NatureScot Research Report 1307 - Investigate the issue of herbicide usage in nature conservation and the potential impacts on biodiversity including adverse effects on non-target species - Phase 1 - Review and evidence screening

Year of publication: 2024

Authors: Dr Sarah K Cook

Cite as: Cook, S.K. 2024. Investigate the issue of herbicide usage in nature conservation and the potential impacts on biodiversity including adverse effects on non-target species - Phase 1 - Review and evidence screening. NatureScot Research Report 1307.

Keywords

herbicide; nature conservation; biodiversity; non-target species

Background

In nature conservation, some plant species may pose a threat to conservation interests, and management plans might include measures to control them or prevent their occurrence. Invasive non-native species (INNS) can reduce populations of other species by direct competition and can alter entire ecosystems. In some situations, native species might need to be controlled if their population is at a level whereby nature conservation objectives are being compromised.  This can occur as a result of inappropriate management practices having been applied historically, or other factors such as adjacent land management or climate change. 
 
However, the use and effect of herbicides for conservation management is not widely documented or widely understood. This lack of knowledge may have implications for management particularly of sensitive features that may be susceptible to negative impacts from herbicides.  This literature review aims to better understand usage of the key herbicides, and the impacts of these on non-target species, and the gaps in evidence for the commonly used herbicides in conservation management situations.

Main findings

• Herbicides are rigorously evaluated in order to be registered but the assessments are targeted primarily to ensure the safety of workers applying the product, consumers of treated produce, residents living adjacent to application sites and others who may be passing-by at the time of treatment. As part of the approvals process a further set of tests is done on how the herbicide might move through the environment once it has been applied to assess the impact on the environment, in particular on water bodies and wildlife.
• Notwithstanding the above, the set of tests is limited in scope, and this literature review has found studies showing harmful effects of herbicides on a range of different organisms, and that these effects are sometimes significant and serious. 
• The amount of peer reviewed literature available is variable depending on the herbicide and the length of time it has been on the market. Some herbicides are subject to public controversy.
• Indirect effects of pesticides are more challenging to identify. Registration may in the future require more evidence to avoid the effects of indirect links from emerging subsequently.  
• There is sufficient evidence to conclude that pesticides have negative indirect effects on some elements of biodiversity.

1. Introduction

Herbicides are used in nature conservation for a variety of reasons. In this study, the most frequent recorded use of herbicides in nature conservation, was to do with where some plant species may pose a threat to conservation interests, and management plans might include measures to control them or prevent their occurrence. Invasive non-native species (INNS) can reduce populations of other species by direct competition and can alter entire ecosystems. In other situations, native species might need to be controlled if their population is at a level whereby nature conservation objectives are being compromised. This can occur as a result of inappropriate management practices having been applied historically, or other factors such as adjacent land management or climate change. Herbicides are also used in nature conservation to keep paths, tracks and roads clear.
 
However, the use and effect of herbicides in conservation management is not widely documented or widely understood. This lack of knowledge may have implications for the management, particularly of sensitive features that may be susceptible to negative impacts from herbicides.This literature review aims to better understand usage of the key herbicides, and the impacts and the gaps in evidence for the commonly used herbicides in conservation management situations.

1.1 Regulatory process for pesticide approval in the UK

Currently GB decisions will be made to the same regulatory standards as in the EU, with all active substances approved as a result of EU legislation being adopted into GB law. Regulation (EC) No 1107/2009 (the Regulation) provides statutory powers to control Plant Protection Products (PPPs). PPPs are pesticides used to control plants and plant pests and includes all herbicides, fungicides, insecticides, soil sterilants and, where used to protect plants, rodenticides. 

The Regulation is underpinned by the Plant Protection Products Regulations 2011 (as amended) and together these regulations mean that only authorised products can be sold, supplied, stored, advertised or used. PPPs can only be used in situations for which their use is currently authorised by the Health & Safety Executive (HSE). The GB approvals register provides a full list of approved active substances to be included in plant protection products in Great Britain, together with details on the specific conditions of approval. 

EU pesticides legislation provides further explanation of certain conditions of approval or specific provisions in separate review reports.  For approvals transferred from EU decisions taken before EU exit (1 January 2021) relevant EU review reports can be consulted for further explanation of some specific points relating to relevant approvals. Review reports for specific active substances can be obtained from the EU Pesticides database.

Decisions on approvals taken in GB after leaving the EU are also included in the GB register, and the supporting documents including statutory publications on active substances will be available for active substance renewals and new active substances.

Approval of an active substance does not include approval for use. It is illegal to use a plant protection product other than in accordance with its product authorisation. Details of authorisations for UK plant protection product can be found in the pesticide product database.

Active substance approvals due to expire before December 2023 have been extended for 3 years to allow time to plan and implement the GB review programme. This programme to review the safety of active substances is currently being developed and HSE will retain the power to review active substance approvals at any time, if new evidence identifies any concerns to human health or the environment.

1.2 Evidence required for the approvals process (dossier)

Active substance data requirements are presented in Regulation (EC) No 283/2013 and product data requirements in Regulation (EC) No 284/2013. The data requirements cover the following sectors, up to 100 specific tests are done to ensure the safety of an active substance. The complete dossier is a very extensive set of documentation comprising the required tests and studies, and a series of supporting documents providing background information on the active substance and its uses.

• Physical and chemical properties of the active substance and the components and ‘recipe’ of the pesticide product.
• Analytical methods - methods of testing for the substance.
• Toxicology – includes effects on mammals.
• Residues - to support all pesticide uses which could result in pesticide residues in food or animal feed.
• Consumer exposure - through dietary intake.
• Non-dietary human exposure - exposure via non-dietary routes, including inhalation, dermal absorption and ingestion.
• Environmental fate and behaviour - how substances behave in soil, surface water, sediment, groundwater and air including:
  • What the active substance breaks down into.
  • How quickly the active substance and metabolites break down.
  • Where the active substance and metabolites move to in the environment.
  • Whether the substances accumulate in the environment.
  • What levels of active substances or metabolites are likely to occur in the environment.
• Ecotoxicology - the risk to non-target organisms, to ensure that there are no significant long-term changes to the population nor to the function of the ecosystem. Including other unacceptable effects such as vertebrate mortalities. Specific groups of organisms covered are:
  • Birds and mammals.
  • Aquatic organisms.
  • Bees and non-target arthropods.
  • Soil organisms.
  • Non-target plants.
• Efficacy - the effectiveness of a substance, including identifying the appropriate dose and mode of action, and identifying any adverse or unintended negative impacts.

This data can be found within three databases developed and maintained by the University of Hertfordshire’s Agriculture & Environment Research Unit (AERU). These databases relate to chemical (natural and synthetic) substances used in agriculture: pesticides (Pesticide Properties Database (PPDB)), bio-pesticides (Bio-Pesticides Database  (BPDB)) and veterinary substances. The databases are comprehensive and relational of pesticide physicochemical, toxicological, eco-toxicological, human health and other related data as provided when a pesticide registration is submitted. The primary data sources used to populate the databases are mainly public domain sources (regulatory bodies), peer reviewed literature and private databases.

1.3 Approvals process

Currently, as all pesticide approvals were done prior to EU exit, the following information is the EU approach. After EU exit GB pesticide approvals and renewals will follow a similar approach. 

Pesticide active substances in PPPs are approved if they meet the requirements and conditions specified in retained Regulation (EC) No 1107/2009. The evaluation of both existing and new active substances follows a phased approach:

1. For each substance an initial draft assessment report (DAR) or renewal assessment report (RAR) is produced by a designated Rapporteur Member State (RMS). Active substances are assessed for their impact on workers applying the product, consumers of treated produce, residents living adjacent to application sites and others who may be passing-by at the time of treatment; whether and how the pesticide might move throughout the environment once it has been applied; the impact on the environment, in particular water bodies and wildlife; and product efficacy.
2. The RMS’s risk assessment is peer reviewed by EFSA in cooperation with all Member States.
3. EFSA drafts a report (“Conclusion”) on the active substance. The EFSA Conclusion informs the European Commission in the approval process, the subsequent assessments of plant protection products by the Member States, and the revision of maximum residue levels in food by EFSA.
4. The European Commission decides whether or not to include the substance in the EU’s list of approved active substances. This determines whether the substance can be used in a plant protection product in the EU.
5. EU Member States assess or re-assess the safety of pesticides containing the active substance that are sold in their territory

Active substances that have been shown to have an acceptable risk to people or the environment are added to the list of approved active substances. The review report may indicate areas for special consideration which will have to be highlighted on the product label.

Active substances are initially approved for a fixed period (usually 10 years). They can be reviewed at any time in light of new scientific evidence becoming available. Older products are routinely reviewed by the manufacturer and the authorities to ensure they meet the most up to date safety standards. This is organised in phased programmes depending on the expiry date of the approval. An application to review an active substance must be submitted at the latest 3 years before the expiry of the current approval. The details of the renewal procedure are set out in Commission Implementing Regulation (EU) No 2020/1740 and the process follows a similar procedure to that stated above.

Although pesticide registration is carried out by the EU, individual member states have control over how pesticides are used in their country.

1.4 Why is there an apparent contradiction between the approval process and some of the findings in the peer-reviewed literature?

Pesticides are assessed and approved with a standard approach following the procedure above. These PPPs are launched and start to be used and their effects are seen more widely. Their use is continually monitored through reviews and renewals and pesticides can be withdrawn due to reasons of safety or because companies took commercial decisions not to support substances through the review or renewal process. Sometimes the rules by which pesticides are assessed may change when new directives come into law e.g. Regulations (EU) No 528/2012 and (EC) No 1107/2009 endocrine disrupters. Isoproturon, for example, was withdrawn under the Drinking Water Directive due to it being found at high levels in water. Neonicotinoid insecticides were withdrawn in response to an increasingly strong body of research suggesting they were lethal for pollinators such as bees. Sometimes label changes are made e.g. products containing clopyralid may now only be applied after 1 March. 

The approvals process only asks for a relatively narrow range of evidence, the longer the chemical has been on the market, the greater the amount of information is available on its effects on non-target species. The approvals process also does not consider “in combination effects” of more than one herbicide. This research is often funded by interested parties as a result of observations in the field or potential concerns over safety.

1.5 Low-risk, basic substances and biopesticides

Biopesticides are low risk crop protection products based on micro-organisms, plant extracts and other natural compounds and have a range of attractive properties for integrated pest management (IPM). Biopesticides are regulated as plant protection products under EU plant protection Regulation 1107/2009. Biopesticides do not exist as a regulatory category alone, the pesticide categories “basic substances” and “low risk substances” were introduced in August 2017, as defined in Regulation 2017/1432, amending Regulation 1107/2009. Biopesticides generally qualify as low-risk active substances and are listed in the GB approvals register.

1.6 Definitions

Within the report there are references to direct and indirect effects of pesticides. For clarity the following definitions have been used (Table 1).

 

Table 1. Definition of direct and indirect links.

-	Evidence requirement to establish direct and indirect links between herbicide use and biodiversity
Direct links between herbicide use and biodiversity	To establish a direct link, evidence is required that a population of non-target species undergoes a significant change in abundance or behaviour as a result of consuming or otherwise coming in contact with a herbicide. Effects can be lethal or sub-lethal with some effects being cumulative.
Indirect links between herbicide use and biodiversity	To establish an indirect link, evidence is required that a population of non-target species undergoes a significant change in abundance or behaviour as a result of a change in resource availability due to herbicide use.  Exposure to endocrine disruptors can have effects over time (e.g. causing egg thinning and poor reproductive success in birds).

 

1.7 Identification of pesticides

NatureScot provided evidence of pesticides consented for use on Sites of Special Scientific Interest (SSSI) from their database and additionally from a survey conducted in February 2021 (Table 2). In recent years the most commonly used herbicide has been glyphosate and a limited amount of triclopyr. In previous years a wider range of herbicides was used but many of these have been withdrawn for use in nature conservation management situations and have not been included in this review (Table 3). Details of the active substance, molecular mass, example product, concentration and field rate of the pesticides used are shown in Table 4.

 

Table 2. Pesticides consented for use on SSSIs as shown in NatureScot’s database.

Active substance	HRAC group	Product	Specified targets
asulam*	18	Asulox	Bracken
citronella	-	Barrier H	Not specified
glyphosate	9	Roundup probiactive, adjuvant mixture B	INNS: Himalayan balsam, Japanese knotweed, rhododendron, giant hogweed, American skunk cabbage, piri bur, cotoneaster Other: Purple moor-grass, thistles, bracken, spear thistle, creeping or field thistle, nettles, curled dock, broad-leaved dock, ragwort, beech, birch, Sitka spruce, western hemlock., gorse, rushes Maintain roads tracks and paths
fluroxypyr + triclopyr	4+4	Doxstar	Docks
triclopyr	4	Timbrel	INNS, e.g. rhododendron
triclopyr + clopyralid	4+4	Grazon Pro	Woody species (e.g. birch/rowan/willow) & rose-bay willowherb, gorse, INNS
MCPA	4	Agroxone	Thistles, rushes, ragwort

* No longer authorised for use in Scotland (as of October 2023)

 

Table 3. Pesticides previously consented but no longer approved for use in natural areas.

Active substance	Product	Target
Imazapyr	Arsenal	Japanese knotweed
Bacillus thuringiensis subsp. israelensis strain AM65-52	Gnatrol SC Vectobac12	Mosquitoes
paraquat	Gramoxone	Spot spraying to base of flood bank General weed control
diquat	Reglone	Canadian pond weed
chlorpyrifos	Cyren	Trees
trifolex-tra	MCPA + MCPB	Creeping thistle

 

Table 4. The active substance, molecular mass, example product, concentration and field rate.

Active substance	Molecular mass (g)	Example product	Concentration (g a.s./L)	Field Rate (L/ha)	Field rate (g a.s./ha)
asulam	230.24	Asulox	400	11	4400
glyphosate	169.1	Roundup bioactive GL	360	Varies max. 6L/ha	1800
glyphosate	169.1	Ecoplug max	720	2 plugs /3 cm stem diameter	Single 283mg plug
triclopyr	256.47	Garlon 4	61.6% w/w	4.48 kg/ha	2759
fluroxypyr	255.03	Starane Hi-load HL	333	0.45	149.85
clopyralid	192.0	Dow Shield 400	400	0.5	200
MCPA	200.62	Agritox 50	500	3.3	1650
citronella oil	154.25	Barrier H	22.9% w/w	25ml/plant	5.73

 

2. Literature Review

2.1 Methods

An initial literature search was undertaken in Web of Science, Science Direct, Google scholar, Google, Open Grey, and Defra science search using the following search parameters;

“pesticide” AND ecotox*) AND (Plant* OR invertebrate* OR insect* OR amphibian OR fish* OR aquatic* OR “water quality”) AND (ecosystem OR mire OR peatland OR wetland OR coast OR loch OR bog)

“pesticide” AND ecotox* AND (Plant* OR bird* OR invertebrate* OR insect* OR "soil biology" OR “water quality” OR mammal OR vertebrate OR bryophyte OR moss OR lichen OR fungi OR fern OR “soil microbiome” OR “soil food web impact” OR “soil functioning impact” OR “herbivores food web impact*” OR pollinators) AND (farmland OR agriculture OR nature OR conservation OR ecosystem OR forest OR peatland OR grassland OR mountain OR heath OR woodland)

The search parameters were refined by limiting the information to the UK, Scotland, Ireland, Wales, Belgium, Denmark, France, Germany, Hungary, Ireland, Lithuania, Netherlands, New Zealand, North America, Sweden and Switzerland as all have similar climates to Scotland.

This literature review has concentrated on the three main herbicides for which there were multiple references relevant to the objectives of this study (Table 5). There was less information available for the remaining four herbicides, fluroxypyr, citronella oil, clopyralid and MCPA. The level of information available is indicated in the table. 
 

Table 5. The amount of information available on the main herbicides identified

-	Asulam	Glyphosate	Triclopyr	Fluroxypyr	Clopyralid	Citronella	MCPA
Drift	H	H	L	-	-	-	-
Herbaceous plants	H	H	L	L	M	-	M
Lichen	-	L	L	-	-	-	L
Ferns and mosses	M	-	-	-	-	-	-
Invertebrates and soil biota	L	M	M	-	-	-	L
Insects	-	L	-	-	-	-	-
Bees	L	M	-	-	-	-	-
Amphibians, fish, aquatic invertebrates	L	M	L	L	-	-	L
Aquatic plants	L	M	L	-	-	-	L
Marine	-	M	-	-	-	-	-
Indirect effects	L	-	-	-	-	-	-

L = Limited, M = moderate, H = High, - no information

This search returned 254 peer reviewed articles, from which 107 were identified as being relevant to this review. Any additional papers cited in this sub-set were also included.

2.2 Ecotoxicological data 

Information was collated on LD₅₀, LC₅₀ and EC₅₀ (see Table 6 below) for all herbicides found in NatureScot’ SSSIs database. The information in Table 6 below is based on the active substance contained in the formulated product which will also contain substances such as adjuvants, water conditioners, emulsifiers and other substances. The information is taken from the University of Hertfordshire Pesticides Properties Databases (Lewis et al., 2016). The data in the table and the results of the literature search allow a qualitative, rather than quantitative, risk assessment, although all the compounds (and authorised products) have been assessed by CRD and deemed to be safe for the approved use. Non-target organisms will be exposed when a pesticide is applied but less so if the product is applied precisely e.g. spot spraying or weed wiper. Non-target organisms can be exposed through spray drift, the movement of pesticides in water either by leaching or run-off, residues or via vapour drift.

The product label will give detailed guidelines on the use of the herbicide and should be read carefully before use. Product labels can contain different guidelines even on products containing the same active substance; for example some products containing glyphosate are approved for use near water whilst others are not.

 

Table 6. Effects of selected herbicides on non-target flora and fauna (Lewis et al., 2016)

-	Mammals	Birds	Fish	Aquatic invertebrates	Aquatic plants	Algae	Honey bee	Earthworms
Low	>2000	>2000	>100	> 100	>10	>10	>100	>1000
Moderate	100-2000	100-2000	0.1 - 100	0.1 - 100	0.01 - 10	0.01 - 10	1 - 100	10 - 1000
High	<100	<100	<0.1	<0.1	<0.01	<0.01	<1	<10 -

LD₅₀ -The amount of a chemical that is lethal to one-half (50%) of the experimental organisms exposed to it. LD50s are usually expressed as the weight of the chemical per unit of body weight (mg/kg).

LC₅₀ -Lethal concentration 50 (LC50) is the amount of a substance suspended in the air required to kill 50% of test organisms during a predetermined observation period.

EC₅₀ -The median effective concentration is the concentration of a substance in an environmental medium expected to produce a certain effect in 50% of test organisms in a given population under a defined set of conditions.

2.3 Asulam

Asulam has been available by emergency authorisation for the control of bracken in rough grazing, moorland, amenity grassland and forestry. However, as of October 2023, its use is no longer authorised in Scotland, and from 2024 in GB.

Asulam is a carbamate herbicide, which inhibits 7,8-dihydropteroate (DHP) synthesis (vitamin synthesis) and induces chlorosis. 

Asulam is highly soluble in water, relatively volatile with a low risk of leaching to groundwater (Lewis et al., 2016). It is not persistent in soils but may be persistent in water especially under dark conditions. Although it is not highly toxic there is a risk of bioaccumulation. It is a skin, eye and respiratory system irritant and may cause undesirable effects on reproduction. It is moderately toxic to birds, and of moderate to low toxicity for aquatic organisms, honeybees and earthworms.

2.3.1 Current status

A review report for asulam, completed in 2012 , concluded that the data requirements for the active substance was insufficient to satisfy the requirements set out in Regulation (EC) No 544/2011 with regards to: 

• the toxicological properties of the metabolite sulphanilamide
• the residue definition for risk assessment
• the processing studies
• the toxicological assessment of impurities.

Concerns were identified regarding the risk to consumers, birds and terrestrial non-target plants. The conclusions in the pesticide peer review (2018) identified the long term risk to birds and wild mammals as an area of critical concern.

The emergency authorisation of asulam for 2021 was considered by the Expert Committee on Pesticides on 9th March 2021 and the emergency authorisation issued on 11th June 2021. The committee noted that the acute risk to birds and mammals, acute and long-term risks to bees, non-target arthropods, soil organisms and processes are all acceptable from the proposed use but there was an unacceptable and high reproductive and long-term risk to birds and mammals. The emergency authorisation states that only ground-based use is permitted on areas with a statutory conservation designation (e.g. SSSI) or agri-environment scheme agreement land. Protection of surface water bodies and non-target plants is included with 5m and 90m buffer zones for horizontal boom sprayers and aerial applications respectively, and the use of low drift nozzles.

Scotland, Wales and Northern Ireland refused to grant approval to use asulam for bracken control in 2023. On the 13th October 2023, UPL, the manufacturers of asulam, announced that a decision had been taken to cease further work on a permanent solution for the use of asulam. Without the support and evidence that UPL would have provided, the application for an emergency authorisation to allow the use of asulam in 2024 could not succeed. This means that there will be no asulam (Asulox) available for bracken control in GB in 2024 and beyond.

2.3.2 Drift

Pesticide drift is an ever-present danger to non-target organisms; this has been recognised, and as a consequence herbicide product labels will often contain restrictions on minimising drift. In the 2020 Extension of Authorisation for Minor Use (EAMU) for asulam, the buffer zone for surface water bodies was increased to 90m and livestock should be excluded from grazing the sprayed areas for one month. 

Work has been carried out showing the effects of asulam drift. An assessment of 23 non-crop plant species and the herbicides asulam, glyphosate and MCPA showed that lethal effects were present up to 6m away from the treated field (Marrs et al., 1989). Effects on flowering, such as the absence of flowering, and reductions in seed production were found up to 10m from the field. The greatest distance at which damage effects were found, including reduction in size, leaf chlorosis, discoloration, was 20m (0.15 % of the field rate). Marrs et al. (1989) also found that some species (herbaceous plants) appeared to be consistently more sensitive than others, e.g. cuckoo flower (Cardamine pratensis), black knapweed (Centaurea nigra), foxglove (Digitalis purpurea), ragged robin (Lychnis flos-cuculi), black medick (Medicago lupulina) and selfheal (Prunella vulgaris).

Breeze et al. (1992), studying drift, calculated the ED₁₀ and ED₅₀ for a selection of wild plant species and identified a range of susceptibilities (Table 7). Herbicide was applied as single droplets of 5µL in a dose range of 0.001 to 1000µg a.s./plant. The highest rate was equivalent to the dose a plant would receive during field spraying. Of the four herbicides tested (asulam, glyphosate, MCPA and mecoprop) asulam was the least toxic to non-target plants with none of the species tested having ED₁₀ of less than 1.0 µg a.s./plant. The model indicated that asulam posed little risk to any of the species from spray drift.

 

Table 7. ED₁₀ and ED₅₀ values with standard errors (μg a.s./plant) for toxicity of asulam to some wild plant species from Breeze et al. (1992)) (1000µg a.s./plant = field dose rate).

Common name	Species	ED10	SE	ED50	SE
Cuckoo flower	Cardamine pratensis	17	9.4	220	52
Common knapweed	Centaurea nigra	71	1.5	88	1.9
Crested dog’s tail	Cynosurus cristatus	3.5	2.2	13	3.4
Hedge bedstraw	Galium mollugo	710	78	980	18
Perforate St John's-wort	Hypericum perforatum	110	0.8	>1000	-
Rough hawkbit	Leontodon hispidus	42	16	170	36
Perennial ryegrass	Lolium perenne	12	5.2	86	20
Common bird's-foot-trefoil	Lotus corniculatus	200	340	960	270
Ragged robin	Lychnis flos-cuculi	3.4	1.3	40	8.2
Meadow buttercup	Ranunculus acris	22	18	250	82
Red clover	Trifolium pratense	6.9	2.7	65	*

* not measured

ED₅₀ - (median effective dose) is the dose of a herbicide that produces a specific effect in 50% of the population that takes that dose

ED₁₀ - is the dose of a herbicide that produces a specific effect in 10% of the population that takes that dose

 

The risk to non-target plants was recognised in the 2021 emergency authorisation for the herbicide product Asulox  (active ingredient asulam), with a 5m buffer zone for horizontal boom sprayers and a 90m for aerial applications, both using low drift nozzles.

2.3.3 Ferns and mosses

Rowntree and Sheffield (2005) investigated effects on eight fern species: Asplenium scolopendrium L. (Hart’s tongue fern), Asplenium trichomanes L. (Maidenhair spleenwort), Athyrium filix-femina (L.) Roth (Lady fern), Blechnum spicant (L.) Roth (Hard fern), Dryopteris dilatata (Hoffm.) A. Gray (Broad buckler fern), Phegopteris connectilis (Michx) Watt (Beech fern),Polystichum aculeatum (L.) Roth (Hard shield fern), Woodsia ilvensis (L.) R. Br. (Oblong woodsia). These were tested at the mature sporophyte stage with three doses of asulam including full application rate (4400 mg a.s./ha) and two doses corresponding to 10 m (440 g a.s./ha) and 50 m (20 g a.s./ha) downwind drift from aerial spray. The damage was assessed over two seasons. The maximum damage occurred one year after treatment with limited signs of recovery only seen by the second season. All species were severely damaged by the full application rate but three species Asplenium scolopendrium, Athyrium filix-femina and Blechnum spicant, were affected by the 10m fall-out dose in the longer term.

Rowntree et al. (2003) tested the sensitivity of the mature gametophyte of 18 moss species to asulam. They showed that concentrations as low as 0.05 g a.s./L inhibited the growth of the most sensitive species tested, and the growth of 14 of the species was significantly inhibited at concentrations of 1 g a.s./L (Table 8). For comparison the full application rate of asulam is 4400g a.s./L.

 

Table 8. Species sensitivity to 24h Asulox exposure after 3 weeks growth (Rowntree et al., 2003)

Sensitive	Intermediate	Tolerant
Polytrichum commune	Calliergonella cuspidata	Campylopus introflexus
Sphagnum contortum	Cratoneuron filicinum	Warnstorphia fluitrans
Sphagnum denticulatum	Dicranella palustris	-
-	Eurhynchium praelongum	-
-	Fissidens adianthoides	-
-	Hypnum jutlandicum	-
-	Philonotis seriata	-
-	Pleurozium schreberi	-
-	Rhytidiadelphus squarrosus	-
-	Rhytidiadelphus triquetus	-
-	Sphagnum magellanicum	-
-	Sphagnum warnstorfi	-
-	Thuidium tamariscinum	-

 

2.3.4 Invertebrates and soil biota

Information on the effects of asulam on invertebrates and soil biota is limited, the herbicide had no or minor effects at 16mg/kg on carbon turnover, but 160mg/kg significantly inhibited respiration of cellulose-amended decomposition processes in soil (Marsh, 1980).

Unpublished results from bracken control trials with asulam, applied at 11L/ha (4400g a.s./ha) at various sites across the UK (unpublished reports by Prof Roy Brown, made available through NatureScot) suggest a rapid decline in soil residues after moderate levels (3-5.5 ppm) in the first 7 days and by month three all values were below 2 ppm for reference. There was no evidence of ongoing chemical residue activity on any sites after year 2 and in most cases after year 1. The same trials show similar profiles and densities for selected soil mesofauna (Acari, carabid beetles, springtails, Enchytraedae) between asulam and control plots. These results have not been peer-reviewed.

2.3.5 Bees

The use of asulam has been shown to lead to the contamination of honey with its degradation product sulfonamide, (Kaufman et al., 2004). This contamination is likely if the herbicide is applied when bees are active and foraging on flowers. No more recent references in this subject area were found, asulam is considered to be of low toxicity to bees (contact and oral acute LD50 >100µg/bee).

2.3.6 Birds

Data were provided to EFSA for the evaluation of asulam as a herbicide for spinach and this included a field study monitoring the foraging behaviour of skylarks in sugar beet fields in Austria, this was not considered appropriate to spinach by ESFA. The long-term risk for insectivorous and herbivorous birds was considered to be a critical area of concern and requiring more supporting data. In the emergency authorisation of asulam (HSE, 2020) use before 1 July is not allowed to protect ground nesting birds. A potentially significant toxicity risk to ground nesting birds towards the end of the breeding season was also identified as one of the conditions.

2.3.7 Amphibians, fish, aquatic invertebrates and other freshwater species

ESFA concluded that asulam is very toxic to aquatic organisms (ESFA, 2018). The risk to aquatic life is mitigated in the pesticide authorisations by a 5m unsprayed horizontal buffer zone distance to surface water bodies from vehicle mounted applications, and a 90m unsprayed zone for aerial applications. In reality the uplands are covered by networks of tiny streams and the risk of asulam reaching water bodies is high.

2.3.8 Aquatic plants

ESFA concluded that asulam is very toxic to aquatic plants (ESFA, 2018). The risk of asulam to aquatic organisms was assessed as low when the risk mitigation requirements are followed (Section 3.1.1).

Macrophytes were exposed to asulam concentrations between 0 and 1260 mg/L. The hazard concentration for Canadian pondweed (Elodea Canadensis) was 42 µg/L, Elodea nuttallii was 5.52 µg/L and curled pondweed (Potamogeton crispus) was 30 µg/L (Arts et al., 2008). The high sensitivity of E. nuttallii to asulam was confirmed in an indoor microcosm study, where biomass of this species was reduced to 71% and 60% of the control at concentrations of 6.0 and 14.7 mg/L asulam, respectively (Van Wijngaarden et al., 2004).

2.3.9 Indirect links between asulam use and invertebrate species

Speight and Whittaker (1987) monitored green dock beetles (Gastophysa viridula) in plots that contained the beetles’ preferred source of food, Rumex obtusifolius (broad leaved dock), over three generations. Plots were sprayed with asulam or left unsprayed. There was 67% reduction in beetles in the sprayed plots, and fewer generations. A third generation was only observed in the unsprayed plots. The authors concluded that the loss of foliage from the R. obtusifolius plants in the treated plots was likely responsible for the reduced reproduction in the first and second generations and lack of a third generation.

2.4 Glyphosate

Glyphosate is used for the control of annual and perennial broadleaved weeds and grasses, and a wide range of other unwanted plants (e.g. bracken, rushes, weed beet, watercress and water lilies) in amenity grass and vegetation, sward destruction in grassland, hard surfaces, forest, forest nursery, farm forestry, land temporarily removed from production, non-crop farm areas, and aquatic situations.

Glyphosate is a broad-spectrum herbicide used for its high effectiveness and low cost. Glyphosate deters plant growth by inhibiting the shikimate pathway that prevents the synthesis of aromatic amino acids. The shikimate pathway is only present in plants, fungi, and some microorganisms. Animals lack the shikimate pathway, which is why glyphosate has historically been considered one of the least toxic herbicides.

Glyphosate as the active substance is often less toxic than the surfactants and/or other adjuvants used with it in commercial products. The herbicide is toxic to most plant species. Consequently, application of glyphosate will pose a risk to all plants outside of the target area, where spray drift is possible. It is highly soluble in water, relatively volatile and does not normally leach to groundwater because glyphosate is readily ionized and the anion is strongly adsorbed to sediments and soils of pH > 3.5. Glyphosate degrades to natural products such as CO2 and phosphate ions with degradation in terrestrial and aquatic systems occurring predominantly via microbial processes (Lewis et al., 2016). It is not persistent in soils but may persist in aquatic systems under certain conditions. It is moderately toxic to birds, most aquatic organisms, earthworms and honeybees.

The glyphosate dossier is now available comprising of 1,507 studies, and 12,000 articles published within the last 10 years. This dossier has been submitted as part of the EU re-approval process which could be searched for additional relevant information.

2.4.1 Current status

A review report for glyphosate, completed in 2017, concluded, in the context of Regulation (EC) No 1107/2009, that products containing glyphosate will still fulfil the safety requirements. 

• Residue levels arising from proposed uses have no effects on human or animal health.
• Glyphosate does not have endocrine disrupting properties through oestrogen, androgen, thyroid or steroidogenesis mode of action. 
• Polyethoxylated tallowamine (CAS No 61791-26-2) should be excluded from the use in PPPs containing glyphosate.
• Some genotoxicity studies on formulations presented positive results, and therefore, this should be addressed during renewal or first authorisation of plant protection products.
• Under the proposed and supported conditions of use there are no unacceptable effects on the environment.

The following conditions should be taken into account when authorising plant protection products containing glyphosate. 

• The protection of the groundwater in vulnerable areas, in particular with respect to non-crop uses.
• The protection of operators and amateur users.
• The risk to terrestrial vertebrates and non-target terrestrial plants.
• The risk to diversity and abundance of non-target terrestrial arthropods and vertebrates via trophic interactions.

Risk mitigation measures should be considered

• Use of glyphosate is minimised in specific areas listed in Article 12(a) of Directive 2009/128/EC. This includes areas used by the general public or vulnerable groups such as public parks and gardens, sports and recreation grounds, school grounds and children’s playgrounds and in the close vicinity of healthcare facilities.

The current EU approval for glyphosate ends on 15 December 2023. In 2019 the glyphosate renewal group applied to renew the approval of glyphosate, the final conclusion of EFSA is expected in mid-2022. The renewal for glyphosate in GB is 15 December 2025.

2.4.2 Drift

Glyphosate drifting from sprayed fields is the main source of exposure of non-target plants. As the field rate is applied through the sprayer, environmental factors such as wind move some small droplets away from the point of application.
 
The drift level (dose) of glyphosate needed to cause phytotoxicity varies between species. Breeze et al. (1992) calculated the ED₁₀ and ED₅₀ for a range of species, there was a range of susceptibilities to a selection of wild plants (Table 9). Herbicide was applied as single droplets of 5µL in a dose range of 0.001 to 1000µg a.s./plant. The highest rate was equivalent to the dose a plant would receive during field spraying. Of the four herbicides tested (asulam, glyphosate, MCPA and mecoprop) glyphosate was the most toxic to non-target plants, seven of the species tested had ED10 of less than 1.0 µg/plant. Breeze et al., calculated that the concentration in general use would be 6.4 g/L and six of the species would be at risk. Two species (wood avens (Geum urbanum) and betony (Stachys officinalis)) were found to be very sensitive to glyphosate with an ED10 of less than 0.1 µg/plant and the authors postulated that damage could be possible at distances greater than 50m from the site of spraying. These distances were greater than those reported by Marrs et al. (1989) for perennial plants and Marrs & Frost (1997) for natural communities. Seedling age and type of sprayer must be considered. When seedlings are considered the distance needs to be at least 20 m, although other sensitive species (crested dog’s-tail (Cynosurus cristatus), perforate St John’s-wort (Hypericum perforatum), primrose (Primula vulgaris) and great mullein (Verbascum thrapsus)) can show symptoms between 20–40m (Breeze et al., 1992; Marrs et al., 1993).

 

Table 9. ED₁₀ and ED₅₀ values with standard errors (μg/plant) for toxicity of glyphosate to some wild plant species from Breeze et al. (1992) (1000µg a.s./plant = field dose rate)

Common name	Species	ED10	SE	ED50	SE
Cuckoo flower	Cardamine pratensis	3.9	1.2	23	3.5
Common knapweed	Centaurea nigra	2.0	0.5	6	0.7
Crested dog’s tail	Cynosurus cristatus	0.3	0.2	0.7	0.2
Hedge bedstraw	Galium mollugo	4.6	0.7	36	2.0
Wood avens	Geum urbanum	0.1	0.1	9.4	6.5
Perforate St John's-wort	Hypericum perforatum	5.6	2.5	14	2.6
Rough hawkbit	Leontodon hispidus	0.5	0.2	5.9	1.1
Perennial ryegrass	Lolium perenne	0.3	0.1	1.3	0.2
Common bird's-foot-trefoil	Lotus corniculatus	8.6	7.9	93	37
Ragged robin	Lychnis flos-cuculi	0.8	0.2	2.4	0.3
Meadow buttercup	Ranunculus acris	0.9	0.4	3.8	0.7
Betony	Stachys officinalis	0.1	0.1	1.7	0.5
Upright hedge-parsley	Torilis japonica	5.8	2.2	16	2.9
Red clover	Trifolium pratense	4.1	1.6	21	3.7

* not measured

ED₅₀ - (median effective dose) is the dose of a herbicide that produces a specific effect in 50% of the population that takes that dose

ED₁₀ - is the dose of a herbicide that produces a specific effect in 10% of the population that takes that dose

 

Stranberg et al. (2019) showed that even with drift reducing nozzles with a wind speed of 4.5m/s, applying glyphosate at label rate of 1440 g a.s./ha, up to 2.8% (40.3 g a.s./ha) of the glyphosate dose reached an experimental area up to 25m away from the sprayer with most sampling plots receiving 0.1% of the dose. This had significant effects on plant flowering by reducing the number of flowers of red clover (Trifolium pratense) and common bird’s-foot-trefoil (Lotus corniculatus). Marrs and Frost (1997) also found various responses to sprayer drift from glyphosate and MCPA. All species were negatively affected in at least one of the three years (Marrs & Frost 1997) and common bird’s-foot-trefoil (L.corniculatus) was among the sensitive species. Studies done simultaneously in Denmark and Canada (Strandberg et al., 2019) showed that exposure to glyphosate resulted in significant reductions in biomass at 1% and 5% of the field rate (1440 g a.s./L) when applied at the 6-8 leaf stage for creeping thistle (Cirsium arvense), dandelion (Taraxacum officinale) and field scabious (Knautia arvensis); and at the 5% dose for night-flowering catchfly (Silene noctiflora), field pansy (Viola arvensis) and field mouse-ear (Cerastium arvense).

Because glyphosate translocates readily from foliage to the growing parts of the plant, good coverage of the target weed is not needed for efficacy. For mature plants of many species within 20m of the spray application, there is minimal damage. A review concluded that limiting glyphosate spray drift to <5g a.s./ha would protect 95% of plants against minor effects, and that reducing spray drift to 1 to 2g a.s./ha would almost completely protect plants in non-target areas. Glyphosate is usually applied at an application rate of 1440g a.s./ha a 1% level of spray drift would correspond to 14.4g glyphosate /ha and (Cederlund, 2017). However, there can be growth stimulation effects of glyphosate at doses below the threshold for phytotoxicity (Hormesis). Subtoxic doses of glyphosate can cause the most profound cases of hormesis, especially in woody plants (Brito et al. 2018). 

Glyphosate drift was also shown to reduce seed production of non-target plants, this was significantly reduced at the 1% (of label rate 1440g a.s./ha) dose in cornflower (Centurea cyanus) when applied at the 6-8 leaf stage and of field pansy (Viola arvensis) when applied at the bud stage (Strandberg et al., 2019). The 5% (of label  rate, 1440g a.s./ha) dose significantly reduced seed production of C. cyanus, V. arvensis and T. officinale following application at both of these growth stages and of broad-leaved willowherb (Epilobium montanum) and field scabious (Knautia arvensis) when applied at the 6-8 leaf stage. In Canada, a significant reduction in seed production of red clover (T. pratense) was obtained with the 5% dose of glyphosate applied at the bud stage but not at the 6-8 leaf stage. Marrs et al. (1993) also showed young plants to be more vulnerable than at later stages.

Glyphosate had an adverse effect on the total number of flowers produced by plants (Strandberg et al., 2019), and high levels of plant mortality were seen when plants were exposed at the 6-8 leaf stage to 5% glyphosate. In total, significant negative effects on total flower production or non-significant trends towards reduced flowering were observed in 57% of all cases, with reductions spanning from 4% to 100%. Overall, trends indicate that the 6-8 leaf stage may be more sensitive than the bud stage but this may miss out species that are sensitive at other life stages.

Standberg et al (2019) reported that flowering was delayed for 24 days for plants that survived exposure to 5% full rate of glyphosate exposure at the 6-8 leaf stage. They postulated that simultaneous effects on a reduction in the number of flowers produced, and a delay in flower production, would result in restricted pollen transfer and out-crossing. Reductions in pollen production would also have implications for visits of pollinating insects to the area. Schmitz et al. (2013) found that flower density of meadow buttercup (Ranunculus acris) was greatly reduced with repeated application of sub-lethal doses of glyphosate.

Petal, anther, and pollen deformations have been reported for Brassica spp. in flowers exposed to low doses of glyphosate (Londo et al., 2014)

Seed germinability was reduced by up to 75% at the 5% rate of glyphosate (of label rate) in C.cyanus, V. arvensis, E. montanum, C. arvensis, T. officinale, C. arvense and K. arvensis (Strandberg et al., 2019). Blackburn and Boutin (2003) also demonstrated a negative effect on the germination rate of seeds produced by plants exposed to glyphosate depending on the degree of seed maturity at exposure.

Helander et al (2012) noted that applications of glyphosate have caused shifts in weed populations towards annual broad-leaved and deep rooted perennial species. In some situations herbicide resistance has occurred, predominantly in situations where multiple applications have been made year on year, but resistance has not been found in the UK. However black-grass (Alopecurus myosuroides) and sterile brome (Anisantha sterilis) populations have shown increased tolerance to glyphosate (Davies et al., 2019; Davies & Neve, 2017). Strandberg et al.  (2019) also considered plant competition and found that low rates of glyphosate were likely to disadvantage more sensitive species which would then experience increased competition from less affected species. This was supported by a field study by Damgaard et al. (2014), who found that low doses of glyphosate altered the competitive interactions between two grass species, common bent  (Agrostis capillaris) and sheep’s fescue (Festuca ovina), with increasing levels of glyphosate, (F. ovina became a better competitor than (A. capillaris). This explained why F. ovina was found to be dominant in field plots treated with higher levels of glyphosate (Damgaard et al. 2014).

Glyphosate has been shown to affect the development of plant disease, the inhibition of the enzyme EPSPS reduces production of shikimic acid pathway–derived compounds, such as some phytoalexins and lignins that plants use to protect themselves from microbial plant pathogens (Hammerschmidt, 2018). This effect is so pronounced that the amount of glyphosate needed to kill a weed is significantly less when plant pathogens are present (as reviewed by Duke 2018). Glyphosate use in silviculture has also been connected to increased blue stain fungi infection in aspen (Populus tremuloides) (Man et al., 2011).

2.4.3 Transfer of herbicide from soil to plants

Glyphosate is translocated within plants, accumulated in roots, and eventually released into the rhizosphere (Laiteinen, 2007). From the soil, glyphosate may also be reabsorbed by the target or non-target plants back through the roots after the initial application. In a review Kanissery (2019) reported that there are a few studies that have investigated the effects of root-zone exposure of glyphosate on crops, for example in rapeseed (Petersen, 2007). The studies indicate there is a likelihood of glyphosate being reabsorbed into the roots of surrounding vegetation. However, most of the conclusions were drawn from observations in hydroponic nutrient solutions. A more recent experiment done under field conditions showed crops sown into glyphosate treated soil had delayed germination and lower biomass during early growth but growth was greater in treated plots at the end of the experiment (10 weeks after sowing (Helander et al., 2019). Glyphosate damage on following crops is not reported as a problem in the UK arable sector.

Information on transfer of herbicide between plant root and their symbiotic fungal hyphae is limited to a single article - although studies are referred to, these are not referenced. The conclusion is that root transfer does not occur because healthy plants have intact barriers to chemical translocation (Chalker-Scot), although root grafts that have already been breached by fungi may serve as conduits for herbicide.

2.4.4 Persistence of herbicide within woody vegetation

According to the manufacturer, once plant material is treated with glyphosate, it is translocated to the growing points where it accumulates at the site of action. Whenever glyphosate is exposed to microbial activity after decomposition of the tree, then degradation will occur, but breakdown of glyphosate within plants is regarded as non-existent, or at least very slow. Once the tree has decayed, there might be residues which could be released to the environment, but little is known about this. Even in the event it were to leach out of the wood, glyphosate would not pose a risk to the wider environment. In soil, glyphosate is tightly bound to soil particles and readily degraded in the environment by naturally occurring micro-organisms; the average half-life in soils is 45 days. The half-life in water is even less at around 4-7 days. Glyphosate is broken down by micro-organisms into AMPA (amino methyl phosphoric acid) and in combination with CO₂ is finally degraded into water, nitrate and phosphate.

Research shows that upon application, glyphosate is absorbed through leaves, stems or roots, and is translocated throughout the plant. This translocation follows the source to sink flow of photosynthates (sucrose and other carbohydrates) through the phloem, and after cycling throughout the plant for at least 72h, glyphosate accumulates in the apical meristems of roots and young leaves. Glyphosate shows little or no metabolism in most plants, and it is easily exudated by the roots of the plants, being eventually released into the rhizosphere, probably through a diffusion process, together with sugars, amino acids and other low molecular weight compounds (Kremer et al., 2005). Exudation through the root is also considered one of the possible mechanisms involved in the environmental destination of herbicides. Glyphosate is strongly adsorbed onto soil particles, degraded by soil microorganisms or absorbed by the roots of adjacent plants.

Eucalypt plants commonly present symptoms of herbicide damage in areas where glyphosate is used.  Tuffi Santos et al., (2008) applied 14C-glyphosate to Signal grass (Brachiaria decumbens) intercropped with two eucalypt clones on a clay and a sandy soil. Samples of the eucalypt were tested at 2, 8, 16 and 24 days. Herbicide symptoms were not observed in any eucalypt plant evaluation. However, 14C-glyphosate was found in all plants, regardless of the soil type, clone or evaluation time, with the highest concentration being found in the sandy soil. Results show radicular exudation of glyphosate by B. decumbens and its absorption by eucalypt plants through roots.

Viti et al. (2019) applied ¹⁴C-glyphosate to seedlings of Pallisade grass (Urochloa brizantha) planted around seedlings of sugarcane and citrus, to evaluate the translocation and root exudation by U. brizantha and its transport in sugarcane and citrus. Twelve days after the application of ¹⁴C-glyphosate to the grass, it was measured in the leaves, culms and roots of the grass, sugarcane and citrus. The highest amount of glyphosate was detected in the U. brizantha leaves, where the applications were carried out. Only traces of glyphosate (0.001%) were detected in soil cultivated with sugarcane. On the other hand, in citrus, U. brizantha exuded 9.46% of the glyphosate by the root system in the soil. The total amount of herbicide found in sugarcane and citrus seedlings was only 0.006 and 0.095%, respectively, in all parts of the plant. The results indicate that the translocation of glyphosate in young plants of U. brizantha associated with citrus was higher in relation to sugarcane, and it was not exuded by the root system of the weed with sugarcane, but presented root exudation with citrus.  However, the amount did not reach what is necessary to affect the dry mass of the agricultural crops. Very little research has been conducted on glyphosate storage and persistence within plant tissues, and to our knowledge, no research has yet been conducted on long-term glyphosate persistence in perennial forest plants beyond one year after treatment. Botten et al. (2021) collected plant tissues from five forest understory perennial species growing in two distinct biogeoclimatic regions of northern British Columbia. Root and shoot samples were collected in July 2018 on forestry cut blocks where a glyphosate-based herbicide was aerially applied at a rate of 3.3–4.0 L/ha (resulting in a concentration of 1.78–2.16 kg a.i./ha), one year, three years, six years, and twelve years before sample collection (corresponding to the treatment years 2017, 2015, 2012, and 2006). Plant tissue was analysed for presence of glyphosate using high performance liquid chromatography – mass spectrometry. The authors collected 216 samples from untreated areas and found that 3.2% contained trace amounts of glyphosate (seven samples). In treated areas they collected 377 samples, 44% contained glyphosate. In the first year after glyphosate application 93% of samples contained glyphosate, three years after, 45% of samples, after six years it was 21% and after twelve years, 2%. The authors found that root tissues generally retained glyphosate residues longer than shoot tissue types (Table 10).

Example amounts of glyphosate after a single year after treatment in rosebay willowherb and raspberry roots were 0.47 µg/g and 0.44 µg/g respectively. After 12 years one sample of  rosebay willowherb contained 0.17 µg/g glyphosate. The quantities of glyphosate contained in plant tissues after 3–12 years are extremely low, this study considered the amounts in relation to MRLs and their effect on animals eating the vegetation. The quantities of glyphosate contained in plant tissues after 3–12 years are extremely low, and the authors said that these should not be considered an immediate hazard, but could be considered when assessing exposure of humans and wildlife to chronic, low-concentrations of glyphosate and other chemicals in the environment.

 

Table 10. Timeline of glyphosate persistence in native plant species after treatment with glyphosate based herbicides in forests of northern British Columbia, Canada (Botten et al., 2021).

Common name	Species	Fraction	Year when residue concentrations dropped below detectable >0.008ppm
Blueberry	Vaccinium caespitosum	Fruit	1-3
Raspberry	R.ideaus	Fruit	1-3
Raspberry	R.ideaus	Shoot	1-3
Red osier dogwood	Cornus sericea	Shoot	3-6
-	Salix spp.	Shoot	6-12
Rosebay willow herb	Chamaenerion angustifolium	Shoot	6-12
Raspberry	R.ideaus	Root	6-12
Rosebay willow herb	Chamaenerion angustifolium	Root	12+

 

Glyphosate and AMPA may persist in perennial plant tissues for an extended duration of time of a year or more (Roy et al., 1989; Mamy et al., 2016; Wood, 2019).

Previous research on perennial forest plants has primarily considered only short-term (much less than one year) persistence of glyphosate in plant tissues. Wood (2019) showed that glyphosate ranging in concentration from 0.077 to 1.050 µg/g could be detected in the tissues of non-targeted perennial forest plants at one year after operational treatment with GBH. Prior to this, Newton et al., (1994) reported 0.162 µg/g glyphosate residue remaining in herbaceous vegetation 346 days after treating the canopy with a high dose of glyphosate. The herbicide is quickly transferred to the growing points and is unlikely to remain in the timber.

2.4.5 Lichen

In a study by Vannini et al., (2015) a lichen (Xanthoria parietina) was shown to take up glyphosate very efficiently and this caused toxic effects reducing photosynthetic efficiency and chlorophyll. The concentration of glyphosate bio-accumulated in X. parietina remained stable throughout the 90 days of the experiment (Vannini et al., 2016).

Glyphosate reduced the abundance of 56% of the lichen species studied in Canada (McMullin et al., 2011). Tolerance to glyphosate varied between lichen species. The most herbicide sensitive species were generally highly branched. The species with the highest mortality were Bryoria furcellata, Cladonia uncialis, and Trapeliopsis granulosa. 

2.4.6 Invertebrates and soil biota

The effects of glyphosate and glyphosate-based products on soil microorganisms have been extensively investigated (von Mérey et al., 2016; Cerdeira and Duke, 2010; Duke et al., 2012, Duke, 2020). Studies of glyphosate effects on soil microflora, using glyphosate levels that could be found in agricultural soil, have generally found minor effects that do not last long (Nguyen et al., 2016; Rose et al., 2016 ; Nguyen et al., 2018). Kepler et al. (2019) showed that plots treated with glyphosate did not differ from untreated plots in overall microbial community composition. Results with glyphosate formulations indicate no long-term effects on nitrogen transformation in soil, even at rates that greatly exceed maximum use rates (von Mérey et al., 2016). In addition, numerous laboratory and field studies have investigated the effects of glyphosate on soil bacteria and fungi (Duke et al. 2012), although some laboratory tests have shown effects on nitrogen fixing bacteria and soil fungi, effects are typically observed only at glyphosate concentrations well above normal field application rates (Duke et al., 2012). In one case, even when applied at 3X the recommended field rates, glyphosate had only small and transient effects on soil microbial communities (Weaver et al. 2007).

In a review by Helander, Saloniemi & Saikkonen (2012) on the effects of glyphosate in northern hemisphere locations, they noted that glyphosate has been observed to accumulate in plant roots from where it is gradually released into the rhizosphere. Because glyphosate blocks the shikimic acid pathway which is present in plants but also in some fungi and bacteria, it affects the microbial activity and can modify the microbial community in the soil in both field and laboratory experiments. If the soil microbial communities responsible for decomposing organic material are negatively affected by herbicide treatments, then the decomposing rate of biomass and nutrient cycling is potentially altered when glyphosate is applied.

In a comprehensive review of multidimensional relationships of herbicides with insect-crop food webs (Sharma et al., 2018), the authors stated that direct topical applications of glyphosate, when tested in laboratories, have some level of adverse effect on one or all life stages of the majority of insects. However in field conditions, several indirect factors including soil moisture, agriculture system; weed and insect relationship; and crop and insect relationship status play a role to reduce the effects. The authors concluded that herbicides as the most commonly applied pesticides in agro-ecosystems impact insects. They can directly influence the life cycle of insect pests and beneficial insects and indirectly impact pollinators by amending their food source.

A study of the effects of glyphosate on non-target insects in farm forestry (Whitehouse & Brown, 1993), found no significant effects on mortality of chafer larvae or adult ground beetles. Laboratory studies (Haughton et al., 2001a &b), investigating the direct effects of glyphosate on non-target spiders (Lepthyphantes tenuis), found that spider mortality was less than 10% after 48 hours and increased slightly to 13% after 72 hours.  Indirect effects were also studied, in field margins which had been sprayed with varying levels of glyphosate (Haughton et al., 2001a).  The abundance of spiders was significantly lower in the sprayed plots compared to an unsprayed control plot.  The reasons for this decline seemed to be increased amounts of dead vegetation and decreasing height of the remaining vegetation.  When the treated plots were revisited after a year there were no differences between the treatments and this was attributed to dispersal and the species being particularly fecund (six generations in the 16 months since the margin was sprayed). The conclusion was that the glyphosate applications only had a within-season indirect effect on the spider.

Behrend and Rypstra (2019) exposed female wolf spiders (Pardosa milvina) carrying egg sacs on a surface treated with the herbicide for seven hours and observed higher levels of activity and greater capture success of their offspring in the following four weeks. These results suggest that herbicides have the potential to adjust the behaviour of individuals in the predator community.

A range of glyphosate based herbicides (GBH) and their active ingredients (AI) were applied to arable soils. Both GBHs and AIs increased the surface activity of springtails (Sminthurinus niger) and this was attributed to GBH/AI toxicities, population variations of Collembolans and concurrent changes in abiotic parameters such as moisture content and soil organic matter (Maderthaner et al., 2020).

Arthropod predators including wolf spiders, Pardosa milvina and Hogna helluo, a ground beetle, Scarites quadriceps were exposed to the field rate of glyphosate based product containing polyethoxylated tallowamine (Evans et al., 2010). The herbicide affected both the locomotion time and distance of the arthropods and decreased the survival in P. milvina. A spider Pardosa agricola and a beetle Poecilus cupreus (Michalkova & Pekar, 2009) inhabiting agroecosystems have been shown to exhibit subtle shifts in behaviour and reproduction during or after exposure to glyphosate. These shifts in locomotion and associated behaviours in predatory arthropods could drive changes in colonisation rates that affect predator population dynamics, foraging success and reproductive rates. These changes may in turn lead to changes in the community structure of these predators, altering their ability to exert top-down effects on herbivorous pest communities (Evans et al., 2010). Polyethoxylated tallowamine has since been banned from use in glyphosate products.

Garden snails (Helix aspersa) were exposed to label doses of glyphosate - concentrations of the herbicide were low from 56 days after application (Druart et al., 2011). In these conditions, no effect was found on survival, growth, development of the genital apparatus, or gametogenesis of H. aspersa. At a 10-fold higher concentration, accumulation of glyphosate occurred in snail tissues exposed to constant loads of herbicide in food, but no damaging effects were found in snails.

Earthworms (Eisenia fetida) were unaffected by glyphosate applied at field rate (von Merey et al., 2016). Likewise Nuutinen et al. (2019) looking at glyphosate applied to simulate cereal stubble showed no adverse effects on earthworms. These results are in line with EFSA evaluation that glyphosate is of low to moderate toxicity and low risk to earthworms. Conversely mesocosms filled with soil and planted with either the grass cock’s-foot (Dactylis glomerata), white clover (Trifolium repens) and dandelions (Taraxacum officinale) were treated with glyphosate at below field rate.  Surface casting activity of vertically burrowing earthworms (Lumbricus terrestris) almost ceased three weeks after herbicide application, while the activity of soil dwelling earthworms (Aporrectodea caliginosa) was not affected. Reproduction of the soil dwellers was reduced by 56% within three months after herbicide application (Gaupp-Berghausen et al., 2015).

2.4.7 Bees

Regulatory studies have shown that glyphosate is non-toxic to honeybees (oral LD₅₀ 100 g/bee) (Lewis et al., 2016). The glyphosate renewal group reported in their dossier (GRG, 2021) low acute and chronic risk to larval and adult honey bees, bumble bees and wild bees. They also stated that that indirect effects to bee populations from in-crop weed control is unlikely because in-crop flowering weeds are not a significant resource for nectar and pollen and the off-crop non-target terrestrial plants community will be protected by in-field no spray zones. In natural situations glyphosate indirect effects may be more likely because flowering plants are more likely to be a significant source of nectar and pollen. There are indications from the literature that glyphosate may perturb the honeybee gut microbiome, and may potentially leave bees more susceptible to pathogens, but effects on overall colony health are unclear (Motta et al., 2018). Therefore, it may be wise to exercise caution while applying glyphosate to, or close to, flowering plants. An ecological risk assessment of Roundup® by Giesy et al. (2000) concluded that the greatest risks to birds and mammals, non-target arthropods and bees were from altered habitat structure and food availability within treated areas rather than direct toxicity.

Vazquez et al., (2018) fed a honey bee brood (Apis mellifera) food containing traces of glyphosate (1.25–5.0 mg per litre of food) under laboratory conditions. They showed that the brood had a higher proportion of larvae with delayed moulting and reduced weight. Their work indicated that glyphosate is a stressor that affects larval development depending on individual and colony susceptibility. The same authors (Vazquez et al., (2020) fed honey bees sugar solution containing glyphosate and showed that the ingestion of 50ng of glyphosate decreased both antennal activity and sleep bout frequency.

Battisi et al., (2021) conducted a meta-analysis on mortality of bees from 34 data sets. The study supports the hypothesis that the exposure of bees to glyphosate and glyphosate -based formulations, in ecologically relevant doses which can be found in the environment (0.0086μg / bee, 0.00013μg / bee, 275ng / a.e, 550ng / a.e,  0.16mL / bee (week)) or in recommended concentrations used in agricultural settings,( 0.5 ng / bee (day), 140μL / bee, 1.5μL / 200mL) might cause lethal effects (mortality) in these insects, since in most categories significant differences were reported between the experimental and control groups. The glyphosate ingestion (food), spraying, and contact with glyphosate had a significant effect on bees. Adult bees or larvae were shown to be susceptible to the active ingredient. Chronic expositions were apparently more harmful to bees. Regarding the studied bee species, both stingless bees Melipona quadrifasciata and Hypotrigona ruspolii, as well as A. mellifera exposed to glyphosate were negatively affected when compared to the control group. Conversely Thompson et al. (2014) demonstrated no effect on larval development, growth and survival and adult survival at glyphosate concentrations of 75, 150 and 300mg a.e./L.

2.4.8 Amphibians, fish, aquatic invertebrates and other freshwater species

Most studies looking at amphibian sensitivity to pesticides focus on the larval stage, however response to pesticides is species and life-stage specific (Bruhl et al., 2011; Barker et al., 2013). Glyphosate-based herbicides have been found to be moderately toxic, and glyphosate (active substance) slightly toxic to amphibian larvae at field rates (Wagner et al., 2013). Generally, absorption of pesticides is greater in amphibians than other vertebrates.

Sublethal concentrations of Glyphosate were shown to cause a slight anaemic and significant immunosuppressive response in common carp (Cyprinus carpio) juveniles (Kondera et al., 2018).

An acute 96 hour LC50 for larvae of the common frog (Rana temporaria) of 10.4mg a.s./L was reported in the literature, from a study conducted with formulated product (which did not contain the surfactant polyethoxylated tallowamine (Wagner et al., 2017). Applications of the herbicide in early spring (when early larvae are present in breeding ponds) are more dangerous for the aquatic development of R. temporaria.

In a critical review of the potential impact of glyphosate based herbicides on amphibians, Wagner et al (2013) stated that the impact on amphibians depended on the herbicide formulation, with different sensitivity of taxa and life stages. Effects on development of larvae were the most sensitive endpoints. As with other contaminants, co-stressors such as predator cues and competition mainly increased the adverse effects. A later review by Breckels et al. (2018) concluded that glyphosate posed potentially negligible or short-lived effects to non-target aquatic organisms including fish, invertebrates and amphibians.  Conversely, certain surfactants used to increase herbicide efficacy, were suggested to be more toxic than the herbicide alone. The effects of surfactants in glyphosate based products have been recognised, and the use of the surfactant polyethoxylated tallowamine was withdrawn from glyphosate products in the EU in 2017.

Low levels of a glyphosate-based herbicide induced significant negative effects on the aquatic invertebrate Daphnia magna, was the conclusion drawn by Cuhra, Traavik & Bohn (2013). They found concentrations below 10 mg/L induced immobility in D. magna within 48 h. both Roundup and glyphosate at 0.05 mg a.s./L reduced juvenile size with 4.05mg a.s./L reducing growth and fecundity, increasing mortality, and increasing abortion.

2.4.9 Aquatic plants

Glyphosate can reach aquatic ecosystems via run-off from agricultural land or spray drift from terrestrial applications, and also following intentional applications to control aquatic vegetation which can result in greater concentrations in aquatic ecosystems than from terrestrial uses.

Glyphosate was taken up by duckweed (Lemna minor), in plants exposed to 3μM of glyphosate for 7 days, the herbicide induced morphological and biochemical changes in non-target plants and exerted toxic effects on aquatic ecosystems even during short-term exposure (Sikorski et al., 2019; Tajnaiová et al., 2020).

Sea clubrush (Bolboschoenus maritimus) plants were able to grow and survive after 20 days of exposure to glyphosate, but the herbicide affected their growth, through a direct interaction with the root system. The reduction in growth was due to a decrease in the net photosynthetic rate. Glyphosate excess caused photo-inhibitory damage. In this study Mateos-Naranjo & Perez-Martin (2013) demonstrated that presence of the herbicide in water as a pollutant could be a source of indirect phytotoxicity for B. maritimus.

Different concentrations of glyphosate-based herbicides were trialled on wetland plant communities over two years of herbicide application (alone and in combination with agricultural fertilizers) and two subsequent years without herbicide (or fertiliser) application. The herbicide was applied to achieve low and high target aqueous glyphosate concentrations of 210 and 880µg a.e./L respectively. The application of glyphosate-based herbicides was found to result in a 25% decrease in macrophyte species richness and cover, and an increase in macrophyte species evenness (the count of individuals of each species) by 25% when compared to the untreated. The glyphosate effects were evident in the first year of herbicide application (2009), and became more pronounced in the second year of herbicide application (2010). However, when herbicides were not applied in the following year, recovery was observed in most endpoints, with the exception being species evenness, for which partial recovery was not observed until 2012 (up to 3 years after treatment) (Mudge & Houlahan, 2019).

2.4.10 Marine

In a review of the effects of glyphosate on marine invertebrates (Matozzo et al., 2020), data obtained in acute toxicity tests indicated that glyphosate and its commercial formulations were lethal at high concentrations (not environmentally realistic). Results of several weeks’ exposure to glyphosate (time exposed to a concentration) can markedly affect biological responses of marine invertebrates.

Glyphosate can be used as a phosphorous source by a range of phytoplankton species while it is toxic to some other species and yet has no effects on others (Wang et al., 2016). The observed differential effects suggest that the continued use of glyphosate in agriculture and the resulting run-off and increasing concentration of this herbicide reaching coastal waters will likely exert significant impact on coastal marine phytoplankton community structure.

Lipok et al. (2010) showed that that the presence of isopropylamine in commercial formulation of herbicides containing glyphosate increases the toxicity of this herbicide towards photosynthesizing aquatic phytoplankton, namely cyanobacteria and algae when compared to the pure active substance of glyphosate alone. The use of the surfactant polyethoxylated tallowamine was withdrawn from glyphosate products in the EU in 2017.

2.5 Triclopyr

Products containing triclopyr are registered for the control of perennial broadleaved weeds, brambles, docks, scrub, common nettle, rhododendron and woody weeds, on amenity grassland, and grassland.  It can also be used to control unwanted standing coppice or scrub and for the prevention of shoot growth on cut stumps. It can be purchased as the single active substance or in mixture with fluroxypyr, clopyralid, 2,4-D or aminopyralid. Mixtures are only approved for use on amenity grassland and grassland.

Triclopyr is a selective herbicide used to treat woody and herbaceous plants but has little or no impact on grasses or conifers. It works by mimicking the auxin hormone and causes exposed plants to grow excessively fast and die over the course of several weeks.

Triclopyr is very soluble in water, and moderately persistent in soil, and constitutes a slight risk to groundwater (Lewis et al., 2016).  Its main soil metabolite is more persistent, and mobile, and the risk of groundwater contamination is greater.  Generally, triclopyr has low toxicity to mammals and birds, bees and other wildlife; but its derivatives and main metabolite are moderately toxic to aquatic fauna.  

Triclopyr in water breaks down faster with light. The half-life of triclopyr in water with light is around 1 day. Without light, it is stable in water with a half-life of 142 days

Triclopyr breaks down relatively quickly in soils. It is mainly broken down by microbes. The soil half-life ranges from 8 to 46 days. In deeper soils with less oxygen, the half-life is longer. Triclopyr is mobile in soils. However, movement studies show that triclopyr was not measured in soils deeper than 15 to 90cm (about 6 to 35 inches). Its movement in soil is affected by the amount of compost and rain, among other factors.

As a systemic herbicide, triclopyr is absorbed through plant leaves and roots. It tends to accumulate in the growing points in a plant. The half-life in plants can vary widely with the type of plant. Barley and wheat plants broke down 85% of triclopyr within 3 days of application. The half-life in grass was between 5 and 20 days. The half-life in plants ranges from 3 to 24 days.

2.5.1 Current status

A review report for triclopyr was completed in 2014, in view of the inclusion of triclopyr in Annex I of Directive 91/414/EEC.  It concluded that triclopyr fulfils the safety requirements of the directive and when used as proposed, under good plant protection practice, has no harmful effects on human or animal health.

The following conditions should be taken into account when authorising plant protection products containing triclopyr: 

• It should only be used as a herbicide and a total application per year of maximum of 480g a.s./ha.
• Groundwater should be protected under vulnerable conditions. Conditions of authorisation should include risk mitigation measures and monitoring programmes should be initiated in vulnerable zones.
• Safety of operators should be highlighted and conditions of use prescribe the application of adequate personal protective equipment.
• Particular attention should be made to the protection of birds, mammals, aquatic organisms and non-target plants. Conditions of authorisation should include risk mitigation measures, where appropriate.
• Further studies were requested to confirm the acute and long term risk assessment for birds and mammals and the risk to aquatic organisms from exposure to the metabolite 6-chloro-2-pyridinol.

The current approval for triclopyr in GB ends on 30 April 2024. The Draft Renewal Assessment Report prepared according to the Commission Regulation (EC) No 1107/2009 was completed in 2018 and a public consultation ended in 2019 in preparation for the renewal of triclopyr approval.

2.5.2 Drift 

Triclopyr drift and dissipation in foliage were assessed following a targeted low-volume foliar or basal bark application (Voinorosky, 2020). Concentrations in foliage were higher following low-volume foliar applications but dissipated to 50% of initial concentrations within a week. The indirect impacts of triclopyr on habitat quality were also examined through litter mass loss and carbon:nitrogen ratios. They reported higher concentrations of nitrogen (lower carbon:nitrogen) in field treated foliage but litter breakdown was unaffected.

2.5.3 Herbaceous plants

Extensive damage was caused to yarrow (Achillea millefolium) by triclopyr application at <50% of the field application rate and rosebay willowherb (Chamerion angustifolium) was killed by the lowest rate used (Isbister et al., 2017). This indicates there may be indirect effects on desirable species, the product label does not contain detailed information on the species controlled and more detailed information will be available from Corteva (manufacturer).

Clopyralid and triclopyr were applied at the recommended rate for managing Broom (Cytisus scoparius). These soils were sampled periodically and sown with C. scoparius seeds. They found that the herbicides were residual in the soil and the longer it had been since the herbicide had been applied, the less it suppressed C. scoparius germination and growth (Tran et al., 2015).

2.5.4 Lichens

In Canadian studies triclopyr reduced the abundance of 40% of the lichen species studied (McMullin et al., 2011). The species with the highest mortality were Bryoria furcellata, Cladonia uncialis, and Trapeliopsis granulosa.

2.5.5 Invertebrates and soil biota

In a study of wood ants (formica sp), triclopyr was shown to have no effect on the studied behaviour traits (visibility, speed, activity and exploration) Karlsten (2017).

In Canada containers filled with soil were sprayed with triclopyr (Garlon 4) at field rate of 4.48kg/ha.Two ground beetles, Pterostichus inanis and, Scaphinotus marginalus; an isopod (Ligidium sp); and a millipede (Harpaphe haydeniana) were added to the container immediately after spraying. The isopod displayed significant mortality rate after seven-day exposure, however the other four species were unaffected (Moran, 1999).

The springtail, Folsomia candida, and soil mite, Oppia nitens, were seen to avoid leaf litter treated at rates above field application rates of triclopyr; survival and reproduction rates of F. candida were unaffected by field rates (Voinorosky, 2020).

The impacts of triclopyr on soil microbial communities were described in two studies, both of which found that triclopyr altered the community composition of bacterial communities (Souza-Alonso et al. 2015, Marileo et al. 2016). Bacterial communities were only measured following two applications of triclopyr (Souza-Alonso et al. 2015). Fungal diversity and community composition were not impacted by triclopyr (Souza-Alonso et al. 2015). These studies taken together suggest that bacterial communities can be impacted by either two applications at the recommended rate or by one application at twice the recommended rate, but that they are unaffected by one application at the recommended rate.

2.5.6 Amphibians, fish, aquatic invertebrates and other freshwater species

In Canada, survival of a range of aquatic insects were differentially affected by triclopyr applied to stream channels; in in-flow-through bioassays with triclopyr ester, 10 of 12 test species showed no significant mortality at concentrations greater than 80 mg/L. Survival of springflies (Isogenoides sp). and the caddisfly (Dolophilodes distinctus) was significantly affected at less than 80mg/L (Kreutzweiser et al, 1992).

Triclopyr has been shown to be genotoxic to the European eel (Anguilla anguilla L.) and caused DNA damage (Guilherme et al., 2015).

No effects were observed on the survival of tadpoles of the northern leopard frog (Lithobates pipiens); although, they were physically smaller after the exposure to triclopyr (Curtis & Bidart, 2017; Yahnke et al., 2017). Tests on embryos and tadpoles of the green frog (Rana clamitans), leopard frog (Rana pipiens), and bull frog (Rana catesbeiana) show that triclopyr had no effect on the hatching, behaviour, or their size but freshly-hatched individuals were very sensitive to the herbicide, and increased mortality or paralysis were observed (Berrill et al., 1994).

2.5.7 Aquatic plants

Triclopyr + fluroxypyr caused 100% growth inhibition of Lemna minor even at significantly lower concentrations (0.01%) than the ready-to-use concentration (Tajnaiova et al., 2020).

2.6 Fluroxypyr

Fluroxypyr is registered for the post-emergent control of certain broad-leaved weeds in grassland and amenity grassland. The herbicide is available in mixtures with florasulam, clopyralid, triclopyr, 2,4-D, dicamba and aminopyralid. Weeds should be small and actively growing.

Fluroxypyr is of low mammalian toxicity, and not expected to adversely impact birds, bees, and soil fauna (Lewis et al., 2016).  However, it is moderately toxic to earthworms and aquatic organisms (and moderately toxic to aquatic plants).

2.6.1 Current status

A review report for fluroxypyr was completed in 2017 (European Commission , 2017a), in view of the approval of fluroxypyr as an active substance in accordance with Regulation (EC) No 1107/2009. It concluded that fluroxypyr fulfils the safety requirements of the directive. Particular attention should be paid to the following and risk mitigation measures need to be included: 

• The potential contamination of groundwater by metabolite fluroxypyr pyridinol, when the active substance is applied in regions with alkaline or vulnerable soil or with vulnerable climatic conditions.
• The risk to aquatic organisms.
• An earlier review identified that a study on the long-term risk for earthworms and soil organisms was required, this was completed and the long-term risk was considered acceptable.

The current approval for fluroxypyr in GB ends on 30 December 2024.

2.6.2 Literature

In Taraxacum vulgare and Trifoium pratense flowering was delayed and the number of flowers reduced by exposure to fluroxypyr (7.2 - 144.0g a.s./ha) (Boutin et al., 2014).

Fluroxypyr, was found to be toxic to green algae (Chlamydomonas reinhardtii) with low concentrations (0.05–0.5mg/L) found to stimulate its growth and high levels (0.75–1.00mg/L) found to inhibit its growth (Zhang et al., 2011).

2.7 Clopyralid

Clopyralid is a selective, post-emergent, systemic herbicide used to control annual and perennial dicotyledons, including corn marigold (Glebionis segetum) and creeping thistle (C. arvense), in grassland and amenity grassland.  In these situations clopyralid is sold in mixture with florasulam, fluroxypyr, MCPA and triclopyr. No applications may be made between 31st August and 1st March.

Clopyralid has a high aqueous solubility, is volatile and there is a high risk of it leaching to groundwater (Lewis et al., 2016). It can be persistent in both soil and water systems depending upon conditions. It has a low mammalian toxicity and is not expected to bioaccumulate. It is moderately toxic to birds, fish, aquatic invertebrates, honeybees and earthworms. It has a low toxicity to aquatic plants and algae.

2.7.1 Current status

A review report for clopyralid was completed in 2018. The context of the peer review was required by Commission Implementing Regulation (EU) No 844/2012. It concluded that the risk to birds and mammals, bees, non-target arthropods other than bees, soil dwelling macro-fauna other than earthworms and soil micro-organisms was low. The risk to aquatic organisms, earthworms and non-target plants was assessed as low for exposure in the field. 

The acute risk assessment resulted in a low acute risk to honeybees. The acute and chronic risk to adult bees and the risk to larvae were assessed as low. No data were available for bumble bees and solitary bees.

The current approval for clopyralid in GB ends on 30 April 2026.

2.7.2 Literature

Of nine species tested (Centaurea cyanus, Silene noctiflora, Viola arvensis, Cerastium arvense, Cirsium arvense, Epilobium montanum, Knautia arvensis, Taraxacum officinale and Trifolium pretense), clopyralid had low effect on biomass of only S. noctiflora (1% dose, bud stage); T. officinale (5% dose, 6-8 leaf stage) and T. pratense (5% dose, bud stage) showed effects (Strandberg et al., 2019). The herbicide also had effects on seed production of T. pratense and C. cyanus. Flowering was delayed for 15 days after application of clopyralid at 1% and 5% of the field rate (80g a.s./L) with some reductions in flower numbers seen in C. arvense, T. pratense and V. arvensis mainly after applications at the 6-8 leaf stage.

Clopyralid and triclopyr were applied at the recommended rate for managing Cytisus scoparius (broom) to a silt loam soil. Soil was collected periodically from the treatments, placed in pots and sown with C. scoparius seeds. They found that the longer it had been since the herbicide had been applied the less it suppressed broom germination and growth (Tran et al., 2015).

Oak seedlings showed spatulated leaf tips within 1 month of application of clopyralid but this symptom then disappeared (Vasic et al., 2009).

2.8 MCPA

Approvals for products containing this active ingredient are restricted to grassland, managed amenity turf and amenity grassland.

This herbicide is available alone or in combination with mecoprop-P, clopyralid, fluroxypyr, 2,4-D, 2,4-DB, dicamba and iron sulphate.  It is approved for the control of annual and perennial broad-leaved weeds. The weeds should be actively growing at application and generally at the seedling stage.

MCPA is moderately toxic to birds, fish, earthworms and aquatic plants but of low toxicity to aquatic organisms and honeybees.

2.8.1 Current status

A review report for MCPA was completed in 2017, in view of the inclusion of MCPA in Annex I of Directive 91/414/EEC. It concluded that MCPA fulfils the safety requirements of the directive. It should only be used as a herbicide. 

Particular attention should be paid to the potential for groundwater contamination, when the active substance is applied in regions with vulnerable soil and/or climatic conditions. Conditions of authorisation should include risk mitigation measures, with particular attention to the protection of aquatic organisms and must ensure that the conditions of authorisation include risk mitigation measures, where appropriate, such as buffer zones.

The current approval for MCPA ends on 31 October 2023.

2.8.2 Literature

Breeze et al (1992) studying spray drift calculated that MCPA would pose little risk to any of the plant species in Table 11. Herbicide was applied as single droplets of 5µL in a dose range of 0.001 to 1000µg a.s./plant. The highest rate was equivalent to the dose a plant would receive during field spraying. Of the four herbicides tested (Asulam, glyphosate, MCPA and mecoprop) MCPA was one of the least toxic to non-target plants, with only Centurea nigra having an ED₁₀ of less than 1.0µg/plant.

 

Table 11. ED₁₀ and ED₅₀ values with standard errors (μg/plant) for toxicity of MCPA to some wild plant species from Breeze et al. (1992) (1000µg a.s./plant = field dose rate).

Common name	Species	ED10	SE	ED50	SE
Cuckoo flower	Cardamine pratensis	4.7	1.3	22	2.9
Common knapweed	Centaurea nigra	0.6	0.2	5.8	0.9
Crested dog’s tail	Cynosurus cristatus	12	50	>1000	-
Hedge bedstraw	Galium mollugo	470	340	930	85
Perforate St John's-wort	Hypericum perforatum	640	0.2	>1000	-
Rough hawkbit	Leontodon hispidus	15	4.1	68	10
Perennial ryegrass	Lolium perenne	3.8	9.8	>1000	-
Common bird's-foot-trefoil	Lotus corniculatus	4.2	8.5	>1000	-
Ragged robin	Lychnis flos-cuculi	25	7.7	69	10
Meadow buttercup	Ranunculus acris	42	18	120	25
Betony	Stachys officinali	1.8	1.5	32	12
Upright hedge-parsley	Torilis japonica	17	8.8	65	15
Red clover	Trifolium pratense	38	13	160	31

* not measured

ED₅₀ - (median effective dose) is the dose of a herbicide that produces a specific effect in 50% of the population that takes that dose

ED₁₀ - is the dose of a herbicide that produces a specific effect in 10% of the population that takes that dose

 

Sub-lethal doses of MCPA were shown to reduce the seed production of Chamomilla recutita and Bilderdykia convolvulus; in Thlaspi arvense seed size was reduced (Andersson, 1994).

In Ramalina fraxinea (cartilage lichen), MCPA induced a reduction in the maximum capacity of Photosystem II (PSII), in chlorophyll content along with chlorophyll degradation, and in adverse effects on cell membrane integrity, and induced oxidative stress (Sujetovienė et al., 2019).

No effect of MCPA was seen on epigeal predator fauna in the polders of the Netherlands (Everts et al., 1989). Plots on arable land in southern England received an annual application of MCPA herbicide for 10 out of 13 years -  no effects of MCPA were recorded (Davis,1965).

Eelgrass (Zostera marina) and a natural community of phytoplankton were assessed after treatment with MCPA.  There were no significant direct effects on eelgrass but phytoplankton production was stimulated (Nielsen & Dahllöf, 2007).

Two hundred crested newts (Triturus cristatus carnifex) were tested at concentrations equivalent to 100, 200, and 400 ppm of the active substance MCPA for 1 year.  Under these experimental conditions, there was no carcinogenic activity of MCPA. (Zavanella et al., 1988).

2.9 Citronella oil

Citronella oil is approved for the control of ragwort (Jacobaea vulgaris) in amenity grassland, green cover on land temporarily removed from production, and grassland. Applications are made as a spot treatment through hand held sprayer.

Citronella is of low toxicity to birds, honeybees, earthworms and mammals but of moderate toxicity to fish and aquatic invertebrates (Lewis et al., 2016).

2.9.1 Current status

A review report for Citronella oil was completed in 2012 in view of the inclusion in Annex I of Directive 91/414/EEC. It concluded that Citronella oil does not have any harmful effects on human or animal health or on groundwater or any unacceptable influence on the environment. 

Conditions of use should include risk mitigation measures with particular attention to: 

• the protection of operators, workers, bystanders and residents, ensuring that conditions of use include the application of adequate personal protective equipment, where appropriate.
• the protection of groundwater, when the substance is applied in regions with vulnerable soil.
• the risk to non-target organisms.
• The current approval for Citronella oil in GB ends on 28 February 2025. It is not classified as a low-risk substance.

Citronella oil has moderate toxicity to fish, aquatic invertebrates and algae, often a greater toxicity than the other herbicides in this review (Table 5). No relevant results were found in the literature search. 

2.10 Summary of the literature review 

The pesticide registration process includes a requirement for detailed studies on the direct effects on non-target organisms, but these are standard test organisms, which include terrestrial vertebrates, invertebrates and aquatic species. Toxicological testing now includes consideration of non-target arthropods including parasitic wasps, predatory mites and also ground beetles, rove beetles, spiders, ladybirds and lacewings.  Other soil dwelling arthropods such as springtails and soil mites may also be tested for specific pesticides.

A rigorous set of tests is done on how the pesticide might move throughout the environment once it has been applied; the impact on the environment, in particular water bodies and wildlife; and product efficacy. This set of tests is limited and this literature review has shown that studies done after the herbicides were placed on the market, show harmful effects on a range of different organisms, and that these effects are sometimes significant and serious.

Indirect links between pesticide use and species occur where the availability of essential resources is changed as a result of pesticide applications. This can lead to population scale changes in non-target species. For example, indirect links as a result of changing food availability do not only affect farmland birds; reduced plant diversity and abundance reduces the species richness of non-target invertebrates. To demonstrate indirect links operating through food chains, evidence is needed that pesticides are the cause of changes in food abundance, which lead to a change in species at the population level (Boatman et al., 2005). There are, however, a number of other factors that impact on the population dynamics of species, which are often linked to pesticide use, although disentangling the relative impact of each is extremely difficult, and highly context dependent.

As described in section 1.4 pesticides are assessed and approved with a standard approach. Additional research reported in peer-reviewed literature is done and is often proportional to the length of time the pesticide has been on the market.

The use of asulam is not approved in the EU, and after ten years of emergency authorisations in GB, as of October 2023, its use in Scotland is no longer permitted.  This means that there will be no asulam (Asulox) available for bracken control in 2024 and beyond. Asulam is a herbicide of low toxicity to a range of broad-leaf non-target plants but is toxic to ferns and mosses. There is no information on its effects on lichens.  Its effect on bees is low, but it has been found in honey. There are concerns over its effects on ground nesting birds, and the long term risk to birds and wild mammals is of critical concern.  Asulam is highly soluble in water and toxic to aquatic organisms.

Glyphosate is a herbicide of great controversy, used extensively throughout agriculture and in nature conservation. It is due for renewal in the EU in December 2023 and is already subject to withdrawals of use in countries such as France where it can no longer be used in alleys between vines and fruit trees, in amenity, nature conservation and forestry, or in crop fields that are ploughed. Germany announced total withdrawal of Glyphosate by 2024. During the renewal of glyphosate in 2017 the use of glyphosate was restricted in public parks and gardens, sports and recreation grounds, school grounds and children’s playgrounds and in the close vicinity of healthcare facilities, but it still remains a herbicide that has ‘no unacceptable effects on the environment’.

Glyphosate has a high level of toxicity on non-target plants and is effective at very low levels on sensitive plants. Even if it does not kill it can reduce growth and flowering because it translocates to the growing point of plants. Prevention of drift should be a priority area for action, but it is unlikely transfer of herbicide from plant to plant will occur. Lichens are also sensitive to the herbicide. Glyphosate can change the soil microbiome over short periods but generally these effects do not last long.  Arachnids are relatively unaffected by the herbicide with low effects on mortality but sometimes activity is increased and behaviour changed, many of these were associated with changes in the vegetation habitat. Earthworms were frequently associated with negative effects. Insects are affected in a similar way to arachnids the herbicide affecting both activity and behaviour again associated with changes in habitat. Bees are affected more by changes in the abundance of flowering plants but there were toxic effects also. Glyphosate is not approved for use in waterbodies but some formulations are approved for uses near water, the active substance glyphosate has often been shown to be less toxic than other substances included in the product. The herbicide does have effects on the aquatic environment and the avoidance of drift is key to avoidance of problems.

Triclopyr has high solubility in water and the risk of groundwater contamination is high and should be protected where it is used. Risk mitigation is key to the protection of water with this herbicide. Some herbaceous plants are sensitive to triclopyr and information on this area should be sought from the manufacturer or through simple pot trials. Lichen is also susceptible to the herbicide. Application of the herbicide to water is not permitted.

Fluroxypyr usually forms a component of a herbicide mixture along with clopyralid and MCPA. All of these herbicides pose a risk to non-target plants and details on susceptible species can be obtained from the manufacturer or could be obtained from simple pot trials. The use of these herbicides is not permitted in water and risk mitigations should be in place to prevent the herbicides reaching water.

Citronella oil has limited label approvals being limited to the control of ragwort. Its use in water is not permitted.

A recently conducted hazard assessment of pesticides and soil invertebrates (Gunstone et al., 2021) reviews 400 studies on the effects of pesticides on non-target invertebrates that have egg, larval, or immature development in the soil. The review encompassed 275 unique species, taxa or combined taxa of soil organisms and 284 different pesticide active ingredients or unique mixtures of active ingredients. Both field and laboratory studies were reviewed, but the data are only available for the combination of both in the two herbicide types relevant to this report. 

Earthworms were the most frequently studied with around half the studies showing negative effects for both glyphosate and the auxin herbicides (includes the herbicides fluroxypyr, triclopyr, MCPA and clopyralid) (Table 12).  23% of studies on beetles and 50% of studies on springtails showed negative effects. Nematodes were the taxa that were most negatively affected by glyphosate (86% of seven studies).  

 

Table 12. The number of tested parameters (# par.) and the percentage that resulted in negative effects (% neg.) to each soil taxa after exposure to different pesticide types.

-	Glyphosate # par.	Glyphosate % neg.	Auxin* # par.	Auxin* % neg.	Mean of all  Herbicides # par.	Mean of all  Herbicides % neg.
Oligochaeta (earthworms)	100	60	38	55	315	72
Enchytraeidae (potworms)	1	0	0	-	49	88
Nematoda (roundworms)	7	86	0	-	27	67
Tardigrade (water bears)	0	-	0	-	0	-
Acari (mites)	3	33	4	0	18	39
Myriapoda (millipedes, centipedes, Pauropoda)	0	-	0	-	2	50
Isopoda (woodlice)	3	33	0	-	9	22
Collembola (springtails)	10	50	4	0	46	65
Protura (coneheads)	0	-	0	-	0	-
Isoptera (termites)	0	-	0	-	0	-
Coleoptera (beetles)	13	23	7	29	59	24
Formicidae (ants)	0	-	3	0	4	0
Bombus spp. (bumble bees)	0	-	0	-	0	-
Ground-nesting bee (non-Bombus)	0	-	0	-	0	-
Parasitic wasp	0	-	0	-	2	100
Mixed taxa	4	25	0	-	10	20
Total parameters and negative effect %	141	55	56	41	541	63
no effect %	-	42	-	59	-	35
positive effect %	-	3	-	0	-	1

 *includes the herbicides fluroxypyr, triclopyr, MCPA and clopyralid

 

Of endpoint categories, structural changes and biochemical biomarkers were the most impacted by herbicides followed by mortality, behaviour, reproduction growth, richness and diversity, abundance, and lastly, biomass (Table 13).  The auxin herbicides had a greater effect on reproduction and biochemical biomarkers whilst glyphosate followed the average trend of all herbicides.

 

Table 13. The number of tested parameters (# par.) and the percentage that resulted in negative effects (% neg.) on each endpoint category tested on soil invertebrates after exposure to different pesticide types (overall mean).

Endpoint Category	Glyphosate # par.	Glyphosate % neg.	Auxin* # par.	Auxin* % neg.	Total Herbicides # par.	Total Herbicides % neg.
Mortality	26	50	6	67	129	67
Abundance	5	40	17	0	56	27
Biomass	4	25	5	0	15	13
Reproduction	25	56	3	100	78	59
Behaviour	31	48	5	60	81	65
Biochemical biomarkers	30	77	11	100	114	86
Growth	15	47	9	22	52	50
Richness & diversity	3	33	0	-	10	30
Structural changes	2	100	0	-	6	100
Total parameters and negative effect %	141	55	56	41	541	63
no effect %	-	42	-	59	-	35
positive effect %	-	3	-	0	-	1

*includes the herbicides fluroxypyr, triclopyr, MCPA and clopyralid

 

3. Needs analysis

3.1 The desired outcome 

The desired outcome is to better understand usage of key herbicides and the impacts and gaps in evidence for commonly used herbicides in conservation management situations. The use and effect of herbicides in conservation management is not well documented or widely understood. This lack of knowledge may have implications for management particularly of sensitive features that may be susceptible to negative effects from herbicides.

3.2 The current situation

If a private manager wants to apply a herbicide within an SSSI and this is listed as an operation requiring consent, they must obtain written consent from NatureScot, or from a relevant regulatory authority. They must specify:

• The nature of the operation (i.e. enough information on all aspects of the operation to allow an assessment of impacts on any of the sites notified features).
• The proposed dates for commencement and completion of the operation.
• The land on which it is proposed to carry out the operation clearly identified to a sufficient level of accuracy.

There were two sources of herbicide application data. Data were taken from NatureScot’s  protected areas database detailing SSSI consents issued (but not necessarily actioned) between 1983 and 2006. A further survey across all NatureScot’s nature reserves during January 2021 identified the most frequently used herbicides, the purpose of their use, application method, target, periodicity of use and the extent of application for the previous 5 years and including proposed treatments for 2021.

The results show that between 1983 and 2006 asulam was the most frequently applied herbicide on SSSIs followed by glyphosate, then products containing MCPA. In 2021 glyphosate was the most frequently applied herbicide on NatureScot nature reserves, 85% of reported usage (Table 14), with triclopyr and clopyralid being the only other herbicides used (15% of usage). Herbicides were not used on 22 reserves.

The majority of herbicide usage by NatureScot was to control INNS, followed by keeping paths/tracks and roads clear. Control of native vegetation was the only other reason for use noted (Table 15).

 

Table 14. Herbicides used on nature reserves

Herbicide	All SSSIs (MIDAS) Consents granted*	NatureScot Nature reserves only Frequency of use	NatureScot Nature reserves only No of reserves
-	1983-2020	2017-2021	2017-2021
Asulam	90	0	0
Glyphosate	51	29	20
MCPA/ MPCA mix	28	0	0
Triclopyr or triclopyr mix	2	4	4
Citronella oil	2	0	0
None	-		22

*consents granted do not necessarily reflect actual use.

 

Table 15. Purpose of herbicide use on nature reserves – 2017-2021 only

Purpose	Frequency of use	No of reserves
To control INNS	17	14
To control native vegetation	7	4
To maintain paths/tracks/roads	10	10

 

A wide range of application methods were used on nature reserves to apply glyphosate (Table 16) with application through a knapsack being the most frequent (44%) followed by injection (23%) and handheld sprayer (10%). Between 1983 and 2006 hand held sprayer and weed wiper were the most common methods of application recorded with knapsack and spot spraying reported next. According to NatureScot’s protected areas database, aerial spraying was consented 43 times between 1983-2020 on SSSIs, predominantly for asulam for bracken control, but no aerial spraying was reported on its nature reserves.

 

Table 16. Application methods used to apply glyphosate on nature reserves – between 2017-2021

Application method - Glyphosate	Frequency of use	No of reserves
Spray - knapsack	17	14
Injection	9	9
Spray - handheld	4	3
Paintbrush	3	1
Ecoplugs	2	2
Spray - vehicle mounted	1	1
Weedwipe/boom - vehicle mounted	1	1
Dabbing or 'glove-of-death'*	1	1
Sponge	1	1
Spray - aerial	0	0
Granules	0	0
Weedwipe/ boom - handheld	0	0
Total	39	33

*glove-of-death is application of glyphosate by wiping with a wetted glove

 

Between 2017-2021 rhododendron was the most frequently treated species, with Japanese knotweed, grass and general weed control being the other top four applications (Table 17). The majority of the target uses were for perennials with a high proportion of woody species.

In comparison, the ‘top ten’ most  frequently treated weeds between 1983 and 2006 were giant hogweed, Japanese knotweed, ‘weeds in the Weeds Act 1959’, bracken, thistles, rushes, rhododendron, nettles and sycamore.

Notably bracken control was not identified in the survey as this was all controlled mechanically.

 

Table 17. Target of herbicide application in nature reserves-2017-2021

Target species/habitat	Frequency of use	No of reserves
Rhododendron	7	7
Japanese knotweed	6	4
Grass	6	6
General weed control	4	4
Himalayan balsam	3	3
Gorse	3	2
Nettles	2	2
Ragwort	2	2
Piri piri	2	2
Giant hogweed	1	1
Rushes	1	1
Bamboo	1	1
Beech	1	1
Birch	1	1
Broom	1	1
Crocosmia	1	1
Cotoneaster	1	1
Rosa rugosa	1	1
Sitka spruce	1	1
Sycamore	1	1
Western hemlock	1	1
Willow	1	1

 

Herbicides were generally applied once per year with eight sites reporting applications made several times per year (Table 18). The pattern of applications was similar to that in the collation of SSSI consents 1983-2006.

 

Table 18. Periodicity of use of herbicides in nature reserves - 2021

Periodicity of use	Periodicity
Once per year	17
Several times per year	8
Once in last 5 years	7
Twice in last 5 years	1
As required by contractor	1

 

The 2017-2021 survey reported that most herbicides are applied in localised areas rather than extensively (Table 19), with the larger areas being treated limited to paths and around electric fences.

 

Table 19. Extent of application of herbicides in nature reserves (area treated)

Extent of application	Frequency of use	No of reserves	Comment
Spot spray	23	18	-
Length <=100m	3	3	All path sides
Length 101m to 300m	1	1	electric-fence
Length 301m to 1000m	0	0	-
length >1000m	3	3	All path sides

 

These surveys have identified the main herbicides used in nature conservation management on protected areas – and believed to be typical of uses for nature conservation management elsewhere. In recent years, they were mainly glyphosate and products containing triclopyr or triclopyr plus a further active ingredient. In the past asulam was used widely, but as of September 2023 there will be no more asulam (Asulox) available for bracken control in Scotland.  

3.3 The knowledge gaps and solutions

3.3.1 Knowledge gaps

The literature review has indicated that all of the pesticides uses except for asulam are fully approved and considered safe by the EU and HSE. As part of the approval process a rigorous set of tests is done on how the pesticide might move throughout the environment once it has been applied; the impact on the environment, in particular water bodies and wildlife; and product efficacy. This set of tests is limited in scope to a range of species chosen to represent specific groups of organisms. After approval, other interested parties may continue to test the effects of the pesticide against a broader range of organisms and these tests may find deleterious impacts that were underestimated or missed entirely in the official regulatory evaluation. In these circumstances, review panels can recommend that the product is removed from the register of approved products, or that it is not renewed in its current form once the original approval expires.

There are gaps in the information available for each pesticide as indicated by the literature review (Table 5). Whilst the effects of herbicides on non-target herbaceous plants, aquatic plants, ferns and mosses are relatively inexpensive and simple to test for, tests for the effects of herbicides on non-target fauna, including soil biota are more problematic and expensive.

Asulam has been authorised for emergency use for bracken control in GB for 10 years pending approval in the EU and GB, as there were concerns over its safety, and alternative approaches were more difficult to implement.  However, as of October 2023, asulam is off the list of available chemicals, as UPL, the manufacturers of Asulam, have decided not to pursue applying for full authorisation in the UK.  This will mean that there will be no asulam (Asulox) available for bracken control in 2024 and beyond.

The information found in the literature search on flora and fauna was specific to selected organisms, and responses of each species could vary considerably depending on how the experiments were done. There is a vast range of non-target organisms that could be evaluated and these need to be more clearly defined. It is important for NatureScot to identify situations in which herbicide use is considered necessary, this could be due to practical reasons such as large areas for treatment, access issues for mechanical or thermal equipment and treatment of invasive perennial weeds e.g. Japanese knotweed. When the use of a herbicide is considered necessary the treatment zone needs to be considered and vulnerable species identified within it. These species could then be tested for their sensitivity to the herbicide in question. This would be relatively simple for herbaceous plants, aquatic plants, ferns and mosses.

The review has indicated how critical the growth stage of the flora is in relation to the timing of the herbicide application. This has two aspects; a) the effect on processes within the plant at the time of application of the herbicide e.g. effects on flowering and seed production and b) the effects on non-target species who use the plant as a food source e.g. bees for nectar.

The review has indicated the importance of preventing pesticide drift to non-target organisms, particularly glyphosate on non-target plants and asulam, triclopyr, fluroxypyr, clopyralid, MCPA and citronella oil to water.

The SSSI consent application form is generic and does not specify the type of information required to make an informed decision on herbicide use. The site must have been through a process to consider IPM alternatives to herbicide application. The legislation for the determination of applications for SSSI consent allow only for consideration of likely damage to the site’s notified features (and not wider biodiversity).  There is also a lack of requirement to state the qualifications of those recommending the pesticide and the pesticide operators. There is a case to be made that this process needs to be more clearly defined, to include information regarding the process that identifies the need for the pesticide, the names and qualifications of those making the decision and a site based risk assessment that identifies both the target, and the key non-target organisms that should be protected. After treatment there should be feedback on the results of the application, both positive and negative, what worked well and how improvements could be made. This information could be collated and shared, building a database and used to inform future pesticide consents. This could be considered in phase 3 of this project.

3.4 Conditions of use 

A wider range of herbicides are approved for use in nature conservation management and can be located through the HSE CRD database using the crop or situation detailed in Table 20.

 

Table 20. Basic crops and situations, and associated definitions, most relevant to nature conservation sites.

Basic Crop or Situation	Definition
Grassland	Land grown for grass production includes short and long-term grass leys and permanent pasture, which may be grazed and /or cut for subsequent animal consumption. Includes use on newly sown leys and moorland for grazing (unless specifically excluded on the label/authorisation). Excludes use on amenity grass (see ‘Amenity grassland’)
Enclosed waters	Any natural or artificial body of water that does not drain to a water course.
Intertidal zones of estuary	The area between the low and high watermarks of a river estuary. Includes beaches.
Land immediately adjacent to aquatic area	The bank of any water course or body of water. Includes sand dunes.
Open waters	Any natural or artificial body of water that drains to a water course or is used as a reservoir for domestic water supplies.
Saltmarsh	Area of vegetated salt water marsh adjacent to the sea or saline river estuary
Cut log	Any felled timber.
Farm forestry	Groups of trees established on arable land or improved grassland, including those planted for short rotation coppicing.
Forest	Groups of trees being grown in their final positions e.g. after planting out from a forest nursery. Trees grown primarily for commercial production, including ancient traditional coppice and farm forestry or from natural regeneration, colonisation or coppicing. Covers all woodland grown for whatever objective, including commercial timber production, amenity and recreation, conservation or landscaping, ancient traditional coppice and farm forestry. This includes restocking of established woodland and new planting on both improved and unimproved land.
Green cover on land not being used for crop production	Areas of land with a vegetation cover that have been removed (temporarily or otherwise) from production. For example some types of set aside. Includes fields or non-crop field margins covered by natural regeneration or by a planted green cover crop that will not be harvested. Includes conservation crops such as wild bird and pollen/nectar mixes and crops grown for game cover. Crops must not be harvested for human or livestock consumption or used for livestock grazing. Does NOT include use in industrial crops or inter-row use within a crop (edible or non-edible). Since this definition covers a wide range of situations, the commercial risk is entirely the grower's if the product label does not specifically refer to the crop/species mix you are treating
Amenity grassland	Areas of semi-natural or planted grassland subject to minimal or non-intensive management. Includes areas that may be accessed by the public, such as golf roughs. May include airfields and predominantly grassed railway embankments and roadside verges. May be floristically rich and irregularly managed so that plants may flower and set seed.
Amenity vegetation	Any areas of semi natural or ornamental vegetation, including trees. May include parks, railway embankments and roadside verges which are predominantly covered in vegetation other than grass. Also includes areas of bare soil around ornamental plants or intended for ornamental planting. Does NOT include hedgerows around arable fields.
Natural vegetation	Areas of natural vegetation not covered by a situation stated separately in this Definitions List.
Hedgerow	Linearly planted trees and/or shrubs maintained to form a boundary, including those surrounding arable fields.

 

Up to date authorisations for pesticide use can be found in the following places:

The pesticide register database contains information on Plant Protection Products with on-label authorisations. The extension of authorisation for minor use (EAMU) in GB and Northern Ireland allows a search of products with ‘off-label” approval.

Different application techniques are evaluated during the authorisation process but there is no overall list of approved application techniques. In the Code of Practice (HSE, 2021) there is a glossary of the most common terms used to identify equipment for, and methods of, applying pesticide at Annex C, but this is not exhaustive as stem injection is not listed.

Pesticides must be used in line with the conditions of the product approval as stated on the label, but for approved uses not specified on the label, conditions given on the relevant notice of approval must be followed. Unless the label places a legal obligation on the use or proscription of a specific type of equipment to apply the pesticide, the product may be applied using methods other than those recommended as long as:

• the equipment you have chosen is suitable for the intended method of applying the pesticide,
• the COSHH assessment, where appropriate, has shown that the proposed method does not involve an increased risk to health or safety compared to the normal method,
• you have assessed the environmental effects of your intended method of applying the pesticide and your assessment shows there is no increased risk to wildlife or the environment,
• the necessary control measures are in place to reduce, as far as is reasonably possible, the risks to people, wildlife and the environment.’

So there is scope to use methods other than those recommended as long as assessments have been completed prior to use.

The current situation (4.2) identified that herbicide use was generally targeted through the use of knapsack sprayers, injection, paintbrush and ecoplugs. These methods target applications to where the herbicide is required, minimising the amount of herbicides used, and minimising the chances of contamination of non-target organisms.

Some application methods being used such as paintbrush and dabbing or 'glove-of-death' are not specified on product labels and require risk assessments as above.

The opportunities for training in pesticides for advisers has developed in recent years as conservation, enhancing the environment and biodiversity and sustainable farming practices  have become key goals in the farming sector. BASIS have recently introduced a BETA Conservation Management qualification which covers all aspects of environmental management; including habitat creation, soil management, IPM and Water Quality.

For those applying pesticides the mandatory unit for a City & Guilds NPTC Certificate of Competence is the ‘Principles of Safe Handling and Application of Pesticides (PA1)’. Additional specialist modules are available including mounted or trailed sprayers, boat mounted equipment, hand held sprayers, pellets or granules, pesticide plugs in tree stumps and pesticide injection equipment. Spray operators can join the National Register of Sprayer Operators (NRoSO) which provides for continuing professional development (CPD).

Conclusions

Pesticides undergo a rigorous evaluation process in order to be registered but the assessments are targeted primarily to ensure the safety of workers applying the product, consumers of treated produce, residents living adjacent to application sites and others who may be passing-by at the time of treatment. Another set of tests is done on how the pesticide might move throughout the environment once it has been applied; the impact on the environment, in particular water bodies and wildlife; and product efficacy. However, this set of tests is limited in scope and does not cover aspects which are considered important within nature conservation. As has been shown through the literature review, studies show harmful effects on a range of different organisms, and that these effects are sometimes significant and serious. The amount of peer reviewed literature is variable depending on the herbicide and the length of time it has been on the market, some herbicides are subject to public controversy e.g. glyphosate.

Indirect effects of pesticides are more challenging to identify, registration may in the future require more evidence to prevent indirect links from emerging subsequently. There is sufficient evidence to conclude that pesticides have negative indirect effects on biodiversity (Mann et al., 2009; McKenzie and Whittingham, 2009; Gilburn et al., 2015).

Herbicide use is under increasing scrutiny for health and environmental reasons. New application methods are resulting in reduced and more targeted use. A greater number of bioprotectants for pest and disease control are reaching the market predominantly for use in protected environments, but development of bioherbicides is slow. Targeted application methods allow a reduction in overall quantity of herbicide used, but these methods should be risk assessed before use. Alternative technologies are being developed and it is important that nature conservation keeps a watching brief on these changes so they can be adopted if applicable.

NatureScot are following guidelines set out in the Sustainable Use Directive (2009/128/EC) which came into effect in 2011. The directive sets a framework to achieve sustainable use of pesticides where one of the requirements is to take all necessary measures to reduce pesticide use and prioritise the use of non-chemical methods for the control of weeds, pests and diseases. Pesticides can be used as the last resort if all other methods are not suitable. More information can be found in the Integrated Pest Management in Nature Conservation Handbook. When considering applications for consent to use herbicide within an SSSI NatureScot could exercise greater control over the choice of treatment and its extent and encourage consideration of alternative methods. The legislation might usefully also be amended to allow NatureScot to take account of the possible risk of herbicide use on all biodiversity within the site and on adjoining land.

Recommendations

The following recommendations are targeted at managed land where biodiversity outcomes matter.  They follow from or are based upon findings in this review, and are listed in the order in which they should be considered:

1. Advisers and land managers should evaluate the risk posed by the pest species (refer to any treatment thresholds) and make a decision about control and whether the aim should be to contain the problem at a non-damaging level or to eliminate it.
2. Advisers and operators should have knowledge of functional groups interacting directly and indirectly with target species.
3. Advisers should develop an audit for the area around the herbicide target. Consider all key habitats and species and draw up a plan to protect them. The audit is for both pesticide and non-chemical interventions.
4. Authorities and advisers should specify/ designate buffer zones around critical areas, and work with local farmers and agronomists to minimise pesticide use surrounding critical areas.
5. Land managers should use or develop integrated pest management (IPM) protocols prior to considering pesticide use to minimise pesticide use. NatureScot provides a Handbook on Integrated Pest Management in nature conservation areas, which contain decision trees for a range of plants that may cause issues on sites.
6. Authorities should work with water companies. They carry out regular testing of watercourses for pesticide loading and then work with farmers within the catchment area to try to minimise movement of pesticides to water.
7. Land managers and authorities should ensure the use only of specified approved pesticides.
8. Land managers and authorities should ensure the use of trained personnel, including both those advising on pesticide applications and those applying pesticides.
9. Consider the use of low drift nozzles and nozzle shields.
10. Develop the use of targeted application techniques.
11. Use external protection for critical species e.g. shields and covers.
12. Develop the use of low risk substances and bio-pesticides (see section 1.5.
13. Develop the use of biocontrol.
14. Authorities and advisers to liaise with chemical manufacturers re specific issues such as new herbicides, application techniques, and herbicide susceptibilities as they may be able to provide more information.

References

Andersson, L., 1994. Effects of MCPA and tribenuron-methyl on seed production and seed size of annual weeds. Swedish Journal of Agricultural Research, 24(2), 49-56.

Arts, G.H., Belgers, J.D.M., Hoekzema, C.H. and Thissen, J.T., 2008. Sensitivity of submersed freshwater macrophytes and endpoints in laboratory toxicity tests. Environmental Pollution, 153(1), 199-206.

Battisti, L., Potrich, M., Sampaio, A.R., de Castilhos Ghisi, N., Costa-Maia, F.M., Abati, R., dos Reis Martinez, C.B. and Sofia, S.H., 2021. Is glyphosate toxic to bees? A meta-analytical review. Science of The Total Environment, 145397.

Behrend, J.E. and Rypstra, A.L., 2018. Contact with a glyphosate-based herbicide has long-term effects on the activity and foraging of an agrobiont wolf spider. Chemosphere, 194, 714-721.

Berrill, M., Bertram, S., McGilliray, L., Kolohon, M. and Pauli, B., 1994. Effects of low concentrations of forest‐use pesticides on frog embryos and tadpoles. Environmental Toxicology and Chemistry: An International Journal, 13(4), 657-664.

Blackburn, L.G. and Boutin, C., 2003. Subtle effects of herbicide use in the context of genetically modified crops: a case study with glyphosate (Roundup®). Ecotoxicology, 12(1), 271-285.

Boatman N.D., Brickle N.W., Hart J.D., Milson T.P., Morris A.J., Murray A.W.A., Murray K.A. and Robertson P.A. 2004. Evidence for the indirect effects of pesticides on farmland birds. Ibis, 146, 131-143.

Botten, N., Wood, L.J. and Werner, J.R., 2021. Glyphosate remains in forest plant tissues for a decade or more. Forest Ecology and Management, 493, 119259.

Boutin, C., Strandberg, B., Carpenter, D., Mathiassen, S.K. and Thomas, P.J., 2014. Herbicide impact on non-target plant reproduction: what are the toxicological and ecological implications? Environmental Pollution, 185, 295-306.

Breckels, R.D. and Kilgour, B.W., 2018. Aquatic herbicide applications for the control of aquatic plants in Canada: effects to nontarget aquatic organisms. Environmental Reviews, 26(3), 333-338.

Breeze V. G., Thomas G., and Butler R., 1992. Use of a model and toxicity data to predict the risks to some wild plant species from drift of four herbicides. Annals of Applied Biology, 121(3), 669-677.

Brito, I.P., Tropaldi, L., Carbonari, C.A. and Velini, E.D., 2018. Hormetic effects of glyphosate on plants. Pest Management Science, 74(5), 1064-1070.

Brühl, C.A., Pieper, S. and Weber, B., 2011. Amphibians at risk? Susceptibility of terrestrial amphibian life stages to pesticides. Environmental Toxicology and Chemistry, 30(11), 2465-2472.

Cederlund, H., 2017. Effects of spray drift of glyphosate on non-target terrestrial plants - a critical review. Environmental Toxicology and Chemistry, 36(11), 2879-2886.

Cerdeira, A.L. and Duke, S.O., 2010. Effects of glyphosate-resistant crop cultivation on soil and water quality. GM Crops, 1(1), 16-24.

Cuhra, M., Traavik, T. and Bøhn, T., 2013. Clone-and age-dependent toxicity of a glyphosate commercial formulation and its active ingredient in Daphnia magna. Ecotoxicology, 22(2), 251-262.

Curtis, A.N. and Bidart, M.G., 2017. Effects of chemical management for invasive plants on the performance of Lithobates pipiens tadpoles. Environmental Toxicology and Chemistry, 36(11), 2958-2964.

Dabney, B.L. and Patiño, R., 2018. Low-dose stimulation of growth of the harmful alga, Prymnesium parvum, by glyphosate and glyphosate-based herbicides. Harmful Algae, 80, 130-139.

Damgaard, C., Strandberg, B., Mathiassen, S.K. and Kudsk, P., 2014. The effect of glyphosate on the growth and competitive effect of perennial grass species in semi-natural grasslands. Journal of Environmental Science and Health, Part B, 49(12), 897-908.

Davies, L.R. and Neve, P., 2017. Interpopulation variability and adaptive potential for reduced glyphosate sensitivity in Alopecurus myosuroides. Weed Research, 57(5), 323-332.

Davies, L.R., Hull, R., Moss, S. and Neve, P., 2019. The first cases of evolving glyphosate resistance in UK poverty brome (Bromus sterilis) populations. Weed Science, 67(1), 41-47.

Davis, B.N., 1965. The immediate and long-term effects of the herbicide MCPA on soil arthropods. Bulletin of Entomological Research, 56(2), 357-366.

Druart, C., Millet, M., Scheifler, R., Delhomme, O. and De Vaufleury, A., 2011. Glyphosate and glufosinate-based herbicides: fate in soil, transfer to, and effects on land snails. Journal of Soils and Sediments, 11(8), 1373-1384.

Duke, S.O., 2018. Interaction of chemical pesticides and their formulation ingredients with microbes associated with plants and plant pests. Journal of Agricultural and Food Chemistry, 66(29), 7553-7561.

Duke, S.O., 2020. Glyphosate: environmental fate and impact. Weed Science, 68(3), 201-207.

Duke, S.O., Lydon, J., Koskinen, W.C., Moorman, T.B., Chaney, R.L. and Hammerschmidt, R., 2012. Glyphosate effects on plant mineral nutrition, crop rhizosphere microbiota, and plant disease in glyphosate-resistant crops. Journal of Agricultural and Food Chemistry, 60(42), 10375-10397.

European Commission, 2017a. Final Review report for the active substance fluroxypyr finalised in the Standing Committee on the Food Chain and Animal Health at its meeting on 17 June 2011 in view of the approval of fluroxypyr as active substance in accordance with Regulation (EC) No 1107/2009. SANCO/11019/2011-rev.5, 23 March 2017.

European Food Safety Authority (EFSA) 2015. Conclusion on the peer review of the pesticide risk assessment of the active substance glyphosate. EFSA Journal, 13 (11), 4302.

European Food Safety Authority (EFSA), Arena, M., Auteri, D., Barmaz, S., Brancato, A., Brocca, D., Bura, L., Chiusolo, A., Court Marques, D., Crivellente, F. and De Lentdecker, C., 2018. Peer review of the pesticide risk assessment of the active substance asulam (variant evaluated asulam‐sodium). EFSA Journal, 16(4), e05251.

Evans, S.C., Shaw, E.M. and Rypstra, A.L., 2010. Exposure to a glyphosate-based herbicide affects agrobiont predatory arthropod behaviour and long-term survival. Ecotoxicology, 19(7), 1249-1257.

Everts, J.W., Aukema, B., Hengeveld, R. and Koeman, J.H., 1989. Side-effects of pesticides on ground-dwelling predatory arthropods in arable ecosystems. Environmental Pollution, 59(3), 203-225.

Gaupp-Berghausen, M., Hofer, M., Rewald, B. and Zaller, J.G., 2015. Glyphosate-based herbicides reduce the activity and reproduction of earthworms and lead to increased soil nutrient concentrations. Scientific Reports, 5(1), pp.1-9.

Giesy, J.P., Dobson, S. and Solomon, K.R., 2000. Ecotoxicological risk assessment for Roundup® herbicide. Reviews of Environmental Contamination and Toxicology, 35-120.

Gilburn, A.S., Bunnefeld, N., Wilson, J.M., Botham, M.S., Brereton, T.M., Fox, R. and Goulson, D., 2015. Are neonicotinoid insecticides driving declines of widespread butterflies?. PeerJ, 3, p.e1402.

Glyphosate renewal group (GRG), 2021. The glyphosate dossier.
 
Guilherme, S., Santos, M.A., Gaivao, I. and Pacheco, M., 2015. Genotoxicity evaluation of the herbicide Garlon® and its active ingredient (triclopyr) in fish (Anguilla anguilla L.) using the comet assay. Environmental Toxicology, 30(9), 1073-1081.

Gunstone, T., Cornelisse, T., Klein, K., Dubey, A. and Donley, N., 2021. Pesticides and Soil Invertebrates: A Hazard Assessment. Frontiers in Environmental Science, 9, 122.

Hammerschmidt, R., 2018. How glyphosate affects plant disease development: it is more than enhanced susceptibility. Pest Management Science, 74(5), 1054-1063.

Haughton, A.J., Bell, J.R., Boatman, N.D. and Wilcox, A., 2001a. The effect of the herbicide glyphosate on non‐target spiders: Part II. Indirect effects on Lepthyphantes tenuis in field margins. Pest Management Science: formerly Pesticide Science, 57(11), 1037-1042.

Haughton, A.J., Bell, J.R., Wilcox, A. and Boatman, N.D., 2001b. The effect of the herbicide glyphosate on non‐target spiders: Part I. Direct effects on Lepthyphantes tenuis under laboratory conditions. Pest Management Science: formerly Pesticide Science, 57(11), 1033-1036.

Helander, M., Pauna, A., Saikkonen, K. and Saloniemi, I., 2019. Glyphosate residues in soil affect crop plant germination and growth. Scientific Reports, 9(1), 1-9.

Helander, M., Saloniemi, I. and Saikkonen, K., 2012. Glyphosate in northern ecosystems. Trends in Plant Science, 17(10), 569-574.

HSE, 2020. Emergency Authorisation Of A Plant Protection Product.

HSE, 2021. Code of Practice for Using Plant Protection Products.

Isbister, K.M., Lamb, E.G. and Stewart, K.J., 2017. Herbicide toxicity testing with non-target boreal plants: The sensitivity of Achillea millefolium L. and Chamerion angustifolium L. to triclopyr and imazapyr. Environmental Management, 60(1), 136-156.

Kanissery, R., Gairhe, B., Kadyampakeni, D., Batuman, O. and Alferez, F., 2019. Glyphosate: Its environmental persistence and impact on crop health and nutrition. Plants, 8(11), 499.

Karlsten, A., 2017. Effects of triclopyr on wood ants (family Formica). Theses Umeå University, Faculty of Science and Technology, Department of Ecology and Environmental Sciences.

Kaufmann, A. and Kaenzig, A., 2004. Contamination of honey by the herbicide asulam and its antibacterial active metabolite sulfanilamide. Food Additives and Contaminants, 21(6), 564-571.

Kondera, E., Teodorczuk, B., Ługowska, K. and Witeska, M., 2018. Effect of glyphosate-based herbicide on hematological and hemopoietic parameters in common carp (Cyprinus carpio L). Fish Physiology and Biochemistry, 44(3), 1011-1018.

Kremer, R., Means, N. and Kim, S., 2005. Glyphosate affects soybean root exudation and rhizosphere micro-organisms. International Journal of Environmental Analytical Chemistry, 85(15), 1165-1174.

Kreutzweiser, D.P., Holmes, S.B. and Behmer, D.J., 1992. Effects of the herbicides hexazinone and triclopyr ester on aquatic insects. Ecotoxicology and Environmental Safety, 23(3), 364-374.

Laitinen, P., Rämö, S. and Siimes, K., 2007. Glyphosate translocation from plants to soil–does this constitute a significant proportion of residues in soil? Plant and Soil, 300(1), 51-60.

Lewis, K.A., Tzilivakis, J., Warner, D.J. and Green, A., 2016. An international database for pesticide risk assessments and management. Human and Ecological Risk Assessment: An International Journal, 22(4), 1050-1064.

Lipok, J., Studnik, H. and Gruyaert, S., 2010. The toxicity of Roundup® 360 SL formulation and its main constituents: glyphosate and isopropylamine towards non-target water photoautotrophs. Ecotoxicology and Environmental Safety, 73(7), 1681-1688.

Londo, J.P., McKinney, J., Schwartz, M., Bollman, M., Sagers, C. and Watrud, L., 2014. Sub-lethal glyphosate exposure alters flowering phenology and causes transient male-sterility in Brassica spp. BMC Plant Biology, 14(1), 1-10.

Maderthaner, M., Weber, M., Takács, E., Mörtl, M., Leisch, F., Römbke, J., Querner, P., Walcher, R., Gruber, E., Székács, A. and Zaller, J.G., 2020. Commercial glyphosate-based herbicides effects on springtails (Collembola) differ from those of their respective active ingredients and vary with soil organic matter content. Environmental Science and Pollution Research, 1-10.

Mamy, L., Barriuso, E. and Gabrielle, B., 2016. Glyphosate fate in soils when arriving in plant residues. Chemosphere, 154, 425-433.

Man, R., Rice, J.A., & MacDonald, G.B., 2013. Performance of planted spruce and natural regeneration after pre-and post-harvest spraying with glyphosate and partial cutting on an Ontario (Canada) boreal mixed wood site. Forestry, 86(4), 475-480.

Man, R., Rice, J.A., Freeman, L. and Stuart, S., 2011. Effects of pre-and post-harvest spray with glyphosate and partial cutting on growth and quality of aspen regeneration in a boreal mixedwood forest. Forest Ecology and Management, 262(7), 1298-1304.

Mann, R.M., Hyne, R.V., Choung, C.B. and Wilson, S.P., 2009. Amphibians and agricultural chemicals: review of the risks in a complex environment. Environmental Pollution, 157(11), 2903-2927.

Marileo, L.G., Jorquera, M.A., Hernández, M., Briceño, G., Mora, M.D.L.L., Demanet, R. and Palma, G., 2016. Changes in bacterial communities by post-emergent herbicides in an Andisol fertilized with urea as revealed by DGGE. Applied Soil Ecology, 101, 141-151.

Marrs, R.H. and Frost, A.J., 1997. A microcosm approach to the detection of the effects of herbicide spray drift in plant communities. Journal of Environmental Management, 50(4), 369-388.

Marrs, R.H., Frost, A.J. and Plant, R.A., 1991. Effects of herbicide spray drift on selected species of nature conservation interest: The effects of plant age and surrounding vegetation structure. Environmental Pollution, 69(2-3), 223-235.

Marrs, R.H., Williams, C.T., Frost, A.J. and Plant, R.A., 1989. Assessment of the effects of herbicide spray drift on a range of plant species of conservation interest. Environmental Pollution, 59(1), 71-86.

Marsh, J.A.P., 1980. Effects of asulam on some microbial activities of three soils. Bulletin of Environmental Contamination and Toxicology, 25, 15–22.

Mateos-Naranjo, E. and Perez-Martin, A., 2013. Effects of sub-lethal glyphosate concentrations on growth and photosynthetic performance of non-target species Bolboschoenus maritimus. Chemosphere, 93 (10), 2631-2638.

Matozzo, V., Fabrello, J. and Marin, M.G., 2020. The effects of glyphosate and its commercial formulations to marine invertebrates: A review. Journal of Marine Science and Engineering, 8(6), 399.

McKenzie, A.J. and Whittingham, M.J., 2009. Why are birds more abundant on organic farms. Journal of Food, Agriculture & Environment, 7(2), 807-814.

McMullin, R.T., Bell, F.W. and Newmaster, S.G., 2012. The effects of triclopyr and glyphosate on lichens. Forest Ecology and Management, 264, 90-97.

Michalková, V. and Pekár, S., 2009. How glyphosate altered the behaviour of agrobiont spiders (Araneae: Lycosidae) and beetles (Coleoptera: Carabidae). Biological Control, 51(3), 444-449.

Moran, P.P.W., 1999. Susceptibility of riparian soil invertebrates to the herbicide triclopyr. University of Oregon.

Motta, E.V., Raymann, K. and Moran, N.A., 2018. Glyphosate perturbs the gut microbiota of honey bees. Proceedings of the National Academy of Sciences, 115(41), 10305-10310.

Mudge, J.F. and Houlahan, J.E., 2019. Wetland macrophyte community response to and recovery from direct application of glyphosate-based herbicides. Ecotoxicology and Environmental Safety, 183, 109475.

Newton, M., Horner, L.M., Cowell, J.E., White, D.E. and Cole, E.C., 1994. Dissipation of glyphosate and aminomethylphosphonic acid in North American forests. Journal of Agricultural and Food Chemistry, 42(8), 1795-1802.

Nguyen, N.K., Dörfler, U., Welzl, G., Munch, J.C., Schroll, R. and Suhadolc, M., 2018. Large variation in glyphosate mineralization in 21 different agricultural soils explained by soil properties. Science of the Total Environment, 627, 544-552.

Nielsen, L.W. and Dahllöf, I., 2007. Direct and indirect effects of the herbicides Glyphosate, Bentazone and MCPA on eelgrass (Zostera marina). Aquatic Toxicology, 82(1), 47-54.

Nuutinen, V., Hagner, M., Jalli, H., Jauhiainen, L., Rämö, S., Sarikka, I. and Uusi-Kämppä, J., 2020. Glyphosate spraying and earthworm Lumbricus terrestris L. activity: Evaluating short-term impact in a glasshouse experiment simulating cereal post-harvest. European Journal of Soil Biology, 96, 103148.

Petersen, I.L., Hansen, H.C.B., Ravn, H.W., Sørensen, J.C. and Sørensen, H., 2007. Metabolic effects in rapeseed (Brassica napus L.) seedlings after root exposure to glyphosate. Pesticide Biochemistry and Physiology, 89(3), 220-229.

Rose, M.T., Cavagnaro, T.R., Scanlan, C.A., Rose, T.J., Vancov, T., Kimber, S., Kennedy, I.R., Kookana, R.S. and Van Zwieten, L., 2016. Impact of herbicides on soil biology and function. Advances in Agronomy, 136, 133-220.

Rowntree, J.K. and Sheffield, E., 2005. Effects of asulam spraying on non-target ferns. Botany, 83(12), 1622-1629.

Rowntree, J.K., Lawton, K.F., Rumsey, F.J. and Sheffield, E., 2003. Exposure to Asulox inhibits the growth of mosses. Annals of Botany, 92(4), 547-556.

Roy, D.N., Konar, S.K., Banerjee, S., Charles, D.A., Thompson, D.G. and Prasad, R., 1989. Persistence, movement, and degradation of glyphosate in selected Canadian boreal forest soils. Journal of Agricultural and Food Chemistry, 37(2), 437-440.

Sharma, A., Jha, P. and Reddy, G.V., 2018. Multidimensional relationships of herbicides with insect-crop food webs. Science of the Total Environment, 643, 1522-1532.

Sikorski, Ł., Baciak, M., Bęś, A. and Adomas, B., 2019. The effects of glyphosate-based herbicide formulations on Lemna minor, a non-target species. Aquatic Toxicology, 209, 70-80.

Souza-Alonso, P., Guisande, A. and González, L., 2015. Structural changes in soil communities after triclopyr application in soils invaded by Acacia dealbata Link. Journal of Environmental Science and Health, Part B, 50(3), 184-189.

Speight, R., Whittaker, J., 1987. Interactions between the chrysomelid beetle Gastrophysa viridula, the weed Rumex obtusifolius and the herbicide asulam. Applied Ecology, 24, 119–129.

Strandberg, B., Boutin, C., Carpenter, C., Mathiassen, S.K., Damgaard, C.F., Sorensen, P.B., Bruus, M., Dupont, Y.L., Bossi, R. and Andersen, D.K., 2019. Pesticide effects on non‐target terrestrial plants at individual, population and ecosystem level (PENTA). Copenhagen (DK): Ministry of Environment and Food of Denmark.

Sujetovienė, G., Gasauskaitė, K. and Žaltauskaitė, J., 2019. Toxicity of a phenoxy herbicide on the lichen Ramalina fraxinea. Toxicological & Environmental Chemistry, 101(9-10), 497-507.

Tajnaiová, L., Vurm, R., Kholomyeva, M., Kobera, M. and Kočí, V., 2020. Determination of the Ecotoxicity of Herbicides Roundup® Classic Pro and Garlon New in Aquatic and Terrestrial Environments. Plants, 9(9), 1203.

Thompson, H.M., Levine, S.L., Doering, J., Norman, S., Manson, P., Sutton, P. and von Mérey, G., 2014. Evaluating exposure and potential effects on honeybee brood (Apis mellifera) development using glyphosate as an example. Integrated Environmental Assessment and Management, 10(3), 463-470.

Tran, H., Harrington, K.C., Robertson, A.W. and Watt, M.S., 2015. Relative persistence of commonly used forestry herbicides for preventing the establishment of broom (Cytisus scoparius) seedlings in New Zealand plantations. New Zealand Journal of Forestry Science, 45(1), 1-13.

Tuffi Santos, L.D., Santos, J.B., Ferreira, F.A., Oliveira, J.A., Bentivenha, S. and Machado, A.F.L., 2008. Radicular exudation of glyphosate by Brachiaria decumbens and its effects on eucalypt plant. Planta Daninha, 26, 369-374.

van Wijngaarden, R.P., Cuppen, J.G., Arts, G.H., Crum, S.J., van den Hoorn, M.W., van den Brink, P.J. and Brock, T.C., 2004. Aquatic risk assessment of a realistic exposure to pesticides used in bulb crops: a microcosm study. Environmental Toxicology and Chemistry: An International Journal, 23(6), 1479-1498.

Vannini, A., Guarnieri, M., Bačkor, M., Bilová, I. and Loppi, S., 2015. Uptake and toxicity of glyphosate in the lichen Xanthoria parietina (L.) Th. Fr. Ecotoxicology and Environmental Safety, 122, 193-197.

Vannini, A., Guarnieri, M., Paoli, L., Sorbo, S., Basile, A. and Loppi, S., 2016. Bioaccumulation, physiological and ultrastructural effects of glyphosate in the lichen Xanthoria parietina (L.) Th. Fr. Chemosphere, 164, 233-240.

Vasic, V., Konstantinovic, B., Jarak, M. and Orlovic, S., 2009. Herbicide weed control in pedunculate oak (Quercus robur) forest regeneration. Forest Vegetation Management–Towards Environmental Sustainability, (35-2009), 14.

Vázquez, D.E., Balbuena, M.S., Chaves, F., Gora, J., Menzel, R. and Farina, W.M., 2020. Sleep in honey bees is affected by the herbicide glyphosate. Scientific Reports, 10(1), 1-8.

Vázquez, D.E., Ilina, N., Pagano, E.A., Zavala, J.A. and Farina, W.M., 2018. Glyphosate affects the larval development of honey bees depending on the susceptibility of colonies. PloS One, 13(10), e0205074.

Viti, M. L., Alves, P. A. T., Mendes, K. F., Pimpinato, R. F., Guimarães, A. C. D., & Tornisielo, V. L. 2019. Translocation and root exudation of glyphosate by Urochloa brizantha and its transport on sugarcane and citrus seedlings. Planta Daninha, 37.

Voinorosky, C., 2020. The Effects of Targeted Triclopyr Application on Habitat Quality in Boreal Saskatchewan Transmission Rights-of-Way (Doctoral dissertation, University of Saskatchewan).

von Mérey, G., Manson, P.S., Mehrsheikh, A., Sutton, P. and Levine, S.L., 2016. Glyphosate and aminomethylphosphonic acid chronic risk assessment for soil biota. Environmental Toxicology and Chemistry, 35(11), 2742-2752.

Wagner, N., Reichenbecher, W., Teichmann, H., Tappeser, B. and Lötters, S., 2013. Questions concerning the potential impact of glyphosate‐based herbicides on amphibians. Environmental Toxicology and Chemistry, 32(8), 1688-1700.

Wagner, N., Veith, M., Lötters, S. and Viertel, B., 2017. Population and life‐stage–specific effects of two herbicide formulations on the aquatic development of European common frogs (Rana temporaria). Environmental Toxicology and Chemistry, 36(1), 190-200.

Wang, C., Lin, X., Li, L. and Lin, S., 2016. Differential growth responses of marine phytoplankton to herbicide glyphosate. PloS One, 11(3), p.e0151633.

Weaver, M.A., Krutz, L.J., Zablotowicz, R.M. and Reddy, K.N., 2007. Effects of glyphosate on soil microbial communities and its mineralization in a Mississippi soil. Pest Management Science: formerly Pesticide Science, 63(4), 388-393.

Whitehouse, D.M. and Brown, V.K., 1993. Herbicides in farm forestry: effects on non-target insects (1), 121-126.

Wood, L.J., 2019. The presence of glyphosate in forest plants with different life strategies one year after application. Canadian Journal of Forest Research, 49(6), 586-594.

Yahnke, A.E., Grue, C.E., Hayes, M.P. and Pearman‐Gillman, S., 2017. Effects of the herbicide triclopyr on metamorphic northern red‐legged frogs. Environmental Toxicology and Chemistry, 36(9), 2316-2326.

Zavanella, T., Zaffaroni, N.P. and Arias, E., 1988. Evaluation of the carcinogenic risk of the phenoxyherbicide MCPA to an urodele amphibian. Ecotoxicology and Environmental Safety, 16(2), 114-122.

Zhang, S., Qiu, C.B., Zhou, Y., Jin, Z.P. and Yang, H., 2011. Bioaccumulation and degradation of pesticide fluroxypyr are associated with toxic tolerance in green alga Chlamydomonas reinhardtii. Ecotoxicology, 20(2), 337-347.

Appendix  - Glossary and Metrics

Glossary

Term	-	Description
Acid equivalent	a.e	Active substance weight includes the salt part of the molecule, while acid equivalent weight does not.
Active ingredient	a.i.	Alternative term for active substance
Active substance	a.s.	Any substance or micro-organism, including a virus, that has a general or specific action: against harmful organisms; or on plants, parts of plants or plant products. Active substances are usually formulated with other materials in a pesticide product.
Adjuvant	-	A substance other than water, without significant pesticidal properties, which enhances or is intended to enhance the effectiveness of a pesticide product (when it has been added to the spray tank with the pesticide product).
EC50	EC50	EC means “effective concentration”. EC values refer to the concentration of a chemical that  the concentration of a substance that inhibits 50% of the growth of algae.
ED50	ED50	ED means “effective dose” ED50 refers to the dose that produces a certain effect in 50 % of test organisms
ED10	ED10	ED10 refers to the dose that produces a certain effect in 10 % of test organisms
Hectare	ha	10,000m2 or 2.47 acres
LC50	LC50	LC means "Lethal Concentration". LC values refer to the concentration of a chemical in water that kills 50% of the test animals during the observation period.
LD50	LD50	LD means “Lethal Dose". LD50 is the amount of a material, given all at once, which causes the death of 50% (one half) of a group of test animals. The LD50 is one way to measure the short-term poisoning potential (acute toxicity) of a material.
Pesticide	-	something that prevents, destroys, or controls a harmful organism ('pest') or disease, or protects plants or plant products during production, storage and transport.
Pesticide label	-	The label is the main source of information on the safe and effective use of a product and contains information on: trade name and marketing company of the product; MAPP (registration) number; product's active ingredients (and their concentrations); any necessary hazard symbols; precautions to be taken during use; medical advice; safe disposal of the product; advice on what pests (including diseases and weeds) the product controls. Field of use restrictions; User restrictions; The crop or situation which may be treated; Maximum individual dose/application rate; Maximum number of treatments or maximum total dose; Maximum area or quantity which may be treated; Latest time of application, harvest or re-entry interval; Operator protection or training requirements; Environmental protection requirements.
w/w	-	The weight concentration of the pesticide is expressed as weight for weight.

 

Metrics

Weight conversion table

ppm	g/L	mg/L	µg/L	ng/L
1	0.001	1	1000	1,000,000
1000	1	1000	1,000,000	1e+9