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 Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Bronson, K. F. Articles by Nichols, R. L. Search for Related Content PubMed Articles by Bronson, K. F. Articles by Nichols, R. L. Agricola Articles by Bronson, K. F. Articles by Nichols, R. L. Related Collections Nitrogen Cotton Field-Scale Studies Remote Sensing Nutrient Management Plant Nutrition

Remote Sensing

Cotton Canopy Reflectance at Landscape Scale as Affected by Nitrogen Fertilization

^a Texas A&M Univ. Texas Agric. Exp. Stn., RR 3, Box 219, Lubbock, TX 79403
^b Texas Coop. Ext., RR 3, Box 213AA, Lubbock, TX 79403
^c Cotton Inc. World Headquarters, 6399 Weston Parkway, Cary, NC 27513

^* Corresponding author (k-bronson{at}tamu.edu)

Received for publication April 1, 2004.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION SUMMARY REFERENCES

 
Multispectral reflectance of crop canopies has potential as an in-season indicator of N status in cotton (Gossypium hirsutum L.). The objectives of this study were to correlate leaf N with reflectance at 16 wavebands from 450 to 1700 nm and to assess the effect of N fertilization on vegetative ratio indices using two bands of reflectance. We also compared regressions of leaf N on ratio indices with partial least squares (PLS) regression using reflectance at 16 wavebands. Reflectance was measured at 50 cm above the canopy at 135 points in a 14-ha field of irrigated cotton at early squaring in the Texas High Plains in 2002 and at an 80-cm height in 2003 and 2004. Leaf N had weak, negative correlation with green reflectance in all 3 yr. Normalized difference vegetative indices (NDVIs) using red (670 nm) or green (550 nm) reflectance were significantly greater in N-fertilized plots than zero-N plots in 2 of 3 yr. However, the NDVIs related poorly or not at all with leaf N, biomass, and lint yield. Leaf N was estimated by PLS regression with three factors having R² of 0.64 in 2002 and 2004 when an N fertilizer response was observed. In 2003, there was no added N effect, and the R² for PLS regression of leaf N was 0.41. The poor correlation between NDVIs and leaf N was not expected, and these results suggest that use of NDVIs to determine need of in-season N may be most successful using well-fertilized areas and the sufficiency index approach.

Abbreviations: DGPS, differential global positioning system • NDVI, normalized difference vegetative index • NDVI-G-P, normalized difference vegetative index–green-passive • NDVI-R-A, normalized difference vegetative index–red-active • NDVI-R-P, normalized difference vegetative index–red-passive • NIR, near infrared • PLS, partial least squares

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION SUMMARY REFERENCES

 
SPECTRAL REFLECTANCE of crop canopies has been reported to be a good estimator of crop biomass or canopy cover. Most of the early work in this area was done with satellite images (Lennington et al., 1984; Wiegand et al., 1990; Quarmby et al., 1992). Typically, vegetative ratio indices of reflectance at red and near infrared (NIR) (or of green and NIR) wavebands are calculated. Tucker (1979) proposed the NDVI as (R_NIR – R_red)/(R_NIR + R_red), where R_NIR and R_red are reflectance in the NIR and red regions, respectively. Recently, proximal sensing to estimate crop N status has received strong interest. Bausch and Duke (1996) measured reflectance of corn (Zea mays L.) under a center pivot at a 10-m height and found that R_NIR/R_green related to leaf N. Researchers in Oklahoma have extensively tested a ground-based spectroradiometer with upward- and downward-facing NIR and red sensors directly over the row in wheat (Triticum aestivum L.). They could relate N uptake, biomass (Stone et al., 1996; Solie et al., 1996), and yield (Raun et al., 2001) with NDVI. Osborne et al. (2002) estimated plant N and P concentration, biomass, and grain yield of corn with hyperspectral reflectance at 3 m above the canopy. Ma et al. (2001) estimated soybean [Glycine max (L.) Merr.] yield with NDVI calculations of in-season measurements of spectral reflectance 2 m above the ground. Xue et al. (2004) used multispectral reflectance at 2 m above the canopy to monitor leaf N in rice (Oryza sativa L.).

Assessing N concentration of cotton with spectral reflectance has in many cases been confined to analysis of individual leaves. Lough and Varco (2001) used spectral reflectance to assess N and K status of cotton leaves harvested from the field. They reported that reflectance at 550 nm and a shift in the edge of red reflectance separated N and K fertilizer treatments, respectively. Saranga et al. (1998) used NIR analysis to assess N status of cotton leaves by punching 2.5-cm-diam. disks out of the leaves sampled in the field. Tarpley et al. (2000) calculated indices from spectral reflectance (350–1050 nm) of individual cotton leaves to estimate leaf N.

Field reflectance studies that assess in-season biomass and plant nutrient concentration in cotton have been relatively few. Maas (1998) used a portable spectroradiometer to estimate cotton ground cover. Plant et al. (2001) took false infrared images from aircraft at 850- to 1500-m altitude above cotton plots that received various N fertilizer rates. Calculations of NDVI from their reflectance data related to N rate and lint yield. Li et al. (2001) reported that NDVI calculated from measurements of spectral reflectance with a hand-held spectroradiometer at 2 m above the canopy correlated well with cotton N uptake, biomass, and lint yield.

In-season remote or proximal sensing of spectral reflectance has the potential to supplement preplant soil-testing data, by providing information on the need for fertilization. In center-pivot- and subsurface-drip-irrigated areas such as the High Plains, in-season fertigation with the irrigation water is commonly practiced; therefore, timely information on the need for fertilizer could be utilized (Chua et al., 2003). Bronson et al. (2003a) reported that simple vegetative ratio indices of NIR to red or green reflectance estimated in-season cotton N status and predicted the need for N fertilization. In that study, 16-band spectral reflectance was measured, but only reflectance at two bands (e.g., 550 and 820 nm) was utilized. With the increasing availability of multispectral and hyperspectral radiometers, the question arises whether better estimates of in-season plant N status can be achieved with reflectance from a large number of wavebands than with reflectance from just two wavebands (e.g., an NIR and a red or green waveband).

The objectives of this study were to: (i) determine the effect of N fertilizer on simple vegetative ratio indices using NIR and red or green reflectance of the canopy of irrigated cotton at landscape scale, (ii) correlate in-season leaf N with 16 bands of multispectral reflectance, and (iii) compare leaf N estimates from regressions on vegetative ratio indices with PLS regression of multispectral reflectance of 16 wavebands.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION SUMMARY REFERENCES

 
This study was performed on an N fertilizer management and irrigation level study on a 14-ha area under a 48-ha center-pivot cotton field at Lamesa, TX, from 2002 to 2004 (Bronson et al., 2004). Three levels of low-energy precision application of irrigation were applied from 63 to 93% estimated evapotranspiration through plastic socks (Lyle and Bordovsky, 1981). ‘Paymaster 2326 Round-Up Ready’ cotton was planted into glyphosate-[isoprophylamine salt of N-(phosphomonomethyl)glycine]-terminated rye (Secale cereale L.) in 1-m rows on 9 May 2002 and on 8 May 2003, and on 4 May 2004, the variety 'FiberMax 989 Round-Up Ready' was planted. The soil series at this site is an Amarillo sandy loam (fine-loamy, mixed, superactive, thermic, Aridic Paleustalf). Plot lengths ranged from 500 to >1000 m, due to circular rows. Nitrogen fertilizer treatments applied each year were zero N, blanket-rate N, and variable-rate N. The experimental design used was a split plot in randomized complete blocks. Row spacing was 1 m. Irrigation level main plots and N treatment subplots were 24 and 8 rows wide, respectively. Average rates of N fertilizer for the variable-rate N treatments were 63, 98, and 95 kg N ha^–1 for 2002, 2003, and 2004 respectively, but varied widely within the field. Blanket N treatment rates were 59, 101, and 100 kg N ha^–1 for 2002, 2003, and 2004, respectively. In all cases, the N fertilizer applied was based on a constant 1100 kg lint ha^–1 yield goal. The recommendation for this yield level is 134 kg N ha^–1 minus the kilograms of extractable NO₃–N per hectare in a 0- to 60-cm layer soil sample (Zhang et al., 1998). Two 0- to 60-cm soil samples were taken each spring at 135 differential global positioning system (DGPS) referenced points in the 14-ha study area and analyzed for KCl-extractable NO₃–N. Inverse distance to a square interpolation of the 0- to 60-cm soil NO₃–N data was used to guide variable-rate N fertilizer applications. A variable-rate liquid fertilizer system with spoke-wheel applicators (Bronson et al., 2003b) was used to apply urea ammonium nitrate (320 g N kg^–1) in two equal splits at about 20 d after planting and at early squaring.

At early squaring of cotton (24 July 2002, 15 July 2003, and 8 July 2004) in all years, multispectral reflectance readings were taken at two locations within 3 m of 135 DGPS-referenced sampling points. The hand-held passive spectroradiometer (Model MSR16R, CropScan, Inc., Rochester, MN) consisted of upward- and downward-facing radiation transducers that have 16 interference filters (450, 470, 500, 530, 550, 570, 600, 630, 650, 670, 700, 780, 820, 870, 1600, and 1700 nm). The bandwidth of the 16 filters ranged from 6.5 to 17.0 nm for 450- to 1700-nm wavelength centers. The radiometer was programmed to take 100 readings from each of the 16 upward and downward sensors, which took just 4 s. The sensor of the radiometer was positioned 50 cm above the row of what we judged to approximately be the average plant height in the field. This relatively low height above the canopy was used because our primary interest was N status of the leaf and not biomass or ground cover. The field of view of the downward sensors is 28°. At 50 cm, reflectance from soil was minimal compared with plant reflectance. Reflectance readings were taken within 2 h of solar noon. However, we did not take readings within 20 min of solar noon because the shadow of the radiometer was on the canopy at that time. The radiometer was calibrated before each use by using an opal glass to provide the same irradiance alternatively to the upward sensors and to the downward sensors, both at a 45° angle to the sun. Percentage reflectance was calculated as reflected irradiance/incoming irradiance for each waveband. The program that calculated percentage reflectance applied sun angle cosine corrections to the millivolt readings of each sensor, based on data, time, latitude, and longitude as well as sensor temperature corrections (Cropscan Inc., 1998).

In 2003 and 2004, reflectance was also measured at 80 cm above the canopy of early squaring cotton with the GreenSeeker hand-held active spectroradiometer (NTech Industries Inc., Ukiah, CA). In 2003 and 2004, the height of the CropScan sensor readings was raised to 80 cm to compare reflectance with the GreenSeeker. The GreenSeeker cannot properly sense reflected light that is generated from its light-emitting diodes if the height above the target is less than 80 cm. The GreenSeeker measures reflectance at two wavelengths, 670 and 780 nm, and has a 60-cm-wide field of view. The operator held the GreenSeeker at 80 cm above the canopy and walked a 4-m distance for two rows of cotton at each of the 135 DGPS points. Approximately 150 reflectance readings were taken by the GreenSeeker in each 4-m pass.

Biomass was sampled at early squaring on 2 m of row (1 m from two rows) near the DGPS-referenced grid-sampling points. Leaves, squares, and stems were dried at 65°C and weighed for dry matter. Leaves were ground to 1 mm and analyzed for N concentration on a dry combustion N analyzer.

A John Deere 484 four-row stripper harvester, equipped with a Micro-Trak optical yield-monitoring system (Eagle Lake, MN), was used to harvest seedcotton in October of each year. Seedcotton weights were adjusted to boll buggy weights, and a single percentage turnout of lint concentration from the local gin was used to calculate kilograms of lint per hectare. The "Create Buffers" routine in ArcView GIS 3.2 (ESRI, 1992) was used to create oval-shaped zones (24 by 8 m) centered on each of the DGPS points. Next, the "Summarize Zones" routine averaged the yield values in each oval zone at each DGPS point and added this column of data to the existing table of 135 records of spectral reflectance, biomass, and leaf N data. The number of yield data values in each 24-m zone from which averages were calculated ranged from 4 to 12.

The following ratio vegetative indices were calculated from the percentage reflectance (R) data of the passive CropsScan sensor: NDVI–green-passive, NDVI-G-P = (R_NIR – R_green)/(R_NIR + R_green), and NDVI–red-passive, NDVI-R-P = (R_NIR – R_red)/(R_NIR + R_red). We calculated the nine possible combinations of each of these two indices using reflectance from the Cropscan sensor's three NIR bands (780, 820, and 870 nm), the three green bands (530, 550, and 570 nm), and the three red bands (630, 650, and 670 nm). Only one NDVI could be calculated with the active GreenSeeker sensor: NDVI–red-active, NDVI-R-A = (R₇₈₀ – R₆₇₀)/(R₇₈₀ + R₆₇₀).

Simple statistics, such as minimum, maximum, mean, standard deviation, and 25 and 75% quantiles, were calculated for leaf N, biomass, and lint yield for the 3-yr data with PROC UNIVARIATE (SAS Inst., 1999a). Analysis of variance of the NDVI measures, leaf N, and biomass was calculated using PROC MIXED (SAS Inst., 1999a). Replicate and interactions of N and water with replicate effects were considered random. Nitrogen, water, and N x water interaction effects were considered fixed. Simple correlation was performed among leaf N, biomass, lint yield, and reflectance at the 16 wavebands using PROC CORR (SAS Inst., 1999a). Simple ordinary least squares regression was used (PROC REG; SAS Inst., 1999a) to estimate leaf N concentration, biomass, and lint yield as a function of NDVI-G-P, NDVI-R-P, and NDVI-R-A. Partial least squares regression with dependent variables leaf N concentration, biomass, and lint yield was performed on three factors extracted from reflectance of all 16 wavebands using PROC PLS and PROC REG (SAS Inst., 1999a). We decided to use three factors in the PLS and PROC REG analysis because there was negligible improvement in explaining the variation in the dependent variables with more than three factors.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION SUMMARY REFERENCES

 
The expected yield of 1100 kg lint ha^–1 was met in the highest-yielding 4 and 63% of DGPS points in 2002 and in 2004, respectively (Table 1). Damaging hail and wind in May of 2003 followed by cool June temperatures resulted in very slow June growth, low early squaring biomass, low lint yields, and lack of response to N fertilizer that year (Bronson et al., 2004). Above-average rainfall in 2004 contributed to high early squaring biomass and high lint yields but a limited lint yield response to irrigation level. Irrigation level did not affect leaf N, NDVI, or biomass measured at early squaring in any year. This was not unexpected as irrigation commenced only shortly before the squaring stage. For this reason, we will not discuss irrigation level furthermore in this study.

It would be valuable to know if the N leaf concentrations measured at early squaring were below established critical levels (i.e., concentration that separates sufficiency from deficiency). Minimum leaf N concentrations at early squaring in the 3 yr ranged from 24.0 to 34.8 g N kg^–1 (Table 1). Plank (1979) and Jones et al. (1991) reported that the low end of the sufficiency range of leaf N at early squaring is 35.0 g N kg^–1. By this critical level, there was only one N-deficient sampling point in 2002 and 2003, and 30% of the points were N deficient in 2004. However, in 2002, the average leaf N concentration in zero-N plots in 2002 was a relatively high 42.6 g N kg^–1, and yet a lint yield response to N was still observed. Therefore, fixed critical values of cotton leaf N concentration may be of limited practical use.

Many workers who monitor in-season N status of crops with either chlorophyll meters or spectroradiometers have suggested the use of well-fertilized reference plots and the calculation of a sufficiency index (Varvel et al., 1997; Hussain et al., 2000; Chua et al., 2003). Sufficiency indices calculated as the NDVIs from zero-N plots divided by the NDVIs of the N-fertilized plots in Table 2 range from 0.945 to 1.00. Calculating sufficiency indices in this manner assumes that the N-fertilized plots were not N limiting. The lowest sufficiency index calculated was in the case of the NDVI-R-P in 2004. Chua et al. (2003) suggested that a critical sufficiency index for irrigated cotton at early squaring should be 0.97 or 0.98. In the present study, a sufficiency index of 0.988 was calculated for the NDVI-R-A in 2004 (Table 2). Since the NDVI-R-A was statistically greater with the N-fertilized plots than with the zero-N plots in 2004, statistical significance may be a more reasonable basis for determining need of in-season N in irrigated cotton compared with a fixed critical sufficiency index.

Leaf N was negatively, but weakly, correlated with green reflectance in all 3 yr (Table 3). Negative correlation of leaf N with leaf reflectance at 550 nm has been widely reported (Blackmer and Schepers, 1994; Blackmer et al., 1996). This means N-deficient leaves (i.e., light green leaves) reflect more green (550 nm) light than N-sufficient leaves (i.e., dark green leaves). In 2004, a significant negative correlation (r = –0.54) between leaf N and red (630 nm) reflectance was observed.

Lint yield was negatively correlated with visible reflectance in 2002 and positively correlated with NIR reflectance the same year (Table 3). In 2003 and 2004, lint yield only correlated with red reflectance (630–670 nm). Correlations between reflectance at 1600 and 1700 nm and biomass and lint yield were inconsistent (Table 3).

Among the vegetative ratio indices, similar regressions with leaf N, biomass, or lint yield were calculated using the NIR reflectance at 780, 820, or 870 nm (data not shown). We will only present NDVI data using NIR reflectance at 870 nm. Among the green wavebands, reflectance at 550 nm gave slightly greater R² values in NDVI-G-P calculations; therefore only this NDVI-G-P will be presented. All red bands between 630 and 670 nm resulted in similar regressions with leaf N, biomass, and lint yield. Only NDVI-R-P calculated with red reflectance at 670 nm will be discussed, as this will also allow us to compare between the CropScan and GreenSeeker units.

Coefficients of determination for regressions of NDVI-G-P or NDVI-R (-P or -A) on leaf N were weak and frequently zero (Table 4). Studies with other crops such as wheat, corn, or rice have reported significant correlation between red or green ratio indices and leaf N (Sembiring et al., 1998; Bausch and Duke, 1996; Takebe et al., 1990). Correlations of NDVI-G-P and NDVI-R (-P and -A) with biomass were not greater than those with leaf N. This is in contrast to other studies and may be related to our relatively low sensor height. Li et al. (2001) reported at this same site that NDVI-R-P calculated from reflectance at 2 m above the cotton canopy was positively correlated with biomass, N accumulation, and lint yield but not with leaf N. Sembiring et al. (1998) reported that NDVIs more consistently related to wheat biomass than leaf N.

Estimates from PLS regression of leaf N on reflectance from each of the 16 wavebands are shown in Fig. 1 . In 2002 and 2004, PLS regressions for leaf N had positive estimates for blue (450–500 nm) and NIR (780–870 nm) reflectance and negative estimates for green (530–570 nm) reflectance. These regression estimates in 2003 were much smaller and with less trend. This is probably associated with the lack of N fertilizer response in leaf N in 2003.

View larger version (24K): [in this window] [in a new window]   Fig. 1. Centered and scaled estimates from partial least squares regression of early squaring leaf N concentration (g N kg–1) on three factors derived from passive spectral reflectance (50 cm above canopy in 2002 and 80 cm in 2003 and 2004) at 16 wavebands (450 to 1700 nm), Lamesa, TX, 2002–2004.

Lint yield estimates were poor in all cases (Table 4). This may have been because the early squaring growth stage is in late June to early July while harvest is in October. Li et al. (2001) reported strong correlation between NDVI and lint yield (r = 0.80) on 20 August or about peak bloom. Biomass estimates are often achievable with ratio indices calculated from spectral reflectance (Bronson et al., 2003a). However, biomass does not always reflect final lint yield in cotton as fruit load and fruit damage by insects may not affect biomass (Oosterhuis, 1990).

A spectroradiometer would be used in a continuous data acquisition mode on a producer's field. More research is needed with continuous ground-based multispectral reflectance measurements in cotton. Leaf nutrient data collection, however, is always going to be limited to point data. The results of this study suggest that improved ratios for two-band reflectance (NIR and green or red) and algorithms or models for multispectral reflectance are needed to better estimate leaf N in cotton at the landscape scale. Alternatively, use of NDVIs to determine in-season need of N may be best done with well-fertilized parts of the field and the sufficiency index approach.

	SUMMARY

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION SUMMARY REFERENCES

 
Regressions between NDVIs using either green or red (passive or active) reflectance related poorly or not at all with leaf N, biomass, or lint yield. However, NDVIs in all cases reflected leaf N response or lack of response to added N fertilizer.

Partial least squares regression estimated leaf N reasonably well (R² = 0.64) in the 2 yr when N fertilizer response was observed. The PLS regression estimates for each waveband followed a similar pattern of negative estimates for green reflectance in these 2 yr. The lack of consistency of the standardized estimates of the PLS regression probably limits their usefulness in predicting in-season leaf N, as does the low R² of 0.41 in the year without an N fertilizer effect. The vegetative ratio indices, although simple to calculate, need to be improved on as well for in-season leaf N estimation in cotton. The better approach may be to use the current NDVI calculation to simply separate N-deficient field areas from well-fertilized areas with the sufficiency index.

	ACKNOWLEDGMENTS

 
The authors would like to thank Jimmy Mabry and Danny Carmichael for their capable field assistance. This study was funded by a Special Initiative of the Texas State Legislature on Precision Agriculture and by Cotton Inc./Texas State Support Committee.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION SUMMARY REFERENCES

 

• Bausch, W.C., and H.R. Duke. 1996. Remote sensing of plant nitrogen status in corn. Trans. ASAE 39(5):1869–1875.
• Blackmer, T.M., and J.S. Schepers. 1994. Techniques for monitoring crop nitrogen status in corn. Commun. Soil Sci. Plant Anal. 25:1791–1800.
• Blackmer, T.M., J.S. Schepers, G.E. Varvel, and E.A. Walter-Shea. 1996. Nitrogen deficiency detection capabilities of light reflection from irrigated corn canopies. Agron. J. 88:1–5.[Abstract/Free Full Text]
• Bronson, K.F., T.T. Chua, J.D. Booker, J.W. Keeling, and R.J. Lascano. 2003a. In-season nitrogen status sensing in irrigated cotton: II. Leaf nitrogen and biomass. Soil Sci. Soc. Am. J. 67:1439–1448.[Abstract/Free Full Text]
• Bronson, K.F., J.W. Keeling, J.D. Booker, J.P. Bordovsky, M.N. Parajulee, T.A. Wheeler, and R.K. Boman. 2004. Variable-rate nitrogen and water for irrigated cotton in the Southern High Plains. In Proc. 2004 Beltwide Cotton Conf., San Antonio, TX. 6–13 Jan. 2004 [CD-ROM]. Natl. Cotton Council of Am., Memphis, TN.
• Bronson, K.F., J.W. Keeling, J.D. Booker, T.T. Chua, T.A. Wheeler, R.K. Boman, and R.J. Lascano. 2003b. Influence of phosphorus fertilizer, landscape position and soil series on cotton lint yield. Agron. J. 95:949–957.[Abstract/Free Full Text]
• Chua, T.T., K.F. Bronson, J.D. Booker, J.W. Keeling, A.R. Mosier, J.P. Bordovsky, R.J. Lascano, C.J. Green, and E. Segarra. 2003. In-season nitrogen status sensing in irrigated cotton: I. Yield and nitrogen-15 recovery. Soil Sci. Soc. Am. J. 67:1428–1438.[Abstract/Free Full Text]
• Cropscan. 1998. Multispectral radiometer user's manual. Cropscan, Inc., Rochester, MN.
• Environmental Systems Research Institute. 1992. ArcView GIS 3.2. ESRI, Redlands, CA.
• Fridgen, J.L., and J.J. Varco. 2004. Dependency of cotton leaf nitrogen, chlorophyll, and reflectance on nitrogen and potassium availability. Agron. J. 96:63–69.[Abstract/Free Full Text]
• Hussain, F., K.F. Bronson, Yadvinder-Singh, Bijay-Singh, and S. Peng. 2000. Use of well-fertilized reference plots for chlorophyll meter-based N management in irrigated rice. Agron. J. 92:875–879.[Abstract/Free Full Text]
• Jones, J.B., Jr., B. Wolf, and H.A. Mills. 1991. Plant analysis handbook. Micro-Macro Publ., Inc., Athens, GA.
• Lennington, R.K., C.T. Sorensen, and R.P. Heydorn. 1984. A mixture model approach for estimating crop areas from Landsat data. Remote Sens. Environ. 14:197–206.
• Li, H., R.J. Lascano, E.M. Barnes, J. Booker, L.T. Wilson, K.F. Bronson, and E. Segarra. 2001. Multispectral reflectance of cotton related to plant growth, soil water, texture, and site elevation. Agron. J. 93:1327–1337.[Abstract/Free Full Text]
• Lough, J.L., and J.J. Varco. 2001. Effects of varying N and K nutrition on the spectral reflectance properties of cotton. In P.C. Robert et al. (ed.) Precision agriculture [CD-ROM]. Proc. Int. Conf., 5th, Minneapolis, MN. 16–19 July 2000. ASA, CSSA, and SSSA, Madison, WI.
• Lyle, W.M., and J.P. Bordovsky. 1981. Low energy precision application (LEPA) [low energy-precision application] irrigation system. Trans. ASAE 24:1241–1245.
• Ma, B.L., L.M. Dwyer, C. Costa, E.R. Cober, and M.J. Morrison. 2001. Early season prediction of soybean yield from canopy reflectance measurements. Agron. J. 93:1227–1234.[Abstract/Free Full Text]
• Maas, S. 1998. Estimating cotton canopy ground cover from remotely sensed scene reflectance. Agron. J. 90:384–388.[Abstract/Free Full Text]
• Oosterhuis, D.M. 1990. Growth and development of a cotton plant. 1990. p. 1–24. In W.N. Wiley and D.M. Osterhuis (ed.) Nitrogen nutrition in cotton: Practical issues. ASA, Madison, WI.
• Osborne, S.L., J.S. Schepers, D.D. Francis, and M.R. Schlemmer. 2002. Detection of phosphorus and nitrogen deficiencies in corn using spectral radiance measurements. Agron. J. 94:1215–1221.[Abstract/Free Full Text]
• Plank, C.O. 1979. Plant analysis handbook for Georgia. Bull. 735. Univ. of Georgia Coop. Ext., Athens.
• Plant, R.E., D.S. Munk, B.R. Roberts, R.N. Vargas, R.L. Travis, D.W. Rains, and R.B. Hutmacher. 2001. Applications of remote sensing to strategic questions in cotton management and research. J. Cotton Sci. 5:30–41.
• Quarmby, N.A., J.R.G. Townshend, J.J. Settel, K.H. White, M. Milnes, T.L. Hindle, and N. Silleos. 1992. Linear mixture modeling applied to AVHRR data for crop area estimation. Int. J. Remote Sens. 13:415–425.
• Raun, W.R., J.B. Solie, G.V. Johnson, M.L. Stone, E.V. Lukina, W.E. Thomason, and J.S. Schepers. 2001. In-season prediction of potential grain yield in winter wheat using canopy reflectance. Agron. J. 93:131–138.[Abstract/Free Full Text]
• Saranga, Y., A. Landa, Y. Shekel, A. Bosak, and U. Kafkafi. 1998. Near-infrared analysis of cotton leaves as a guide for nitrogen fertilization. Agron. J. 90:16–21.[Abstract/Free Full Text]
• SAS Institute. 1999a. The SAS system for Windows. Version 8.0. SAS Inst., Cary, NC.
• SAS Institute. 1999b. SAS/STAT user's guide. Version 8.0. SAS Inst., Cary, NC.
• Sembiring, H., W.R. Raun, G.V. Johnson, M.L. Stone, J.B. Solie, and S.B. Phillips. 1998. Detection of nitrogen and phosphorus nutrient status in winter wheat using spectral radiance. J. Plant Nutr. 21:1207–1233.
• Solie, J.B., W.R. Raun, R.W. Whitney, M.L. Stone, and J.D. Ringer. 1996. Optical sensor based field element size and sensing strategy for nitrogen application. Trans. ASAE 39:1983–1992.
• Stone, M.L., J.B. Solie, W.R. Raun, R.W. Whitney, S.L. Taylor, and J.D. Ringer. 1996. Use of spectral radiance for correcting in-season fertilizer nitrogen deficiencies in winter wheat. Trans. ASAE 39:1623–1631.
• Takebe, M., T. Yoneyama, K. Inada, and T. Murakami. 1990. Spectral reflectance ratio of rice canopy for estimating crop nitrogen status. Plant Soil 122:295–297.
• 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]
• Tucker, C.J. 1979. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ. 8:127–150.[CrossRef]
• Varvel, G.E., J.S. Schepers, and D.D. Francis. 1997. Ability for in-season correction of nitrogen deficiency in corn using chlorophyll meters. Soil Sci. Soc. Am. J. 61:1233–1239.[Abstract/Free Full Text]
• Wiegand, C.L., A.H. Gerbermann, K.P. Gallo, B.L. Blad, and D. Dusek. 1990. Multisite analysis of spectral-biophysical data for corn. Remote Sens. Environ. 33:1–16.
• Xue, L., W. Cao, W. Luo, T. Dai, and Y. Zhu. 2004. Estimating leaf nitrogen status in rice with canopy spectral reflectance. Agron. J. 96:135–142.[Abstract/Free Full Text]
• Zhang, H., B. Raun, J. Hattey, G. Johnson, and N. Basta. 1998. OSU soil test interpretations. Publ. no. F-2225. Oklahoma Coop. Ext. Serv., Oklahoma State Univ., Stillwater.

This article has been cited by other articles:

				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. P. Bordovsky, J. W. Keeling, T. A. Wheeler, R. K. Boman, M. N. Parajulee, E. Segarra, and R. L. Nichols Site-Specific Irrigation and Nitrogen Management for Cotton Production in the Southern High Plains Agron. J., January 5, 2006; 98(1): 212 - 219. [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 Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Bronson, K. F. Articles by Nichols, R. L. Search for Related Content PubMed Articles by Bronson, K. F. Articles by Nichols, R. L. Agricola Articles by Bronson, K. F. Articles by Nichols, R. L. Related Collections Nitrogen Cotton Field-Scale Studies Remote Sensing Nutrient Management Plant Nutrition