A SIMPLE SPECTRAL INDEX USING REFLECTANCE OF 735 nm IN MAPPING OF NITROGEN STATUS OF RICE CANOPY

摘要

Nitrogen (N) is the most important nutrient to increase and stabilize yield of paddy rice and the spatial distribution of N status of rice plants within field is the primary information needed for precision management of N fertilizer. This study developed a simple spectral index using dR/dlambda 735 to map N status of rice canopy and then validated the spectral model using a simplified imaging system. Results indicated that the first derivative values of canopy reflectance spectra at 735 nm (dR/dλ |_(735)) were linearly related to plant N concentration (R~2=0.679, P ＜ 0.001) measured at the panicle formation stage of the first cropping season of 2001. The linear relationship of N-dR/dλ |_(735) was better than that of N-NDVI relationship (R~2=0.471, P=0.0023), and remained valid when pooling more data from different locations of different cropping seasons and years (R~2=0.514, P＜0.001). The composite regression equation provided fair results (r=0.554~(**)) in predicting plant N concentration by using another set of spectral data from varied locations and seasons. A simplified imaging system using a CCD (charge-coupled device) cameras and band filters of 730 nm and 740 nm was assembled and mounted on a mobile lifter and a man-carry helicopter to take spectral images of rice canopies for mapping the N status within fields. Results showed that the simplified imaging system was able to provide field maps of canopy N concentration with reasonable accuracy (r=0.465 to 0.912, RMSE= 0.100 to 0.550).