A snapshot of wheat and barley agronomy trials in the northern grains region of NSW

Matthew Gardner¹, Rohan Brill² and Guy McMullen¹

 

  ¹NSW DPI Tamworth
² NSW DPI Coonamble
 

Background

Broadly, this paper reports on a range of agronomy trials conducted throughout the northern grains region from 2008 to 2012. Primarily trials are associated with the Variety Specific Agronomy Packages (VSAP) project that aims to close the yield gap between seasonal potential and achieved yields through optimising the agronomy of key wheat, barley and canola varieties. This paper is divided into four sections:

1. Plant Growth Regulators (PGRs) and lodging management
2. Effects of delayed harvest
3. Nitrogen management in barley
4. Time of Sowing and New Varieties

1. Plant growth regulators and lodging management

Plant growth regulators (PGRs) have been used routinely in high input, high yielding cereal systems in Europe and NZ for some time to shorten plant height and reduce the incidence of lodging. Lodging results in significant losses in crop production due to reduced movement of water, nutrients and translocation of plant stored carbohydrates through the stem into the head. Lodging also reduces grain quality; increases harvest losses and the actual time and cost of the harvesting process. Although Gibberellin inhibitors and Ethylene producers are the two main PGR groups, the research presented here only investigated Gibberellin inhibitor products. These products act by blocking gibberellin biosynthesis to reduce internode length in stems thereby decreasing plant height. There are a number of phases in this pathway and different PGRs act at different points. For example chlormequat (Cycocel ®) acts early in the hormone synthesis pathway while more recently developed products such as trinexapac-ethyl (Moddus®) act on later stages.

PGRs have also been reported to have a yield enhancement effect by improving the proportion of crop dry matter that is partitioned into grain yield. This effect has been related to a reduction in the plant resources required for stem elongation with these resources then available for grain-fill. Some PGRs have also been associated with increased root growth resulting in improved water extraction from soil. Yield responses to PGRs can be highly variable with responses ranging from -40% to +20% depending on product choice, application time, crop or variety and growing season conditions.

What did we do?

In the 2011 and 2012 seasons a series of trials were conducted to investigate the capacity of PGRs to reduce lodging in Commander (high yielding with poor straw strength) barley. In both seasons Commander and Oxford (high yielding with good straw strength) were grown at a target plant population of 120 plants/m²  with four treatments of: nil PGR, Cycocel (0.2 L/ha), Moddus (1.0 L/ha) and a combination of Cycocel (0.2 L/ha) + Moddus (1.0 L/ha). PGRs were applied in each season during stem elongation (GS31) at a 100 L/ha water rate. In 2011, sites were established at Tamworth and Spring Ridge, while in 2012 sites were at Moree and Breeza. Please note that at the time of writing, Moddus is not commercially available and is not currently registered in Australia for these use patterns.

What did we find?

Although in 2011 the Tamworth site had lower lodging compared to Spring Ridge the trends were similar (Table 1). The lodging severity for Commander was approximately 3 times what was observed for Oxford, again highlighting the importance of variety selection in lodging management (Table 1). The combination of Cycocel + Moddus was the most effective PGR treatment at reducing the severity of lodging compared to the control treatment (Nil PGR) (Table 1).

The ability of PGRs to reduce the severity of lodging appears related to their capacity to restrict plant height (Figure 1a). At Spring Ridge (2011) and Moree (2012) the Cycocel + Moddus treatment was the most effective at reducing plant height (Figure 1a and 2a). As a single product, Moddus restricted plant height to a greater extent than Cycocel® at both sites (Figure 1a and 2a). There was a large difference between the amount of height reduction measured at the two sites with the maximum height reduction being 7 cm at Spring Ridge in 2011 compared to 34 cm at Moree in 2012. At the Spring Ridge site the treatments containing Moddus had no impact on yield compared to the nil treatment, whereas, the Cycocel treatment significantly increased the yield of Commander by 8 % compared to the nil treatment (Figure 1b). The large reduction in plant height at Moree for the Moddus and Cycocel + Moddus treatments resulted in a significant reduction in yield of 8 and 13 %, respectively (Figure 2b).

Figure 1: The effect of three PGR treatments on a) plant height and grain yield (LH) b) of Commander compared to a nil PGR control at Spring Ridge in 2011(RH).

Figure 2: The effect of three PGR treatments on a) plant height and grain yield(LH) b) of Commander and Oxford compared to a nil PGR control at Moree in 2012 (RH).

In 2008 a trial conducted at the Tamworth Agricultural Institute found a significant yield increase with early applications (GS25) of a PGR combination of Moddus and Cycocel in the absence of lodging. These increases occurred in both durum (EGA Bellaroi) and bread wheat (EGA Gregory) while no effect was found in Gairdner or Fleet barley. The later application timing resulted in no significant differences compared to the control.

PGR conclusions

In all trials PGR treatments were shown to reduce lodging to some degree, which was most likely a function of the reduced plant height obtained from PGR applications. Reductions in plant height associated with PGR application were generally moderate (3 to 15 cm) with the exception of the Moree site in 2012 where height reductions up to 34 cm were recorded. Yield responses to PGR application ranged from -13% to +16% in the barley and wheat trials across seasons. The grain yield and plant height results highlight the variability in responses to PGR application, which makes it difficult to accurately predict economic benefit of using PGRs within a cropping system. Of the PGR treatments the combined Cycocel® + Moddus® treatment resulted in the most consistent reduction in plant height, which is likely due to the two products blocking Giberellin production at different parts of the synthesis pathway. Further research is needed to understand the influence that PGRs are having on crop structure, tiller formation, root growth, development and soil water extraction in winter cereal crops in the northern grains region.

2. Effect of delayed harvest

The 2010 and 2011 seasons experienced above average rainfall during grain-fill into harvest in many regions of central and northern NSW. Wet weather can cause significant delays in harvest, either from paddocks being too wet for machinery traffic and/or high grain moisture. In addition to rainfall, the temperature often falls during wet weather, which increases the danger of pre-harvest sprouting. Pre-harvest sprouting is a serious problem in cereals resulting in downgrading of grain and heavy financial penalties to the grower. Varieties with dormant genotypes can be used to reduce downgrading caused by rain and, in combination with improved harvesting practices, the risk of weather damage (pre-harvest sprouting) from rain delayed harvest can be minimised. In barley it must be noted that dormancy which persists beyond harvest is highly undesirable because it prevents malting of newly received barley. In current Australian barley varieties, levels of dormancy are very low, and generally are only expressed when cool, wet weather conditions occur.

What did we do?

A trial was conducted in 2011 to investigate the impact of delaying harvest due to wet weather conditions on barley grain yield and quality and whether there was any difference between 12 varieties. This experiment was repeated in 2012 but expanded to include 16 wheat and 15 barley varieties in separate trials. Both the wheat and barley trials were planted on the 9^th May.

Each trial had four harvest dates with the first aimed at removing grain before any weather damage had occurred, whereas harvest dates beyond this point were timed to occur after rainfall events. Harvest dates and the rainfall between harvests are outlined in Table 2. Grain quality parameters were determined by a sub-sample from each plot. Unfortunately, grain quality data was not available for the 2012 trial at the time of writing this paper.

What did we find?

The yield penalty associated with delaying harvest from date 1 (11 Nov) to date 4 (7 Dec) varied significantly between barley varieties in 2011. Fitzroy and Buloke incurred the greatest yield penalty (2.2 and 2.1 t/ha, respectively) of the 12 varieties, whereas, Roe, Vlamingh and Grout (0.5, 0.7 and 0.7 t/ha, respectively) had the smallest losses (Figure 4). Commander, Gairdner, Hindmarsh and Shepherd all had similar yield losses (average 1.2 t/ha) from delaying harvest at Tamworth by 26 days, during which 215 mm of rainfall occurred.

Delaying harvest had a significant effect on all grain quality attributes (Table 3). Protein increased by up to 0.8% between either harvests 1 or 2 and harvests 3 and 4. The 1000 grain weight was similar for harvests 1 and 2, which were significantly higher (2.1 g) than both harvests 3 and 4. Retention significantly increased from harvests 1 and 2 to harvest 3 with a further significant increase associated with delaying to harvest 4. Screenings were negligible across the 4 harvest times but did significantly decline across the later harvest times. Test weight significantly declined by 3.1 kg/hL between harvest 1 and 2 before declining a further 2.7 kg/hL for harvests 3 and 4, which were similar.

**Any values designated with different letters within a column are significantly different (P<0.05).

In 2012, yield loss between harvest date 1 and harvest date 4 was on average 0.9 and 0.6 t/ha for the barley and wheat trials in 2012, respectively. Similar to what was observed in 2011 in barley there was a significant difference in yield loss with delayed harvest between varieties for both wheat and barley. Fathom, Oxford and Bass were least affected by delaying harvest while Navigator appeared to be the most affected. LongReach Crusader and EGA Gregory had greater yield losses with delayed harvest than all other varieties except for the experimental line LPB07-0548. Once grain quality data is available it will assist in further explaining differences between varieties. 

Delayed harvest conclusions

The 2010 – 11 seasons represented wet harvests that resulted in significant downgrades in grain quality and price received for grain at the silo. These trials have given an indication of what the penalties are for weather delayed harvest in barley and wheat and the differences that exist between varieties. In reality growers have little choice in the matter and would harvest if conditions, paddock or grain, would permit. However, if weather is imminent, this information may allow growers to prioritise the harvest of paddocks containing varieties which suffer the greatest negative impacts (yield and quality). Equally, this information may be useful for growers to consider at the start of the season when selecting varieties for sowing to spread their risk against unfavourable weather conditions during harvest.

3. Nitrogen management in barley

The ability to achieve malt classification, when a suitable variety has been selected, can result in significant financial advantages in some seasons when producing barley. To meet malt specifications growers should target protein levels of 10.5 – 12% to achieve maximum yield and still meet receival standards. As the rate of N supply is increased, yield will generally increase to a maximum level, whereas protein may continue to increase with further N application. Drier or wetter than expected seasonal conditions can significantly change yield potential mid-season, which consequently changes N requirements to meet target protein contents. Therefore, the flexibility of delaying N application to in-crop timings can be a risk management strategy for growers to adapt to changing seasonal conditions. When considering in-crop N applications it is critical to know starting soil N levels. Many paddocks may have high starting soil N levels, well in excess of what is required to achieve realistic target yields and maintain grain protein levels suitable for the production of malting barley.

What did we do?

In 2012 there were two groups of trials. The first group had three N sites, located at Cryon (near Walgett), Bithramere (near Tamworth) and Moree in the northern grains region of New South Wales, which had starting soil N of 61, 63 and 95 kg N/ha (0-120 cm), respectively. Commander, Bass, Navigator and Gairdner barley, were grown at a plant populations of 100 plants/m² at all three trials sites in 2011. In each trial 4 rates of N were applied at sowing including 0, 40, 80 and 120 kg N/ha as granular urea (46 % N). Two additional N treatments were implemented, 80 kg N/ha applied at growth stage 31 (GS31 - stem elongation) and a split application treatment where 40 kg N/ha was applied at sowing with a further 40 kg N/ha applied at GS31. The in-crop application of N was applied as 50% diluted liquid UAN, applied through streamer bars at 100 L/ha water rate.

The second group of trials had two N sites, located near Coonamble and Trangie in the Central Western grains region of New South Wales, with starting N values of 33 and 50 kg N/ha, respectively. Commander, Bass, Hindmarsh, Wimmera and Buloke barley, were grown at Trangie, with 3 rates of N applied at sowing including 0, 40 and 80 kg N/ha. Commander, Bass, Hindmarsh, Schooner and Buloke barley, were grown at Coonamble, with 4 rates of N applied at sowing including 0, 25, 50 and 100 kg N/ha. Two additional N treatments were implemented at Coonamble, 50 kg N/ha applied at GS 31and a split application treatment where 25 kg N/ha was applied at sowing with a further 25 kg N/ha applied at GS 31. The in-crop application of N was applied as granular urea.   

What did we find?

For the purposes of this paper only the Bithramere and Moree sites are presented. At both sites Commander  had the highest yield on average compared to the other varieties. Bass and Navigator had similar yield to Gairdner in these trials.  There was a significant N response at both sites, although much stronger at the Moree site. The split N application gave similar yields as the 80 kg N/ha up front treatment at both sites and for the respective varieties except with Commander at Bithramere where the split N treatment provided an 8% yield benefit (Figure 6). Delaying N until stem elongation resulted in a significant decrease in yield for Gairdner, Commander and Bass at Moree, whereas, at Bithramere there was no significant difference between the delayed N treatment and the 80 kg N/ha upfront. The protein responses were relatively linear at both sites but there were significant differences between treatments. Commander had the lowest protein content of all varieties, virtually across all N rates (Figure 6c and 6d). Gairdner has been found in previous NSW DPI trials to achieve approximately 1-1.5% higher protein than Commander with the same N input, which was again the case at Moree and Bithramere in 2011. Bass on average had the highest protein across the two sites and was approximately 0.4% higher than Gairdner. Delaying or split applications of N at Moree significantly reduced grain protein compared to the upfront application of 80 kg N/ha.

At Trangie Commander had significantly greater yield at the 40 kg N/ha treatment and was similar to Hindmarsh at the 80 kg N/ha. Hindmarsh was the only variety to have a significant increase in yield with each increase in N application. Commander had significantly lower protein than Wimmera, Buloke and Bass at all N treatments. There was no significant difference in the yield response to N treatments for Wimmera, Buloke or Bass. The protein concentration for Hindmarsh was similar to Commander at the 40 kg N/ha treatment, but higher for all other N treatments.

At Coonamble, Commander had significantly greater yield than all other varieties at the highest N rate but was similar to Buloke and Hindmarsh at the 25 and 50 kg N/ha treatments, respectively. Schooner had significantly lower grain yields than all other varieties for the 50 and 100 kg N/ha treatments but had significantly higher proteins across all N treatments. Commander had significantly lower protein than all other varieties for all N treatments except for Hindmarsh at the 25 kg N/ha treatment. There was no significant difference in the yield response to N treatments for Buloke or Bass except for the 25 kg N/ha treatment where Buloke had higher yield and protein. The delayed or split N treatments had no significant impact on protein or yield compared to the 50 kg N/ha applied at sowing.

Barley N management conclusions

There has been a trend for new barley varieties to achieve lower proteins; however, Navigator, Bass and Wimmera appear to be more protein responsive to N applications compared to Commander. The low protein of Commander has generally been an advantage to meet malt specifications, however, over the past couple of seasons extremely low proteins have been achieved throughout the region (<9%) suggesting that yield may have been sacrificed at this level. The higher protein achievement of Bass and Navigator indicates that growers may need to be careful growing these varieties on paddocks with high levels of residual N as it may jeopardise achieving malt specifications. Further analysis of grain quality will determine whether Bass, Wimmera or Navigator produce high screenings and low tests weights under high N situations similar to Gairdner.

4. Time of Sowing and New varieties 

The autumn break in NSW occurs anywhere between March and June, with the reliability of the break being more inconsistent in northern NSW compared to southern NSW. Wheat varieties available to growers provide the opportunity to plant wheat crops from late March until late June and still have the crop flowering when the risks of frost and heat stress are acceptable. Between mid September to the first week in October is the optimum flowering window for cereal crops at Tamworth to avoid excessive frost risk (>10%) and limit exposure to heat stress later in the season. The optimum flowering time for Trangie is from the start of September until the third week in September. Trials were conducted at Tamworth and Trangie to determine the yield and quality of a range of wheat varieties across 3 different sowing times.

What did we do?

The 2012 barley time of sowing (TOS) trials included 18 varieties with varying maturities and agronomic traits at Tamworth and 15 varieties at Trangie, including both commercially available lines and advanced breeder lines. These varieties were sown on 3 separate occasions; 26^th April, 20^th May and the 20^th June at Tamworth and 30^th April, 20^th May and 12^th June at Trangie. For the purposes of this paper, yield and anthesis dates for 11 varieties will be presented for both Trangie and Tamworth.

What did we find?

In 2012 at Trangie there was a significant decline in grain yield for each delay in planting time. On average TOS 1 and 2 both flowered within the optimum flowering window for Trangie (Table 4), while TOS 3 flowered 7 days past the optimum window. Fathom, had the highest yields from TOS 1, but was similar to Hindmarsh, which was surprising as it usually performs better from a mid May planting time. Commander, Navigator and IGB 1101 all performed similar to Hindmarsh. Commander and IGB 1101 were the better performing varieties for yield from TOS 2, while Commander, Oxford and Fathom had similar yields for the TOS 3. This was interesting as there was a 10 day difference in the maturity of Fathom compared to Oxford yet similar yields. Hindmarsh and IGB 1101 had similar maturities and for TOS 1 and 2 were 11 and 6 days quicker than the average anthesis date, while Fathom quickened its maturity for the final two planting times resulting in similar maturity to Hindmarsh and IGB 1101. Navigator and Urambie were the longest maturity varieties at all 3 planting times and were 10, 5 and 6 days longer than the average flowering time.

Delaying planting date from the 26^th April (TOS 1) to the 20^th May (TOS 2) had no significant effect on yield when averaged across 11 barley varieties in 2012 (Table 5). TOS 3 however, resulted in a 33% yield loss compared to TOS 2 (Table 5). The first TOS flowered 18 days earlier than the optimum flowering window, while TOS 2 and 3 flowered in the optimum window. Oxford and Fairview were the highest yielding varieties from TOS 1, both of which are mid to long season varieties.  Fathom, Grange and IGB 1101 also achieved in excess of 6 t/ha from TOS 1, which is interesting given that both Fathom and IGB1101 are quicker varieties. Fathom achieved the highest grain yield for TOS 2, while Commander, Navigator, Grange, Oxford and Hindmarsh all had yields between 6.25 and 6.43 t/ha. From TOS 3 Fairview, Fathom, Grange and IGB 1101 were the better performing varieties. Oxford and Navigator are both longer season options, however not as long as Navigator and Urambie, which had similar anthesis dates. Grange is slightly quicker than Commander while Bass and Fairview are similar to Commander in their anthesis dates. IGB 1101, is a similar maturity to Hindmarsh for which it is a likely replacement.

Time of sowing conclusions

The barley TOS trial showed Fathom to be a high yielding variety across all three planting times at both Trangie and Tamworth with a similar to slightly longer maturity than Hindmarsh. Grange and Fairview also performed well across a range of planting times at Tamworth while at Trangie Commander showed the ability to alter flowering time and yielded among the top varieties at all planting times. Oxford was a good yielding variety, particularly from the earlier planting at Tamworth but surprisingly was one of the top yielders from a mid June sowing at Trangie. IGB 1101 appears to be a suitable Hindmarsh replacement with slightly better yields and similar maturity. Of the two new malt varieties released in 2012; Bass has a similar maturity to Commander but is slightly slower from a late planting while Navigator was one of longest varieties trialled being similar to Urambie.   

Acknowledgments

The authors would like to thank all trial co-operators including Dave Denyer, Scott Carrigan, Paul and Charles Tattam and Angus Murchison  for providing trial sites. Technical assistance provided by Rod Bambach, Jan Hosking, Patrick Mortell, Stephen Morphett, Jim Perfrement, Rob Pither and Jayne Jenkins is gratefully acknowledged.

Contact details

Matthew Gardner, NSW Department of Primary Industries
Ph: 02 6763 1138
Fx: 02 6761 2222
Email: matthew.gardner@dpi.nsw.gov.au