Nitrogen release dynamics of enhanced efficiency fertilisers (EEFs): placement, soil factors and plant uptake

Take home message

Enhanced efficiency nitrogen fertilisers (EEFs) control and delay the release of nitrogen (N) into soil, potentially leading to improved crop uptake and reduced environmental losses. However, in these studies we found that banded application of EEFs did not demonstrate any improvements in maize grain yields relative to conventional urea fertiliser.

The main responses to EEFs were as follows:

• Nitrification inhibitors (NIs) significantly reduced N₂O emissions compared to urea and delayed nitrification but did not increase crop N uptake. Residual N in the soil was generally shallower (≤ 0.3m) than from urea and provided no additional N to a subsequent maize crop
• Urease inhibitors (UIs) did not improve N uptake by the initial maize crop and did not have any greater residual benefit than urea. UIs increased the depth of NO₃^- movement compared to urea, with most residual N located > 0.3m in the profile
• Polymer coated urea (PCUs) delayed N release, then provided a low and continuous supply of N throughout the season. This N was not effectively captured by the maize crop, with either no yield advantages or in some cases poorer yields compared to urea

Despite EEFs performing as designed, no clear productivity or yield advantages were found relative to conventional N fertilisers (i.e., urea) in banded applications. The slower rate of N release and / or availability appears to be poorly synchronised with crop N demand.

Introduction

Improved nitrogen use efficiency (NUE) in Australia’s cropping industries is critical for

1. Reducing fertiliser costs to growers
2. Improving agronomic efficiency and crop yields and
3. Reducing environmental impacts (e.g., nitrous oxide [N₂O] emissions, leaching & runoff losses).

Enhanced efficiency fertilizer (EEF) technology (i.e., inhibitors and controlled-release products) are potential tools for improving NUE in cropping environments. However, the effective use of EEFs is hampered by a poor understanding of how these technologies behave in soil and how this translates to agronomic outcomes in the field.  Furthermore, most EEF products have been developed and tested in temperate environments where they are broadcast or incorporated with conventional tillage. Thus, a better understanding of the mechanisms at work when different EEF technologies are band-applied under Australian field conditions is required.

This study evaluated several urea-based EEFs, including urease inhibitors (UI - Green Urea NV^TM), nitrification inhibitors (NI - ENTEC®) and 90 day controlled-release polymer coated urea (PCU - N90®), benchmarking their performance against granular urea under field conditions on Vertosol soils. The N release dynamics of these products were quantified and their potential to increase crop production and reduce N₂O emissions were investigated.

Methods

This paper presents data from a series of field experiments (2018-2020) established at The University of Queensland Gatton campus. These experiments provide an integrated assessment of EEF performance benchmarked against granular urea under irrigated field conditions in southern Queensland, with maize as the test crop.

Maize production

In September 2017, a forage sorghum crop (cv. Pioneer® Super Sweet Sudan) was planted, and sequentially cut and baled twice, to deplete the indigenous soil-N prior to the start of the experiments. The first experiment examined EEFs (UI, NI and PCU) applied to maize at rates of 125 kg N ha^-1 (sub-optimal relative to crop N demand) and 250 kg N ha^-1 (optimal or supra-optimal). To confirm these rates were sub/supra optimal for the maize crop, six rates of urea (0, 62.5, 125, 175, 250 and 300 kg N ha^-1) were tested to develop a reference N response curve, against which EEFs were benchmarked. A second experiment in the same season (adjacent to the first experiment block) compared N₂O emissions from NI and PCU against those from urea at an intermediate rate of 150 kg N ha^-1. Treatments in both experiments were replicated four times in randomised block designs.

Fertilisers were band-applied at a soil depth of 10cm with a spacing of 75cm, immediately prior to sowing of the maize crop. Maize hybrid PAC 606IT was planted at a rate of 70,000 plants ha^-1 at a depth of 6cm, adjacent to fertiliser bands.

N₂O emissions

Nitrous oxide emissions were measured over the entire maize crop season using a fully-automated chamber measuring system (see Grace et al. 2020). Chambers were positioned next to plant rows to account for N₂O emissions integrated across the fertiliser band and adjacent row. Grain yield at 14% moisture was determined at time of harvest, after which soil samples were collected from fertiliser treatments (250 kg N ha^-1) to quantify mineral N remaining in the soil. Soil samples were carefully taken to represent the band, near-band and inter-band locations (see Dang et al. 2021).

Residual N

Soil samples were collected using a 50mm diameter tube pushed by a hydraulic sampling rig and partitioned into 0–0.1, 0.1–0.3, 0.3–0.6 and 0.6–0.9m increments. Samples were dried at 40^oC, ground and analysed for concentrations of nitrate (NO₃^-) and ammonium (NH₄⁺). The total mass of NO₃^- and NH₄⁺ was calculated by multiplying concentrations with bulk density and the depth of the soil layer.

In 2019, the same maize hybrid PAC 606IT was planted over the treatment plots without application of N fertiliser, to assess the residual value of fertiliser N applied during the previous season. The crop was sown at a rate of 65,000 plants ha^-1 at a depth of 6cm into the space between the rows from the previous maize crop. Grain yield at 14% moisture was obtained at time of harvest.

Three-dimensional soil dynamics

Following the maize crop harvest in June 2019, wheat was sown to further draw down the soil N. An experiment was then established in January 2020 to investigate the release and distribution dynamics of N from banded fertilisers (urea, NI and PCU).

Fertiliser products were band-applied at a soil depth of 10cm with a spacing of 75cm on 8 January 2020 at a rate of 150 kg N ha^-1. Soil samples were collected 7, 19, 46, 78 and 145 days after the onset of N release (i.e., once fertiliser bands were wet). During sampling, a profile (approx. 0.4m deep, 0.3m wide) perpendicular to the fertiliser band was exposed in each plot in central bands. Once the fertiliser band was located, a 5cm diameter soil coring tube was inserted into the face of the soil profile, such that the fertiliser band was in the centre of the tube. Additional tubes were then inserted directly above (5cm) and below (10cm) the fertiliser band on all sampling times, with two additional samples collected adjacent to the band for the first three samplings (see Martinez et al. 2021). After field sampling, soils were stored at 4^oC until processing was completed within 48 hours of collection. Samples were manually mixed to ensure homogeneity and analysed for NO₃^-, NH₄⁺, and soil pH. Samples collected on the fertiliser band were also analysed for urea-N.

Results

Maize production

Prior to the 2018 maize crop, NO₃^- (< 1 mg kg^-1) and NH₄⁺ (2-4 mg kg^-1) concentrations were low at all soil depths. Maize yield and grain N content was therefore highly responsive to N fertilisation (Table 1). Grain yield increased significantly with increasing rates of applied urea up to 175 kg N ha^-1, after which only very small increases were observed. A Mitscherlich equation provided the best model fit for the relationship between N rate and yield, with maximum potential yield of 11.9 t ha^-1 and 90% relative yield of 10.7 t ha^-1 at 175 kg N ha^-1 (Fig. 1).

Grain yields from EEF treatments did not improve compared to urea (Figure 1 and Table 1). While an average yield increase of 19% was recorded as N rates increased from 125 to 250 kg N ha^-1, increases were not statistically significant for the NI treatment. Furthermore, the PCU produced significantly lower yields at both the low (8.1 t ha^-1) and high (10.3 t ha^-1) application rates compared to urea (9.7 and 11.4 t ha^-1, respectively; Figure 1).

Table 1. Grain yield for maize grown in 2018 and 2019 at increasing rates of urea (0, 62.5, 125, 175, 250 and 300 kg N ha^-1) and EEFs (UI, NI, PCU) applied at 125 and 250 kg N ha^-1.

Values within a column followed by the same letter are not significantly different at p=0.05. UI = urease inhibitor; NI = nitrification inhibitor; PCU = polymer coated urea

N₂O emissions

The majority of N₂O emissions occurred in the first two months after sowing / fertiliser application, when soil mineral N was highest. The highest daily N₂O fluxes from all treatments occurred after an 80mm rainfall event in February, with a smaller peak also occurring in response to a 50mm irrigation event that was applied to the emission trial area during the postharvest fallow period (Figure 2a). The EEFs produced significantly lower cumulative N₂O fluxes than urea, but differences between urea and PCU were not significant (Figure 2b). The PCU product continued to produce spikes in emissions in response to rainfall events later in the growing season (Figure 2a). Conversely, the NI produced significantly lower cumulative N₂O fluxes than urea and the PCU (Figure 2b). The calculated N₂O emissions factors ranged between a high of 1.2% of applied urea-N to a low of 0.2% for the NI. The PCU had an emission factor of 1.0%, indicative of little improvement compared to the urea.

Figure 2. N₂O emissions from EEFs and urea-N, both applied at 150 kg N ha^-1 during the 2018 growing season (a) pattern and (b) cumulative. 0N = nil N applied; NI = nitrification inhibitor; PCU = polymer coated urea

Residual N

Soil samples collected after the 2018 maize harvest revealed increasing amounts of residual NO₃^- in the soil as fertiliser N rates increased above 125 kg N ha^-1 (Figure 3). Interestingly, while increasing rates of applied urea significantly (p < 0.001) increased the residual mass of NO₃^- in the soil, the mass of NH₄⁺ remained largely unchanged (Figure 3a). The mass of residual NO₃^- in soil was statistically similar for all EEFs except the UI, which had more NO₃^- than urea at an application rate of 250 kg N ha^-1 (Figure 3b). At 250 kg N ha^-1, NH₄⁺ concentrations were greater for the PCU treatment compared to urea.

Figure 3. Residual soil profile mineral N (kg N ha^-1) for (a) increasing rates of applied urea (0-300 kg N ha^-1) and (b) EEFs applied at 250 kg N ha^-1. Vertical bars represent l.s.d. at 5% significance: NO₃-N (solid), NH₄-N (dashed), NS (not significant). UI = urease inhibitor; NI = nitrification inhibitor; PCU = polymer coated urea

Despite wet seasonal conditions (419mm of rainfall and irrigation), NO₃^- was largely confined to the band and near-band (0.05–0.10m) positions, with relatively small amounts found in the mid-row positions (i.e., 0.35–0.40m from the nearest band), with similar patterns observed for urea and EEFs (data not shown). There were noteworthy differences in the depth distribution of NO₃^- for the different N fertilisers (Figure 4), with the mass of NO₃^- in the 0–0.3m for PCU (85%) > NI (70%) > urea (55%) > UI (30%). Relative NO₃^- mass in the 0–0.1m soil depth followed similar patterns, with PCU (33%) > NI (23%) > urea (15%) > UI (10%). Movement of NO₃^- below 0.3m was highest for the UI (70%), with relatively lower NO₃^- mass observed for urea (45%), NI (30%) and PCU (15%).

Figure 4. Relative NO₃^- mass (%) depth distribution for urea in comparison to EEFs (UI, NI and PCU). UI = urease inhibitor; NI = nitrification inhibitor; PCU = polymer coated urea

In the subsequent maize crop grown in 2019, there was no grain yield response to residual N in plots treated with rates up to 125 kg N ha^-1 (applied in 2018; Table 1), which was consistent with the lack of residual N seen in the soil profiles (Fig. 3). However, significant improvements in grain yield were observed as the initial N application rate (2018) increased to 175 kg N ha^-1 and above. Consistent with this, grain yield and N content responses from EEF products was greater at an initial N application rate of 250 kg N ha^-1 compared to the lower rate of application (125 kg N ha^-1). Nevertheless, none of the EEF products resulted in significantly higher grain yields or N contents relative to urea at either low (125 kg N ha^-1) or high (250 kg N ha^-1) rates of application. Notably, the PCU yielded 28.1% lower than urea when both were applied at a rate of 125 kg N ha^-1.

Three-dimensional soil N dynamics around N bands

Irrespective of fertiliser type, NH₄⁺ was the predominant soil mineral N species at 7 days after onset of N release, after which the proportion of mineral N found as NO₃^- increased for both the urea and PCU treatments (Figure 5). Over the first 46 days, the concentration of NH₄⁺ in the PCU treatment was relatively low (compared to urea and NI) and was distributed over a much smaller zone around the fertiliser band. In the NI treatment, a greater proportion of mineral N in the soil remained as NH₄⁺ for the first 46 days, after which NO₃^- became the predominant N form, peaking at 78 days. Nitrate concentrations peaked at 46 days in the urea and PCU treatments and declined considerably after this time (Figure 5).

In-band soil pH was significantly higher in all fertiliser treatments relative to the unfertilised control at 7 days, although that of the PCU was significantly lower than the urea and NI treatments (Figure 5).  Soil pH continued to increase for PCU until 19 days, while only minor changes were evident for granular urea and the NI between 7 and 19 days. Subsequently, the pH of soil for the urea treatment declined up to 46 days, after which no further change was recorded. For the NI treatment, the pH remained higher than the urea and PCU treatments at 46 days, with this difference being significant compared to urea. A minimum pH of 6.5 was reached at 78 days for this N-fertiliser treatment. In comparison, the pH of soil treated with PCU demonstrated a continuing decline from 19 days until at least 145 days. The acidification of soil resulting from nitrification in and around N-fertiliser bands was substantially greater in the PCU treatment compared to urea, with the PCU treatment recording a significantly lower in-band pH (5.0) by 145 days relative to granular urea (pH 6.9; Figure 5).

Figure 5. Vertical and lateral distribution of NH₄⁺ and NO₃^- (mg N kg^-1) and changes in soil pH in and around fertiliser bands of urea and EEFs at 7, 19, 46, 78 and 145 days after onset of N release. The total amount of incident rainfall and irrigation (mm) between sampling occasions is shown.