Detection of Phosphorus and Nitrogen Deficiencies in Corn Using Spectral Radiance Measurements

techniques to estimate nutrient status could decrease the amount of labor needed for sampling, and could Applications of remote sensing in crop production are becoming reduce the cost associated with sampling and analysis. increasingly popular due in part to an increased concern with pollution of surface and ground waters due to over-fertilization of agricultural Destructive tissue testing is a common way to assess lands and the need to compensate for spatial variability in a field. crop N and P status. Nondestructive methods have been Past research in this area has focused primarily on N stress in crops. developed to monitor crop N status. Blackmer and Other stresses and the interactions have not been fully evaluated. A Schepers (1994) found that the chlorophyll meter was field experiment was conducted to determine wavelengths and/or a useful method of monitoring corn N status, compared combinations of wavelengths that are indicative of P and N deficiency with measuring leaf N concentration, which requires and also the interaction between these in corn (Zea mays L.). The destructive sampling. While the chlorophyll meter is a field experiment was a randomized complete block design with four good indicator of in-season N status, the technique rereplications using a factorial arrangement of treatments in an irrigated quires time and labor for data collection. The use of continuous corn system. The treatment included four N rates (0, 67, 134, and 269 kg N ha 1) and four P rates (0, 22, 45, and 67 kg P ha 1). remote sensing could help eliminate the need for extenSpectral radiance measurements were taken at various growth stages sive field sampling while still providing a good detection in increments from 350 to 1000 nm and correlated with plant N and of deficiencies. P concentration, plant biomass, grain N and P concentration, and Remote sensing is simply obtaining information about grain yield. Reflectance in the near-infrared (NIR) and blue regions an object, area, or phenomenon by analyzing data acwas found to predict early season P stress between growth stages V6 quired by a device that is not in contact with the object, and V8. Late season detection of P stress was not achieved. Plant N area, or phenomenon (Lillesand and Kiefer, 1987). Reconcentration was best predicted using reflectance in the red and cently, researchers have evaluated remote sensing techgreen regions of the spectrum, while grain yield was estimated using niques for estimating the N status of growing crops by reflectance in the NIR region, with the particular wavelengths of importance changing with growth stage. determining the appropriate wavelength or combination of wavelengths to characterize crop N deficiency. Blackmer et al. (1994) stated that light reflectance near 550 nm (green) was best for separating N treatment N is the most limiting nutrient in production differences, and could be used to detect N deficiencies of nonleguminous crops in central Nebraska and in corn. Everitt et al. (1985) studied the relationship of the Great Plains. As cropping practices become more plant leaf N concentration and leaf reflectance from 500 intensive, other nutrients will likely become limiting as well. The second most limiting nutrient for corn producto 750 nm, concluding that buffalograss (Buchloe engelm tion is often P. Current methods for estimating the Poaceae) receiving no fertilizer N had highest reflecamount of P available to growing crops include soil tance readings. Walburg et al. (1982) demonstrated that sampling or in-season plant sampling, both of which can N treatments affected reflectance in both the red and be costly and labor intensive. The use of remote sensing near-infrared (NIR) regions of the spectrum, with red reflectance increasing and NIR reflectance decreasing for N-deficient corn canopies. Blackmer et al. (1996) S.L. Osborne, Dep. of Agronomy, Univ. of Nebraska, Lincoln, NE reported that reflected radiation near 550 and 710 nm 68583, currently at USDA-ARS, Northern Grain Insects Res. Lab., was better for detecting N deficiencies compared with 2923 Medary Ave., Brookings, SD 57006; and J.S. Schepers, D.D. Francis, and M.R. Schlemmer, Dep. of Agronomy and USDA-ARS, reflectance at other wavelengths. They found that a ratio Univ. of Nebraska-Lincoln, Lincoln, NE 68583. Received 3 Apr. 2000. of the 550 to 600 nm band to the 800 to 900 nm band *Corresponding author ([email protected]). could distinguish between N treatments in irrigated corn canopies. Stone et al. (1996) demonstrated that total Published in Agron. J. 94:1215–1221 (2002).