Efficient use of nitrogen fertiliser in Riverina irrigated cropping – could mid-row banding help?

Take home messages

• Placing fertiliser into a concentrated mid-row band at sowing can offer a low risk alternative for meeting crop nitrogen (N) requirements.
• Losses via volatilisation, leaching and denitrification will potentially be minimised as urea will be preserved in the ammonium form which is not vulnerable to these losses.
• A large proportion of the mid-row banded N was preserved as ammonium until 91 days after sowing (DAS), successfully preserving N through the waterlogging event, even though almost all nitrate was denitrified.
• When waterlogged, mid-row banded N achieved similar grain yield, grain N and apparent nitrogen recovery efficiency (ANRE) as topdressing after the waterlogging event, but less than a particularly efficient topdressing that occurred before the water-logging event. When dry, the treatment effect was the same.

Background

Irrigated winter crops in the Riverina need a substantial nitrogen (N) supply for high, water productive yields. A 5t/ha wheat crop needs a supply of 240kg N/ha. For continuous cropping systems, often more than 150kg N/ha of that must come from fertiliser nitrogen (N).

In wet winters the saturated soil profile of a pre-watered or rice stubble crop increases the risk of prolonged periods of waterlogging in winter from relatively modest rainfall events. When N is in the nitrate form it can quickly be leached below the root zone or denitrified as nitrous oxide gas, resulting in a large loss of fertiliser N (Zerulla et al., 2001). Denitrification is particularly a risk on clay soils in the Riverina, where soil internal drainage rates are low. As such, waterlogging will be a strong focus of this trial.  

Placing fertiliser into a concentrated mid-row band at sowing can offer a low risk alternative for meeting the crops N requirements. Much of the N can be preserved in the ammonium form, as nitrification is inhibited when ammonium concentration reaches 3000ppm or when the soil pH >8 (Wetselaar et al., 1973). Angus et al. (2014) found that high concentrations of banded N were toxic to microbes responsible for nitrification. Brar (2013) recorded that banding also resulted in a corresponding increase in soil pH, further inhibiting nitrification. As a result, banding should provide a slow-release of N throughout the growing season. Losses via volatilisation, leaching and denitrification will potentially be minimised as urea will be preserved in the ammonium form which is not vulnerable to these losses. In addition, roots have been shown to enclose the band of fertiliser, making them well-placed to intercept mobile nitrate once it becomes available (Wetselaar et al., 1972; Angus et al., 2014; Sandral et al., 2017).

Method

The trial was conducted in a rice bay that was cropped with a 5t/ha barley crop in 2016, but not pre-watered. Wheat (var. Mace) was sown on the 16 of May at 85kg/ha to a depth of 3 to 5cm, with 150kg/ha MAP + 1.0% Zinc. A Bettinson disc drill was used with 18cm row spacings. Every third row was blocked to seed and MAP, and only urea was placed in this row (at about 7cm depth) for the 104N mid-row banding (MRB) and 150N MRB treatments. No fertiliser was placed in the mid-row band for the other treatments.

The early topdressing treatment was applied at GS31 on the 28 July, immediately prior to the waterlogging treatments were applied on the 29 July. The late topdressing treatment was applied between GS32-33 on the 16 of August prior to a small rainfall event.

The site was not pre-watered. Levees were constructed after sowing, to exclude water from the ‘Dry’ plots during the winter waterlogging event. Waterlogged treatments were irrigated for 10 days. Two spring irrigations occurred on the 19 September and 20 October to set up the crop for the nominated yield potential of 5t/ha.

A split plot design was used, with waterlogging treatments in the main plots and N treatments in the sub-plots. There were four replicates. Each plot was approximately 40m long and 8m wide.

Results

Yields of all treatments were modest, driven primarily by a small number of grains per head. This is not surprising, as neither the dry nor wet irrigation treatments had ideal conditions. The dry treatment suffered from substantial water stress before the first irrigation on 19 September as the paddock was not pre-irrigated. The wet treatment was exposed to 10 days’ inundation beginning 29 July.

A large proportion of the mid-row banded N was preserved as ammonium through the waterlogging event, even though almost all nitrate was denitrified (Figure 1). About half of the initial concentration was still present at 91 days after sowing (DAS) after the waterlogging event, but only 5 to 8% of the mid-row banded N was available on 3 October, 131 DAS, so the bulk of it was exhausted by flowering in late-September, when the number of grains per head was set.

The mid-row ammonium concentration reduced a similar amount from before to after the waterlogging event, for both the water-logged and unwater-logged treatment, suggesting that little or no ammonium was lost due to waterlogging.

Figure 1. The concentration of N in the sampled mid-row band, present as either NH4+ or NO3- (mg/kg) at four different times of sampling, for two mid-row banded N treatments (104N MRB and 150N MRB), for two waterlogging treatments after the imposition of the water-logging event, Moulamein 2017. Columns with the same letter within each time of measurement were not significantly different (P<0.05).

Both MRB treatments had similar apparent nitrogen recovery efficiency (ANRE) to late topdressed treatments (NTL), of 18 to 21% but less than early topdressed (NTE) treatments when waterlogged. In a non-waterlogged environment MRB treatments had an ANRE of 19 to 31% which reflects previously reported efficiencies of 21 to 28% in an Australian high rainfall environment (Angus et al.,2014) and 24 to 31% in south-east China (Chen, 2016). The topdressed treatments had an ANRE of 30 to 41% which is above the reported rates of 20% in Australia and 27 to 30% in China. The trial topdressing figures are more similar to the global estimate of 33 to 34% reported by Ladha et al. (2005) and Raun & Johnson (1999). Waterlogging significantly reduced treatment efficiencies, resulting in an ANRE of 18 to 22% for MRB treatments and 21 to 36% for topdressed treatments.

The waterlogging event both reduced chlorophyll content of plants (CCCI) and a normalised measure of the canopy chlorophyll and biomass (NDRE) and also eliminated all N treatment effects (comparing CCCI 2 wet with dry and NDRE 2 wet with dry, immediately after the waterlogging event). They were all the same immediately after the waterlogging event (CCCI 2 and NDRE 2). The topdressed N treatments recovered better after the waterlogging event than the mid-row banded N treatments (comparing ∆CCCI and ∆NDRE, which is the change from measurement 2 to measurement 3).

It appears that both topdressed treatments were particularly efficient, as ANRE was relatively high. With reference to the high ANRE of the NTE treatment, we suspect that topdressed N was preserved as ammonium during the subsequent waterlogging and hence was not lost via denitrification.

Table 1. The grain yield (mt/ha), grain N (kg/ha), ANRE (%), yield components, CCCI and NDRE and change in CCCI and NDRE (∆CCCI and ∆NDRE) after the waterlogging event, for five different N treatments (15N, 150NTE, 150NTL, 104NMRB and 150NMRB) and two irrigation treatments (dry and wet), Moulamein 2017.

Conclusions

Mid row banding preserved much of the N as ammonium until 91 DAS, successfully preserving the N from a waterlogging event. Only 5 to 8% of the mid-row N was in the mid-row by 131 DAS, just after flowering. When waterlogged, mid-row banded N achieved similar grain yield, grain N and ANRE as topdressing after the waterlogging event, but less than an efficient topdressing before the water-logging event. When dry, the treatment effect was the same.

Wheat with mid-row banded N showed less recovery after the water-logging event, and had fewer grains per head than either of the topdressed treatments. Hence, the N supply from MRB N to flowering and beyond appeared less than that of topdressed N.

Mid-row banding of N showed an acceptable response in both waterlogged and non-waterlogged conditions, but similar or less than the highly effective N topdressing treatments.

References

Angus, JF, Gupta, VVSR, Pitson, GD, and Good, AJ (2014) Effects of banded ammonia and urea fertiliser on soil properties and the growth and yield of wheat, Crop and Pasture Science, 65, 337-352

Brar, N. K. (2013). Nitrogen management in wheat sown into rice crop residues. Wagga Wagga: Charles Sturt University.

Chen, Z., Wang, H., Liu, X., Liu, Y., Gao, S., & Zhou, J. (2016). The effect of N fertilizer placement on the fate of urea-15N and yield of winter wheat in southeast China . PLoS ONE, 11(4): e0153701.

Ladha, J. K., Pathak, H., Krupnik, T. J., Six, J., & van Kessel, C. (2005). Efficiency of fertilizer nitrogen. Advances in Agronomy, 87:86-156.

Raun, W. R., & Johnson, G. V. (1999). Improving nitrogen use efficiency for cereal production. Agronomy Journal, 91:357-363.

Sandral, G. A., Tavakkoli, E., Harris, F., Koetz, E., & Angus, J. (2017). Improving nitrogen fertiliser use efficiency in wheat using mid-row banding. Ballarat: Agronomy Australia Conference.

Wetselaar R, Passioura JB, Rose DA, Jakobsen P (1973) Banding nitrogen fertilizers in soil: principles and practice. Chimie & Industrie – Genie Chimeque 106: 567–572.

Zerulla W, Barth T, Dressel J, Erhardt K, Horchler von Locquenghien K, Pasda G, Rädle M, Wissemeier AH (2001) 3,4-Dimethylpyrazolphosphat (DMPP)—a new nitrification inhibitor for agriculture and horticulture Biol. Fertil. Soils, 34: 79–84.

Acknowledgements

The support of GRDC, Murray LLS and Murray Irrigation is acknowledged for the conduct of the 2017-18 experiments. The extensive services of CeRRF and valuable input of John Angus, Sam North and John Blackman is acknowledged for the 2015 - 2018 experiments.

Contact details

Leigh Vial
‘North Dale’, Moulamein, NSW, 2733
0403 489848
leigh.k.vial@gmail.com
@leigh_vial

Laura Kaylock
20 Noorong Street, Barham, NSW, 2732
0431 236045
laura.kaylock@wmlig.org
@WesternMurray