Managing herbicide resistance

Author: Peter Boutsalis (School of Agriculture, Food & Wine, University of Adelaide and Plant Science Consulting P/L), Christopher Preston, Samuel Kleemann and Gurjeet Gill (School of Agriculture, Food & Wine, University of Adelaide). | Date: 26 Jul 2016

ɸExtra technical comment by Protech Consulting Pty Ltd

Herbicide resistant grass weeds in NSW

GRDC funded random weed surveys of cropping regions across southern Australia are being regularly conducted. The methodology involves collecting weed seeds from paddocks chosen randomly at pre-determined distances. Weed seeds are collected and tested in pot trials the following winter. Random surveys have shown that the greatest herbicide resistance in New South Wales (NSW) is in annual ryegrass. The incidence of resistance in the Forbes-Grenfell-Cowra-Wellington-Peak Hill regions as identified by John Broster of Charles Sturt University (CSU) is presented in Table 1. Resistance to Group A, B and D herbicides was detected. Opportunities for using these latter herbicides however still remain if herbicide resistance testing is conducted to check for herbicide efficacy.

In wild oats, more than half the samples collected exhibited resistance to Topik®, whereas other mode of action herbicides were effective.

Scatter map showing showing paddocks surveyed in NSW

Figure 1: Survey regions in NSW. Each point represents a paddock surveyed.

Table 1: The incidence of herbicide resistance around Forbes-Grenfell-Cowra-Wellington-Peak Hill in 2013 (50 paddocks). The data represents the percentage (%) of samples (one sample represents one paddock) where 20 per cent or greater survival was recorded in a pot trial. Each herbicide was applied at the field rate with the recommended adjuvants. Trials conducted by John Broster, CSU, Wagga Wagga.

Trifluralin Hoegrass® Hussar® Intervix® Select® Glyphosate Boxer Gold® Sakura®
Ryegrass 23 82 56 33 19 0 0 0
Wild Oats Topik®

Getting the best out of pre-emergent herbicides

In the absence of effective post-emergent herbicides, ryegrass management has to rely on pre-emergent herbicides and non-chemical tactics. In getting the best out of pre-emergent herbicides, it is important to understand some of their characteristics and how they will perform under different conditions.

Trifluralin and Stomp® (pendimethalin) have low water solubility so tend to stay where they are applied, therefore, they need to be placed in close proximity to the weed seed. These herbicides are also volatile, so need incorporation shortly after they are applied to avoid losses. They bind tightly to organic matter including stubble. If there is too much stubble, some will need to be removed to get these herbicides to work effectively.

Boxer Gold® (prosulfocarb + S-metolachlor) has high water solubility and will move readily through the soil. It typically requires 5-10mm of rainfall over a week to activate the herbicide. If heavy rainfall occurs after application, some crop damage may occur. Wheat is more sensitive than barley, so damage will be greater in wheat crops. Boxer Gold® has medium binding to organic matter, so will move more readily in low organic matter soils. If heavy rainfall occurs after application, some crop damage may occur. Boxer Gold® has relatively short persistence, so late emerging weeds will be a problem in high rainfall zones.

Sakura® (pyroxasulfone) has lower water solubility, making it less likely to move in soil. It requires more rainfall than Boxer Gold® to activate; 10-15mm. Sakura® is not bound tightly to soil, but its low water solubility means that it is normally not highly mobile. However, in soils with low organic matter or after high rainfall events, some crop damage may occur. Sakura® is active for an extended period of time.

Avadex® Xtra (triallate) on its own will only control ryegrass at high concentrations. It is volatile and requires incorporation. Avadex® is more mobile in soil than trifluralin and binds less tightly to organic matter. Avadex® is primarily absorbed through the coleoptile rather than the roots, so controls deeper emerging weeds.

Rustler® and Imtrade Edge 900WG® are the only propyzamide products registered in canola. Propyzamide is similar to Sakura® in its behaviour. It has low water solubility and medium binding to organic matter in the soil. This means it usually does not move far through the profile, but can do so with heavy rain. Canola tends to be sown shallower than wheat, so the herbicide is closer to the crop. Therefore, Rustler® damage to canola is more likely with high rainfall.

We conducted a series of six trials in high rainfall zones to determine how to get the most out of pre-emergent herbicides. Pre-emergent herbicides generally perform better in lower rainfall zones, because there is not extended germination of ryegrass that occurs in high rainfall zones. Therefore, these trials demonstrate the full value of the products. The trials also examined season long control of annual ryegrass by counting the number of seed heads present at harvest. As a rule of thumb, it is necessary to reduce ryegrass populations by 97per cent or more in order to keep the seedbank the same.

The results are in Figure 2 as a box and whiskers plot where the mean is the line in the middle of the box and the whiskers are the range of results. Some of these trials were conducted with populations that are resistant to trifluralin, affecting the performance of this herbicide.

Sakura® on average performed better in these high rainfall environments than did Boxer Gold® or trifluralin. Adding Avadex® Xtra to any of the other pre-emergent herbicides improved control. New registrations for Boxer Gold® will provide increased flexibility in gaining higher control of annual ryegrass populations.

Box and whisker plot showing season long control of annual ryegrass by pre-emergent herbicides

Figure 2: Season-long control of annual ryegrass by pre-emergent herbicides in six trials conducted in high rainfall zones.

Key resistance management considerations

Cross-resistance patterns

Herbicide cross-resistance patterns differ between species. Ryegrass resistant to one ‘FOP’ herbicide is almost always resistant to all other FOPs, even if they have never been used. However, brome can be resistant to Targa® but not to Verdict®, or resistant to Select® but not to Verdict®. Barley grass also tends to follow the same pattern as brome. Furthermore, in wild oats, there are cases where Targa® can control Topik® and Axial® resistant wild oats. Other cross-resistance patterns can give resistance to Topik®, Verdict® and Select® but not to Axial®. In broadleaf weeds such as wild radish, Indian hedge mustard and sowthistle, diverse resistance patterns can occur between one or more herbicides such as Ally®, Gleanɸ, Eclipse®, Intervix®, etc. There are situations where exclusive use of one Group B herbicide (e.g. Gleanɸ) has resulted in broad cross-resistance in a broadleaf weed species so that when another Group B herbicide is used for the first time (even if in a different chemical class e.g. Eclipse® or Intervix®), the weed is highly resistant. Cross-resistance patterns in the same species can also differ between neighbouring paddocks depending on the origin of the resistance. Therefore, resistance testing can allow growers to identify suitable herbicides that are still highly effective on their weed populations.

ɸGlean is no longer a registered product however there are plenty of other chlorsulfuron products registered.

Rotating herbicides

Rotating herbicides aids in reducing the risk of resistance to any one herbicide. If a herbicide is working well, it can be difficult to convince a grower to change to another. Heavy use of Group B herbicides in the upper Yorke Peninsula in South Australia (SA) led to the rapid development of resistance in Indian hedge mustard to all Group B chemistries. Subsequently, rotating and relying heavily on Group I chemistry proved effective in the short-term, but soon after mustard with multiple resistance to Group B and I herbicides became prevalent. Herbicide modes of action must be rotated to keep chemistries working for as long as possible. Even if Group Bs are still effective, rotate with Group C, F, G, H, and I herbicides to reduce the risk of resistance to Group B herbicides.

Herbicide rate

Some herbicides have a rate range in the label. Using higher label rates can often improve weed control. Weeds with weak resistance mechanisms can often be killed with higher rates. This is particularly common for clethodim and glyphosate resistance. Ensuring optimum coverage is essential to maximise weed control. Using high rates with poor coverage can have the same effect as applying a lower rate. To maximise coverage and herbicide performance, it is important to spray during ideal weather conditions with the correct nozzles, speed, water volume, water quality and adjuvant.

Herbicide resistance testing

The only way to determine which herbicides still control a weed population is to conduct a test. Some growers find it convenient to assume they have herbicide resistance to all Group A and B herbicides and use products such as Sakura® or Boxer Gold®, because there is no confirmed field resistance to either product. However, this can be a costly exercise and there may be cheaper older herbicides that still work but have been overlooked. For more information visit Plant Science Consulting Website

Growth stage

Are weeds resistant at all growth stages? The answer to this question is ‘not always’. The type of resistance mechanism present and the growth stage influence herbicide efficacy. Numerous trials have shown that herbicide resistant weeds can sometimes be killed or heavily damaged if treated at the seedling stage. Clethodim has been shown to control some ‘DIM’ resistant ryegrass biotypes at the 2-leaf stage but not at the tillering stage. Improved control of Group A resistant brome has been shown when 2-leaf seedlings are treated vs. poor control of tillering plants. Group I resistant wild radish can often be better controlled at the 2-3 leaf stage than at the 5-6 leaf stage. A common strategy by some growers is to delay application of post-emergent herbicides to maximise germination from the seedbank in order to ‘treat all the weeds’. This strategy is sound where there is no herbicide resistance present. However, reduced control of older plants that are herbicide resistant can occur. For weeds with staggered germination, particularly brome, wild oats and wild radish, multiple herbicide timings (with different chemistries) should be used.

Non-herbicide techniques

Strategies that do not rely on herbicides alone can ‘take the pressure off herbicides’ and reduce the risk of resistance. In most cases, relying on herbicides alone will lead to resistance. Extensive research is occurring in the area of non-herbicide weed control techniques which are often termed integrated weed management (IWM) strategies. Maximising crop competition with weeds is an important factor. Weeds often occupy gaps in crops. Cereals are usually more competitive than broadleaf crops. Barley tends to be the most competitive cereal, whereas durum is the least competitive. Significant differences between varieties of the same crop species can also occur so growing the best suited variety for your region is important. Seed with poor crop vigour will not compete effectively with weeds. Only conducting a germination test will not test for vigour; vigour testing is important. Other factors which can optimise crop seed germination, emergence and development include correct sowing time, correct seeding depth, fertiliser timing and placement, pest and disease management.

If a crop is infested with weeds, urgent action is needed because replenishing the seedbank can be disastrous. Tactics to sacrifice part of the crop to prevent the replenishment of the seedbank can include cutting for hay, silage or green manuring. Harvesting the crop and removing or killing weed seed by the use of chaff carts, Harrington Seed Destructor (HSD) or placing the chaff in windrows for burning will also reduce the weed burden.

Glyphosate resistance

Nationally, the number of species and individual cases of confirmed glyphosate resistance continues to increase with the greatest cases in ryegrass. Most cases of glyphosate resistant ryegrass have been detected in winter grain crops (300 cases), fencelines and roadsides (100 cases each). Resistance in new species such as sowthistle and wild radish is of particular concern.

Scatter plot showing increase in confimed cases of glyphosate resistance in winter weeds over time.

Figure 3: The increase in confirmed cases of glyphosate resistance in winter weeds between 1996 and 2014. Data from Christopher Preston, Glyphosate Sustainability Working Group.

Reduced efficacy of glyphosate

There are numerous reasons for the poor performance of glyphosate, a common one being herbicide resistance. Resistance to glyphosate can range from weak resistance to strong resistance. Plants with weak resistance may be controlled with higher label rates, but this strategy should not be overused because weeds can develop resistance to very high rates. One Victorian roadside population has survived 20L/ha glyphosate in pot trials.

Table 2: Per cent survival (%) of a selection of grower resistance tests from 2013 and 2014 treated with glyphosate (540g ai/L). Data ranked according to per cent survival at 1000ml/ha.

Town State 1000 1500 2000
Wagin WA 5 0 0
Yendon NSW 5 5 0
Griffith NSW 5 5 5
Dowerin WA 5 0 0
Lake Grace WA 10 0 0
Yendon NSW 20 20 0
Yendon NSW 20 0 0
Deniliquin NSW 20 5 0
Temora NSW 20 5 0
Goomalling WA 20 0 0
Goomalling WA 20 20 0
Calingiri WA 20 0 0
Griffith NSW 20 5 5
Griffith NSW 25 0 0
Yendon NSW 40 20 0
Badgingarra WA 50 5 0
Griffith NSW 55 5 5
Ballidu WA 80 70 70
Griffith NSW 80 60 0
Nhill Vic 80 80 5
Naracoorte SA 90 55 0
Calingiri WA 100 90 90

Data courtesy of Peter Boutsalis, Plant Science Consulting

Glyphosate is usually absorbed within 24 hours of application and moves readily in the phloem of actively growing plants. Application of glyphosate in the morning can result in greater uptake than application at night. Greater glyphosate activity is usually observed in actively growing young plants. On larger plants, higher rates are required to maintain good efficacy. Herbicide uptake can be restricted when plants are stressed. Factors that can cause stress include frost, moisture (drought or waterlogging), temperature, nutrition and pest damage. Pot trials have shown that glyphosate activity is often reduced as ambient temperature increases. In ryegrass it has been observed that the optimum daily temperatures for glyphosate activity range between the low teens and mid-twenties. These findings have been observed in unstressed two to three leaved ryegrass growing in pots.

Factors that limit the contact of glyphosate with a target weed include poor coverage (water rates, nozzle selection, applying on very dense populations or wet leaves), poor water quality or application onto dust covered plants. Glyphosate is readily bound to soil particles present as dust or in dirty water.

Applying glyphosate on herbicide resistant weeds that are stressed or exposed to sub-optimum herbicide concentrations (reduced coverage, dust, etc.) can result in poor control. Combinations of factors that reduce glyphosate efficacy on plants with weak resistance mechanisms can elevate the resistance levels. Testing for glyphosate resistance via seed or plants (Quick-Test) can aid in making future weed control decisions. A test can highlight the presence of weak glyphosate resistance and whether higher rates could be effective. The identification of strong resistance can aid in convincing growers to adopt alternative strategies to combat the problem.

Increased resistance occurs when two mechanisms of glyphosate resistance occur in the same plant.

Scatter plot with trend lines showing survival rate of weed resistant herbicide. Box plot with box and whiskers showing range of average survival marked on the trend line.

Figure 4: Glyphosate resistance mechanisms are additive. Dose response of ryegrass populations with a target site mutation, SLR 77, (●), the translocation resistance mechanism, NLR 70, (■), and the F1 cross between SLR 77 and NLR 70 (♦) compared with the susceptible population VLR1 (○).

Improving glyphosate efficacy

Using higher label rates can often improve weed control. Weeds with weak glyphosate resistance mechanisms can often be killed with higher label rates. Additionally, higher rates can help counteract poor application, improve control of older plants, stressed plants or overcome reduced efficacy caused by using poor quality water or treating plants covered by dust. Higher label rates can also improve glyphosate activity of plants exposed to higher temperatures that can arise in early autumn or late spring.

Our findings have shown that although glyphosate resistant weeds are resistant at all growth stages, seedlings are more sensitive than multi-tillered plants. Numerous trials have shown that herbicide resistant weeds are often killed or heavily damaged if treated at the seedling stage. A common strategy by some growers is to delay application of glyphosate to maximise germination from the seedbank in order to 'treat all the weeds'. This strategy can be effective if the weeds are not herbicide resistant or stressed. However, reduced control of older plants that are herbicide resistant can occur if rates are not sufficiently high or weeds are stressed. In weed species that exhibit staggered germination such as brome, wild oats and wild radish, multiple herbicide timings are recommended. The type of resistance mechanism(s) present and more importantly the level of resistance it confers can also influence glyphosate efficacy.


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. 

Contact details

Peter Boutsalis
University of Adelaide, Waite Campus, Glen Osmond SA 5064
0400 664 460

Herbicide resistance testing:Peter Boutsalis, Plant Science Consulting
Herbicide resistance website

GRDC Project code: UA00113, UA00121, UCS00020