Nitrogen strategies for N banking

Nitrogen strategies for N banking

Take home messages.

  • Some growers struggle with making informed nitrogen decisions.
  • Two decision support systems, Yield Prophet® and N banks, offer support.
  • Both strategies produce ideal outcomes, however Yield Prophet may be an ideal tool for ‘active’ managers, whereas N banks may be ideal for ‘passive’ managers.
  • In a multi-year N management trial (2018-2023), the YP50% and NB125 strategies provided high gross margins, whilst mitigating environmental N loss through lower (but positive) partial N balances.

Background

Two key factors affect wheat yield potential in Australia, rainfall and nitrogen (N). Outside of received rainfall (Hochman et al. 2017), which can’t be controlled, wheat yields only reach half of their potential, due, in major part, to receiving insufficient N (Hochman and Horan, 2018). Grain yields for other non-legume crops (barley, canola, oats) are also likely to demonstrate similar limitations. Under-fertilisation can also leave the soil with a negative N balance (exporting more N in grain than what is being returned with legumes and fertiliser (see survey by Norton and Elaina vanderMark 2016). This can lead to the mining of soil organic N and associated soil organic matter causing N mineralisation decline, which is estimated to half every 20 to 30 years in continuous cropping systems (Dalal and Mayer 1986, Clarke and Russell 1976, and Heenan et al 2004). Remedying insufficient N supply to grain crops is possible and would have a substantial impact on the profitability of Australian farming enterprises; some data show that reducing N deficiency in wheat would increase Australian wheat yields by 40 per cent (Hochman and Horan 2018, Hunt et al. 2021).

Legumes are commonly incorporated into cropping rotations either as pastures, brown manures, or for grain. However, grain legumes do not leave enough N behind to support the yield of subsequent crops, and not every cropping rotation will incorporate a legume pasture or brown manure. Yields of subsequent crops are therefore reliant on the input of additional N through fertiliser, yet growers can be hesitant to apply higher levels of N fertiliser due to concerns relating to risks to the crop (‘haying off’), environment loss (leaching, volatilisation, denitrification), and economic factors (the cost of urea, for example) (Hunt et al. 2021). Because of these risks, and possibly the emphasis placed on them, growers in southern Australia tend to under-fertilise, most probably due to the uncertainty of rainfall and final yield which determines final N demand.

Nitrogen management decisions are a key area of concern for Australian growers. Over 50 per cent of growers surveyed as part of the GRDC’s RiskWi$e project reported that making decisions relating to N management is challenging, with a subset of those growers classifying the decisions as ‘very difficult’ (Figure 1); this is where Decision Support Systems (DSS) aid growers in identifying strategies for approaching N decisions. The use of a DSS tools facilitates the important step of thinking slow rather than fast in the decision-making process and this improves the probability of favourable outcomes.

Growers have access to two key DSS that can assist with N decisions, Yield Prophet and N banks. Yield Prophet helps match N input to the potential yield for a given season. This predictive management tool relies on APSIM modelling, and as such, the quality of output is reflective of the quality of input; the trade-off being that the cost (or time required) of collecting input data may be too high for some growers. Yield Prophet may best be considered a DSS for growers who want to take a more ‘active’ approach to N management. On the other hand, some growers may prefer to employ a lower time- and money-cost method by using N banks, a simpler, more ‘passive’ strategy that capitalises on year-to-year soil N carry-over to support crop growth.

Survey results summarising grower perceptions of the difficulty associated with various decisions encountered in farming. Survey results provided by Brown et al. (2024) as part of reporting for the GRDC RiskWi$e project.

Figure 1. Survey results summarising grower perceptions of the difficulty associated with various decisions encountered in farming. Survey results provided by Brown et al. (2024) as part of reporting for the GRDC RiskWi$e project.

As part of the GRDC RiskWi$e project BCG is testing N banks and Yield Profit in a network of N management trials to establish the risk and reward profiles of both systems. A multi-year trial was established in 2018 at Curyo, Victoria to compare different N management strategies, including Yield Prophet and N banks. Findings for several of the N management strategies from 2018 to 2023 are summarised here.

Method

N management strategy set up

A multi-year experiment using a randomised complete block design was established in 2018 to evaluate the performance of different N management systems. Four different systems were tested:

  1. Matching N fertiliser to seasonal yield potential (Yield Prophet and Yield Prophet Lite, YP)
  2. Maintaining a base level of fertility using N fertiliser (N banks)
  3. Replacing the amount of N removed in grain each year with fertiliser in the next season replacement)
  4. Applying national average N fertiliser rate (45 kg/ha) each season (national average, NA)

All systems were compared to a nil control which received only starter fertiliser (7 kg N/ha per year in MAP). Within the Yield Prophet and N bank systems there were different treatments targeting different yield potentials (Table 1). In the Yield Prophet treatment before 2021, water limited potential yield was determined at different levels of probability; the amount of N required to achieve these yields was applied, assuming a requirement of 40 kg/ha N per t/ha wheat yield and 80kg/ha N per t/ha canola yield. From 2021 onward, Yield Prophet Lite was used in a similar way. For the N bank treatments there were different target levels of N fertility (N bank targets). N fertiliser rates in these treatments were calculated as the N bank target value minus soil mineral N (kg/ha) measured prior to sowing. Gross margins were calculated based on the 2022 SAGIT Gross Margin Guide (SAGIT 2022). Average N fertiliser cost was based on calculations from historical SAGIT Gross Margin Guides over the six-year period (2018-2023).

Table 1. Nitrogen management systems and treatments used in the experiments.

System

Trt

Description

Nil

Nil

No nitrogen applied other than in starter fertiliser

Replacement

R

Amount of N removed in grain applied as fertiliser N

National average

NA

National average N fertiliser (45kg/ha N) applied each season

Nitrogen banks

(kg/ha N)

NB100

Soil mineral N + fertiliser = 100kg/ha N

NB125

Soil mineral N + fertiliser = 125kg/ha N

NB150

Soil mineral N + fertiliser = 150kg/ha N

Yield Prophet® probabilities based on seasonal finish

YP100%

Lowest yielding finish on record (decile 1 in Yield Prophet® Lite)

YP75%

Lower yielding quartile finish (decile 2-3 in Yield Prophet® Lite)

YP50%

Median finish (decile 4 – 7 in Yield Prophet® Lite)

YP25%

Higher yielding quartile finish (decile 8 – 9 in Yield Prophet® Lite)

N mining experiment

Within the trial outlined above, an N mining experiment was conducted in 2023 to study background N fertility in soil and how this affects N response in subsequent crops. N mining studies were conducted in the western-most sub-plot of treatment plots for the two most extreme treatments, Nil and YP25%. While the trial was sown to lentil in 2023, the N mining sub-plots were sown to canola. The sub-plots were treated with five different rates of N: 0 kg/ha, 50 kg/ha, 100 kg/ha, 150 kg/ha, and 200 kg/ha. This N was applied by hand in the form of urea; the arrangement of sub-plots was randomised. Soil chemistry and canola yield data were collected to ascertain current levels of N and carbon (C), and the impact on crop productivity.

Results and discussion

N management strategies

Grain Yield: Grain yield data over the six trial years for each N management treatment are shown in Figure 2. On average the YP25% and YP50% are the most promising strategies in closing the yield gap (3.1 t/ha and 3.2 t/ha, respectively), and both treatments were significantly higher than all other treatments (p < 0.001, LSD = 0.2), except for the difference between YP25% and NB125 (2.9 t/ha) where the average grain yield for YP25% was higher, but non-significant.

There were no significant differences in mean grain yield between the N bank treatments, NB100, NB125, and NB150, (2.8 t/ha, 2.9 t/ha, 2.9 t/ha, respectively). However, considering the partial N balance for all three treatments, only the NB125 showed a lower positive balance (38 kg/ha) relative to NB100 and NB150 (-32 kg/ha and 81 kg/ha, respectively) over the six-year period. This finding suggests that of the N bank treatments, the NB125 strategy is the best option for this environment.

YP25% showed a relatively high six-year cumulative partial N balance to YP50% (114 k/ha and 45 kg/ha, respectively), indicating a higher chance of N loss under this treatment in this environment. Given the relatively minor difference in grain yield between YP25% and YP50%, and the substantial difference in partial N balance, YP50% is the optimal N strategy for this environment.

Box-and-whisker plots of six-year average grain yield (t/ha) data for each treatment. Lower and upper box thresholds are the 25th and 75th percentiles; the median is depicted by the horizontal line within each box, whereas the mean is depicted by an ‘x’. Whiskers extending from the boxes represent the extreme percentiles.

Figure 2. Grain yield (t/ha). Box-and-whisker plots of six-year average grain yield (t/ha) data for each treatment. Lower and upper box thresholds are the 25th and 75th percentiles; the median is depicted by the horizontal line within each box, whereas the mean is depicted by an ‘x’. Whiskers extending from the boxes represent the extreme percentiles.

Gross margin: The highest average gross margin across the six years was achieved by YP50% ($791), closely followed by YP75%, YP25%, NB125, and NB100 which were all statistically assessed as not different from each other (Figure 3). YP50% and YP75% were the only high return N strategies that avoided a negative gross margin over the six year period and consequently had lower down side risk. Lower gross margins were achieved with the Replacement, NB150, and National Average strategies ($619, $667, and $668, respectively, p < 0.001, LSD $110) (Figure 3).

Taking into consideration grain yield, cumulative partial N balance average over six years, and downside risk, YP50% has been shown to be the optimal N strategy in this environment. Where decision makers are looking for a simpler strategy, NB125 is the best simplified N strategy.

Box-and-whisker plots of six-year average gross margin ($/ha) data for each treatment. Lower and upper box thresholds are the 25th and 75th percentiles; the median is depicted by the horizontal line within each box, whereas the mean is depicted by an ‘x’. Whiskers extending from the boxes represent the extreme percentiles.

Figure 3. Box-and-whisker plots of six-year average gross margin ($/ha) data for each treatment. Lower and upper box thresholds are the 25th and 75th percentiles; the median is depicted by the horizontal line within each box, whereas the mean is depicted by an ‘x’. Whiskers extending from the boxes represent the extreme percentiles.

N mining experiment

Starting N: Soil chemistry samples taken prior to the commencement of the experiment showed the Nil and YP25% treatments contained similar amounts of mineral N (53 kg N/ha and 57 kg N/ha, respectively).

N and C balance: Five-year cumulative N studies (2018-2022) examined in the laboratory showed the Nil treatment was depleted by -134 kg N/ha and soil organic carbon by -3.3 t/ha. Conversely the YP25% treatment had accumulated 93 kg/ha N and 0.12 t/ha C.

2023 canola yield: Canola yield data (Figure 5) in 2023 show that to achieve the same yield across treatments, a significant amount of additional N is required where N has been mined compared to treatments with a positive N balance. For example, to achieve similar yields for a treatment with a large negative N balance compared to those with a positive N balance (Figure 4), an additional amount of approximately 50 kg/ha of N would be required to achieve a similar yield outcome (Figure 5). This estimate is denoted by the difference between the dashed and dotted lines in Figure 5.

The cumulative partial N balance (kg/ha) for each treatment across six years.

Figure 4. The cumulative partial N balance (kg/ha) for each treatment across six years.

The long-term effect of mining soil organic N.  R = 66.8, regression F pr. = <0.001, difference F pr. = 0.006. Dashed and dotted lines represent line of best fit for each treatment.

Figure 5. The long-term effect of mining soil organic N. R = 66.8, regression F pr. = <0.001, difference F pr. = 0.006. Dashed and dotted lines represent line of best fit for each treatment.

Conclusion

The findings from this research show N rate strategies that ran a small net positive partial N balance performed better than strategies that resulted in a negative partial N balance. This highlights the importance of N fertiliser additions being considered with a longer-term perspective. Supporting evidence for this longer-term view is provided in a meta-analysis of 15N farming systems data by Yonk et al (2022), who showed approximately 44 per cent of fertiliser N is taken up by the crop in the year of application and a further 22 per cent is taken up in the two to three years that follow, providing a total crop recover of 66 per cent from the initial application. The main point here is not to focus on the recovery percentages but rather the proof that recovery of fertiliser N can occur up to three years after application.

In this environment, YP50% was the optimal N strategy for maximising crop yield, whilst minimising the downside financial risk and likely N loss to the environment. The NB125 treatment which was considered the best performing N bank strategy had a very good six-year average grain yield, similar gross margin outcome to YP50% and a small positive partial N balance. The advantage of this strategy is the simplified decision-making process which is likely to be appealing to some grain growers.

These findings illustrate that both the Yield Prophet and N bank approaches are sound decision-making systems to maximise returns on a short- and longer-term basis. Grower preference for a particular strategy is likely driven by preference for simplicity verses complexity in N management. Importantly, the N mining experiment results reinforce that N mining can reduce yield within five years, and that growers cannot avoid paying the price for under-fertilising; the price will either be paid as you go where a slightly positive partial N balance is provided or in the future where mining occurs.

Acknowledgements

This research has been funded by the GRDC since 2022, initially through the National Grower Network project, N banking strategies to manage variable and unpredictable nitrogen demand in the MRZ of the Southern Region (PROC-9176566), and currently as part of the National RiskWi$e project (PROC-9176569). From 2018 to 2019, the research was funded by La Trobe University funded through the Securing Food, Water and the Environment Research Focus Area. Funding was subsequently provided by the Mallee Catchment Management Authority from 2019 to 2022, through the Australian Government’s National Landcare Program and GRDC, through National Grower Network. We thank Paul Barclay for hosting the experiment.

Useful resources

Basso, B., Shuai, G., Zhang, J. et al. Yield stability analysis reveals sources of large-scale nitrogen loss from the US Midwest. Sci Rep 9, 5774 (2019).

Brown, B, Azeem, M, Llewellyn, R (2024) An overview of Australian grain grower risky decision processes.

Clarke AL and Russell JS (1977) Crop Sequential Practices. In: J.S. Russell & E.L. Greacen (Eds): Soil factors in crop production in a semi-arid environment. University of Queensland Press, St Lucia.

Dalal RC Mayer RJ (1986) Long term trends in fertility of soils under continuous cultivation and cereal cropping in southern Queensland. II. Total organic carbon and its rate of loss from the soil profile. Soil Research 24, 281-292.

Heenan DP, Chan KY, Knight PG, (2004) Long-term impact of rotation, tillage and stubble management on the loss of soil organic carbon and nitrogen from a Chromic Luvisol, Soil and Tillage Research, Volume 76, Issue 1, pages 59-68,.

Hochman, Z, Horan, H (2018) Causes of wheat yield gaps and opportunities to advance the water-limited yield frontier in Australia. Field Crops Research 228, 20-30.

Hunt, JR, Kirkegaard, JA, Maddern, K, Murray, J (2021) Strategies for long term management of N across farming systems. GRDC Available at Strategies for Long Term N Management.

Norton R and vanderMark E (2016). Nitrogen performance indicators on southern Australian grain farms. Proceedings of the 2016 International Nitrogen Initiative Conference, "Solutions to improve nitrogen use efficiency for the world", 4 –8 December 2016, Melbourne, Australia.

SAGIT (2022) 2022 Gross Margin Guide

Vonk, W. J., Hijbeek, R., Glendining, M. J., Powlson, D. S., Bhogal, A.,Merbach, I., Silva, J. V., Poffenbarger, H. J., Dhillon, J., Sieling, K., & ten Berge, H. F. M. (2022). The legacy effect of synthetic N fertiliser. European Journal of Soil Science, 73(3), e13238.

Contact details

Yolanda Plowman
73 Cumming Avenue, Birchip VIC 3487
0447 755 312
Yolanda.plowman@bcg.org.au
@YolandaPlowman

GRDC Project Code: BWD2303-004BGX,