Grains Research and Development

Date: 30.06.2014

The most profitable phosphorus according to science

Author: Bob Freebairn

 Month 2009
Table 1 Monthly rainfall (mm)
 January  14  10  8  35
 February  37  163  70  179
 March  38  42  83  128
 April  60  49  25  37
 May  5  45  34  60
 June  83  32  12  44
 July  30  58  17  43
 August  10  47  57  15
 September  13  50  23  33
 October  13  67  56  7
 November  20  85  139  18
 December  75  98  102  14
 Stored PAW mma
 84  140  202  147
 In-crop rainfall
 154  300  199  303
 Effective rainfall
 238  440  401  350
   a  Stored moisture measured at sowing via five grainmetric soil cores

Switching to alternatives to conventional fertilisers such as monoammonium phosphate (MAP) to correct phosphorus deficiency is unlikely to deliver any hoped-for savings, according to recent research by the New South Wales Department of Primary Industries (DPI).

The research was part of the GRDC-funded ‘More profit from crop nutrition’ program and was led by Colin McMaster, a R&D agronomist at the NSW DPI in Cowra.

Trials were undertaken on ‘Duran’, a property at Gunningbland (near Forbes), NSW, managed by Jim Cronin. The data was reviewed by Barry Haskins, from AgGrow Agronomy and Research, and Dr Simon Speirs and Rohan Brill, from the NSW DPI. Dr Remy van de Ven, from the NSW DPI, provided biometrical support.

‘Duran’ is considered a typical central-west NSW environment, with an average annual rainfall of about 525 millimetres and each month receiving about 44mm on average (Table 1).

Research study

Research from 2009 to 2011 tested a range of phosphate fertiliser products (Table 2) over three consecutive winter crops: two sown to wheat and one to canola. The fourth season relied on residual phosphorus with no fertiliser phosphorus applied in 2012.

 Phosphorous source
 % phosphorous
Table 2 Fertiliser product phosphorus content
 High-analysis – MAP (granular)
 Rock phosphate (granular)
 Phosphoric acid – Easy NP (liquid)
 Polyphosphate (liquid)
   Note: Basal applications of nitrogen applied as urea
 to balance all treatments

Nitrogen was added where necessary to ensure each product was assessed for phosphorus responsiveness on equal terms.

The trials were conducted partly in response to grower interest in liquid phosphorus fertilisers, which have become popular on calcareous, alkaline soils in South Australia. On these soils liquid phosphorus has been shown to be more efficiently used by crops than phosphorus from solid forms such as MAP.

However, most of the NSW cropping belt does not have calcareous soils and pH can vary from alkaline to quite acidic. The benefits of liquid versus granular phosphorus fertilisers had not been adequately assessed in central NSW on such soils prior to this research.

The trial site was a grey vertosol (clay soil throughout the profile and self-mulching) with only low levels of free lime.

The soil phosphorus level at the site was 15 parts per million Colwell (significantly deficient) with a low phosphorus buffering index (PBI) of 106ppm. The low PBI indicates minimal phosphorus tie-up, with unused fertiliser phosphorus returning to the available soil phosphorus pool.


Grain yield

Grain yield increased as the MAP fertiliser rate increased. The five, 10 and 20 kilograms-per-hectare phosphorus rates increased grain yield over the duration of the research on average by 329, 469 and 631kg/ha per year respectively (Figure 1).

A graph showing average grain yield response of fertiliser treatments over a three-year period

Figure 1 Average grain yield response of fertiliser treatments over three-year period.

Averaged across three years there was no significant grain yield gain from rock phosphate fertiliser.

Biological products such as ‘microbe friendly’ seed treatments and ‘biological inoculants’ were also assessed when used in conjunction with rock phosphate. They are commonly advocated for use with rock phosphate, but this study found that rock phosphate with these additives when applied at equivalent phosphorus rates did not match the yield responses achieved from MAP (Figure 2).

A graph showing average grain yield response over a three-year period comparing MAP to rock phosphate and biological inoculants (systems trial)

Figure 2 Average grain yield response over three-year period comparing MAP to rock phosphate and biological inoculants ('systems trial').

Liquid forms of phosphate performed well in the research, with responses similar to MAP at the various phosphate rates assessed. But they retail at much higher cost for kilograms per hectare of phosphorus supplied, therefore reducing their comparative profitability.

Residual soil phosphorus (Colwell)

Residual phosphorus levels gradually increased as the rate of MAP (phosphorus rate) application increased: five, 10 and 20kg/ha of phosphorus increased residual levels by 1.8, 2.6 and 6.2 milligrams/kg respectively.

Liquid phosphate fertilisers assessed had similar residual response to MAP treatment. Rock phosphate did not increase soil phosphorus levels.

Mr McMaster concludes that soil phosphorus levels are unlikely to significantly increase, and could decline over time if fertiliser rates are reduced in order to afford more expensive forms of phosphorus.

If crop removal of phosphorus is greater than fertiliser phosphorus input – irrespective of the fertiliser product – soil phosphorus levels are almost certain to decline.

Higher rates of applied phosphorus from using MAP and liquid phosphate fertilisers result in a gradual build-up of soil phosphorus levels, indicating future value to crops or pastures.

In the research study, rock phosphate used at the various rates did not increase soil phosphorus levels.

Yield response

Grain yield was significantly increased from the accumulation of three years of MAP and liquid phosphorus fertilisers. The highest yield was achieved where MAP had been used at 20kg/ha over the previous three years, delivering an increase of 574kg/ha over the nil phosphorus treatment.

Five and 10kg/ha of phosphorus applied as MAP produced yield benefits of 346 and 312kg/ha respectively over nil phosphorus treatment.

As mentioned, there was no significant yield benefit with the rock phosphate fertiliser.

Photo of man crouching in field

Colin McMaster

PHOTO: Bob Freebairn


The main conclusion from the research is to regularly monitor soil phosphorus levels with an overall strategy to base fertiliser use levels on available phosphorus and likely removal rates in grain harvested off the paddock.

However, when higher rates of phosphorus are applied than what is used by the crop – commonly when 20kg/ha of phosphorus in forms such as MAP is applied – soil phosphorus levels can build significantly for future crop use.

Mr McMaster says the potential of a soil to lock up phosphorus is estimated by the PBI, a test used with standard soil analysis. The majority of soils in central and southern NSW (as well as many other cropping areas throughout Australia) have low PBI values, indicating that much of the applied phosphorus will be plant available over time, as shown in the study.

A combination of paddock history, crop type, sowing date (early or late sown) and soil test results have proven to be beneficial tools in predicting individual paddock responsiveness to applied available phosphorus.

Over the four years of research on the phosphorus-responsive site MAP proved the most profitable fertiliser, returning increased profit of $380/ha, $430/ha and $510/ha for five, 10 and 20kg/ha of applied phosphorus respectively.

More information:

Colin McMaster,
0427 940 847,

Watch a related video from GroundCover TV, Episode 13: GCTV13: Phosphorus Uptake


Close eye on research lifts WUE and saves the sheep


Cover crops and drains help win a dry argument

GRDC Project Code DAN00168

Region South, North