Incitec Fertilizers
1 12 Merrigal Street, Griffith NSW 2680, Phone 02 6964 6242, Fax 02 6964 6243 e-mail: yash.dang@incitec.com.au
2 PO Box 260, North Geelong Vic 3215, Ph 03 5273 7806, Fax 03 5273 7810 e-mail: charlie.walker@incitec.com.au
Whole field monitoring in the Griffith region of NSW indicated poor apparent nitrogen use efficiency (ANUE) in maize in the 1997/98 and 1998/99 seasons. In 1999/00 experiments were conducted in two fields to assess the fate of fertilizer and residual soil nitrogen (N) in three sections of a field (head, mid & tail). A yield map of the field developed from yield monitoring clearly showed a low yielding control strip placed across the field. Adjacent to the control was a uniformly yielding strip that received pre-plant 120 kg N/ha. The rest of the field treated with 240 kg N/ha split pre-plant & water run showed acceptable ANUE & yield mid-field. The tail end of the field showed low ANUE & yield suggesting significant N losses: most likely denitrification, due to water logging. Surprisingly, the head ditch end of the field also showed lower than acceptable ANUE suggesting that water run N had stripped out as nitrate N further down the field or had denitrified.
An experiment during 2000/01, investigated improving ANUE through method of N application. Application methods used were all pre-plant, split pre-plant & side dress or split pre-plant & multiple water run. Increasing rates of pre-plant N significantly increased maize yield up to 240 kg N/ha. Comparing maize yield and yield map for N applied @ 240 kg N/ha in different methods of N applications, showed that N evenly split applied as pre-plant & side dress urea was slightly better than split as pre-plant & water run and all pre-plant N (non-significant). A non-significant trend was identified that split soil application of nitrogen was able to smooth out across field yield variation when compared to split soil application and water run urea.
Introduction
Nitrogen being costliest input in agriculture and growing concern over its impact on water quality justifies more efficient N use (Angus et al. 1993). Most N recommendations were developed for whole-field using averaged yields and soil information commonly resulting in unrealistic match of demand and supply in a substantially variable field. To increase NUE, it is necessary to pinpoint under performing or problem areas and adjust N inputs and application strategies according to site-specific requirements.
Yield maps highlight large in-field variations on a fine scale and help in making the decision for the following season. Conventional sampling is too time consuming and expensive to achieve required resolution of data. However, yield maps developed using latest electronic technology can estimate data values at unsampled locations to help growers measure what is happening to the crop and soil and assist in making more effective decisions.
Methods and materials
In 1997/98 and 1997/98 seasons, 20 soil cores (0-10 cm and 10-60 cm) were taken before the sowing of maize crop and mixed separately for each depth from whole field in the maize growing area of Griffith, NSW and analysed for NO3-N, NH4-N (Rayment and Higginson, 1992). At maturity, maize grain yield was recorded and grain samples were analysed for total N. The ANUE was calculated using the amount of available soil N, fertilizer N and crop nitrogen requirement using LINK total nitrogen management model (Nutrient Advantage, Incitec Ltd.)
In 1999/00, experiments were conducted in two fields near Griffith, NSW in south eastern Australia to assess the fate of fertilizer and residual soil N in three sections of a field (head ditch end, mid field & tail drain end). Both soils were neutral to alkaline at the surface and showed little changes in pH, NO3-N or NH4-N with depth (Table 1).
Table 1. Some properties of the soils
1999/00 |
2000/01 | |||||
Site 1 |
Site 2 |
Site 3 | ||||
Depth (cm) |
0-10 |
10-60 |
0-10 |
10-60 |
0-10 |
10-60 |
Treatments included no applied fertilizer N, pre-plant N (anhydrous ammonia (82 % N) as Cold-Flo ™; Big N) @ 120 kg N/ha & 240 kg N/ha evenly split as 120 kg N/ha pre-plant, the balance as 4 water run urea treatments. The plots were 7.2 m wide and 500 m long. Plant samples were taken at anthesis and physiological maturity of crop growth and analysed for total N. Water samples were obtained during an irrigation run from the entry and exit points of the field. Samples were analysed for NO3-N with no difference in concentration recorded. At maturity, grain yield was determined by harvesting 3 rows, each 1m long at the top, mid and tail end of each plot and yield was corrected to 14% moisture. Subsamples of grain were analysed for total N. Soil samples (0-15, 15-30, 30-45, 45-60 and 60-90 cm) were taken after the harvest of maize crop from the centre of beds and analysed for NH4-N and NO3-N.
In 2000/01, a experiment was conducted to examine the response of maize to the application of several rates of N as anhydrous ammonia as a pre-plant band application and compares with split N applications as pre-plant anhydrous ammonia with balance N in irrigation water (water run urea, WRU), side dress with urea (SDU) or side dress anhydrous ammonia (SDA) as given in Table 2.
Table 2. Treatments applied in 2000/01 season
Pre-plant N |
Post-plant N |
Strip number |
Pre plant |
Post-plant N dates | |||
N0 |
|
1 & 10 |
13.10.00 in all the treatments as Big N |
- |
- |
- |
- |
SD, side dress; WR. Water run; Big N, anhydrous ammonia as cold flow | |||||||
Side dressed urea or anhydrous ammonia was applied on 26th December and 4 WRU one each on 15th December 2000, 26th December 2000, 6th January 2001 and 16th January 2001. Each plot was 7.2m wide and 350 m long. Plant samples were taken from three sections of the each treatment (head, mid and tail) at anthesis & physiological maturity and analysed for total N. At maturity, grain yield was determined by harvesting 3 rows, each 1m long at the top, mid and tail end of each plot and yield was corrected to 14% moisture. Subsamples of grain were analysed for total N.
Statistical analysis were performed in split plot design with method of N application as main plot and three sections of the treatment as sub-plot using Genstat 4.2.
Results and discussion
The ANUE in maize on whole-field scale in the 1997/98 and 1998/99 seasons ranged between 25-35%, which was low as compared to 40-55% as reported by Power et al. (1973); Muirhead et al. (1984).
The experiment conducted during 1999/00, assessed the fate of soil and fertilizer N in three sections of the field (head, mid and tail). The available soil N contents in the three sections of the paddock were 142 in head ditch (HD), 218 mid (MID) and 227kg N/ha, in tail drain end (TD). As expected, with the increase in available soil N from top to tail end, the grain yield also increased from 6.6 to 8.7 t/ha.
On the whole-field basis, the increasing applied N rates increased grain yield but ANUE decreased at 240 kg N/ha (Table 3).
Table 3. Maize yield, grain N and apparent nitrogen use efficiency (1999/00)
Control |
Pre-plant |
Split pre & post | ||||
Yield (t/ha) |
Grain N |
Yield (t/ha) |
Grain N |
Yield (t/ha) |
Grain N | |
Head ditch end |
6.6 |
68 |
11.5 |
148 |
10.0 |
111 |
Figures in parenthesis are the percent apparent nitrogen use efficiency.
The TD end of the field tended to contribute most to lowest ANUE. Unexpectedly, the HD end of the field exhibited comparatively low ANUE where N was split between pre-plant and water run.
Figure 1. Yield map 2000
A yield map of the field clearly showed a low yielding control strip across the field (Figure 1). Adjacent to the control was a uniformly yielding strip that received pre-plant 120 kg N/ha. The rest of the field treated with 240 kg N/ha split pre-plant & water run showed acceptable ANUE and yield in MID. Tail drain end showed low ANUE and yield suggesting significant losses of N most likely as a result of denitrification due to water logging. Surprisingly, the HD end of the field also showed lower than acceptable ANUE suggesting that water run N had stripped out as nitrate N further down the field or denitrified. Similar observations are common in cotton field with yellow head ditches which is a combination of stripping, leaching and denitrification (C. Dowling, Pers. Comm.). Soil samples taken after the harvest of maize also confirmed deposition (or lower levels of N loss processes) of NO3-N in the middle (Table 4).
Table 4. Nitrate-N (mg/kg) in soil profile (0-90 cm) and estimated available N (kg/ha) after the harvest of maize (1999/00)
Depth |
Control |
Pre-plant |
Split pre & post | |
Head ditch end |
0-10 |
3.2 |
5.1 |
6.2 |
On the basis of the results during 1999/00, it was thought that the safest strategy for growers may well be place more emphasis on N application banded into beds or hills at or below furrow depth. This is because N banded at depth should protect against losses through denitrification and would be less likely rise to the soil surface in the evaporative stream in the nitrate form and become stranded on the surface of the bed. With water-run urea in a bed or hill irrigation system, there is potential for hydraulic pressure to push nitrate to the soil surface in the absence of significant rainfall. It was thought that the mechanical placement of N well under the soil surface through side dressing may reduce the risk of denitrification losses to the atmosphere being dislocated from higher levels of crop residues and slightly cooler.
The experiment conducted during 2000/01 season, placed emphasis on applying various rates of pre-plant N and comparing with split N applications as pre-plant N with balance N in irrigation water, side dress as urea or anhydrous ammonia. The dry matter yield and N uptake at anthesis (first week of January) significantly increased with increasing rates of pre-plant N (data not shown). Pre-plant N @ 240 kg/ha was at par with 240 kg N/ha applied as half pre-plant and half as water run. Total N uptake and dry matter yield with N applied in split application as water run N was higher than in split application as side dress urea or anhydrous ammonia, this could be due the fact that 2 splits of N in water run were applied during December 2000 as compared to one time side dress on 26th December, a week before the sampling.
Increasing pre-plant N rates significantly increased maize yield up to 240 kg N/ha while the increase in yield with 300 kg N/ha was not significant (Table 5). Comparing maize yield at 240 kg N/ha in different methods of N application, showed that N applied half pre-plant and half as side dress urea was slightly better than half pre-plant and half water run and pre-plant banded N.
Table 5. Effect of method of N application on maize yield (2000/01)
Maize yield (t/ha) | ||||||
Pre-plant N |
Post-plant N |
Head ditch end |
Mid field |
Tail drain end |
Mean |
Whole-field |
N0 |
|
1.34 |
1.37 |
1.55 |
1.42 |
1.36 |
SD, side dress; WR. Water run; Big N, anhydrous ammonia as cold flow | ||||||
The interaction between application strategy and location within field revealed an interesting non-significant trend that warrants further investigation (Figure 2). Where 4 application strategies were compared, soil applications showed the most even yields from one end of the field to the other. Where 50% of N was applied as water run urea, the trend of higher yield MID compared to HD and TD was again observed. While non-significant, the observation strengthens the argument that soil applied nitrogen with a 50:50 split (pre-plant / post sowing) offers the highest yield across the field.
Figure 2. The effect of N application strategy on maize yield in 3 sections of an irrigated field
A yield map from right with first strip as a border followed by a low yielding control strip across the field (Figure 3). The subsequent strips followed the treatments as given in Table 2. Yield map clearly showed uniform stripes in all pre-plant N and side dress urea or anhydrous ammonia as compared to high yielding MID and relatively low yielding TD for water run urea in the experimental as well as rest of the paddock.
The yield monitoring and yield maps from both the years suggest that application of N as water run urea is effective method of applying N when the length of row was limited to approximately 200 m. These results are in accordance with Muirhead et al. (1985). They reported 63% ANUE with water run urea up to 100 m length of row. However, when the length of row is more than 200 m, the side dress urea may be the safest strategy for the growers to apply N during grand growth period to meet the demand of the growing plants.
Acknowledgment
We wish to acknowledge the generosity of Mr A Irvin for making land available and carrying out the field operations for this study.
References
Angus, J.F., Bowden, J.W. and Keating, B.A. (1993). Modelling nutrient responses in the field. Plant and Soil 155/156: 57-66.
Muirhead, W.A., Melhuish, F.M. and White, R.J.G. (1985). Comparision of several nitrogen fertilizers applied in surface irrigation systems. I. Crop response. Fertilizer Res. 6: 97-109.
Power, J.F., Alessi, J., Reichman, G.A. and Grunes, D.L. (1973). Recovery, residual effects and fate of nitrogen fertilizer sources in a semiarid region. Agron. J. 65: 765-768.
Rayment, G.E. and Higginson, F.R. (1992). Australian Laboratory Handbook of Soil and Water Chemical Methods. Inkata Press.
About authors
Yash Dang:
Before joining Incitec as Market Development Agronomist in November 1999, I worked with Queensland Department of Natural Resources and reported on conditions and trends in natural resources management for Condamine-Balonne Catchment. I have also worked on nutritional aspects of field crops for more than 18 years in India and Australia which includes a PhD from the University of Queensland in Soil Chemistry and Plant Nutrition.
Charlie Walker:
As Market Development Agronomist for Incitec’s Southern Region, I spend a large portion of my time on projects associated with improving outcomes from the use of nitrogen in field crops. Apart from this work, I am also involved in projects in rice, cereals in arid environments of Vic & SA, wheat in the long season cool climate zones of southern Australia and a long term continuous cropping experiment in the Wimmera. Prior to working in southern Australia, I spent 3 years on the Darling Downs with Incitec, 1 year in Moree and 3 years as a sales agronomist in Tamworth, NSW.
How Incitec plans to use the technology?
Given the higher gross returns per unit area in irrigated agriculture, it is foreseen that this is the area that precision agriculture (PA) will be most rapidly adopted. While the traditional diagnosis provided by PA was variable rate application, this approach actually challenges the need for VRT where (usually) the largest variable is removed (water). Hence, through simple farmer conducted experiments, we are able to confirm financial responses to changed application strategies and adopt these strategies where justified. Another possibility is also the identification of the potential for denitrification where water logging occurs and subsequent changes to water management or land forming.
