An informed approach to phosphorus management in 2022

Take home messages

• Opportunities are available for reformed phosphorus rates under high fertiliser prices, but background knowledge is key.
• Gross margin analysis with phosphorus application rates is sensitive to soil available P, yield potential, fertiliser, and grain prices.
• On phosphorus responsive soil types, return from fertiliser (P) investment is normally greatest and most stable with cereal phases.

Background

Fertiliser prices for phosphorus (P) inputs have more than doubled since those used for the start of the 2021 season and for a three-year rolling price average. Currently, these high fertiliser prices are coupled with high grain prices which offsets potential decreases in partial gross margins, but in the current global scenario, there is high uncertainty if grain prices will hold until the end of 2022. Higher inputs costs will naturally generate a mindset of simply reducing these input rates, but it is important to have background knowledge supporting these decisions so yield returns aren’t compromised. Combined with high fertiliser prices there have been the observations that P replacement programs have resulted in mining of phosphorus in some soil types. This paper aims to outline gross margin scenarios under a range of fertiliser and grain prices which could be vastly different to those set up in previous seasons. Importantly, the gross margin analysis will be performed using a range of different background P levels, soil type characteristics and yield potentials. Identification of likely paddock responsiveness and the variability in that response across the paddock is important. Several tools are available to assist with this determination which will be explained.

Method

Through various research projects across the last 10 years, both Agronomy Solutions and Trengove Consulting have obtained over 50 replicated field trials across the broadacre regions of South Australia, with most of them within the last 5 years (>40). Most of these trials have assessed wheat and barley responses to P applications across a range of soil types x climate years, but we have also been gaining valuable information on pulse P requirements. This dataset is highly valuable to assess gross margin scenarios under a range of conditions and the accuracy of various data layers in predicting P requirements.

Typically, a response parameter is used when assessing grain yield relationships from a range of increasing P applications using relative yield, which compares the yield of the nil treatment with the maximum yield obtained as determined through curve fitting analysis. For this paper, we have used the P rate which is associated with the greatest partial gross margin (PGM) return when factoring in fertiliser prices and returns from grain yields. We have used this dataset to test the accuracy of various data layers in predicting PGM under current conditions and, from the most accurate data layers, looked at the effect of changing fertiliser to grain price ratios for expected 2022 scenarios. Determination of PGM has used recent price trends of MAP at $1250, Wheat (APW) at $400 and Barley (F1) at $295. This field trial dataset is concentrated in the Yorke Peninsula and Mid North regions of South Australia but is applicable to wider regions where soil types vary in alkalinity within paddocks driven by the presence of carbonates. Widespread data from the recently completed GRDC-funded project (PROC9176604) has been utilised to provide an indication of current P levels across the broadacre cropping region of SA and the potential impact of decreasing P rates in response to high fertiliser prices.

Results and discussion

Current soil P levels

Reviewing the large soil test database from PROC9176604 reveals the overall P status of the broadacre cropping regions of SA and VIC. Over 1300 surface samples were collected in 2019 and 2020 with both Colwell P and DGT P levels placed in deficient, marginal, and sufficient categories (Table 1) built on published data (Moody 2007, Mason et al 2010). The PBI value for each site was used to determine a critical Colwell P position and the Colwell P value then compared to this target. DGT P critical levels of 60µg/L for wheat were established in Mason et al. 2010 and have continued to be stable with incorporation of a substantial amount of field trials post 2010. As a summary, there was some discrepancy between the two tests, but over half (52%) of sites were above critical DGT levels and as much as 73% of sites were sufficient in P using Colwell P. It is assumed that sites with soil P levels lower than the critical value or range will result in a yield penalty if no P is applied. Using soil test results to make a P recommendation for the sites sampled shows that there are 83% for Colwell P and 73% for DGT of sites that require <10kg P/ha to maximise yields. The high proportion of sites with low requirements for P provides opportunities in 2022 for identification of soil types where residual P is adequate and lowering P rates won’t generate yield penalties.

Table 1: Soil P test results (Colwell P and DGT P) through the southern broadacre cropping region sampled in 2019 and 2020 placed in deficient, marginal, and sufficient categories with associated determinations of required P rates to maximise yields.

Site soil characteristics driving recent P responses

The intensive field trial dataset produced by Trengove Consulting from 2019 to 2021, where 33 replicated field P response trials have been established on various soil type x NDVI/grain yield zones, is a powerful tool to test previously reported data layers (Colwell P, DGT P) as an accurate guide for P requirements, but also other accessible data layers which might assist P decisions moving forward. Of the 33 sites, 64% recorded non-significant (p>0.05) responses to applied P (Table 2), which is in line with the broader analysis reported in Table 1. Of the 12 responsive sites, at current prices, the average P rate required to maximise PGM was 20kg P/ha, which highlights the continued importance of identification of P responsive soil types. Responsive soil types are characterised by soil pH (CaCl2) between 7.5-7.8, higher PBI values (P retention) driven by the presence of soil carbonate and low comparative NDVI values (Table 2). These broad averages support recent opinions that replacement P programs must incorporate a soil P retention factor which can have a greater influence on residual P availability compared to P offtake through crop yields.

Table 2: Summary of soil characteristics averaged across the 12 significant (p<0.05) responsive P sites compared to 21 nonresponsive sites through Yorke Peninsula and Mid-North regions of SA. PGM was calculated based off MAP at $1250, Wheat (APW) at $400 and Barley (F1) at $295.

Relationships between P rates at maximum PGM and several data layers were used to assess the applicability of soil data layers for accurately determining where reducing P inputs in 2022 may not be a profitable approach. Of the soil P tests alone, DGT P (R2 = 0.71) was superior at splitting apart profitable high P rates at current prices and sites where reduction in P rates would not cause a decrease in PGM. This data set highlights the importance of including PBI with Colwell P interpretation. In this case, we have simply divided the Colwell P value at each site by the PBI value obtained to create an index value which greatly improved Colwell P interpretation (R2 = 0.73 from R2 = 0.44). Using Colwell P alone as a relatively cheap data layer for intensive sampling strategies (for example, grid sampling) will add confusion and should be avoided unless a PBI measure is made at similar intensities.

The most accurate combined data layer to provide a P rate requirement for max PGM was an index of the soil pH and NDVI near GS30 across different zones within a paddock (Figure 1). The index simply divides soil pH with the NDVI normalised to the paddock average outlining that relatively high soil pH coupled with low NDVI values in a zone will generate higher index values and a high likelihood of higher P requirements. In association with higher soil pH coupled with poor early vigour will be the presence of soil carbonate, higher PBI values and lower residual P. The index is yet to be tested on soil types where high PBI is driven by other soil attributes such as Al or Fe, where there is a tendency of soil pH to be <6 in these soils (for example, ferrosols on Kangaroo Island). For these areas, a normalised NDVI index alone could be appropriate.

Figure 1. Relationships between the P rate associated with max PGM for P response trials (2019-2021) with DGT P, Colwell P, Colwell P/PBI and pHnNDVI.

Partial gross margin analysis for fluctuating fertiliser and grain prices

While there is some clarity with fertiliser prices for the 2022 season, there is difficulty in predicting the grain price when it comes to locking in prices towards the end of 2022. At current grain prices, the identification of P responsive sites as outlined previously still pays but what happens if grain prices fall? Using an accurate data layer (DGT P or pHnNDVI), we can present the influence of changing fertiliser and grain prices on optimal P rates for max PGM (Table 3). Based off 2021 fertiliser prices as a comparison and expected 2022 prices, this analysis suggests economic P rates will be slightly less than half of that required in 2021.

Table 3: Sensitivity analysis of optimal P rates required for max PGM (kg/ha) for moving MAP prices at three decile grain prices (1, 5, 9) using either the pHnNDVI index or DGT P as a guide of deficiency (see Figure 1). Grain price deciles from 2010 onwards, source: Mercado.

Opportunities for 2022 – Time of Sowing (TOS)

Recent SAGIT funded project (AS216) outlined the effect of TOS on P requirements through trials established on P responsive sites between 2017 and 2018 due to the prevalence of earlier sowing times. Results outlined that if adequate soil moisture was present at April sowing times, P rates can be reduced dramatically without any impact on yield. This benefit diminished if either low moisture was prevalent in April or sowing times moved to mid-May and beyond, with June sowing times producing uneconomic linear but relatively flat responses. Under high soil moisture and warm temperatures, crop root systems develop effectively and therefore exploration of residual P is high, placing less reliance on fertiliser P inputs. Diffusion rates of P in these conditions are also optimised. Data from Trengove Consulting supports this theory, as the 2020 field trial data set revealed a lower pHnNDVI with optimal P rate relationship (Figure 2). With early May sowing, the high soil moisture present due to favourable early rainfall meant the early TOS effect was present (Table 4). This is a potential option for 2022 if wet conditions in April prevail.

Figure 2 and Table 4: Influence of high rainfall and high soil moisture at the 2020 sites compared to 2019 and 2021 and the impact of lower P requirements at P deficiency indices.

Alternative sources of P

With higher fertiliser P prices, there is a natural tendency to look at alternative sources of P at associated lower costs if they are available. For products that need broadcast application (for example, chicken litter, biosolids) the efficiency of P in these products if applied alone range from approximately 20-50% of the same amount of MAP applied with or below the seed. Due to limited movement of P in soils, particularly those prone to deficiency, the broadcast approach without incorporation can limit root accessibility for that current year. These alternative sources of P should be used in conjunction with an amount of applied P banded. For all other sources of fertiliser P, it is advised to check the chemistries of the product and seek trial information of product performance before use.

Conclusion

High P fertiliser price is currently slightly offset by high grain prices, but with uncertainty if these grain prices will continue into 2022, it is advised to revise P applications in 2022 due to major impacts on optimal P rates required to maximise gross margins. Several data layers are available to assist with identifying areas where P rates can be safely cut back and those that will still return a profit with increased grain yields through adequate P applications.

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 authors would like to thank them for their continued support. The authors also thank SAGIT for their support. We would like to acknowledge the growers involved in SAGIT funded project TC219 and TC221 and SAGIT for funding support for projects AS216, TC219 and TC221.

References

Moody PW (2007) Interpretation of a single-point P buffering index for adjusting critical levels of the Colwell soil P test. Australian Journal of Soil Research 45(1), 55-62.

Mason S, McNeill A, McLaughlin MJ, Zhang H (2010) Prediction of wheat response to an application of phosphorus under field conditions using diffusive gradients in thin-films (DGT) and extraction methods. Plant and Soil 337 (1-2), 243-258.

Contact details

Sean Mason
Agronomy Solutions Pty Ltd
U3/11 Ridley Street, Hindmarsh SA 5007
0422 066 635
sean@agronomysolutions.com.au
@AgSolutionsOz