HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Fridgen, J. L. Articles by Varco, J. J. Search for Related Content PubMed Articles by Fridgen, J. L. Articles by Varco, J. J. Agricola Articles by Fridgen, J. L. Articles by Varco, J. J. Related Collections Cotton Nutrient Management Plant Analysis Plant Nutrition Nitrogen

COTTON

Dependency of Cotton Leaf Nitrogen, Chlorophyll, and Reflectance on Nitrogen and Potassium Availability

Dep. of Plant and Soil Sci., Mississippi State Univ., Mississippi State, MS 39762

^* Corresponding author (jvarco{at}pss.msstate.edu).

Received for publication January 13, 2003.

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
In-season assessment of cotton (Gossypium hirsutum L.) leaf N and K concentration using remote sensing techniques is needed to address spatial and temporal variation of these two nutrients. The objective of this study was to evaluate the effects of varying N and K availability on cotton chlorophyll concentration and detection of leaf N utilizing reflectance properties. Fertilizer N rates of 0, 45, 90, 135, and 180 kg ha^–1 in factorial combination with K rates of 0 and 112 kg ha^–1 were applied to cotton under field conditions. Recently matured leaves on the main stem were collected at first bloom and peak bloom physiological stages of growth in 1999 and 2000 to determine leaf N and K concentrations, chlorophyll concentration, and spectral reflectance. Nutrient stress anomalies from spectral reflectance data were predicted using partial least-squares regression. Partial least squares yielded a better predictability of leaf N concentration at first bloom and peak bloom when K was adequately supplied. The greatest predictability of leaf N was observed at peak bloom in 2000 with a maximum r² of 0.77 and minimum standard error of prediction of 2.72. A red-edge shift to longer wavelengths with increased N supply was observed when K was sufficient. Utilization of leaf N concentration sampled at a coarse resolution in combination with timely and appropriate imagery may enhance nutrient management capabilities in precision agriculture so long as other nutrients are not limiting.

Abbreviations: DAP, days after planting • NIR, near infrared • PLS, partial least squares • SEP, standard error of prediction • _re, red edge

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
REMOTE SENSING has the capability of delineating stress anomalies within crop production fields as well as mapping spatial and temporal variation in crop growth. Although leaf greenness is discernible with the naked eye, reflectance in the visible region is generally not nutrient specific. Difficulties in diagnosing specific deficiencies with remote sensing are encountered when multiple nutrient deficiencies are present (Masoni et al., 1996). However, spectral reflectance characteristics of healthy vs. unhealthy plants at various stages of growth may help identify specific nutrient stresses under conditions where other environmental factors (including other nutrients) are not limiting.

Deficiencies of N and K influence leaf coloring as well as physiological efficiency of cotton (Maples et al., 1988; Stromberg, 1960). A deficiency of N in cotton generally results in lower plant biomass production and premature senescence, as evidenced by yellowing or chlorosis of older leaves. A deficiency of K in modern faster-fruiting cultivars is expressed in the upper canopy as interveinal chlorosis along margins of younger leaves beginning from early to late fruit formation (Maples et al., 1988). Thus, changes in the physiological status of plants may be estimated from measured changes in leaf pigmentation (e.g., chlorophyll a, chlorophyll b, etc.). Less-than-optimum N availability is well correlated with reduced total chlorophyll concentration in cotton leaves (Everitt et al., 1985; Longstreth and Nobel, 1980; Thomas and Gausman, 1977). Studies have also demonstrated K nutritional effects on chlorophyll concentration. Oosterhuis and Bednarz (1997) reported that chlorophyll a and total leaf chlorophyll concentrations were reduced in cotton when K was deficient. Conversely, high levels of K nutrition promoted formation of chlorophyll a and b in cucumber (Cucumis sativus cv. Brunex) leaves (Lamrani et al., 1996).

Specific detection of N nutritional status of crops utilizing remote sensing techniques is supported by reports of successful discrimination of N rates based on leaf and canopy reflectance (Blackmer et al., 1994; Chappelle et al., 1992; Gausman, 1982; Thomas and Gausman, 1977), chlorophyll meter readings (Wood et al., 1992), and chlorophyll content (Everitt et al., 1985; Longstreth and Nobel, 1980; Thomas and Gausman, 1977). Thomas and Gausman (1977) concluded that chlorophyll content was the most important independent factor affecting leaf reflectance of cotton for varying N nutrition. Recently, Buscaglia and Varco (2002) found more accurate prediction of cotton leaf N using reflectance at 550 nm compared with chlorophyll meter readings. Gausman (1982) reported that differently pigmented cotton leaves (green, red, light red or bronze, and yellow-green) showed dissimilar reflectance properties in the visible light region.

Little research has been conducted in utilizing remote sensing techniques for the discrimination of leaf N concentration relative to K availability. Remote sensing capabilities for delineating stress anomalies have concentrated on N primarily due to its importance in crop production, relationship to chlorophyll, and influence on reflectance patterns as previously stated. However, spectral differences relative to N availability might be influenced by K availability, which may not be visually evident until later physiological stages of growth (Lough, 2000). Masoni et al. (1996) reported difficulty in utilizing plant spectral properties in the detection of specific mineral deficiencies, unless the plant species and specific nutrient deficiency were known. Therefore, detection of a specific nutrient deficiency may be possible using leaf reflectance in the visible and near-infrared (NIR) regions, given no other nutrients are deficient.

A plausible goal of adapting remote sensing technologies is to predict nutritional status from visible/NIR reflectance spectra of leaves early enough in the growing season to allow for variable-rate sidedress fertilization. Prediction of nutrient stress anomalies from reflectance data requires the application of a method that considers the number of variables involved and tests for multicollinearity. Traditional multiple-regression techniques do not compensate for collinearity and often increase the risk of overfitting if reflectance at each wavelength is considered as an explanatory (X) variable (Helland, 1988). Partial least-squares (PLS) regression constructs predictive models when the number of factors exceeds sample numbers and are highly collinear (Tobias, 2000). Sudduth and Hummel (1991) used several regression techniques (i.e., PLS, principal-component regression, multiple linear regression, and stepwise multiple linear regression) to predict soil organic matter content from visible/NIR spectral reflectance data. They concluded that PLS regression had the most promise for the prediction of soil organic C content from visible/NIR reflectance. With PLS regression, the data values of both the independent and dependent variables influence the formation of the factors (Garthwaite, 1994; Tobias, 2000).

Reflectance approaching the NIR region of the spectrum is particularly sensitive to changes in leaf chlorophyll content (Carter, 1993; Everitt et al., 1985; Lichtenthaler et al., 1996). Near a wavelength of 700 nm, commonly referred to as red edge (_re), is the long wavelength limit of the visible spectrum of chlorophyll absorption and is highlighted by a reflectance derivative curve (Filella and Penuelas, 1994; Horler et al., 1983). It marks the boundary of unique optical properties of plant tissue, chlorophyll absorption of red wavelengths, and leaf scattering of NIR wavelengths (Curran et al., 1991; Horler et al., 1983). It is hypothesized that an increase in chlorophyll concentration causes a deepening and broadening of chlorophyll absorption. Shifting of the _re may be useful in detecting leaf N levels related to variations in chlorophyll content.

Red-edge determination has gained popularity for the indication of chlorophyll content in leaves and its potential relationship to N nutrition of plants (Filella and Penuelas, 1994; Horler et al., 1983; Lichtenthaler et al., 1996; Penuelas and Filella, 1998). Horler et al. (1983) determined total chlorophyll and the red-edge wavelength to be highly correlated for several plant species [e.g., winter wheat (Triticum aestivum L.), r = 0.92; maize (Zea mays L.), r = 0.84]. Lichtenthaler et al. (1996) also found the red-edge position of the reflectance spectrum to be highly correlated (r = 0.97) with chlorophyll content of tobacco (Nicotianna tabacum L.) leaves. Filella and Penuelas (1994) concluded that pepper (Capsicum annuum L.) plants fertilized with high N rates, whose chlorophyll content was progressively increasing, showed a shift in the _re to longer wavelengths throughout the growth cycle. Conversely, plants under N stress, whose chlorophyll content was decreasing, showed a red-edge shift toward shorter wavelengths. More recently, Tarpley et al. (2000) reported on the use of single wavelength ratios to predict leaf N in cotton. Wavelength ratios that included reflectance within the red-edge region (700 or 716 nm) and NIR regions (755–920 nm and 1000 nm) provided good precision and accuracy. Slightly less accurate were ratios that included reflectance at 555 or 590 nm in combination with various wavelengths in the NIR region (755–1000 nm). Buscaglia and Varco (2002) noted improved leaf N prediction at flowering using a wavelength of 728 nm. Therefore, through determination of the red-edge shift from leaf reflectance, chlorophyll content may be monitored, thus providing insight to leaf N status.

The objective of this study was to evaluate the effects of varying N and K availability on cotton chlorophyll concentration and detection of leaf N utilizing reflectance properties.

	MATERIALS AND METHODS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Site Description and Cultural Practices
A 5 x 4 factorial study of N and K rates was established in 1998 at the Mississippi State Plant Science Research Center, Mississippi State, MS. Data for this study were collected in 1999 and 2000. The experimental site consisted of a Marietta fine sandy loam (fine-loamy, siliceous, thermic Fluvaquentic Eutrochrept) soil. Initial soil test levels as determined by the Mississippi Soil Test Method (Rasberry and Lancaster, 1977) of the experimental area averaged 77 mg P kg^–1 (High plus), 96 mg K kg^–1 (Low), pH of 8.27 (1:2 soil/deionized H₂O), and a cation exchange capacity of 31 cmol_c kg^–1. Mississippi Extension Service fertilizer recommendations to optimize cotton production at these soil test levels are no P, 84 kg K ha^–1, and 135 to 157 kg N ha^–1. Plots consisted of six rows 9.14 m in length, with a row spacing of 0.97 m. Treatments consisted of a factorial combination of fertilizer N (ammonium nitrate) rates of 0, 45, 90, 135, and 180 kg ha^–1 with fertilizer K (muriate of potash) rates of 0, 56, 112, and 168 kg ha^–1 arranged in a randomized complete block design with four replications. For this study, the combination of all N rates with 0 and 112 kg K ha^–1 was used. Potassium treatments were broadcast before planting (13 Apr. 1999 and 7 Mar. 2000) and incorporated with tillage. Nitrogen fertilizer rates were applied (broadcast) 50% at planting and the remaining 50% applied as a sidedress application 35 d after planting (DAP) in 1999 and 39 DAP in 2000.

Data Collection and Analysis
Leaf Nitrogen and Potassium Concentration and Plant Height
Twelve most recently matured leaves (fully expanded leaves capable of maximum photosynthetic potential on main stem at third to fourth node from terminal) within each plot were randomly sampled at 68 DAP (first bloom, >50% of plants have at least one flower) and 86 DAP (peak bloom) in 1999 and 65 DAP (first bloom) and 79 DAP (peak bloom) in 2000 for leaf tissue analysis. Samples were rinsed with deionized water, oven-dried at 65°C, and ground in a Wiley Mill (1 mm). Tissue samples for leaf K were prepared for elemental analysis using a modified dry ash procedure (Plank, 1992). Leaf K concentrations were determined using atomic absorption spectrometry. Total leaf N was determined on 4 to 6 mg of oven-dried samples using a Carlo Erba N/C 1500 dry combustion analyzer (Carlo Erba, Milan, Italy). Plant height was measured at both physiological stages of growth in 1999 and 2000. Plants within a 1-m length of Rows 3 and 4 were measured from the soil surface to the terminal of each plant.

Chlorophyll
Leaf disks (six disks per plot) were randomly sampled from recently matured leaves at 71 and 88 DAP in 1999 and 65 and 79 DAP in 2000 using a 13.5-mm-diam. cork borer. Chlorophyll was extracted from leaf disks using Hiscox and Israelstam's (1979) procedure with modifications as follows: Leaf disks were individually collected in plastic vials containing 10 mL of dimethyl sulfoxide (DMSO) (Sigma Chem. Co., St. Louis, MO). Vials were placed on ice to freeze DMSO and preserve samples until analysis. All vials were placed in an incubator (65°C) until all visible green pigmentation was removed (approximately 45 min).

One milliliter of each chlorophyll extract was transferred to a semimicro (methacrylate, 1.5 mL) cuvet (Fisher Sci., Pittsburgh, PA) for determination of chlorophyll absorbency. Absorption measurements were determined using a Hewlett Packard 8452A Diode Array Spectrophotometer (Hewlett Packard, Palo Alto, CA), scanning from 400 to 800 nm in 2-nm increments. Absorption measurements were used to quantify chlorophyll a, chlorophyll b, and total chlorophyll concentrations based on equations reported by Barnes et al. (1992).

Leaf Reflectance
Three recently matured leaves within each plot were randomly selected 71 and 87 DAP in 1999 and 65 and 79 DAP in 2000 for the determination of spectral reflectance for each treatment. Leaves were sampled separate from leaves used for chlorophyll analysis in 1999 but were the same in 2000. Leaf reflectance was measured via a LI-COR 1800 spectroradiometer equipped with an integrating sphere (spectral range 300–1100 nm, LI-COR, Lincoln, NE), scanning from 400 to 800 nm with a resolution of 2 nm. The integrating sphere was mounted on a tripod and, due to limited portability, was maintained stationary while leaves were detached plot by plot and measured immediately. A quartz fiber-optic probe connected the spectrometer to the integrating sphere, and a 10-W halogen lamp served as the radiation source. Each leaf was mounted on the integrating sphere such that primary venation was avoided, and a reflectance scan was taken of the adaxial surface. A standard reflectance scan of pressed BaSO₄ was also taken per leaf. Reflectance was calculated using the ratio of reflected radiation from the leaf surface divided by reflected radiation from the standard. Measurements were taken on cloudless to partly cloudy days beginning at 1000 h and completion by 1400 h.

Statistical Analysis
An analysis of variance (ANOVA) was performed to test N and K effects on leaf N and K and chlorophyll content. Mean separation was determined using LSD at a P = 0.05. Regression analysis was performed to determine relationships between Leaf N and chlorophyll a content. Analyses were performed using SAS (SAS Inst., 1999).

Partial least squares regression was implemented to analyze leaf spectral reflectance data sets relative to leaf N concentration. The analysis was performed using the PLS procedure in SAS (SAS Inst., 1999). A three-fold cross-validation procedure was implemented to select the number of factors to include in the model for a given reflectance data set (Garthwaite, 1994). The data were randomly broken into three groups where one group was omitted and the other two groups (calibration data set) were used to construct a predictive equation for Y from the omitted group (validation data set). This equation was used to predict Y-values for the omitted group and compared with this group's actual values. This procedure was repeated until each group was omitted once, and then the standard error of prediction (SEP) was calculated across all groups. Partial least-squares regression coefficients were calculated for models beginning with one factor until a maximum of 12 factors were added to the model. The number of factors used in the final PLS regression model was determined by the model that obtained a minimum global SEP (see Fig. 1 for graphical illustration).

View larger version (20K): [in this window] [in a new window]   Fig. 1. Graphical illustration of the standard error of prediction (SEP) vs. the number of factors used in the partial least-squares regression model.

	RESULTS AND DISCUSSION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
No N x K interaction was detected for leaf N and K, chlorophyll a, or total chlorophyll concentrations at either physiological stage of growth in 1999 and 2000. Therefore, N and K means of the variables presented in the discussion are averaged across K and N levels, respectively.

Leaf Tissue Nitrogen and Potassium and Plant Height
Nitrogen fertilizer rates induced a broad variation in leaf N concentration at first bloom and peak bloom in 1999 and 2000 regardless of the amount of K applied (Table 1). Leaf N concentration at peak bloom was lower than what was observed at first bloom in both years due to N translocation from leaves to developing bolls. Leaf N concentrations for all N treatments in 1999 were adequate to above adequate at first bloom compared with reported sufficiency levels (Sabbe and MacKenzie, 1973; Wood et al., 1992). However, leaf N concentrations for treatments 0 and 45 kg N ha^–1 were likely deficient at first bloom in 2000. Peak-bloom leaf N concentrations were adequate in 1999 and 2000 (with the exception of 0 kg N ha^–1) compared with reported sufficiency levels (Sabbe and MacKenzie, 1973).

Chlorophyll
Regression equations relating chlorophyll a concentration and leaf N concentration at first bloom and peak bloom for both years are shown in Table 3. As expected, chlorophyll a and total chlorophyll concentrations increased as leaf N concentration increased regardless of K nutrition at first bloom in 1999 and 2000 (total chlorophyll data not shown). Chlorophyll a and total chlorophyll concentrations also increased as leaf N concentration increased for both K treatments at peak bloom.

View larger version (23K): [in this window] [in a new window]   Fig. 2. Mean spectral reflectance of recently matured cotton leaves for each N rate at first bloom in 2000 with 112 kg K ha–1.

View larger version (22K): [in this window] [in a new window]   Fig. 3. Predicted vs. actual leaf N concentrations at first bloom in 2000 obtained using partial least-squares regression to relate spectral reflectance data in the visible/near-infrared wavelength range to the reference leaf N values. The graphed line represents a 1:1 relationship. SEP, standard error of prediction.

View larger version (22K): [in this window] [in a new window]   Fig. 4. Predicted vs. actual leaf N concentrations at peak bloom in 2000 obtained using partial least-squares regression to spectral reflectance data in the visible/near-infrared wavelength range to the reference leaf N values. The graphed line represents a 1:1 relationship. SEP, standard error of prediction.

Red-Edge Shift
The sharp rise in reflectance at the short-wavelength edge (680–740 nm) of the NIR spectral range is known as the _re. The _re shift corresponds to the movement of the _re wavelength, defined as a dR/d wavelength maximum or where d²R/d² = 0, between N treatments. Figure 5 illustrates the relationship between the reflectance spectra and the _re shift of healthy (135 kg N ha^–1) and deficient (0 kg N ha^–1) cotton leaves with an adequate supply of K at first bloom and peak bloom in 2000. Deficient leaves had greater reflectance in the visible wavelengths and showed a shift of the _re toward shorter wavelengths compared with healthy leaves. Results in Table 6 indicate a significant shift in the _re relative to N availability, as long as K was not limiting (with the exception of peak bloom in 2000). Thus, with an increase in N supply and adequate K nutrition, _re shifted to a longer wavelength. Although a significant shift was indicated between deficient and healthy cotton leaves with an adequate supply of K, a spectral resolution of less than 2 nm may be needed to make _re shift a more reliable technique in the detection of nutrient stress anomalies.

View larger version (27K): [in this window] [in a new window]   Fig. 5. Reflectance spectra and red-edge (re) shift for healthy and N-deficient leaves with an adequate supply of K at (left) first bloom and (right) peak bloom in 2000.

	CONCLUSIONS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

A distinct nutrient deficiency, given that no other nutrients are deficient, may be detected by leaf reflectance in the visible to NIR range. The PLS regression model yielded a greater predictive capability of leaf N concentration when K was not deficient throughout the growing season. Data also presented here show that the _re positioning showed a definite relationship to N status and may be a useful indicator of leaf N relative to chlorophyll concentrations. However, the ability to use leaf spectral properties and _re as an indicator of leaf N may depend on the availability of other nutrients, such as K.

	ACKNOWLEDGMENTS

 
We greatly acknowledge the additional support from the Mississippi Space Commerce Initiative (MSCI) Fellowship program during this study. We would also like to express our appreciation to John M. Thompson and H.J. Buscaglia for their research assistance and to all the student workers for their field assistance.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Contribution of the Mississippi Agric. and Forestry Exp. Stn., Journal Paper no. J10383.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Barnes, J.D., L. Balaguer, E. Mnarique, S. Elvira, and A.W. Davison. 1992. A reappraisal for the use of DMSO for the extraction and determination of chlorophylls a and b in lichens and higher plants. Environ. Exp. Bot. 32:85–100.
• Blackmer, T.M., J.S. Schepers, and G.E. Varvel. 1994. Light reflectance compared with other nitrogen stress measurements in corn leaves. Agron. J. 86:934–938.[Abstract/Free Full Text]
• Buscaglia, H.J., and J.J. Varco. 2002. Early detection of cotton leaf nitrogen status using leaf reflectance. J. Plant Nutr. 25:2067–2080.
• Carter, G.A. 1993. Responses of leaf spectral reflectance to plant stresses. Am. J. Bot. 80:239–243.[ISI]
• Chappelle, E.W., M.S. Kim, and J.E. McMurtrey, III. 1992. Ratio analysis of reflectance spectra (RARS): An algorithm for the remote estimation of the concentrations of chlorophyll a, chlorophyll b, and carotenoids in soybean leaves. Remote Sens. Environ. 39:239–247.
• Curran, P.J., J.L. Dungan, B.A. Macler, and S.E. Plummer. 1991. The effect of a red leaf pigment on the relationship between red edge and chlorophyll concentration. Remote Sens. Environ. 35:69–76.
• Everitt, J.H., A.J. Richardson, and H.W. Gausman. 1985. Leaf reflectance–nitrogen– chlorophyll relations in buffelgrass. Photogramm. Eng. Remote Sens. 51:463–466.
• Filella, I., and J. Penuelas. 1994. The red edge position and shape as indicators of plant chlorophyll content, biomass, and hydric status. Int. J. Remote Sens. 15:1459–1470.
• Garthwaite, P.H. 1994. An interpretation of partial least squares. J. Am. Stat. Assoc. 89:122–127.[ISI]
• Gausman, H.W. 1982. Visible light reflectance, transmittance, and absorption of differently pigmented cotton leaves. Remote Sens. Environ. 13:233–238.
• Helland, I.S. 1988. On the structure of partial least square regression. Commun. Stat.—Simula. 17:581–607.
• Hiscox, J.D., and G.F. Israelstam. 1979. A method for the extraction of chlorophyll from leaf tissue without maceration. Can. J. Bot. 57:1332–1334.
• Horler, D.N.H., M. Dockray, and J. Barber. 1983. The red edge of plant leaf reflectance. Int. J. Remote Sens. 4:273–288.
• Hsu, H.H. 1976. Potassium soil test calibration for cotton. M.S. thesis. Mississippi State Univ., Mississippi State.
• Jarrel, W.M., and R.B. Beverly. 1981. The dilution effect in plant nutrition studies. Adv. Agron. 34:197–224.
• Lamrani, Z., A. Belakbir, J.M. Ruiz, L. Ragala, I. Lopez-Cantarero, and L. Romero. 1996. Influence of nitrogen, phosphorus, and potassium on pigment concentration in cucumber leaves. Commun. Soil Sci. Plant Anal. 27:1001–1012.
• Lichtenthaler, H.K., A. Gitelson, and M. Lang. 1996. Non-destructive determination of chlorophyll content of leaves of a green and an aurea mutant of tobacco by reflectance measurements. J. Plant Physiol. 148:483–493.[ISI]
• Longstreth, D.J., and P.S. Nobel. 1980. Nutrient influences on leaf photosynthesis. Effects of nitrogen, phosphorus, and potassium for Gossypium hirsutum L. Plant Physiol. 65:541–543.[Abstract/Free Full Text]
• Lough, J.L. 2000. Effects of varying nitrogen and potassium nutrition on spectral reflectance and physiological properties of cotton. M.S. thesis. Mississippi State Univ., Mississippi State.
• Maples, R.L., W.R. Thompson, Jr., and J. Varvil. 1988. Potassium deficiency in cotton takes on a new look. Better Crops Plant Food 73:6–9.
• Masoni, A., L. Ercoli, and M. Mariotti. 1996. Spectral properties of leaves deficient in iron, sulfur, magnesium, and manganese. Agron. J. 88:937–943.[Abstract/Free Full Text]
• Microcal Software. 1994. Microcal Origin version 3.5. Microcal Software, Northampton, MA.
• Nichols, S.P. 1983. The effect of selected soil factors on the potassium concentration of cotton and soybeans. M.S. thesis. Mississippi State Univ., Mississippi State.
• Oosterhuis, D.M., and C.W. Bednarz. 1997. Physiological changes during the development of potassium deficiency in cotton. p. 347–351. In T. Ando et al. (ed.) Plant nutrition for sustainable food production and environment. Kluwer Academic Publ., Dordrecht, the Netherlands.
• Penuelas, J., and I. Filella. 1998. Visible and near-infrared reflectance techniques for diagnosing plant physiological status. Trends Plant Sci. 3:151–156.
• Plank, C.O. 1992. Plant analysis reference procedures for the southern region of the United States. S. Coop. Ser. Bull. 368. Georgia Agric. Exp. Stn., Athens.
• Rasberry, F.P., and J.D. Lancaster. 1977. A comparative evaluation of the Mississippi soil test method for determining available manganese, magnesium, and calcium. Commun. Soil Sci. Plant Anal. 8:327–339.
• Sabbe, W.E., and A.J. MacKenzie. 1973. Plant analysis as an aid to cotton fertilization. p. 299–313. In L.M. Walsh and J.D. Beaton (ed.) Soil testing and plant analysis. SSSA Book Ser. 3. SSSA, Madison, WI.
• SAS Institute. 1999. SAS/STAT user's guide, Version 8. Vol. 3. SAS Inst., Cary, NC.
• Stromberg, L.K. 1960. Need for potassium fertilizer on cotton determined by leaf and soil analysis. Calif. Agric. 14:4–5.
• Sudduth, K.A., and J.W. Hummel. 1991. Evaluation of reflectance methods for soil organic matter sensing. Trans. ASAE 34:1900–1909.
• Tarpley, L., K.R. Reddy, and G.F. Sassenrath-Cole. 2000. Reflectance indices with precision and accuracy in predicting cotton leaf nitrogen concentration. Crop Sci. 40:1814–1819.[Abstract/Free Full Text]
• Thomas, J.R., and H.W. Gausman. 1977. Leaf reflectance vs. leaf chlorophyll and carotenoid concentrations for eight crops. Agron. J. 69:799–802.[Abstract/Free Full Text]
• Tobias, R.D. 2000. An introduction to partial least squares regression. SAS Inst., Cary, NC.
• Wood, C.W., P.W. Tracy, D.W. Reeves, and K.L. Edmisten. 1992. Determination of cotton nitrogen status with a hand-held chlorophyll meter. J. Plant Nutr. 15:1435–1448.[ISI]

This article has been cited by other articles:

				M. T. Gomez-Casero, F. Lopez-Granados, J. M. Pena-Barragan, M. Jurado-Exposito, L. Garcia-Torres, and R. Fernandez-Escobar Assessing Nitrogen and Potassium Deficiencies in Olive Orchards through Discriminant Analysis of Hyperspectral Data J. Amer. Soc. Hort. Sci., September 1, 2007; 132(5): 611 - 618. [Abstract] [Full Text] [PDF]

				J. K. Kruse, N. E. Christians, and M. H. Chaplin Remote Sensing of Nitrogen Stress in Creeping Bentgrass Agron. J., October 31, 2006; 98(6): 1640 - 1645. [Abstract] [Full Text] [PDF]

				K. F. Bronson, J. D. Booker, J. W. Keeling, R. K. Boman, T. A. Wheeler, R. J. Lascano, and R. L. Nichols Cotton Canopy Reflectance at Landscape Scale as Affected by Nitrogen Fertilization Agron. J., April 27, 2005; 97(3): 654 - 660. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Fridgen, J. L. Articles by Varco, J. J. Search for Related Content PubMed Articles by Fridgen, J. L. Articles by Varco, J. J. Agricola Articles by Fridgen, J. L. Articles by Varco, J. J. Related Collections Cotton Nutrient Management Plant Analysis Plant Nutrition Nitrogen