Analysis Of Leaf Optical Characteristics Based On The Global Leaf Spectral Property Database

Leaf Area Index(LAI)is a key parameter to represent vegetation structural features.Green leaf is one of the most important parts of vegetation canopy.The vegetation spectrum under common condition is mainly controlled by foliage structure.Long time series LAI remote sensing inversion algorithms,like MODIS LAI/FPAR algorithms,use only a few leaf spectras to represent the global leaf spectral characteristics throughout the year.while due to the variation of leaf spectra,it may introduce uncertainties to LAI remote sensing products.Therefore,it's essential to consider leaf spectral features in the LAI inversion algorithms.An amount of spectrum databases containing leaf spectrum of different vegetation species,locations and time phase and corresponding biochemical parameters have been constructed to provide support for the analysis of spectral characteristics of leaves.This paper mainly uses the leaf spectral database LOPEX '93,ANGERS ' 03,Spectral library of typical ground objects in China and field experimental data to analyze the effects of spectral characteristics of different plant species and different climate zones on MODIS reflectance of specific channels by parameters(e.g.red edge parameters,yellow edge parameters and spectral index).Finally,based on MODIS LAI inversion algorithm and look up table,we explored the effects of leaf spectral variation on LAI inversion processes and further to provide prior information for the development of new LAI inversion algorithm to obtain high accuracy LAI products.The result suggests that: the leaf spectral has a wide range of variation.There exist significant differences of leaf spectra features between different vegetation types.Even the same vegetation type,leaf spectra features under different climate zone or phenology stage present regular changes.The difference between maple leaf and phoenix tree's leaf could reach 28 nm at most,red edge position of marine climate zone is 13 nm larger than that of continental monsoon climate in average.The corn leaf yellow edge gradient gradually increased with growth process.the leaf spectra differences mainly have influences on the reflectance in red and green band(green band is most sensitive to leaf spectra variation);In LAI inversion algorithms,the error could reach 3 if only considering vegetation types rather than leaf spectra differences.