The objectives of this project are to quantify the size and efficiency of response to mid-row banded nitrogen in irrigation systems within the southern Riverina of Australia. This will include the assessment of the effect of mid-row banded N on growth, yield and apparent nitrogen recovery efficiency (ANRE), of wheat as compared to the current best-practice method of topdressing at GS31 and the control.

Another key objective of this project is to assess the loss of banded nitrogen during waterlogging events to determine if denitrification losses are reduced in banded nitrogen and compared to topdressed nitrogen and the control.

Mid-row banding nitrogen at 133 kgN.ha⁻¹ gave a similar response to topdressing the same rate at GS31. It increased yield by 1.38 t/ha and grain N by 38 kgN.ha⁻¹ to give an ANRE of 30%. Mid-row banded N created a highly-concentrated band of ammonium which appeared to mostly survive a water-logging event.

Mid-row banding N appears to be a valid application method for nitrogen in a southern Riverina irrigated context, allowing all nitrogen requirements to be placed in the soil at sowing. This is applicable where a higher yield potential is known with some confidence, and seeding equipment can handle larger amounts of fertilizer at sowing. Hence, it may also be a useful technique in higher rainfall zones, where yield potential is high and waterlogging is more common.

Please see details for Moulamein site.

Please see details for Moulamein site.

To investigate the feasibility of mid-row banding under a variety of irrigation systems commonly found in the southern Riverina, three locations were chosen to host trials in 2018:

1. North Dale: Supplementary irrigated lasered contour, (rice harvested April 2018)
2. Collinjen: Fully irrigated border-check
3. Royal Park: Supplementary irrigated lasered contour (rice harvested April 2017)

Within these sites, three nitrogen treatments were tested to compare the nitrogen use efficiency of mid-row banded nitrogen to topdressed nitrogen:

1. DAP 1% Zn @ 100kg.ha⁻¹ in the seed row (15N)
2. DAP 1% Zn @ 100 kg.ha⁻¹ in the seed row + 290 kg.ha⁻¹ Urea topdressed at GS31 (150NT)
3. DAP 1% Zn @ 100 kg.ha⁻¹ in the seed row + 290 kg.ha⁻¹ Urea banded in the mid-row[1] at sowing (150NMRB)

To continue investigating the implications of losses to denitrification on heavy Sodosols within the region, there was also a waterlogging treatment imposed at North Dale:

• No waterlogging event imposed (‘Dry’)
• A waterlogging event imposed, by irrigating on August 5^th, then draining on August 16^th (‘Wet’)

A randomized block design was used for site 2 & 3, investigating nitrogen treatments, whilst site 1 used a split plot design to investigate the interaction between waterlogging and nitrogen treatments, with waterlogging treatments in the main plots and nitrogen treatments in the sub-plots. All sites had four replicates with plots measuring 20m long and 8m wide.

To measure the effect on growth and yield, and interaction between waterlogging and nitrogen treatments, the following plant measurements were taken:

• NDVI, NDRE, CCCI (September 17^th)

• SPAD meter (September 17^th)
• Biomass cuts (September 19^th & October 18^th)
• Head number (November 28^th)
• Grain yield, protein & 1000 grain weight (December 10^th)

The following soil measurements were taken:

• Soil test pre-sowing (May 7^th)
  • Three samples per site at 0-10cm, 10-20cm & 20-60cm (12 sample points per sample)
• Soil mineral N in the location of the mid-row band pre-waterlogging (July 18^th), post-topdressing (September 13^th), grain fill (October 23^rd) and post-harvest (January 31^st)
  • Two sample depths per plot (2-10cm & 10-20cm) and 5 sample points per plot
• Soil matric potential every 12 hours using the Watermark gypsum block sensors

Management

Collinjen was pre-watered about a month before sowing, but neither Royal Park nor North Dale was not pre-watered (North Dale had rice the preceding summer/autumn). Consequently Collinjen suffered less drought effects from the extremely low growing season rainfall. Wheat (var. Mace) was sown using a modified Bettinson disc drill with 25-27cm row spacing on the 15-16^th of May at a rate of 80 kg.ha⁻¹ and depth of 3-5cm. Fertiliser (DAP 1% Zn) was treated with 4L/t of Impact for fungal suppression and sown at 100 kg.ha⁻¹. Weed, insect and fungus control was conducted as per the surrounding field.

At North Dale levees were constructed after sowing to exclude water from the ‘dry’ plots during the winter waterlogging event beginning on the 5^th of August. One spring irrigation followed this in late September. Collinjen was fully irrigated with two irrigation events in spring. Royal Park received a single irrigation in spring.

Urea topdressing occurred at GS31 for all sites, which occurred from 18-29^th of August, 11 days after the waterlogging event at North Dale. At this time, the North Dale site was also assessed for duck damage incurred during water-logging and the variability between plots was quantified.

During grain-fill Russian Wheat Aphid also impacted both the North Dale and Royal Park sites, with Royal Park being treated subsequently. Again, this damage was assessed and noted for consideration when analysing results.

Statistical analysis was conducted using the Statistix software package for analysis of variance (ANOVA). Two-way ANOVA was used to assess the effects of nitrogen treatments and water-logging treatments on wheat yield, N uptake and N in the soil. Significances among the treatments were compared by the least significant difference at P<0.05 level.