorithm; variable importance; RANDOM FORESTS; HYPERSPECTRAL IMAGERY; ABOVEGROUND BIOMASS; CYPERUS-PAPYRUS; HABITAT LOSS; CLASSIFICATION; COVER;
D O I
10.1117/12.970469
中图分类号
S2 [农业工程];
学科分类号
0828 ;
摘要
The saturation problem associated with the use of NDVI for biomass estimation in high canopy density vegetation is a well-known phenomenon. Recent field spectroscopy experiments have shown that narrow band vegetation indices computed from the red edge and the NIR shoulder can improve the estimation of biomass in such situations. However, the wide scale unavailability of high spectral resolution satellite sensors with red edge bands has not seen the up-scaling of these techniques to spaceborne remote sensing of high density biomass. This paper explored the possibility of estimating biomass in a densely vegetated wetland area using indices computed from Worldview-2 imagery, which contains a red edge band centred at 725 nm. Indices derived from the red edge band and the NIR shoulder yielded higher accuracies (R-2 = 0.71) for biomass estimation as compared to indices computed from other portions of the electromagnetic spectrum. Predicting biomass on an independent test data set using the Random forest algorithm and 3 NDVIs computed from the red edge and NIR bands yielded a root mean square error of prediction (RMSEP) of 441g/m(2) (13 % of observed mean biomass) as compared to the traditional spectral bands. The results demonstrate the utility of Worldview-2 imagery in estimating and ultimately mapping vegetation biomass at high density - a previously challenging task with broad band satellite sensors.
引用
收藏
页数:9
相关论文
共 45 条
[1]  
Abdel-Rahman E. M., 2009, IEEE INT GEOSC REM S, pIV
[2]  
Adam E. M., INT J REMOT IN PRESS
[3]   Multispectral and hyperspectral remote sensing for identification and mapping of wetland vegetation: a review [J].
Adam, Elhadi ;
Mutanga, Onisimo ;
Rugege, Denis .
WETLANDS ECOLOGY AND MANAGEMENT, 2010, 18 (03) :281-296
[4]   Spectral discrimination of papyrus vegetation (Cyperus papyrus L.) in swamp wetlands using field spectrometry [J].
Adam, Elhadi ;
Mutanga, Onisimo .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2009, 64 (06) :612-620
[5]  
[Anonymous], 2006, P INT PREC FOR S
[6]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[7]   Evaluation of Random Forest and Adaboost tree-based ensemble classification and spectral band selection for ecotope mapping using airborne hyperspectral imagery [J].
Chan, Jonathan Cheung-Wai ;
Paelinckx, Desire .
REMOTE SENSING OF ENVIRONMENT, 2008, 112 (06) :2999-3011
[8]   Estimating aboveground biomass of grassland having a high canopy cover: an exploratory analysis of in situ hyperspectral data [J].
Chen, Jin ;
Gu, Song ;
Shen, Miaogen ;
Tang, Yanhong ;
Matsushita, Bunkei .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2009, 30 (24) :6497-6517
[9]   Estimation of green grass/herb biomass from airborne hyperspectral imagery using spectral indices and partial least squares regression [J].
Cho, Moses Azong ;
Skidmore, Andrew ;
Corsi, Fabio ;
van Wieren, Sipke E. ;
Sobhan, Istiak .
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION, 2007, 9 (04) :414-424
[10]   Remote sensing estimates of boreal and temperate forest woody biomass: carbon pools, sources, and sinks [J].
Dong, JR ;
Kaufmann, RK ;
Myneni, RB ;
Tucker, CJ ;
Kauppi, PE ;
Liski, J ;
Buermann, W ;
Alexeyev, V ;
Hughes, MK .
REMOTE SENSING OF ENVIRONMENT, 2003, 84 (03) :393-410