Can applying nitrogen fertiliser in the fallow period increase fertiliser efficiency in wheat?

Take home messages

• Applying nitrogen (N) during the fallow period did not consistently improve the efficiency of applied N when compared to in-crop applications.
• Applying N during the fallow period often did not result in any penalty when compared to in-crop applications.
• Overwhelmingly, the greatest response in yields and protein was to the rate of N applied.
• Growers should primarily focus on applying the most appropriate rate of N, with the timing a secondary concern.

Background

It is generally accepted that cropping paddocks are declining in fertility, which includes the available nitrogen (N) (Laycock, Bell 2009). This has been exacerbated by a gradual shift to continuous cropping systems, dropping the pasture or lucerne phase, which has been the traditional time for rebuilding soil fertility and N levels, particularly deeper in the soil profile. While our current farming systems tend to include a legume, the N contribution from this phase is rarely enough to support the full requirements of the subsequent crop. Similarly, for those farmers who do use N fertiliser the amount applied is more often than not outweighed by N exported in grains (Norton, 2016).

With this decline in fertility, it would be easy to assume that adding fertiliser N should reliably provide yield and grain quality benefits. Yet growers have often been disappointed in responses to applied N, with grain yields not reflecting what is applied, and/or grain protein being under par. A recent study by Daniel 2018 suggests N recovery in increased yields from applied N can be as low as 10%. This is far short of the often quoted N use efficiency of circa 40-50%. Such low efficiencies would have a significant downward pressure on the economic argument to apply N fertiliser.

So what if the N we are applying is simply in the wrong spot at the wrong time? Dowling (2014) has described this as ‘positional availability’, where active root mass is at distance from mineral N sources for a significant period of the crop growth.

Winter cropping in the northern region relies heavily on fallow rainfall to maximise production. Accumulating moisture in the profile results in the subsequent crop drawing its requirements from deeper in the soil during drier periods of growth. Typically, this occurs in spring, which is also the period of most rapid biomass accumulation, grain fill and protein accumulation.

In our current wheat farming systems, N is typically applied at any time from seeding to stem elongation. This may influence the depth in the soil to which the N will be distributed.

Physical placement of N is generally limited to 0-10 cm when applied with the planting equipment, whilst any topdressing application places N on the soil surface. Any deeper movement of N during the growing period relies on rainfall events. It is therefore easy to imagine that when in-crop rainfall is limited, there is limited opportunity to move nitrates any deeper into the profile.

It is plausible that N in the topsoil may become ‘perched’ or positionally unavailable at times when the crop requires it, and temporarily may not have the ability to use it because it is accessing moisture from deeper in the profile.

Could the conversion of applied N to increased yields be improved by N being distributed deeper in the soil profile? Could this be achieved by applying N fertiliser in the fallow period and allowing the rainfall to transport N deeper into the soil profile as it recharges the soil profile?

Grains Orana Alliance (GOA) has run eleven trials since 2015 investigating this hypothesis. The trials look at whether applying N fertiliser at the beginning of a fallow period has resulted in improved efficiency compared to when it is applied at sowing or top-dressed at Z30 stage of the cereal crop.

Aims

To assess if applying N during the fallow period would result in improved yield and protein responses over that of when N was applied at the more traditional timings of sowing or topdressing after crop begins elongation (Z30+).

Methodology

Eleven trials were established between 2015 and 2017 across the GOA region. Trial sites were selected for low soil N status, confirmed by soil testing.

The trial design was a full factorial design with 32 treatments and three replicates. The treatments included;

• A higher and a lower biomass variety, EGA Gregory and Longreach Lancer respectively
• Four to five N application timings: early fallow, mid fallow*, sowing, split* and topdressing.
• Four N rates: 0, 50, 100 and 200 kg/ha applied as granular urea (46% N)

*Initial trials in 2015 incorporated a split timing with fertiliser N equally divided between early fallow and topdressing. This treatment was replaced by a mid fallow timing in the 2016 & 2017 trials

All N was applied as urea and was drilled for the fallow and sowing N treatments (except the sowing treatments in 2017) using a plot seeder to a depth to ensure reasonable soil cover (typically 5-8cm). All other treatments were broadcast. The 2017 sowing treatments were broadcast ahead of seeding and incorporated by the sowing process. Topdressing treatments were hand broadcast ahead of predicted rain events after the crops had reached Z30. In all cases sufficient incorporation rainfall was received.

A number of treatments were soil tested prior to sowing to assess any differences in N distribution. The two wheat varieties, EGA Gregory and Longreach Lancer, were sown and managed to best management practices to assess if there were varietal differences between responses. Plots were harvested by plot header and assessed for yield, protein, screenings, test weight and moisture. Statistical analysis was provided by the GRDC Statistics for the Australian Grains Industry (SAGI) project team.

An across sites analysis for each of the traits yield, protein and screenings was undertaken using a linear mixed model framework, with fixed effects included for site-year, variety, the rate of applied N, the timing of N application and the respective interactions between the effects. Predictions of the significant effects (α=0.05) were provided from the model as empirical best linear unbiased estimators (eBLUEs). All analyses were conducted using the ASReml package in R.

Error bars within graphs below illustrate the standard error of the predicted value, if the bars overlap between any treatments within a site being compared, the reader can assume there is no statistical difference.

Results and discussion

Impact on grain yield

The range of locations and seasons meant that yields ranged from 2 to 8 t/ha. All sites were N responsive, however the magnitude of response was much greater in 2016 than in 2017 and 2015, which both had much lower rainfall. In 2016, yields increased up to the highest rates of applied N, reflecting the wet season, while in 2015 and 2017, most of the yield response was from the addition of the first 50 kg N/ha, reflecting the dry seasons.

There was no significant interaction with variety with the response to the rate or timing of N applied, as such the results shown in Figure 1 are combined for both varieties.

In the majority of trials there was no yield advantage in applying N at any one rate either at the early or mid-fallow timings when compared to the same rate applied at either sowing or topdressing as illustrated in Figure 1 below.

Figure 1. Yield (t/ha) response to four rates of applied N at four application timings, combined by both varieties.(Note: Split treatment at Burraway and Nyngan only, replacing mid fallow treatment applied at other locations)

There are however a few exceptions-

• Peak Hill 2017- early and mid-timings both outperformed the top dressing timing but not the sowing timing.
• Tullamore 2016- early and mid-timings both outperformed the top dressing timing but again not sowing timings, but only at the highest rate of 200 kg/ha N

There were however and number of cases where the early or mid-fallow timings underperformed when compared to the sowing or top dressing timings also illustrated in Figure 1.

Impact on grain quality

The low protein levels in the treatments where N was not applied confirms the low N status of the sites selected.

Similar to the yield responses, applying N in either of the fallow timings generally did not significantly improve grain protein compared to applying the same N rate at sowing or topdressing.

The one exception was Fifield 2017 where applying 100 or 200 kg N/ha at early or mid-fallow had a significant protein advantage over application at sowing or as a topdressing, where protein levels were increased by 1.2% and 2.4% respectively. In 4 out of the 6 sites there was a protein improvement by the topdressing application of N over that of when the same N rate was applied at other timings, which is not unexpected (note that generally these came with a yield reduction).

Figure 2. Protein (%) response to four rates of applied N and four application timings (yields combined for both varieties) (Note: Split treatment at Burraway and Nyngan only, replacing mid fallow treatment applied at other locations)

Screenings

Similar to the yield and protein responses applying N at either of the fallow timings had little impact on screenings (Figure 3). In 2016 and 2017 the screenings tended to decrease with increasing N rates, where both seasons had relatively ‘soft’ finishes, on the other hand in 2015 screenings tended to increase with increasing N rates, in a season with a dry finish.

Figure 3. Screenings (%) response to four rates of applied N and four application timings (combined results of 2 varieties) (Note: Split treatment at Burraway and Nyngan only, replacing mid fallow treatment applied at other locations).

Discussion

The three years of this trial had very different seasonal conditions. 2015 season had an average start and dry finish, 2016 was a very wet year while 2017 was dry for most of the season with rain and cooler conditions in the spring and late stages of crop maturity, however all seasons had typical fallow rainfall.

With the exception of the Forbes 2017 site, all sites demonstrated N responsiveness with the magnitude of response dictated by the seasonal conditions/in-crop rainfall. However, in terms of application timing, there was no clear or consistent yield or grain quality benefit to application of N in the fallow when compared to application at sowing.

These findings are similar to that found by trials undertaken by Northern Growers Alliance (Daniel et al 2018) where in 10 out of 11 trials they found no advantage of application of N in the fallow compared to application at planting.

One possible reason for this might be that N moves less in the soil than what is considered the current convention. Soil testing by GOA within these trials failed to show any significant amount of either vertical or lateral movement of N despite reasonable amounts of rainfall. Daniel et al (2018) found in their investigations that most of the N applied in the fallow was still in the top 15 cm of soil at planting. A key assumption for this research was that rainfall in the fallow would be sufficient to move N deeper into the profile however this appears not to be the case, and upon reflection there is likely to be close correlation between the relatively low fallow efficiencies experienced in the Central West and the lower than expected N movement.

It is possible that there are still benefits to be gained by increasing the N deeper in the profile, however it is evident that applying N (broadcast or drilled) in the fallow does not achieve this objective. There may be benefit in looking into options for rapid N transfer deeper into the soil, or looking at increasing the overall soil fertility, where the fertiliser applied in one crop, will also have benefit for the subsequent crop.

However, it is worth highlighting that although the fallow application timings did not often result in any benefit in terms of improved yields, they sometimes did result in reduced relative efficiencies when applied at high rates. For example, a small number of cases at Collie 2016 and Tullamore 2016 showed that applying N in the fallow period resulted in less yield than achieved when applied at sowing or topdressing.  The explanation of any potential reasons for these reductions are not detailed in this paper, other than to say that situations were observed where banding high rates of N in the fallow caused penalties in terms of N response.

Conclusion

Overall, applying N fertiliser in the fallow periods did not consistently result in improved N efficiencies in terms of improved yields or grain quality over that of more conventional timings. However, some situations were observed where fallow applications out yielded topdressing timings, however the improved efficiency was tempered by the higher protein levels in the topdressing timings. Conversely there were a small number of cases where fallow application at high rates resulted in poorer outcomes than those of later N application timings.

In almost every case the overwhelming response was to N rate, regardless of its timing. Suggesting growers should focus more on applying the appropriate amount of fertiliser and be less concerned with the timing. If applying N in the fallow period facilitates more convenient input of N into the farming system, growers should not be concerned about doing so.

References

Bell, M.,(2009) Preserving soil fertility, Groundcover (59), GRDC.

Daniel, R., (2017) Nitrogen management in wheat 2016 - method, timing and variety

Daniel, R., Norton, R., Mitchel, A., Baily. L., Kilby, D., and Duric, B. (2018) Nitrogen use (in)efficiency in wheat – key messages from 2014-2017. Proceedings from the Goondiwindi GRDC Grains Research Update March 2018

Dowling, C., (2014) The fundamentals of increasing nitrogen use efficiency (NUE)

Laycock, J., (date unknown) Soil Nitrogen reserves decline.

Norton, R., (2016) Are you running down nitrogen? Do you know?

Acknowledgements

The research undertaken as part of this project is made possible by the significant contributions of growers through both trial cooperation and the support of the GRDC, the author would like to thank them for their continued support. We would also like to thank the team at SAGI for their assistance in interpreting the results and statistical analysis.

Contact details

Maurie Street
Grain Orana Alliance
Dubbo NSW
Mb: 0400 066201
Email: maurie.street@grainorana.com.au

Ben O’Brien
Grain Orana Alliance
Dubbo NSW
Mb: 0409 697 860
Email: ben.obrien@grainorana.com.au

Varieties displaying this symbol beside them are protected under the Plant Breeders Rights Act 1994