Grains Research and Development

Date: 10.02.2015

Biology and management of summer weeds

Author: Gurjeet Gill, Duc The Ngo, Ben Fleet and Chris Preston

Gurjeet Gill, Duc The Ngo, Ben Fleet and Chris Preston,

School of Agriculture, Food & Wine,University of Adelaide.

GRDC project code: UA00134, UA00149

Keywords: fleabane, windmill grass, Feathertop Rhodes grass, weed management.

Take home messages

  • Fleabane continues to be an important summer weed in South Australia but its severity varies considerably between seasons.
  • Normal summer weed control rates of glyphosate tend to fairly ineffective and quite variable on fleabane; use of double-knock with paraquat provides consistently high level of weed control.
  • Windmill grass and Feathertop Rhodes grass are continuing to spread across the agricultural landscape in South Australia. At this stage these weed species are mainly present along roadsides but their seeds can be easily dispersed by wind.
  • Windmill grass and Feathertop Rhodes grass possess many biological features (e.g. high glyphosate tolerance, short life-cycle, rapid growth rates) that could make these species serious summer weeds in southern Australia.


Summer weeds are an on-going problem across all winter cropping regions of Australia. Of course the level of weed infestation and species encountered vary over sites and seasons. The benefits of conserving soil water over the summer-autumn period by controlling summer weeds have been seen repeatedly on many farms as below-average spring rainfall has been a regular occurrence in southern Australia. In summers with infrequent and small amounts of rainfall, farmers often wonder if the weeds would simply die from water stress or do they have to spray them out with herbicides.

Summer weed species

Some summer weed species (e.g. caltrop, heliotrope and melons) have highly persistent seed-banks and occur in fields on a regular basis. In contrast, other species such as fleabane and common sowthistle have very short-lived seed-banks and rely on frequent good seasons for seed production. A few years ago, the combination of wet springs and good summer rainfall led to serious infestations of fleabane on many farms in South Australia. The subsequent run of dry spring and summers led to a decline in fleabane populations on farms to the point that researchers have had difficulty in regularly finding suitable trial sites. Over the last 2-3 years, common sowthistle has appeared as a serious summer weed in many paddocks. This species has two distinct periods of establishment; one in late autumn/early winter and another in spring/early summer. Absence of seed dormancy in sowthistle enables it to rapidly establish after rainfall events in spring/summer. Sowthistle appears to be highly tolerant to heat and water stress, which enables it to persist over the summer months. It is also very tolerant to glyphosate when plants are water stressed. Management of sowthistle is becoming more difficult due to high incidence of group B resistance and recently some populations have been shown to have developed group I resistance (Preston, pers. comm.).  Sowthistle has also evolved resistance to glyphosate in the Northern region. Marshmallow is another winter weed species, which appears to be becoming more common over summer months.

In this paper, we will focus on our recent research on fleabane establishment and control as well as on the biology of windmill grass and Feathertop Rhodes grass.

Results and discussion


As stated earlier, fleabane seeds possess little to no seed dormancy so the germination in the field is largely controlled by temperature and moisture. In SA fleabane seeds start germinating in late winter to late spring period but initial seedling growth rate is very slow. Research undertaken has shown that fleabane seedlings are highly sensitive to crop competition and under moisture stress conditions in spring there tends to be extremely high seedling mortality. If there are bare patches in the paddock due to seeding problems or setback due to other factors, then these areas would be ideal for fleabane establishment.

Fleabane has higher tolerance to glyphosate than many other weed species. Often glyphosate rate of two L/ha or greater is needed to control fleabane; double-knock with paraquat can provide consistent weed control but obviously at a greater cost to the grower. The level of tolerance of fleabane to glyphosate is highly dependent on its growth stage. Plants at seedling stage (rosette) can be easily killed with low rates of glyphosate but a four-fold greater dose might be required once the weed has reached stem elongation stage. However, our results have also shown large differences in glyphosate efficacy on fleabane over seasons. For example, glyphosate alone @ 2 L/ha provided modest 55% weed kill in 2012 but gave 97% control in 2014. Application of double-knock with paraquat after glyphosate treatment tends to ensure high weed kill across seasonal conditions.

Table 1. The response of fleabane to glyphosate in field trials over two different years.

Glyphosate (L/ha) Glyphosate alone Glyphosate + double-knock
  2012 trial 2014 trial 2012 trial 2014 trial
1 30 - 54 -
2 55 97 82 100
3 89 - 95 -
4 93 100 97 100
2 L/ha + 2,4-D 1.4 L/ha 50 84 84 100

Windmill grass and Feathertop Rhodes (FTR) grass

These two Chloris species are major weed species in Queensland and have been rapidly spreading along the transport corridors in SA. At present both of these species are mainly present along roadsides but they can be easily dispersed by wind, especially FTR grass. Both of these are C4 species, which means they can have very rapid growth rates under warm and sunny conditions.

As shown in Table 2, seed germination in both of these species is strongly stimulated by exposure to light. Therefore, burial by tillage is expected to inhibit germination in these weed species but may extend the persistence of their seed-bank. Our studies showed that seed dormancy in both species was relatively short and all viable seeds were able to germinate in spring and early summer (Figure 1).

Table 2. Effect of light on germination of windmill grass (CT) and Feathertop Rhodes grass (CV) seed collected in June 2013 and tested in January 2014. Different letters after the means indicate significant differences at P=0.05.

Population Germinability (%)
24 h dark 12 h light/dark
Mean S.E* Mean S.E
CT2 15.0 b 5.0 44.0 c 3.7
CT3 2.0 a 1.2 77.0 d 3.0
CV4 35.0 c 3.4 78.0 d 1.2
CV5 17.0 b 5.3 72.0 d 6.5

*S.E: standard error of mean

Figure 1. Germinability of windmill grass - C. truncata (a) and Feathertop Rhodes grass - C. virgata (b) seed collected in June 2013, and tested monthly from June 2013 to February 2014. Vertical bars indicate standard error of the mean.

Figure 1. Germinability of windmill grass - C. truncata (a) and Feathertop Rhodes grass - C. virgata (b) seed collected in June 2013, and tested monthly from June 2013 to February 2014. Vertical bars indicate standard error of the mean.

Some additional findings on the behaviour of windmill and Feathertop Rhodes grass include:

  • Short persistence of the seed-bank; complete exhaustion within 12 months
  • Base temperature for germination was 10oC for windmill grass but was much lower at 4oC for FTR grass. Therefore, FTR grass could germinate earlier under cooler conditions than windmill grass.
  • Windmill grass is a short-lived perennial species which can regrow from the crown of the original plant. This is an advantage for this species in rapidly establishing once ambient temperature is favourable for its growth.
  • Even though both windmill and FTR grass are C4 grasses, FTR grass has a much greater growth rate and can produce 5-6 fold greater shoot biomass than windmill grass. If this species was to become established in paddocks in SA then it could rapidly use up precious soil water over summer.
  • Both of these weed species are known for being difficult to control with glyphosate. Recent studies at University of Adelaide have confirmed presence of resistance to glyphosate in some populations from NSW.
  • In some fields, windmill grass was found to be growing well along fence lines where glyphosate had been used for weed control and had effectively removed other weed species.
  • Growers are urged to carefully monitor their properties for incursion by these wind dispersed species.


Research reported here on fleabane was undertaken in a current GRDC funded project (UA00134, UA00149); research on windmill grass and Feathertop Rhodes grass is part of a PhD project funded by VIED and UA.

Contact details

Gurjeet Gill
University of Adelaide
(08) 8313 7744