Publications

Wang, FM, Huang, JF, Wang, XZ (2009). Normalized Difference Ratio Pigment Index for Estimating Chlorophyll and Cartenoid Contents of in Leaves of Rice. SPECTROSCOPY AND SPECTRAL ANALYSIS, 29(4), 1064-1068.

Abstract
The objective of the present study was (1) to construct new pigment indices based on making full use of all the spectral bands in the 350-2500 nm region and (2) to compare the performance of these new pigment indices with that of the published normalized difference ratio pigment indices in estimating pigment content of rice. The 252 leaves of rice were sampled at different development stages, representing a wide range of pigment contents. The hyperspectral reflectance of leaves of rice and the corresponding chlorophyll contents and carotenoid contents were measured. A rigorous method using all the wavebands in the range of 350-2500 nm was applied to generate all possible two-band normalized difference pigment indices, and then the linear models between these indices and chlorophyll and carotenoid contents were constructed. Finally, the index with the highest determination coefficients was selected as the optimal index for corresponding pigment. The model was tested and these selected indices were compared with the published indices. The result indicated that the indices E(R-1729-R-707)/(R-1729+R-707), (R-1554-R-572)/(R-1554+R-572), (R-1729-R-706)/(R-1729+R-706), (R-1536-R-707)/(R-1536+R-707)] can relatively accurately estimate chlorophyll and cartenoid contents. The reference bands of the these new indices are mainly located in short wave infrared spectral region, which indicate that the spectral bands in the short wave infrared region are significant to construct normalized ratio pigment index., while the index bands of these indices are mostly in the region near 700 nm, and the longer bands of green region in the next place. Compared to the published indices, in general, the new indices give more accurate estimation of chlorophyll and carotenoid contents, or they behave the same as the best published index.

DOI:
10.3964/j.issn.1000-0593(2009)04-1064-05

ISSN:
1000-0593