Take home messages

• With ever increasing levels of herbicide resistance, growers need to employ non-herbicide strategies such as increasing crop competition, to aid in the battle against weeds
• Increasing sowing rates to increase crop competitiveness may be cheaper and easier to adopt than alternates such as reducing row spacings
• Growers’ fears of reduced yields or increased screenings from higher seeding rates has been shown in this study to be uncommon
• Lower sowing rates often resulted in lower yields than higher rates
• Increasing crop population had either no impact on screenings or reduced them
• There is little risk to yield or screenings by increasing crop competitiveness by raising sowing rates
• In 99% of yield and 97% of screenings comparisons there was no negative impacts from moving from the lowest population to the next highest (60 to 100 or 30 to 70 plants/m²)

Background

The farmers’ battle with weeds is ageless, with the weapons used continuously evolving. Primitive farmers hand weeded, horses and steel saw the start of mechanised control, with development of herbicides in the mid-20^th century creating a revolution in our farming system.

Weed control has steadily improved over time but this trajectory is starting to slow. Through increasing herbicide resistance, weeds are adapting to survive in our modern farming practices, and systems that rely heavily on herbicides are being challenged.

This was reinforced at recent GRDC National Grower Network (NGN) meetings in Central NSW in 2024 where controlling weeds rated among the highest constraints to production for growers. Improving control of weeds with advanced herbicide resistance is central to farm enterprise sustainability.

Release of new herbicide chemistries cannot be relied upon for weed control going forward, so growers need to look toward alternative, non-chemical tactics as part of an integrated weed control program.

WeedSmart is an industry collaboration aimed at educating growers on options to combat herbicide resistance. Based on significant research and development they have developed the ‘Big 6’ strategy, which comprises of 6 integrated approaches to improve weed control. Increasing crop competition is one of the ‘Big 6’ and is based on a large body of research that shows that a competitive crop can reduce weed numbers and hence seed set, when compared to an un-competitive crop. The Big 6 states that this can be achieved by reducing row spacing, selecting more competitive crops or cultivars, and early sowing. Another option is to increase crop density by increasing sowing rate. Increasing sowing rates is relatively easy, cheap and achievable, compared to changing row spacings via machinery modifications or purchase.

Many growers in the lower rainfall areas of Central West (CW) NSW are concerned that increasing sowing rates may result in lower yields or increased screenings, colloquially referred to as ‘haying off’, ‘blowing up’ the crop or ‘a blowout’ in screenings.

Growers in these lower rainfall areas are commonly targeting below recommended sowing rates as a mechanism to combat these perceived risks. It is less clear if growers appreciate that this action could be influencing weeds or affecting yields and grain quality.

Growers in the medium to high rainfall zones tend to sow at higher rates, but it could be questioned if improved weed control outcomes could be achieved through further increases and that concerns over yield and quality may also be preventing adoption.

Grain Orana Alliance (GOA), with GRDC support, has been conducting wheat population trials in CW NSW, since 2018. Initially this research looked at increasing seeding rates in late sowing scenarios, toward the end of the drought of 2017–19 when late breaks and low stored moisture were common.

These experiments were a precursor to the main body of trials that ran from 2020 till 2024, looking specifically at the interaction between wheat plant population and yield/quality stability over key varieties. The aim was to give growers confidence, in low rainfall environments, that increasing seeding rates to improve crop competition for weed control does not lead to increased risk of lower yields or increased screenings. Further to this, it is suggested that if yields and quality prove to be stable in the low rainfall areas of CW NSW, then growers in higher rainfall areas should also be comfortable with the concept.

Research

This paper details the findings of 12 trials, conducted from 2018 to 2024 in CW NSW. The trials included a range of common wheat varieties, tested at a range of sowing rates and subsequent plant populations to measure the impact on crop yield and quality. These trials were managed to be weed free, so any differences in crop performance would be a function of the population rather than of weed burdens.

The trial was small plot, full factorial randomised complete block design with 4 replicates. Results were analysed by ANOVA with a 95% confidence level.

Varieties

The varieties tested were selected as they were commonly grown in the district and included some contrasting plant types, such as tillering and crop height. The varieties were primarily mid-maturity types as only one time of sowing was possible at each site. Details of the varieties grown in each year is detailed in Table 1.

Table 1. Varieties and the number of trials in each year that each variety was tested.

Sowing rates and plant populations

Four plant populations were targeted (detailed in Table 2). The sowing rates for each variety, averaged over trialling years are listed in Table 3.

To achieve the decided population each variety had sowing rate adjusted to reflect differences in seed size (grains/kg) and germination percentage. As targeted populations increased, adjustments were made to reflect a lower establishment rate.

The low-end targeted populations aimed to cover the lower sowing rates commonly used by growers. The high sowing rates were sufficiently high to thoroughly test concerns over negative outcomes often raised with higher rates but would not be considered commercially relevant.

Table 2. Plant populations targeted over the years of the study.

Table 3. Actual sowing rates (averaged over all years) for each variety and target plant population (kg/ha).

The sites

Concerns around the impact of increasing sowing rate on yield and grain quality are greatest in lower rainfall environments where hotter and drier grain filling conditions are more common. Additionally, low sowing rates are most common in these areas and more expensive or complex weed control options are often not affordable.

For this reason, trial locations were targeted at the lower rainfall zones, generally on the outer fringe of the cropping areas of CW NSW. In 2018, sites were not necessarily selected on these criteria however the low rainfall received along with low levels of stored soil moisture represented what can happen in drier years, even in traditionally higher rainfall locations.

A range of seasonal conditions has been experienced. Both 2018 and 2023 were characterised as drier years with low starting soil moisture, and 2020-22 and 2024 were wetter years with good levels of stored soil moisture. Site details of the long-term average rainfall and the rainfall received in the trial year, both in-crop and over the year are detailed in Table 4.

Table 4. Long term average (LTA) and trial year monthly rainfall, and annual (total) and in crop (April-October rainfall (mm).

Results

Due to the significant amount of information generated by these trials the full details and results of each trial are not included here and can be accessed here.

The findings from this study are presented using comparisons. Each comparison compares statistically analysed responses within any one variety, a location and a year for the 4 plant populations tested. This approach allows for 90 comparisons to be made over the 5 years and 12 trials conducted.

To summarise the impacts on yields and screenings across the population ranges, results have been grouped into 4 characterisations:

• Increasing: yield or screenings increased with increasing populations.
• Decreasing: yield or screenings decreased with increasing populations.
• Nil: no change in yield or screenings from changes in populations.
• Inconsistent: impacts on yield and screenings were variable under increasing populations.

The following categories are considered a positive outcome:

• Yield: either increasing or nil effect.
• Screenings: either decreasing or nil effect.

Established plant populations

Seeding rates were adjusted for seed size, germination percentage and a predicted establishment rate. In most cases the establishment achieved was close to the target plant population, with clear contrasts between the 4 established populations for all sites.

Crop competitiveness

Vegetation index (VI), measured by a handheld normalised differential vegetation index (NDVI) meter was undertaken across all sites and it is suggested to be used as a proxy measure for crop competitiveness.

The effects of plant population on VI is illustrated in Figure 1. In all instances increasing plant populations increased the mean VI (not statistically analysed), suggesting increased competitiveness from increased plant populations. Some varieties had higher VI than others (at the date of measurement). Furthermore, the rate of increases in VI in response to increasing plant populations was also different between varieties.

Impacts on yield

The effect of increasing plant population on yields is set out in Figure 2.

• In 71% of comparisons (64 out of 90), yields either increased as populations increased or there was nil effect.
  • In 42% of comparisons (38 out of 90) yields increased.
  • In 29% of comparisons (26 of the 90) there was nil impact of plant populations.
• In 21% of comparisons (19 out of 90) responses were inconsistent to population increases.
• In 8% of comparisons (7 out of 90), yields decreased.

Impacts on screenings

The effects of increasing plant population on screenings is set out in Figure 3.

• In 68% of comparisons (61 out of 90), screenings either decreased as populations increased or there was nil effect.
  • In 19% of comparisons (17 out of 90) screenings decreased when populations increased.
  • In 49% of comparisons (44 out of 90), plant populations had nil impact on screenings.
• In 20% of comparisons (18 out of 90), the impact was inconsistent.
• In 12% of comparisons (11 out of) screenings increased as populations increased.

Figure 1. Vegetative index (VI) of varieties across locations and years in response to changes in targeted plant population.

Figure 2. Yield (t/ha) of varieties across locations and years in response to changes to targeted plant population. Data points within any one year, site or variety sharing the same lettering indicate they are not significantly different (P 0.05)

Figure 3. Screenings (%) of varieties across locations and years in response to changes to targeted plant population. Data points within any one year, site or variety sharing the same lettering indicate they are not significantly different (P 0.05)

Discussions

Vegetative index

Crop VI increased consistently with the increasing populations tested suggesting any increase in sowing rate has the potential to improve crop competition against weeds. However, it is suggested the most effective gains would be achieved from increases from the lowest populations tested.

Varietal differences were also measured for VI with some varieties showing higher VIs for the same sowing date, suggesting some may be better for weed competition earlier in the crop cycle which is when weeds first establish and compete the most.

Additionally, the rate of VI change with increasing plant populations also differed between varieties suggesting the competitive nature of some may be more responsive to increasing populations than others.

Yield

In the context of increasing plant populations primarily to increase weed competitiveness the impact on yields was shown to be a good outcome. Overall, 71% of comparisons showed increasing plant populations (and weed competitiveness) to result in either no impact on yield or to increase them. Yield increased in 42% of comparisons with an average of 0.67 t/ha improvement or up to 1.7t/ha in some comparisons in response to increases in populations. Yields were improved most frequently when the population was increased from 30 to 70 plants/m². However, in many comparisons yield gains were achieved over multiple increases in population.

The mechanism by which this improvement was achieved was not investigated in this study but may likely be related to higher populations achieving increased biomass and tiller production, enabling the crop to take full advantage of the resources (e.g. soil, nutrition and water) available.

In most comparisons, further increases in yields in response to populations >70 plants/m² were not common. This agrees with many other bodies of research.

Nineteen yield comparisons (20%) had inconsistent effects. Yields went up, down or were neutral in response to increasing populations. However, in all but one of these, the yield at the highest populations (150–180 plants/m²) was not different to or higher than the lowest population, challenging the perceived risk that high sowing rates result in lower yields. In this study, sowing wheat at a high sowing rate was equal to or better than a low sowing rate (30 plants/m²).

In the 8% of comparisons where yield reductions occurred, all were sown very late (5 July) in the 2018 drought year. In 4 of these treatments, yield reductions were <130 kg/ha. In the 3 other comparisons yield reductions were up to ~430 kg/ha (or ~22%). The lowest yield was achieved from the highest, but arguably extreme, population of 180 plants/m², established from sowing rates of 96 to 108kg/ha. For the other 3 more moderate populations (60, 100 and 140 plants/m²) tested yields were not different. Furthermore, with the delayed sowing date of 5^th July, these rates could be considered even more extreme.

As such, growers increasingly targeting very low populations should not interpret this as evidence that increasing rates to a more moderate level is risky for yield and screenings.

Only one comparison out of 90 demonstrated a yield decrease with increasing plant populations from very low to more moderate populations.

Screenings

In 68% of screenings comparisons, there was either no impact from increasing plant population or a reduction in screenings, suggesting growers could either benefit from or find no impact on wheat quality from increasing sowing rates.

For 20% (n=18) of comparisons there were inconsistent impacts on screenings, but the effect was often small, varying only ±2%. In only 2 comparisons did screenings vary more than 5%, both in Suntop. In 15 comparisons screenings were lower at the highest populations compared to the lowest.

In 12% (n=11) of comparisons screenings increased with population. In only 2 comparisons did the variation in screenings exceed 4%, and one with 8.4% variation across the range of populations. In only 4 out of 11 comparisons where screenings increased with population, would bin grade have changed in response to changes in screenings alone. As the data has shown in most comparisons the impact of increasing populations has either been beneficial or neutral. Of the comparisons with variable or negative outcomes the impacts have been relatively small in terms of the level of influence over economics, where binned grade would have changed in response.

In a more practical consideration of the plant populations tested, there was almost no evidence of downside risks to yield and screenings when considering moving from lower populations to more moderate ones. In 99% of yield comparisons and 97% of screenings comparisons there was no negative impact from moving from the lowest population to the next highest bracket (60 to 100 or 30 to 70 plants/m²). In the very dry 2018 conditions there was only one case where there was a yield decline (~60 kg/ha) when moving from 60 to 100 plants/m² and one case where screenings increased, which was less than 2%. This should give growers in lower rainfall environments confidence to move towards the higher target plant populations recommended by NSW DPIRD in these areas, i.e.  70 – 90 plants/m².

Summary

Increasing plant populations through increased sowing rates consistently increased crop vegetation index which implies improved competitiveness. Regardless of what seeding rates growers in the lower rainfall zones are currently using and considering many are employing low rates now, any increase to them will likely bring improved competition against weeds.

In this series of trials, increasing plant population showed in most cases (68% and 71%) to either have a positive or neutral influence on both yield and screenings. In a further ~20% of cases, both yields and screenings showed inconsistent responses to increasing populations, but the interrogation of the data does not support the likelihood of the negative outcomes’ growers are often concerned with.

Of the cases where there was a negative impact from increasing populations on yield or screenings the impacts were relatively small, particularly when compared to the potential upside from increasing plant populations.

A clear learning from this set of 12 trials is that growers should increase sowing rates to >30 plants/m², as yield loss at this sowing rate was common, even in lower rainfall year environments. Increasing populations above 70 plants/m² may result in higher yields but the main benefit is improved weed competition. Therefore, as recommended in the WeedSmart Big 6, growers can increase their sowing rates without excessive risks to yields and screenings.

Acknowledgements

The research undertaken as part of this project is made possible by the significant contributions of growers through both trial cooperation and the support of the GRDC, the author would like to thank them for their continued support.

GOA would like to acknowledge the support from growers and advisors in this study through identification and provision of trial sites, but also their engagement and guidance in this work.

Contact details

Maurie Street
Grain Orana Alliance
maurie.street@grainorana.com.au

Ben O'Brien
Grain Orana Alliance
ben.obrien@grainorana.com.au

Date published
February 2025

Varieties displaying this symbol are protected under the Plant Breeders Rights Act 1994.