| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Eastern Cereal and Oilseed Res. Cent. (ECORC), Cent. Exp. Farm, Res. Branch, Agric. and Agri-Food Canada, 960 Carling Ave., Ottawa, ON, Canada, K1A 0C6
b Univ. of Paso Fundo, Paso Fundo, RS, 99001-970, Brazil
* Corresponding author (mab{at}agr.gc.ca)
Received for publication May 18, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND GENERAL CONCLUSIONSREFERENCES |
|---|
at V6 were analyzed. Relationships of parameters collected early in the growing season vs. grain yield, harvest index, and total plant N uptake at maturity were determined. In 2 yr (2000 and 2002), grain yields increased significantly with fertilizer rates up to 120 kg N ha–1. While soil mineral N and plant N concentrations differentiated 0 N from preplant N at 40 kg N ha–1, both leaf chlorophyll and canopy reflectance measured at V6 stage responded linearly to fertilizer N up to 120 kg N ha–1. We concluded that these leaf and canopy optical measurements could be used as crop-based indicators for early-season N amendment.
Abbreviations: CAN, calcium ammonium nitrate • CHU, crop heat unit(s) • HI, harvest index • NDVI, normalized difference vegetation index • PSNC, presidedress soil nitrate concentration
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND GENERAL CONCLUSIONSREFERENCES |
|---|
In the humid climate regions, such as in eastern Ontario, N fertilization for corn is generally applied once shortly before or at planting and may again be sidedressed from the V6 to V8 growth stages (Ritchie et al., 1992). In some cases, the entire N addition was applied as sidedress due to restrictions at planting. In this humid environment, corn yield response to N amendments is poorly correlated with soil mineral N at preplant or presidedress (Ma and Dwyer, 1999). Postemergence application of N serves to minimize early-season N losses in wet years (Scharf et al., 2002), which occur with some frequency in the region. However, delays in N application can lead to irreversible yield losses; for example, in one case, grain yield was reduced by 12% when N application was postponed to V6 (Binder et al., 2000). Scharf et al. (2002) found that yield remained responsive to N application until silking, but full yield was not achieved when applications were delayed until then.
Application of some or all of the N fertilizer after planting may allow for a more precise matching of the quantity of N required to local conditions. On one hand, the required quantity could be based on estimates of soil N availability obtained through soil testing or predictions based on climatic, soil, and genotypic factors for a given season and location (Scharf et al., 2002). All of these factors have been shown to influence the response of corn to N fertilizer applications (Magdoff et al., 1984; Scharf et al., 2002; Andraski and Bundy, 2002). On the other hand, the required quantity could be based on a diagnosis of crop N needs through in-season crop tissue testing or crop canopy optical sensing. Consequently, knowledge of both abiotic and biotic factors, related to soil N availability and crop N needs, respectively, are prerequisites to developing best management practices to maximize yield response to fertilizer N (Muchow, 1998).
Presidedress soil nitrate concentration (PSNC) has shown promise as a means of quantifying and improving N management for corn production (Magdoff et al., 1984; Binford et al., 1992). A critical value of PSNC was defined to be 20 to 30 mg NO–3–N kg–1 soil (Magdoff et al., 1984). In southwest Ontario, Vyn et al. (1999) reported a positive correlation between PSNC and subsequent corn yields. Andraski and Bundy (2002) suggested that adjusting N application rates for corn using the PSNC was more profitable than not making these adjustments. However, in humid environments, such as eastern Ontario, it is difficult to use PSNC as a relevant guide for N application. Furthermore, collecting a representative number of samples to account for spatial variability is time consuming and costly (Ma and Dwyer, 1999).
A number of nondestructive methods have been used to monitor in-season N status of crop plants, including leaf greenness and canopy reflectance. Some instantaneous diagnostic techniques to monitor crop N status, such as the chlorophyll meter (Minolta SPAD-502), were found to be good predictors of corn grain yield (Wood et al., 1992; Blackmer and Schepers, 1995; Waskom et al., 1996). The optimal measurement times for in-season N amendment were reported to be at V10 and tasseling stages (Waskom et al., 1996). However, compared with soil indicators, leaf greenness was unable to quantify excess available N at early development stages (Schröder et al., 2000). The narrow range of leaf chlorophyll readings at V6 made it difficult to separate N-deficient from N-sufficient regions of a field (Dwyer et al., 1991), and a large number of observations were required (Costa et al., 2001).
The use of remote-sensing techniques such as measuring canopy light reflectance could help to eliminate the need for extensive field sampling (Gilabert et al., 1996; Ma et al., 1996). Crop reflectance is defined as the ratio of the amount of radiation reflected by an individual leaf or canopy to the amount of incident radiation (Schröder et al., 2000). Several researchers have used multispectral canopy reflectance to measure the N status and to predict grain yield potential in different crops, including corn (Bausch and Duke, 1996; Ma et al., 1996; Osborne et al., 2002b), rice (Oryza sativa L.) (Casanova et al., 1998), soybean [Glycine max (L.) Merr.] (Ma et al., 2001), cotton (Gossypium hirsutum L.) (Bronson et al., 2003), and wheat (Triticum aestivum L.) (Aparicio et al., 2000; Flowers et al., 2003). Similarly, Osborne et al. (2002a) used the spectral reflectance to detect P and N deficiencies in corn. Such methods may be limited by location-specific conditions (Schröder et al., 2000). For example, weather variables and soil background reflectance variables may influence equipment performance.
None of these methods presents a perfect solution. While these tools have been studied and tested separately, comparative studies of their reliability, efficiency, and interrelationships have seldom been undertaken. Whether the N status assessments from a young crop can be used to estimate grain yield is equally unclear. Furthermore, as N is lost from cropping systems through a number of pathways, a search for N best management practices cannot limit itself to a single solution (Binder et al., 2000).
Techniques that provide a rapid and accurate assessment of soil and plant N status on a regular and short-term basis would be useful for managing in-season N amendments (Bausch and Duke, 1996). We hypothesize that the precision of N application can be improved by fine-tuning the application rates according to crop-based indicators of N status. Consequently, this study was initiated to (i) determine the appropriate rates and timing of N applications in the humid environment of eastern Ontario, (ii) evaluate the ability of nondestructive plant-based methods in comparison with the PSNC test in discriminating fertilization N application rates near sidedress time, and (iii) document how yearly variations in environmental conditions affect the ability of different approaches to assess corn N status.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND GENERAL CONCLUSIONSREFERENCES |
|---|
|
concentrations each year. In 2001 and 2002 growing seasons, six continuous plants were sampled from the middle portion of the second row from each plot at the V6 growth stage. Dry matter of all the plants was recorded after drying at 70°C for >72 h. Samples were ground to pass a 1-mm sieve and analyzed for total N concentration by micro-Kjeldahl method. When plants in most plots reached V6 to V8 stages, sidedressing with CAN was applied on the soil surface in double bands (5 cm away from plant rows) in the designated treatments and incorporated through cultivation immediately after application. Plant phenological events were recorded from emergence to physiological maturity (black layer or zero milk line). Total precipitation during the growing season was recorded, and crop heat units (CHU) accumulated from planting to physiological maturity (R6) were determined following Brown and Bootsma (1993). Minimum nighttime temperatures below 4.4°C were set to 4.4°C, and maximum daytime temperatures below 10°C were set to 10°C. Crop heat units are calculated separately from the daily average of Ymax, representing daytime temperature relationships:
 | [1] |
 | [2] |
 | [3] |
Biweekly simultaneous measurements of canopy reflectance and leaf chlorophyll content were taken from V5 to V8. Additional measurements were also made at V10, silking (R1), and 6 wk after silking (R4). Before silking, leaf chlorophyll readings (SPAD–502 Chlorophyll Meter, Minolta Camera Co. Ltd., Tokyo, Japan) were taken on the topmost fully expanded leaves of 30 plants per plot. At R1 and R4, the readings were taken on ear leaves of five plants per plot.
Canopy reflectance measurements were taken using a hand-held multispectral radiometer (MSR16, CropScan Inc., Rochester, MN), which records percentage light reflectance in 11 bands (460, 507, 559, 613, 661, 706, 769, 813, 850, 900, and 950 nm). Each band has a half-peak band of approximately 5 to 15 nm, depending on the specific pass band. The sensing method used is band-limited optical interference filters and photodiodes. The band-limited optical filters only pass wavelengths of irradiance in the pass-band range to the active surface of the detecting photodiode. The photodiode outputs current in direct proportion to the number of photons striking the photodiode. This electrical current was converted to a voltage and amplified by the circuitry in the radiometer. The data logger controller measured and logged these sensor millivolt readings. Data of percentage reflectance at each pass band were processed subsequently by a computer program using the calibration and correction constants through a minicomputer connected to the sensor. The sensor head was mounted on an adjustable pole. The pole was hand held at an angle between 45° and 60° relative to the ground surface. The sensor head receptor facing the center of the plot (between Rows 3 and 4) was parallel to the ground surface at a perpendicular height of about 3 m above the ground. At each sampling date, duplicate measurements were taken within each plot. It took approximately 50 min to complete measurements for the experiment. Measurements were taken from 1000 to 1100 h on a sunny day with no winds. The sensor readings were used to derive five different vegetation indices as follows:
 | [4] |
 | [5] |
 | [6] |
 | [7] |
 | [8] |
At R6 growth stage, a subsample of five continuous plants was taken from the middle portion of the fifth row in each plot and oven-dried to constant weight at 70°C. The dried grain and stover samples were ground to pass a 1-mm screen for the determination of N concentrations. Harvest index (HI) was calculated as the ratio of grain dry matter to the total aboveground biomass. Grain yield was harvested by combining the central two rows (13.7-m2 area) of each plot and reported as kg ha–1 on a 155 g water kg–1 basis.
All the data each year were first subjected to the analysis of variance using the GLM procedure of SAS (SAS Inst., 1996). Analysis of variance was performed separately for the treatments containing only preplant fertilizer N (0, 40, 80, 120, and 250 kg N ha–1) before sidedress and for all eight N treatments thereafter. When variables were found to be significant, the nature of response was tested using orthogonal polynomial contrasts (linear, quadratic, and cubic) on the four equally spaced treatments (i.e., 0, 40, 80, and 120 kg N ha–1). Treatment means were separated by the ANOVA-protected least significant difference (LSD0.05) test.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND GENERAL CONCLUSIONSREFERENCES |
|---|
|
|
|
Values of NDVI increased from approximately 0.12 at V5 to 0.82 by R1. The NDVI measurements obtained at the R1 growth stage correlated more strongly with grain yield (data not shown) than those measured before V8. This relationship indicates that maintenance of crop health and greenness after the sidedress stage is important, and greenness, especially at silking, is crucial for higher grain yield. Growth conditions and N supply after V8 played a more important role with respect to grain yield than those preceding the V8 stage.
Canopy reflectance was not responsive to N fertilization regimes until the V6 stage as was leaf chlorophyll meter readings. Measurements taken after V6 were sufficient to differentiate crop N status based on N treatments. Unlike the chlorophyll meter, measurements of canopy reflectance were much quicker, taking roughly one-fifth the time for chlorophyll meter readings. Between V6 and V8 stages, the response of leaf chlorophyll meter readings or canopy reflectance to preplant N rates was linear up to 120 kg N kg–1, suggesting their suitability as crop-based indicators for corn N management. Both chlorophyll meter readings and NDVI measured at silking were highly correlated with yield (r 
0.9) at harvest, which agrees with earlier studies (Ma et al., 1996).
Whole-Plant Nitrogen Concentrations at Presidedress Stage
Plant samples at V6 growth stage were not analyzed in 2000. In 2001, there was no effect of N treatment on plant dry weight at V6. In general, whole-plant N concentrations ranged from 37.6 ± 1.4 to 38.6 ± 3.0 g kg–1 in all preplant N treatments, which was significantly higher than in the 0 N control (36.2 ± 2.7 g kg–1). However, significant hybrid-by-N application rates (P < 0.05) occurred in 2001 (Table 2). For Pioneer 3905, whole-plant N concentrations increased linearly with preplant N application rates up to 120 kg ha–1 while for Pioneer 39F06 Bt, response of whole-plant N concentration to preplant N application rates was significant only up to 80 kg N ha–1 (Fig. 3). In 2002, effects of hybrid and preplant N applications on whole-plant N concentration were independent; the 250 kg ha–1 slow-release N treatment showed the highest plant dry weight, N concentration, and N content. The N concentration for the slow-release N treatment was significantly higher (32.3 g kg–1) than the 120 kg N ha–1 treatment (28.7 g kg–1), and the effect of preplant N application on the whole-plant N concentrations was linear and more consistent in Pioneer 39F06 Bt than in Pioneer 3905 (Fig. 3).
|
Presidedress Soil Nitrate Nitrogen
Nitrogen fertilization regimes did not affect PSNC in 2000 (averaged 14.1 µg kg–1) or in 2001 (11.2 µg kg–1); however, in 2002, differences in PSNC did exist between the five preplant N treatments (Table 2; Fig. 4). Application rates of 120 and 250 kg N ha–1 resulted in similar and significantly greater PSNC than the 0 or 40 kg N ha–1 treatments. Soil NH4+ concentrations were very low in all treatments (data not shown). As the preplant N was banded in this study, PSNC was expected to vary greatly; by V6 growth stage, fertilizer N would have not been evenly distributed in the soil profile as it would via broadcast, i.e., banding added spatial variability in soil sampling compared with broadcasting method. The large variations in PSNC within a treatment and across years (Fig. 4) highlighted the requirement of large sample size and difficulty to use it as an indicator for corn N amendments in this humid region. Our results contrast with previous studies in the use of PSNC for corn sidedressing recommendation in semiarid environments (Magdoff et al., 1984; Blackmer et al., 1989; Binford et al., 1992; Klausner et al., 1993; Vyn et al., 1999), perhaps partially being associated with the fact that nowadays, preplant fertilizer N banded at planting is more common than broadcast applications.
|
Response of Grain Yield to Applied Nitrogen
In 2000, the application of 250 kg slow-release N ha–1 resulted in greater grain yield than other preplant N treatments, except for N applied at 120 kg ha–1 (Fig. 5). Nitrogen treatments of 80 kg N ha–1, applied either as preplant, or sidedressed, or split equally as preplant and sidedress, produced similar grain yields. The 0 N control treatments had the lowest grain yield. The effect of N treatment on grain yield in 2001 was not significant because of the unusually dry conditions. In 2002, grain yield was significantly different among N treatments. As expected, 0 N control treatments produced the lowest yield. In general, grain yield increased with N applications up to 120 kg N ha–1. These results indicate that N applications exceeding 120 kg N ha–1 in this environment are not efficient, and thus N might have been wasted.
|
It is very likely that grain yield is affected to a larger extent by the conditions after sidedressing. In 2001, the response of yield to N rates was confounded by drought effects. Although the crop-based indicators differentiated N treatments at V6 in the other 2 yr, NDVI, chlorophyll meter readings, or whole-plant N concentrations at V6 were not strongly correlated with grain yield. This indicates that under adverse environmental conditions, the early crop N status is not fully reflected into final grain yield. Besides climatic effects, the change in N response according to the crop development stage and the pattern of soil N-supplying power (NO–3–N release) affect the accuracy of early grain yield prediction. During the first 3 wk after emergence, corn takes up soil mineral N at a rate less than 0.5 kg ha–1 d–1 (Blackmer et al., 1989). It begins to rapidly take up N during the middle of the vegetative growth period, peaking near silking (Hanway, 1960, as cited by Binder et al., 2000; Ma and Dwyer, 1998). Excess soil N during the initial stage might move from the rooting zone and ultimately be lost (Blackmer et al., 1989). Thus, applying N at sidedress (V6–V8) should be one of the best ways of supplying N to meet the higher demand (Binder et al., 2000).
Nitrogen Uptake and Partitioning
The aboveground N uptake measured at crop maturity was greatly influenced by N treatment in all 3 yr (Fig. 6). Although in 2002, the effect of N fertilization regime was not observed in grain yield, total N uptake in grain and stover was significantly affected. From year to year, slow-release N at 250 kg ha–1 consistently resulted in higher N accumulation in grain and stover than other preplant N treatments, except for preplant N at 120 kg ha–1 or a combination of preplant and sidedress N of 40 + 160 kg ha–1 (Fig. 6). It is interesting to note that in all years, sidedressed N at 80 kg N ha–1 resulted in significantly greater total N uptake at crop maturity than the same amount of N applied as preplant (Fig. 6). This indicated that sidedressed N was more efficiently utilized, especially in grains, which also agrees with Subedi and Ma (2005) in that the later N was applied during the growing stage, the greater the proportion of N reaching the grain. On average, grain contained 63, 75, and 73% of total aboveground plant N in 2000, 2001, and 2002, respectively. The lower percentage of N in 2000 was mainly attributed to an overall lower HI as a result of the late planting and shortage of CHU during the growing season, forcing the crop to mature before dry matter and N were fully partitioned to grains.
|
|
SUMMARY AND GENERAL CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND GENERAL CONCLUSIONSREFERENCES |
|---|
In general, our data demonstrated that application of N above 120 kg ha–1 had no yield and grain N concentration advantage in this particular environment. Narrow margins in grain yield difference existed among the ranges of N treatments in 2000 and 2002, and no difference was observed in 2001. The reason for the narrow difference in grain yields between the highest and lowest amounts of N applications was not clear. Probably, it was associated with lower efficiency of applied N and larger soil N-supplying power for the region. The fact that the higher rate of N treatments had smaller HI and vice versa indicates that additional N increased total biomass (derived from HI and yield), but partitioning of biomass to grain yield was unchanged or even negatively affected.
The benefit of a slow-release fertilizer could not be realized in terms of grain yield and N uptake. Rather, there was an excess mineral N release from the system, especially in the wet years. The N uptake data also clearly indicated that sidedressed N at the same amount was more efficient than preplant N application (e.g., when comparing 80 + 0 vs. 0 + 80). Preplant N application could lead to the risk of the unused mineral N leaching out of the rooting zone and/or emitting N gas; up to V6 growth stage (accounting for one-third of its life cycle), the corn crop takes up only a small portion (up to 15%) of its total N accumulation. On the other hand, by V8, crop demand for N had increased exponentially; thus, the sidedressed N was rapidly taken up by responsive plants resulting in minimal N loss. It has been well established that corn begins to rapidly take up N after V6, with the maximum rate of uptake occurring near silking (Ma and Dwyer, 1998; Binder et al., 2000).
The variations in weather patterns, particularly rainfall, influenced corn response to N treatment assessed by different approaches. Under drought-stressed conditions such as that occurring in 2001, all of the indicators tested poorly determined the N response in corn. Similarly, it has been observed that environmental conditions after V6 growth stage played more dominant roles in the determination of final grain yield than the indicators at an early stage.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND GENERAL CONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. M. Samborski, N. Tremblay, and E. Fallon Strategies to Make Use of Plant Sensors-Based Diagnostic Information for Nitrogen Recommendations Agron. J., July 7, 2009; 101(4): 800 - 816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. C. Scharf and J. A. Lory Calibrating Reflectance Measurements to Predict Optimal Sidedress Nitrogen Rate for Corn Agron. J., May 8, 2009; 101(3): 615 - 625. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. D. Subedi and B. L. Ma Dry Matter and Nitrogen Partitioning Patterns in Bt and Non-Bt Near-Isoline Maize Hybrids Crop Sci., May 31, 2007; 47(3): 1186 - 1192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. L. Ma, K. D. Subedi, and T. Q. Zhang Pre-Sidedress Nitrate Test and Other Crop-Based Indicators for Fresh Market and Processing Sweet Corn Agron. J., January 1, 2007; 99(1): 174 - 183. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. D. Subedi, B. L. Ma, and D. L. Smith Response of a Leafy and Non-Leafy Maize Hybrid to Population Densities and Fertilizer Nitrogen Levels Crop Sci., July 25, 2006; 46(5): 1860 - 1869. [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 | |||
0.05).
