Selection of Optimum Reflectance Ratios for Estimating Leaf Nitrogen and Chlorophyll Concentrations of Field-Grown Cotton

often result in immoderate growth, increased input costs, and potential adverse environmental impacts, esLeaf N and chlorophyll (Chl) concentrations of cotton (Gossypium pecially water quality (Jaynes et al., 2001). Therefore, hirsutum L.) are important indicators of plant N status. Laboratory one of the goals of farm managers is to accurately detect determinations of plant tissue N are time consuming and costly. Measurements of leaf reflectance may provide a rapid and accurate means plant N status and provide N fertilizer in a timely manner of estimating leaf N and Chl. Studies were conducted to determine to improve yield, increase N use efficiency and profit, the relationships between leaf hyperspectral reflectance (400–2500 nm) and minimize N losses to the environment. Changes in and Chl or N concentration in field-grown cotton. One study consisted leaf N concentrations depend on not only environments, of four N rates of 0, 56, 112, and 168 kg N ha 1, and another study but also the stage of crop development (Oosterhuis et consisted of four mepiquat chloride (MC) rates of 0, 0.59, 1.17, and al., 1983). Recently, Bell et al. (2003) reported that the 2.34 L MC ha 1. Chlorophyll and N concentrations and reflectance of critical leaf-blade N concentrations associated with seed uppermost, fully expanded mainstem leaves were measured throughcotton yield loss were 5.4% at the pin-head square, 4.3% out the growing seasons. Reflectance at 556 and 710 nm increased sigat early flower, and 4.1% at 3 wk after flower. Therefore, nificantly as N fertilizer rate decreased. Averaged across years and multiple determinations of leaf N concentration are resampling dates, the percentage increase in reflectance at these two wavelengths was 8, 10, and 19% greater in the 112, 56, and 0 kg quired for N recommendations to optimize cotton yields. N ha 1 treatments, respectively, compared with the 168 kg N ha 1 Leaf N concentration is an important indicator for treatment. The effect of MC on leaf reflectance was more complex diagnosing plant N status (Gerik et al., 1994; Bell et than the N effect. In both the N and MC studies, a linear relationship al., 2003). Traditional methods of determining tissue was found between leaf N and a simple ratio of leaf reflectance at nutrient concentrations in a laboratory are time consum517 and 413 nm (R517/R413) (r2 0.65–0.78***). Leaf Chl concentration ing and costly. Furthermore, by the time the symptoms was associated closely with reflectance ratios of either R708/R915 or of plant nutrient deficiency become clearly visible, many R551/R915 (r2 0.67–0.76***). Our results suggest leaf reflectance can physiological processes may have been severely disbe used for real-time monitoring of cotton plant N status and N rupted by nutrient stress. fertilizer management in the field. Recent advances in remote sensing, coupled with lower cost of acquiring images, have allowed the collection of timely information on crop growth and physioN fertilization management is an imporlogical parameters temporally and spatially as affected tant issue in cotton production systems. It is more by environmental stresses. Such information can be used difficult to balance demand and supply of cotton plant for in-season crop nutrient assessment and management N nutrition compared with other nutrient fertilizers be(Filella et al., 1995; Daughtry et al., 2000; Zarco-Tejada cause of the complexity of N cycling in the soil and et al., 2000a, 2000b; Afanasyev et al., 2001). Several the indeterminate growth habit of cotton (Gerik et al., studies have assessed N status and other physiological 1998). Both deficient and excessive N negatively affects parameters of field crops using leaf or canopy spectral lint yield and fiber quality (Gerik et al., 1998; MacKenreflectance parameters (Gausman, 1982; Chappelle et zin and van Schaik, 1963). Insufficient N supply deal., 1992; Blackmer et al., 1994; Thomas and Gausman, creases leaf area (Fernandez et al., 1996; Reddy et al., 1977; Peñuelas and Filella, 1998; Peñuelas and Inoue, 1997; van Delden, 2001), photosynthesis (Ciompi et al., 2000; Zhao et al., 2003). Nitrogen deficiency causes a 1996; Reddy et al., 1997; Lu et al., 2001), and biomass decrease in leaf Chl concentration, resulting in an inproduction (Fritschi et al., 2003), resulting in lower crease in leaf reflectance in the visible spectral region yields (Howard et al., 2001; Fritschi et al., 2003). On (400–700 nm) (Buscaglia and Varco, 2002; Carter and the other hand, excessive applications of N fertilizer Estep, 2002; Read et al., 2002; Zhao et al., 2003). However, several other stresses may also result in increased D. Zhao, K.R. Reddy, V.G. Kakani, and S. Koti, Dep. of Plant and reflectance due to reduced amounts of Chl (Carter and Soil Sci., Box 9555, Mississippi State Univ., Mississippi State, MS Knapp, 2001). Furthermore, diagnosing a specific nutri39762; and J.J. Read, USDA-ARS, Crop Sci. Res. Lab., P.O. Box ent deficiency with remotely sensed data can be difficult 5367, Mississippi State, MS 39762. Contribution from Dep. of Plant when plants are subjected to deficiencies of multiple and Soil Sci., Mississippi State Univ., Mississippi Agric. and Forestry Exp. Stn. Received 8 Jan. 2004. *Corresponding author (krreddy@ ra.msstate.edu). Abbreviations: Chl, chlorophyll; DAS, days after sowing; DW, dry weight; FF, first flower; FS, first square; MC, mepiquat chloride; R i , Published in Agron. J. 97:89–98 (2005). © American Society of Agronomy reflectance at i nanometers; RD, reflectance difference; RS, reflectance sensitivity. 677 S. Segoe Rd., Madison, WI 53711 USA