Previous PageTable Of ContentsNext Page

Variable response to phosphorous fertilisers in wheat and chickpea crops in central Queensland

Richard Routley1, Graham Spackman2 and Maurice Conway1

1 DPI&F, LMB 6, Emerald, Qld, 4720, Email richard.routley@dpi.qld.gov.au, maurice.conway@dpi.qld.gov.au
2 Graham Spackman & Associates, 85 Edgerton St, Emerald, Qld, 4720, Email spackman@tpg.com.au

Abstract

Many central Queensland (Qld) grain producers routinely apply low rates of phosphorus (P) fertiliser with the seed at planting to winter grain crops, even though observed yield responses to this practice have been variable. As fertiliser prices rise, it is increasingly important that starter P application is targeted to situations where the chance of positive economic response is high. Fourteen on-farm P response field trials were established in wheat and chickpea crops throughout the central highlands in the winter of 2006. Phosphorus treatments were nil, low (farmer P rate, mean 3.9 kg P/ha) and high (double farmer rate, mean 7.8 kg P/ha).

Mean yield response to applied P was greater in chickpea (8.3% low rate and 14.1% high rate) than in wheat (4.3% and 6.9%), although the magnitude of response at individual sites varied markedly. The relationship between yield response and soil Colwell P level and fallow length was examined and in general, responses were greater for low soil Colwell P levels and long fallows. There were a number of cases however, where little or no yield response was observed where Colwell P was < 10 mg P/kg. Results suggest that more effective indicators of the likelihood of crop response to P fertiliser applications need to be developed for this environment.

Introduction

Yield responses to phosphorus (P) fertilisers have been well documented across the northern cereal growing region of Australia (Strong et al. 1978). A number of factors interact to determine the likelihood and magnitude of grain yield response to applied P, including soil plant-available P supply, soil P buffering capacity, seasonal conditions, yield potential, crop type and fallow length (Holford 1997).

Despite well known limitations (Holford 1997), soil available P concentration measured using the Colwell bicarbonate extraction method in the 0-10 cm soil layer is the most commonly used indicator of potential P responsiveness, on the alkaline vertosols that make up a large proportion of the grain cropping soils of the region. Critical soil Colwell P concentrations (i.e. levels below which a response to P fertiliser application is likely to occur) are generally considered to be between 10 and 30 mg P/kg on these soils, depending on crop type and yield potential (Moody and Bollard 1999).

Chickpea is generally considered to be highly efficient in extracting P from low P soils due to the production of acidic root exudates which dissolve insoluble soil P. For this reason lower critical Colwell P levels are often used with chickpea than with wheat and other cereals. Responsiveness to applied P fertilisers is often greater after long fallow periods, an effect attributed to a decline in vesicular arbuscular mycorrhiza (VAM) levels, and this response is likely to be greater in crops with a high VAM dependence such as chickpea (Thompson 1993).

Despite the range of variables affecting response to P fertiliser application and the unreliability of observed grain yield responses to P application (G Spackman pers. comm.), the application of low rates of P fertiliser to cereal crops, generally applied with the seed at planting, is a common farm practice in the central Qld cropping region. A recent survey showed that 62% of graingrowers in the region used P fertiliser and that the mean P application rate to wheat crops was 3.8 kg P/ha (Mowat et al. 2007). Many producers have indicated that they consider P fertiliser application to be relatively cheap ‘insurance’ against the possibility of yield penalties associated with P deficiencies.

As the cost of P fertiliser increases however, it becomes increasingly important to develop reliable indicators of the likelihood of achieving economic responses to P fertiliser application. This paper reports the results of a preliminary study designed to quantify the yield response and economic benefits of P fertiliser application to wheat and chickpea crops in central Qld, and to examine the relationship between of a range of commonly used indicators of potential P responsiveness and actual P response.

Methods

Phosphorous fertiliser rate trials were established at 14 on-farm sites in the central highlands region of central Qld during the winter cropping season of 2006. Treatments at each site were nil P applied (P0), the ‘normal’ rate of starter P fertiliser used by that farmer (P1) and double that rate (P2). The P rates used at each site are included in Table 1. A randomised complete block design with three replicates was used at each site. Plot size was 1 x planter width (usually 12 m) x approx 400 m in length.

Seven sites were planted to wheat and seven to chickpea during the period 3 May to 1 Jun 2006. Planting and all other operations were carried out by the cooperating farmers using commercial equipment and normal management practices. In all cases, P fertiliser was applied in close proximity to the seed during the planting operation. Grain yield was determined by harvesting one commercial header width (~10m) from the centre of each plot and determining grain weight using a weigh bin. Grain weight was corrected to 12.5% moisture content.

Each site was sampled prior to sowing by taking ten 50 mm soil cores to a depth of 200 mm which were composited and sub-sampled for determination of bicarbonate extractable P (Colwell 1963) in the 0-10 and 10–20 cm depth layers.

Results

A statistically significant (P<0.05) increase in grain yield in response to application of P was observed at three of the seven wheat sites (Table 1). Across all wheat sites, the mean improvement in grain yield for the P1 treatment was 4.3%, and 6.9% for the P2 treatment. Statistically significant grain yield responses were observed at five of the seven chickpea sites. Across all chickpea sites, the mean improvement in grain yield for the P1 treatment was 8.3%, and 14.1 % for the P2 treatment

Table 1. Soil bicarbonate-extractable P (Bicarb P), grain yield, % yield response and margin over fertiliser cost (MOFC) for each P treatment (P0=nil P, P1 = low P rate, P2=high P rate),.

(Yield values within a site followed by different letters are significantly different at P<0.05)

Site

Crop

Bicarb P (mg/kg)

P applied
(kgP/ha)

Grain Yield

(t/ha)

% Yield response

MOFC#

($/ha)

   

0-10

10-20

P1

P2

P0

P1

P2

lsd

P1

P2

P1

P2

1

W

9.0

<5

6.70

13.40

1.74 a

2.04 b

1.91 b

0.14

17.0%

9.9%

49

2

2

W

5.6

<5

3.55

7.10

2.69

2.65

2.73

0.31

-1.7%

1.3%

(20)

(12)

3

W

16.0

8.1

3.90

7.80

2.54

2.45

2.60

0.26

-3.6%

2.6%

(32)

(7)

4

W

8.4

<5

4.15

8.31

2.88

2.91

3.03

0.31

1.0%

5.0%

(5)

10

5

W

6.5

<5

4.10

8.19

2.75 a

2.90 ab

3.10 b

0.35

5.4%

12.5%

23

56

6

W

17.0

8.1

5.03

10.06

0.93

0.99

0.99

0.06

6.1%

6.7%

(1)

(14)

7

W

6.5

<5

3.40

6.80

2.51 a

2.66 b

2.76 c

0.08

6.0%

10.3%

25

40

Mean

  

9.9

4.1

4.40

8.81

2.29

2.37

2.45

  

4.3%

6.9%

6

11

8

CP

5.9

<5

3.55

7.10

1.62

1.70

1.70

0.19

4.6%

4.9%

35

28

9

CP

5.8

<5

4.10

8.19

1.20 a

1.48 b

1.72 c

0.20

23.6%

43.9%

158

293

10

CP

16.0

8.1

4.38

8.76

1.27 a

1.32ab

1.39 b

0.09

3.9%

9.0%

18

44

11

CP

9.1

<5

2.19

4.38

1.12

1.14

1.02

0.23

1.5%

-9.1%

4

(73)

12

CP

7.6

<5

3.12

6.24

1.27 a

1.30ab

1.34 b

0.04

2.5%

6.0%

10

28

13

CP

7.6

 <5

3.12

6.24

1.54 a

1.74 b

2.08 c

0.08

12.8%

34.7%

109

304

14

CP

16.0

<5

3.40

6.80

1.59 a

1.74 b

1.74 b

0.11

9.0%

9.0%

77

67

Mean

  

9.7

3.3

3.41

6.82

1.37

1.49

1.57

  

8.3%

14.1%

59

99

# Margin Over Fertiliser Cost = Value of grain yield increase - Cost of fertiliser applied. Calculated using grain prices (Wheat: $230/tonne, Chickpea: $600/tonne) and fertiliser prices ($2.80/kg P) applying at the time.

The relationship between soil bicarbonate-extractable P levels, P application rate and fallow length, and grain yield in the fertilised treatments relative to the yield of the untreated controls, is shown in Figure 1. The only statistically significant relationship (P<0.05) was a linear relationship between fallow length and relative yield of chickpea, where fallow length explained 47.2% of the variation in relative yield of the fertilised treatments.

a. Soil Colwell P

b. P rate

Fig 1. Effect of various parameters on Relative Grain Yield (yield as a proportion of P0 treatment) for wheat (solid markers & regression lines) and chickpea (hollow markers and broken regression line)

 

c. Fallow length

Discussion

For both wheat and chickpea, only minor yield responses to applied P were observed at sites where soil Colwell P (0-10 cm) levels were high (>15 mg P/kg), while at low Colwell P sites, yield response ranged from slightly negative to highly positive (up to 45%). This poor relationship between soil Colwell P (0-10 cm) level and the magnitude of any yield response to applied P, particularly in the case of wheat, suggests that Colwell P (0-10 cm) is an unreliable indicator of crop P response. It is interesting to note that in all three wheat crops where yield response to P was observed, that Colwell P levels in the 10 – 20 cm soil layer were very low (<5mg P/kg, the limit of detection). We suggest that there is a need to develop critical soil test values for yield response based on soil P to at least 20 cm depth, particularly as deep planting practices often used result in seed being placed at >10 cm depth and to account for the P stratification associated with zero tillage systems.

The generally higher yield and margin over fertiliser cost (MOFC) responses to P application observed in the chickpea crops, compared to the wheat crops, suggests that attention to P fertility management in chickpea is necessary to optimise yields and profitability. In particular, high yield responses were observed in all chickpea crops grown after fallows of greater than 6 months in duration, supporting the current understanding of the VAM dependency of chickpea and its impact on P uptake.

The lack of responsiveness of grain yield to P application rate suggests that the low application rates (compared with crop removal rates) that are currently in common use are sufficient given the placement practices used (close to the seed at planting). Other research suggests that the stranding of immobile P in dry soil close to the surface will limit responses to higher P rates applied in this manner (M Bell, pers. comm.).

This preliminary study has demonstrated that a significant return on investment can be achieved from P fertiliser application to wheat and chickpea in central Qld and has helped confirm our understanding of some of the factors that contribute to achieving reliable responses. However the variability in responses that were observed and the poor relationship between response and the most commonly used indicator of potential response (Colwell P 0-10cm) indicates a need to develop more reliable indicators of responsiveness. Further research is in progress to develop these indicators and to investigate application practices (including timing of application and placement) that may improve responsiveness in the central Qld environment.

References

Holford ICR (1997) Soil phosphorus: its measurement, and its uptake by plants. Australian Journal of Soil Research 53:227-39

Moody, PW and Bollard, MDA (1999). Phosphorus. In Soil analysis, an interpretation manual. Eds KI Peverill, LA Sparrow and DJ Reuter. CSIRO Publishing, Melbourne.

Mowat J, Routley RA and Shepherd, A (2007). CQSFS2 – An evaluation of project achievements, benefits and outcomes. DPI&F. Emerald.

Strong, WM, Best EK and Cooper JE (1978). Yield response of dryland wheat to nitrogen and phosphorus fertilisers. Qld Wheat Research Institute Biennial Report, 1976 – 78. DPI&F. Toowoomba.

Thompson, J (1993). What is the potential for management of mycorrhizas in agriculture? In Management of Mycorrhizas in Agriculture, Horticulture and Forestry. Eds AD Robson, LK Abbott and N Malajczuk. Kluwer, Dordecht.

Previous PageTop Of PageNext Page

...

Search Site