Home
Publications
Asa
2012 16th AAC
Nutrition
Why Crops Fail To Produce Grain Yield Responses To P...

Why crops fail to produce grain yield responses to phosphorus fertilisers?

RD Armstrong¹, K Dunsford¹, MJ McLaughlin², S Mason² and T McBeath³

^1. Future Farming Systems Research, Department of Primary Industries Victoria, Horsham, Vic 3400.
Email: Roger.Armstrong@dpi.vic.gov.au^2. The University of Adelaide, PMB 1, Waite Campus, Glen Osmond, SA 5064, Australia. Email: Mike.McLachlan@csiro.au ^3. CSIRO Sustainable Agriculture Flagship, CSIRO Ecosystem Sciences, PMB 2, Glen Osmond, SA 5064, Australia Email: T.McBeath@csiro.au

Abstract

Phosphorus fertiliser is widely applied to grain crops growing on alkaline soils throughout Australia. However widespread anecdotal evidence suggests that early vegetative responses to applied P often fail to translate to grain yield response. An experiment was conducted on a P-deficient Vertosol in the Wimmera. The hypothesis tested was that the failure of wheat to produce grain yield responses to P application is the result of a combination of soil moisture and N supply limitations. A factorial design consisting of P (0 or 16 kg P/ha); nitrogen (N) (0 and 50 kg/ha) and irrigation (plus or minus) applied throughout the growing season was imposed. Significant dry matter responses to P and N application were recorded during early vegetative stages (although tissue P concentrations were regarded as marginal) and these responses continued to maturity. Phosphorus uptake continued during grain fill even under dryland conditions but there was no significant grain yield response to P unless N was also applied, regardless of soil water supply. The trial suggested that the importance of N and its interactions with soil water availability to grain yield response to P may have been underestimated.

Key Words

Phosphorus, nitrogen, water stress

Introduction

Anecdotal evidence from field studies conducted on alkaline soils, ranging from the Eyre Peninsula in SA (R Holloway pers, comm.), through to Victoria (Armstrong et al 2008) and northern Australia (M Bell pers, comm.) suggests that dry matter responses to applied P fertilisers during early growth often fail to translate to grain yield responses. An analogy to this lack of grain yield response to P is the ‘haying off’ effect observed with luxury supply of N during early crop growth followed by a ‘dry finish’ (van Herwaarden et al. 1998). In contrast to N, plants are particularly sensitive to P deficiency during early growth stages because P significantly influences root growth and the subsequent ability to access soil water and other nutrients (Grant et al. 2001). Furthermore P is an immobile nutrient and soil moisture will affect fertiliser granule dissolution and the subsequent diffusion of P (Kovar and Claassen 2005) as well as the ability of the crop roots to access soil and fertiliser P, which is often concentrated in the topsoil (Pinkerton and Simpson 1986). This paper reports a field study that examined the effect of soil water and N availability on the response of wheat to fertiliser P. Understanding why grain yield fails to respond to P and the relationship with seasonal conditions may explain the reputed poor ability of some soil tests (e.g. Colwell P) to predict crop responsiveness to P.

Methods

The trial was conducted at Murtoa (Vertosol; average annual rainfall 420 mm) in the Victorian Wimmera. The trial consisted of a randomised block design of 2 P levels (0 and 16 kg P/ha) and 2 N levels (0, 50 kg N/ha), replicated 4 times. These main plots consisted of 14 m long plots with 8 rows (19.5 cm row spacing) with a buffer drill strike on either side. Within each main plot, irrigation subplots (8 rows x 2.4 m long) were imposed. Irrigation was applied using gravity fed drip irrigation, with 40 mm equivalent precipitation applied over 2 to 3 hours (to minimise runoff from plots). Three irrigation treatments were applied: (1) dryland: no irrigation (2) irrigated: 40 mm irrigation applied 16^th June (one week after emergence), 13^th September and 13^th October and (3) ‘dry finish’ (40 mm irrigation applied 16^th June and rainout shelters erected on 25^th October (2 weeks post-anthesis). The rainout tents were constructed of ‘Solarweave’ plastic, open at either end (1.2 m high x 3.5 m long). Phosphorus fertiliser was applied as double superphosphate (banded with seed) and N (as urea) incorporated at sowing. Wheat (cv. Clearfield JNZ) was sown at 70 kg/ha on 18^th May. Weed, mice and rust control was achieved using typical regional practice.

Crop biomass was sampled at mid tillering, anthesis and grain maturity (6 rows x 0. 5 m). The samples were dried at 70^oC for 48 hours and weighed. Plant samples were then ground (< 0.5 mm) and analysed for tissue P (boiling acid digest followed by ICP analysis) and N concentration (Leco). Soil water was sampled in all subplots by hydraulically inserting a thin walled tube (50 mm diameter) to 140 cm depth. The core was subsampled at 20 cm increments, weighed then dried at 105^oC to determine water content. Root mass was measured at anthesis by inserting 2 cores per plot to 140 cm, and then washing sampled roots from soil (over 0.5 mm sieve), and determining the dry weight of the sample. Grain yield components were measured at maturity.ub headings are important to distinguish from main headings and normal text. Normal text is Times New Roman, 11 point, with the title larger (14 point, bold) and address smaller (9 point). Of course, there are many other ways to present data (Fischer 1985; Muchow and Carberry 1989). The examples here are just a guide and have been adapted from those prepared for the recent International Crop Science Congress.

Results

Soil analysis showed insufficient soil available P compared to a critical Colwell P of 38 (Moody 2007) derived from the PBI (Table 1) as well as the DGT-P test. There was no indication of subsoil physicochemical constraints (e.g. salinity, data not presented). Despite being sown to chickpeas in 2010, soil nitrate at sowing was low at 34 kg N/ha.

Table 1: Summary of soil characteristics (0-10 cm) at Murtoa (Vertosol) at sowing.

Site	DGT P (ug/L)	Colwell P (mg/kg)	PBI	Total-P (mg/kg)	Total C (%)	pH (1: 5 CaCl₂)	Total N (%)	Profile NO₃-N (kg/ha)	Profile PAWC (mm)
Murtoa	45	29	151	211	1.31	7.3	0.088	34.2	80.2

Rainfall in January and February was very high (Decile 10), followed by below average rainfall until July when rainfall was Decile 5. From August to harvest rainfall was below average (Table 2).

Table 2: Monthly rainfall at Murtoa (with respective long-term averages) in mm. GSR = growing season rainfall

Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	GSR
					Rainfall (mm)
Murtoa	166	52	23	13	16	16	49	37	28	34	49	43	228
LT mean	24	23	22	30	42	45	45	45	43	43	31	29	294

Early dry matter production was stimulated by applying either P or N fertiliser, with the response to N much greater when post emergence irrigation was applied in both the irrigation and dry finish treatments (Table 3). The stimulatory effect of greater N and soil water on dry matter production was maintained until grain maturity whereas that of P application was not evident by maturity (except under a dry finish). Grain yield was significantly greater (P < 0.01) when N was applied, especially in the absence of P but neither irrigation treatment nor P fertiliser per se stimulated grain yield. Neither harvest index nor grain size was affected by N or P fertilisation or irrigation treatment.

Crop P uptake up to anthesis was increased by P and N fertiliser application (P < 0.01), especially when irrigation was applied (Table 4). Although shoot P concentrations were increased by the application of P fertiliser and irrigation (average of 0.32% P), levels were still ‘marginal to adequate’ for the growth of wheat (Reuter and Robinson 1997). P uptake continued to increase up to maturity, even where no supplementary irrigation was applied and was significantly increased by P application, irrigation and N (in the irrigated treatment only). Grain protein was deficient across all treatments (mean of 9.7%), even when N was applied.

Root mass was concentrated in the topsoil (0-20 cm), but was not significantly (P > 0.05) affected by P, N or irrigation treatment (data not presented). There was a trend (P < 0.1) for N application to increase root growth in intermediate depths (20- 80 cm) when supplementary irrigation was applied, whereas P application increased root growth under dryland conditions at greater soil depths (80 -120 cm). There was little root growth beyond 100 cm, resulting in an average of 38 mm soil water remaining at maturity regardless of treatment. Water/rainfall use efficiency ranged from as little as 7.5 to 11.6 in irrigated treatments up to 11.5 to 13.8 kg grain/mm/ha in dryland plots .

Table 3: Effect of P and N fertilisation and soil water availability on dry matter and yield.

Fertiliser	Irrigation	Mid till dm (kg/ha)	Anthesis dm (kg/ha)	Maturity dm (kg/ha)	Grain yield (kg/ha)	Harvest Index	1000 seed wgt (g)
0N0	dryland	419	4407	8895	3225	0.361	41.8
	irrigated	442	4426	8035	2845	0.353	42.2
	dry finish	501	4693	8929	3151	0.351	42.5
P16N0	dryland	776	5214	9198	3102	0.337	41.8
	irrigated	638	5560	9745	3389	0.348	41.0
	dry finish	928	6618	11429	4037	0.353	42.4
P0N50	dryland	673	5515	10354	3733	0.361	41.9
	irrigated	835	7037	13000	4381	0.338	42.3
	dry finish	561	6080	10657	3819	0.357	42.5
P16N50	dryland	924	7266	9201	3240	0.352	42.1
	irrigated	1209	7556	11963	3752	0.342	42.2
	dry finish	1016	7301	11623	3958	0.340	42.1
Lsd (5%)		182	703	1425	465	n.s.	n.s.
		P ;N; N*Irri	P; N*Irri	N;PN;NIrri	N; P*N

Table 4: Effect of P and N fertilisation and soil water availability on P and N uptake.

Fertiliser	Irrigation	Mid till P (kg/ha)	Anthesis P (kg/ha)	Maturity P (kg/ha)	HI-P	Maturity N (kg/ha)	Grain protein (%)
P0N0	dryland	1.19	6.59	11.32	0.91	70.39	9.86
	irrigated	1.20	7.44	11.23	0.90	56.81	9.09
	dry finish	1.46	7.48	10.70	0.92	60.20	8.84
P16N0	dryland	2.31	7.12	12.19	0.90	66.95	9.48
	irrigated	2.12	9.31	14.50	0.91	67.92	9.02
	dry finish	2.89	10.27	13.54	0.91	82.25	9.22
P0N50	dryland	1.83	7.26	11.20	0.90	82.52	10.90
	irrigated	2.59	11.43	15.29	0.90	96.54	9.66
	dry finish	1.66	8.08	11.87	0.91	82.68	9.86
P16N50	dryland	2.70	9.84	11.19	0.90	77.23	10.45
	irrigated	3.83	11.62	15.60	0.90	81.56	9.73
	dry finish	3.25	10.47	12.25	0.90	89.42	9.79
Lsd (5%)		0.56	1.44	1.90	n.s.	8.77	0.39
		P ; N; N*Irri	PNIrri	P; Irri; N*Irri		N	N

Conclusions

Despite pronounced responses to P fertiliser during vegetative growth stages, grain yield was increased only when N was also applied, regardless of soil water supply. Computer simulation studies suggest that rainfall (quantity and timing) can markedly affect the P and N response of wheat during vegetative growth stages in this region (Dunbabin et al. 2009). This has been verified recently (McBeath et al. 2012) who, using ³³P labelling techniques showed that adequate soil water can significantly increase the uptake of both soil and fertiliser P during early vegetative stages of wheat. In the current study irrigation applied soon after sowing influenced subsequent crop growth (and response to applied N and P) via both increasing seedling establishment following ‘patchy’ emergence (resulting from highly variable soil moisture in the topsoil) as well as influencing the availability of, and demand for, nutrients to the young crop. Phosphorus deficiency restricts root growth during early growth stages (Grant et al. 2001), limiting the ability of the crop to effectively access water and other nutrients during grain fill. Adequate N supply however appeared to be critical to realising any benefit on grain yield of applying P.

Despite the application of relatively high rates of P (16 kg P/ha), tissue P concentrations of wheat were still marginal during early growth stages. Granular P fertilisers, such as the superphosphate used in this experiment, can be ineffective due to chemical fixation in highly alkaline soils (Hollaway et al 2001) but previous research indicates that this is unlikely on Vertosols in the Wimmera (Armstrong et al. 2008). Soil water in the topsoil at sowing at this site was relatively low (and quite variable) but there was nearly a ‘full profile’ at lower depths. As a consequence we expected that irrigation may have ensured adequate crop uptake of soil and fertiliser P, at least in early vegetative growth, but this did not occur. However P uptake by the wheat continued throughout the season across all water treatments, so that this initial P deficiency may have been alleviated later in the season.

Nitrogen application significantly increased dry matter production throughout the season as well as grain yield. The response to N application was enhanced when P and/or irrigation was applied. However grain protein levels were relatively low even when N was applied, suggesting that grain yield responses may have been even greater if further N was applied. The ‘Millennium Drought’ saw a marked reduction in N applications and presumably a run down in soil N reserves to below the critical level of 0.110% N in the 0-10cm (Tuohey and Robson 1980) at this site (Table 1). This trial suggests that N management will be important to realising grain yield potential across a range of seasonal conditions in the coming years.

Acknowledgements

This research was co funded by the GRDC (Project DAV00095). We are very grateful to Richard Adler for use of his property and M Munn, R Perris, A Purdue and J Elliott (DPI) for their technical assistance.

References

Armstrong R, Nuttall J, Holloway B, Lombi B and McLaughlinM (2008). How effective are fluid phosphorus fertilisers in the Wimmera and Mallee region of Victoria? Proc. 14th Australian Agronomy Conference, 21-25 September 2008, Adelaide, South Australia. Australian Society of Agronomy.

Dunbabin, V.M., Armstrong, R.D., Officer, S.J. and Norton, R.M. (2009). Identifying fertiliser management strategies to maximise nitrogen and phosphorus acquisition by wheat in two contrasting soils from Victoria, Australia. Aust. J. Soil Res. 47, 74-90.

Grant , CA, Flaten DN, Tomasiewicz DJ, and Sheppard SC (2001). The importance of early season phosphorus nutrition. Canadian. J . Plant Sci. 81,211-224.

Holloway R, Bertrand I, Frischke A, Brace D, McLaughlin M and Shepperd W (2001). Improving fertiliser efficiency on calcareous and alkaline soils with fluid sources of P, N and Zn. Plant & Soil 236,209-219.

Kovar JL and Claassen N (2005) Soil-root interactions & Phosphorus nutrition of plants. In ‘Phosphorus: Agriculture and the environment.’ Ed. LK Al-Amoodi pp. 379-414. ASA/CSSA/SSSA Madison, WI.

McBeath, TM, McLaughlin MJ, Kirby JK and Armstrong RD (2012) Plant & Soil Online First.

Moody P (2007) Interpretation of a single-point P buffering index for adjusting critical levels of the Colwell soil P test. Aust. J Soil Res 45,55-62.

Pinkerton A and Simpson JR (1986). Interactions of surface drying and subsurface nutrients affecting plant growth on acidic soil profiles from an old pasture. Aust .J. Exper. Agric. 26,681-9.

Reuter DJ and Robinson JB (1997). Plant Analysis:An Interpretation Manual. CSIRO Publishing, Melbourne.

Romer W and Schilling G (1986) Plant requirements for phosphorus of the wheat plant during various stages of its life cycle. Plant & Soil 91,221-229.

Tuohey CL and Robson AD (1980). The effect of cropping after medic and non-medic pastures on total soil Nitrogen, and on the grain yield and Nitrogen content of wheat.. Aust. J. Exp. Agric. Anim. Husb. 20,220-228.

van Herwaarden AF, Farquhar GD, Angus JF, Richards RA, Howe GN (1998). `Haying-off’, the negative grain yield response of dryland wheat to nitrogen fertiliser I. Biomass, grain yield, and water use Aust J Agric Res 49, 1067-81.

...

Search Site

The Regional Institute
Help/Feedback
phone: 61 2 4369 6006
email: admin@regional.org.au
Sitemap