Photosynthetic pigments are essential for plant development. Quantifying these pigments in great extensions of agricultural crops is an important objective in remote sensing for agricultural purposes. This information can be used to produce a more accurate estimation of the physiological state of the vegetation, for species discrimination and productivity estimation. The aim of the present study was to (a) evaluate the potential for estimating chlorophyll content of crop canopies, using narrow band spectral indexes, and (b) in this respect compare the performances of NDVI (a multispectral wide band index) and two narrow band vegetation indexes (R750/700 and R750/550). Experiments were carried out under greenhouse conditions whereby soybean [Glycine max (L.), Merril] was monitored with a high-resolution spectroradiometer (10 nm at 365-1,126 nm range) during the phenological cycle of the crop. Chlorophyll (a, b and total) contents were determined weekly in the laboratory. A statistical correlation analysis was performed between narrow band spectral indexes against chlorophyll content and r² coefficients near 0.84 were obtained. For NDVI r² was around 0.51. These analyses showed that R750/700 and R750/550 ratios are very useful indexes for chlorophyll determination and very effective compared with NDVI (one of the wide band indexes widely used). Thus, it can be stated that hyperspectral remote sensing has great potential for providing a reliable estimate of photosynthetic pigment content at the canopy level through evaluated indexes and other such indexes that might arise. Thus, further studies should be carried out for evaluating other indexes at the canopy level, both in the laboratory and under field conditions, using spectroradiometers and hyperspectral images, aimed at providing information for agricultural purposes.
canopy chemistry; hyperspectral remote sensing; remote sensing in agriculture; spectroradiometry
