Journal list menu

Volume 3, Issue 1 e20025
ORIGINAL RESEARCH ARTICLE
Open Access

Wheat grain yield and nitrogen uptake as influenced by fertilizer placement depth

Ryan Bryant-Schlobohm

Corresponding Author

Ryan Bryant-Schlobohm

Oklahoma State University, 371 Agriculture Hall, Stillwater, OK, 74078 USA

Correspondence

Ryan Bryant-Schlobohm, Oklahoma State University, 371 Agriculture Hall, Stillwater, OK 74078, USA.

Email: [email protected]

Search for more papers by this author
Jagmandeep Dhillon

Jagmandeep Dhillon

Oklahoma State University, 371 Agriculture Hall, Stillwater, OK, 74078 USA

Search for more papers by this author
Gwendolyn B. Wehmeyer

Gwendolyn B. Wehmeyer

Oklahoma State University, 371 Agriculture Hall, Stillwater, OK, 74078 USA

Search for more papers by this author
William R. Raun

William R. Raun

Oklahoma State University, 371 Agriculture Hall, Stillwater, OK, 74078 USA

Search for more papers by this author
First published: 04 March 2020
Citations: 5

[Correction added on 09 March 2020, after first online publication: spelling of “Gwendolyn B. Wehmyer” changed to “Gwendolyn B. Wehmeyer”.]

Abstract

Global nitrogen use efficiency (NUE) for cereal production is estimated to be only 33%. Providing producers with efficient methods to increase the effectiveness of their N applications is integral to agricultural sustainability and environmental quality. This study was conducted to evaluate the effect of urea ammonium nitrate (UAN) injected at different depths on grain yield and uptake of N in grain. Liquid UAN (28–0–0) was applied in bands at depths of 5 and 10 cm, along with surface applications, all at various N rates around Feekes growth stage 5. Placement depth had the most significant impact on yield at low N rates. Subsurface application at 10 cm was most beneficial in low N no-till (NT) soils, whereas surface treatments produced higher yields in low N environments of conventional till (CT) systems. Three of the four locations experienced higher rates of N uptake from subsurface applications when compared with surface treatments. No difference in grain N uptake was apparent between application depths of 5 and 10 cm. Subsurface N applications were beneficial in reducing rates of ammonia volatilization from urea-based fertilizers. While there was no clear separation between 5 and 10 cm application depths, subsurface depths of 10 cm provide the most significant promise in benefiting yield in low N environments of NT soils and increasing grain N across CT and NT systems.

Abbreviations

  • CT
  • conventional tillage
  • NT
  • no-till
  • NDVI
  • normalized difference vegetation index
  • NUE
  • nitrogen use efficiency
  • UAN
  • urea ammonium nitrate
  • 1 INTRODUCTION

    Agricultural intensification demands enhanced nutrient use efficiencies. Managing inputs in a manner that promotes maximized production with substantial sustainability initiatives are imperative in today's agroecosystems. Nitrogen plays an imperative role in the physiological mechanisms that determine the yield and quality of grain and is often one of the most limiting plant nutrients (Thomason et al., 2002). Additionally, N is often the subject of environmental quality concerns, as denitrification, leaching, and volatilization are significant pathways of loss in which excess, reactive N floods the environment (Aulakh, Rennie, & Paul, 1982; Chichester & Richardson, 1992; Fowler & Brydon, 1989; Hargrove, Kissel, & Fenn, 1977). Interactions between environmental conditions and producer practices are critical avenues that dictate N loss and ultimate nitrogen use efficiency (NUE) of cropping systems (Di & Cameron, 2002; Freney, Simpson, & Denmead, 1981; Wallenstein, Myrold, Firestone, & Voytek, 2006). As global NUE of cereal production is estimated to be 33% (Raun & Johnson, 1999), identifying practical methods to stimulate effective rates of plant N uptake is critical to improving the economic viability and environmental sustainability of modern agriculture. Complexing this issue is yearly environmental variability which necessitates optimum N rates be dependent on annual environmental conditions (Raun et al., 2019). Solie, Monroe, Raun, and Stone (2012) notes the benefit normalized difference vegetation index (NDVI) data provides in determining inseason N rates, however, proper application method in critical to maximizing NUE.

    Global wheat (Triticum aestivum L.) production exceeded 771 million Mg in 2017 (FAO, 2019). As a staple cereal throughout the world, enhancing the NUE of wheat production can provide immense environmental and economic benefit. Various production practices have demonstrated the ability to promote greater NUE. Midseason applications of N award the benefits of enhanced N utilization by the plant (Dhillon, Figueiredo, Eickhoff, & Raun, 2020a; Malhi & Nyborg, 1983; Olson, 1986; Olson & Swallow, 1984; Raun et al., 2017a). Furthermore, subsurface placement of N in concentrated bands leads to decreased N loss and higher plant uptake (Carter & Rennie, 1984; Dhillon et al., 2020b; Malhi, Nyborg, & Solberg, 1996; Nkebiwe, Weinmann, Bar-Tal, & Müller, 2016; Rao & Dao, 1996; Rees. et al., 1996; Sharpe, Harper, Giddens, & Langdale, 1988; Tomar & Soper, 1981).

    Ammonia volatilization and immobilization play a substantial role in diminishing plant-available N from urea-based fertilizers, these effects can be compounded depending on interactions with placement method and tillage system (Bandel, Dzienia, & Stanford, 1980; Rice & Smith, 1984). As previously established, subsurface applications of N prove beneficial in enhancing grain N uptake, a major avenue to enhancing NUE and collecting premiums for grain protein. Rees et al. (1996) demonstrated improved N recovery in wheat from incorporated fertilizer and subsurface bands in comparison to surface applications. Malhi et al. (1996) further documented the benefit subsurface bands of N provide in enhancing N recovering within no-till (NT) systems. However, little work has been conducted in evaluating different subsurface placement depths of midseason urea ammonium nitrate (UAN) in winter wheat. The objective of this study is to evaluate two placement depths of UAN at various midseason N rates in conventional and conservation tillage environments to identify proper placement depths for enhancing N uptake, yield, and NUE.

    2 MATERIALS AND METHODS

    Four field trials were initiated in the 2015–2016 growing season to evaluate the appropriate depth to apply UAN in winter wheat with relevance to the interactive effects’ depth displayed on grain yield, grain N uptake, and NUE. Within Oklahoma, two conventional tillage (CT) and two conservation tillage (NT) locations were employed. The experiment sites at Hennessey (HEN, NT, 36°6′59.2″N, 97°54′4.3″W), Lahoma (LAH, CT, 36°23′23.0″N, 98°6′42.2″W), Lake Carl Blackwell (LCB, CT, 36°8′38.7″N, 97°17′1.3″W), and Perkins (PRK, NT, 35°59′35.5″N, 97°2′38.7′W) were located on a Bethany silt loam (fine, mixed, superactive, thermic Pachic Paleustolls), a Grant silt loam (fine-silty, mixed, superactive, thermic Udic Argiustolls), a Pulaski fine sandy loam (coarse-loamy, mixed, superactive, nonacid, thermic Udic Ustifluvents) and a Konawa (fine-loamy, mixed, active, thermic Ultic Haplustalfs), and Teller fine sandy loam (fine-loamy, mixed, active, thermic Udic Argiustolls), respectively. A randomized complete block design with three replications and 14 treatments was utilized at all sites. Trial size measured 42.7 by 24.4 m, with each plot measuring 6.1 by 3.0 m. Replications were separated with 3.0 by 42.7 m allies.

    Core Ideas

    • Winter wheat grain yield can benefit from 10 cm application depths of UAN in low N NT systems.
    • Application depths of 10 cm increases NT and CT grain N, and increases yield in low N NT.

    Iba, a winter wheat variety from Oklahoma Foundation Seed, was used at all locations with seeding rates of 67, 82, 82, and 84 kg ha−1 at HEN, LAH, LCB, and PRK, respectively. All sites received a blanket application of 45 kg N ha−1, with the exception of the 0 N check (treatment 14). Conventional producer practices were employed for weed control and disease management. Midseason N was applied to all trials when wheat approached Feekes growth stage 5 with a rolling coulter applicator, utilizing guide wheels to control depth. Coulters were spaced 43 cm apart and N was applied in bands on the surface, and at subsurface depths of 5 and 10 cm at rates of 34, 67, 100, and 134 kg N ha−1.

    Weekly NDVI readings were taken with a GreenSeeker throughout the spring growing season leading up to reproductive growth to gauge plant response to placement method of applied N. Readings fall between a scale of 0−1, with NDVI data approaching 1 signifying higher quantities of green biomass. Grain harvest was achieved with a Massey Ferguson self-propelled combine, harvesting the center 1.8 m of each plot. Grain subsamples were collected from each plot and placed in a drier at 60 °C for 24 h. Subsamples were subsequently ground down and rolled to a fine flour, at which point 150 mg of each sample was analyzed for total N utilizing a LECO Combustion Carbon/Nitrogen Analyzer. To determine NUE, the following formula was applied:
    Grain N uptake of treated plot Grain N uptake of 0 N check N rate applied

    Statistical analysis was conducted with SAS version 9.4 (SAS Institute, 2013). Within SAS, LSD option within PROC GLM was used for mean separations at alpha = .05. Additionally, single degree of freedom contrasts were employed to evaluate treatment linear relationships and treatment differences. Each site was independently analyzed to avoid overlooking treatment response due to different environmental effect (Raun et al., 2017b).

    3 RESULTS

    3.1 Yield

    Focusing on LAH, the surface N rate of 34 kg ha−1 produced significantly higher grain yield than the 5- and 10-cm application depths (Table 1). Moving up to 67 kg ha−1, 10-cm application depth was not significantly different from the surface treatment in this CT system, with both being statistically greater than the 5-cm application depth. Subsurface treatments were significantly higher than the surface treatment at midseason rates of 100 kg N ha–1. Single degree of freedom contrasts (Table 2) for LAH were significant and in favor of surface treatments when comparison was made to 5-cm application depths (p = .0161). No difference was evident between surface and 10-cm treatments (p = .7710), however, 10-cm depths were significant in comparison to application depths of 5 cm (p = .0313). Influence of midseason N placement had an alternative effect in the NT system of PRK. Subsurface depth of 10 cm was significantly higher than the surface treatment at 34 kg ha−1. Yield appeared to be maximized at PRK following 67 kg ha−1, as no statistical differences were apparent. No significant contrasts resulted from comparisons of placement methods at PRK. Grain yield from LCB and HEN provided limited results, as weed pressure and flooding limited response at these locations, respectively.

    TABLE 1. Effect of midseason N placement depth and rate on grain yield, means separated at α = .05
    Treatment information Grain yield
    Pre-plant N rate Midseason application depth Midseason N rate HEN NT LAH CT LCB CT PRK NT
    kg ha−1 cm kg ha−1 Mg ha−1
    45 n/a 0 5.22Aa 1.74DE 2.31B 3.54E
    45 0 34 4.46BA 3.73A 2.98BA 4.24ED
    45 5 34 4.23BA 2.47C 3.23BA 4.68DC
    45 10 34 4.77A 2.72BC 3.33BA 5.02BC
    45 0 67 3.66BA 3.65A 4.16A 5.37BAC
    45 5 67 4.62BA 2.33DC 3.05BA 5.34BAC
    45 10 67 4.65BA 3.28BA 3.73BA 5.55BA
    45 0 100 4.00BA 2.70BC 2.83BA 5.85A
    45 5 100 4.50BA 3.78A 4.18A 5.43BAC
    45 10 100 4.68BA 3.94A 2.84BA 5.75BA
    45 0 134 4.53BA 3.97A 3.15BA 5.81A
    45 5 134 2.93B 3.71A 2.69B 5.94A
    45 10 134 4.00BA 3.91A 3.27BA 5.61BA
    0 n/a 0 3.95BA 1.29E 2.36B 2.58F
    MSE 1.184 0.175 0.743 0.216
    SED 0.920 0.341 0.703 0.380
    CV 25.32 13.55 27.35 9.20
    • Note. HEN, Hennessey; LAH, Lahoma; LCB, Lake Carl Blackwell; PRK, Perkins; NT, no-till; CT, conventional till; n/a, not applicable; MSE, means square error; SED, standard error of the difference of two equally replicated means; CV, coefficient of variation.
    • a Numbers with same letters are not statistically different at α = .05.
    TABLE 2. Single df contrasts comparing surface and subsurface treatments
    Grain yield Grain N NDVI, F7
    Contrasts HEN LAH LCB PRK HEN LAH LCB PRK HEN LAH LCB PRK
    Linear surface ns ns ns ** * ** ns ** ns ns ns **
    Linear 5 cm ns ** ns ** ** ** ** ** ns * ns ns
    Linear 10 cm ns ** ns ns ** ** ** ** ns *** ns *
    Surface vs. 5 cm ns * ns ns ns ** ** ** ns ** ns ns
    Surface vs. 10 cm ns ns ns ns ns ** * * ns * ns ns
    5 cm vs. 10 cm ns * ns ns ns ns ns ns ns * ns ns
    • Note. NDVI, normalized difference vegetation index; HEN, Hennessey; LAH, Lahoma; LCB, Lake Carl Blackwell; PRK, Perkins; ns, not significant.
    • *Significant at the .05 probability level.
    • **Significant at the .01 probability level.
    • ***Significant at the .10 probability level.

    3.2 Grain nitrogen uptake

    Application depths of 5 and 10 cm provided significant improvements over surface treatments at LAH, LCB, and PRK to grain N (Table 3). Additionally, PRK subsurface treatments were statistically superior to surface applications at 100 kg N ha−1. Contrast data for influence of application depth to grain N shows a similar trend (Table 2). For LAH, LCB, and PRK, 5-cm depths improved grain N over surface treatments (p = .0001, p = .0031, p = .0033, respectively). Likewise, 10-cm applications depths were superior to surface treatments at LAH, LCB, and PRK (p = .0016, p = .0101, p = .0338, respectively). Contrasts between 5- and 10-cm depths were not significant at any location.

    TABLE 3. Effect of midseason N placement depth and rate on grain N uptake, means separated at α = .05
    Treatment information Grain N
    Pre-plant N rate Midseason application depth Midseason N rate HEN NT LAH CT LCB CT PRK NT
    kg ha−1 cm kg ha−1 %
    45 n/a 0 1.56FEa 1.43G 1.66BEDC 1.53F
    45 0 34 1.59E 1.51GF 1.61EDC 1.61FE
    45 5 34 1.74DEC 1.67EF 1.63EDC 1.70DE
    45 10 34 1.75DEC 1.70EF 1.58ED 1.68FE
    45 0 67 1.90BDAC 1.70EF 1.45ED 1.86C
    45 5 67 1.71DE 2.11BAC 1.69BEDC 1.86C
    45 10 67 1.73DEC 1.86ED 1.71BEDC 1.83DC
    45 0 100 1.88BDC 2.01DC 1.81BDC 1.88C
    45 5 100 1.97BAC 2.09BAC 1.80BDC 2.08BA
    45 10 100 1.98BAC 2.13BAC 1.96BAC 2.03B
    45 0 134 1.94BDAC 2.03BDC 1.40E 2.02B
    45 5 134 2.14A 2.23A 2.29A 2.19A
    45 10 134 2.08BA 2.22BA 2.00BA 2.17BA
    0 n/a 0 1.33F 1.43G 1.67BEDC 1.61FE
    MSE 0.022 0.014 0.048 0.008
    SED 0.123 0.096 0.179 0.071
    CV 8.32 6.22 12.67 4.66
    • Note. HEN, Hennessey; LAH, Lahoma; LCB, Lake Carl Blackwell; PRK, Perkins; NT, no-till; CT, conventional till; n/a, not applicable; MSE, means square error; SED, standard error of the difference of two equally replicated means; CV, coefficient of variation.
    • a Numbers with same letters are not statistically different at α = .05.

    3.3 Normalized difference vegetation index

    With the exception of LAH, limited differences were present in NDVI data (Table 4), which limited our ability to draw definitive conclusions from GreenSeeker data in regards to interactions between tillage system and application depth. At midseason N rates of 34, 67, and 134 kg ha−1, surface applications had significantly higher NDVI readings at F7. No difference in NDVI data was evident between 5- and 10-cm depths. Contrast results for application depth and NDVI at LAH show a similar trend (Table 2). As surface treatments were significantly greater than both 5 and 10-cm application depths (p < .0001 and p = .0267, respectively). Furthermore, contrast data shows 10-cm depths having a greater NDVI reading at F7 compared to 5-cm treatments (p = .0251).

    TABLE 4. Effect of midseason N placement depth and rate on normalized difference vegetation index (NDVI) at Feekes 7, means separated at α = .05
    Treatment information NDVI, Feekes 7
    Pre-plant N rate Midseason application depth Midseason N rate
    kg ha−1 cm kg ha−1 HEN NT LAH CT LCB CT PRK NT
    45 n/a 0 0.70BAa 0.38E 0.59BA 0.45C
    45 0 34 0.72BA 0.51BA 0.69BA 0.53B
    45 5 34 0.72BA 0.40ED 0.62BA 0.57BA
    45 10 34 0.70BA 0.43EDC 0.64BA 0.56B
    45 0 67 0.73A 0.51BA 0.65BA 0.58BA
    45 5 67 0.63B 0.39E 0.61BA 0.58BA
    45 10 67 0.69BA 0.45DC 0.64BA 0.56B
    45 0 100 0.66BA 0.43EDC 0.67BA 0.64A
    45 5 100 0.69BA 0.46BC 0.67BA 0.59BA
    45 10 100 0.65BA 0.48BAC 0.71A 0.59BA
    45 0 134 0.65BA 0.53A 0.65BA 0.64A
    45 5 134 0.69BA 0.46BDC 0.73A 0.60BA
    45 10 134 0.64BA 0.48BAC 0.70BA 0.64A
    0 n/a 0 0.47C 0.31F 0.55B 0.36D
    MSE 0.004 0.001 0.009 0.002
    SED 0.049 0.025 0.077 0.037
    CV 8.98 7.33 14.40 8.05
    • Note. HEN, Hennessey; LAH, Lahoma; LCB, Lake Carl Blackwell; PRK, Perkins; NT, no-till; CT, conventional till; n/a, not applicable; MSE, means square error; SED, standard error of the difference of two equally replicated means; CV, coefficient of variation.
    • a Numbers with same letters are not statistically different at α = .05

    3.4 Nitrogen use efficiency

    As a result of combining variable yield data with grain N, NUE levels do not show any remarkable statistical differences (Table 5). Excluding HEN and LCB due to flooding and weed pressure, respectively, LAH and PRK show a trend for 10-cm application depths to have numerically superior levels of NUE in comparison to surface treatments, with the exception of N rates of 34 and 67 kg ha−1 at LAH.

    TABLE 5. Nitrogen use efficiency (NUE) as influenced by midseason N rate and application depth
    Treatment information NUE
    Pre-plant N rate Midseason application depth Midseason N rate HEN NT LAH CT LCB CT PRK NT
    kg ha−1 cm kg ha−1 %
    45 n/a 0 80Aa 9D <0A 27A
    45 0 34 47BAC 45A 9A 34BA
    45 5 34 47BAC 26BC 15A 48BA
    45 10 34 58BA 32BAC 15A 53A
    45 0 67 55BA 37BAC 18A 52A
    45 5 67 33BC 25BDC 9A 52A
    45 10 67 27BC 35BAC 21A 53A
    45 0 100 35BC 23DC 7A 47BA
    45 5 100 42BC 40BA 24A 49BA
    45 10 100 31BC 43A 10A 51A
    45 0 134 26BC 34BAC 2A 42BA
    45 5 134 19C 35BAC 11A 50A
    45 10 134 35BC 37BAC 14 45BA
    0 n/a 0 n/a n/a n/a n/a
    MSE 393.597 95.279 548.701 176.538
    Mean 41 32 11 46
    CV 48.19 30.16 208.50 28.68
    • Note. HEN, Hennessey; LAH, Lahoma; LCB, Lake Carl Blackwell; PRK, Perkins; NT, no-till; CT, conventional till; n/a, not applicable; MSE, means square error; CV, coefficient of variation.
    • a Numbers with same letters are not statistically different at α = .05.

    4 DISCUSSION

    Surface applications of UAN provided higher yields in CT systems when N was limited at low rates. Conversely, at low N rates in NT systems, subsurface application depths of 10 cm provided substantial increase to yield over surface applications. These results are similar to what was found by Rao and Dao (1996) where subsurface applications were more beneficial to yield in NT systems, with variable impact in CT soils. Results of this study can be explained by Doran (1980), who's research concluded elevated microbial populations in the upper 7.5 cm of NT systems as a result of shifts in soil properties influencing microbial growth, ultimately increasing rates of immobilization of surface-applied N within this zone. A reversal of these populations is found in CT systems, where rates of microbial immobilization increase substantially between 7.5 and 15 cm in the rhizosphere. Work by Helgason, Walley, and Germida (2009) and Mathew, Feng, Githinhi, Ankumah, and Balkcom (2012) further documented the shift of surface microbial activity NT soils encounter. Although not a subject of the study, increased rates of microbial immobilization in the upper surface of NT soils can likely be attributed to lower yields of surface applied UAN at PRK. Nitrogen is at risk of short-term microbial immobilization when applied to the surface of NT soils (Rice & Smith, 1984). This immobilization can exacerbate plant N deficiency during yield determining growth stages in N limited environments, thus leading to lower yields than would be encountered in a mirrored CT system. Furthermore, the shift of microbial populations to the lower 7.5–15 cm of CT soils is a logical explanation for surface-applied UAN providing higher yields than subsurface applications at LAH. Although surface applied UAN is subject to volatilization, the rates of microbial immobilization likely outpaced volatilization rates in CT soils, leading to higher yields at low N rates for surface-applied N. Adequate availability of N at higher rates of this study did not provide yield separations, as volatilization and immobilization did not likely reach levels which diminished plant available N.

    Placement of N had a substantial influence on grain N. Subsurface applications had a greater impact on increasing grain N than did surface treatments across NT and CT systems. Similarly, Schlegel, Dhuyvetter, and Havlin (2003) and Nkebiwe et al. (2016) identified the benefit subsurface N placement provides to increased grain N over surface applications. Alternatively, Dhillon et al. (2020b) noted that timing of N application had a more significant impact than application method on grain yield, grain N uptake, and NUE. Pronounced differences were not evident between 5- and 10-cm application depths. Results of grain N uptake diverged from what was witnessed with interactions to grain yield. Grain yield is strongly influenced at an earlier growth stage than what final grain N concentration is. Prior to anthesis, the majority of wheat aboveground N is accumulated, as peak nutrient uptake occurs just prior to stem elongation (Kalen & Sadler, 1990; Oscarson, Lundborg, Larsson, & Larsson, 1995). Morris et al. (2006) notes that wheat possesses the ability to recover from early season N stress if adequate N is available for uptake before initiation of stem elongation, or Feekes growth stage 5. However, N deficiencies following stem elongation will lead to reduced spike formation, ultimately limiting yield potential (Frederick & Camberato, 1995). As wheat is able to uptake N following anthesis (Oscarson et al., 1995; Ransom et al., 2016; Van Sanford & MacKown, 1987), there is a longer window for N uptake to influence concentrations of grain N. Bly and Woodard (2003) concluded that grain N can be increased through post-pollination by applications of foliar N, as Fowler and Brydon (1989) note that delayed availability of N is associated with increased grain N.

    In this study, N is likely recovered from initial immobilization in active microbial zones of the soil and made available for grain N uptake at later growth stages. This initial immobilization is what separated grain yield in CT and NT at low N rates due to decreased availability prior to critical growth stages at which yield is determined. Ultimately, subsurface applications of N play a significant role in increasing grain N content, as subsurface placement reduces risk for volatilization of urea-based fertilizers. No clear separation is witnessed between 5- and 10-cm depths for grain N uptake, however, the benefit 10-cm applications provided in increasing yield at low N rates in NT systems shows their worth. In support of 10-cm depths more than 5 cm, Rochette et al. (2014) notes placement of N at depths >7.5 cm provides the highest rates of retention in avoiding volatilization. Further research to solidify 10-cm depths will prove beneficial, as subsurface placement of midseason UAN at depths of 10 cm suggests enhanced grain N in CT and NT systems, and elevated yields in low N environments of NT soils.

    ACKNOWLEDGMENTS

    Special thanks is to be given to the numerous graduate students and staff of the Oklahoma State University Soil Fertility program for their help and support of this project. Additionally, the authors would like to thank the Oklahoma Agricultural Experiment Station and the OSU Department of Plant and Soil Sciences for research support.

      CONFLICT OF INTEREST

      The authors declare no conflict of interest.