Mid-Season Recovery to Nitrogen Stress in Winter Wheat

K.B. Morris, K.L. Martin, K.W. Freeman, R.K. Teal, K. Girma, D.B. Arnall, P.J. Hodgen, J. Mosali, W.R. Raun*, and J.B. Solie

Department of Plant and Soil Sciences and Department of Biosystems and Agricultural Engineering, Oklahoma State University, Stillwater, OK 74078. Contribution from the Oklahoma Agricultural Experiment Station. *Corresponding author, wrr@mail.pss.okstate.edu.

 ABSTRACT

Although spring applied nitrogen (N) has been shown to be more efficient, delaying all N applications until mid-season and the resultant effect on maximum yields has not been thoroughly evaluated.  This experiment was conducted to determine if potential yield reductions from early season N stress can be corrected using in-season N applications. Data from three experimental sites and two growing seasons (6 site-year combinations) were used to evaluate 3 preplant N rates (0, 45, and 90 kg ha^-1) and a range of in-season topdress N rates.  Topdress N amounts were determined using a GreenSeeker^TM hand held sensor and an algorithm developed at Oklahoma State University.  Even when early season N stress was present (0-N preplant), N applied topdress at Feekes 5 resulted in maximum or near maximum yields at 4 of 6 site-year combinations when compared to other treatments receiving both preplant and topdress N.

INTRODUCTION

As environmental and economical issues become a concern, it is important for action to be taken to address these important issues.  Fertilizer is one of the major controversial environmental issues in today’s world.  Vidal et al. (1) stated that the application of N at rates exceeding plant utilization represents an unnecessary input cost for wheat producers and can harm aquatic and terrestrial environments. 

Current methods of determining N fertilization rates in cereal production systems are determined by subtracting soil test N from a specified yield goal-based N requirement.  The yield goal represents the best achievable yield in the last 4 to 5 years (2, 3).  There are, however, more precise and efficient ways of obtaining fertilizer recommendations to maximize yield and minimize cost.  Following extensive soil sampling, optical sensor measurements of plants, and geostatistical analysis, several authors reported that the spatial scale of N availability was at 1m² and that each square meter needed to be treated independently (4, 5, 6).  When N management decisions are made on areas of 1 m², the variability that is present at that resolution can be detected using optical sensors (measuring normalized difference vegetative index, NDVI) and treated accordingly with foliar application of N (7, 8, 9), which increases nitrogen use efficiency (NUE) (8).   Recently, methods for estimating winter wheat N requirements based on early-season estimates of N uptake and potential yield were developed (6, 10).  Remote sensing collected by a modified daytime-lighting reflectance-sensor was used to estimate early-season plant N uptake.  The estimate was based on a relationship between NDVI and plant N uptake between Feekes physiological growth stage 4 (leaf sheaths lengthen) and 6 (first node of stem visible) (7, 8, 11).  NDVI was calculated using the following equation:  NDVI = (ρ_NIR – ρ_Red)/(ρ_NIR + ρ_Red)

ρ_NIR = Fraction of emitted NIR radiation returned from the sensed area  (reflectance)

ρ_Red = Fraction of emitted Red radiation returned from the sensed area (reflectance) 

Increasing NUE by just 20% would result in savings exceeding $4.7 billion per year (9).  Improving NUE will decrease the risk of NO₃-N contamination of inland surface and ground water (8, 9), as well as the hypoxia in specific oceanic zones, which are believed to be caused by excess N fertilizer (9, 12).

Raun et al. (6) stated that measuring the quantitative response to fertilizer N is achievable for a given area. This is why the N fertilization optimization algorithm (NFOA) was developed.  It determines the prescribed N rate needed for each 1 m² based on predicted yield potential without added N fertilizer (YP₀) and the specific response index (RI) for each field.  Johnson and Raun (13) defined RI as the amount of yield response to expect from an application of fertilizer N compared to yield with no additional N, and that may range from 1 to as high as 4.  Raun et al. (6) explained that the NFOA accounts for spatially variable potential yield, early season N uptake, and responsiveness of the crop to N input.  The algorithm calculations are as follows:  

1.)    Predict YP₀ from the equation for grain yield and in-season estimate of yield (INSEY), where:  INSEY = NDVI (Feekes 4-6)/ days from planting where growing degree days (GDD) > 0 [GDD= (T_min + T_max)/2 – 4.4^o C, where T_min and T_max represent daily ambient high and low temperatures].

Lukina et al. (10), showed that a single equation could be used to predict grain yield over a wide production range (0.5-6.0 Mg ha^-1), diverse sites, and with differing planting and harvest dates.  Dividing NDVI at Feekes 5 (excellent predictor of early-season plant N uptake) by the days from planting to the NDVI sensing date resulted in an index that would approximate N uptake per day.

2.)    Predict the magnitude of response to N fertilization, in-season RI (RI_NDVI), computed as: NDVI collected from growing winter wheat anytime from Feekes 4 to Feekes 6 in non-limiting fertilized plots divided by NDVI in a parallel strip receiving the farmer preplant N rate. 

The RI_NDVI has been found to be highly correlated with the RI at harvest (RI_Harvest), which is similarly computed by dividing the highest mean grain yield of the N rich treatment from the mean grain yield of 0-N treatment (check plot) (14).  The farmer preplant N rate could range anywhere from zero to a rate for non-N limiting conditions. (6).

3.)    Determine the predicted yield with additional N (YP_N) based both on RI_NDVI and the YP₀ as follows:  YP_N = YP₀ * RI_NDVI

The RI_NDVI was limited so as not to exceed 3.0 and YP_N was similarly limited not to exceed the maximum obtainable yield (YP_max).  The YP_max was determined by the farmer, or by measuring the maximum NDVI in the N rich strip (N applied at adequate but not excessive rates preplant) (W. Raun, J. Solie, personal communication, July 2004, and reported on http://nue.okstate.edu) and using that value to calculate the YP_max  using the yield potential equation.  The YP_max can also be defined as a biological maximum for a specific cereal crop grown within a specific region and under defined management practices (e.g., YP_max for dry land winter wheat produced in central Oklahoma would be 7.0 Mg ha^-1).  The RI_NDVI was capped at 3.0 as in-season applications of N would unlikely lead to YP_N being more than three times greater than baseline YP₀.

4.)  Calculate predicted grain N uptake (PNG) at YP_N(GNUP_YPN), average percent N in the grain multiplied by YP_N:  GNUP_YPN = YP_N * PNG

5.)  Calculate PNG at YP₀, average percent N in the grain multiplied by YP₀:  GNUP_YP0= YP₀ * PNG.

6.)  Determine in-season fertilizer N requirement (FNR):

FNR = (GNUP_YPN – GNUP_YP0)/0.60

A divisor of 0.60 in the above equation is used because the theoretical maximum NUE of an in-season N application is approximately 60%.

The use of active growing days from planting and NDVI (estimate of total N uptake and or biomass) in computing INSEY allows integration of the effects of both winter and spring growing conditions and date of planting. The INSEY index is essentially the rate of N uptake (kilograms of forage N assimilated per day) by the plant. This approach is consistent with work showing the relationship between above ground plant dry weight and cumulative GDD (15).   Further analyses showed that a reliable INSEY could be obtained by dividing NDVI by the days from planting to sensing date (where GDD > 0) (6, 16).  Mullen et al. (16) also stated that the INSEY was used to estimate N uptake in the grain based on a predicted yield level.  Finally, using predicted wheat N uptake (measured by NDVI) at Feekes 5 (excellent predictor of early-season plant N uptake) and projected grain N uptake from INSEY, topdress fertilizer N rates have been determined (grain N uptake minus early season plant N uptake) (10).

Johnson et al. (17) defined the harvest response index (RI_Harvest):

RI_Harvest = (highest mean yield N-treatment)/(mean yield check treatment).

The use of RI_Harvest does not allow for in-season adjustment of N.  In-season sensor measurements of NDVI as an indicator of wheat N uptake between plots receiving N and those not receiving N can be used in the same way using the following equation:

RI_NDVI = (highest mean NDVI N treatment)/(mean NDVI check treatment).  

Mullen et al. (16) concluded that basing fertilizer N rates on INSEY and RI_NDVI may help optimize in-season fertilizer application, which in turn could increase NUE and yield.  The objective of this work was to determine if RI_NDVI could accurately predict RI_Harvest at Feekes growth stages 5, 9, 10.5, and 11.2.  They also found that RI_NDVI measured at Feekes 5 was highly correlated to RI_Harvest.  Mullen et al. (16) recognized that after remote sensing data is collected, yield enhancing and limiting factors may occur that result in underestimation or overestimation of RI_Harvest by RI_NDVI.  For example, in 1999, early spring rains after a dry fall planting period improved post sensing growing conditions.  Timely rainfall may have increased the N response resulting in a larger RI_Harvest than predicted by RI_NDVI.  The objectives of this work were to determine if potential yield reductions from early stress can be corrected by using in-season fertilizer applications, and to evaluate the relationship between RI_NDVI and RI_Harvest over years and locations.

MATERIALS AND METHODS

Three experimental sites were selected for this study.  The Covington site was located at a cooperating farmer field.  The soil at this site was Kirkland- Renfrow silt loam, fine, mixed, superactive, thermic Udertic Paleustolls.  The Stillwater Research Station Lake Carl Blackwell, Oklahoma had Port-oscar silt loam, fine-silty, mixed, super active, thermic Cumulic Haplustolls soil.  The Tipton site had Tillman-Hollister silt loam; fine-loamy, mixed, thermic, Pachic urgiusoll soil. 

A randomized complete block design was employed with 15 treatments and 4 replications.  Plot size was 3.05 m x 6.1 m with 3.05 m alleys.  Three preplant N rates (0, 45, and 90 kg ha^-1) were applied to plots as ammonium nitrate (34-0-0).  Topdress N application rates were determined utilizing the NFOA (6) with four different RI values.  The RI values evaluated were 1.0, 1.3, 1.6, and 2.0.  Algorithms differed for 2003 and 2004 whereby the coefficient of variation (CV) was used in 2004 to alter yield potential achievable with N fertilization (18).   Response index was calculated as reported by Johnson and Raun (13).  Spectral reflectance was measured using a GreenSeeker^TM Hand Held Optical Sensor (N-tech Industries) that collected NDVI measurements.  This device uses a patented technique to measure crop reflectance and calculate NDVI.  The unit senses a 0.6 x 0.01 m spot when held at a distance of approximately 0.6 to 1.0 m from the illuminated surface.  The sensed dimensions remain approximately constant over the height range of the sensor.  The sensor unit has self-contained illumination in both the red (650 + 10 nm FWHM) and NIR (770 + 15 nm FWHM) bands (FWHM = full width at half maximum).  The device measures the fraction of the emitted light in the sensed area that is returned to the sensor; the fractions are used within the sensor to compute NDVI.

The sensor unit is designed to be “hand-held” and measurements are taken as the sensor is passed over the crop surface.  The sensor samples at a very high rate (approximately 1000 measurements per second), and averages measurements between outputs.  The sensor outputs NDVI at a rate of 10 readings per second.  Reflectance readings were collected throughout the growing season.  The NDVI readings taken for the topdress N fertilization application from all experiments were collected post-dormancy.  The date when readings were collected generally corresponded to Feekes growth stage 5 (pseudo-stem, formed by sheaths of leaves strongly erect) (11).  Topdress N was foliar applied to the whole plot using urea ammonium nitrate (UAN, 28-0-0) with a Solo backpack sprayer (amounts were calculated and then measured with a graduated cylinder).  For the smaller rates, a pulse modulated sprayer designed by OSU was used.

Winter wheat grain was harvested using a self-propelled Massey-Ferguson 8XP combine.  An area of 2.0 x 6.1 meters was harvested from the middle of each plot, and a Harvest Master yield-monitoring computer installed on the combine recorded yield data.  A sub-sample of grain was taken and dried in a forced-air oven at 66^oC, ground to pass a 100 µm screen, and analyzed for total N content using a Carlo-Erba NA-1500 Dry Combustion analyzer (19).  Statistical analysis was preformed using SAS (20).  Treatment structure for 2002-2004 is reported in Table 1.  Initial soil samples, chemical characteristics and classification of soils are reported in Table 2.  Field activities and dates are listed in Table 3. 

RESULTS

Covington, 2003

At Covington in 2003 where no topdress N was applied, there was a linear increase in wheat grain yield with increasing preplant N (treatment 1 = 3170 kg ha^-1, treatment 6 = 4527 kg ha^-1, treatment 11 = 5234 kg ha^-1, Table 4).  At this site, there was also an increase in wheat grain yield for topdress N rates whether or not preplant N had been applied.  However, the yield increases from topdress N diminished with increasing preplant N.  Maximum yields were not achieved at this site from mid-season topdress N applications in plots receiving no preplant N when compared to the plot that achieved maximum yield, which was not the N- rich plot (treatment 11).  The plot that achieved maximum yield was treatment 10.  It should be noted that even with early N stress, topdress N rates (treatment 5 = 5271 kg ha^-1) did produce an equal yield to the preplant non N limiting plot (treatment 11 = 5234 kg ha^-1), but that was still less than maximum yield (treatment 10  = 5875 kg ha^-1).  The “catch-up” effect being evaluated in this work states the following: Can maximum yields be produced when no N is applied preplant and N applications are delayed until February or March?  At this site, it was not possible to “catch-up” where no N was applied preplant plus a mid-season topdress N application (treatment 10 = 5875 kg ha^-1, Table 4).

Recent work has shown that when CVs are < 18, “catch-up” is possible (catch-up: waiting to apply all nitrogen topdress and still achieving maximum yields) (18).  Consistent with this work, CVs were all > 18 at this site indicating that “catch up” was not going to be possible and that was confirmed.

Nitrogen use efficiency was the greatest for the 0 N preplant treatments plus mid-season applied N (treatments, 1-5), but it should be noted that NUE’s were generally quite high at this site (Table 4). 

The RI estimated using in season NDVI readings was under estimated at this site (RI_NDVI = 1.27 and RI_Harvest = 1.7, Table 4).  It is possible that the N rich treatment (11) may not have received enough preplant N to accurately estimate RI_NDVI.  RI_NDVI over time for 2003 did not change much from Feekes 3 to Feekes 5 (RI_NDVI=1.18, 1.24, and 1.27, Table 3).  The RI_Harvest was much higher than the RI_NDVI values (RI_Harvest= 1.7, Table 4).

Lake Carl Blackwell, 2003

At Lake Carl Blackwell in 2003, there was a linear increase in grain yield for N applied preplant (treatment 1 = 3207 kg ha^-1, treatment 6 = 3579 kg ha^-1, treatment 11 = 4276 kg ha^-1, Table 5).  There was an increase in grain yield for topdress N rates for the 45 kg ha^-1 preplant rates, but no increase from topdress N where 90 kg ha^-1 was applied preplant.  Maximum yields were achieved at this site from mid-season topdress N applications in plots receiving 0 preplant N in comparison to the maximum yielding plots (treatment 4 = 4453 kg ha^-1, and treatment 5 = 4453 kg ha^-1, Table 5).  At this site, maximum yields were achievable with no preplant N plus a topdress rate for the maximum yielding plot (treatment 10 = 4546 kg ha^-1) and also for the N rich plot (treatment 11 = 4276 kg ha^-1).

The CVs (9, 9, and 7) from sensor readings in treatments 1, 6, and 11 at the time topdress N was applied tended to decline as preplant N increased.  Consistent with previous work (18), CVs were < 18, and it was expected that “catch-up” would be possible, which was confirmed (4453 kg ha^-1, treatments 4 and 5 versus 4537 kg ha^-1, treatment 15, Table 5).

The NUE at this site varied across all treatments.  In general, the 0 preplant plus topdress treatments had the highest NUE with the exception of treatment 11 (Table 5). 

The RI estimated using in season NDVI readings slightly underestimated RI_Harvest at this site (RI_NDVI = 1.14 and RI_Harvest = 1.3, Table 5).  RI_NDVI over time for 2003 did not change much from Feekes 3 to Feekes 4 (RI_NDVI=1.15, 1.13, and 1.14, Table 3).  The RI_Harvest were higher than the RI_NDVI’s (RI_Harvest= 1.3, Table 5).

Tipton, 2003

At Tipton in 2003 where 0 topdress N was applied, there was a linear response to N (treatment 1 = 1357 kg ha^-1, treatment 6 = 1264 kg ha^-1, treatment 11 = 2082 kg ha^-1).  Also, there was an increase in wheat grain yield whether or not preplant N was applied (Table 6).  Maximum yields were not achieved at this site from mid-season topdress N applications in plots receiving 0 preplant N in comparison with the maximum yielding plot (treatment 13).  At this site, it was not possible to “catch-up” with no N preplant plus a topdress N application, even though CVs were relatively low (Table 6).  However, it should be noted that “catch-up” was possible if treatment 5 was compared to treatment 11 (treatment 5 = 2278 kg ha^-1 and treatment 11 = 2082 kg ha^-1).

Response index estimated using in season NDVI readings was the same as RI_Harvest (RI_NDVI = 1.49 and RI_Harvest = 1.5) (Table 6). 

Covington, 2004

At Covington in 2004, there was a linear increase in wheat grain yield with increased N where no topdress was applied (treatment 1 = 1985 kg ha^-1, treatment 6 = 2846 kg ha^-1, and treatment 11 = 3751 kg ha^-1, Table 7).  At this site, there was a significant increase in wheat grain yield, and N rates required to maximize yields diminished as preplant N rates increased.  It should be noted that the highest yielding plot was a 0 preplant rate (treatment 5, Table 7), thus suggesting that “catch-up” was possible with respect to maximum yield and also with the N rich plot.

The CVs, for treatments 1, 6, and 11 were 18, 21, and 15 respectively.  Some of the CVs were < 18 indicating that maximum yields could be achieved even when early season N stress was present.

The NUE’s were generally higher for the 90 kg ha^-1 preplant treatments (Table 7), likely because this was a N responsive site.

The RI_NDVI was slightly over estimated at this site (RI_NDVI = 2.02 and RI_Harvest = 1.89) (Table 7).  RI_NDVI over time for 2004 did not change from Feekes 3 to Feekes 4 (RI_NDVI = 1.47, 2.19, 2.18, 2.02, 1.9 and 1.9, Table 3).  The RI_Harvest slightly differed from the RI_NDVI (RI_NDVI = 2.02 and RI_Harvest = 1.89, Table 3).

Lake Carl Blackwell, 2004

At Lake Carl Blackwell in 2004, there was a linear increase in wheat grain yield where 0 topdress N was applied (treatment 1 = 3047 kg ha^-1, treatment = 3502 kg ha^-1, and treatment 11 = 3766 kg ha^-1, Table 8).  Wheat grain yield increased as topdress N rates increased.  Maximum yields were achieved at this site from mid-season topdress N applications in plots receiving 0 preplant N (treatment 3 = 3675 kg ha^-1 versus treatment 11 = 3766 kg ha^-1), thus “catch-up” was possible for this site with 0 preplant plus topdress application in accordance with the highest yielding plot (Table 8).

The CVs varied once again at this site for treatments 1, 6, and 11 (18, 19, and 16, respectfully).  Consistent with CVs being less than 18, “catch –up” was possible at this site.

The NUE’s were generally higher for the 0 and 45 preplant kg ha^-1 treatments (Table 8).

Response index estimated using in-season NDVI readings was the same as RI_Harvest (RI_NDVI = 1.24 and RI_Harvest = 1.24) (Table 8). RI_NDVI over time for 2004 varied from Feekes 3 to Feekes 8 (RI_NDVI=1.10, 1.10, 1.11, 1.24, and 1.22).  The RI_Harvest was the same as the RI_NDVI at fertlization (Table 3).

Tipton, 2004

At Tipton in 2004, there was a linear increase in wheat grain yield where 0 topdress N was applied.  There was a significant increase in grain yield with applied topdress N rates for the 0 and 45 kg ha^-1 preplant rates with no increase from topdress N for the 90 kg ha^-1 preplant rates.  Maximum yields were achieved at this site from mid-season topdress N applications in plots receiving 0 preplant N with respect to the maximum yielding plot (treatment 8 = 4845 kg ha^-1).  Catch-up was also possible with respect to the N rich plot (treatment 11), since treatment 5 out yielded the N rich plot.  At this site, “catch-up” was possible with 0 preplant N plus topdresss N applications in accordance with the highest yielding plots (Table 9).  

The NUE’s were generally higher for the plots receiving 0-N preplant plus topdress N (Table 9).

Response index estimated using in season NDVI readings was under estimated for this site (RI_NDVI = 1.49 and RI_Harvest = 1.68, Table 9).  RI_NDVI over time for 2004 changed from Feekes 3 to Feekes 5 (RI_NDVI=.92, 1.09, and 1.49, Table 3).  The RI_Harvest was different than the RI_NDVI was at fertilization (Table 9).

DISCUSSION

In this study, 6 locations had a linear increase in wheat grain yield from topdress N applied to plots receiving 0-N preplant (Tables 4-9).  Also, 4 of the 6 sites had a increase in wheat grain yield for the topdress N rates whether or not preplant N had been applied (Tables 4-9).  Melaj et al. (21) stated that N uptake increased around the time of maximum crop growth, so application of fertilizer at tillering would increase N fertilizer recovery by the crop.  Early season plant N uptake can lead to increased plant N volatilization (22).  Boman et al. (23) states that a management strategy to reduce N loss would be to apply enough fertilizer N in the fall to establish the crop and apply the remaining N requirement in the late winter or early spring before rapid growth begins.  Warm soil temperatures after this time would coincide with rapid wheat growth and also increase nutrient demand.

If N application is made prior to the period of rapid uptake and growth, there is a potential for increased N uptake and N use efficiency (24, 25). At all three locations in 2003, the highest NUE’s were found where preplant N was applied.  In 2004, 0-N preplant plus topdress N treatment generally had improved NUE’s.  At two sites where early N stress was severe, preplant N applications were superior to 0-N preplant plus topdress N.  Woolfolk et al. (26) and Gauer et al.(27) agree that increasing grain protein by applying higher fertilizer N rates is relatively inefficient (NUE decreases with increasing N level), especially under dry soil conditions.  In our work, there was one exception at Lake Carl Blackwell, 2003 that was treatment 11 (Table 5).  Treatment 11’s NUE was almost the same as the highest NUE for the site (treatment 2 = 53 and treatment 11 = 52). Wuest and Cassman, (28, 29) indicated that a late-season N application has greater uptake efficiency and is more effective in increasing grain N levels than N applied at planting.  Alternatively, they noted that preplant N was more effective in increasing grain yields.

This work addresses an interesting question.  Can N applications be delayed until mid-season in winter wheat without decreasing wheat grain yields?  The majority of farmers in this region of the wheat belt apply all of their fertilizer N at planting.  Although topdress N applications have become more popular, it is still a common practice to apply anhydrous ammonia in the fall at rates exceeding 110 kg N ha^-1.  The ease of applying liquid UAN topdress and the advent of larger, 20-30 m wide applicators has assisted the extension of delaying fertilizer N until late February. 

Because maximum yields were achieved at 4 of 6 sites where all N applied was delayed until Feekes 5, this work has use in both efficiency and environmental implications.  Also, all 6 sites were able to catch-up with respect to the N-rich plots, however, it should be noted that 5 of the 6 N rich plots were not the maximum yielding plots.  RI over time showed little to no change for 2003 across all three sites.  However, in 2004, two of the three sites had a change in RI over time.  By delaying fertilizer N applications until post dormancy, there is decreased risk of NO₃-N leaching and/or surface fertilizer N runoff when preplant applications are made to the surface without incorporation.  Also, by applying fertilizer N to the foliage in late February, increased use efficiency can be realized (foliar N uptake) when compared to preplant soil applied N (N subject to NO₃-N leaching, immobilization, denitrification, surface volatilization, and early season plant N loss). 

These results are not yet definitive concerning whether or not all N should be delayed until mid-season.  The reason for this is because exceptional growing conditions occurred, whereby timely rainfall was received immediately following topdress fertilizer N applied, especially for 2004.  Although not explicitly evaluated in other work conducted in Oklahoma, there have been dry springs where delayed topdress fertilizer N was not beneficial and maximum yields were not produced.  This was evident in many of the GreenSeeker sensor experiments conducted by OSU in 1999, 2000, 2001, whereby the topdress N plots never achieved the same yields as what was found in the N Rich Strip (N applied at adequate but not excessive rates preplant) (W. Raun, J. Solie, personal communication, July 2004, and reported on http://nue.okstate.edu). 

CONCLUSIONS

In 2002-2004, obtaining maximum wheat grain yields from topdress N applications in 0-N preplant plots was possible at four of the six locations.  At three of the six sites, the RI estimated using in-season NDVI readings was underestimated.  Even when early season N stress was present (0-N preplant), N applied topdress at Feekes 5 resulted in maximum or near maximum yields at 4 of 6 sites when compared to other treatments receiving both preplant and topdress N.  Furthermore, when compared to the conventional 90 kg ha^-1 preplant N, mid-season N applied (0-N preplant) resulted in maximum yields at all 6 sites.  When plot CVs (estimate of plant stand) were > 18, maximum yields could not be achieved when N fertilization was delayed until mid-season.  Alternatively, when plot CVs were < 18, delaying all N fertilization until mid-season resulted in maximum yields and increased NUE.

BIBLIOGRAPHY

1.         Vidal, I.; Longeri, L.; Hetier, J.M.  Nitrogen uptake and chlorophyll meter measurements in spring wheat.  Nutrient Cycling in Agroecosystems. 1999, 55, 1-6.

2.         Raun, W.R.; Johnson, G.V.; Stone, M.L.; Thomason, W.E.; Lukina, E.V.  In-season prediction of yield potential in winter wheat. Better Crops. 1999, 83, 24-25.

3.         Raun, W.R.; Solie, J.B.; Johnson, G.V.; Stone, M.L.; Lukina, E.V.; Thomason, W.E.  In-season prediction of potential grain yield in winter wheat using canopy reflectance.  Agron. J. 2001, 93, 131-138.

4.         Raun, W.R.; Solie, J.B.; Johnson, G.V.; Stone, M.L.; Whitney, R.W.; Lees, H.L.; Sembiring, H.; Phillips, S.B.  Micro-variability in soil test, plant nutrient, and yield parameters in bermudagrass.  Soil Sci. Soc. Am. J. 1998, 62, 683-690.

5.         Solie, J.B.; Raun, W.R.; Stone, M.L.  Submeter spatial variability of selected soil and bermudagrass production variables.  Soil Sci. Soc. Am. J.  1999, 63, 1724-1733.

6.         Raun, W.R.; Solie, J.B.; Johnson, G.V.; Stone, M.L.; Mullen, R.W.; Freeman, K.W.; Thomason, W.E.; Lukina, E.V.  Improving nitrogen use efficiency in cereal grain production with optical sensing and variable rate application.  Agron. J. 2002, 94, 815-820.

7.         Solie, J.B.; Raun, W.R.; Whitney, R.W.; Stone, M.L.; Ringer, J.D.  Optical sensor based field element size and sensing strategy for nitrogen application. Trans. ASAE 1996, 39, 1983-1992.

8.         Stone, M.L.; Solie, J.B.; Raun, W.R.; Whitney, R.W.; Taylor, S.L.; Ringer, J.D.  Use of spectral radiance for correcting in-season fertilizer nitrogen deficiencies in winter wheat. Am. Soc. Agricul. Engineers.  1996, 39, 1623-1631.

9.         Raun, W.R.; Johnson, G.V.  Improving nitrogen use efficiency for cereal production.  Agron. J. 1999, 91, 357-363.

10.       10.       Lukina, E.V.; Freeman, K.W.; Wynn, K.J.; Thomason, W.E.; Mullen, R.W.; Klatt, A.R.; Johnson, G.V.; Elliott, R.L.; Stone, M.L.; Solie, J.B.; Raun, W.R.  Nitrogen fertilization optimization algorithm based on in-season estimates of yield and plant nitrogen uptake. J. Plant Nutr. 2001, 24, 885–898.

11.       Large, EC.  Growth Stages in Cereals.  Plant Pathol. 1954, 3, 28-129.

12.       12.       Malakoff, D.  Death by suffocation in the gulf of Mexico. Science. 1998, 281, 190-192.

13.       Johnson, G.V.; Raun, W.R.  Using a response index to improve nitrogen management for winter wheat. J. Plant Nutr. 2003, 26, 249-262.

14.       14.       Mullen, R.W.; Thomason, W.E.; Johnson, G.V.; Freeman, K.W.; Stone, M.L.; Solie, J.B.; Raun, W.R.  Use of an in-season response index to predict potential yield increases from applied nitrogen.  In Agronomy Abstracts (CD-ROM).  ASA, CSSA, and SSSA, Madison, WI.  2001.

15.       Rickman, R.W.; Waldman, S.E.; Klepper, B.  MODWht3: A development – driven wheat growth simulation. Agron. J. 1996, 88, 176–185.

16.       Mullen, R.W.; Freeman, K.W.; Raun, W.R.; Johnson, G.V.; Stone, M.L.; Solie, J.B.  Use of an in-season response index to identify the potential to increase wheat yield with additional nitrogen. Agron. J. 2003, 95, 347-351.

17.       Johnson, G.V.; Raun, W.R.; Mullen, R.W.  Nitrogen use efficiency as influenced by crop response index.  In Agronomy Abstracts,  ASA, CSSA, and SSSA, Madison, WI.  2000, 291.

18.       http://nue.okstate.edu.  (accessed July 2004).

19.       Schepers, J.S.; Francis, D.D.; Thompson, M.T.  Simultaneous determination of total C, total N, and ¹⁵N on soil and plant material.  Commun. Soil Sci. Plant Anal. 1989, 20, 949-959.

20.       SAS Institute.  The SAS system for windows version 8.02. SAS Inst., Cary, NC. 2001.

21.       21.      Melaj, M.A.; Echeverria, H.E.; Lopez, S.C.; Studdert, G.; Andrade, F.; Barbaro, N.O.  Timing of nitrogen fertilization in wheat under conventional and no-tillage systems.  Agron. J. 2003, 95, 1525-1531.

22.       Lees, H.L.; Raun, W.R.; Johnson, G.V.  Increased plant N loss with increasing nitrogen applied in winter wheat observed with ¹⁵N.  J. Plant Nutr. 2000, 23, 219-230.

23.       Bowman, R.K.; Westerman, R.L.; Jojola, M.E.  Spring-applied nitrogen fertilizer influence on winter wheat and residual soil nitrate.  J. Prod. Agric.  1995, 8, 584-589.

24.       Sowers, K.E.; Pan, W.L.; Miller, B.C.; Smith, J.L.  Nitrogen use efficiency of  split nitrogen applications of soft white winter wheat.  Agron. J. 1994, 86, 942-948.

25.       Johnston, A.M.; Fowler, D.B.  No-till winter wheat production: Response to spring applied nitrogen fertilizer from and placement.  Agron. J.  1991,  83, 722-728.

26.       Woolfolk, C.W.; Raun, W.R.; Johnson, G.V.; Thomason, W.E.; Mullen, R.W.; Wynn, K.J.; Freeman, K.W.  Influence of late-season nitrogen applications on yield and grain nitrogen in winter wheat.  Agron. J. 2002, 94, 429-434.

27.       Gauer, L.E.; Grant, C.A.; Gehl, D.T.; Bailey, L.D.  Effects of nitrogen fertilization on grain protein content, nitrogen uptake, and nitrogen use efficiency of six spring wheat (Triticum aesticum L.) cultivars, in relation to estimated moisture supply.  Can. J. Plant Sci. 1992, 72, 235-241.

28.       Wuest, S.B.; Cassman, K.G.  Fertilizer-nitorgen use efficiency of irrigated         wheat: 1. uptake efficiency of preplant versus late-season application.  Agron. J. 1992, 84, 682-688.

29.       Wuest, S.B.; Cassman, K.G.  Fertilizer-nitrogen use efficiency of irrigated wheat: 2.  partitioning efficiency of preplant versus late-season application.  Agron. J. 1992, 84, 689-694.

Table 1. Treatment Structure for all 3-experimental sites, (Covington, Lake Carl Blackwell, and Tipton OK, 2002-2004.)

* RI is the actual response index determined for that field.

^† Response Index was adjusted as a function of CV in 2003-2004.

Table 2. Initial surface (0-15 cm) soil chemical characteristics and classification at Covington, Lake Carl Blackwell, and Tipton, Oklahoma.

*pH – 1:1 soil: water, K and P – Mehlich III, NH₄-N and NO₃-N – 2M KCL, Total N and Organic Carbon– dry combustion.

Table 3. Field activities, planting dates, seeding rates, Pre-plant nitrogen dates, Foliar nitrogen dates, sensor reading dates, Growing Degree Days, RI_NDVI, RI_Harvest, and harvest dates.  (Covington, Lake Carl Blackwell, and Tipton OK, 2002-2004.)

* Covington had 18-46-0 @ 56 kg ha^-1 banded with seed (2002).

* Covington had 11-52-0 @ 50 kg ha^-1 banded with seed (2003).

* Lake Carl Blackwell had 0-46-0 @45 kg ha^-1 preplant incorporated (2003 and 2004)

* F =Feekes growth stages, determined by (Large 1954)

* GDD = Growing Degree Days: T_max+ T_min/2- 4.4˚ C.

Table 4. Treatment, preplant nitrogen, topdress nitrogen RI factor, topdress nitrogen applied, total nitrogen applied, yield, grain nitrogen uptake, and % NUE, for Covington, 2003.

Table 5. Treatment, preplant nitrogen, topdress nitrogen RI factor, topdress nitrogen applied, total nitrogen applied, yield, grain nitrogen uptake, and % NUE, for Lake Carl Blackwell, 2003.

Table 6. Treatment, preplant nitrogen, topdress nitrogen RI factor, topdress nitrogen applied, total nitrogen applied, yield, grain nitrogen uptake, and % NUE, for Tipton, 2003.

Table 7. Treatment, preplant nitrogen, topdress nitrogen RI factor, topdress nitrogen applied, total nitrogen applied, yield, grain nitrogen uptake, and % NUE, for Covington, 2004.

Table 8. Treatment, preplant nitrogen, topdress nitrogen RI factor, topdress nitrogen applied, total nitrogen applied, yield, grain nitrogen uptake, and % NUE, for Lake Carl Blackwell, 2004.

Table 9. Treatment, preplant nitrogen, topdress nitrogen RI factor, topdress nitrogen applied, total nitrogen applied, yield, grain nitrogen uptake, and % NUE, for Tipton, 2004.