Comparing herbicide resistance in New Zealand and Australia

ABSTRACT

Although New Zealand is developing an increasing number of herbicide-resistant weed populations, it has a much lower incidence of herbicide resistance than Australia. Understanding the reasons for these differences may help with future management of herbicide resistance in both countries. Australia is much larger than New Zealand so greater areas of weeds are exposed to herbicides annually in Australia. Another difference is the high frequency of Lolium rigidum in Australian agricultural systems, a species almost absent in New Zealand, and many of the cases of resistance in Australia involve this species. However, species closely related to L. rigidum are increasingly being found resistant to herbicides in New Zealand. Higher rainfall in most parts of New Zealand allows much more crop rotation than is possible in Australia, and greater crop yields give New Zealand farmers more flexibility to use higher application rates and more expensive herbicides in rotation than is feasible in Australia. Dry conditions result in more use of summer fallows in Australia using glyphosate which has caused some of the problems. Selection pressure for resistance still occurs in New Zealand, so herbicide and crop rotation may just be delaying the appearance of resistance in this country.

KEYWORDS:

• Herbicide resistance
• Australian weeds
• New Zealand weeds
• arable weeds
• glyphosate resistance

Introduction

Continuous use of herbicides with the same mechanism of action over many years can lead to herbicide-resistant biotypes of weed species developing within farms (Busi et al. 2021). In a recent review, we showed that cases of herbicide resistance within New Zealand have been increasing steadily (Ghanizadeh and Harrington 2021). The first case in this country involved biotypes of Chenopodium album that evolved resistance to atrazine within maize crops and was first noticed in late 1979 (Rahman et al. 1983). Herbicide resistance was first detected in Australia at about the same time, with a population of Lolium rigidum found to be resistant to diclofop-methyl in 1980 within a legume seed crop (Heap and Knight 1982).

Most papers published about herbicide resistance make use of the International Herbicide-Resistant Weed Database (Heap 2022) when claims are made about the number of herbicide resistance cases in various countries around the world. For a case of herbicide resistance to be listed on this database, generally a dose response experiment needs to be conducted comparing susceptible with resistant populations, and simple screening experiments are not sufficient (Heap 2005). Also, the cases then need to be reported to the database, which does not always happen. Thus, the number of cases recorded on the database is presumably an underestimate of the true number of herbicide resistance cases known by researchers. For example, the estimate for New Zealand on the database in 2022 was 22 cases, compared with the 25 cases reported in 2021 by Ghanizadeh and Harrington (2021), as some of the cases listed in that review were from work in which just single rates of application were used to determine the presence of resistance. But despite these issues, it is generally considered that the number of cases listed on this database gives a reasonable estimate of the extent of herbicide resistance in a country.

Compared with the 22 cases listed for New Zealand in 2022, there were 154 different cases of herbicide resistance reported from Australia (Heap 2022), even though the first cases in each country occurred at about the same time. The objectives of this review are to discuss possible reasons for this 7-fold difference in the number of herbicide resistance cases that have been reported, and also to determine whether New Zealand may soon have as many cases as Australia. We also wish to determine whether there are any lessons to be learned from what is happening in New Zealand that has kept resistance from developing as rapidly as in Australia.

Differences in funding and testing

The larger size of Australia compared with New Zealand may be relevant to the difference in herbicide resistance that has been reported. With Australia at 7.74 million km² compared with New Zealand at 0.27 million km² (WorldData 2022), this means there is 29 times the land area in Australia where weeds can be growing and being exposed to herbicide. Although a proportion of Australia’s land mass is too dry for agriculture, there are still large differences between the two countries in the area dedicated to agriculture. For example, the area of wheat grown in Australia in 2021 was approximately 12.9 million ha (Statista 2022) compared with 43,000 ha in New Zealand (Figure.NZ 2022). The area of grazed pasture in Australia is estimated at 373 million ha (though only 47.5 million ha are improved pastures) (ABARES 2022) whereas in New Zealand the total grazed pasture area is approximately 10 million ha, with 7.7 million ha of improved pastures (Hipgrave 2021).

The current population of Australia is 25,739,000, compared with 5,123,000 in New Zealand (WorldData 2022), so with five times the population, there is likely to be more resources such as more scientists and more universities to investigate problems. Much of the research in Australia conducted on herbicide resistance has been funded by the Grains Research and Development Corporation (GRDC), which receives its funding from levies collected from grain growers and also from the Australian government (GRDC 2022). Although New Zealand has a parallel entity, the Foundation for Arable Research (FAR), there are far fewer grain farmers from which levies can be collected, and much less government funding. Thus until recently, the few weed scientists present in New Zealand obtained little funding to work specifically on herbicide resistance. Research outputs on the topic have increased since a 5-year government funded programme on herbicide resistance began in New Zealand in late 2018, funded by the Ministry of Business, Innovation and Employment (MBIE) with co-funding from other organisations such as FAR.

Partly due to greater funding in Australia and partly caused by increased problems caused by herbicide resistance, there are much better testing services in Australia than New Zealand for farmers wanting to confirm that they have herbicide resistance. For example, Charles Sturt University has operated a commercial testing service since 1991 (Broster et al. 2019) and Peter Boutsalis has also been offering herbicide resistance testing with his Plant Science Consulting business for over 25 years at the University of Adelaide (Boutsalis 2001). The Australian Herbicide Research Initiative (AHRI) based at the University of Western Australia is a research institution dedicated to herbicide resistance research, being first set up in 1999 by Steve Powles with funding from GRDC, and they have also been involved with extensive testing for herbicide resistance (Busi et al. 2021).

In contrast, no formal commercial herbicide testing programme has been available in New Zealand to date, apart from a service offered at Massey University for a few years that was not sustainable due to lack of interest from farmers. Random free testing has recently been occurring as part of the MBIE-funded research, which has shown that herbicide resistance is more common in New Zealand arable cropping than had been realised (Buddenhagen et al. 2021).

Differences in cropping systems

Although a greater area is used for growing wheat in Australia than in New Zealand, the lower rainfall in Australia and lower fertility soils in many areas result in lower average yields in Australia. In 2021, the average wheat yield in Australia was 2.6 t/ha (Statista 2022) compared with 9.9 t/ha in New Zealand (NZGSTA 2020). According to Llewellyn et al. (2016), the average weed-free wheat yield varies from 1.25 t/ha in the eastern West Australia area to 4.21 t/ha in Tasmania and the high-rainfall regions of Victoria. The higher productivity of crops such as wheat under New Zealand conditions has several implications for development of herbicide resistance. The lower return per hectare in many parts of Australia puts pressure on farmers there to use only the least expensive herbicides and often at lower application rates per hectare than are used in New Zealand (Preston et al. 1999). The continual use of low rates of a limited number of herbicides makes development of herbicide resistance more likely (Neve and Powles 2005). In New Zealand, higher herbicide rates and rotation of more expensive herbicides with different modes of action are more feasible. Also, a greater range of alternative crops can be grown in rotation with wheat in New Zealand than in many parts of Australia, making it easier to use alternative herbicides to deal with the build-up of any problem weed species (Sharma et al. 2021). Broster et al. (2012) made similar assertions about why there was less herbicide resistance found in Tasmanian cereal crops than elsewhere in Australia, with higher herbicide rates and greater crop rotation occurring there than in drier parts of Australia.

Another difference between Australian and New Zealand cropping systems is the high prevalence of L. rigidum as a weed throughout Australia, a species rarely found in New Zealand (Champion et al. 2012). Many of the cases of herbicide resistance in Australia involve this species, including the first ever case of herbicide resistance in Australia (Heap 2022). Randomised surveys conducted in various parts of Australia have shown how L. rigidum is the main weed species causing herbicide resistance problems in cereal crops (Boutsalis et al. 2012; Owen et al. 2014; Harries et al. 2020; Broster et al. 2022). Assessments of L. rigidum populations from across Australia where resistance had been reported emphasised how this most damaging weed has also become resistant to a large number of herbicides (Busi et al. 2021).

However, several of the recent cases of herbicide resistance detected in New Zealand have involved the closely related L. multiflorum and/or L. perenne, with resistance detected to glyphosate (Ghanizadeh et al. 2013), amitrole and glufosinate (Ghanizadeh et al. 2015), haloxyfop, iodosulfuron, chlorsulfuron, pyroxsulam and thifensulfuron (Gunnarsson et al. 2017; Ghanizadeh et al. 2022b), clethodim (Ghanizadeh et al. 2019) and pinoxaden (Ghanizadeh et al. 2022a). These species are also all commonly sown as pasture species, resulting in high densities of L. rigidum seedlings being found in agricultural land within Australia and likewise high densities of L. perenne and L. multiflorum seedlings in New Zealand agriculture. Note that Broster et al. (2012) considered that some of the resistant ryegrass populations reported in Australian surveys are likely to be L. perenne, L. multiflorum or hybrids of these species as they are difficult to differentiate from L. rigidum.

The dryness that occurs in many parts of Australia has led to summer fallows often being used during dry periods prior to planting crops, a practice seldom used in New Zealand. To help conserve moisture in the soil during this fallow, weeds are usually controlled with herbicides, and glyphosate is the herbicide most commonly used (Chauhan and Jha 2020). This use of glyphosate has led to a number of cases of glyphosate-resistance developing in weed species (Heap 2022).

The crops grown in the two countries differ to some extent which can also result in different herbicide resistance issues. Herbicide resistance has developed in weeds within crops such as canola, cotton, sorghum, chickpeas, sugarcane and lupins in Australia (Heap 2022), whereas these crops are not very common in New Zealand. Conversely, several of the herbicide resistance problems that have developed in New Zealand have been in maize, with C. album becoming resistant to atrazine (Rahman et al. 1983) then also to dicamba (Rahman et al. 2014), to atrazine in Persicaria persicaria (Rahman and Patterson 1987) and more recently Digitaria sanguinalis has become resistant to nicosulfuron (Buddenhagen et al. 2021). Resistance of Solanum nigrum to cyanazine has also been reported in New Zealand sweet corn crops (Harrington et al. 2001). No resistance of weeds to herbicides has been recorded for maize or sweet corn in Australia (Heap 2022).

Another difference between the countries is that genetically modified crops are allowed to be used in most states in Australia if authorised by a government body that oversees their use (including compulsory stewardship programmes) (OGTR 2021b) but are banned in New Zealand. In USA, where few restrictions exist on use of such crops, many of the cases of resistance to glyphosate in weed species have developed due to overuse of this herbicide in crops which are genetically modified to be tolerant of glyphosate, creating high selection pressure for resistance (Brunharo et al. 2022). In Australia, two crops are licensed to be grown commercially which have been genetically modified to tolerate herbicides: cotton and canola (OGTR 2021b). Tolerance to glyphosate is the main trait present, though some cultivars also have tolerance to dicamba or glufosinate, and over 99% of cotton grown in Australia is now genetically modified (OGTR 2021a). Cultivars of cotton tolerant of glyphosate have now been used since 2000, and at least five weed species have become resistant to glyphosate within cotton crops (Werth et al. 2021). Glyphosate-tolerant canola is not grown as extensively, and there are also canola cultivars available into which tolerance of other herbicides has been introduced either transgenically or conventionally to help reduce herbicide resistance developing in weeds (Dhammu et al. 2020).

Growing crops in New Zealand for which genes have been introduced through conventional breeding is permitted. Several Cleancrop^TM brassica crops which were developed within this country are available that are tolerant of chlorsulfuron (Dumbleton et al. 2012). To date, no weeds are known to have developed resistance to this herbicide through use of these crops within New Zealand, though resistance to chlorsulfuron can occur fairly readily (Heap 2022).

Similarities in glyphosate use

As already discussed, weeds have developed resistance to glyphosate in Australia due to extensive use of this herbicide in summer fallows and glyphosate-tolerant cotton crops. The first cases of any weed species in the world becoming resistant to glyphosate were reported from Australia after L. rigidum developed resistance to glyphosate in 1995 in two parts of Australia after different uses of the herbicide. One case was in a New South Wales orchard following 15 years of applying the herbicide two to three times annually (Powles et al. 1998). The other case was in Northern Victoria where glyphosate had been used for direct drilling in a continuous cropping system also for 15 years (Pratley et al. 1999). Since then, glyphosate resistance in this weed has also been reported from many other parts of Australia including in numerous vineyards, often with multiple resistance to other herbicide groups (Preston et al. 2009).

Although New Zealand has very little summer fallowing and no glyphosate-tolerant crops, it does have many orchards and vineyards. Resistance to glyphosate was first reported from vineyards in New Zealand in 2013, with resistance detected in both L. multiflorum and L. perenne in several vineyards in the Marlborough region following heavy reliance on glyphosate without many other herbicides being used during the year (Ghanizadeh et al. 2013). Resistance to glyphosate in these two species has been found to be more widespread in Marlborough vineyards since then (Buddenhagen et al. 2022), with some also resistant to amitrole or glufosinate (Ghanizadeh et al. 2015). Vineyards in other parts of New Zealand also now have glyphosate-resistant ryegrasses (Buddenhagen et al. 2022).

To date, 38 cases of resistance to glyphosate in Australia have been reported to the International Herbicide-Resistant Weed Database, and although many of these are cases in fallows and vineyards, a number are also reported from golf courses, around buildings, along fencelines and on roadsides (Heap 2022). Glyphosate is also used in New Zealand for controlling weeds near buildings, fencelines and roadsides but no cases of resistance have yet been reported from anywhere other than vineyards. Given the delay of 15 years between the first cases being reported in Australian and New Zealand vineyards, presumably cases will soon be detected elsewhere in New Zealand. As with Australia, herbicide resistance strategies (Preston 2010) have been developed over time in New Zealand to advise herbicide users on how to avoid resistance, including a strategy on glyphosate use (Harrington et al. 2021).

Differences in pasture systems

Weed control practices in Australian pastures tend to differ from those used in New Zealand, and so the herbicide resistance problems differ as well. Of the seven cases of herbicide resistance reported in Australian pastures by Heap (2022), six involve grass weeds. The practice of applying simazine to permanent pastures in winter to selectively control annual grass weeds has led to resistance to this herbicide developing in Vulpia bromoides (Ashworth et al. 2016), a practice not used in New Zealand pastures although is used in some lucerne crops. The main grass weed actively controlled in New Zealand pastures is barley grass (Critesion spp, formerly Hordeum spp), for which no herbicide resistance has been reported to date within this country. Although there is resistance to herbicides in barley grass within Australia, it is through use of paraquat within lucerne (Powles 1986) or ACCase-inhibiting herbicides in crops such as field peas (Shergill et al. 2016) that resistance has developed. Although this resistant barley grass will then grow in pastures, these herbicides are not traditionally used in pasture to control the weed grass as they would also kill pasture grasses, and also barley grass is often considered a useful pasture species in Australia. Likewise, there are many herbicide-resistant L. rigidum plants that grow in Australian pastures for which resistance has developed to selective herbicides in crops (Heap 2022). However, this grass species is also a desirable pasture species in grazed situations (Bajwa et al. 2021) so this resistance poses no problems while it is growing within pastures.

Grass weeds that are sprayed with herbicide in Australian pastures are larger, invasive, non-palatable species such as Sporobolus fertilis, Nassella trichotoma and Eragrostis curvula, all of which have developed resistance to flupropanate, with cross-resistance to 2,2-DPA (both Group 15 herbicides) in S. fertilis (Ramasamy et al. 2008; Powells 2022). Flupropanate has only recently been introduced to New Zealand for control of N. trichotoma (Lusk et al. 2017) and no resistance has been reported to it yet, though a population of the closely related Nassella neesiana was reported as resistant to 2,2-DPA in New Zealand in 1992 (Hartley 1994).

More widespread herbicide resistance problems within New Zealand pastures have developed in broad-leaved weed species following many years of phenoxy herbicide application. Carduus nutans developed resistance to MCPB, MCPA and 2,4-D in a number of sheep farms during the 1980s (Harrington 1989), and Ranunculus acris also developed resistance to MCPB and MCPA in dairy pastures at this time (Bourdôt and Hurrell 1988). R. acris later developed resistance to flumetsulam and thifensulfuron as well (Lusk et al. 2015). Carduus pycnocephalus has also been recorded as developing resistance to phenoxy herbicides in New Zealand (Ghanizadeh and Harrington 2019). Despite phenoxy herbicides also being used in Australian pastures, no resistance to this group of herbicides has been recorded. The only broad-leaved pasture weed in Australian pastures to be listed as resistant to selective herbicides to date is Echium plantagineum which had developed resistance to chlorsulfuron by 1997 (Heap 2022).

The reason for New Zealand having more cases of resistance to phenoxy herbicides in pasture weeds than Australia despite the much larger area of pastures in Australia presumably relates back to the greater returns per hectare from pasture production in New Zealand because of the moister climate. For example, the average dairy production in New Zealand was 1200 kg milk solids (MS)/ha/year in 2020/2021 (DairyNZ 2022), compared with 520 kg MS/ha/year in New South Wales in the same year (Dairy Australia 2021). Thus it is more feasible to have higher costs of production in New Zealand, allowing more frequent spraying of pastures for weed control and using higher application rates. In Australia, the main use of phenoxy herbicides such as 2,4-D is for spray-grazing, in which low rates are applied to pastures to increase palatability of weeds to animals, and the weeds are then grazed hard (Dowling et al. 2000). The less frequent use of herbicides in Australian pastures and the lower application rates may not exert sufficient selection pressure for herbicide resistance to develop. Worldwide, there have been relatively few cases of weeds developing resistance to phenoxy herbicides compared with some other herbicide groups, especially given that they have been on the market since the late 1940s (Busi et al. 2018).

The high returns per hectare from dairy production in New Zealand encourages many dairy farmers to grow maize as a supplementary crop (Densley et al. 2011). The costs involved with optimising production from supplementary crops by applying herbicides are usually considered minimal compared with the importance of having the extra feed obtained from weed-free maize to increase dairy production. Thus the profitability of the New Zealand dairy industry may result in more maize-growing and herbicide use in maize than in Australia, leading to the greater herbicide resistance problems in maize in New Zealand.

Differences in impacts of resistance

Listing a case of a weed species developing resistance to a herbicide on the International Herbicide-Resistant Weed Database does not reflect how much impact it has on agriculture and society in that country. The greater number of cases listed for Australia compared with New Zealand suggests that resistance is more of a problem in Australia. Information about each case on the database can have follow-up information inserted regarding the extent of the problem in a country, though usually there is little indication of the impact caused by the case.

For Australia, the greatest impact of herbicide resistance does appear to be for the grain cropping industry, and L. rigidum is one of the main drivers for resistance problems (Busi et al. 2021). The impact of the widespread nature of multiple-herbicide resistance within the Australian cereal cropping regions has caused major changes in farming practices in these areas. Many farmers are now trying to remove weed seeds at harvest using chaff carts to collect them, or baling the chaff and feeding this to animals away from cropped land, or by putting this chaff into narrow windrows then burning this (Walsh et al. 2018). Growers are more likely to use herbicide rotation, increase herbicide rates and use double knockdown methods where a second treatment of herbicide is used to kill survivors from the first application (Height et al. 2022). Land use has changed to deal with herbicide resistance, with a move away from having any pasture in the rotation and increased use of oilseed crops such as canola in which a greater range of pre-emergence herbicides can be used to deal with resistant grass weeds (Harries et al. 2020).

Thus far, herbicide resistance has had a minimal impact on farming practices in New Zealand (Ghanizadeh and Harrington 2021). Resistance to herbicides used in maize crops has usually just resulted in changes in the herbicides chosen by growers due to the wide range of products available with different modes of action. Dairy farms with phenoxy-resistant R. acris generally just switched to using flumetsulam, though problems may soon occur as more farms develop resistance to both MCPA and flumetsulam (Jackman et al. 2022). Low rates of clopyralid were added to phenoxy herbicides where C. nutans developed resistance, though this probably damaged white clover (Trifolium repens) in many of the affected pastures (Rahman et al. 1994). Resistance of Solanum spp. to paraquat in sweet potato (Ipomoea batatas) crops (Lewthwaite and Triggs 2009) has not affected this minor industry much, with more use now being made of bentazone mixed in with the paraquat to help solve that problem. The Soliva sessilis in New Zealand turf resistant to triclopyr, clopyralid and dicamba (Ghanizadeh et al. 2021b) is still a fairly minor problem, although it is spreading and bentazone is being used to help deal with that weed.

The increasing frequency with which glyphosate-resistant Lolium spp. are being found in vineyards (Buddenhagen et al. 2022) is causing concern for that industry, with discussions currently underway as to how weed control systems can be modified to stop this becoming more of a problem. Residual herbicides such as indaziflam and flumioxazin are now permitted in these crops to help deal with the problem (Harrington et al. 2021). Likewise, the recent surveys that have identified a number of herbicide resistance issues in the New Zealand arable industry (Buddenhagen et al. 2021) will not initially cause major changes in growing practices, but planning is underway to give growers advice on how to prevent the problems becoming more serious (Espig et al. 2022).

Concluding comments

Although Australia and New Zealand recorded their first cases of herbicide resistance at about the same time, Australia has since then developed many more cases of herbicide resistance than New Zealand. Especially in the Australian arable industry with multiple resistance to herbicides in L. rigidum, the impact of this resistance has led to many changes in management practices to try coping with the problem, including systems to destroy the seeds of resistant weeds. New Zealand has been less affected by herbicide resistance so far, perhaps due to its smaller size, and probably due to the greater flexibility in crop and herbicide rotation that can be used in this country due to its more favourable climate producing higher financial returns per hectare.

However, evidence is increasing that greater resistance problems are developing within the New Zealand arable industry and also the grape-growing industry. Although there is generally good crop and herbicide rotation in the arable industry, this is thought to merely delay the onset of herbicide resistance rather than avoiding it entirely (Busi et al. 2020). Issues within the grape-growing industry have occurred due to an over-reliance on glyphosate for weed control which is now being addressed. The lack of a commercial testing service within New Zealand hampers the monitoring of resistance problems, though the recent 5-year funding boost for herbicide resistance research has allowed more information to be gathered on the state of resistance within this country (Buddenhagen et al. 2021, 2022). Progress has also been made in recent years on several aspects of managing herbicide resistance in New Zealand, such as predicting which weeds will develop resistance next (Ngow et al. 2020; Hulme and Liu 2021; Hulme 2022), finding mutations responsible for resistance so that quick tests can be developed (Ghanizadeh et al. 2021a, 2022b), and also with understanding what affects farmers’ decisions on whether to adopt management practices to avoid resistance (Espig et al. 2022). Once this funding ends in late 2023, it will be important that a herbicide resistance testing service becomes available to replace the free sampling currently occurring as part of the research project so that the incidence of new cases of herbicide resistance can be confirmed if they are suspected. Funding will be required for further research into herbicide resistance in New Zealand so that detailed strategies for dealing with new cases can be developed as they occur in an attempt to prevent the situation in New Zealand escalating to becoming like Australia. Strategies can be devised in some cases though using research findings from overseas countries such as Australia.

Acknowledgement

The authors wish to thank Dr Ian Heap for allowing us to make extensive use of his International Herbicide-Resistant Weed Database, and also the two anonymous peer reviewers who helped to improve the paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Financial support for the authors was provided by the Endeavour fund (C10X1806, Managing Herbicide Resistance) from the New Zealand Ministry for Business Innovation and Employment.