| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Texas A&M Univ., Rt. 3, Box 219, Lubbock, TX 79403
b Texas A&M Univ., USDA-ARS, 3810 4th Street, Lubbock, TX 79415
c USDA-ARS, U.S. Water Conserv. Lab., 4331 E. Broadway, Phoenix, AZ
d Texas A&M Univ., 1509 Aggie Drive, Beaumont, TX 77713
e Dep. of Soil Sci., North Carolina State Univ., Raleigh, NC 27695
* Corresponding author (r-lascano{at}tamu.edu)
Received for publication October 16, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ET, evapotranspiration • LAI, leaf area index • MIR, midinfrared • NDVI, normalized difference vegetative index • NIR, near infrared • PFB, plant fresh biomass • PWC, plant water content • SE, site elevation • SWC, soil water content
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Spectral leaf reflectance as a function of wavelength between 400 and 2500 nm was used to determine soybean [Glycine max (L.) Merr.] (Kollenkark et al., 1982) and cotton (Bowman, 1989) plant water status as well as corn leaf water stress (Bausch, 1993) and N deficiency (Blackmer et al., 1996). Soybean spectral reflectance at 760 to 840 nm was correlated with leaf area index (LAI), fresh and dry biomass, percent soil cover, and grain yield (Kollenkark et al., 1982). Spectral reflectance at 810 and 1665 nm was correlated with leaf relative water content and total water potential (Bowman, 1989). The reflected wavelengths most sensitive to detecting N deficiencies in a corn canopy were 550 and 710 nm based on their correlation with grain yield (Walburg et al., 1982; Blackmer et al., 1996). Both water and N stresses altered plant reflectance and lowered normalized difference vegetative index (NDVI) values (Plant et al., 2000). Cotton lint yield was significantly correlated with soil water and sand contents (Li et al., 2001a), red and near-infrared (NIR) reflectance, and NDVI (Wiegand et al., 1994; Plant et al., 2000; Li et al., 2001b), and temporal patterns of cotton reflectance were related to plant growth at different stages (Plant et al., 2000; Li et al., 2001b). These results suggest that it may be possible to use spatial and temporal patterns of plant reflectance and spectral index for in-season water and N management. However, the difficulty is that the spectral signal for a given crop is complex because it is simultaneously affected by many soil and plant variables.
Soil texture, moisture, plant cover, and landscape surface roughness could affect soil and plant reflectance in the visible and NIR wavelength regions (Kollenkark et al., 1982; Jackson, 1984, 1987; Begue, 1993; Asner, 1998). Plant reflectance, NIR/red reflectance ratio, and the NDVI were different with stress events, sun angles, cultivars, and irrigation and fertilization treatments (Daughtry et al., 1992; Wiegand et al., 1994; Begue, 1993; Blackmer et al., 1996; Stone et al., 1996). Stress events, such as drought, reduced spectral estimates of absorbed radiation and NDVI in corn and soybean canopies (Daughtry et al., 1992). Weather variability also affected canopy spectral characteristics because NDVI traits measured on grassland had a negative and nonlinear relationship with annual precipitation (Paruelo and Lauenroth, 1998). Further, landscape factors also caused variation of crop reflectance signal (Asner, 1998; Pachepsky and Ritchie, 1998). There is experimental evidence that field heterogeneity, such as SE, may also affect water and N distribution patterns and that dynamic soil and crop variables should then be measured as a function of space and time (van Es et al., 1989; van Es and van Es, 1993; Lascano et al., 1999; Cassel et al., 2000; Li et al., 2001a).
Understanding the relationship among crop reflectance, water and N inputs, and field heterogeneity would be useful for further evaluation of remote sensing as a tool for irrigation and fertilization monitoring. For this study, we hypothesized that water and N supply and field heterogeneity would affect cotton spectral reflectance and agronomic variables. The objectives of the study were to (i) assess cotton–soil reflectance related to plant biomass, plant water content (PWC), N uptake, lint yield, and soil properties with different rates of irrigation and N fertilizer; (ii) determine the effects of the treatments on spectral and agronomic response variables; (iii) develop NDVI maps; and (iv) determine the linear correlation and cross correlation between spectral reflectance and vegetation index, agronomic parameters, soil water and texture, and topography for in-season N management.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The 2-yr field study started in May 1998. The experimental area was 32 m wide and 700 m long across an altitude between 890.8 and 894.6 m above sea level. The experimental treatments consisted of irrigation at 50% and 75% of cotton potential evapotranspiration (ET) and N fertilization at rates of 0, 90, and 135 kg ha-1 arranged in an incomplete block of size-2 design and selected to minimize effects of field heterogeneity on outcome of treatments (van Es et al., 1989; van Es and van Es, 1993). There were four replicates for the N treatment reference (control) and N at a rate of 135 kg ha-1 and five replicates for N at a rate of 90 kg ha-1. Plot size was twelve 1-m rows wide and 50 m long. Cotton (‘Roundup Ready 2326’) was seeded at a rate of 16.8 kg ha-1 in early May each year. Fertilizer N {urea [(NH2)2CO], 32–0–0} was fractionally applied into soils on rows by a chisel at emergence at a rate of 45 kg ha-1; at bloom (beginning of July) for the 90 kg ha-1 rate; and at emergence, bloom, and first square stage (mid-July), respectively, for the 135 kg ha-1 rate. In-season insect and weed controls were applied according to regional recommendations.
Rainfall was 91 mm during the growing season in 1998, and in 1999, precipitation was 130 mm in June and dry the remaining period. Irrigation was applied using a LEPA (Low Energy Precision Application) irrigation system (Lyle and Bordovsky, 1981), and irrigation rates varied between 5.1 and 10.1 (50% ET) and 7.6 and 15.2 (75% ET) mm every 3 d, depending on rain amounts and plant requirements at different growth stages. Total irrigation applied was 215 and 323 mm in 1998 and 190 and 285 mm in 1999 for the 50 and 75% ET, respectively.
Multispectral Reflectance and Agronomic Measurements
Multispectral plant/soil reflectance across the wavelength between 447 and 1752 nm was measured using a portable MSR16 (8 up sensors and 8 down sensors) radiometer (CropScan, Rochester, MN). Measurement distance was 2 m from sensors (looking straight down) to crop canopy top. With a 31.1° field of view, the sensor viewed a 1.0-m-diam. ground area. On each main plot, plant and soil composite reflectance was measured on harvest rows from four areas: Two were 25 m apart on Row 6 and the other two also 25 m apart on Row 7. To maintain the integrity of the spectral measurement areas, plants were not sampled from these two rows. Reflectance was measured twice per week within a 15 to 30° solar zenith angle throughout the growing season. There were 16 to 20 readings per N treatment, totaling 104 readings at each measurement event. The sensor output consisted of reflectance in the visible, NIR, and midinfrared (MIR) bands at the center wavelength of 460, 559, 660, 661, 710, 810, 830, and 1650 nm, respectively.
A 0.5- by 0.5-m Spectralon panel (Labsphere, Orth Sutton, NH) sintered with polytetrafluoroethylene powder (a very bright reflector), with stable and known reflectance factors (typically between 98.6–99.1%), was used as a reference reflectance standard surface to calibrate the field reflectance measurements. This calibration was because field measurements of reflectance factors generally involve the measurements of radiance from a reference panel or a target with the radiometer (Jackson et al., 1987, 1992). In addition, a dry soil reference was measured on a 0.6- by 0.5-m soil tray (fairly constant reflectance with time) to further verify the calibration, as indicated by Jackson et al. (1987), to account for any drift in the calibration factors and any unexpected changes due to equipment problems that could occur during the season. Spectralon and soil tray readings were taken twice simultaneously at the beginning and the end of each field reflectance measurement event, as suggested by Jackson et al. (1987).
Leaf area index was simultaneously measured with a LAI-2000 Plant Canopy Analyzer (LI-COR, Lincoln, NE) at the reflectance measurement positions using the method of Hicks and Lascano (1995). Monthly soil water content (SWC) was measured in 0.3-m increments to a 1.8-m depth using a neutron probe (Model 503 Hydroprobe, CPN Corp., Martinez, CA). Neutron readings were field-calibrated with gravimetric SWC measured at the same soil depth from samples taken three times during the growing season. In addition, monthly surface (0–0.06 m) SWC was measured at reflectance measurement sites in 1999 with a frequency-domain sensor (ThetaProbe, Delta-T Devices Ltd., Cambridge, England) using the method of Lascano et al. (1999). Site elevation was measured using a SL 2001/3001 L-Band Receiver (Satloc, Scottsdale, AZ), which was calibrated with USDA-NRCS measurements.
In each plot, 16 to 32 plants were sampled from four 5- by 25-m areas, beside reflectance measurement locations, on a 10-d interval from early June to harvest. Root length and fresh and dry matter of roots, leaves, stems, bolls, and seeds were measured. Plant samples were dried at 70°C until constant weight and ground to 0.5 mm. Total plant N was determined using a LECO FP-528 Analyzer (Leco Corp., St. Joseph, MI). Soil was sampled on the row within 0.5 m of each neutron access tube to a depth of 1.8 m in 0.3-m increment at emergence, bloom, and harvest. Soil samples were air-dried and sieved to 2 mm. Soil texture (n = 52 per ET) was measured using the hydrometer method (Gee and Bauder, 1986), with soils sampled in July 1998. The 0.1 M KCl extractable NO3–N (Technicon Ind. Syst., 1986) was measured using a Technicon Auto-Analyzer II C (Technicon Instruments Corp., Tarrytown, NY). Cotton lint was hand-harvested from four 1- by 4-m areas at reflectance measurement locations in early October of each year.
Reflectance Discrimination and Calculations
The spectral radiometer readings were calibrated with a reflectance correction factor, which was calculated by applying the dark millivolt readings at 25°C, temperature sensitivity constants, up-facing sensor cosine parameters, and sun angle values (CropScan). Then, the converted spectral reflectance was expressed as percentage of reflectance, a ratio of output in millivolt of down and up sensors. This was multiplied by a correction factor (a ratio of the known Spectralon reflectance factors to Spectralon readings) to obtain corrected field reflectance values. Soil reflectance was discriminated in the red band between 648 and 674 nm, and plant reflectance was discriminated in the NIR band between 797 and 829 nm.
Total N uptake was calculated by multiplying total N content by plant dry matter. Plant water content was determined by difference between fresh and dry biomass. The SWC was obtained by converting neutron probe readings using field-specific calibration equations. The ratio vegetative index was expressed as NIR/red reflectance (Jackson, 1984; Sembiring et al., 1998). The NDVI was determined by the ratio of differencing and combining reflectance measured in NIR and red bands (RED) as described by Jackson (1984), Sembiring et al. (1998), and Plant et al. (2000):
 | [1] |
The spatial correlation between two variables x (soil texture or SE) and y (cotton reflectance, NDVI, or lint yield) was determined with the cross correlation covariance [Cxy(h)] and the cross covariance function [
xy(h)], as described by Cassel et al. (2000) as follows:
 | [2] |
 | [3] |
and 
are the mean of all measurements of x and y, respectively (Cassel et al., 2000). In Eq. [3], 
x and y are the standard deviations of x and y, respectively.
Data Statistics and Mapping
Regression analysis of relationships between spectral and agronomic variables and SE were evaluated using the General Linear Models procedure (SAS Inst., 1990). The effects of irrigation and N fertilizer on multispectral cotton reflectance, SWC, PWC, total N uptake, and cotton lint yield were tested using the PROC MIXED model procedure (Little et al., 1996). The cross correlation covariance and cross covariance function of the reflectance, lint yield, soil texture, and topography were determined using the AutoRegressive Integrated Moving Average (ARIMA) procedure (SAS Inst., 1993). Cotton NDVI maps were generated using data measured in a 2- by 6-m grid across the plots at plant maturity in August each year using MapInfo 6.0 (MapInfo Corp., Troy, NY).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
) and crop development (Fig. 1)
, measured in the 75% ET plots. The reflectance data measured in the 50% ET plots showed similar patterns (results not shown). As N treatments had no effect on reflectance data (Table 1), the means of all N treatments were plotted in Fig. 1. Sensor outputs are typically in the green, red, NIR, and MIR portions at center wavelength of 460, 559, 661, 810, and 1650 nm, respectively (Fig. 1). In June, red (637–674 nm) reflectance was higher (Fig. 1a), which corresponded to a higher percentage of exposed soil surface at the early vegetative stage. As cotton plants grew, red radiance decreased and reflected NIR (797–829 nm) increased quickly from June to August (Fig. 1a). The peak of NIR reflectance was measured in mid-August at plant maturity. As a result of a decrease in LAI (from 2.5 to 1.8) near the open-boll growth stage in September, the red increased, and reflected NIR decreased simultaneously (Fig. 1a).
|
|
The NIR and red reflectance-based NDVI, calculated with Eq. [1], showed a temporal pattern comparable to that of LAI (Fig. 2) . Because there was no effect of N on NIR and NDVI (Table 1), data plotted in Fig. 2 were the mean of all N treatments. The LAI varied between 0.1 and 2.7 during the two growing seasons. In 1998, the maximum average NDVI (n = 52) was 0.69 (range of 0.54–0.82) and 0.76 (range of 0.57–0.83) at the 50 and 75% ET, respectively, determined on 17 August. Both NDVI and LAI (determined in 1998) increased proportionally with increasing irrigation level (Fig. 2). The peak of NDVI curve was situated towards the end of August. High NDVI was a result of an increase in the NIR band and a decrease in the red band. The NDVI was low at early vegetative growth stage and harvest because the composite reflectance was high in the red band due to less plant ground cover.
|
|
Mixed Effects of Irrigation on Spectral Reflectance
Mean cotton lint yields were 704 and 962 kg ha-1 in 1998 and 819 and 924 kg ha-1 in 1999 at 50 and 75% ET irrigation levels, respectively. Lint yield varied more in 1998 than 1999; standard deviations were 118 and 186 kg ha-1 in 1998 compared with 146 and 144 kg ha-1 in 1999 at 50 and 75% ET, respectively. The fixed effect of irrigation, determined with a mixed-model procedure, was significant on all measured spectral and agronomic parameters. The fixed effect of N fertilizer was only significant (P < 0.0281) on lint yield in 1999 (Table 1). Interaction between irrigation and N fertilization was significant only on reflectance in the NIR band (P < 0.0152) in 1998, but the interaction was very significant on red and NIR reflectance as well as NDVI in 1999 (Table 1).
The random effects of the mixed-model variance components (block and treatments) were not significant (Table 2). However, the model residual was significant on all measured spectral and agronomic parameters. The Z-statistics for these model variance components are based on asymptotic normality. Except for the N uptake in 1999, all Z-statistics values varied between 6.36 and 6.56, with P = 0.0001 (Table 2). The model residual effects consisted of the covariance of other independent variables such as SE, slope length, and soil texture.
|
Variations of spectral reflectance with SWC and SE are shown in Fig. 4 when the spectral and SWC measurements were taken across all plots at plant maturity in mid-August 1999. The surface SWC (0–0.06 m), which would influence spectral soil and cotton plant reflectance, ranged between 0.060 and 0.273 m3 m-3 (Fig. 4a and 4d). The NIR reflectance (center 810 nm) was significantly higher (Table 1) on 75% ET plots (44.5%) than on 50% ET plots (41.9%). Correspondingly, the narrow red reflectance (center 661 nm) was significantly lower (Table 1) on the 75% ET (6.8%) than the 50% ET plots (8.3%). Compared with the 75% ET, the MIR reflectance (center 1650 nm) was higher at the 50% ET (24.9 vs. 22.1%). From south to north across the field, the consistent trend was that the NIR reflectance was higher (48 vs. 41%) on lower positions than on upslopes while red and MIR reflectance increased (6.8 to 10.6% and 22.3 to 30.8%, respectively) towards the northern upslope areas (Fig. 4a and 4b). Spectral soil and cotton plant characteristics seem to be linked to irrigation level and SE variability.
|
Correlation of Cotton Spectral and Agronomic Variables and Field Heterogeneity
Spectral reflectance in the blue, green, red, and MIR bands was positively correlated with SE and clay content, but the NIR reflectance, NDVI, and NIR/red ratio had a negative correlation relationship with these two field-heterogeneity variables (Table 3). The SE and sand content were also closely linked to SWC, cotton lint yield, and total N uptake (2-yr mean). The negative or positive correlation coefficient (r) indicated that higher lint yield and N uptake and more SWC accumulation were associated with low elevation and high sand content (Table 4). The reflectance in the visible, NIR, and MIR bands was highly correlated with each other (-0.67 < r < 0.98). There was a negative correlation of blue, green, red, and MIR reflectance with lint yield, N uptake, and SWC. Red and MIR reflectance was also negatively correlated with PWC (Table 4). Conversely, the NIR reflectance, NDVI, and NIR/red ratio were positively correlated with these parameters (Table 4).
|
|
|
0.74) with higher frequency (15%) compared with the 50% ET plots. The heterogeneous distribution of the NDVI in the area with the same irrigation and N fertilization level reflected the impact of SE and soil texture on soil water distribution, N uptake, and plant growth.
|
|
xy(h)], calculated with Eq. [3], varied between -0.68 and 0.64. Considering the 95% confidence level as described by Cassel et al. (2000), the NIR reflectance (Fig. 8a) and NDVI (Fig. 8b) were positively cross-correlated with SWC across a lag distance of ±30 m (or separation distance of 60 m). The NDVI and sand (Fig. 8c) and red reflectance and clay (Fig. 8d) were also positively cross-correlated with each other across a distance between 60 and 75 m. The cross correlation distance between SE and NIR reflectance, NDVI, and red and MIR reflectance varied between 60 and 80 m (Fig. 8e, 8f, 8g, and 8h). As the lag distance increased, these cross correlation functions positively or negatively approached zero at a lag distance ±180 m (Fig. 8).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Sand content appears to have more impact on spectral reflectance, soil water, and lint yield than does clay (Table 4). Surface soils (0–0.3 m) in depressed areas contained on average 10% more sand than on northern upslope and summit where soils had 8% more clay (127–234 g kg-1) than on depressed areas (116–210 g kg-1). Sand has accumulated in the surface layer in depressed areas where the surface soil dried, but the subsoil contained more water (Fig. 4) where there was a higher plant density. As a result, NIR increased in depressed areas where surface soil contained more sand and less clay, which led to a negative correlation between visible reflectance and sand and SE (Table 3). Because higher sand content was found on downslope areas where NIR reflectance increased, correlations between sand and visible reflectance were negative (Table 3).
Increasing SWC resulted in an increase of PWC, PFB, N uptake (Table 4), and NIR reflectance but a decrease of reflectance in the visible and MIR bands (Fig. 4). Jackson (1984) and Bowman (1989) reported that NIR and MIR reflectance varied primarily as a function of leaf water content. Reflectance in the NIR band is determined by plant cell structure and chlorophyll and leaf N content, which were positively correlated (Bausch and Duke, 1996). Our results suggested that a decrease in visible reflectance and an increase in NIR reflectance would indicate a higher water and N content in plants based on the r values (Table 4).
Treatments and Field Heterogeneity
The amount of irrigation and interaction between irrigation and N fertilizer also appeared to significantly affect spectral plant and soil reflectance, N uptake, and lint yield (Table 1). Soil NO3–N amounts could explain the lack of cotton response to N treatments. In this field, N fertilizer at a rate of 190 kg ha-1 yr-1 has been applied to the soil for the last 5 yr. The soil NO3–N (including residual and fertilizer N) in the effective root zone (0–0.9 m) at bloom was 179 ± 83 kg ha-1 at the 50% ET and 174 ± 76 kg ha-1 at the 75% ET in July 1998, and the soil NO3–N was declined but remained moderately high (103 ± 30 kg ha-1 at the 50% ET and 120 ± 42 kg ha-1 at the 75% ET) in early May (before seeding) 1999. Therefore, larger amounts of soil NO3–N was one of the factors causing no effect of N treatments on the spectral and agronomic parameters in 1998 (Table 1), and the effect of N rates on lint yield and interaction between the treatments on the red and NIR reflectance and NVDI was significant (Table 1).
Poor rain infiltration and resulting runoff might have influenced cotton response to the treatments. Significant effects of N fertilizer and interaction between irrigation and N fertilizer on lint yield could be expected in a wet year such as in 1999 (Table 1). The 1998 growing season was exceptionally dry (total rain was 22% of the 30-yr average), but the 1999 season was unusually wet in June (190% of the 30-yr average). The SWC in the rooting depth (0–0.9 m) in June was on average 0.14 m3 m-3 in 1999 compared with 0.09 m3 m-3 in 1998. As a result, the lint yield with 50% ET increased 14% in 1999 compared with 1998. Soil water and N supply were strongly linked to cotton production (Wiegand et al., 1994; Plant et al., 2000; Li et al., 2001a), and annual rain difference contributed to grass NDVI variability (Paruelo and Lauenroth, 1998).
Changes in cotton spectral characteristics, soil water, N uptake, and lint yield were largely controlled by SE rather than by soil texture and NO3–N. The significance of the mixed-model residual (Table 2) indicated the degree of influence of covariables such as SE, slope length, and sand and clay content on cotton spectral characteristics, soil water, N uptake, and lint yield. Our results suggested that real-time reflectance in the NIR and red bands and NDVI could represent real-time soil conditions and plant development. An in-season variable water and N application related to local field heterogeneity would be determined by combining a NDVI map (Fig. 7) with a cross correlation distance (Fig. 8). For example, the amount and locations for N application can be determined from a real-time NDVI map, and the cross correlation distance between NDVI and SE can be used as a variable rate application distance.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Yabaji, J. W. Nusz, K. F. Bronson, A. Malapati, J. D. Booker, R. L. Nichols, and T. L. Thompson Nitrogen Management for Subsurface Drip Irrigated Cotton: Ammonium Thiosulfate, Timing, and Canopy Reflectance Soil Sci. Soc. Am. J., March 1, 2009; 73(2): 589 - 597. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. G. Sullivan and C. C. Holbrook Using Ground-Based Reflectance Measurements as Selection Criteria for Drought- and Aflatoxin-Resistant Peanut Genotypes Crop Sci., May 31, 2007; 47(3): 1040 - 1050. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Pena-Yewtukhiw, G. J. Schwab, and L. W. Murdock Univariate Distribution Analysis to Evaluate Variable Rate Fertilization Agron. J., April 11, 2006; 98(3): 554 - 561. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Clay, K.-I. Kim, J. Chang, S. A. Clay, and K. Dalsted Characterizing Water and Nitrogen Stress in Corn Using Remote Sensing Agron. J., April 11, 2006; 98(3): 579 - 587. [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]
|
|
|
|
|
|
J. Ko, S. J. Maas, R. J. Lascano, and D. Wanjura Modification of the GRAMI Model for Cotton Agron. J., September 19, 2005; 97(5): 1374 - 1379. [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]
|
|
|
|
|
|
T. T. Chua, K. F. Bronson, J. D. Booker, J. W. Keeling, A. R. Mosier, J. P. Bordovsky, R. J. Lascano, C. J. Green, and E. Segarra In-Season Nitrogen Status Sensing in Irrigated Cotton: I. Yields and Nitrogen-15 Recovery Soil Sci. Soc. Am. J., September 1, 2003; 67(5): 1428 - 1438. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. F. Bronson, T. T. Chua, J. D. Booker, J. W. Keeling, and R. J. Lascano In-Season Nitrogen Status Sensing in Irrigated Cotton: II. Leaf Nitrogen and Biomass Soil Sci. Soc. Am. J., September 1, 2003; 67(5): 1439 - 1448. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. F. Bronson, J. W. Keeling, J. D. Booker, T. T. Chua, T. A. Wheeler, R. K. Boman, and R. J. Lascano Influence of Landscape Position, Soil Series, and Phosphorus Fertilizer on Cotton Lint Yield Agron. J., July 1, 2003; 95(4): 949 - 957. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Li, R. J. Lascano, J. Booker, L. T. Wilson, K. F. Bronson, and E. Segarra State-Space Description of Field Heterogeneity: Water and Nitrogen Use in Cotton Soil Sci. Soc. Am. J., March 1, 2002; 66(2): 585 - 595. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
