Influence of nitrogen and seed rate on the performance of plump grain barley varieties in lower rainfall areas (2005-2007)

Figure 1. Grain yield response to increasing barley establishment density in a) 2005, b) 2006 and c) 2007.

Grain quality

Varietal differences in grain quality were observed in at least two out of every three trials (Table 3), but the ranking of varieties varied across sites. In 50% of trials the grain of the later sown barley had a lower hectolitre weight, higher screenings, higher grain protein, was brighter and harder than the grain of the barley sown three to four weeks earlier. The change in grain quality with delayed seeding was not influenced by variety.

Applied N influenced grain quality in at least two out of every three trials except for grain hardness (Table 3). The three varieties responded similar to increasing N (Table 3). On average there was a 0.9mg decrease in kernel weight, 0.6kg/hL decrease in hectolitre weight, 4% increase in screenings, 1.5% increase in grain protein, a <0.5“L*” decrease in grain brightness and a slight increase in grain hardness across sites when N was increased from 20 to 60 kg N/ha. Screenings was the only grain quality trait whose increase due to applied N was likely to be influenced by date of seeding. At sites where there was an interaction, the increase in screenings due to N was slightly larger at the second date of seeding than the first.

Seed rate influenced grain quality, with the three varieties responding similarly to increasing plant density (Table 2). On average increasing plant density decreased kernel weight (<2mg) and hectolitre weight (<1kg/hL), increased screenings (<5%), decreased grain protein (<1%) and slightly decreased grain brightness (<0.5‘L*’) with little effect on grain hardness. The change in grain quality was only occasionally influenced by when the crop was sown.

Table 3. Likelihood of time of sowing, N rate, seeding rate and their interaction being significant (p<0.05) on grain quality traits across thirteen trials. Significance: rarely = occurred in ≤20% trials, occasionally = 20-40% trials, 50% chance = 40-60% trials, often = 60-80% trials and consistently = ≥80% trials significant.

Economics of changes in crop management

An economic analysis calculating return (yield by price minus costs) was done for each time of sowing, variety, N rate, seed rate and site, using the assumptions in Table 4 and the 2016/17 GIWA Barley receival standards for Western Australia. The per cent of replicate samples meeting BFED1 receival specifications and average return ($/ha) across all 13 sites is presented in Figure 3 for each time of sowing, rate of applied N, seed rate, variety and season averaged across sites and Figure 4 for response to seed rate averaged across times of sowing, varieties and rates of applied N. It should be noted that these trials were not designed to determine the optimum N or plant density for each variety. The treatments chosen were designed to facilitate the discrimination of varietal performance rather than be economic rates of N application and seed rate. Operating costs do not take into account differences in disease control that may be required and the analysis assumes all varieties can be delivered to the same bin. On average (across all replicate samples), 30% of samples met Malt1, 13% of samples Malt2 and 57% of samples were downgraded to feed. Across years, 63% of samples were received as malt (Malt1 + Malt2) in 2005, only 20% in 2006 and 52% in 2007 (Figure 3).

Table 4. Assumptions used in the economic analysis of the thirteen agronomy trials.

Figure 3. a) per cent of replicate samples meeting BFED1 specifications; and b) average return (yield by price minus costs) for each time of sowing, rate of applied N, seed rate, variety and year averaged across sites.

There was a trend towards lower returns when seeding was delayed (Figure 3), but the difference in return between the first and second date of seeding was only significant in four of the 13 sites (Table 3). This reduction in return was associated with a lower grain yield and a higher proportion of the samples downgraded to feed. Varietal returns were generally similar across sites and varietal return was not influenced by date of seeding. Receival as malt barley was similar across varieties, across treatments and varied with season (Figure 3).

Applied N consistently influenced return, with the three varieties responding similarly to increasing N (Table 3). Despite the small yield response to applied N, return decreased with increasing N due to the lower receival as malt barley (Figure 3) and the extra cost of the additional N. The decrease in return with increasing N was generally larger with the second date of seeding than the first (data not shown).

Seed rate consistently influenced return (Figures 3 and 4), with the three varieties responding similar to increasing plant density (Table 3). Seed rate did not influence the per cent of samples received as malt (Figure 3). The increased yield associated by doubling the target density from 75 to 150 plants/m² in this trial series was offset by the increased cost of the seed, resulting in a similar return for the 75 and 150 plants/m² seed rate treatments and a similar per cent received as malt (Figure 3). Increasing the target density above 150 plants/m² did not increase yield, did not influence the per cent of samples received as malt, but increased costs (due to the extra seed) and resulted in a lower return (Figures 3 and 4).

Figure 4. Change in return ($/ha) as the target density increases above the 75 plants/m² seed rate treatment in a) 2005, b) 2006 and c) 2007 averaged across times of sowing, rates of applied N and variety.

Conclusion

In this lower rainfall study the optimum plant density that maximised return was achieved by targeting an establishment density between 75 and 150 plants/m² (Figures 2 and 4). This is similar to and slightly lower than the optimum plant density for malt barley varieties suggested by Paynter (2016) and Paynter et al. (2016a, 2016b, 2016c) in higher rainfall areas. Sowing to establish more than 150 plants/m² resulted in lower returns in this study.

The three varieties sown did not differ in how they responded to increasing seed rate (Table 2 and 3). Therefore the same target density would be recommended for Hamelin, Stirling and Vlamingh if they were still in commercial production for delivery as malt varieties. The suggested target density for those varieties in this study would be 90 to 110 plants/m².

The optimum plant density for grain yield was not correlated with the site mean yield and was slightly higher but not largely changed with delayed seeding (Figure 2). This is similar to the observation of Paynter et al. (2016c) and indicates that growers do not need to change their target density (therefore seeding rate) if seeding is delayed by two to four weeks. In saying that, delayed seeding influenced hectolitre weight, screenings, grain protein concentration and grain brightness more than seed rate (data not shown). As a consequence delayed seeding rather than increased seed rate was more likely to cause downgrading to feed (Figure 4). So whilst growers don’t have to adjust the seeding rate, any delay in seeding date from May to June will likely impact on profitability.

Applied N regularly affected grain yield (often increasing it), grain quality (increasing the per cent of grain received as feed) and return (mostly decreasing profit) (Tables 2 and 3; Figure 4) but did not influence the optimum plant density for grain yield (Figure 2) or influence the response to increasing plant density (Tables 2 and 3). Paynter et al. (2016a) suggested, however, that the optimum plant density for grain yield decreased with increasing N, although the N rate that maximised returns was 30 kg N/ha, not 0 or 90 kg N/ha. In this study the extra yield generated did not compensate for the extra fertiliser cost and the increased downgrading to feed.

Whilst this study reports on the performance of three malt varieties that are no longer segregated in Western Australia, the DAFWA barley agronomy team with co-funding from the GRDC is undertaking new studies with current or potential malt varieties (ie. Compass and La Trobe) in lower rainfall areas. Unfortunately the early autumn and higher growing season rainfall received in 2016 resulted in grain yields (if not affected by frost) higher than our target grain yield of <1.5t/ha. The 2016 low rainfall seed rate series is to be repeated in 2017. In the meantime, the target plant density for malt barley growers in lower rainfall areas (<1.5t/ha) should be at or above 100 plants/m². Whilst lower target densities achieved a similar return in this study, lower densities have been shown to decrease the yield potential of barley in other studies (Paynter 2016 and Paynter 2016a, 2016b, 2016c) and decrease the competitiveness of barley with weeds (Paynter and Hills 2009), without enhancing the prospects of meeting malt barley specifications significantly. The method for calculating the seed rate in kg/ha based on the 1000 kernel weight of the seed batch and the chosen target density is found in Table 5.

References

Paynter BH (2016). Compass and La Trobe: head to head from 25 to 400 plants/m² at York in 2014. 2016 GRDC WA Grains Research Update, Perth WA, 29 Feb and 01 Mar 2016.

Paynter BH, Graham N, Malik R and Hills AL (2016a). National barley agronomy trial series (2012-2014) – Western Australian observations on the interaction between variety, nitrogen and seed rate. 2016 GRDC WA Grains Research Update, Perth WA, 29 Feb and 01 Mar 2016.

Paynter BH and Hills AL (2009). Barley and rigid ryegrass (Lolium rigidum) competition is influenced by crop cultivar and density. Weed Technology 23:40–48.

Paynter BH, Hills AL and Malik R (2016b). How much does seed rate (changing from 50 to 400 plants/m²) influence barley’s performance? 2016 GRDC WA Grains Research Update, Perth WA, 29 Feb and 01 Mar 2016.

Paynter BH, Hills AL and Malik R (2016c). How much is the seed rate (changing from 50 to 400 plants/m²) response of barley influenced by date of seeding? 2016 GRDC WA Grains Research Update, Perth WA, 29 Feb and 01 Mar 2016.

Acknowledgments

Technical support was provided by Sue Cartledge, David Dodge and the DAFWA Technical Support Units at Wongan Hills, Merredin, Katanning and Esperance. This research was located at the farms of Mike Kalajzic, Cadoux; Bruce Mills, Corrigin; Lawry Pitman; Corrigin; Rory Graham, Salmon Gums; Geoff Sanderson, Grass Patch and the DAFWA research farms at Merredin and Newdegate. DAFWA funded this research with co-funding from the GRDC. Rick Graham, NSW DPI and Dion Nicol, DAFWA provided feedback on a draft of this paper.