Agronomy Journal Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 January 2007
Published in Agron J 99:174-183 (2007)
DOI: 10.2134/agronj2006.0028
© 2007 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ma, B. L.
Right arrow Articles by Zhang, T. Q.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Ma, B. L.
Right arrow Articles by Zhang, T. Q.
Agricola
Right arrow Articles by Ma, B. L.
Right arrow Articles by Zhang, T. Q.
Related Collections
Right arrow Best Management Practices
Right arrow Nitrogen
Right arrow Maize

Nitrogen Management

Pre-Sidedress Nitrate Test and Other Crop-Based Indicators for Fresh Market and Processing Sweet Corn

B. L. Maa,*, K. D. Subedia and T. Q. Zhangb

a Agriculture and Agri-Food Canada, Eastern Cereal and Oilseed Research Centre (ECORC), Central Experimental Farm, 960 Carling Ave., Ottawa, ON, Canada K1A 0C6
b Agriculture and Agri-Food Canada, Greenhouse and Processing Crops Research Centre, 2585 County Rd. 20, Harrow, ON, Canada N0R 1G0

* Corresponding author (mab{at}agr.gc.ca)

Received for publication January 31, 2006.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
Commercial sweet corn (Zea mays L.) production requires significant quantities of fertilizer N, leading to inefficient N use and negative environmental impact. A field experiment was conducted for 4 yr (2001–2004) in Ottawa, Canada, to assess and compare presidedress soil nitrate test (PSNT) with some crop-based measurements (canopy reflectance, leaf chlorophyll and plant total N) for improved N management. A fresh market sweet corn (FMSC, hybrid ‘Temptation’) grown from 2001 to 2003, and a processing sweet corn (PSC, hybrid ‘Hollywood’) from 2002 to 2004, both received five fertilizer N rates (0, 50, 100, 150, and 200 kg N ha–1). Soil samples taken from the V4 to V8 growth stages were analyzed for NO3––N. Leaf chlorophyll content (SPAD) and canopy reflectance were also measured for FMSC at the same time. All N treatments affected the number of marketable ears, kernel dry weight and total biomass production. However, in most cases, there was no difference between N treatments from 100 to 200 kg ha–1. The PSNT NO3––N increased linearly with the fertilizer N rates, and there were significant positive correlations between PSNT at V4 to V6 and the number of marketable ears. It was evident that PSNT, plant N concentration at V6, SPAD and canopy reflectance all differentiated sweet corn N response similarly, and they were highly correlated with one another. We concluded that PSNT at V4 to V6 was effective in predicting sweet corn N requirement in this cool and short-growing region.

Abbreviations: DM, dry matter • DW, dry weight • FMSC, fresh-market sweet corn • NDVI, normalized difference vegetative index • PSC, processing sweet corn • PSNT, presidedress nitrate test • SPAD, leaf chlorophyll content


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
NITROGEN FERTILIZER for sweet corn production is usually broadcast and incorporated into the soil near planting in the spring. To increase fertilizer N use efficiency and reduce N loading to the environment, fertilizer N can also be applied as a combination of preplant and sidedress applications (Ma et al., 2005) near V6 stage (Ritchie et al., 1993). Applying N fertilizer without information about N-supplying capacity of the soil can contribute to leaching of nitrate (NO3–) and groundwater pollution (Heckman et al., 1995). On the other hand, lack of soil mineral N during the critical stages of corn development may lead to significant yield reductions. For example, Uhart and Andrade (1995) reported that N stress did not affect corn grain yield if N was available 15 to 20 d before flowering, while Subedi and Ma (2005) observed that restriction of N supply before V8 growth stage causes an irreversible adverse effect in final yield of field corn.

There is a growing environmental concern for excessive N application in the FMSC production. Nitrogen fertilizer recommendations for sweet corn are generally based on yield goals. Nitrogen application based on yield goals established before planting may result in under- or overapplication of N because of weather-induced variations in yield potentials from year to year (Derby et al., 2004) and soil N mineralization (Ma et al., 1999). Matching soil N supply with plant demand is one of the main nutrient management challenges of crop production (Heckman, 2002). There is less time for leaching or denitrification loss when N is applied after plant emergence (Vetsch and Randall, 2004). Studies have shown that sidedress N applied during early growth (i.e., close to the time of increasing demand by the crop) is used more efficiently than preplant applied N (Magdoff et al., 1984; Magdoff, 1991; Ma et al., 2005). The general consensus is that N should be applied near the time when it is needed by the crop.

The rate of N uptake by corn is relatively low before entering the period of rapid growth at the V6 stage (Ma and Dwyer, 1998). In the northern climate, sufficient soil moisture and rapid increase in temperatures from late May to late June leads to increased soil N mineralization (Ma et al., 1999), while the highest rates of N uptake by corn do not occur, yet (Magdoff, 1991; Ma et al., 1999). The NO3––N that accumulates during the early part of the growing season is usually still present in the root zone during peak demand period (Magdoff, 1991). Some testing procedures have been developed to improve N fertility management for field corn production. Magdoff et al. (1984) proposed the presidedress soil NO3––N test, which recommends the amount of fertilizer N requirement based on the soil test during crop growing season. This approach involves: (i) time (when corn is 20–30 cm tall), and (ii) soil depth (top 30 cm) of sampling. Using the PSNT to make fertilizer N recommendations can minimize overapplication of fertilizer N by farmers while at the same time guarding against N-deficient corn (Magdoff, 1991; Heckman et al., 1995), allowing for a more precise matching of the quantity of N required under local conditions. Andraski and Bundy (2002) concluded that adjusting N application rates for corn using PSNT is more profitable than not making such adjustment. Therefore, PSNT shows promise as a means of quantifying and improving N management for corn production (Magdoff et al., 1984; Binford et al., 1992). Although presidedress analysis of NO3––N was developed as a useful soil test for humid regions where soil NO3––N is low at planting and soil N mineralization is important (Schepers and Meisinger, 1994), it has been difficult to implement this test in eastern Ontario (Ma and Dwyer, 1999) due to large number of samples required to account for spatial variability and the narrow window for soil sampling. Additional limitations to the PSNT in humid environments, such as eastern Canada include frequent rainfall, large spatial variability and rapid changes in soil NO3––N concentrations (temporal variation), making it difficult to use PSNT as a relevant guide for N application in grain or silage corn production. The time, labor and cost of conducting PSNT tests limits the number of field sites that can be tested (Heckman, 2002). More importantly, soil NO3– tests must be performed at critical crop growth stages, and the results must be obtained rapidly to make decisions about the amount of fertilizer application.

Recently, various nondestructive and instantaneous diagnostic techniques such as the chlorophyll meter (Minolta SPAD-502 Chlorophyll Meter, Minolta Camera Co. Ltd., Japan) have been used to predict N requirements of grain corn (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 tasselling stages (Waskom et al., 1996). However, compared to soil indicators, leaf chlorophyll content 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 the 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 eliminate the need for extensive field sampling (Gilabert et al., 1996; Ma et al., 1996, 2005). 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). Using portable commercial radiometers such as CropScan (CropsScan Inc., Rochester, MN, USA; Ma et al., 1996, 2001, 2005) or GreenSeeker (Ntech Industries Ltd., Ukiah, CA, USA), a large amount of canopy reflectance data across a field can be collected in about one-fifth the time required for the same information by chlorophyll readings (Ma et al., 2001, 2005). Scharf et al. (2002) concluded that corn color measured in aerial photographs can be used to predict sidedress N need. Nevertheless, there are some limitations with the measurements by canopy reflectance such as location-specific conditions, weather variables and soil background, which may influence equipment performance (Schröder et al., 2000). Ma et al. (2005) compared the PSNT and crop-based indicators for N management in field corn and found that all of the tests differentiated zero- or low N treatments from high N treatment, but the correlation of grain yield with any of these parameters at V6 to V8 was weak.

Most of the above observations are related to field corn, and there is a scarcity of literature in sweet corn N management, especially for the fresh market production in a cool- and short-growing season region. Because of the short growing season and harvesting at an earlier stage of crop development, the N requirement pattern for sweet corn differs from that of field corn. The N application rates for sweet corn are often in excess of crop demand. Because the crop is harvested before physiological maturity, it does not fully utilize the applied N fertilizer. This results in high levels of NO3––N left in the soil following crop harvest, which is available for leaching, leading to environmental problems and a less profitable yield. Therefore, to increase the fertilizer use efficiency in sweet corn, growers need to improve the methods for determining optimum N rate. Heckman et al. (1995) found that soil NO3––N concentration of >25 mg kg–1 were associated with relative yields ≥92%, and the success rate for the PSNT critical level was 85% for predicting whether sidedress was needed in sweet corn. They concluded that although PSNT was quite accurate in identifying N-sufficient sites, it has not been adequately tested in sweet corn to guide N fertilizer applications. Sainz Rozas et al. (2000) also found that relative yield of field corn was highly correlated with NO3––N concentration (0–30 cm) at the V6 stage. However, because of the short season production of sweet corn, the PSNT has rarely been tested. Similarly, there is limited use of the crop-based indicators for N management in sweet corn. Therefore, a field experiment was conducted to assess and compare PSNT with some crop-based measurements (canopy reflectance, leaf chlorophyll and plant total N) for N management in sweet corn production. The specific objectives of this study were to: (i) determine the appropriate sampling time and depth of PSNT for sweet corn, (ii) establish the threshold values of PSNT in a short growing region such as eastern Ontario for sweet corn, and (iii) assess the reliability of other crop-based indicators as compared to PSNT for the detection of N status in sweet corn.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
A field experiment was performed for 4 yr (2001–2004) in Ottawa (45°17' N, 75°45' W), Canada, in a nonirrigated cropping system. The soils in the experimental sites varied from silt loam to sandy loams. The soil texture, pH and nutrient concentrations are presented in Table 1.


View this table:
[in this window]
[in a new window]

  		Table 1. Soil texture, organic matter, pH and extractable N, P, and K of the soils at 0- to 30-cm depth before planting from the experimental sites for 4 yr.

 
The experiment was a randomized complete block design with four replications in each year. A FMSC hybrid Temptation was grown in 2001, 2002, and 2003, while a PSC hybrid Hollywood was grown in 2002, 2003, and 2004. In 2002 and 2003, both types of the sweet corn hybrids were planted in the same field but in separate blocks with different planting dates to avoid outcross. The five N treatments (0, 50, 100, 150, and 200 kg N ha–1) for each type of sweet corn were randomized in each block. Each year, the plot was 9 m x 6.1 m with eight rows of corn spaced 76 cm apart. Phosphorus and K fertilizers were applied according to soil test recommendations at the time of land preparation. Nitrogen fertilizer (ammonium nitrate, 33.5% N) treatments were broadcast and incorporated into the seedbed before planting. The experiment was planted at a population density of 66 000 plants ha–1 from 17 May to 6 June; PSC was planted about 14 d later than the FMSC in 2002 and 4 d later in 2003. The broad-spectrum herbicide Primextra II Magnum [s-metolachlor/benoxacor/atrazine (2-chloro-4-ethylamino-6-isopropylamino-1,3,5-triazine)], at a rate of 1.75 L ha–1 was applied before planting for weed control. Postemergence hand-hoeing was performed to remove any weeds that escaped the herbicides.

Measurements
Preplant soil samples (0–30 cm) were taken to determine soil pH (in water), mineral N (NO3– and NH4+), soil test P (Bray 1) and K (NH4OAc method) (Table 1). In addition, soil samples were also taken for PSNT at V4, V5, V6, and V7 or V8 stages from 0- to 30-cm depth. While sampling at V6 growth stage, additional soil samples were taken from a depth of 30 to 60 cm. At each sampling, soils taken in two cores per plot were mixed thoroughly, and representative fresh subsamples were taken for analysis. Mineral N concentrations including nitrate (NO3–) and ammonium (NH4+) extracted from the fresh soil with 2 M KCl were analyzed colorimetrically (Keeney and Nelson, 1982) on a TRAACS 800 Auto-Analyzer (Bran-Luebbe, Analyzing Technologies, Elmsford, NY). The NO3––N concentrations from 0- to 60-cm depth was calculated as the weighted means of soil NO3––N concentrations from 0- to 30- and 30- to 60-cm depths.

At V6, five representative plants from the center of Row 3 of each plot were sampled for the determination of plant N concentration. The samples were oven dried at 80°C, and then ground to pass by 1-mm screen. Subsamples (1 g each) were analyzed for total N using a micro-Kjeldhal method. For FMSC at V6 and V8 growth stages in 2001 and 2002, leaf greenness was measured on the topmost fully expanded leaves of 30 plants from the two middle rows of each plot by using a chlorophyll meter (SPAD–502). Canopy reflectance measurements were also simultaneously taken at V6 and V8 growth stages with a hand-held multispectral radiometer (MSR16, CropScan, Rochester, MN), which records percentage light reflectance in 11 bands (460, 507, 559, 613, 661, 706, 769, 813, 850, 900, and 950 nm), approximately at 50 nm intervals. The canopy reflectance measurement at the V6 stage was taken between 10:00 and 14:00 h, in clear sky conditions. The sensor readings were used to derive a normalized difference vegetation index (NDVI) as NDVI = (IR813 – R613)/(IR813 + R613) (where R = red, IR = infrared; Ma et al., 2001).

The time intervals required for initiation of various leaves and to silking and harvest were recorded. Sweet corn ears were harvested at approximately 4 wk after silking, which was very close to the maturity date determined using the criteria of the silks turning dark brown. Thirty plants from the two central rows of each plot were harvested to measure fresh marketable yield and quality related parameters (sweetness, ear length and diameter). Total fresh weight of the 30 plants was recorded and the total number of ears on the 30 plants counted and recorded. A subsample of five plants was taken from the 30 plants and their fresh weight was recorded. Ears were separated from the remaining 25 plants and dehusked. For the FMSC, each ear was assessed for their marketability based on its length and stage of kernel development (i.e., ≥15 cm with filled kernels). The total number of marketable ears was determined based on the sample of 30 plants. Fresh and dry weight (DW) of the marketable ears were then determined. Based on the fresh and DW of the 30 plants, the total biomass yields were estimated on a per hectare basis. The length of the ears from the subsample plants was measured. The ear was then broken into two halves at the middle and their diameter was measured as the distance between the kernel tips. To determine kernel sweetness (i.e., % sugar content), ears were squeezed and the juice was tested using the Atago PR-32 Digital Refractometer (Ancansco Inc., Toronto, ON, Canada). For the PSC, the harvesting of 30 plants and subsampling of five representative plants was similar to that of FMSC. However, the number of marketable ears, lengths and diameters of the ears from the five plant samples were not determined. After taking the fresh weight of the ears from the five plants, all kernels were removed from the cob with a knife. Similarly, the sugar content of the kernel at harvest was also determined.

Data Analysis
Experimental data were subjected to analysis of variance (ANOVA) using the general linear models of SAS (SAS Institute, 1996). The ANOVA was performed to evaluate the effect of N treatments on the response variables measured. Treatment means were separated by the ANOVA-protected least significant difference (LSD0.05) test. Partial correlations were also analyzed between various parameters.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSION
 REFERENCES
 
Growing Conditions
Weather conditions varied considerably over the four growing seasons (Table 2). A prolonged drought occurred from June through August in 2001, resulting in <50% of 45-yr average rainfall. Wet and cool growing conditions prevailed in 2002 during the emergence and seedling establishment stage. A heavy rainfall (55 mm within 7 h) during the seedling stage (June 27) may partially explain the low recovery of soil NO3––N in 2002 as discussed below. There was also a period of brief drought bracketing the silking and early grain filling period. The whole growing season in 2003 and 2004 received evenly distributed precipitation, and the crop did not experience moisture stress. Mean temperatures during the growing period (from planting to harvest) were consistently >15°C in all years.


View this table:
[in this window]
[in a new window]

  		Table 2. Monthly total rainfall and average temperature during the growing season from 2001 to 2004 and the recent 40-yr average in Ottawa, Canada.

 
Phenology
In all years, it required 58 to 71 d from planting to reach 50% silking, and 85 to 100 d from planting to reach harvest for fresh market corn. A similar duration was required by the processing sweet corn (PSC, harvest for kernels instead of the whole ears); however, it might have accumulated slightly more heat units as this corn was planted 4 to 14 d later after the FMSC had been planted. Thus, the PSC was grown under relatively warmer days than the FMSC.

Response of Marketable Ears to Nitrogen Treatments
Ear length was not affected by N treatment in each year. The mean ear lengths were 17.5, 18.7 and 14.4 cm in 2001, 2002, and 2003, respectively. Ear diameter was not affected by N treatment in 2001 (3.9 cm) and 2003 (4.8 cm). However, in 2002, ear diameter was affected by N rate, because it was 4.9 cm for N rates of 100 and 200 kg ha–1 and 4.7 cm for 0 and 50 kg N ha–1. The kernel sugar content was not affected by N treatment in any year. The mean sugar contents were 16.7% (Brix °) in 2001, 25.4% in 2002, and 25.8% in 2003.

The crop in 2001 produced fewer total and marketable ears than in 2002 and 2003, which is attributed to the prolonged drought. Nitrogen treatment in 2001 had no effect on the total number of ears, but the number of marketable ears was highly responsive to N rates (Fig. 1 ). Nitrogen fertilization at 100 to 200 kg ha–1 produced similar numbers of marketable ears, and was significantly greater than the other two N treatments. The unfertilized treatment had the lowest number of marketable ears. A similar response to N treatment was observed for total and ear dry matter (DM) such that total DM was not affected by N rates, but there was a significant treatment effect on ear DM.


Figure 1
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 1. Effect of N treatments on the number of total and marketable ears, total and ear dry matter (DM) of fresh-market sweet corn (Temptation) in 2001, 2002, and 2003. The points followed by different letters indicates significant differences (P ≤ 0.05) among N rates.

 
In 2002, the total number of ears for the 150 and 200 kg N ha–1 treatments was significantly greater than for the other N treatments (Fig. 1). A similar response was observed for the number of marketable ears: fertilizer at 0 and 50 kg N ha–1 treatments had smaller number of marketable ears than the other treatments. Response of total and ear DM to N treatments was similar to those of marketable ears. Larger response of total DM and marketable ears in 2002 than that of 2001 or 2003 was probably due to the fact that in 2002, there was little N released from soil N mineralization because of cool weather conditions during the early growth (Table 2).

The number of ears and total DM production was very high, but the response to N treatment was smaller in 2003 as compared to those in 2002. All the treatments (receiving fertilizer N from 50 to 200 kg N ha–1), except the 0 N control treatments, produced similar number of ears, total DM and DM of the marketable ears, and the 0 N treatment resulted in smaller ear numbers and less DM than the other N treatments.

Pre-Sidedress Nitrate Test and Marketable Ears
The PSNT analyzed from V4 to V6 growth stages for the FMSC showed that the NO3––N concentrations varied greatly in different years, and N treatments had a large effect. Analysis of variance indicated that N treatment had significant effects on PSNT in all years, but the NO3––N concentrations were higher and N treatment differences were greater in 2001 and 2003 compared with those of 2002 (Fig. 2 ). In 2001, there was a linear increase in NO3––N concentrations with increasing amounts of N applied, although the increase for the 0- to 60-cm depth was much smaller than for 0- to 30-cm depth at V6 growth stage. Even though a statistically significant difference in PSNT levels was observed in 2002, the overall NO3––N concentrations were small and differences among the N treatments were small. Still, the pattern was representative of the amount of N applied, such that the high N treatment (200 kg N ha–1) always had the highest NO3––N concentration. The small values of NO3––N measured in 2002 were probably associated with the record 1-d rainfall (55 mm within 7 h) in late June before the first sampling, which may have led to more NO3––N loss via leaching, runoff and/or denitrification. Cool temperature coupled with excess rainfall may have also inhibited soil N release from mineralization. In 2003, NO3––N concentrations at 0- to 30-cm depth increased with increasing N rate, while the results followed the similar trend as previous years and increment for the 0- to 60-cm depths was smaller than the 0- to 30-cm samples at V6 (Fig. 2). Within a year, NO3––N concentrations at the 0- to 30-cm depth varied from V4 to V6 growth stages, but the trend remained the same: the increased NO3––N concentrations followed the increase in fertilizer N rates.


Figure 2
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 2. Presidedress nitrate N (mg NO3––N kg–1) in soil at various growth stages of FMSC (Temptation) as affected by fertilizer N treatments in: (a) 2001, (b) 2002, and (c) 2003. Regression equations for V6 growth stage at the 0 to 30 cm (top left within each year) and 0 to 60 cm (top or bottom right within each year) depths are inserted.

 
There was a positive correlation between PSNT (0–30 cm) and the number of marketable ears of FMSC, at various growth stages, especially at V6 (Table 3). There was fluctuation in NO3––N concentrations from V4 to V8 stages, but they were strongly correlated each other (data not shown). The correlation of NO3––N concentration from 0- to 60-cm depth at V6 with the number of marketable ears was generally weaker than the PSNT (0–30 cm depth), indicating that there was a trade-off between NO3––N concentrations of the top layers where fertilizer was applied and where N released from soil microbial activity took place and the deep layer. An alternative explanation could be that the majority of NO3––N through fertilization and soil mineralization has not leached to the deep layers by the presidedress stage. The concentration of NH4+–N was not significantly affected by the N treatment in any of the years and growth stages it was measured. As reported earlier (Ma and Dwyer, 1999), NH4+–N accounts for only a small portion of soil mineral N at sidedress time in this region.


View this table:
[in this window]
[in a new window]

  		Table 3. Correlation coefficients (r) between different parameters measured with the number of marketable ears in a FMSC (Temptation) and with the kernel dry weight in a PSC (Hollywood) in 3 yr (n = 20).

 
Tissue Nitrogen Concentration and Marketable Ears
The tissue N concentrations at V6 varied considerably each year (Fig. 3a ), with N treatment having significant effects on plant N concentration at V6. Nitrogen treatment at 100 kg N ha–1 or higher resulted in higher plant N concentrations than those of 0 or 50 kg N ha–1 treatment in 2001 and 2003. While in 2002, there was no difference in plant N concentrations from 0 to 150 kg N ha–1 treatment, the highest N treatment (200 kg N ha–1) had higher plant N concentrations than any of the other N treatments. Plant N concentration at V6 was strongly correlated with the number of marketable ears, especially in 2001 (Table 3).


Figure 3
View larger version (39K):
[in this window]
[in a new window]

  		Fig. 3. Plant total N concentration (g kg–1) of (a) a FMSC (Temptation) and (b) a processing sweet corn (PSC) (Hollywood) at the V6 growth stage in 2001, 2002, and 2003. The bars followed by different letters indicate significant differences (P ≤ 0.05) within a year.

 
Crop-Based Indicators and Marketable Ears
Analysis of variances showed that N treatments significantly affected (P ≤ 0.05) SPAD-determined leaf chlorophyll content and canopy reflectance, expressed as NDVI. At the V6 to V8 growth stages, plants receiving 0 or 50 kg N ha–1 always had lower SPAD and NDVI values than those receiving higher fertilizer N rates. The SPAD readings differentiated between N treatments, and increased linearly with N treatments up to 100 or 150 kg N ha–1 at the V8 stage in both years (Fig. 4 ). The smaller R2 values at V6 stage may indicate the quantitative SPAD-N rate relationship has not been fully established at this stage, and it also highlights the need for further research. The SPAD readings at both V6 and V8 growth stages were positively correlated with the number of marketable ears in 2001 and 2002 (Table 3). Similarly, there was a significant positive correlation between fertilizer N rates and NDVI values in 2002 (Fig. 4), although a power function fit the data better than the linear fitting. In both cases, the relationship was stronger at V8 than at V6. The nonlinearity between NDVI and N rates above 100 kg N ha–1 may indicate NDVI saturations (e.g., Gitelson, 2004) at the high N rates, which is the current research topic in this area. The NDVI measured at V8 was also strongly associated with the number of marketable ears in 2002 (Table 3). These results indicate that the different crop-based measurements were equal in measuring N treatments at V6 to V8 and similar to PSNT and tissue N test in predicting yield responses to N. The correlations between PSNT at V4 to V6 and SPAD readings at V6 or V8 were very strong (r = 0.81–0.99). Similarly, plant N concentration at V6 was also strongly associated with the SPAD readings at V6 (r = 0.74) and V8 (r = 0.97). However, at V6 stage, NDVI was weakly associated (r = 0.35) with the tissue N concentration (data not shown).


Figure 4
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 4. Relationships of leaf chlorophyll content (SPAD) and canopy reflectance (NDVI) measured at V6 and V8 stages of fresh-market sweet corn (Temptation) with different fertilizer N rates applied at planting in (a) 2001, (b) 2002, and (c) 2002.

 
Response of Kernel Yield of Processing Sweet Corn to Nitrogen
At harvest, the whole plant moisture content of the PSC was 82 to 84%, and kernel moisture content ranged from 77 to 80%. Nitrogen treatments had no effect on kernel moisture at harvest, but whole plant moisture was significantly lower in 0 N, and the highest in the 200 kg N ha–1 treatments. Kernel sugar content was not influenced by N treatments, which was similar to that for FMSC.

Both total DM and kernel DM of the PSC were much greater in 2003 than in the other 2 yr (Fig. 5 ). In 2002, kernel DM and total DM were significantly greater with 150 and 200 kg N ha–1 than the 0 and 50 kg N ha–1 treatments. However, kernel yield for the 50 to 200 kg N ha–1 treatments was not significantly different. In 2003, kernel DM was not different among the N treatments, but total DM was significantly affected by N treatment. The total DM was significantly lower at 150 kg N ha–1 than at 200 kg N ha–1, but the rest of the treatments had a similar DM. Unlike in the previous 2 yr, total DM was not affected by N treatment in 2004, but there was a significant N effect on kernel DM. Nitrogen treatment at 100 kg ha–1 had higher kernel DM than that of 0 N treatments, but this was not different from the other N rates.


Figure 5
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 5. Effect of N treatments on the total dry matter (DM) production and kernel DM yield of a PSC (Hollywood) in (a) 2002, (b) 2003, and (c) 2004. The points followed by different letters indicate significant differences (P ≤ 0.05) among N rates.

 
Analysis of variance showed that N treatment and sampling depth had significant (P ≤ 0.05) effects on the PSNT. Similar to those for FMSC, concentrations of NO3––N for PSC experiment were smaller in 2002 than in 2003 and 2004 (Fig. 6 ). As a result, the numerical difference of PSNT values among the N treatments was also smaller. In general, of all the presidedress stages, there was a linear increase in NO3––N concentration with the amount of N application until V6 growth stage although in several occasions (e.g., for both 0–30- and 0–60-cm depths in 2003 and for 0–30-cm depths in 2004), a quadratic fitting was statistically better than the linear function, indicating a plateau of soil NO3––N at 100 kg ha–1 or higher N rates. In both 2003 and 2004, the NO3––N concentration in the 30- to 60-cm depth at V6 was much lower than in the 0- to 30-cm depth. As a result, NO3––N for the 0- to 60-cm depths was also low. This indicates that there was not as much leaching of the applied N by V6 growth stage. Similar to the FMSC, the NH4+–N analyzed during the presidedress stage (V4–V8) was not affected by the N treatment in the PSC.


Figure 6
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 6. Presidedress nitrate N (mg NO3––N kg–1) in soil at various growth stages of PSC (Hollywood) as affected by fertilizer N treatments in (a) 2002, (b) 2003, and (c) 2004. Regression equations for V6 growth stage at the 0 to 30 cm (top left within each year) and 0 to 60 cm (top or bottom right within each year) depths are inserted.

 
Correlation coefficients between PSNT and kernel DM of the PSC were smaller than that of the number of marketable ears in FMSC (Table 3). The relationships were stronger in 2002 than in 2003 and 2004. Kernel yield was positively correlated with the plant N concentration at V6 in 2003 (Table 3). Nitrogen application at 100 kg ha–1 or higher resulted in greater plant total N, but significant difference in plant total N at V6 was observed only between the highest N and other N rates in 2002 (Fig. 3b).

Comparison of Methods for Assessing Crop Nitrogen Status
Although there were year-to-year variations in marketable yields and N responses due to differences in soil types and weather conditions, marketable ears of the FMSC and kernel yield of PSC were responsive to N treatments. In all cases, differences in yields between 100 and up to 200 kg N ha–1 treatments were small to nonexistent, which is common for grain corn production in this environment (Ma et al., 2005; Subedi and Ma, 2005). Across years and corn types, it appears that reasonable optimum N rate for this region is about 100 kg N ha–1 or slightly less. The nonsignificant treatment differences for NO3––N in the soils at 30- to 60-cm depth in all years except for 2002 indicated that the excess applied N was not lost to the deeper soil profile by the time of V6 growth stage. These results are in agreement with those reported by Magdoff (1991), who established that the NO3––N that accumulates early in the growing season is usually still present in the root zone during peak demand period. The pattern of PSNT between 0- and 30- and 0- to 60-cm depths has clearly indicated that the plant available NO3––N is still in the soil at the V6 growth stage. However, it was interesting that the amount of available NO3––N at presidedress stage was not necessarily correlated with greater yields of sweet corn at harvest. For example, in 2001, NO3––N concentrations at all presidedress stages were much higher than in other years (Fig. 2), but the total biomass yield and marketable ears were much smaller than in other years. The crop may have suffered from severe drought stress due to shortage of precipitation in 2001 (Table 2). As a result, sufficient N application did not result in the higher yield potential. The NO3––N concentrations in all presidedress stages of 2002 were lower than the critical values of 20 to 30 mg kg–1 as defined by Magdoff et al. (1984) for grain corn production in Vermont, but marketable ears and total biomass yield were much greater than in 2001 (Fig. 1). The excess rainfall before initial soil sampling and relatively higher NO3––N concentrations in the 30- to 60-cm depth of 2002 may indicate that NO3––N has leached out of the root zone at V6. Another evidence of NO3– leaching was that R2 values of the relationship between marketable ears and PSNT was greater for 0- to 60- than 0- to 30-cm depth. However, the leached N may still have been available to corn plants at late stages because by flowering, corn roots reach over 100-cm soil depth (Dwyer et al., 1996). As a result, a linear response of total DM or ear DM up to 150 kg N ha–1 was established. The PSNT value for 0 to 30 cm at V6 was near the critical level of 20 mg kg–1, a value that agrees with that reported by Magdoff et al. (1984). Although the PSNT provides a guide of N status of soil at presidedress stage, final yield is influenced more by environmental factors such as drought. The weaker correlation of PSNT against kernel yield in 2003 could be the reason that much greater amount of N was released from soil organic N mineralization than 2001 and 2002, which may shift the balance toward excess N, suggesting that PSNT might work best under less than ideal growing conditions. However, further study is needed on different soil types and following different preceding crops to test this hypothesis.

In general, PSNT worked well in all years, but the threshold values could have changed according to yield potential realized in each year, which is influenced primarily by rainfall and other environmental conditions (Calvino et al., 2003). The soil sampling at a depth of 30 to 60 cm did not correlate with crop N status, but PSNT (0–30-cm depth) from V4 to V6 were equally important predictors. This supports the statement by Meisinger et al. (1992) that soil NO3––N at V6 stage represents the net balance between production (mineralization from soil organic matter, manure and fertilizer) and loss (leaching, denitrification and immobilization) because little or no N uptake occurs before that stage.

For the reasons explained above, although PSNT works in most environments, it is difficult to determine a threshold value of PSNT for sweet corn in the humid environment of eastern Ontario. The PSNT (NO3––N concentration at the 0- to 30-cm depth when corn is 15 to 30 cm tall) is suggested for use in a number of U.S. states (Randall et al., 1999) and Ontario of Canada. In this study, an approximate concentration above 20 mg kg–1 NO3––N in 0- to 30-cm depth or 12 to 15 mg kg–1 NO3––N in the 0- to 60-cm depth at the V6 stage of sweet corn produced similar number of marketable ears and total biomass yield in 2001, 2003, and 2004. However, in 2002, despite lower values of NO3––N at all presidedress stages, the number of marketable ears and total biomass yields had responded significantly to N treatment. This indicates that PSNT may not always be a true predictor of final yield. Villar-Mir et al. (2002) monitored several commercial cornfields for NO3––N levels, crop N uptake and crop productivity and concluded that soil PSNT + N fertilizer was neither related to plant N uptake nor a final plant biomass and grain yield predictor. They also observed an excess of N in the system. Perhaps, temporal variations of N release in the soil profile (Ma et al., 1999) may have caused the negating of the validity of the PSNT + N relationship to plant N uptake. As Mamo et al. (2003) have suggested, that variations in soil NO3––N levels at different sampling dates must also be considered with site-specific N management, especially in humid conditions. In addition, integration of other environmental variables such as soil moisture status into the prediction equations may increase the predictive capacity of PSNT or crop-based indicators. Such study is underway on grain corn N management in eastern Canada.

As previously demonstrated in field corn (Ma et al., 2005), crop-based indicators (i.e., SPAD and NDVI readings) have indicated N status of plant tissue and differentiated N treatment effects in sweet corn. Both marketable ears and total biomass yield were correlated with SPAD and NDVI values measured at presidedress stage. Because such measurements are quicker (on-the-go) and require less labor (get results on site without laboratory analysis), these crop-based measurements can be used as alternatives in predicting the N requirement of sweet corn for fresh market production. However, the prediction of N rates by these tools is not straightforward, which requires further verifications and extensive studies on the algorithms of crop-based measurements and N needs suitable for FMSC and FSC in the cool and short growing season regions.


   CONCLUSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
REFERENCES
 
This study evaluated the four methods of assessing crop N status at presidedress stage (i.e., PSNT, tissue N concentration, SPAD and NDVI). The outcomes of each method were very similar and corroborated each other, such that they can be used to differentiate soil and plant N status during the initial developmental stage. However, there were instances that soil or plant N status at V6 stage was not always reflected in final yield, suggesting that environmental conditions after V6 played a central role in the overall yield formation of sweet corn. The crop-based indicators can be equally effective to PSNT or plant tissue N concentration, and have the potential in this short and cool growing region to be used as a quick guide for the N status of sweet corn at presidedress stage. We believe that enough evidence exists to justify further investigation into the use of these crop based indicators.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge the excellent technical assistance of L. Evenson, D. Balchin and V. Deslauriers for field and laboratory technical assistance. ECORC contribution no. 05-588.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ma, B. L.
Right arrow Articles by Zhang, T. Q.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Ma, B. L.
Right arrow Articles by Zhang, T. Q.
Agricola
Right arrow Articles by Ma, B. L.
Right arrow Articles by Zhang, T. Q.
Related Collections
Right arrow Best Management Practices
Right arrow Nitrogen
Right arrow Maize

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