DOI: 10.1016/j.jag.2013.04.004
Scopus记录号: 2-s2.0-84897585295
论文题名: Comparative analysis of different uni- and multi-variate methods for estimation of vegetation water content using hyper-spectral measurements
作者: Mirzaie M ; , Darvishzadeh R ; , Shakiba A ; , Matkan A ; A ; , Atzberger C ; , Skidmore A
刊名: International Journal of Applied Earth Observation and Geoinformation
ISSN: 15698432
出版年: 2014
卷: 26, 期: 1 起始页码: 1
结束页码: 11
语种: 英语
英文关键词: ANN
; Hyper-spectral data
; Narrow band indices
; PCR
; PLSR
; Vegetation water content (VWC)
Scopus关键词: comparative study
; environmental monitoring
; least squares method
; model validation
; plant water relations
; principal component analysis
; risk assessment
; spectral analysis
; vegetation
; water content
英文摘要: Assessment of vegetation water content is critical for monitoring vegetation condition, detecting plantwater stress, assessing the risk of forest fires and evaluating water status for irrigation. The main objectiveof this study was to investigate the performance of various mono- and multi-variate statistical methodsfor estimating vegetation water content (VWC) from hyper-spectral data. Hyper-spectral data is influ-enced by multi-collinearity because of a large number of (independent) spectral bands being modeledby a small number of (dependent) biophysical variables. Therefore, some full spectrum methods that areknown to be suitable for analyzing multi-collinear data set were chosen. Canopy spectral reflectance wasobtained with a GER 3700 spectro-radiometer (400-2400 nm) in a laboratory setting and VWC was mea-sured by calculating wet/dry weight difference per unit of ground area (g/m2) of each plant canopy (n = 95).Three multivariate statistical methods were applied to estimate VWC: (1) partial least square regression,(2) artificial neural network and (3) principal component regression. They were selected to minimizethe problem related to multi-collinearity. For comparison, uni-variate techniques including narrow bandratio water index (RWI), normalized difference water index (NDWI), second soil adjusted vegetation index(SAVI2) and transferred soil adjusted vegetation index (TSAVI) were applied. For each type of vegetationindex, all two-band combinations were evaluated to determine the best band combination. Validationof the methods was based on the cross validation procedure and using three statistical indicators: R2,RMSE and relative RMSE. The cross-validated results identified PLSR as the regression model providing themost accurate estimates of VWC among the various methods. The result revealed that this model is highlyrecommended for use with multi-collinear datasets (R2CV= 0.94, RRMSECV= 0.23). Principal componentregression exhibited the lowest accuracy among the multivariate models (R2CV= 0.78, RRMSECV= 0.41). © 2013 Elsevier B.V. Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/79777
Appears in Collections: 气候变化事实与影响
There are no files associated with this item.
作者单位: RS and GIS Center, Shahid Beheshti University, Evin, Tehran, Iran; University of Tehran, Faculty of Geography, RS and GIS Department, Vesal Avenue, Tehran, Iran; University of Natural Resources and Life Sciences (BOKU), Vienna, Institute of Surveying, Remote Sensing and Land Information, Peter Jordan-Strasse 82, 1190 Vienna, Austria; Faculty of Geo-information Science and Earth Observation (ITC), University of Twente, Hengelosestraat 99, P.O. Box 6, 7500 AA Enschede, Netherlands
Recommended Citation:
Mirzaie M,, Darvishzadeh R,, Shakiba A,et al. Comparative analysis of different uni- and multi-variate methods for estimation of vegetation water content using hyper-spectral measurements[J]. International Journal of Applied Earth Observation and Geoinformation,2014-01-01,26(1)