EVALUATING WHEAT NITROGEN STATUS WITH CANOPY REFLECTANCE INDEXES AND DISCRIMINANT-ANALYSIS

Nitrogen deficiencies can seriously reduce yield and economic returns for farmers, Remote sensing could provide inexpensive, large-area estimates of N status and be used to monitor N status since leaf chlorophyll (Chl) A content is mainly determined by N availability. The objective was to determine if remote sensing of wheat (Triticum aestivum L.) Chi A content would provide a rapid estimation of wheat N status. We measured the reflectance of a wheat crop submitted to five different fertilization treatments throughout the growth cycle, We tested several empirical reflectance indices of pigment content: reflectance at 550 nm (R550), reflectance at 680 nn (R680), three parameters of the red edge [wavelength (lambda(re)), amplitude in the first derivative of the reflectance spectra (dR(re)), and sum of amplitudes between 680 and 780 nn in the first derivative of the reflectance spectra (Sigma dR(680-780nm))], and pigment simple ratio (PSR) and normalized pigment chlorophyll index (NPCl) (indices of carotenoid/Chl ratio). We also measured leaf CN A and N content, and leaf area index. There were significant correlations between canopy Chi A content and R550, R680, and all the red edge parameters. The NPCl and PSR followed phenological evolution of the carotenoids/Chl A ratio and separated the different treatments. BY discriminant analysis based on the pigment indices reflectance at 430 nn (R430), R550, R680, lambda(re), dR(re), and NPCl, each reflectance spectrum can be assigned to a different N status class. Thus, the use of these optical techniques offers a potential for assessing N status of wheat.