Protocols for the development of the guidance on the risk assessment of PPP to non‐target arthropods, in‐soil organisms and non‐target terrestrial plants

Abstract

The present document defines the problem formulation, evidence needs and methodological approach agreed by the Environmental Risk Assessment Working Group and the EFSA Panel on Plant Protection Products and their Residues for the Tier 1 risk assessment of the terrestrial ecotoxicology guidance document. This revision falls within the remit of mandate M‐2024‐00086, which specifically requests development of three guidance documents: one on the risk assessment for non‐target arthropods other than bees, one for in‐soil organisms and one for non‐target terrestrial plants.

SUMMARY

The current document reports on the problem formulation, evidence needs and methodological approach specified by the Environmental Risk Assessment Working Group (ERA Working Group) and the EFSA Panel on Plant Protection Products and their Residues (PPR Panel) for the revision of the terrestrial ecotoxicology guidance document (European Commission, 2002). This revision falls within the remit of mandate M‐2024‐00086, specifically requesting development of guidance on risk assessment for non‐target arthropods (NTAs) other than bees, in‐soil organisms and non‐target terrestrial plants (NTTPs).

The protocol follows the principles outlined in the EFSA Guidance on protocol development for EFSA generic scientific assessments. Although developing a protocol is not mandatory for the revision of guidance documents, the ERA Working Group and the PPR Panel considered it useful for such a complex mandate. The protocol will focus on Tier 1 risk assessment.

For each of the three guidance documents, the terms of reference of the mandate have been translated into an assessment question and several sub‐questions related to exposure characterisation, hazard characterisation and risk assessment.

For each sub‐question, based on the relative priority identified by the Working Group and the sensitivity of the different aspects considered, the methodological approach to gather additional data was identified.

1. INTRODUCTION

The current document reports the problem formulation, evidence needs and approach specified by the Environmental Risk Assessment Working Group (ERA Working Group) and the EFSA Panel on Plant Protection Products and their Residues (PPR Panel) for the revision of the terrestrial ecotoxicology Guidance Document (European Commission, 2002). This revision falls within the remit of mandate M‐2024‐00086, specifically requesting development of guidance on risk assessment for non‐target arthropods (NTAs) other than bees, in‐soil organisms and non‐target terrestrial plants (NTTPs) (EFSA, 2025a).

The protocol follows the principles outlined in the EFSA Guidance on protocol development for EFSA generic scientific assessments (EFSA Scientific Committee, 2023). This latter guidance acknowledges that protocols must be adapted to the mandate requirements to ensure the delivery of fit‐for‐purpose and efficient scientific advice. Such adaptation involves tailoring both the level of detail in the protocol and the approach to publishing and disseminating it for each specific mandate.

Although developing a protocol is not mandatory for the revision of guidance documents, the ERA Working Group and the PPR Panel opted to develop one for the Tier 1 risk assessment of the three guidance documents. This focus, however, does not imply that revision and/or inclusion of higher tier risk assessment options will be excluded from the revision process; their update will be considered and addressed in the revised guidance documents if deemed necessary and relevant, taking into account scientific advancements, regulatory requirements and data availability.

It is also further emphasised that the planning of the revision of the three guidance documents depends on ongoing projects, including AENEAS, ¹ EESE, ² PERA ³ and other related initiatives. Moreover, regarding the hazard identification and characterisation, recommendations of the roadmap of the European Commission ⁴ towards phasing out animal testing will be considered to the extent possible.

The protocol has been endorsed by the PPR Panel as a draft protocol, shared for public consultation and finalised considered the comments received. Deviations from the planned method may occur during the development of the guidance documents and will be documented accordingly.

1.1. Terms of Reference as provided by the requestor

The Terms of Reference for mandate M‐2024‐0086 Terrestrial Ecotoxicology Guidance are the following:

‘EFSA is requested to review and/or update the risk assessment methodology, for the assessment of active substances and plant protection products falling under Part A of the Annexes to Regulations (EU) No 283/2013 and No 284/2013, respectively, as regards the potential effects on non‐target arthropods other than bees, on in‐soil organisms (non‐target soil meso‐ and macrofauna and effects on soil nitrogen transformation) and on non‐target terrestrial plants outside the treatment area’.

Separate mandate to develop guidance for assessing potential indirect effects on biodiversity through trophic interactions under agro‐environmental conditions. A dedicated output is in preparation. It is well recognised that (EFSA, 2025b), EFSA suggested clearly considering elements accounting for indirect effects when proposing SPGs to risk managers.

1.2. How to navigate this protocol

This protocol consists of (i) an introduction explaining the general process and the terms of reference of the mandate, (ii) cross‐cutting sections applicable to the review process of all three guidance documents, clarifying terminology, definitions and explaining the principles of the protocol development and (iii) three guidance‐specific sections (see Figure 1).

FIGURE 1.

Visualisation of the main parts of the document explaining how to better navigate within the document.

As part of the cross‐cutting sections, the problem formulation was defined. This was done by articulating the term of reference of the mandate into an assessment question and related sub‐questions. A gap analysis was also used as a method for further defining the sub‐questions. Once gaps were identified, they were prioritised and the methods for gathering data identified.

The specific sections (5, 6, 7) make reference to the assessment question and sub‐questions, as well as to the problem formulation, by detailing for each group of organisms the gap analysis, the priority assigned to each gap and the method selected for filling that gap based on the assigned relative priority.

Conceptual models were developed for each organism group to visualise the sequence from PPP emission to organism‐level exposure. These conceptual models are presented in Sections 5, 6 and 7, respectively.

2. BACKGROUND, ANALYSIS OF PREVIOUS EFSA SCIENTIFIC OPINIONS ON TERRESTRIAL ORGANISMS AND NEW DEFINITIONS

2.1. Biological and regulatory basis for separate protocols across terrestrial non‐target organisms in the remit of the mandate

The three guidance documents within the remit of the above‐mentioned mandate are inevitably interrelated. Traditionally, the risk assessment for NTAs has focused on arthropods living above‐ground, either on soil surface (ground‐ or surface‐dwelling arthropods) or on plants (plant‐ or foliage‐dwelling arthropods), while the risk assessment for in‐soil organisms has covered those organisms living below ground, in the soil or the soil litter layer. The NTTP risk assessment focuses on non‐crop plants living outside of the field.

From a biological standpoint, the division between above‐ground NTAs and in‐soil organisms is somewhat arbitrary, as some NTAs spend part of their life cycle in the soil and part above ground (e.g. many coleopteran species). Moreover, certain invertebrates (e.g. isopods, chilopods, diplopods) that live above ground are not routinely considered in the NTA risk assessment at Tier 1. However, it is logical to separate the primary producers (plants, here referred to as NTTPs) from NTAs and soil organisms, especially considering that NTTPs need to be assessed in the off‐field area only, according to the mandate.

The risk assessment of PPPs is geared towards encompassing the different exposure routes and the non‐target organisms potentially exposed. This is why the regulatory framework tends to simplify the large complexity of the natural system by creating this arbitrary separation, since addressing the risk of PPPs to all the above‐mentioned organisms within a single guidance document would be extremely complex. However, while the overlap between guidance documents should be acknowledged, as some routes of exposure may be relevant across organism groups, others may be specific to certain non‐target groups. By this compartmentalisation, the ERA Working Group aims to design risk assessment schemes which ensure that SPGs are properly addressed as well as feasible to implement.

The risk assessment approach, particularly its tiered structure, is designed to accommodate this complexity. Lower tier assessments are intentionally simple and conservative, enabling early identification and screening of substances unlikely to pose a risk. As tiers progress, more realism is introduced to reflect the true complexity of exposure and ecological interactions. Therefore, while this clear separation of organism groups is maintained at the lower tier of the risk assessment, pending on the feasibility and the needs, the WG could still reconsider this approach in the revised guidance documents for higher tier studies.

While the protocol and future related guidance documents will still focus on animal testing for hazard identification, they will also integrate new approach methodologies (NAMs) should those be considered ready to use. To this respect, one of the projects under PERA FPA EUBA/EFSA/PREV/2023/01 aims at collecting NAMs for the terrestrial compartments and categorising their level of readiness. It is acknowledged that NTAs, in‐soil organisms and NTTPs might be exposed spatially and temporally to multiple residues (e.g. mix of insecticides, fungicides and herbicides) in agricultural areas. However, the protocol and future guidance documents do not address the risk assessment of combinations of more than one PPP of unknown composition or when PPPs are applied sequentially within one season. In this respect, the Guidance documents are developed under the remit of the Regulation 1107/2009, in which only single product/a.s. assessments are implemented. In line with the approach followed in the guidance on the risk assessment for bees (EFSA, 2023a) and Birds and Mammals (EFSA, 2023b), whenever it is necessary to assess effects of mixtures according to Regulation (EC) No 1107/2009 (i.e. in case of formulated products containing multiple active substances, synergists/safeners/co‐formulants of particular concern, racemic mixtures and mixtures of active substances with metabolites of concern) the principles of the EFSA guidance on mixtures (EFSA Scientific Committee, 2019) can be used.

2.2. Cross‐cutting considerations in exposure assessment across the three protocols

As part of the problem formulation for the revision of the terrestrial ecotoxicology guidance for the three organism groups (NTAs, in‐soil organisms and NTTPs), several commonalities were identified in how exposure is conceptualised and assessed across them. Although the three groups occupy different ecological niches and interact with the environment through specific behaviours, they share several exposure‐relevant environmental matrices (e.g. soil, pore water, plant surfaces, nectar, pollen, air and plant litter) which act as key interfaces for exposure.

Similarly, they also share exposure routes, such as contact via natural substrate (plant parts or soil) and dietary uptake.

In the current document, a structured effort was made to align terminology related to sources of emission, exposure pathways, exposure‐relevant environmental matrices and exposure routes, with the goal of promoting consistency and improving comparability across protocols and guidance documents. This effort builds upon the content of the respective EFSA Scientific Opinions (EFSA PPR Panel, 2014, 2015, 2017) and existing guidance documents on exposure (e.g. EFSA, 2017), which served as the starting point for the development of the exposure assessment framework in each protocol. However, these documents differed in how exposure is framed and described. The harmonised approach described in this document aims to ensure a coherent risk assessment while still allowing flexibility to reflect group‐specific ecological traits and exposure dynamics.

The current protocol focuses on open‐field uses. Applications under protected structures, such as greenhouses, walk‐in tunnels and open protective structures, are not specifically addressed in this protocol. The relevant EFSA guidance might be considered for the evaluation of those uses (EFSA, 2014).

2.3. Comparison of the exposure assessment across the EFSA Scientific Opinions on terrestrial organisms

The three EFSA Scientific Opinions on NTAs (EFSA PPR Panel, 2015), in‐soil organisms (EFSA PPR Panel, 2017) and NTTPs (EFSA PPR Panel, 2014) provide foundational considerations for exposure assessment, though they vary in structure, scope and terminology. All three documents acknowledge the importance of identifying sources of emission, relevant environmental compartments and exposure‐relevant matrices, as well as the routes through which organisms become exposed. Despite this shared conceptual basis, the level of detail and harmonisation across the Scientific Opinions remains limited.

Among them, the in‐soil organisms Scientific Opinion offers the most structured and quantitatively model‐driven approach. As the most recent of the three Opinions, it integrates methodological developments that were underway at the time, most notably the tiered framework outlined in the EFSA Guidance Document on the assessment of exposure of soil organisms (EFSA, 2017). Although this guidance document is now available, it has not yet been formally implemented in the EU regulatory context. The tiered exposure scheme presented in the Opinion mirrors the later guidance and includes standardised scenario‐based PEC (Predicted Environmental Concentration) calculations supported by tools such as the simple analytical model PERSAM, and the FOCUS models PEARL and PELMO. The in‐soil organisms Opinion defines clear exposure routes for soil‐dwelling organisms, primarily contact and dietary (oral uptake), and identifies the environmental compartments relevant for exposure as total soil, pore water, plant litter and soil organic matter. These compartments are described in connection with how different soil organisms interact with their surroundings (e.g. ingestion of organic matter and microorganisms, contact with pore water). Although the primary focus is on in‐field exposure, the Opinion also includes detailed considerations for off‐field exposure pathways, such as spray drift deposition, vapour drift, dust drift and run‐off. The proposed use of 90th percentile PECs considering all agricultural fields within a defined geographical area, such as a regulatory zone, is described in the in‐soil organisms Scientific Opinion and related exposure guidance as part of a structured, scenario‐based tiered exposure assessment.

The EFSA PPR Panel (2015) Scientific Opinion on NTAs also addresses both in‐field and off‐field exposure, providing a detailed ecological consideration of arthropod functional groups (e.g. pollinators, predators, herbivores) and life stages. It identifies four distinct spatial ‘exposure situations’ that reflect where and how NTAs may come into contact with pesticide residues: (i) in‐field on crop, (ii) in‐field on soil, (iii) off‐field on vegetation and (iv) off‐field on soil. These situations account for differences in residue deposition and organism behaviour and are used to guide the conceptualisation of exposure. The Scientific Opinion describes multiple contributors to exposure, including sources of emission and exposure pathways. Both contact and dietary exposure routes are recognised as relevant, particularly via residues on soil surface, plant surfaces, pollen, nectar and prey. The term ‘exposure situations’ used in the EFSA PPR Panel (2015) Opinion serves as an ecologically grounded concept for describing where and how NTAs may be exposed to pesticide residues. While the link to the broader ‘exposure scenarios’ introduced later in the document (see Section 2.4) is not explicitly defined, the protocol effectively builds on these situations when structuring exposure scenarios for risk assessment for NTAs (see Section 5.2.1.3).

Although it does not propose an operational exposure model, EFSA PPR Panel (2015) presents a set of quantitative theoretical considerations to support future modelling efforts. These include estimating the proportion of the applied dose intercepted by the crop canopy, dissipation of residues via volatilisation, degradation or penetration into plant tissues, and the redistribution of residues due to plant growth, wash‐off or canopy expansion. These processes are discussed as key modifiers of residue availability over time and space, particularly for mobile species that interact with multiple compartments throughout their life cycle. Such considerations provide a solid conceptual basis for the development of more refined exposure models for NTAs.

In contrast, the NTTP Scientific Opinion is conceptually broad, but less developed in terms of exposure assessment methodology. It includes both in‐field and off‐field NTTPs within its scope of protection, acknowledging that plants growing within treated fields, such as rare arable weeds, may contribute to ecosystem functions, particularly by supporting higher trophic levels. However, the exposure assessment is primarily structured for off‐field scenarios, where a range of emission and transport processes are considered. The Opinion outlines general principles for calculating exposure via droplet and vapour drift deposition, particulate drift and run‐off, and refers to established methodologies for these purposes. However, it does not provide an integrated exposure model or a tiered framework, instead relies on simplified, route‐specific estimates. As a result, the exposure assessment remains more descriptive and less operational than the model‐based or theoretically structured approaches presented for in‐soil organisms and NTAs.

Harmonisation of the exposure assessment framework across the three organism groups is essential to ensure consistency, scientific coherence and regulatory transparency in the development of the revised terrestrial guidance documents. Although each organism group presents specific ecological traits and exposure dynamics, the underlying structure of the protocols and related guidance documents relies on a common organisation of problem formulation, exposure characterisation, hazard assessment and risk assessment elements. Developing visual conceptual models using a harmonised terminology (e.g. distinguishing source of emission, exposure pathway, exposure‐relevant environmental matrix and exposure route) facilitates comparability across protocols, avoids duplication and supports a consistent narrative across sections. Moreover, aligning the conceptual framing of exposure allows the models to be integrated more effectively into the guidance document structure, ensuring that cross‐referencing and methodological parallels between protocols are scientifically justified and clearly communicated to risk assessors and stakeholders.

Table 1 provides a set of key commonalities and differences in exposure assessment framework, highlighting where the Opinions converge and where further harmonisation is needed.

TABLE 1.

Summary of key commonalities and differences across the three EFSA PPR Panel Scientific Opinions on terrestrial organisms.

Key commonalities

All Scientific Opinions acknowledge, explicitly or implicitly, the need for a tiered assessment approach.

Exposure is described as resulting from environmental transport processes following emission, including spray drift, volatilisation and run‐off, which influence the distribution and accessibility of residues to non‐target organisms.

Spray drift and deposition are recognised as important exposure pathways.

All Scientific Opinions address multiple exposure routes (e.g. contact, foliar and root uptake, dietary).

Exposure assessment goal: the 90th percentile Predicted Environmental Concentration (PEC), representing a spatial percentile across treated fields within a regulatory zone, is mentioned as a preferred regulatory target.

Key differences

Only the in‐soil organisms Scientific Opinion provides a formally structured, tiered framework with operational modelling tools aligned to EFSA (2017) soil exposure guidance.

The non‐target arthropod Scientific Opinion includes the most ecologically detailed treatment of exposure across spatial settings and life stages. Although it lacks an operational modelling tool, it provides theoretical considerations to support quantitative exposure estimation.

Exposure‐relevant environmental matrices are systematically defined only in the in‐soil and the non‐target arthropod Scientific Opinions.

2.4. Harmonisation of exposure terminology across the three EFSA PPR Panel Scientific Opinions

While each of the three EFSA Scientific Opinions describes relevant processes and interactions between PPPs and non‐target organisms, the terminology used is often inconsistent or implicit. To ensure a harmonised approach across the three protocols, four core concepts were introduced – source of emission, exposure pathway, exposure‐relevant environmental matrix and exposure route – and defined in a way that promotes clarity, comparability and the development of shared conceptual models.

In line with this effort, a common definition of exposure scenario is also proposed to consistently frame how exposure is conceptualised and quantified in the context of environmental risk assessment for non‐target organisms. An exposure scenario is a conceptual and quantitative representation of the environmental context and organism‐specific ecological traits relevant to the estimation of exposure to PPP residues. It comprises a coherent set of agronomic and environmental conditions, such as soil type, climate, crop/plant presence and landscape structure (e.g. in‐field vs. off‐field) that influence the distribution and persistence of residues in the environment. The scenario also reflects the ecological traits of the non‐target organisms, including habitat preferences and behavioural patterns, that affect their interaction with contaminated environmental matrices. Within the risk assessment process, exposure scenarios serve to structure the estimation of ecotoxicologically relevant exposure concentrations in environmental matrices and support the quantification of organism exposure via relevant routes of exposure.

In the revision of the terrestrial guidance documents, the aim is to build on EFSA's existing scenario‐based approach, as exemplified in the revised guidance on the risk assessment on bees (EFSA, 2023a). Although that guidance does not always provide explicit definitions, it effectively employs exposure scenarios structured around representative spatial and functional compartments of the agricultural landscape where exposure may occur. To ensure clarity and consistency, scenarios are grouped into in‐field and off‐field contexts:

• Treated crop (in‐field): cultivated area where the PPP is applied according to the GAP.

• Weeds in the treated field (in‐field): refers to weeds present within the treated crop area, which may also receive residues directly from PPP applications.

• Succeeding crop (in‐field, ‘temporal’): refers to annual crops sown or planted after a previously treated crop on the same field.

• Field margins (off‐field): strips of non‐crop vegetation bordering the treated field.

• Adjacent crop (off‐field): an agricultural crop in a neighbouring field located directly next to the treated crop area.

The identified scenarios may nevertheless be amended and/or expanded if a need is identified at a later stage.

It should be noted that, although some approaches (e.g. Adriaanse et al., 2022) distinguish three zones (in‐crop, in‐field off‐crop and off‐field), this protocol retains two spatial contexts only: in‐field and off‐field. For feasibility and consistency with SPGs, we do not introduce a separate in‐field off‐crop strip; instead, the in‐crop evaluation is taken as protective for any in‐field off‐crop areas. This is justified because PPP deposition generally peaks within the crop (both for foliar‐ and ground‐dwelling organisms) and declines beyond the last crop row, so in‐crop exposure represents a conservative bound for in‐field off‐crop strips. See also EFSA (2025b).

The same exposure scenarios will be adopted, where applicable, across the three guidance documents to ensure that the environmental risk assessment is grounded in a consistent framework. This alignment not only promotes conceptual coherence but may also enable the reuse or adaptation of exposure models across guidance. The relevance of each scenario for the three groups of organisms, why it matters and under what conditions it may be excluded a priori, is discussed in the subsequent Sections (5.3.1.3 for NTAs, 6.2.1.3 for in‐soil organisms and 7.3.1.3 for NTTPs).

In the revised guidance documents, exposure scenarios are intended to ensure biological realism while providing a feasible risk assessment scheme and maintaining regulatory consistency across protocols.

The following sections provide explanations of each of the four components.

2.4.1. Source of emission

The source of emission refers to the starting point at which the substance is released into the environment, marking the beginning of potential environmental exposure. This typically coincides with the application method of a PPP, and determines how, where and when the active substance first enters environmental compartments.

Clearly defining the source of emission is essential for distinguishing the initial entry of the substance from subsequent environmental transport processes, and for understanding which organisms may be exposed under different spatial configurations (e.g. in‐field vs. off‐field). Importantly, a well‐defined source of emission enables a clearer delineation between emission control measures (e.g. nozzle type, boom height, treated area) and exposure modelling approaches (e.g. spray drift curves, run‐off and erosion entries). This separation enhances the robustness and transparency of the exposure assessment framework.

For the purposes of this protocol and following guidance documents, the focus remains on conventional application techniques, including conventional spraying applications, granule applications and seed treatments, which underpin the structure of standard exposure scenarios used in current regulatory risk assessment (e.g. EFSA, 2017; FOCUS models). These represent the most used application methods in current agricultural practice and are typically reflected in standard exposure scenarios (broadcast/treatment of the whole field) used in the EU regulatory risk assessment. In this context, it is noted that the EFSA Guidance Document for predicting environmental concentrations of active substances of plant protection products and transformation products of these active substances in soil (EFSA, 2017) explicitly applies also to crops grown on ridges and to non‐uniform applications in annual crops (e.g. row, band, strip or spot applications) for the exposure assessment of the in‐field scenario. Consequently, soil exposure in such situations is already covered for the assessment of both NTAs and in‐soil organisms and does not require the development of additional scenario types within the future Guidance Documents.

However, the mandate for the revision of the terrestrial guidance explicitly highlights the need to reflect recent scientific and technological developments. In this context, emerging application technologies and practices – including unmanned aerial vehicles (UAVs or drones), sensor‐based delivery systems and site‐specific application strategies such as band spraying or targeted spraying – are receiving growing attention in the field and are increasingly referenced in scientific literature and regulatory discussions. Although these technologies may offer improved targeting and reduced off‐target emissions, they also introduce new spatial and temporal dynamics that are not fully captured by current standard exposure scenarios.

Moreover, the EU Compendium of Risk Mitigation Measures (European Commission, 2024) lists several of these newer techniques, such as low‐drift nozzles and GPS‐controlled banded applications, acknowledging their potential to reduce environmental emissions and therefore alter exposure pathways and magnitudes. The consideration of the Compendium is explicitly foreseen in the Terms of Reference of the current mandate and may inform the development of revised scenarios and model parameters.

In parallel, the European Precision Application Task Force (EUPAF) is currently supporting these developments by coordinating a multi‐stakeholder effort to categorise and describe precision use scenarios and propose consistent descriptors for inclusion in GAP tables. EUPAF's work is expected to inform the design of tailored risk assessments, particularly regarding exposure estimation for non‐target organisms.

Moreover, information from a systematic literature search and experimental data from field studies under the PERA FPA EUBA/EFSA/PREV/2023/01 may further support these advancements by generating refined drift deposition values and vegetation interception data applicable to both conventional and innovative spraying systems, depending on the scope and outcomes of the planned experimental activities.

Taken together, while this protocol maintains its operational focus on conventional application methods, it not only acknowledges but is also designed to integrate relevant updates emerging from recent and ongoing developments. In particular, when results from the above‐mentioned activities – including those under the PERA FPA EUBA/EFSA/PREV/2023/01 and the work of the European Precision Application Task Force (EUPAF) – become available during the revision process, they will be incorporated into the exposure assessment framework where appropriate. This will help ensure that exposure scenarios remain scientifically robust and reflective of both current and evolving agricultural practices, in line with the objectives of the mandate.

Application methods that are not frequently used, e.g. soil injection, soil fumigation, are not specifically addressed in this protocol, since no additional data will be gathered. Nevertheless, those types of applications will be further considered in the future revision of the three guidance documents, i.e. NTAs, in‐soil organisms and NTTPs.

2.4.2. Exposure pathway

In the context of environmental risk assessment for PPPs, the exposure pathway refers to the sequence of the environmental transport processes that describe how a substance moves from its source of emission to exposure‐relevant environmental matrices or directly to the non‐target organism itself, for example through overspray. It encompasses all relevant mechanisms that influence the distribution, behaviour and fate of PPP residues across environmental compartments.

The exposure pathway is conceptually distinct from the source of emission, which marks the initial release of the PPP into the environment, and from the exposure route, which refers to the mode of entry into the organism (defined separately below).

Key processes that constitute exposure pathways for pesticides include:

• Overspray: the direct deposition of spray droplets onto organisms present in the treated area or in adjacent habitats at the time of application. This should not be interpreted as over‐application or the unintentional spraying outside intended field boundaries.

• Drift movement of PPP residues, including droplets, vapours or particles, that occurs during or shortly after application, resulting in off‐target transport. Drift is a major contributor to off‐field exposure for NTTPs, NTAs and, indirectly, for in‐soil organisms. It includes:

  • ○
    spray drift, which is the (horizontal) transport of liquid droplets during application, typically occurring from boom sprayers, air‐blast or handheld devices. It can be further classified as:

    • sedimentary (or gravitational) drift: transport of spray droplets involving relatively large particles that settle rapidly and therefore occur relatively close to the treated field.
    • airborne drift: involving fine droplets that can remain suspended in the air and be transported over longer distances by wind. Unlike vapour drift, the pesticide is still present as liquid droplets.
  • ○
    dust drift: movement of solid particles which may be released during application or sowing of solid formulations or sowing of treated seeds.
  • ○
    vapour drift: short‐range transport of pesticide vapours generated by rapid evaporation of fine spray droplets that remain airborne immediately after application, before or during deposition.

• Volatilisation: the post‐application evaporation (i.e. vertical gaseous transport) of PPPs from treated or contaminated surfaces – including soil, plant foliage (e.g. crop or non‐target plants) or plant litter – into the atmosphere. Volatilised residues may undergo secondary atmospheric transport and contribute to indirect exposure via re‐deposition onto nearby surfaces or surfaces at greater distances.

• Deposition represents the step of settling of airborne PPPs and/or their residues on environmental surfaces such as soil, water or vegetation. Deposition may occur following drift or after volatilisation and subsequent atmospheric transport. This process determines which matrices and, consequently, organisms may be exposed and can be distinguished into:

  • ○
    2‐D deposition, referring to residues settling on horizontal surfaces (e.g. soil, short grass or water).
  • ○
    3‐D deposition, referring to residues settling on vertical or complex structures (e.g. plant canopies and hedgerows).

• Run‐off and erosion: the surface transport of PPP residues from treated areas to adjacent environmental compartments, driven primarily by rainfall, irrigation or snowmelt. These processes relocate residues dissolved in water (via run‐off) or bound to dislodged soil particles (via erosion). Run‐off and erosion can result in the deposition of residues onto off‐field surfaces, such as field margins, buffer strips or soil surfaces close to non‐target vegetation. The extent of exposure depends on site characteristics (e.g. slope, soil texture, vegetation cover, land‐use practices), the timing of the PPP application relative to precipitation and the physico‐chemical properties of the PPP.

• Infiltration and shallow vertical movement: refer to the process by which PPPs, following deposition on the soil surface (both in‐field and off‐field), enter and redistribute within the upper soil profile, typically within the biologically active topsoil layer (0–20 cm). Infiltration occurs as dissolved residues percolate with water in the soil matrix or cracks/macropores, or as particle‐bound residues percolating into macropores or being incorporated into surface soils. Subsequent vertical movement is governed by water flow, sorption–desorption dynamics, degradation rates, soil structure and organic matter content. This process directly influences residue concentrations in soil pore water and solids, thereby shaping exposure levels for in‐soil organisms (e.g. earthworms, collembolans) and determining the availability of residues for root uptake by plants. Unlike deeper leaching, which is primarily relevant for groundwater risk, shallow vertical movement is an ecologically relevant exposure pathway with direct implications for non‐target soil biota and plants. These processes are inherently accounted for within the regulatory exposure models used to quantify predicted environmental concentrations (PECs) in soil and pore water, ensuring that their contribution to residue distribution and bioavailability is reflected in the exposure assessment.

• Redistribution: the post‐depositional movement of PPP residues within or between environmental compartments due to processes such as plant growth and internal translocation, wash‐off from rainfall or irrigation, mechanical mixing of soil (e.g. tillage) or bioturbation by soil fauna. Redistribution alters the spatial and temporal profile of exposure and can reintroduce PPPs into organism‐accessible zones. Of these processes, only wash‐off from rainfall or irrigation and mechanical mixing of soil (e.g. tillage) are explicitly represented in soil exposure models and are therefore considered in the estimation of PECs in soil.

While these transport processes determine where residues move, they do not include the chemical or biological transformations (e.g. degradation, metabolite formation) that alter the substance's identity or persistence. For clarity, in this document the term environmental compartment refers to the main environmental media considered in exposure modelling (e.g. air and soil). The term matrix is used to denote specific physical or biological substrates relevant to exposure assessment, including subcomponents within compartments (e.g. soil pore water, soil solids) and exposed biological or structural surfaces (e.g. litter, leaf surfaces or vegetation). While compartments are the basic modelling units for substance fate and transport, where processes such as degradation, sorption and metabolite formation are parameterised, matrices represent the actual substrates where organisms may be exposed and residues measured. These transformation processes are already accounted for within the environmental fate modelling of each compartment; they influence the magnitude and duration of exposure but fall outside the formal definition of the exposure pathway.

By clearly delineating the environmental transport mechanisms, the concept of exposure pathway provides the essential link between application practices and exposure scenarios, supporting more transparent and harmonised risk assessments across organism groups.

2.4.3. Exposure‐relevant environmental matrix

The exposure‐relevant environmental matrix refers to the specific environmental medium, such as soil, air, water, plant surfaces or plant litter, that carries residues of a PPP in a form that is potentially accessible to non‐target organisms. It serves as the immediate interface between environmental transport processes and biological exposure routes.

The matrix defines the location, form and availability of residues at the point of organism contact and plays a key role in determining exposure magnitude and duration. The relevance of each matrix depends not only on the biology, behaviour and ecological niche of the non‐target organism, but also on the inherent properties of the matrix (e.g. soil structure, moisture content, organic matter content and microbial activity, when considering soil) as well as on the physico‐chemical properties of the PPP, which influence whether residues can reach and persist in that matrix. These characteristics influence how residues are retained, degraded or transformed within the matrix, thereby shaping the likelihood, magnitude and timing of organism exposure.

Organism‐specific (non‐exhaustive) examples:

• In‐soil organisms:

  • ○
    Soil: exposure may occur through direct contact with bulk soil (solid mineral and organic components) or soil pore water (the aqueous phase between particles). Exposure can also occur via soil intake. For soft‐bodied organisms like earthworms, pore water is particularly relevant due to high dermal permeability and continuous skin contact with moist soil. Soil ingestion is also a relevant exposure route in these organisms. In contrast, hard‐bodied organisms may experience more significant exposure via contact with soil particles, although there is evidence showing the importance of soil pore water as an important exposure route to these organisms as well. Therefore, it is important to also consider latter exposure as well, particularly for compounds that have high water solubility. The relative contribution of each matrix also depends on the sorption behaviour and mobility of the PPP (e.g. hydrophilic substances may partition more into pore water).
  • ○
    Plant litter: surface organic matter may hold residues and serve as a food source or substrate.
  • ○
    Plant roots (in specific cases): for organisms that interact directly with root tissues or exudates, roots may be a point of contact for residues that have been systematically translocated.
• NTAs:

  • ○
    Plant structures (external surfaces): leaves, stems and flowers provide physical contact areas where residues may deposit and which arthropods may use for foraging, oviposition or shelter. While not a medium in the classical sense (being defined 2D and not 3D), these surface layers function as ecologically relevant exposure interfaces. Importantly, residues on plant surfaces are typically quantified as mass per unit area (e.g. μg/cm²), in contrast to environmental 3‐D media like soil or water, where residues are measured as mass per volume or mass per mass (e.g. μg/L or μg/kg). This difference in residue expression reflects the surface‐based nature of exposure and supports the treatment of plant surfaces as functionally equivalent to 3‐D matrices in risk assessment.
  • ○
    Plant litter serves as a microhabitat or foraging ground where residues may be encountered.
  • ○
    Soil refers primarily to the soil surface and uppermost substrate layer where ground‐dwelling arthropods may interact with residues via direct contact, resting, sheltering or pupation.
  • ○
    Food sources: plant material, prey items, nectar and pollen may contain residues that result in dietary exposure.
• NTTPs:

  • ○
    Air: medium for transporting PPP residues (e.g. spray drift) that may settle onto or taken up by adjacent vegetation.
  • ○
    Soil: represents the root‐accessible zone of the soil profile (primarily the upper 20–30 cm) where PPP residues may become available for uptake via roots. Root exposure primarily occurs via residues dissolved in the pore water, while the solid soil phase may act as an indirect reservoir, gradually replenishing the dissolved fraction depending on sorption–desorption dynamics and soil properties.

Understanding which matrices are relevant for each organism group, and how residues are present and quantified in them, allows risk assessors to more clearly determine the conditions under which exposure occurs. This understanding is essential for the design of biologically realistic exposure settings that underpin formal exposure scenarios, supporting ecologically meaningful risk assessments and the harmonisation of conceptual models across in‐soil organisms, NTAs and NTTPs. In addition, it may be helpful for defining efficient risk mitigation measures.

To facilitate the interpretation of the concepts introduced above, Table 2 reports the main exposure pathways for PPPs and their links with environmental compartments and exposure‐relevant matrices. It provides an overview of how residues are transported, where they are distributed and which matrices serve as the immediate interface for exposure of non‐target organisms.

TABLE 2.

Overview of the main exposure pathways relevant for non‐target terrestrial organisms and their correspondence with environmental compartments and exposure‐relevant matrices.

Exposure pathway	Description (transport process)	Main environmental compartment(s)	Exposure‐relevant matrices/interfaces
Spray drift sedimentary (gravitational) airborne	Horizontal movement of droplets during application. Sedimentary drift involves large droplets depositing close to the treated field; airborne drift involves fine droplets transported further by wind.	Air as transport medium; receiving compartments only following deposition	Receiving surfaces following deposition
Dust drift	Movement of solid particles (e.g. during sowing or handling of solid applications) that can deposit onto adjacent areas.	Air as transport medium; receiving compartments only following deposition	Receiving surfaces following deposition
Vapour drift	Short‐range transport of pesticide vapours generated by rapid evaporation of fine spray droplets.	Air as transport medium; receiving compartments only following deposition	Receiving surfaces following deposition
Deposition	Settling of airborne residues (droplets, particles, vapours) onto environmental surfaces. 2‐D = horizontal (soil, herbaceous vegetation); 3‐D = vertical/complex (hedgerows, trees).	Transition between air and receiving compartments/matrices	2‐D deposition = residues settling on horizontal surfaces (soil, herbaceous vegetation) 3‐D deposition = residues settling on vertical/complex structures (e.g. hedgerows, trees).
Run‐off and erosion	Surface transport of residues (dissolved or particle‐bound) from treated areas due to rainfall, irrigation or snowmelt.	Soil (source) → Off‐field soil	Soil surface
Volatilisation	Post‐application evaporation of PPPs from contaminated surfaces into the atmosphere. Volatilised residues may later redeposit locally.	Soil, vegetation, plant litter (source) → Air (receiving)	Air (gaseous phase), plant and soil surfaces (for re‐deposition)

2.4.4. Exposure route

The exposure route refers to the biological mechanism through which a non‐target organism interacts with or takes up PPP residues present in or on an exposure‐relevant environmental matrix. It defines the interface between the organism and that matrix, through which the substance may subsequently cross via processes such as absorption, ingestion or uptake. This concept is distinct from the environmental exposure pathway, which describes how the residue reaches the exposure‐relevant environmental matrix.

Although defined primarily from a biological perspective, the relative importance and magnitude of each exposure route can be influenced by the physico‐chemical properties of the substance (e.g. solubility, volatility, lipophilicity), as these characteristics affect the bioavailability, transfer and uptake efficiency of residues within the relevant matrices. The exposure route itself, however, is determined by the traits of each organism group, such as physiology, morphology, behaviour, habitat use and life stage. Identifying these routes is essential for developing realistic exposure scenarios, selecting appropriate systems for assessing ecotoxicological effects and ensuring consistency and harmonisation across risk assessments for different organism groups.

Organism‐specific (non‐exhaustive) examples:

• In‐soil organisms (e.g. earthworms, Collembola):

  • ○
    Contact: Residues within the soil matrix, pore water or litter may interact with the external surface of the organism during movement, burrowing or soil surface activity. This interface may enable passive uptake depending on permeability and residue solubility.
  • ○
    Dietary (oral ingestion): Organisms may ingest residues when feeding on contaminated soil particles or pore water, decomposing organic matter, microorganisms and microfauna or plant litter.
• NTAs:

  • ○
    Contact: Occurs through direct deposition of residues on the organism's body during PPP application (overspray), or via contact with treated surfaces such as leaves, stems, soil or plant debris/litter.
  • ○
    Dietary (oral ingestion): Results from consumption of contaminated food sources such as, plant material, prey, nectar, pollen, contaminated guttation fluid or honeydew. This is especially relevant for pollinators (excluding bees), parasitoids, herbivorous and predatory arthropods.
  • ○
    Inhalation (or respiratory uptake via tracheal system in arthropods): Volatile substances or aerosolised residues may be taken up via the spiracles. This route is limited to compounds with sufficient vapour pressure and is most relevant shortly after application when airborne levels are highest.
• NTTPs:

  • ○
    Root uptake: occurs through absorption of residues from pore water by root tissues.
  • ○
    Above‐ground tissue uptake: absorption of residues by above‐ground plant parts (e.g. leaves, stems) following deposition of drift particles, or vapour‐phase substances. This route includes direct penetration of drift droplets through the cuticle and uptake of vapour‐phase substances via stomata.

Although exposure routes are specific to each group, the principle of identifying and describing the biological interface between organisms and PPP residues in a certain matrix is shared across all three groups. Including exposure routes in the cross‐cutting considerations ensures that each protocol applies a consistent conceptual approach to exposure characterisation. This supports the development of coherent and biologically realistic exposure scenarios, helps align test systems and endpoint selection and enables transparent comparisons of exposure dynamics across in‐soil organisms, non‐target arthropods and non‐target terrestrial plants.

3. PROBLEM FORMULATION

3.1. Objective

The aim of this work is to illustrate the conceptual model that will be applied in the revised guidance for NTAs, in‐soil organisms and NTTPs within Tier 1 risk assessment schemes for PPP of chemical origin. The protocol defines the evidence needs and approaches for addressing data gaps, thereby establishing the basis for the forthcoming revision of the three Guidance Documents.

The protocol focuses on the problem formulation for Tier 1 risk assessment schemes, the related exposure scenarios, the risk assessment parameters to the extent possible and the priority assigned to each of them (see Section 4).

3.2. Assessment question and sub‐questions across the three groups of organisms

The Agent, Pathway, Receptor, Intervention and Output (APRIO) paradigm (EFSA Scientific Committee, 2023) was used to help make the assessment questions and sub‐questions derived from the mandate operational, that is, answerable through the scientific assessment and mutually understandable between risk assessors and risk managers.

In the context of the Terms of Reference for this mandate, the APRIO elements were defined as described in Table 3.

TABLE 3.

APRIO elements related to the Terms of Reference of the mandate.

APRIO elements	Definition (EFSA Scientific Committee, 2023)	In this mandate
Agent (A)	Anything that can cause an effect on a receptor	Pesticide active substance and its metabolite(s) and degradation product(s) and formulated product(s) of chemical origin
Pathway (P)	Any way in which an agent interacts with its receptor: It is the sequence of events leading the agent to cause an effect on the receptor. It can simply cover the route of exposure or represent, for instance, the steps of introduction and spread when assessing a pathogen.	Relevant exposure scenarios and the corresponding routes of exposure
Receptor (R)	Anything that experiences the effect of the agent: The receptor can also experience a secondary consequence to the exposure to the agent (e.g. farmers changing cropping practices because of the crops being affected by a pest).	NTAs/In‐soil organisms/NTTPs
Intervention (I)	Any intentional measure aimed at changing directly or indirectly the exposure and/or the consequence of the exposure to the agent	Not applicable for this protocol since it is limited to Tier 1, only
Output (O)	Form of the answer to the assessment question or sub‐question, the result of an assessment process	Risk assessment

The mandate Terms of Reference have been translated into assessment questions (AQs) and sub‐questions (SQs) based on the risk assessment paradigm: exposure characterisation, hazard identification and characterisation and risk assessment (see Table 4).

TABLE 4.

Assessment question (AQ) and sub‐questions (SQs) to be answered for the revision of the terrestrial guidance.

AQ 1 – How to conduct Tier 1 risk characterisation for non‐target arthropods/in‐soil organisms/non‐target terrestrial plants exposed to plant protection products of chemical origin through various exposure scenarios and routes within the European Union?
1 – Exposure characterisation
SQ 1.1	What are the key pathways through which PPP reach exposure‐relevant environmental matrices?
SQ 1.2	What are the exposure‐relevant environmental matrices leading to exposure of NTAs/in‐soil organisms/NTTPs?
SQ 1.3.	Which are the exposure scenarios for assessing the risks of PPP pesticide risks of NTAs/in‐soil organisms/NTTPs?
SQ 1.4.	Through which exposure routes do NTAs/in‐soil organisms/NTTPs take up PPP residues (active substance, metabolite(s) and degradation product(s)) from the exposure‐relevant environmental matrices?
SQ 1.5	How is the exposure quantified in each relevant environmental matrix considering the relevant route of exposure and exposure scenarios?
2. Hazard identification and characterisation
SQ 2.1.	What is the species and life stage sensitivity of the NTAs/in‐soil organisms/NTTPs to PPPs? This should consider potential differences in life stages of a species, differences across species of the same functional and/or taxonomical group and differences between those groups.
SQ 2.2	Which is the relevant exposure duration for a proper hazard characterisation for NTAs/in‐soil organisms/NTTPs at Tier 1?
SQ 2.3	What are the relevant (eco)toxicological types of effects to be considered at Tier 1 and how to characterise them?
SQ 2.4	What type of information will be derived from Tier 1 studies (e.g. single endpoint, type of endpoint and level of effect, e.g. ECx, NOEC or whole dose–response)?
3 – Risk characterization (hereafter referred to as Risk assessment)
SQ 3.1	How to link the predicted exposure with the effect as characterised for each exposure scenario and risk case 5 ?
SQ 3.2	How to assess the outcome of the risk assessment in relation to the agreed specific protection goals?

3.3. Conceptual model

To ensure that the assessment process is coherent and transparent across organism groups, a conceptual model (see Figure 2) was developed to link the exposure and hazard components of the risk assessment. The same assessment question and sub‐questions apply across the protocols of the three guidance documents, i.e. NTAs other than bees, in‐soil organisms and NTTPs.

FIGURE 2.

Conceptual model linking the assessment question and sub‐questions.

As illustrated in the conceptual model in Figure 2, the arrow preceding the last exposure sub‐question indicates that the exposure‐related sub‐questions must be addressed sequentially. Sub‐question 1.5 is essential to ensure that exposure characterisation leads to a quantitative outcome that can be used in the risk characterisation. While sub‐questions 1.1–1.4 define the key components of exposure (i.e. pathways, matrices, scenarios and routes as described in Section 2.4), sub‐question 1.5 integrates these elements to define how exposure is quantified. It therefore represents the final step of the exposure pillar and logically follows the preceding sub‐questions. Similarly, the risk assessment sub‐questions should be answered in a sequential order. In contrast, the sub‐questions related to hazard are not interdependent and can be addressed in parallel, as illustrated by the radial representation in Figure 2. The combined answers to all exposure and hazard‐related sub‐questions, inform the sub‐questions related to risk assessment and, ultimately, lead to answering the assessment question. Specific conceptual model for NTAs, in‐soil organisms and NTTPs outlining how the sub‐questions are proposed to be addressed is presented in Figures 8, 12, 17.

FIGURE 8.

Applied conceptual model linking exposure (in blue) and hazard (in red) sub‐assessment questions for non‐target arthropods.

FIGURE 12.

Applied conceptual model for in‐soil organisms where sub‐questions and related answers are interlinked. The gradient orange box indicates that the relevance of the oral route of exposure will be further investigated, and it is therefore not yet confirmed. The dashed orange lines are related to the relevance of the oral exposure route for the different in‐soil organisms.

FIGURE 17.

Applied conceptual model for non‐target terrestrial plants where sub‐questions and related answers were are interlinked.

3.4. Gap analysis

To address the assessment sub‐questions outlined in Table 3, a gap analysis was performed for each of the three groups of organisms. This analysis identified additional, supplementary sub‐questions that must be addressed to complete the assessment. It was used to prioritise the elements needed to address each sub‐question identified in Table 3.

Several relevant documents were consulted for defining the gaps:

• Reports from the ESCORT 2 (Candolfi et al., 2001) and ESCORT 3 (Alix et al., 2012) workshops, organised by SETAC (Society of Environmental Toxicology and Chemistry);

• The EFSA PPR Panel (2014) Scientific Opinion on NTTPs;

• The EFSA PPR Panel (2015) Scientific Opinion on NTAs;

• The EFSA PPR Panel (2017) Scientific Opinion on in‐soil organisms;

• The EFSA Technical Reports on general recurring issues in ecotoxicology (EFSA, 2015, 2019);

• The report from the 2025 workshop co‐organised by EFSA and the European Commission on the revision of the terrestrial ecotoxicology guidance document and the development of an approach on indirect effects (EFSA, 2025a);

• The Deficit analysis ⁶ (Byers et al., 2025) for the revision of the guidance on NTTPs conducted by UBA (Umweltbundesamt) with the support of other Member State representatives.

Moreover, the ERA Working Group was also consulted, and additional gaps might be identified and added based on expert judgement during the process of development of the revised Guidance documents.

4. PRIORITISATION OF ASPECTS RELATED TO EACH SUB‐QUESTION

Once the gaps, which may be considered part of the sub‐question, were identified, the relative priority for addressing each gap was defined using a risk matrix. For exposure sub‐questions two aspects were considered:

• Likelihood of occurrence, e.g. how often a pathway is expected to occur under realistic agricultural and environmental conditions.

• Relevance, e.g. as the potential of a pathway to result in ecotoxicologically meaningful exposure of non‐target organisms to PPP residues or metabolites/degradation products.

For hazard and risk assessment sub‐questions, instead, the following were considered:

• Impact on risk assessment: the chance that the identified gap can occur or affect the outcome of hazard characterisation under realistic assessment conditions.

• Relevance: the extent to which the identified gap could significantly influence the accuracy or ecotoxicological robustness or significance of the hazard characterisation.

The score to assign to each gap in terms of likelihood or impact on risk assessment and relevance was based on regulatory experience and expert judgement. The resulting relative priority together with a consideration of the sensitivity of the item, e.g. based on preliminary comments received by stakeholders, dictates the choice of the method for filling the gaps. The higher the relative priority and the sensitivity of a gap, the more robust were the methods chosen for addressing it. Gaps with higher priority will require more robust and systematic approaches to ensure that the data generated can adequately support the risk assessment framework. Gaps with lower priority will not be disregarded, but their consideration in the guidance development may be less comprehensive and more pragmatic or left to a case‐by‐case consideration during individual risk assessments.

It is important to note that the analysis and protocol only outline those aspects that are not currently addressed in the Tier 1 risk assessment or those that, although currently addressed, need revision and therefore nevertheless represent a gap.

5. NON‐TARGET ARTHROPODS

5.1. Gap analysis

Figure 3 gives an overview of the gaps identified in the current risk assessment of NTAs related to the exposure and the hazard identification and characterisation.

FIGURE 3.

Gap analysis for exposure assessment (green boxes) and effect assessment (blue box) for non‐target arthropods using a fish‐bone diagram. All considerations apply to both active substances and their metabolites/degradation products.

The list of gaps identified are listed below and are already categorised based on the sub‐question they contribute to.

Exposure pathways (sub‐question 1.1)

• Need to update spray drift deposition values for conventional applications and to reflect advances in application technology and farming practices.

• Lack of assessment of exposure to dust released from solid formulations, such as abraded seed coatings and granules, including dust deposition in‐field and dust drift to off‐field areas.

• Lack of assessment of exposure through vapour drift for off‐field exposure.

• Lack of assessment of exposure through volatilisation.

Exposure‐relevant environmental matrices (sub‐question 1.2)

Current exposure characterisation for NTAs does not adequately consider a variety of environmental matrices that may act as relevant sources of PPP residues. These include:

• Plant structures, tissues and secretions (e.g. pollen, nectar, guttation fluid) and insect‐produced honeydew.

• Plant litter and decaying material.

• Contaminated prey or hosts.

• Other residue‐bearing substrates (e.g. spider webs, animal‐created structures).

Many NTAs interact with these matrices via feeding or other ecological behaviours, potentially resulting in direct or indirect exposure.

Exposure scenarios (sub‐question 1.3)

Certain exposure scenarios are not explicitly covered in Tier 1:

• Weeds in the treated field.

• Field margins and adjacent crops unintentionally contaminated with PPP residues via drift or volatilisation.

• Succeeding crops exposed to persistent and plant‐mobile PPP residues remaining in soil.

Exposure routes (sub‐question 1.4)

Several relevant exposure routes are currently omitted from Tier 1 assessments, including:

• Contact exposure through direct overspray or drift for NTAs present in crops or weeds and field margins/adjacent crops, respectively.

• Dietary (oral) exposure through feeding on treated foliage or contaminated food sources.

• Inhalation exposure to volatilised substances.

Quantification of exposure (sub‐question 1.5)

• Lack of empirical evidence supporting the default vegetation distribution factor (VDF) used to estimate off‐field exposure of NTAs in vegetated field edges.

• Insufficient residue data for exposure‐relevant environmental matrices, particularly in off‐field vegetation, following PPP application.

• Limited consideration of PPP wash‐off and translocation processes within plants, which can affect residue distribution and bioavailability in plant structures (both in‐field and off‐field).

Sensitivity of NTA species (sub‐question 2.1)

• Limited representativeness and sensitivity of the indicator species currently used in Tier 1 risk assessments.

• Incomplete coverage of sensitive life stages; current test methods do not address eggs.

Relevant exposure duration (sub‐question 2.2)

• Absence of chronic toxicity assessment. Tier 1 focuses on acute endpoints, potentially underestimating long‐term effects.

Relevant ecotoxicological type of effects (sub‐question 2.3)

• Insufficient assessment of sublethal endpoints, particularly reproductive effects, in Tier 1 testing.

Information to be derived from Tier 1 studies (sub‐question 2.4)

• No explicit gaps were identified regarding the type of information to be derived from Tier 1 studies (e.g. single endpoint versus whole dose–response). Nevertheless, pending on the SPGs selected by risk managers, the level of effect and the type of endpoint, e.g. EC_x/ER_x should be further considered.

Test methods (sub‐questions 2.2, 2.3 and 2.4)

• Lack of suitable Tier 1 testing methods to address certain exposure routes (e.g. oral or overspray exposure) and to account for PPPs with specific or non‐standard modes of action (e.g. insect growth regulators, physically acting substances, seed treatments, feeding inhibitors).

5.2. Risk matrix

For exposure characterisation, the analysis aimed to rank and/or prioritise the gaps related to the exposure pathways, environmental matrices, scenarios and exposure routes identified in Section 5.1.

For hazard identification and characterisation, the analysis focused on ranking and prioritising the gaps related to species sensitivity, exposure duration, ecotoxicological type and level of effects and information to be derived from Tier 1 studies, identified in 5.1.

5.2.1. Exposure characterisation

5.2.1.1. Exposure pathways (sub‐question 1.1)

For exposure pathways, likelihood and relevance were further defined as follows:

• Likelihood referred to the frequency with which the pathway is expected to occur under realistic agricultural and environmental conditions.

• Relevance was defined as the potential of a pathway to result in ecologically meaningful exposure of NTAs to PPP residues.

The analysis was carried out separately for in‐field and off‐field areas based on expert knowledge from members of the ERA working group. The evaluation considered current knowledge on environmental fate processes and NTA ecological traits.

The outcomes of this evaluation are presented in Figure 4 for in‐field and off‐field areas.

FIGURE 4.

The diagram illustrates the overall priority level attributed to each potential exposure pathway (from low to high, x‐axis). The overall priority level is obtained by combining the score attributed to likelihood and relevance. ⁷ This score is represented by the number of bricks in each colour (i.e. 1 brick = minimal; 2 bricks = moderate, 3 bricks = large).

The ERA Working Group decided to include only those pathways reaching at least an overall medium priority level in the protocol, thus excluding all others. The excluded pathways are:

• FRoot uptake and translocation (treated crop, spray applications post‐emergence applications): Expected to have a much lower contribution to exposure compared with direct deposition.

• Volatilisation and re‐deposition: Low contribution compared to direct deposition pathways. Depends on substance vapour pressure and environmental conditions.

• Vapour drift: Limited to substances with high vapour pressure. Contribution influenced by post‐application environmental conditions and substance properties.

The approach for these pathways will be considered, where appropriate, during the development of the revised guidance document.

5.2.1.2. Exposure‐relevant environmental matrices (sub‐question 1.2)

Not all environmental matrices contribute equally to risk assessment. Their relevance for exposure to NTAs and the likelihood of contamination under standard agricultural conditions vary.

In the case of exposure‐relevant matrices, relevance and likelihood were further defined as follows:

• Relevance (i.e. potential for ecologically meaningful exposure) was defined as the capacity of a matrix to contribute significantly to NTA exposure.

• Likelihood (i.e. frequency of contamination) was defined as how often the matrix is expected to contain PPP residues following application under typical agricultural conditions.

The results are summarised in Figure 5.

FIGURE 5.

Overall priority levels for potential environmental exposure matrices for non‐target arthropods. The diagram illustrates the overall priority level attributed to the potential environmental matrices (from low to high, x‐axis). The overall priority level is obtained by combining the score (see footnote 7) attributed to likelihood and relevance. This score is represented by the number of bricks in each colour (i.e. 1 brick = minimal; 2 bricks = moderate, 3 bricks = large). *Plant litter/debris will be primarily covered in the in‐soil organisms Guidance Document. However, leaf litter is also recognised as an ecologically relevant matrix for surface‐dwelling NTAs, serving both as a habitat and as a food source. Further data and literature on litter occurrence and ingestion by NTAs will be collected for future consideration.

The ERA Working Group decided:

• Not to collect further evidence to perform a Tier 1 exposure assessment for those matrices with an overall low priority: surface water accumulations, stem flow and throughfall water, air, spider webs, faecal material and animal‐created structures. It should be noted that plant litter/debris will be considered for in‐soil organisms (see Sections 6.2.1.2 and 6.3.1.2).

• To gather further evidence before making a final decision on their inclusion in the revised guidance document for those matrices categorised with an overall medium priority: guttation fluid, extrafloral nectar and honeydew, and surface water films.

5.2.1.3. Exposure scenario (sub‐question 1.3)

The identification and prioritisation of exposure scenarios was not conducted in line with the approach explained in Sections 5.2.1.1 and 5.2.1.2. In this case, harmonisation with other existing risk assessment schemes was considered. In alignment with the EFSA (2023a) Bee Guidance Document, several exposure scenarios will be distinguished within in‐field and off‐field areas. For in‐field areas, these scenarios will include treated crops, weeds within the treated field and succeeding crops. For the off‐field areas, the scenarios will include field margins and adjacent crops. This potential expansion of exposure scenarios will be evaluated for each of the exposure routes described above: contact exposure (via residues deposited or reaching environmental matrices and direct overspray) and dietary (oral) exposure. Plant‐ and ground‐dwelling NTAs living in in‐field and off‐field areas will be considered.

5.2.1.4. Exposure routes (sub‐question 1.4)

The ERA Working Group agreed that all exposure routes identified in the EFSA PPR Panel (2015) Scientific Opinion should be covered in the revised Guidance Document, as they are all relevant and likely to occur in the case of conventional PPP application. These include:

• Dietary (oral) exposure.

• Contact exposure, either through direct overspray or via contaminated matrices (including some others in addition to the ones currently considered) or surfaces resulting from PPP application.

The relevance of exposure via inhalation was discussed by the ERA Working Group. While the experts acknowledged that inhalation could contribute to overall residue exposure, they concluded that this exposure route is less relevant and should be given lower priority for inclusion in the revised Guidance Document. This conclusion was primarily based on two considerations as outlined in the FOCUS Air Group Report Pesticides in Air (2008). First, the underlying processes of vapour drift and volatilisation occur only for a limited group of substances with sufficiently high vapour pressure. According to the FOCUS Air Group Report Pesticides in Air (2008), a substance must have a vapour pressure > 10⁻² Pa for these processes to be significant. Both processes are therefore restricted to a relatively small subset of active substances currently in use.

Second, even when relevant, the magnitude of exposure from these processes is generally small compared with the main deposition routes. The FOCUS Air Group Report Pesticides in Air (2008) demonstrated that deposition of volatilised residues is quantitatively much less important than spray drift, especially at short distances from the treated field (e.g. 1 metre for arable crops, 3 metres for vines and tree crops). These emissions are often so low that they cannot be effectively captured by standard drift collectors, which further demonstrates their limited contribution to overall exposure.

On this basis, the ERA Working Group agreed that inhalation exposure should be acknowledged in the revised Guidance Document but treated as a low‐priority route for Tier 1, compared with oral and contact exposure. Such exposure routes could be covered in higher tier assessments using field studies. However, the possibility of performing quantitative exposure assessments for highly volatile substances will be considered during Guidance development, ensuring that the cases where volatilisation may play a more significant role in NTA exposure are appropriately addressed.

Table 5 summarises the key exposure scenarios to be considered in the Tier 1 risk assessment (using the same terminology as in EFSA, 2013, 2023a, 2023b) and categorising exposure according to the main routes: dietary (oral), contact and inhalation.

TABLE 5.

Scenarios to be addressed in Tier 1 risk assessment for non‐target arthropods.

Scenario			Process description
Dietary (oral) exposure
In‐field areas	Treated field	Crop	Foliage‐dwelling arthropods ingesting contaminated crop‐related matrices (e.g. leaves, tissues, nectar, pollen) or prey/hosts
		Soil surface	Ground‐dwelling arthropods ingesting prey/hosts
	Weeds in the field		Foliage‐dwelling arthropods ingesting contaminated weed‐related matrices (e.g. leaves, tissues, nectar, pollen) or prey/hosts
	Succeeding crop		Foliage‐ and ground‐dwelling arthropods ingesting relevant matrices contaminated through root uptake and translocation of residues into the succeeding crop
Off‐field areas	Field margin/Adjacent crop	Off‐field plants	Foliage‐dwelling arthropods ingesting contaminated plant‐related matrices (e.g. leaves, tissues, nectar, pollen) or prey/hosts
		Soil surface	Ground‐dwelling arthropods ingesting prey/hosts
Contact exposure a
In‐field areas	Treated crop	Crop	Foliage‐dwelling arthropods oversprayed during PPP applications or coming into contact to dried or fresh residues on crop surfaces
		Soil surface	Ground‐dwelling arthropods oversprayed during PPP applications or coming into contact to dried or fresh residues on soil surfaces
	Weeds in the field		Foliage‐dwelling arthropods oversprayed during PPP applications or coming into contact to dried or fresh residues on weed surfaces
Off‐field areas	Field margin/Adjacent crop	Off‐field plants	Foliage‐dwelling arthropods coming into contact to spray droplets during PPP application or coming into contact to dried or fresh residues on non‐target plant surfaces
		Soil surface	Ground‐dwelling arthropods coming into contact with spray drift droplets during PPP application or coming into contact to dried or fresh residues on soil surfaces

^a It comprises contact exposure from direct overspray and from matrices containing PPP residues.

5.2.2. Hazard identification and characterisation

The risk matrix was applied to assign the identified gaps a relative priority, based on the combination of their relevance and likelihood of occurrence or potential impact, as defined in Section 4. The outcome of the analysis is presented in Figure 6.

FIGURE 6.

The diagram illustrates the overall priority level attributed to the identified gaps related to hazard characterisation (from low to high, x‐axis). The overall priority level is obtained by combining the score (see footnote 7) attributed to likelihood and relevance. This score is represented by the number of bricks in each colour (i.e. 1 brick = minimal; 2 bricks = moderate, 3 bricks = large).

The ERA Working Group identified several gaps that could significantly affect hazard characterisation, potentially leading to an underestimation of risks and therefore all considered to have high relevance and high likelihood. Those gaps are:

• Species representativeness and sensitivity of the standard indicator species currently used in Tier 1 risk assessments, including the limited coverage of potentially sensitive life stages (further details are provided below).

• Absence of hazard characterisation resulting from long‐term exposure.

• Insufficient assessment of sublethal endpoints.

• Lack of suitable Tier 1 testing methods to address certain exposure routes, (oral, overspray) and PPPs with specific or non‐standard modes of action (e.g. insect growth regulators, physically acting substances, seed treatments, feeding inhibitors) (further details are provided below).

Sensitivity and representativeness of NTA species

The Tier 1 risk assessment aims to strike a balance between practicality and protection by using a limited number of representative test species. Currently, Tier 1 testing for NTAs is based on two standard species: Typhlodromus pyri (predatory mite) and Aphidius rhopalosiphi (parasitic wasp). Their selection was originally informed by sensitivity studies (Candolfi et al., 1999; Vogt, 2000). However, subsequent evaluations have revealed shortcomings in both their taxonomic and functional representativeness, advocating for an analysis of existing data to better understand sensitivity, vulnerability and a better coverage of different feeding guilds (De Lange et al., 2012; DEFRA, 2005, 2007; EFSA PPR Panel, 2015; Sowa et al., 2025, 2026; Uhl & Brühl, 2019).

While current Tier 1 protocols include representatives of certain taxonomic groups, significant groups, for instance, Lepidoptera, which contribute to pollination and represent an important herbivorous guild, are not represented. Likewise, decomposers, soil surface‐dwelling organisms and specialist feeders are still poorly represented.

Recognising these gaps, the EFSA PPR Panel (2015) emphasised species selection for Tier 1 testing as a critical priority in the revision of the Guidance Document. It is recommended that testing should explicitly consider the most sensitive life stages, expand the range of species to encompass broader taxonomic and functional diversity, and systematically address uncertainties associated with interspecies differences in sensitivity (Sowa et al., 2026).

Current test methods and limitations

A comprehensive list of available and potentially relevant laboratory test methods was presented in the EFSA PPR Panel (2015) Scientific Opinion. Appendix G details IOBC published standard test methods for evaluating PPP side‐effects on NTAs, as recommended by the current Guidance Document. Additionally, it includes test protocols submitted during peer‐reviewed PPP active substance evaluations in the EU or identified from open literature that address traits, taxonomic groups, endpoints or exposure pathways missing from standard IOBC methods. The information was compiled from public literature, but non‐systematic methods were used.

Despite the availability of these test methods, several limitations in current test systems have been identified that compromise the consistency and reliability of hazard assessment across different NTA species.

• Test system design: Lack of uniformity across species, major variations in substrate types, life stages tested, exposure duration and endpoints.

• Substrate and exposure methods: NTAs are tested on various substrates with different exposure routes and intensities depending on PPP application, ranging from dried residues to direct overspray and ‘extended’ tests using natural substrates compared to standard glass plate tests.

• Endpoint assessment: Measured endpoints vary; most reproduction studies focus on short‐term PPP exposure rather than exposure throughout the entire reproductive period.

• Life‐stage selection: Highly variable; larvae and adults of holometabolous arthropods differ in habitat preference and feeding behaviour.

• Current testing designs are not suited for supporting toxicokinetic‐toxicodynamic (TK‐TD) modelling approaches, which could characterise hazard independently of exposure time and facilitate mechanistic understanding.

Therefore, the ERA Working Group agreed on the need to extend the scope of Tier 1 testing, including additional species, sensitive life stages and related modifications of the test methods.

5.3. Methods for answering the sub‐questions

Methods for addressing the sub‐questions related to exposure, hazard and risk assessment are outlined in Sections 5.3.1, 5.3.2, 5.3.3, respectively. An integrated conceptual model is included at the end of Section 5.

5.3.1. Exposure characterisation sub‐questions

5.3.1.1. Exposure pathways (sub‐question 1.1)

To comprehensively address this sub‐question, it is important to distinguish between in‐field and off‐field areas, as the relevant exposure pathways differ depending on spatial context and the environmental fate of the PPP during and after application.

In‐field

The key exposure pathway for ground‐ and foliage‐dwelling NTAs living in in‐field areas is the direct deposition of PPP residues onto environmentally relevant matrices (e.g. soil surface, plant material) or directly onto the arthropod's cuticle through overspray.

Other pathways, including volatilisation, root uptake and translocation into plants following spray application (post‐emergence), and dust deposition have been assigned lower priority as shown in Figure 4.

Volatilisation (see Section 2.4.2 for the definition) from treated or contaminated in‐field surfaces may occur after application. However, for in‐field NTA exposure, direct deposition/overspray is expected to remain the dominant exposure pathway at Tier 1. Therefore, volatilisation is not further developed as a separate in‐field exposure pathway in this protocol. This does not mean that it is excluded but that it may be considered more pragmatically or on a case‐by‐case basis, where relevant.

Redistribution of residues (see Section 2.4.2 for the definition) following initial deposition (e.g. wash‐off by rain or irrigation) may lead to PPPs entering the soil matrix. These processes are already implicitly accounted for in PEC_soil and PEC_porewater calculations through the standard exposure models, as wash‐off from the plant canopy contributes to the residues transfer to soil. Therefore, this pathway does not need to be considered separately at Tier 1. For plant surfaces, however, wash‐off is not considered at Tier 1, because it would reduce estimated residue levels and therefore not represent a worst‐case assumption for organisms contacting or feeding on vegetation.

Foliar and root uptake and translocation in the treated field following spray applications (post‐emergence, for root uptake) and root uptake and translocation in the treated field following the application of solid formulations become relevant for substances that are mobile in plants. In addition, root uptake and translocation in succeeding crops are considered relevant for any persistent active substance or metabolite that is present in the soil. When foliar and root uptake occurs, this may lead to contamination of herbivorous prey/hosts feeding on them and contamination of their products (e.g. honeydew). The likelihood of these translocation pathways may vary depending on factors such as soil properties, substance characteristics, plant phenology and growth stage. Although these processes are recognised and have been attributed to a certain likelihood in Section 5.2.1.1, their quantitative contribution to overall exposure is still uncertain. Therefore, a systematic literature review and experimental work are being conducted in the context of the ‘PERA (FPA EUBA/EFSA/PREV/2023/01) literature searches and experimental work on bioavailability in plants. The ERA Working Group will decide whether to incorporate these pathways into the Tier 1 exposure assessment based on the outcomes of both the literature review and experimental work performed in the context of the PERA project.

The current in‐field NTA risk assessment only considers direct deposition on crop material and soil surface, assuming homogeneous exposure based on application rate. The ERA Working Group agreed that this pathway will be retained in the revised Guidance Document. In addition to direct deposition, direct overspray will be incorporated into Tier 1 exposure assessments. Section 5.3.1.5 provides further details on the quantification of these pathways.

Off‐field

The main off‐field exposure pathway for NTAs is spray drift, both sedimentary and the finer airborne droplet fraction (see Section 2.4.2 for the definition); see Section 5.3.1.5 for exposure quantification.

Beyond spray drift, dust drift (e.g. from sowing treated seeds or spreading granules) (see Section 2.4.2 for the definition) may also contribute to off‐field exposure.

The ERA Working Group agreed that those pathways should also be considered in the revised Guidance Document. Section 5.3.1.5 provides more detail on how dust drift and spray drift will be incorporated in Tier 1 exposure assessments.

Volatilisation followed by atmospheric transport and re‐deposition may also contribute to off‐field exposure, particularly for semi‐volatile and volatile substances. Volatilisation may be more pronounced shortly after application but can persist under favourable environmental conditions. The EFSA Scientific Opinion (2017) refers to the EVA 2 model (FOCUS, 2008) for estimating vapour drift deposition, although it notes its limitations. More recently, EVA 3.0 (https://www.bvl.bund.de/SharedDocs/Downloads/04_Pflanzenschutzmittel/zul_umwelt_eva_prog‐EN.html?nn=11010942) has become available and may offer improved assessment capabilities. For the purposes of this protocol, volatilisation/re‐deposition is acknowledged as a potentially relevant off‐field pathway but is not prioritised for additional data collection or detailed methodological development at this stage. In line with the general prioritisation approach described in Section 4, lower priority does not mean that the pathway is disregarded. Rather, its consideration may be less comprehensive, more pragmatic or addressed on a case‐by‐case basis during guidance development or individual risk assessments, where relevant.

Surface run‐off/erosion (see Section 2.4.2 for the definition) can cause the displacement of residues from treated fields to adjacent off‐field areas, such as field margins. The ERA Working Group considered that the exposure into soil resulting from these processes will be covered by the in‐soil organisms guidance document (refer to Section 6.3.1.5), whereas the exposure on the soil surface is retained within the scope of Guidance Document for NTAs.

5.3.1.2. Exposure‐relevant environmental matrices (sub‐question 1.2)

The current risk assessment scheme explicitly covers only plant surfaces and soil surfaces, while the other matrices listed in Table 6 are excluded from the exposure assessment.

TABLE 6.

Potential exposure routes and scenarios for which an exposure model should be developed for the revised guidance document for non‐target arthropods.

Exposure route		Main environmental matrix	Exposure scenario	Main exposure pathways
Contact	Direct overspray	Air	Treated field Weeds in the treated field	Direct application of PPP
			Field margin Adjacent crop	Drift a
	Environmental matrices with PPP residues	Plant surfaces Soil surface	Treated field Weeds in the treated field	Direct application of PPP
			Field margin Adjacent crop	Drift a
Dietary (oral) b		Plant surfaces Plant tissues Nectar/pollen Soil surface Prey/hosts Guttation fluid c Extrafloral nectar c Honeydew c Surface water films c	Treated field Weeds in the treated field	Direct application of PPP Foliar and root uptake followed by translocation (spray and solid formulations, respectively) c
			Field margin Adjacent crop	Drift a
			Succeeding crop	Root uptake followed by translocation (plant‐mobile PPPs)

^a Drift includes spray drift (sedimentary and airborne) and dust drift.

^b Several functional groups will be considered: Herbivores, predators and parasitoids, others (e.g. pollinators, decomposers, scavengers).

^c Relevance and/or likelihood to be ascertained.

Environmental matrices that will be covered in the revised Guidance, as agreed by the ERA Working Group and indicated in Table 6, are: plant and soil surfaces (already considered in the current risk assessment framework), prey/host food items, plant tissues, air vapour phase (to cover contact exposure from overspray, see Section 5.3.1.4), pollen and nectar.

For some other matrices (see Table 6), it is uncertain to which extent they can be contaminated with significant amounts of PPP residues, and whether they contribute significantly to the exposure of NTAs either via contact, dietary or inhalation routes. These include honeydew produced by phloem‐feeding insects, contaminated water, including guttation fluid and extrafloral nectar.

Ingestion of honeydew has been suggested as a route of exposure to NTAs as honeydew excreted by phloem‐feeding insects can contain PPP residues (Calvo Agudo, 2021). Some non‐exhaustive information on crops in which honeydew can be present is included in Appendix A of the supplementary document of the EFSA (2023a) Guidance Document on the Risk Assessment of Bees.

Extrafloral nectaries are produced by several crops, including apricots, beans (including broad beans, horse beans, dry beans), castor oil plants, cherries, cowpea, elder, hemp, leguminous vegetables, lentil, okra, peaches and nectarines, pumpkins, squash and gourds, cotton (seed), sesame seed, sunflower and vetches (Appendix A of EFSA, 2023a). Extrafloral nectar is consumed by NTAs (e.g. Limburg & Rosenheim, 2001; Pemberton & Vandenber, 1993) and can eventually be contaminated with PPP residues (Bredeson & Lundgren, 2018).

Guttation refers to the process of exudation of liquid droplets from the tips, edges and axial surfaces of plant leaves. It occurs in a wide range of plants when environmental conditions favour rapid absorption of water and low transpiration (Sanjay, 2016), such as are present in the early morning or late evening. Guttation fluid originates in the sap of the plant and consists primarily of water, but also contains any other substances present in the sap, such as pesticides residues. There are examples in literature where pesticides were found to be present at high concentrations in guttation fluid (e.g. Nikolakis et al., 2014; Pistorius et al., 2012; Tapparo et al., 2011). A recent review (Urbaneja‐Bernat et al., 2024) highlights the nutritional value of plant guttation droplets, present in several perennial and annual crops, as a reliable resource for NTAs.

In order to investigate the ecological relevance of exposure via the consumption of contaminated water, including guttation fluid, honeydew and extrafloral nectaries, it is necessary to understand (i) when and where these matrices are likely to be present relative to NTAs activity patterns, (ii) which NTA families/species actually interact with these matrices in agricultural systems and (iii) what proportion of an NTA's diet these matrices could represent.

Based on the exclusion of these matrices in the revised EFSA Revised Guidance Document on the Risk Assessment (2023a) of PPPs on Bees due to insufficient data, and considering the specialised nature of honeydew, guttation fluid and extrafloral nectaries as environmental matrices, substantial data limitations are anticipated that may prevent robust quantitative exposure assessment for Tier 1 risk assessment. Key data gaps are expected in relation to limited information on PPP residue concentrations in these matrices, sparse documentation of NTA consumption rates and behavioural interactions and insufficient characterisation of likelihood of those matrices across different agricultural systems and seasons. The high degree of variability anticipated in matrix occurrence and utilisation patterns among diverse NTA groups may further complicate the development of standardised exposure scenarios suitable for Tier 1 assessment. Given these anticipated constraints, a streamlined assessment approach using narrative literature searches and WG expert judgement is proposed as more resource‐efficient than comprehensive systematic reviews, allowing for early identification of data adequacy and matrix relevance while avoiding extensive investment in potentially data‐poor areas.

5.3.1.3. Exposure scenarios (sub‐question 1.3)

The current risk assessment distinguishes between two main exposure scenarios: in‐field and off‐field. The ERA Working Group decided for the time being, however, to follow the approach of the EFSA bee Guidance Documents (EFSA, 2013, 2023a) and to consider the following scenarios in the revised Guidance Document for NTAs:

• Treated field

• Weeds in the treated field

• Succeeding crop in in‐field areas

• Field margin

• Adjacent crop

The identified scenarios may nevertheless be amended and/or expanded if a need is identified at a later stage.

Treated field

Under the revised Guidance Document, this scenario will correspond to the in‐field scenario described in the existing Guidance (European Commission, 2002). Both plant‐dwelling and ground‐dwelling NTAs will be considered in this scenario. This scenario represents the area directly subjected to pesticide application, where exposure is expected to be the highest. Exposure may occur through multiple routes (e.g. direct overspray, contact with residues on foliage or soil and dietary uptake via contaminated matrices), and the assessment will account for the relevance of these pathways depending on the feeding guild of the NTAs in question.

Weeds in the treated field scenario

NTAs may be exposed to PPP residues in treated fields while active in areas where weeds are present, with particular concern for pollinators when weeds are in flower due to their flower‐visiting behaviour. Therefore, in line with EFSA (2023a, 2023b) this scenario will be included in the revised guidance. The WG will further decide in which situations it can be considered relevant and those where it can be excluded.

Succeeding crop scenario

Exposure resulting from pesticide residues that persist in soil from a previous application – subsequently taken up by the roots of a succeeding annual crop and translocated through the plant's vascular system – is not addressed in EFSA PPR Panel (2015). This so‐called ‘succeeding crop’ scenario is included in both bee Guidance Documents (EFSA, 2013, 2023a, 2023b), as bees may be exposed to contaminated pollen and nectar from a flowering, untreated crop grown in rotation with a previously treated crop.

It would be logical to incorporate this additional scenario into the exposure assessment for NTAs, especially for persistent substances. Unlike bees, NTAs may be exposed not only via nectar and pollen but also to other matrices such as plant tissues, the soil surface and food sources, including prey items. Therefore, the relevance of these additional matrices needs to be established in the context of the succeeding crop scenario.

Key questions for further development of this scenario include:

• In addition to pollen and nectar (already considered in EFSA, 2023a), which environmental matrices should be included in the assessment if a succeeding crop scenario is triggered?

• The analysis of available field‐relevant residue levels of PPPs in pollen and nectar of flowering succeeding annual crops suggests that a conservative assumption can be made, that predicted environmental concentration in pore water = concentration in pollen = concentration in nectar. For matrices other than pollen and nectar, is it possible to make the same assumption? If not, how should concentrations in other exposure‐relevant matrices be estimated?

• Is there evidence that concentrations in environmental matrices (other than pollen and nectar) may exceed the concentrations in the directly treated crop? If concentrations in the succeeding crop scenario are consistently lower than those in the treated crop across all relevant matrices, this scenario could potentially be excluded for simplification purposes, noting that differences in overall crop attractiveness, including palatability, to NTAs may also need consideration.

To address those questions, it is proposed to use existing residue datasets, e.g. the initial residue database (Annex H of the EFSA (2023a) bee Guidance Document) and the related to residue in foliage of the birds and mammals Guidance Document (EFSA, 2023b) to identify the presence and magnitude of residues in non‐floral tissues that can result in exposure to NTAs and the outcome of the systematic literature review and the experimental work on plant translocation envisaged under the PERA project.

Field margin and adjacent crop scenarios

The ERA Working Group agreed that both off‐field scenarios (field margin and adjacent crop) should be assessed for NTAs, given their potential wide home ranges and dispersal capabilities (e.g. Sivakoff et al., 2012). Although exposure in these scenarios is generally lower than in the treated field, the taxonomic composition and ecological relevance of the off‐field area differ from those of in‐field area. Moreover, the SPGs differ for in‐field and off‐field areas. Both plant‐dwelling and ground‐dwelling NTAs can be present, and exposure may occur via contact with residues on vegetation or soil, contact with spray droplets and dietary uptake from contaminated matrices.

5.3.1.4. Exposure routes (sub‐question 1.4)

Three primary exposure routes for NTAs to PPPs were identified in EFSA PPR Panel (2015) based on their biological relevance and their distinct exposure dynamics: (i) (physical) contact with residues on plant surfaces or on soil surface, (ii) direct overspray and (iii) oral (dietary) exposure through ingestion of contaminated food. Only contact (physical) with contaminated PPP residues is addressed in the current risk assessment scheme.

Direct overspray exposes arthropods to the highest residue loads during and immediately after application. Subsequent contact with contaminated matrices (e.g. soil, litter or plant surfaces) involves exposure to potentially lower residue levels, as concentrations decline through degradation and dissipation over time. However, such exposures can occur over extended periods because residues persist in these matrices and remain available for contact. In the conceptual model, both exposure routes appear under ‘contact’.

Dietary (oral) exposure, via consumption of contaminated food sources including sprayed vegetation, prey items, nectar, pollen or honeydew, was also highlighted as a significant route in EFSA PPR Panel (2015).

The ERA Working Group agreed to include contact (physical) from direct overspray and oral (dietary) exposure via contaminated matrices in the revised Guidance Document, in addition to the exposure via contact with contaminated matrices, which is already considered in the current scheme. Details on how these routes will also be incorporated in the Tier 1 exposure assessments are provided in Section 5.3.1.5. The ERA Working Group will further consider the combined oral and contact exposure routes (e.g. endoparasitoid larvae developing in contaminated hosts) during the implementation of the protocol.

5.3.1.5. Exposure quantification (sub‐question 1.5)

The EFSA PPR Panel (2015) Scientific Opinion did not propose an operational exposure model, but it provided a set of quantitative theoretical considerations for the different exposure routes and scenarios within a tiered framework. As combining these routes into one overall risk assessment was not possible, the Panel recommended assessing all exposure routes individually. The ERA Working Group may wish to take these theoretical considerations into account when developing exposure models for the agreed routes of exposure.

Proposed exposure models for the revised guidance document for non‐target arthropods

The exposure routes considered relevant in EFSA PPR Panel (2015), i.e. contact via direct overspray or through contaminated matrices and dietary (oral), will be incorporated into the exposure assessments of the revised Guidance Document (Table 6). While some of these exposure routes are already covered in the current risk assessment, the exposure estimates will be revised and refined (e.g. contact with contaminated matrices using updated models and data).

As presented in Section 5.3.1.4, exposure scenarios have been aligned to those for the risk assessment of bees according to EFSA (2013, 2023a), i.e. treated field, weeds in the treated field, field margin, adjacent crop and succeeding crop.

In the context of the revision of the Guidance Document, the following sections will set out the approaches proposed by the WG for the development of different exposure models.

Contact exposure from environmental matrices containing PPP residues in in‐field areas (treated field and weeds in the treated field scenarios)

Contact exposure of NTAs to dried and fresh PPP residues on contaminated matrices in in‐field areas is determined by the amount of PPP that reaches the relevant environmental matrices: the crop canopy, the canopy of weeds within the field and the soil surface of the treated area. For Tier 1 exposure assessments, conservative assumptions are applied. Each exposure scenario is assessed independently, assuming that the entire application rate reaches the corresponding environmental matrix (crop, plant or soil surface) without accounting for transfer between matrices. This approach reflects the fact that different NTA groups are associated with distinct matrices and ecological niches (e.g. foliage‐dwelling arthropods are primarily exposed via residues on plant, whereas ground‐dwelling arthropods are exposed via residues on soil surfaces). Wash‐off from plant surfaces is not included as a dissipation process for canopy exposure but is accounted for as a pathway contributing to soil exposure. These assumptions provide a worst‐case exposure level and thereby ensure a protective risk assessment.

It should be acknowledged that the direction of spraying (e.g. downward vs. upward/sideways) may influence deposition patterns and thus affect the quantification of exposure and the relevance of the focal species considered. At higher tiers, such aspects could be further differentiated. As outlined in the EFSA PPR Panel (2015), upward or sideways application techniques may lead to more even or two‐sided deposition within the canopy, while downward‐directed sprays concentrate residues on the upper layers. This has implications not only for exposure quantification (e.g. choice of maximum leaf area index (LAI) values) but also for ecological relevance, as some NTAs may primarily inhabit the lower canopy or understory.

Future guidance development could introduce refinements based on spray technique, canopy distribution and habitat use of focal species, to be applied at Tier 2 and higher tiers, in line with the scheme proposed in EFSA PPR Panel (2015).

The following model is proposed for estimating contact exposure from contaminated surfaces in the treated field for both plant‐dwelling and ground‐dwelling NTAs (on the soil surface):

$$\text{PEQcon}=\mathrm{AR}\times \mathrm{MAF}\times \mathrm{BSF}\times \text{EFco}$$

where,

PEQcon is the predicted environmental quantity from contact (μg/individual).

AR is the application rate (g/ha).

MAF is the multiple application factor (unitless). Unlike bees, for which a MAF is not applied in EFSA (2013, 2023a, 2023b), because repeated contact is unlikely, NTAs may be exposed multiple times, since many species remain within or close to treated fields for extended periods. The MAF addresses the realistic scenario in which pesticides are applied several times during a growing season, potentially leading to residue build‐up if dissipation is not completed within the application interval.

Two methodological approaches are available for accounting for multiple applications between the deterministic and the probabilistic approach.

Deterministic approach

This methodology is implemented within the existing terrestrial ecotoxicology guidance (European Commission, 2002), as well as in the birds and mammals guidance (EFSA, 2023b).

It calculates the MAF from the degradation rate constant, the number of applications and the application interval, using the following exponential decay model:

$$\mathrm{MAF}=\left(1–{\mathrm{e}}^{-\mathrm{nki}}\right)/\left(1–{\mathrm{e}}^{-\mathrm{ki}}\right)$$

where:

n = number of applications.

k = degradation rate constant (ln2/DT50).

i = interval between application (days).

This equation assumes first‐order degradation kinetics and accounts for the fact that each successive application adds to the residual amount remaining from previous applications.

MAF values for various DT50: spray interval ratios and up to eight applications are reported in Appendix V of ESCORT 2 (Candolfi et al., 2001). When dissipation data are not available, default values representing the 90th percentile of a database of 32 PPPs (see Gonzalez‐Valero et al., 2000) are applied. Default values differ between soil and leaf substrates, reflecting the generally longer DT50 observed in soil compared to plant surfaces, due to differences in environmental conditions and degradation processes.

In the current Birds and Mammals guidance (EFSA, 2023b), default DT50 values are applied: 10 days for plant matrices in screening and Tier 1 assessments.

Probabilistic approach

The EFSA bee Guidance Document (2023a) employs a different methodology for accounting for multiple applications in dietary exposure assessment. This approach uses stochastic residue per unit dose (RUD) values combined with DT50 values and parameters for number of applications (n), interval (i) and application window (w), as described in section 5.3.4 of the Supplementary Document of the EFSA (2023a) bee Guidance Document, it represents a potentially more robust approach than the traditional MAF calculation. Current default DT50 values in the Bee Guidance are 10 days for plant matrices, 2 days for pollen and 3 days for nectar.

The ERA Working Group will decide which of the two approaches will be used in the exposure model.

BSF is the NTA body surface factor (cm²/individual), representing the contact area available for exposure to PPP residues. This parameter requires morphological data on arthropod body dimensions and surface characteristics. Data collection is underway through the AENEAS project, which will provide initial morphological measurements for key NTA species. However, further data will be needed for broader taxonomic coverage. Morphological data will be obtained through systematic literature review (e.g. taxonomic publications, morphometric studies, ecotoxicological test guidelines, regulatory dossiers) and accessible databases such as the Global Biodiversity Information Facility (GBIF), museum collections and specialised arthropod repositories. Derivation of generalised BSF which would allow extrapolation across species will also be evaluated, based on available data. Pending availability of sufficient BSF data, default conservative values might be defined for Tier 1.

EFco is the contact exposure factor (unitless). Its role is to account for the differences in exposure across scenarios. EFco values are derived from deposition factors. For the weed scenario, the deposition factor is linked to crop interception and therefore depends on the crop growth stage. The values proposed in Appendix B of the EFSA (2023a, 2023b); Bee Guidance Document for this scenario can also be applied to NTAs.

DT₅₀ data collection to support parameterisation ⁸

Regardless of which methodological approach is selected for MAF estimation, both methods would benefit from refined, substance‐specific and matrix‐specific DT50 values to replace the current default values. To support this evaluation, DT50 values are available from several existing sources:

• EFSA (2023a, 2023b) bee Guidance Document. It contains DT₅₀ values for pollen and nectar (Annex H of the Supplementary document) and a data analysis for DT₅₀ values for other environmental matrices such as plant tissues (section 5.3.10 of the Supplementary document). Moreover, it contains residue data for different plant matrices.

• EFSA (2023a, 2023b) birds and mammals Guidance Document. It references residue decline databases for different food matrices (including plant materials and arthropods). The databases cited in Appendix K, i.e. Lewis and Tzilivakis (2017), Ebeling and Wang (2018) and Lahr et al., (2018), will be considered for the data collection.

• Appendix F of the External report of EFSA (Benaki Phytopathological Institute et al., 2017): Collection and analysis of pesticide residue data for pollen and nectar; It contains DT₅₀ values for pollen and nectar and other environmental matrices. Moreover, it contains residue data for different plant matrices.

• Brewer et al. (2025) A global nectar and pollen pesticide residue database, with an interface tool for calculating residue per unit dose for different pesticide application methods.

It is expected that these existing datasets will provide sufficient information for parameter derivation. The available data will be re‐analysed to determine appropriate default DT₅₀ values following the grouping approach defined in Annexes of both the EFSA (2023a, 2023b) bee and birds and mammals Guidance Documents, where broad groups of matrices have been established for DT₅₀ derivation. Extrapolation within these groups is considered acceptable, but not between groups. It should be noted that these defined groups do not cover all relevant matrices for the assessment proposed in the protocol and future guidance for NTAs. Additional data collection will be considered if gaps remain after this analysis.

This model aligns with the current exposure assessment approach but differs slightly from the methodology described in the EFSA PPR Panel (2015) for the in‐field, on‐crop scenario, which included consideration of the LAI. In the Scientific Opinion, for Tier 1 assessments, a default LAI value of 1 was recommended, corresponding to canopy coverage (i.e. total leaf surface area equals ground surface area). Under this assumption, all NTAs are considered equally exposed to the dose deposited on the uppermost vegetation layer, regardless of their actual position within the canopy. Consequently, applying a LAI of 1 renders the exposure estimate equivalent to the simplified Tier 1 model currently used and proposed above.

The ERA Working Group might consider whether a specific exposure scenario for weeds in the treated field needs to be developed for this exposure route, or whether this is already adequately covered by the proposed scenario of the treated crop.

Contact exposure from environmental matrices containing PPP residues in off‐field areas

Contact exposure of NTAs to PPPs in off‐field areas is primarily driven by the drift movement of PPP residues – comprising spray drift and dust drift – that reaches off‐field habitats (i.e. field margin and adjacent crop scenarios), followed by the deposition of these residues onto relevant environmental matrices: the plant canopy in the off‐field, on‐plant scenario and the soil surface in the off‐field, on‐soil scenario. Drift is typically expressed as a percentage of the in‐field application rate.

Spray drift

Spray drift deposition varies depending on crop category and decreases exponentially with increasing distance from the treated field. Typically, the overall 90th percentile drift data is used to estimate off‐field drift deposition values (% drift). This approach was based on the recommendations of the FOCUS surface water group, which has defined a reasonable worst‐case scenario for drift estimation as being represented by an overall 90th percentile probability.

In this respect, while default drift values (e.g. Rautmann et al., 2001) are still used in regulatory assessment, they are acknowledged to be outdated (Adriaanse et al., 2022). As part of the activities under PERA (FPA EUBA/EFSA/PREV/2023/01), the selected consortium will carry out:

• A systematic literature review aimed at gathering and collating data to describe environmental exposure across different application techniques and environmental matrices (results expected by the end of 2025) and

• New field studies designed to refine drift values and interception data for 2‐D and 3‐D vegetation structures (data expected by the end of 2026).

Existing drift data will be reviewed and integrated with new information/data from PERA by the ERA Working Group. On this basis, the Working Group will identify the most appropriate realistic worst‐case deposition values. Once consolidated, these spray and particulate drift values will be used to derive environmental exposure estimates by multiplying them by the application rate.

Dust drift

Dust drift represents another important exposure pathway, referring to the aerial transport of solid particles such as abraded dust from treated seeds or granules released during the application of solid formulations. Although empirical data remain limited, the relevance of this route has been recognised in previous guidance documents (EFSA, 2013, 2023a), where parameterisation for some crops/crop categories has been proposed. Similarly to spray drift, particulate drift is expressed as % of the in‐field application rate, allowing for analogous integration into exposure models. It is noted, however, that the crop categories are not the same as for spray drift. Additionally, a draft Guidance Document specific for seed treatment applications is currently under preparation by SANCO (European Commission, 2012). In this document, dust drift proportion is not linked to the in‐field application rate, rather to other aspects. While the document is still a draft, its potential implications are acknowledged. Nevertheless, the ERA Working Group will follow regulatory and scientific developments in this area and will incorporate updated approaches as appropriate.

Proposed exposure model

For Tier 1 exposure assessments, conservative assumptions are applied. No consideration of potential dissipation processes that may reduce residues in the assessed matrix, such as wash‐off from plant surfaces in the on‐plant scenario, degradation, volatilisation from the receiving surface or redistribution to other environmental compartments, is included. These assumptions allow for a worst‐case exposure level and thereby support a protective risk assessment. This does not refer to volatilisation from the treated area followed by atmospheric transport and re‐deposition onto off‐field areas, which is considered separately as a potential off‐field exposure pathway where relevant. The following model is proposed for estimating contact exposure in off‐field areas:

$$\text{PEQcon}=\mathrm{AR}\times \mathrm{MAF}\times \mathrm{BSF}\times \text{EFco}$$

where,

PEQcon is the predicted environmental quantity from contact (μg/individual).

AR is the application rate (g/ha).

MAF is the multiple application factor (unitless); see above for further explanation.

BSF is the arthropod body surface factor (cm²/individual).

EFco is the contact exposure factor (unitless). The role of the EFco parameter is to quantify the differences in the exposure via the differentdifferent scenarios. The parameters for EFco are derived from deposition factors. The deposition to the field margin as well as to the adjacent crop is related to the spray drift/dust drift.

The EFco parameter represents the proportion of the application rate that deposits in off‐field areas due to drift. As drift percentages are crop category and crop growth dependent, separate EFco values must be derived for each crop group and grow stage (i.e. crop height). Currently, the ESCORT 2 (Candolfi et al., 2001) approach applies a single drift value per crop category at a given distance from the field (e.g. 2.77% at 1 m for field crops). For dust drift, EFSA (2013, 2023a, 2023b), uses average deposition values reflecting the specific dimensions of the assessed scenarios. However, single percentage values at fixed distances could also be derived for dust drift, facilitating integration into the exposure model.

Depending on how MAF is calculated, either this parameter or the EFco will need to be adjusted in case of multiple applications to target the overall 90th percentile of the exposure.

Two options will be considered for the development of the revised NTA Guidance Document:

1. Fixed deposition values per crop category, crop growth stage and spray direction. As in ESCORT 2, deposition values (expressed as a percentage of the in‐field application rate) at fixed distances from the field edge will be defined per crop category and growth stage to characterise off‐field exposure.

2. Development of standardised Tier 1 off‐field scenarios. These will reflect not only crop category and crop growth stage, but also spatial characteristics such as distance from the treated field, habitat width and shape. One representative scenario per combination of crop category and growth stage is foreseen, though variation between regulatory zones may also be taken into account if significant. The outcomes of the EESE project will be considered during the development of these scenarios. Once established, corresponding deposition values (e.g. 90th percentile) will need to be agreed upon.

For both approaches, the ERA Working Group is expected to define deposition values that reflect realistic worst‐case conditions for each off‐field scenario. Figure 7 includes illustrative examples about possible considerations.

FIGURE 7.

The left sketch illustrates the current exposure characterisation according to European Commission (2002). Deposition is expressed as the quantity reaching a certain distance from the edge of the treated field. In this example (field crop category), a distance of 1 metre corresponds to a deposition of 2.77% of the application rate. The right sketch is an illustrative example of a two‐dimensional off‐field scenario (field margin, FM) with a width of 1 metre, located 0.1 metres (10 cm) from the downwind side of the treated field (i.e. adjacent to the last crop row). The percentage values shown represent some statistical deposition estimates from a spray application to a field crop. In this example, the respective values for the edges of the field margin and the average over its one metre width are indicated (Rautmann et al., 2001). Alternative statistical values could also be derived.

Vegetation distribution factor

The current risk assessment for NTAs (European Commission, 2002) uses a vegetation distribution factor (VDF). The purpose of using a VDF is to relate results of lower tier effect studies to the ‘real’ off‐field environment where (leaf‐dwelling) organisms are assumed to be less exposed because of a different vegetation structure and a larger dilution of exposure than in the in‐field. In current practice, the VDF is set to 10 by default when the off‐field risk assessment is based on toxicity endpoints obtained in a test design with two‐dimensional exposure (where organisms are exposed on flat surfaces like glass plates or leaf discs). For toxicity endpoints obtained in a test design with three‐dimensional exposure (where organisms are exposed on whole sprayed plants), the VDF is 1.

The terrestrial ecotoxicology guidance document (European Commission, 2002) considers this figure unreliable and states that more data should be used as soon as they become available. Several reviews of the VDF value and attempts to derive an appropriate default figure for the VDF are available (see Appendix E of the EFSA PPR Panel (2015) for further details).

In EFSA PPR Panel (2015), it was proposed not to use a VDF for the exposure assessment in Tier 1 and to adjust the exposure at higher tiers to account for the distribution of the substance over the leaves by using a so‐called deposition distribution factor (DDF).

At the general experts meeting in ecotoxicology in 2017, some Member States proposed two options: (i) not to use a VDF at Tier 1 or (ii) to use a VDF of 3–5 at higher tiers (EFSA Supporting Publication, 2019).

The need for a VDF is closely linked to how spray drift deposition on off‐field vegetation is characterised. The ongoing PERA project (FPA EUBA/EFSA/PREV/2023/01) is generating data on spray deposition that will help refine drift input values for three‐dimensional vegetation structures. Once this data become available, they will inform whether a separate empirical VDF remains necessary or whether its underlying processes (i.e. interception and distribution of drift by off‐field vegetation) are adequately represented through improved drift deposition modelling.

If, following the completion of the PERA project, the ERA Working Group determines that a VDF is still needed, the following methodology could be implemented to develop a scientifically sound VDF based on gathering existing data from multiple sources:

• Gathering PPP residue studies and comparing measured ground depositions on 2‐dimensional surfaces (e.g. Petri dishes) and depositions on 3‐dimensional surfaces (e.g. canopy). This comparison would provide direct evidence of how vegetation structure affects actual exposure patterns.

• Analysing regulatory toxicity data: In regulatory practice, ‘2D’ extended laboratory tests (exposure on sprayed glass plates or leaf disks), as well as ‘3D’ extended laboratory tests (sprayed whole plants), are performed. Assuming direct proportionality between the LR₅₀ values and the exposure to PPPs in these tests conducted during regulatory studies, the ratio between the toxicity in those studies could give additional information on the numeric value of a VDF. To derive an estimate for a VDF, LR₅₀ values published in Draft Assessment Reports and Renewal Assessment Reports and EFSA Conclusions for active substances could be compared between test types for the same species, as detailed in Appendix E of EFSA PPR Panel (2015).

Contact via direct overspray in in‐field areas

Direct overspray of NTAs occurs when spray droplets from PPP applications come into direct contact with arthropods present in the field during the application event. This exposure pathway represents the most immediate contact scenario, occurring prior to any interaction with contaminated environmental matrices such as plant surfaces or soil.

This route of exposure is not considered in the current risk assessment, but it is proposed in the EFSA PPR Panel (2015) Scientific Opinion and has been prioritised by the ERA Working Group (see Sections 5.2.1.3 and 5.3.1.3) to be incorporated into the Tier 1 exposure assessment.

It is assumed that NTA are present and fully exposed during application, with 100% spray interception efficiency and complete transfer of intercepted material to the arthropod body. No consideration is given to potential exposure reduction factors such as behavioural avoidance responses, physical protection from crop canopy or microhabitat features, spatial heterogeneity in spray deposition or arthropod escape behaviours. The model employs deliberately conservative assumptions regarding arthropod presence, spray interception and transfer efficiency to account for the inherent uncertainties in exposure estimation. These assumptions allow for a worst‐case exposure level and provide a protective screening assessment.

The following model is proposed for estimating contact exposure in in‐field areas:

$$\text{PEQove}=\mathrm{AR}\times \mathrm{MAF}\times \mathrm{BSF}\times \mathrm{IE}$$

where:

PEQove is the predicted environmental quantity from direct overspray (μg/individual);

MAF is the multiple application factor (unitless);

AR is the application rate converted to surface density (μg/cm²);

BSF is the arthropod body surface factor (cm²/individual);

IE is the interception efficiency (unitless, conservatively set to 1; to be refined at higher tiers).

This model assumes that the entire spray flux directed towards the NTA cross‐sectional area is intercepted and retained, with no losses due to aerodynamic deflection, droplet bounce‐off or incomplete surface coverage.

It is proposed that at least two generic model species ⁹ be defined for exposure assessment in Tier 1 direct overspray exposure assessments: one representing leaf‐dwelling and one representing soil surface‐dwelling NTAs. These represent the main microhabitats where contact with spray droplets is most likely. Each generic model species should be parameterised to represent conservative exposure scenarios that account for the range of sizes and surface area characteristics found within its respective habitat type.

Contact via direct overspray in off‐field areas

For off‐field areas, drift values should be applied to the exposure model presented above. Therefore,

$$\text{PEQove},\mathrm{off}-\text{field}=\mathrm{AR}\times \mathrm{MAF}\times \mathrm{BSF}\times \mathrm{IE}\times \text{EFove}$$

where:

PEQove is the predicted environmental quantity from direct overspray (μg/individual);

AR is the application rate converted to surface density (μg/cm²);

MAF is the multiple application factor (unitless);

BSF is the arthropod body surface factor (cm²/individual);

IE is the interception efficiency (unitless, conservatively set to 1; to be refined at higher tiers).

EFove is the overspray exposure factor (unitless). It accounts for differences in exposure across scenarios. Deposition to the field margin and adjacent crop is determined by spray drift and dust drift processes, and the EFove values are derived from the corresponding deposition factors.

The ERA Working Group will determine whether the same generic model species defined for in‐field assessments should be applied to off‐field scenarios, or whether different model species should be defined to better reflect the arthropod communities and ecological characteristics typical of off‐field environments.

Dietary (oral) exposure parameters/factors

Dietary exposure assumes that NTAs come into contact with PPPs through the consumption of food sources or food items containing residues (e.g. plant leaves and tissues, nectar, pollen, prey/hosts, guttation fluid). The degree of exposure varies depending on the feeding habits of the NTA.

At this stage, it was considered premature to prescribe a fixed exposure model, therefore defined equations are not proposed in the protocol for the treated crop, weeds in the treated field, field margin and adjacent crop scenario. Instead, it identifies key parameters or factors needing data collection that the ERA Working Group will consider when developing the dietary exposure model during protocol implementation, based on available data and practical feasibility.

• Spray drift: Use of updated values from ongoing projects (PERA FPA EUBA/EFSA/PREV/2023/01) (see Section 5.3.1.1).

• Crop interception: In the context of the revision of the bee Guidance Document (EFSA, 2023a), crop interception values were updated. Crop interception is used to estimate spray deposition on weeds in the treated field. Since deposition values are available only for a limited number of crops/crop categories, other crops have to be grouped with the existing categories. It is proposed to use the same category groups as in the revised EFSA bee Guidance Document (EFSA, 2023a).

• Residue Unit Dose: RUD values for different matrices (e.g. pollen, nectar, arthropods, grasses) are available in the EFSA birds and mammals (EFSA, 2023a) and bee Guidance Documents (EFSA, 2013, 2023a) and will be used when relevant. RUD values for relevant environmental matrices not available in both Guidance Documents will need to be obtained from different sources, and a data collection will be performed accordingly. In this context, results from ongoing projects such as AENEAS may provide additional insights and data to inform the selection of appropriate RUD values once finalised.

• Dust formation factor: It is proposed to apply a factor of 0.1 in all cases, expressing the conservative assumption that 10% of the granules will form dust, as proposed in the revised EFSA bee Guidance Document (EFSA, 2023a). Therefore, no additional evidence/data will be collected.

• Sugar content of nectar: It is proposed to use the existing data on the sugar content of crops grown in the EU that were collected in the context of the revision of the bee Guidance Document (EFSA, 2023a). Considering the collected data, the crops were allocated into four sugar content categories, and the sugar content values to be considered for the risk assessments were defined.

• Food consumption and diet composition of NTAs: A systematic review will be conducted to review the available data from the scientific literature on the consumption rates of the relevant environmental matrices by NTAs belonging to different functional groups/feeding guilds. This review will serve: (i) to inform the selection of generic model species with relatively high energy requirements per unit body weight for each functional group, following EFSA PPR Panel (2015) recommendations and (ii) to parameterise consumption rates for dietary exposure modelling.

• Proportion of diet: Describes the proportion of each food item containing PPP residues. To be estimated or assumed based on species functional group/feeding guild, from the systematic review on food consumption and diet composition.

It is noted that the dietary exposure parameters listed above do not currently account for the potential bioaccumulation of active substances in prey organisms. For predatory NTAs, consumption of contaminated prey items that have bioaccumulated residues from their own food source could result in exposure concentrations exceeding those estimated from primary matrix residues alone. The ERA Working Group will consider whether this represents a relevant exposure gap that needs to be addressed in the dietary exposure model for predatory NTAs during guidance development.

Functional groups

Separate dietary exposure assessments will be conducted for different functional groups of NTAs:

• Herbivores consuming contaminated plant leaves and tissues

• Predators and parasitoids consuming contaminated prey or hosts

• Pollinators (excluding bees) consuming contaminated nectar and pollen

• Other groups e.g. decomposers and scavengers

The final list of functional groups to be assessed will be defined by the ERA Working Group, based on ecological relevance, exposure potential and data availability. For each functional group, it is recommended to define generic model species (GMS), i.e. species with relatively high energy requirements per unit body weight, in line with the recommendations of the EFSA PPR Panel (2015) Scientific Opinion. The selection of GMS should also consider species traits that are likely to affect exposure vulnerability, such as foraging behaviour, habitat use, dietary preferences and life history characteristics. To support this process, it is proposed to investigate existing databases (e.g. Riedel, Romeis & Meisslle, 2015) and information from ongoing projects, such as AENEAS.

Exposure scenarios

For the Tier 1 dietary exposure assessment, separate exposure calculations are proposed for the following scenarios:

• In‐field scenarios

  • ○
    Treated crop
  • ○
    Weeds in the treated crop
  • ○
    Succeeding crop
• Off‐field scenarios

  • ○
    Field margin
  • ○
    Adjacent crop

Succeeding crop scenario

A separate exposure model will be developed for succeeding crop scenarios, where NTAs may be exposed to plant‐mobile, soil persistent PPPs taken up and translocated into the new crop.

Proposed model:

$$\text{PEQdi}=\text{PECpw}\times \text{CMPm}$$

where:

PEQdi is the predicted environmental quantity by diet (μg/individual or μg/immature stage/developmental period or μg/individual‐immature stage/day);

PECpw is the predicted environmental concentration in pore water (mg/L = mg/kg);

CMPm is the consumption of plant tissues by the NTA (mg/individual or mg/immature stage/individual, developmental period or μg/individual‐immature stage/day).

It is noted that the CMPm parameter is most appropriate for chewing herbivores and may not adequately represent exposure of sap‐feeding insects, which ingest vascular fluids rather than solid plant tissues. The ERA Working Group will consider the need to develop an appropriate exposure metric for sap‐feeding herbivores.

This assumes that plant uptake is driven by pore water concentrations and that consumption is proportional to the internal distribution of the substance in plant tissues (EFSA, 2023a).

Plant tissue consumption by NTAs will be estimated from the systematic literature review mentioned above.

Dietary exposure via consumption of guttation water

The ecological relevance of guttation water as a dietary exposure route is currently uncertain. The ERA Working Group will first determine whether existing data are sufficient to assess its relevance and, if so, whether consumption by NTAs can be quantified. A two‐step approach is proposed:

1. Narrative review to determine the ecological relevance of guttation water as a dietary exposure route and whether data sources are likely sufficient to support quantification.

2. If the narrative review demonstrates sufficient evidence, conduct a systematic literature review to estimate consumption values for relevant NTA groups.

Because operationalisation is uncertain at this stage, the protocol does not present a mandatory exposure equation in the main text. If the narrative review demonstrates sufficient data and the ERA Working Group decides to proceed, an exposure model will be developed. The ERA Working Group may consider the exposure model for dietary uptake of guttation water developed for bees in the EFSA (2023a, 2023b) Bee Guidance Document, assessing whether its assumptions and parameters can be adapted for NTAs.

It is proposed to conduct a systematic literature review to estimate consumption values if the narrative review determines that guttation water is considered an ecologically relevant exposure route, and sufficient data are available to quantify it.

Crop attractiveness for non‐bee pollinator NTAs

The ERA Working Group will first determine in a narrative review whether existing data are sufficient to assess crop attractiveness for pollen and nectar collection by pollinators other than bees.

If the narrative review demonstrates sufficient evidence, then a systematic literature review will be conducted. The methodology already used in the EFSA (2023a) bee Guidance Document will be considered.

Additional considerations

All the considerations made above apply to both active substances and degradation products. Regarding metabolite (endogenous), data on metabolism across insect species may be collected to understand how processes leading to metabolism are conserved across insect species. This could allow to understand if endogenous metabolites need separate considerations or can be considered addressed by the risk assessment of the parent compound. Regarding metabolites and degradation products formed in plants, the approach currently used for bees will be considered by the ERA Working Group during the development of the revised guidance.

In addition, the ERA Working Group, pending on the data availability, might decide to include a screening exposure assessment based on worst‐case assumptions, followed by a Tier 1 assessment incorporating more realistic conditions – such as deposition, crop interception and dissipation. This approach would also need to be consistently applied across the other exposure scenarios.

5.3.1.6. Overview of methods for exposure sub‐questions

Table 7 summarises the main outcomes of sub‐questions 1–5, highlighting gaps, relative priorities and methods for addressing those gaps.

TABLE 7.

Overview of relative priority and methods for addressing sub‐questions related to exposure of non‐target arthropods. In bold activities that should be conducted to gather additional data. In bold, activities that should be conducted to gather additional data.

Sub‐question (SQ)		Priority	Method
SQ 1.1 – Exposure pathways	Foliar uptake (treated crop)	Medium	Systematic literature review and experimental work under PERA FPA EUBA/EFSA/PREV/2023/01
	Root uptake (treated crop, spray formulations)	Medium a	Systematic literature review and experimental work under PERA FPA EUBA/EFSA/PREV/2023/01
	Root uptake (treated crop, solid formulations)	Medium	Systematic literature review and experimental work under PERA FPA EUBA/EFSA/PREV/2023/01
	Root uptake (succeeding crop)	Medium	Systematic literature review and Experimental work under PERA FPA EUBA/EFSA/PREV/2023/01
	Volatilisation and re‐deposition	Low	No data gathering needed
	Sedimentary/airborne/dust drift	High	Systematic literature review and experimental work under PERA FPA EUBA/EFSA/PREV/2023/01
	Surface run‐off/erosion	Low	No data gathering needed
SQ 1.2 – Environmental matrices	Plant and soil surface	High‐	No data gathering needed, since already covered in the existing guidance. If needed, available residue data submitted in dossiers will be re‐analysed.
	Prey/host food items	High	Available residue data on insects used in the EFSA (2023) birds and mammals and from AENEAS will checked and used, as needed. If sufficient data is not available, additional data will be collected.
	Plant tissues	High	Available residue data on plant tissues from dossiers and existing database for residue in pollen and nectars will be checked and used, as needed. Moreover, residue data used in the context of the revision of the Birds and Mammals (EFSA, 2023b). Guidance will be used.
	Pollen and nectar	High	Available data used in the EFSA bee Guidance Document (EFSA, 2023b) will be checked and used, as needed
	Honeydew, contaminated water (including guttation fluid) and extrafloral nectar	Medium	Narrative literature review and WG expert judgement
SQ 1.3 – Exposure scenarios	Treated field	High	Once data are gathered as outlined for SQ 1.1 and SQ 1.2, an exposure model will be developed based on ERA Working Group judgement and expertise
	Weeds in the treated field
	Succeeding crop
	Field margin
	Adjacent crop
SQ 1.4 – Exposure routes & SQ 1.5 – Exposure quantification	Contact from residues in environmental relevant matrices	High	Body surface factor (BSF) Data from AENEAS project complemented by a systematic literature review, if needed. Multiple application factor DT50 data collection from different sources (see Section 5.3.1.5) to support parameterisation Vegetation distribution factor Analysis of data from dossiers (residue studies and 2‐D and 3‐D toxicity studies with NTAs) and data obtained from PERA (FPA EUBA/EFSA/PREV/2023/01)
	Contact via overspray	High	Selection of generic model species for exposure assessment Formal expert judgement although based on BSF data from AENEAS project complemented by a systematic literature review, if needed.
	Dietary	High	Selection of model species Through formal expert judgement although based on existing evidence (e.g. Riedel et al. (2017) database) and information from AENEAS Food consumption and diet composition A systematic literature review will be performed for different feeding guilds

^a Pre‐emergence applications only.

5.3.2. Hazard identification and characterisation sub‐questions

5.3.2.1. Current test methods and limitations (sub‐questions 2.1, 2.2 and 2.3)

As reported in Section 5.2.2, a number of limitations have been identified regarding the available test methods. It should be noted that in the currently available standardised test guidelines, NTAs are only exposed to freshly dried residues. In these guidelines, the endpoints are expressed in mass a.s. or PPP/ha. Overspray test guidelines do exist for surface‐dwelling species (Pardosa sp. and Poecilus cupreus; Candolfi et al., 2001), but no test guidelines are currently available in which foliar NTAs are oversprayed with PPP. Similarly, no test guidelines exist in which NTAs are fed with contaminated food items (see also Section 5.3.2.1 for a broader overview of current test methods and their limitations). The ERA Working Group will explore whether the systematic literature review described in that section could help identify suitable test guidelines or alternative test systems to address hazard characterisation for these additional exposure routes. To address the limitations indicated in Section 5.2.2 and update the knowledge base, a two‐step approach is recommended:

1. Systematic literature review to identify new test methods, evaluate advances in NTA protocols and compile an updated inventory addressing the gaps identified in the 2015 assessment. The review will also aim to capture:

   1. Protocols and methods covering different life stages (e.g. egg, all immature stages including larval instars, pupal for holometabolous species, nymphal instar including as neanids for heterometabolous species, adult);
   2. Varying exposure periods;
   3. Different exposure routes, e.g. overspray and dietary;
   4. Approaches for assessing substances with non‐standard modes of action (e.g. insect growth regulators, physical mode of action);
   5. Toxicity data reporting mortality (LR₅₀) and sublethal effect (ER₅₀) endpoints under different exposure durations;
   6. Information on reproductive sensitivity and its variation with exposure duration and life stage.

The review could be completed by a survey administered to known Contract Research Organisations since they may have developed/adopted new or amended test protocols not yet available in the literature.

• iiBased on the literature review findings, identify the need for standardised testing protocols by evaluating current inconsistencies and gaps in substrate selection, exposure duration, endpoint assessment, life stage selection and suitability of methods for substances with non‐standard modes of action. Where suitable test methods are lacking for relevant species or life stages, the ERA Working Group may propose the development of new protocols to ensure these gaps are addressed.

5.3.2.2. Sensitivity of NTA species (sub‐question 2.1)

To address the gaps identified related to sensitivity of NTA species, the ERA Working Group proposed to assess the sensitivity to PPPs across three hierarchical levels:

1. Developmental difference: sensitivity variation among life stages within a species (e.g. larvae vs. adults).

2. Interspecies differences: sensitivity variation among species within the same taxonomic or functional group.

3. Inter‐group differences: sensitivity variation between major taxonomic and functional groups.

Information on available test methods and identified gaps (see Section 5.3.2.1) will be used to support the mapping of data availability, prioritisation of candidate species and assessment of practical feasibility within the decision framework.

To assess sensitivity differences across these three levels, the following approaches will be implemented (see also Section 5.3.2.1):

1. Consideration of outcomes for ongoing research projects.

AENEAS project:

• Identification of representative taxa linked to ecosystem services, including pest control and pollination, mainly in arable crops.

• Selection of key driver NTA species through vulnerability analysis and synthesis of sensitivity data from literature and regulatory databases.

• Evaluation of interspecies sensitivity across PPPs with different modes of action through laboratory assays.

EESE project:

• Compilation of an ecotoxicological database for NTAs, combining trait and toxicity data from existing sources.

• Ecotoxicity testing of additional NTA species, with emphasis on soil‐surface dwellers and soil‐mediated exposure.

• Analysis of data using SSDs to derive assessment factors (AF).

• Complementary literature‐based analysis (if needed).

• iiResults from the ongoing EFSA‐funded projects will be the main source of data. However, if additional information is required to adequately address sensitivity differences (e.g. species to cover by the above projects), targeted literature searches will be conducted to complement existing findings:

• Conduct systematic review of peer‐reviewed literature reporting quantitative toxicity data for NTAs exposed to PPPs under controlled laboratory conditions, and following the principles of EFSA guidance (EFSA, 2010).

• Investigate regulatory dossiers and relevant databases (e.g. PPDB, ¹⁰ US EPA ECOTOX Knowledgebase, ¹¹ and extract data as needed.

• Consider outcome of ongoing EU funded projects, i.e. PollinERA, ¹² WildPosh, ¹³ EcoStack, ¹⁴ Syberac ¹⁵ ;

• Where datasets are sufficiently comparable and contain adequate replication, perform meta‐analyses to quantitatively synthesise toxicity estimates across studies and derive summary effect sizes. This will allow a quantitative assessment of sensitivity differences between NTA groups and support evidence‐based conclusions.

5.3.2.3. Relevant exposure duration (sub‐question 2.2) and ecotoxicological type of effects and their characterisation (sub‐question 2.3)

Sub‐questions 2.2 and 2.3 are addressed together here because the duration of exposure strongly affects ecologically relevant endpoints, including reproduction, and assessing them jointly allows a coherent evaluation of chronic and sublethal effects in Tier 1 hazard characterisation, including potential consideration of time‐reinforced toxicity (TRT). The concept of TRT, as introduced in the EFSA bee guidance (EFSA, 2023a), will be considered for NTAs. The ERA Working Group will assess whether TRT is scientifically applicable to the relevant NTA test species and whether its incorporation at Tier 1 is feasible, drawing on the outcomes of the systematic literature review described in Section 5.2.1.2.

Most of current NTA ecotoxicity tests are short‐term contact studies in which the treatment exposure typically covers only a limited fraction of a species' life span, which may lead to underestimation of cumulative, delayed or reproductive effects and misrepresentation of potential population‐level impacts. Evidence indicates that PPPs can accumulate in organisms and trigger delayed or secondary effects, such as reduced pupation success or reproductive impairment. Chronic low‐dose exposure to insecticides has also been shown to result in delayed mortality in NTAs. Furthermore, acute lethality alone does not capture the full spectrum of effects influencing NTA populations. Sublethal endpoints, particularly reproduction, are ecologically meaningful: reductions in fecundity, egg viability or offspring development can affect population dynamics and ecosystem services.

To address these limitations, the EFSA PPR Panel (2015) recommended the development of long‐term test designs extending over a substantial part of the species' life history and including endpoints on both survival and reproduction. Species with short generation times and straightforward laboratory maintenance were identified as suitable candidates for multi‐generation testing, efficiently capturing both acute and chronic effects.

Building on these recommendations, the ERA Working Group proposes the following approach for Tier 1 hazard assessment:

• Use findings from the comprehensive systematic literature review: The systematic literature review described in Section 5.3.2.1 will provide the evidence base for understanding how sensitivity varies with exposure duration and life stage, including data on acute, chronic and multi‐generational tests capturing both mortality and reproductive endpoints. If the review and ongoing EFSA‐funded projects (AENEAS, EESE) do not provide sufficient data, targeted supplementary literature searches may be conducted to address specific data gaps. The ERA Working Group will evaluate whether the results of the EESE project on ecotoxicity testing can provide complementary information once the results become available.

• Assessment of current test methods for selected species: Evaluate whether the methods identified in Section 5.3.2.1 adequately cover long‐term and multi‐generational effects, including reproductive endpoints, for the species and life stages prioritised in Section 5.3.2.2. For endoparasitoids, this includes consideration of host‐mediated exposure affecting pre‐imaginal stages (larvae and pupae developing within contaminated hosts), in addition to adult exposure via contact and dietary routes. The ERA Working Group will determine whether and how host‐mediated exposure should be acknowledged or incorporated at Tier 1 to ensure protection of reproductive and developmental effects in parasitoids. Practical feasibility, including ease of laboratory culture and species life span, should be considered when evaluating method applicability.

The combined assessment of exposure duration, reproductive sensitivity and prioritised species and life stages (considering ecological relevance, functional group and practical feasibility, including life span) will inform the development of the Tier 1 risk assessment framework. This ensures that Tier 1 testing requirements are defined consistently, incorporating both acute and sublethal endpoints under ecologically meaningful exposure conditions.

Regarding the exposure estimates, it is proposed not to differentiate between acute and chronic at this stage, since from a hazard point the appropriate exposure duration may be different pending on the tested species.

The operationalisation of chronic and sublethal endpoints, including reproductive and behavioural effects, at Tier 1 is a priority objective of the current revision. The ERA Working Group will work towards their explicit incorporation into the revised guidance, building on the inventory of available test methods described in Section 5.3.2.1 and drawing on the outcomes of ongoing research projects (AENEAS, EESE).

5.3.2.4. Information to be derived from Tier 1 studies (sub‐question 2.4)

The current regulatory framework (European Commission, 2002) bases hazard assessment on single effect point estimates:

• LR₅₀ – the rate (g/ha or L/ha) causing 50% mortality.

• ER₅₀ – the rate (g/ha or L/ha) causing 50% reduction in reproduction.

These values are derived from glass plate studies using two standard species (Candolfi et al., 2001), with LR₅₀ values typically used in Tier 1 risk assessment. Current test guidelines do not provide explicit recommendations on the number of tested rates or on the spacing between rates. While guidance exists on the number of replicates (e.g. T. pyri: five replicates with 20 protonymphs each; A. rhopalosiphi: maximum four replicates with 10 wasps each), the absence of standardised rate selection criteria may compromise the robustness and comparability of LR₅₀/ER₅₀ estimates.

Standard practices based on established OECD ecotoxicity test guidelines typically involve at least five concentrations/rates with two‐fold spacing (1.5–2.2 across guidelines), balancing statistical robustness and practical feasibility.

The ERA Working Group will evaluate whether to maintain reliance on single‐point estimates and which kind of effect level to consider based also on the SPGs (LR₅₀/ER₅₀ or any other LR_x/ER_X) or adopt full dose–response relationships, as recommended in the EFSA Bee Guidance Document (EFSA, 2023a). The dose–response approach offers several advantages, including more robust hazard characterisation and the ability to account for predicted effect levels across a range of exposures.

The following methodology is proposed for the information to be derived from Tier 1 studies: Building on the literature review described in Section 5.3.2.3, LR₅₀ and ER₅₀ data from standard glass plate studies and any equivalent Tier 1 tests will be collated, with attention to study design, replication, rate spacing and data quality. This will focus specifically on evaluating the robustness and comparability of effect estimates to inform the choice between single‐point estimates and dose–response modelling.

Based on this evaluation, the ERA Working Group will determine the optimal approach and propose harmonised criteria for test design (e.g. number of rates, spacing, replication) and, if applicable, guidance on dose–response modelling and integration of equivalent studies.

5.3.2.5. Overview of methods for hazard sub‐questions

The priorities to address the sub‐questions relative to hazard characterisation and the proposed methods are listed in Table 8.

TABLE 8.

Overview of relative priority and methods for addressing sub‐questions related to hazard of non‐target arthropods. In bold, activities that should be conducted to gather additional data.

Sub‐question (SQ)	Priority	Method
SQ 2.1, 2.2, 2.3 ‐ Test methods (Relevant for sub‐questions 2.1, 2.2 & 2.3)	High	Systematic literature review to identify new test methods, evaluate NTA protocol advances and compile updated inventory addressing EFSA PPR Panel (2015) gaps
SQ 2.1 – Sensitivity of NTA species	High	Results from the ongoing EFSA‐funded projects, i.e. AENEAS, EESE, PERA FPA EUBA/EFSA/PREV/2023/01, will be evaluated first. If data are not deemed sufficient, additional complementary steps may be implemented, e.g. complementary and targeted literature reviews as outlined above
SQ 2.2 – Exposure duration	High	Refer to the systematic literature search described above
SQ 2.3 – Ecotoxicological type of effects	High	Lethal and sublethal effect data (including mortality, reproduction and other relevant endpoints) from existing datasets identified through the systematic literature search described above (SQ 2.1, 2.2, 2.3) to develop decision framework for incorporating sublethal endpoints
SQ 2.4 – Information to be derived from Tier 1 studies	High	Evaluation of endpoint methodologies from published Tier 1 studies and dose–response analyses to recommend point estimates vs. dose–response approaches

5.3.3. Risk characterisation sub‐questions

5.3.3.1. Linking the predicted exposure with the effect as characterised for each scenario and risk case

As indicated in Section 5.2.1.4, three different exposure routes will be considered in the revised Guidance Document.

The hazard characterisation for NTAs will need to consider the different exposure routes (contact with residues, direct overspray and oral exposure). Since the proposed exposure models express PEQ values in μg/individual (or μg/individual/day), while the currently available effect endpoints are expressed as mass a.s./ha from residue studies, a consistent approach for linking the two will need to be developed. The ERA Working Group will therefore consider how existing endpoints can be translated into a compatible metric, and whether additional test designs are required to cover all relevant exposure routes. Moreover, based on available hazard methods and biology of tested species, the exposure can be estimated considering the hazard time window so that a clear link between exposure and effects can be established without the need to necessarily differentiate between acute and chronic exposure.

It is noted that some aspects of the exposure estimation are specific for the active substance (e.g. the amount of active substance ingested by the NTA), while for hazard characterisation the formulated product is used in most cases. Linking these two aspects may not be straightforward for PPPs containing more than one active substance. Therefore, the ERA Working Group will develop a specific approach to cover this issue.

5.3.3.2. Methods for assessing the outcome of the risk assessment in relation to the agreed SPGs

Two risk assessment approaches can be applied:

• The single‐endpoint approach

• The combined‐effects approach

The single‐endpoint approach represents the traditional method and is the one currently implemented in the EU ecotoxicological guidance for NTAs (European Commission, 2002). At Tier 1, this approach relies on a hazard quotient, calculated as the ratio of exposure (application or drift rates) to the LR₅₀ values obtained from the two standard species (A. rhopalosiphi and T. pyri). PPPs are considered to pose a potential risk if the hazard quotient is equal to or greater than 2 (trigger value defined in European Commission, 2002). Trigger values should be scientifically underpinned by calibration of lower tier results against higher tier studies, thereby ensuring that the agreed SPGs are consistently implemented at all tiers and an equivalent level of protection across tiers is maintained.

The combined‐effects approach can be estimated by first quantifying the effects of a PPP at the individual level for all relevant exposure routes and risk cases (see Table 9), then extrapolating these outcomes to the population level through appropriate modelling frameworks. In a further step, the effects observed on separate endpoints (e.g. mortality, reproduction) can be aggregated by applying the concept of response addition. This integrated methodology allows for a more realistic assessment of overall effects, since it captures the combined consequences of multiple stressors and endpoints rather than relying on a single measure. Such an approach has been explicitly suggested in the revised EFSA bee guidance (EFSA, 2023a), where individual‐to‐population extrapolation and response addition are key elements for linking predicted effects with the agreed SPGs for bees.

TABLE 9.

Overview of the proposed risk cases for non‐target arthropods.

Risk case	Exposure scenarios	Comments
Contact exposure	Treated crop In‐field weeds Field margin Adjacent crop	Contact exposure from direct overspray and via contaminated matrices should be considered
Oral (dietary) exposure	Treated crop In‐field weeds Field margin Adjacent crop Succeeding crop	Separate risk cases will be defined for each relevant functional group (e.g. predators/parasitoids, pollinators, herbivores)

The choice between the single‐endpoint and combined‐effects approach will be decided by the ERA Working Group. A proposal to support this decision could include:

• Complexity of the exposure scenario – considering the number of relevant endpoints, risk cases and exposure routes. For NTAs, three exposure routes are considered: contact exposure from contaminated matrices, contact exposure from direct overspray and oral (dietary) exposure.

• Availability and quality of data – evaluating whether sufficient and reliable effect data exist for the endpoints and exposure routes required by each approach (e.g. LRx/ERx for the single‐endpoint approach, and multiple individual‐level endpoints for the combined‐effects approach). Assessing whether the combined‐effects approach adds significant value in capturing realistic risks from multiple endpoints or exposure routes, or whether the single‐endpoint provides sufficient protection at Tier 1.

• Availability of data to calibrate the different tiers of the risk assessment to make sure that the SPGs, as decided by risk managers, are consistently implemented at all tiers.

6. IN‐SOIL ORGANISMS

6.1. Gap analysis

Figure 9 gives an overview of the gaps identified in the current risk assessment of in‐soil organisms. Gaps related to exposure characterisation are marled in blue, those related to hazard in green and the ones on risk assessment are labelled in yellow.

FIGURE 9.

Gap analysis for the risk assessment of in‐soil organisms presented using a fish‐bone diagram.

The list of gaps identified are listed below and are already categorised based on the sub‐question they belong to.

Exposure pathways (sub‐question 1.1):

• Sedimentary spray drift

• Dust drift

• Vapour drift

• Run‐off (erosion)

• Release of residues from seed coating

• Granules

• Volatilisation & re‐deposition

Environmental relevant matrix (Sub‐question 1.2)

• Exposure through litter layer

Exposure scenario (sub‐question 1.3)

• Field margin (not specified in Figure 9)

• Adjacent crop (not specified in Figure 9)

Exposure route (sub‐question 1.4)

• Oral

Exposure quantification (sub‐question 1.5)

• Bioavailability of a substance

Sensitivity (sub‐question 2.1)

• Species sensitivity including difference between several species of the same taxon or group and differences between organisms belonging to different taxonomic groups

• Difference in sensitivity of different life stages of the same species

Test methods (sub‐question 2.2, 2.3 and 2.4)

• Information on methods for addressing the effects of PPP on soil microorganisms, covering structural and functional endpoints.

6.2. Risk matrix

The risk matrix (see Figure 10) was used to rank the gaps in a transparent manner and assign them relative priority based on the combination of their relevance and likelihood of occurrence or potential impact (see Section 4).

FIGURE 10.

The diagrams illustrate the overall priority level attributed to each exposure and hazard and risk assessment related gap identified (from low to high, x‐axis). The overall priority level is obtained by combining the score (see footnote 7) attributed to likelihood or impact on risk assessment and relevance. This score is represented by the number of bricks in each colour (i.e. 1 brick = minimal; 2 bricks = moderate, 3 bricks = large). One‐direction error bars, using the same colour coding, reflect the upper limits of the scores whenever their quantification was uncertain, and further evidence is needed to clarify the relevance or likelihood.

Each of the aspects shown in Figure 10 is discussed below in Section 6.2.1 for exposure characterisation and in Section 6.2.2 for hazard and 6.3.3 for risk assessment.

The gap related to new application techniques is relevant for all organisms in the remit of this group and was therefore covered in Section 2.4.1.

6.2.1. Exposure characterisation

6.2.1.1. Exposure pathways (sub‐question 1.1)

As outlined in Figure 10, in addition to the direct deposition into the soil, the following pathways should be considered in Tier 1 risk assessment:

• Sedimentary spray drift

• Dust drift

Both have been assigned high priority, especially when the off‐field scenario will be considered (see Section 4).

In addition, the following pathways also need to be considered further, as they were ranked as having medium or medium to high priority:

• Surface run‐off

• Release from treated seeds

Finally, vapour drift (see Section 2.4.2) was also among the possible exposure pathways leading to exposure of in‐soil organisms to PPPs. The WG agreed to assign to this pathway low priority considering that vapour drift generally results in lower deposition levels and is relevant for only a limited number of substances, i.e. substances with high vapour pressure. Thus, it has overall been considered to have low relevance and low likelihood.

6.2.1.2. Exposure‐relevant environmental matrices (sub‐question 1.2)

In addition to total soil and soil pore water, exposure via the litter layer has been given potentially ¹⁶ high priority. This aspect was prioritised because data are scarce and outdated, and exposure through litter could potentially be higher than exposure through soil, particularly for soil organisms such as detritivores.

6.2.1.3. Exposure scenario (sub‐question 1.3)

The current risk assessment framework under SANCO Guidance (European Commission, 2002) considers only an exposure scenario within the treated area (in‐field). However, the EFSA PPR Panel (2015) Scientific Opinion proposed Specific Protection Goals also for off‐field areas, which the ERA Working Group subsequently endorsed for further development. Therefore, a clear definition of exposure scenarios needs to be developed for both the cropped area (in‐field) and off‐field environments. This development should also include defining the relevant exposure pathways for off‐field situations.

From the perspective of exposure assessment for in‐soil organisms, field margins and adjacent crops are merged into a single off‐field exposure scenario. This decision is justified because, regardless of whether the off‐field zone is cropped or uncropped, the key process driving soil contamination is the same: the deposition of residues via drift and/or run‐off/erosion from the treated field. The main difference lies in the above‐ground vegetation (wild plants vs. cultivated crops), which is not directly relevant for in‐soil exposure. For this reason, keeping a single ‘field margin/adjacent crop’ scenario ensures a simplified, harmonised framework for Tier 1 while still covering the full range of possible off‐field soil contamination situations. Should the need be identified, further differentiation may be introduced at higher tiers.

6.2.1.4. Exposure routes (sub‐question 1.4)

Regarding exposure routes, the ERA Working Group decided to only consider the contact route i.e. absorption through the external surface of the organism of pesticide residues present in the soil, as is currently done.

Although in‐soil organisms may be exposed to pesticides both through contact between the soil/litter and the animal skin and through oral via food (soil/litter)/preys and microorganisms ingestion (for example for earthworms which ingest a high quantity of soil (Medina‐Sausa et al., 2019), for Tier 1 toxicity tests, the contact exposure for the majority of in‐soil organisms is still considered the most relevant for a number of reasons:

• A good correlation between results from field studies and laboratory toxicity studies has been reported in literature (Environment and Climate Change Canada, 2022) for earthworms. Therefore, since in higher tier tests, effects resulting from multiple exposure routes are considered, it can be reasonably concluded that the contact route of exposure assessed in the laboratory studies is somehow sufficiently representative.

• In addition, test methods to assess the effects resulting from contact exposure are available, while attempts to develop test methods for exposure via contaminated food have shown high complexity and a number of drawbacks (Natal‐da‐Luz et al., 2019).

• The relevance of exposure through contaminated food depends on several factors, including the physico‐chemical properties of the substance and its toxicokinetic and toxicodynamic properties. Therefore, for the majority of in‐soil organisms the dietary route is considered to be better addressed in intermediate and higher tier studies and not at Tier 1 (which is the only tier covered by the protocol).

It should also be noted that, specifically in the case of earthworms, a certain level of exposure through soil ingestion may also be assumed in the currently available test methods for hazard characterisation.

The only exception to this overall conclusion is represented by litter fragmenters like isopods or anecic earthworms, who feed on litter. Consequently, considering only contact exposure through soil may significantly underestimate their exposure to PPPs and thus the risk as a result of that exposure. As also reported in the Opinion, the importance of exposure through litter via oral uptake should be further investigated.

The exposure via root uptake might be relevant for arbuscular mycorrhizae (AMF) (hyphae, spores in the soil). PPPs may reach roots via soil solution and be taken up by plants. Once PPPs reach the roots, this may affect AMF structures directly or indirectly by acting on the host plants. However, exposure through soil and soil pore water is generally assumed to be more relevant and often leads to higher exposure levels because (i) extraradical AMF hyphae face higher direct pesticide exposure than intraradical forms; (ii) those extraradical forms are more often studied and considered the most sensitive structures and also considered more sensitive than other endpoints (Hage‐Ahmed et al., 2019). Therefore, although root uptake is considered a relevant exposure route, it has not been prioritised because it is adequately covered by the contact route, which results in worst‐case exposure for the most sensitive organisms (AMF).

As part of the main exposure route that will be addressed in the revised Guidance Document, the bioavailability of PPPs to the organisms should be elucidated and thus has been given high priority. In particular, it was considered important to further understand the relationship (e.g. linear) between substance availability in the soil, taking into account the influence of soil parameters (e.g. organic matter, pH, CEC), its uptake into the organisms and the resulting ecotoxicological response.

According to Regulation (EU) No 284/2013, information on the fate and behaviour of plant protection products (PPPs) may in principle be required with the formulated product. However, such studies are rarely considered necessary. The environmental behaviour of the active substance and its relevant metabolites is comprehensively addressed under Regulation (EU) No 283/2013, and PECs are estimated based on the active substance and metabolite properties. The formulated product is generally not expected to significantly alter the environmental fate and behaviour processes of the active substance, such as degradation, sorption–desorption, volatilisation or distribution between environmental compartments.

For the risk assessment of in‐soil organisms, exposure modelling is generally conducted using active substance data and taking into account the GAP to derive PEC values in the relevant environmental matrices. However, in some cases, exposure estimates may also be expressed in terms of the plant protection product, particularly for formulations containing multiple active substances or co‐formulants of known environmental concern. The use of formulated product data remains primarily relevant in the context of toxicity testing, to capture mixture or formulation‐specific effects that cannot be predicted from active substance data alone. Where the available evidence indicates that the formulation or any of its co‐formulants may substantially influence exposure or toxicity, consideration at the PPP level may be warranted.

6.2.2. Hazard identification and characterisation

6.2.2.1. Species and life stage sensitivity (sub‐question 2.1)

High priority has been given to investigation of species sensitivity within a specific taxon, between taxa of the same group and between different groups. Considering the diversity of soil organisms and since the sensitivity of species may vary to a high degree, it is important to understand it properly for a fit‐for purpose and protective hazard characterisation and consequent risk assessment.

To complement information on species sensitivity, the sensitivity of different life stages within a species will also be explored. Normally early life stages tend to be more sensitive than late life stages (McKim, 1985; Mohammed, 2013; Pelosi et al., 2014). As for the species sensitivity above, this will also be useful for proper hazard and risk characterisation.

6.2.2.2. Ecotoxicological test methods (sub‐question 2.2, 2.3 and 2.4)

Invertebrates

Based on the SPGs selected by risk managers and data to address the gap on species and life stages sensitivity, methods for invertebrate species other than the current standard species may be needed. This could be the case for litter fragmenters (e.g. isopods).

Soil microorganisms

Considering the criticism related to the currently used test methods for investigating the impact of PPPs on soil microorganisms and the proposed SPGs (EFSA, 2025b), an inventory of ecotoxicological methods for microorganisms, including both for structural and functional endpoints, should be prioritised.

6.3. Methods for answering the sub‐questions

Methods for addressing the sub‐questions related to exposure, hazard and risk assessment are outlined in Sections 6.3.1, 6.3.2 and 6.3.3, respectively. An integrated conceptual model (Figure 12) is included at the end of Section 6.

6.3.1. Exposure characterisation sub‐questions

6.3.1.1. Exposure pathways (sub‐question 1.1)

The relevant exposure pathway for in‐soil organisms is the direct deposition of PPP residue onto the soil surface and, where present, onto the plant litter layer, followed by subsequent redistribution within the soil system. Moreover, (i) sedimentary spray drift/dust drift, (ii) surface run‐off/erosion, whereby residues are transported with water and soil particles and deposited on soil or litter in adjacent off‐field areas and (iii) localised release/hotspots from treated seeds into surrounding soil are also potentially relevant pathways in particular situations that are explained in Section 6.3.1.5. Currently, only direct deposition onto the soil surface is addressed in the risk assessment framework. See Section 6.3.1.5 for the description of how data will be gathered for these exposure pathways.

6.3.1.2. Exposure‐relevant environmental matrices (sub‐question 1.2)

The key environmental matrices relevant for the exposure assessment of in‐soil organisms are total soil, soil pore water and, potentially, plant litter and treated seeds. These matrices serve as the primary interfaces through which PPP residues can become accessible to in‐soil organisms via contact or ingestion.

The role of plant litter as an environmental matrix has received less attention but is gaining increasing recognition. An earlier analysis by Beulke et al. (2015) concluded that the stable litter layers in cropped fields are generally limited. However, the EFSA PPR Panel (2017) Scientific Opinion noted that, when present, litter can intercept a significant proportion of the PPP residues before they reach the underlying soil. This interception can result in locally elevated exposure concentrations within the litter layer, which may be particularly relevant for litter‐dwelling organisms and species that forage or burrow through this matrix. Therefore, acknowledging the key ecological importance of the litter layer (EFSA PPR Panel, 2010) and the fact that its presence may be promoted as one of the good agricultural practices, e.g. no tillage, sustaining a good soil quality, the ERA Working Group decided to first update the information on litter layer presence in agricultural areas. This update involves conducting a follow‐up systematic literature search following the methodology of Beulke et al. (2015). The outcome of this literature review will provide further insights into litter layer occurrence and better inform the decision regarding its consideration in risk assessment for all relevant exposure scenarios.

6.3.1.3. Exposure scenarios (sub‐question 1.3)

In the current regulatory risk assessment, conducted according to the SANCO Guidance (European Commission, 2002), only exposure in the cropped area is considered (in‐field assessment). This exposure scenario is currently covered for annual and permanent crops by the EFSA (2017) Guidance Document and accompanying tools PERSAM, PEARL and PELMO.

However, the EFSA PPR Panel (2017) Scientific Opinion proposed SPGs for both in‐field and off‐field areas. In the strategy for the definition of SPGs, the ERA Working Group reproposed the SPGs options as included in the Opinion (EFSA, 2025a). Pending the decision of risk managers, off‐field exposure scenarios may therefore be required. As stated in the EFSA PPR Panel (2017), the potential for external recovery (immigration) may not be of high relevance for many in‐soil organisms that move slowly. However, at a smaller scale, internal recovery via e.g. dispersal from more undisturbed areas, e.g. field edge with a hedgerow or combination of tree rows and crop alleys, to the field may occur (Alvarez et al., 2000; D'Hervilly et al., 2024; Frampton et al., 2007). These data emphasise the need to protect the off‐field area in order to allow recolonisation of the in‐field.

It should be noted that exposure scenarios normally cover a number of factors such as agro‐ecological conditions, agronomic practices, landscape structure and species distribution. All those aspects will be considered while developing scenarios in the context of the EESE project (OC/EFSA/PREV/2023/02), with the aim of maximising the relevance and realism of the PPP risk assessment in the context of EU agriculture. However, keeping in mind the tiered approach, realistic worst‐case scenarios are typically considered in higher tier studies while the Tier 1 is meant to be more conservative and requiring less effort (Boesten et al., 2007).

The following scenarios are proposed:

• Treated crop. Residues may deposit directly on bare soil or accumulate first on litter, which can act as a temporary sink. From there, they may migrate into the soil matrix through leaching, degradation or bioturbation, contributing to exposure via both the total soil, litter layer and pore water phases.

• Field margins/adjacent crop. Soil in off‐field areas may be exposed through sedimentary/dust drift or run‐off/erosion deposition. When vegetation is present in off‐field areas, whether as natural vegetation or as an adjacent crop, it can intercept part of the spray/dust drift or airborne drift, thereby reducing the amount of active substance reaching the soil. Consequently, exposure levels in such vegetated off‐field areas are expected to be lower than in bare or sparsely vegetated margins. Nevertheless, the highest abundance of in‐soil organisms, particularly earthworms, is typically found in grassy field margins, which may still represent a biologically relevant worst‐case scenario where there could also be litter present.

For Tier 1 assessment, the two possible off‐field situations (with or without vegetation) are therefore combined into a single, conservative scenario. If further information becomes available (e.g. from the EESE project) or evidence indicates that vegetation cover significantly modifies exposure, these situations could be differentiated in higher tier assessments.

6.3.1.4. Exposure routes (sub‐question 1.4)

In the current risk assessment, contact is the only route of exposure considered. As reported in Section 6.2.1.4, the ERA Working Group decided to only consider that route in Tier 1 risk assessment. However, pending the literature review on the occurrence of litter and the potential sensitivity of litter‐dwelling organisms, the oral route following litter ingestion may become relevant.

An important aspect related to the exposure route and the way a substance reaches the receptor is its bioavailability. Based on the soil properties and the physico‐chemical properties of a chemical, a number of processes leading to sequestration of part of the chemical may occur, rendering it unavailable to soil organisms (Lanno et al., 2004).

Lipophilic substances (logP > 2) may bind to the organic matter OM of the artificial soil used in laboratory toxicity tests (normally 5% or 10%) and thus be less bioavailable to the organisms. In light of this, the SANCO Guidance (European Commission, 2002), suggested reducing the toxicity endpoint for earthworms by a factor of two. In 2015, the issue was discussed between EFSA and representatives from Member States (EFSA, 2015), due to inconsistent application of the extra assessment factor in the case of in‐soil organisms other than earthworms when soils with less than 10% organic matter are used in the toxicity studies. In general, the factor of two does not seem to be scientifically underpinned, as it implies a linear relationship between the organic matter content and toxicity. Moreover, it is well known that several other parameters (e.g. soil pH and texture) beyond the organic matter can significantly influence bioavailability and thus observed toxicity. To help understand the aspects influencing the bioavailability of PPPs to soil organisms, a systematic literature review complemented by experimental work will be conducted under PERA FPA EUBA/EFSA/PREV/2023/01. The results of the systematic literature review should be available in November 2025. The experimental part will be finalised at the end of 2026.

6.3.1.5. Exposure quantification (sub‐question 1.5)

Table 10 provides an overview of the combination of environmental matrices, and exposure pathways and routes for the two exposure scenarios defined in Section 6.3.1.3.

TABLE 10.

Overview of the identified exposure scenarios for in‐soil organisms.

Exposure scenario	Relevant exposure pathways	Exposure‐relevant environmental matrix	Relevant exposure route
Treated crop	Direct deposition onto soil and/or litter (where present) via spray application	Total soil Soil pore water Litter	Contact a
	Release of residues from seed coating or granules into surrounding soil	Total soil Soil pore water	Contact a
Field margin/adjacent crop	Sedimentary spray drift Dust drift Run‐off/erosion	Total soil Soil pore water Litter	Contact a

^a Oral exposure route is excluded from the time being. However, based on the outcome of ongoing projects and the Systematic Literature Review that will be conducted this may be reconsidered, as needed.

Contact exposure in treated crop

The exposure of total soil and soil pore water for the treated crop scenario, resulting from deposition into the soil, is comprehensively addressed in the EFSA (2017) Guidance Document and supporting tools, i.e. PERSAM, PEARL and PELMO. These models simulate the spatial and temporal distribution of PPP residues in the solid and aqueous soil phases under various environmental conditions, enabling the estimation of bioavailable concentrations in the topsoil layers where most in‐soil organisms reside. PECs are estimated for both total soil and pore water, and are typically estimated at different soil depths: 1, 5, 10 and 20 cm.

For the estimation of the concentration of a PPP in the litter layer, once its relevance has been clarified (see Section 6.2.1.2 above), the EFSA PPR Panel (2017) Scientific Opinion already noted that dedicated models would be needed to fully characterise exposure through this matrix. Nevertheless, in the absence of such models, a pragmatic approach for a simple estimation of the peak concentration could be based on the scenarios for permanent crops, as described in EFSA (2017). This would use the application rate, the crop interception at the time of application and a bulk density of litter assuming an organic matter content of 100% (i.e. 0.126 kg/L) to estimate concentrations in the litter layer. This approach will be further considered by the WG. Moreover, the aforementioned literature search on the relevance of the litter matrix will include information on the actual organic matter content of litter in order to derive a more realistic bulk density, thereby refining the exposure estimates.

Contact exposure from treated seeds and granules

The exposure of in‐soil organisms following application as seed dressing may lead to higher local concentrations than those arising from spray application and subsequent re‐deposition into soil. This is particularly relevant for large seeds, where residues can be concentrated in the small volume of soil surrounding each seed, potentially creating local ‘hotspots’. Currently, no validated harmonised method is available at EU level to fully characterise the soil exposure from treated seeds. The EFSA (2017) Guidance assumes residues are homogeneously distributed in the topsoil and does not account for localised concentrations around individual seeds. At the national level, the Dutch Ctgb (2025, Evaluation Manual ¹⁷ ) has operationalised a method where, for large seeds, PECs are calculated based upon the 5 cm homogeneous layer concept but adapted to a 3‐D spherical volume around seeds. This approach provides conservative local exposure estimates and uses explicit crop‐specific input values (e.g. seed/tuber size, seed/tuber density, sowing rates) for standard scenarios. However, it is acknowledged that such localised PEC estimates represent a conservative simplification and may not fully reflect the spatial and temporal exposure of mobile soil organisms under field conditions. The draft SANCO Guidance on seed treatments (European Commission, 2012) distinguishes between small seeds (< 0.5 cm), for which the PECsoil approach is considered adequate and large seeds (≥ 0.5 cm), where a PEC sphere approach may be used to address potential localised exposure. Although the equations are conceptually similar to those described by the Ctgb, SANCO outlines a straightforward adaptation for both Tier 1 and Tier 2 PECsoil calculations. Considering this, and the fact that SANCO guidance aims at harmonisation across Member States, the WG proposes to follow the forthcoming SANCO document once formally adopted. If this latter is not adopted in time for the revision of the terrestrial guidance in the remit of the protocol, the WG will further consider which approach can be further proposed building from what is already available.

For granules and small seeds (< 0.5 cm), EFSA (2017) indicates that when incorporated evenly into the soil, exposure can be assessed with the same PECsoil approach as for sprays in PERSAM. Where drilling in rows or placement between rows leads to localised concentrations, exposure should instead be averaged over the actual incorporation depth, which is accounted for in PERSAM and numerical models. However, when applications are made directly into the furrow at the time of sowing (e.g. in ridge–furrow systems), the distribution of residues becomes highly non‐uniform. EFSA (2017), Appendix A.4) notes that for granules and treated seeds placed at a specific soil depth, the maximum concentration occurs locally at that depth. These situations may therefore lead to localised hotspots that are not adequately represented by simple averaging and may require refinement at higher tiers or the use of scenario‐specific placement depths (z_nc) in PERSAM or numerical models to better capture local exposure. All these possible options will be further analysed by the WG during the guidance development.

Contact exposure in‐field margin/adjacent crop

Sedimentary spray drift/Dust drift

Exposure in the field margin/adjacent crop is driven by the deposition of PPP residues onto soil or plant litter that reaches off‐field habitats resulting from sedimentary spray drift or dust drift.

See Section 5.3.1.5 regarding spray drift and the activities planned for gathering additional data, i.e. a systematic literature review and experimental work.

Dust drift

Dust drift may represent another possible exposure pathway for the field margin/adjacent crop scenario. See Section 5.3.1.5 regarding dust drift and the activities planned for gathering additional data.

Run‐off/erosion

First, the relevance of run‐off/erosion for off‐field exposure is something that needs to be clarified. To address this, the WG proposes that the contribution of run‐off and erosion to off‐field exposure could be characterised using the FOCUS PRZM model (FOCUS, 2001), which simulates run‐off and erosion losses as a function of chemical properties (e.g. Koc) and environmental conditions. The simulations should be conducted with conservative yet realistic input values, including soil adsorption (Koc/KFoc), normalised degradation half‐lives (DegT₅₀), slope, soil type, crop cover, application timing and rainfall scenarios. By running PRZM under these conditions, losses of the active substance from the treated field can be simulated in a way that captures both the influence of substance properties and local agronomic and environmental factors. The predicted losses (expressed as mass per unit area) can then be translated into deposition onto off‐field habitats. From these values, predicted environmental concentrations (PECs) as a result of run‐off/erosion can be derived for both soil and litter in the off‐field scenario. This step provides a direct link between modelled field run‐off outputs and the compartments relevant for in‐soil organisms, thereby allowing a more robust evaluation of whether run‐off is a significant contributor to the overall exposure compared with spray or dust drift.

If, based on the preliminary investigation, run‐off is confirmed to be relevant and likely, the Opinion (EFSA PPR Panel, 2017) recommends that quantification of exposure via run‐off be based on the modelling approaches developed under the FOCUS surface water guidance (FOCUS, 2001), particularly using PRZM at Step 3, which simulates run‐off and erosion losses as a function of chemical properties (e.g. Koc) and environmental conditions. Although originally developed for aquatic exposure, the models and default values can be used to estimate the fraction of PPP lost from the field via run‐off and its potential deposition onto terrestrial off‐field areas such as vegetated buffer strips (FOCUS, 2007). These areas may function as receiving environments for run‐off residues, thereby becoming relevant for exposure assessment of non‐target in‐soil organisms.

In the absence of appropriate off‐field exposure scenarios that apply to a given percentile of the concentration distribution, it is advised to base the calculation of soil concentrations in the off‐field area on the scenarios for in‐field exposure as described in EFSA (2017). As mentioned in EFSA (2010), the exposure estimate should preferably apply to a given percentile of the concentration distribution (usually the 90th percentile) of the treated fields. Developing an exposure scenario for a given percentile requires simulating the concentration distribution in the entire target area (e.g. EFSA, 2017). The model for simulating this concentration distribution should preferably include all relevant exposure routes (i.e. spray‐drift deposition, vapour‐drift deposition, dust‐drift deposition and surface run‐off). Since such models are not yet available for regulatory purposes at the European level, the simplifying assumption is made that the individual exposure routes can be assessed separately. Results of the different entry routes should then be summed up, which is a conservative assumption because it neglects the different dynamic behaviour of the processes. As described above, volatilisation followed by re‐deposition is overall deprioritised. However, if that should be considered relevant for particular PPPs, exposure can be estimated as described above and summed with the other entry routes in case a model where all relevant exposure routes are combined, is developed and used, in particular as higher tier option.

Regarding litter, exposure can be estimated analogously to the treated crop scenario by considering the appropriate concentration reaching the off‐field areas via spray or run‐off. It should be noted that litter occurrence will be further investigated. This may also mean that information on the relevance of litter ingestion may also become available. If that is the case, the dietary route through litter ingestion will be further considered and similarly identification of generic model species.

All the considerations made above regarding methods for filling identified gaps for exposure‐related sub‐question apply to active substances and degradation products (i.e. metabolites).

6.3.1.6. Overview of methods for exposure sub‐questions

Table 11 summarises the main outcomes of sub‐questions 1–5, highlighting gaps, relative priorities and methods for addressing these gaps.

TABLE 11.

Overview of gaps, relative priority and methods for addressing sub‐questions (SQs) related to exposure of in‐soil organisms. In bold activities that should be conducted to gather additional data.

Sub‐question		Priority	Method
SQ1 – Pathways	Deposition into soil	Already addressed	–
	Sedimentary/particulate spray drift	High	Systematic literature review and Experimental work under contract PERA FPA UBA/EFSA/PREV/2023/01
	Surface run‐off/erosion	Potentially high	Preliminary investigation for its relevance via modelling
	Volatilisation	Low	No need for data gathering
	Release from seeds into surrounding soil	High	No need for data gathering since method is already available, although not currently implemented
SQ2 – Environmental matrix	Total soil	–	Already addressed in the existing risk assessment
	Soil pore water	–	Already addressed in the existing risk assessment
	Plant litter	Potentially high	Systematic literature review to clarify its likelihood
SQ3 – Exposure scenario	Treated crop	–	Already addressed
	Treated seeds	High	Once data are gathered as outlined for SQ1, an exposure model will be developed.
	Field margins/adjacent crop unintentionally contaminated	High	Once data are gathered as outlined for SQ1, an exposure model will be developed.
SQ4 – Exposure route	Contact to soil (homogeneous)	High	Already addressed
	Contact to soil (localised/hotspot)	High	No need for data gathering since method is already available, although not currently implemented
	Dietary through soil ingestion or prey	Low	No need for data gathering
	Dietary through litter ingestion		Data on the litter occurrence will be gathered. Based on the outcome, litter ingestion will be considered.
SQ5 – Exposure quantification	Bioavailability	High	Systematic literature review and Experimental work

6.3.2. Hazard identification and characterisation sub‐questions

6.3.2.1. Species, life stage and method sensitivity (sub‐question 2.1)

Soil invertebrates

In the current risk assessment, according to EC (2002), the hazard to in‐soil organisms is characterised by testing one species of earthworm, one species of collembola and one species of soil mite. It should be noted that an OECD reproduction test guideline also exist for enchytraeids. However, it is well known that a large proportion of terrestrial biodiversity is represented by organisms living below ground whose communities may include several hundred to over 1000 species per site (Römbke et al., 2016). To cover for possible interspecies differences, the risk assessment according to EC (2002) proposes an assessment factor (AF) of 5 which, together with intra‐ and interspecies difference in sensitivity, should also cover uncertainties related to the extrapolation from lab to field situations. In the scientific Opinion (EFSA PPR Panel, 2017), SPGs have been proposed for earthworms, microarthropods, macroarthropods, nematodes and isopods. Therefore, it is crucial to understand the sensitivity of the currently tested species compared to (1) other species of the same group (i.e. Eisenia sp. vs. earthworms and enchytraeids, Folsomia sp. vs. collembola, Hypoaspis sp. vs. soil mites) and (2) other organisms belonging to other groups (e.g. other microarthropods, macroarthropods, nematodes and isopods). In addition, it is important to investigate the sensitivity of early life stages of invertebrates and arthropods dwelling in the soil compared to the juvenile/adult life stages that are normally tested. This analysis should also elucidate the sensitivity of litter fragmenters which feed on litter compared to other in‐soil organisms, as well as compare this route with other routes of exposure in toxicity testing.

Some hazard data on different species of in‐soil organisms was collected and reported in the Scientific Opinion (EFSA PPR Panel, 2017). However, the literature review was not systematic and did not address the sensitivity of early life stages. Therefore, for robust hazard identification and characterisation, further investigation of the sensitivity of soil invertebrates is essential. This will be addressed in the context of the PERA FPA EUBA/EFSA/PREV/2023/01. The final results of the PERA FPA systematic data collection and analysis should be available in the second quarter of 2027. Based on the risk managers' decision as to the SPGs for soil organisms, the results of the data collection on species sensitivity and the inventory of methods available for testing soil invertebrates, the WG will decide whether different or additional species to the current standard species should be tested. It is noted that an inventory of relevant methods was included in Appendix F of the opinion and can be complemented/updated with the results of the aforementioned systematic literature search, as needed.

Soil microorganisms other than fungi

Currently, only a functional test investigating the impact of pesticides on the nitrogen formation rate in the soil, according to OECD TG 216, ¹⁸ is required. This guideline aims to assess overall nitrate production, which combines multiple processes (e.g. ammonification, nitrification, denitrification, immobilisation). One of the main criticisms of this assay is its poor sensitivity, often related to this functional redundancy in nitrogen formation within the test. Nitrification in soil involves many different bacterial groups, which show different levels of diversification and thus of redundancy. Consequently, a lack of effects in a test according to OECD TG 216 does not address the concern that a shift in the soil microbial community is still possible although nitrate levels remain unchanged (Pedrinho et al., 2024; Prosser, 2007; Thiour‐Mauprivez et al., 2019). The SPGs proposed in the opinion and brought forward in the SPG strategy (EFSA, 2025b) by the Panel and its related WG consider the in‐soil organism community as the ecological entity to be protected, and the structure/diversity of that community as the attribute to be protected. In light of this, and pending risk managers' decision, an inventory of methods for testing effects on microorganisms where structural and functional endpoints will be considered, will be assembled and presented. This will be done by checking outcomes of ongoing or recent projects, like Microsoil (Schlich et al., 2025; ARISTO, ¹⁹ DEFRA, 2025). Moreover, the results of an ongoing update of the systematic literature search on the effects of chemicals on soil microorganisms (Puglisi, 2012) under the project CT PLANTS 2025 38 ‐ procedure EOI/EFSA/2022/01 will be screened to support this inventory.

Soil Fungi

As reported in the section above, the current risk assessment and the most up‐to‐date data requirements according to Regulation 283/2014 and 284/2014 only recommend soil nitrogen transformation rate as the sole test to assess the potential effects of pesticides on soil microorganisms. The primary focus of that test is the assessment of the impact on nitrogen transformation, which also includes fungal‐mediated processes. However, in addition to the consideration made above regarding the main limitations, a test in accordance with OECD TG 216 does not assess other fungal‐specific endpoints which may be much more sensitive, specifically in the case of broad‐spectrum fungicides.

The PPR Panel in the Scientific Opinion (EFSA PPR Panel, 2017) indicated the importance of AMF as service providing units for various ecosystem services and reported on the need to complement the currently used testing strategy with a test investigating the effects on arbuscular mycorrhizae. This test would allow better coverage of any potential effects on soil microorganisms, particularly in the case of fungicides. Moreover, it could also be useful for understanding possible indirect effects on AMF occurring as a result of direct effects on host plants, which may include a large majority of crops (Pagano et al., 2023). Although an ISO method exists assessing the effects of chemicals on spore germination of AMF, the sensitivity of this endpoint compared to others related to both the pre‐symbiotic and the symbiotic phases is poorly understood. The investigation of additional species and endpoints both related to the pre and symbiotic phases, e.g. spore viability, root colonisation, spore production, etc., has been carried out since the publication of the Opinion (Lunardi et al., 2025; Mallmann et al., 2026). A recent project of the German Environment Agency (Natal‐da‐Luz et al., 2026) aiming at ring testing possible methods showed limitations, however, mainly related to the lack of reproducibility of the selected endpoints and tested species (R. irregolaris). Results of the project, however, also showed high sensitivity of soil fungi compared to currently tested species. Therefore, based on the SPGs defined by risk managers and the overall testing strategy for microorganisms, the WG will further consider whether a test with AMF should be included in the Tier 1 testing battery, and, if so, which endpoints should be recommended.

6.3.2.2. Relevant exposure duration for a proper hazard characterisation to in‐soil organisms (sub‐question 2.2)

Soil invertebrates

Regulations 283/2013 and 284/2013, and related Communications (2023), ²⁰ require only a chronic toxicity test over 56 days in the case of earthworms and 28 days and 14 days in the case of collembola and soil mites, respectively. While in the past an acute toxicity test with earthworms was also required, this was no longer considered necessary as it was deemed less relevant to the risk assessment than the chronic test. The main reason for the exclusion of the acute toxicity test could be that, while mortality and possible acute effects are detectable in a chronic test, sublethal effects, including effects on reproduction occurring over long‐term exposure are not captured in an acute toxicity test. Therefore, the WG confirmed chronic exposure as the most relevant.

Soil Microorganisms

Regarding soil microorganisms, the soil nitrogen transformation test is performed for at least 28 days. The exposure duration can be extended up to 100 days when effects > 25% compared to control(s) are observed at the end of the 28 day exposure. As reported under SQ1, pending on the method for assessing effects on soil microorganisms, an appropriate exposure duration for the new soil microorganism tests will also be proposed. The update of the systematic literature review as conducted by Puglisi (2012) and the outcome of the Microsoil and ERAMYC projects (Schlich et al., 2025; Natal‐da‐Luz et al., 2026) will support the choice of the most appropriate methods for regulatory purposes. Additionally, outcome of other projects like ARISTO will be considered.

6.3.2.3. Relevant ecotoxicological type of effects to be considered at Tier 1 and their characterisation (sub‐question 2.3)

Soil invertebrates

The ecotoxicologically relevant type of effects for soil invertebrates are effects on reproduction, assessed in terms of number of juveniles since those types of effects may be expected to also impact the population in the field. Regarding the percentage of effect, in line with Regulations 283/2013 and 284/2013 and the technical report (EFSA, 2015), 10% is considered as the biologically relevant effect threshold.

It has been questioned whether an adverse negative effect on body weight in soil invertebrates should also be considered for the purpose of endpoint setting. Based on experience, effects on reproduction are considered more relevant as they are expected to impact the maintenance of the population in the field. However, if major effects on body weight are observed, those should not be neglected, and appropriate consideration should be made, e.g. effects observed in isolation or accompanied by other effects.

If an increase in bodyweight is observed, caution should be taken in interpreting it as an adverse effect. Although certain substances may have a stimulatory effect and may thus lead to an increase in bodyweight (e.g. obesogens), the impact of an increase in body weight on the population is not yet well understood.

Similarly to bodyweight, the number of cocoons can also be measured and reported. As shown in Figure 11, cocoons are precursors of the juvenile stage, thus, it is reasonable to assume that an effect on cocoons would cascade down to result in effects on the number of juveniles. Consequently, the number of juveniles is still considered the most relevant ecotoxicological endpoint. However, effects on number of cocoons should still be contextualised based on the time and dose of those effects compared to effects on juveniles.

FIGURE 11.

Life cycle of an earthworm © Rick Kollath (all rights reserved).

In order to fully substantiate all the considerations made above, the results of the systematic literature search and dossiers investigations done under PERA will be further considered.

Soil microorganisms

For soil microorganisms, the relevant type of effects will depend on the method(s) selected for conducting the Tier 1 risk assessment. The recently published reports (Schlich et al., 2025; Natal‐da‐Luz et al., 2026) highlighted the microbial respiration test (MicrorespTM) and the enzymatic activity test (ISO 20130) and the spore germination test for arbuscular mycorrhizae (ISO 10832) as sensitive methods. Moreover, the update of the literature review, conducted in 2012 (Puglisi, 2012) under the project CT PLANTS 2025 38 – procedure EOI/EFSA/2022/01 will further support the choice of the most appropriate methods for regulatory use.

Similarly, for AMF, available information, as specified in Section 6.3.2.1, will be further considered to understand whether a test with AMF, e.g. spore germination test according to ISO 10832, is ready to be used, including information on the species to be tested and possible endpoints.

6.3.2.4. Information from Tier 1 studies (e.g. single endpoint or whole dose–response) (sub‐question 2.4)

Available OECD test methods for earthworms, collembola and soil mites, allow both the derivation of a single endpoint, e.g. EC10 or any other ECx pending also on agreed SPGs or the derivation of a full dose response. This latter may be used in the case the WG decides, in line with what was developed for bees, to conduct the risk assessment relying on the concept of dose–response relationships. ²¹ This latter method has been proposed instead of the one currently used, according to SANCO (European Commission, 2002), which is based on the concept of point estimates, e.g. ECx. Therefore, if the WG decides on the full dose–response option, the choice of test methods will also depend on the possibilities of deriving full dose–response.

6.3.2.5. Overview of methods for hazard sub‐questions

The Table 12 summarises the main outcomes of sub‐questions 1 to 4, highlighting gaps, relative priorities and methods for addressing these gaps. As reported in Section 6.3.1.6, also in the case of hazard, all the considerations apply to both active substances and their degradation products.

TABLE 12.

Overview of gaps, relative priority and methods for addressing sub‐questions (SQs) related to hazard characterisation of in‐soil organisms. In bold activities that should be conducted to gather additional data.

Sub‐question		Priority	Method
SQ1 – Species, life stage and method sensitivity	Soil invertebrates	High	Systematic literature review under contract PERA FPA EUBA/EFSA/PREV/2023/01
	Soil microorganisms	High	Review of existing reports (Schlich et al., 2025; Natal‐da‐Luz et al., 2026); Systematic literature review under contract CT PLANTS 2025 38 ‐ procedure EOI/EFSA/2022/01
SQ2 – Exposure duration in ecotoxicological studies	Soil invertebrates	–	Already addressed in the existing risk assessment
	Soil microorganisms	High	Review of existing reports (Schlich et al., 2025; Natal‐da‐Luz et al., 2026); Systematic literature review under contract CT PLANTS 2025 38 ‐ procedure EOI/EFSA/2022/01
SQ3 – Ecotoxicological type of effect	Soil invertebrates	–	Already addressed in the existing risk assessment
	Soil microorganisms	High	Review of existing reports (Schlich et al., 2025; Natal‐da‐Luz et al., 2026); Systematic literature review under contract CT PLANTS 2025 38 ‐ procedure EOI/EFSA/2022/01
SQ4 – Type of Information form Tier 1 ecotoxicological studies	Soil invertebrates	–	Already addressed in the existing risk assessment
	Soil microorganisms	High	Review of existing reports (Schlich et al., 2025; Natal‐da‐Luz et al., 2026); Systematic literature review under contract CT PLANTS 2025 38 ‐ procedure EOI/EFSA/2022/01

6.3.3. Risk characterisation sub‐questions

Based on the answers to previous sub‐question and in particular on the exposure scenario, exposure route, the risk cases reported in Table 13 have been identified.

TABLE 13.

Overview of the proposed risk cases for in‐soil organisms.

Exposure scenario	Risk case
Treated crop	Chronic‐contact – invertebrates (both soil and litter dwellers, pending on the data collection on litter)
	Chronic‐contact – microorganism
	Chronic – oral (litter fragmenters, only if relevant based on data availability)
Field margin‐Adjacent crop	Chronic‐contact – invertebrates
	Chronic‐contact – microorganisms
	Chronic – oral (litter fragmenters, only if relevant based on data availability)

Additional risk cases could be identified if, based on the outcome of the method for addressing sub‐questions 1.2 and 2.1, it will result that the dietary route of exposure should be considered at Tier 1 for litter fragmenters. As explained above, this will mainly depend on the sensitivity of those species compared to other standard species and on the occurrence of litter.

6.3.3.1. Linking the predicted exposure with the effect as characterised for each scenario and risk case

The EFSA PPR Panel Opinion on in‐soil organisms included a section on linking exposure and effect, referring in particular to the criss‐cross‐model (EFSA PPR Panel, 2012, 2017).

One important aspect of the link between exposure and effects is a clear definition of the interface between exposure and effect, i.e. the ecotoxicologically relevant concentration or exposure quantity (ERC/EREQ). This in simple words means that the ecotoxicological endpoint needs to be expressed in terms of the same type of ERC/EREQ, including duration, as the endpoint in the exposure tiers.

As already specified in the opinion, the relevant ERC is generally the concentration in soil pore water. The Opinion also specified that the concentration in the total soil may be more relevant for hard‐bodied organisms. Moreover, depending on the physchem properties of the substance under assessment, e.g. logP, the concentration in total soil may be more relevant than the concentration in pore water.

While on the exposure side, the available models (PERSAM; PEARL; PELMO) already deliver a PEC for both total soil and soil pore water at different soil depth, the currently available standard test methods, (OECD 222, 232 and 226) only refer to concentration in total soil and only the nominal concentration is generally reported. The opinion already recommended (i) to measure the concentrations over the duration of toxicity studies, as is done for studies with aquatic organisms, (ii) to develop a way to estimate the concentration in pore water, if only the concentration in total soil was assessed and (iii) to explore the possibility of using natural soils to make laboratory testing more realistic in terms of environmental conditions.

In general, for a proper estimation of the hazard in a laboratory test, it is advisable to measure the concentration in pore water and total soil over the duration of the study. Additional recommendations on time points for measurement, and ways to express endpoints will be defined in the guidance. In the cases where only the concentration in total soil is available, e.g. old studies, different methods for the estimation of the concentration in pore water will be further considered and proposed. With regard to the use of natural soils, standard natural soils are available, i.e. LUFA, which, in terms of content of organic matter, may be much more similar to arable soils. Results of the literature review investigating the bioavailability of PPPs to in‐soil organisms under different soil conditions conducted in the context of PERA will be consulted to better understand if natural standard soils may be recommended over the use of artificial soil in specific circumstances, e.g. for lipophilic substances.

Regarding exposure via the litter layer, if the evidence collected as suggested in Section 6.3.1.2 indicates that litter is likely, both contact and dietary exposure via litter should be further developed. In this case, an extensive literature review should be conducted to see whether there are methods for testing toxicity to in‐soil organisms through litter under laboratory conditions, or to understand whether available standard test methods can be modified to test that route of exposure. The test unit of the endpoint resulting from exposure to litter should determine the ERC in case of litter exposure.

6.3.3.2. Methods for assessing the outcome of the risk assessment in relation to the agreed SPGs

Two risk assessment approaches are possible:

• the single‐endpoint approach (= to the current approach)

• the approach combining all the predicted effects

The first approach is the traditional way of comparing single route of exposure and single endpoints, i.e. ECx based on agreed SPGs, to an assessment factor (e.g. trigger values) to consider interspecies differences and extrapolate from lab to field conditions. The trigger values should normally be scientifically underpinned by calibrating lower tiers against higher tiers and select trigger values that can guarantee the same level of protection across tiers.

If this approach is selected, the different tiers of the risk assessment would need to be calibrated against a surrogate reference tier (i.e. a representation, as accurate as possible, of the real situation in the field, i.e. reference tier) in order to make sure that the agreed SPGs are consistently implemented at all tiers. This exercise is, however, complex, as it implies a considerable amount of data at the species/functional group level should be available both from the surrogate reference tier, i.e. field tests in the case of in‐soil organisms and at the lower tiers. This may be the case for earthworms, and to some extent collembola and soil mites, but no data are available on other species of in‐soil organisms. Similarly, no higher tier data is available in soil microorganisms.

The second approach recommends quantifying effects at the individual level for all possible risk cases; combining all calculations of individual level effects with an extrapolation from individual to population levels and the aggregation of effects of a PPP on separate endpoints using the concept of response addition (EFSA, 2023a). The predicted effects at population levels can then be directly compared to the SPG. For in‐soil organisms, this could be quite simple based on the risk cases identified.

6.3.3.3. Overview of methods for risk assessment sub‐questions

Table 14 summarises the main outcomes of sub‐questions 1 and 2, highlighting gaps, relative priorities and methods for addressing these gaps.

TABLE 14.

Overview of gaps, relative priority and methods for addressing sub‐questions (SQs) related to risk assessment of in‐soil organisms. In bold activities that should be conducted to gather additional data.

Sub‐question		Priority	Method
SQ1 – Linking exposure and effect	Definition of the ERC by investigating (i) how to measure the concentrations over the duration of a studies (ii) a way to estimate the concentration in pore water, if only the concentration in total soil was assessed and (iii) the possibility of using natural soils to make laboratory test more realistic in terms of environmental conditions	High	Systematic literature review and experimental work under contract PERA FPA EUBA/EFSA/PREV/2023/01; Review of existing methods for recommendations on how to measure concentration over time in ecotoxicological tests and for estimation of concentration in pore water
SQ2 – Methods for assessing the outcome of the risk assessment in relation to the agreed SPGs	Single‐endpoint approach		Already addressed in existing risk assessment
	Approach combining all the predicted effects	High	The method developed in EFSA (2023a) will be considered and implemented as needed

7. NON‐TARGET TERRESTRIAL PLANTS

7.1. Gap analysis

The gap analysis revealed several issues that the WG considered meriting further investigations. An overview of the identified gaps related to the current risk assessment is visualised in Figure 13.

FIGURE 13.

Gap analysis for non‐target terrestrial plants presented using a fish‐bone diagram.

Apart from airborne spray drift and stomatal uptake, all of these gaps were already identified in the Scientific opinion (EFSA PPR Panel, 2014).

The gaps identified are listed below and are already categorised based on the sub‐question they contribute to.

Exposure pathways (sub‐question 1.1)

• Dust drift (particulate drift)

• Airborne spray drift

• Vapour drift

• Surface run‐off

Exposure route (sub‐question 1.4)

• Root uptake

• Stomatal uptake

Species and life stage sensitivity (sub‐question 2.1)

• Limited coverage of species/group of species in Tier 1 testing, therefore more sensitive groups may not be sufficiently in current scheme

• Limited coverage of potentially sensitive life stages by Tier 1 endpoints, therefore the difference in sensitivity of different life stages (e.g. reproductive stage) may not be covered

Testing methods (sub‐question 2.2)

• Pending SPGs, the current available test methods may not be sufficient, in particular as regards the time scale to cover potentially sensitive life stages.

Risk Assessment

• No consideration of potentially toxic metabolites/degradation or transformation products are available in the current risk assessment scheme.

7.2. Risk matrix

The risk matrix was used to ranking the gaps in a transparent manner and assign them a relative priority based on the combination of the likelihood and relevance (see Section 4) (Figure 14).

FIGURE 14.

The diagram illustrates the overall priority level attributed to each exposure and hazard and risk assessment related gap identified (from low to high, x‐axis). The overall priority level is obtained by combining the score (see footnote 7) attributed to likelihood or impact to risk assessment and relevance. This score is represented by the number of bricks in each colour (i.e. 1 brick = minimal; 2 bricks = moderate, 3 bricks = large). One‐direction error bars, using the same colour coding, reflect the upper limits of the scores whenever their quantification was uncertain and further evidence is needed to clarify the relevance.

Apart from airborne spray drift and stomatal uptake, these gaps are already discussed in the Scientific Opinion (EFSA PPR Panel, 2014).

7.2.1. Exposure characterisation

7.2.1.1. Exposure pathways (sub‐question 1.1)

The relatively low relevance and low likelihood of vapour drift was agreed upon because the majority of pesticide active substances have low vapour pressure and any evaporated molecules tend to disperse over long distances while diluting and degrading in the atmosphere.

The medium to high relevance and medium likelihood of run‐off was agreed considering that surface run‐off occurs only under a combination of specific conditions (e.g. certain rainfall intensity, slope, soil type), meaning that this pathway is highly scenario‐dependent.

Existing data (e.g. see in EFSA, 2013) indicate that dust drift from solid formulations is highly likely, and the particulates do contain the active substance(s). However, it should be noted that solid formulations of herbicides, which represent the most toxic group of PPPs towards plants, are relatively uncommon.

The relevance of the finer airborne spray drift fraction, particularly involving small droplets capable of long‐distance transport, is considered somewhat limited in comparison to the spray drift fraction depositing closer to the treated area (i.e. sedimentary spray drift). The fine droplets that consist of airborne spray drift are subject to dilution, atmospheric decomposition and limited deposition. Tall vegetation such as trees, shrubs or hedgerows may intercept part of this airborne fraction, but they can also enhance vertical air movement and turbulence, which may result in fine droplets being transported further downwind. The overall effect depends on factors such as vegetation density, height and prevailing wind conditions, meaning that such habitats should be regarded not as a consistent sink for airborne drift, but rather as a modifying factor that can either reduce or redistribute deposition (Azimonti et al., 2024; Felsot et al., 2010; Holterman et al., 2025). Nonetheless, the likelihood of the formation of those small droplets is high, as it is controlled primarily by the physical factors of the spraying process.

7.2.1.2. Exposure routes (sub‐question 1.4)

The most relevant exposure route which is not considered in the current assessment scheme is root uptake of PPP residues from soil pore water. This pathway is critical because most plants take up water in large amounts from the soil, and residues dissolved in the pore water are readily available for absorption. In contrast, residues that are strongly adsorbed to the solid phase of the soil (i.e. bulk soil) are unlikely to be directly accessible by plant roots. Nevertheless, those bound residues may desorb over time and migrate into the soil pore water, thereby becoming bioavailable. The likelihood and extent of root uptake are influenced by substance chemical properties, soil properties (e.g. organic matter content, texture, moisture), as well as plant traits such as phenology and growth stage. To better characterise these dependencies, a systematic literature review and complementary experimental work are planned under the PERA FPA EUBA/EFSA/PREV/2023/01. Given its importance, root uptake from soil pore water should be explicitly recognised as a key route of exposure of NTTPs and integrated into future risk assessment frameworks.

Following the deprioritisation of the vapour drift exposure pathway, the exposure route of stomatal uptake of gaseous PPPs in the vapour phase was also assigned low priority.

7.2.2. Hazard identification and characterisation

7.2.2.1. Species and life stage sensitivity (sub‐question 2.1)

As regards hazard characterisation, the Scientific Opinion highlighted the variability in sensitivity of different species of NTTPs and considerable variability in sensitivity of different life stages. Therefore, the need for a consideration to extend the scope of Tier 1 testing (additional species, additional endpoints and related modification of the methods) was agreed. Pending on the information collected to address this gap, risk assessment methods to address the sensitivity of species and endpoints will be considered. The WG considered that these aspects are highly relevant and can have a high impact on risk assessment.

7.2.3. Metabolites, degradation and transformation products

Currently, the risk assessment according to SANCO, 2002 is only performed for the PPP and there is no consideration of metabolites/degradation and transformation products. In the Scientific Opinion, it is indicated that degradation products in soil and metabolites in plant compartments should be considered. Indeed, toxic metabolites can be formed in both compartments, therefore addressing the risk from degradation and transformation products is relevant. Although the likelihood of the formation of toxicologically relevant metabolites is somewhat limited, if they are formed, the impact on the risk assessment could be high.

7.3. Methods for answering the sub‐questions

Methods for addressing the sub‐questions related to exposure, hazard and risk assessment are outlined in Sections 7.3.1, 7.3.2, 7.3.3, respectively. An integrated conceptual model (Figure 12) is included at the end of Section 7.

7.3.1. Exposure characterisation sub‐questions

7.3.1.1. Exposure pathways (sub‐question 1.1)

Currently (according to European Commission, 2002), only the sedimentary spray drift is considered. It is agreed that this pathway will also be considered in the revised guidance document.

Besides the sedimentary spray drift, considering the prioritisation as presented above, the WG agreed that the following exposure pathways will be considered:

• airborne spray drift

• dust drift

• surface run‐off

For the methods proposed for this sub‐question, see Sections 7.3.1.5 and 7.3.1.6.

7.3.1.2. Exposure‐relevant environmental matrices (sub‐question 1.2)

Terrestrial plants interact primarily with two environmental compartments which are their life‐supporting media: air and soil. Both compartments can be contaminated by pesticides from PPP applications, acting as sources of residue uptake. Therefore, it is proposed to consider these two matrices:

• Air

• Soil

For the methods proposed for this sub‐question, see Sections 7.3.1.5 and 7.3.1.6.

7.3.1.3. Exposure scenarios (sub‐question 1.3)

PPPs deposited from the air can be taken up by NTTPs. Sedimentary spray drift and dust drift are generally relevant pathways for any off‐field habitats, regardless of vegetation structure. It should be noted that in the context of risk assessment for NTTPs, it is considered that the off‐field habitat consists of natural, semi‐natural vegetation that is the type of vegetation to be protected (i.e. cropped habitats like adjacent crops are not considered in the ERA for non‐target terrestrial plants). Similar to sedimentary spray drift, root uptake of PPP residues occurring in soil is also independent from the habitat or any environmental scenario. However, the relevance of airborne spray drift may vary depending on the vertical structure of the receiving habitat. In low‐growing plant communities (e.g. grasslands, herbaceous margins), airborne spray drift is less likely to result in significant deposition due to the reduced interception surface at canopy height. In this document, that type of habitat is called a ‘2‐D field margin’, forming the basis for the 2‐dimensional (2‐D) NTTP off‐field exposure scenario. In contrast, taller vegetation structures such as shrubs, hedgerows or trees are more likely to intercept fine spray droplets. For such structures, airborne spray drift becomes a potentially relevant exposure pathway. These types of habitats are described as the ‘3‐D field margin’, corresponding to the 3‐dimensional (3‐D) off‐field exposure scenario. Therefore, two distinct exposure scenarios are proposed for NTTPs:

• 2‐D field margin scenario – representing habitats with low, horizontally structured (semi‐) natural vegetation.

• 3‐D field margin scenario – representing habitats with vertically structured (semi‐) natural vegetation.

If necessary, the WG will consider introducing further definitions, e.g. based on height, for these two scenarios.

The proportion of the PPP mass intercepted by plant surfaces, as well as the proportion reaching the soil, depends on the physical characteristics of the habitat.

7.3.1.4. Exposure routes (sub‐question 1.4)

Non‐target terrestrial plants can be exposed to PPP residues through distinct biological uptake mechanisms, depending on the environmental matrix in which residues are present. PPP residues present in the air can be deposited onto plant surfaces. Once deposited, residues may enter the plant via cuticular or stomatal absorption – a process referred to in this document as foliar uptake. PPP residues present in soil pore water can be taken up by roots and translocated within the plant via the vascular system. This is referred to as root uptake.

Therefore, it is proposed to consider the following exposure routes:

• Foliar uptake (from deposited residues on aerial plant parts).

• Root uptake (from residues dissolved in soil pore water).

7.3.1.5. Exposure quantification (sub‐question 1.5)

According to the above, the key processes as summarised in Table 15, will be quantified.

TABLE 15.

Key processes to be considered for the revised Guidance Document for non‐target terrestrial plants.

No.	Exposure route	Environmental matrix	Exposure scenario	Exposure pathways involved in the process
1	Foliar uptake	Air	2‐D field margin scenario	Sedimentary spray/dust drift
2	Foliar uptake	Air	3‐D field margin scenario	Sedimentary spray/dust drift Airborne spray drift
3	Root uptake	Soil	2‐D field margin scenario	Sedimentary spray/dust drift Surface run‐off
4	Root uptake	Soil	3‐D field margin scenario

1. Foliar uptake from air in 2‐D field margin scenario

It is assumed that the foliar uptake of PPPs by NTTPs from spray applications is directly related to the sedimentary spray drift that reaches the off‐field habitats and deposits to plant surfaces.

See text under 5.3.1.5 regarding spray drift and the activities planned for gathering additional data, i.e. a systematic literature review and experimental work.

Dust (particulate) drift represents another important exposure pathway. See text under 5.3.1.5 regarding dust drift and the activities planned for gathering additional data.

For simulating the exposure pathway, the following model is proposed:

$$\text{PEQfol}=\mathrm{AR}\times \mathrm{MAF}\times \mathrm{EF}$$

where,

PEQfol is the predicted exposure quantity stemming from foliar uptake, g/ha.

AR is the application rate, g/ha.

MAF is the multiapplication factor, − (unitless).

EF is the exposure factor, − (unitless).

Notably, the current guidance (European Commission, 2002), proposes the use of a MAF only for NTAs, but not p for NTTPs. For further considerations and proposed actions on the MAF, see Section 5.3.1.5.

The EF parameter represents the proportion of the application rate that deposits to the field margin considering the drift. See text under 5.3.1.5 regarding drift and the activities planned for gathering additional data and see text under 5.3.1.5 and Figure 15 regarding explanation for EF and the proposed actions.

• 2Foliar uptake from air in 3‐D field margin scenario

FIGURE 15.

Envisaged work plan for addressing airborne spray drift for non‐target terrestrial plants.

Foliar uptake by NTTPs in the 3‐D field margin scenario stems either from dust drift for solid formulations or from two types of spray drift from spray applications; sedimentary spray drift and airborne spray drift. For dust drift and sedimentary spray drift, the same methods as described above in point (1) will be applied. It is acknowledged, however, that robust deposition curves for 3‐D vegetation structures in off‐field habitats are currently lacking. Although efforts have been made at EU level, such as within the SETAC‐DRAW initiative (2017), ²² a harmonised EU‐level protocol for conducting spray drift studies in complex 3‐D vegetation (e.g. field margins, hedgerows) is not yet established. Existing methodological references include ISO 22866 (2005) ²³ and national guidance such as that developed by the UK HSE (LERAP). ²⁴ In addition, the use of a VDF has been discussed in EFSA PPR Panel (2015) for NTA assessments, as a means to account for interception and dilution within structured vegetation. For explanation and proposed action on VDF, see Section 5.3.1.5.

Airborne spray drift is currently neither considered in the Terrestrial Guidance Document nor explicitly addressed in the EFSA Scientific Opinion. The relevance of this exposure pathway will first be investigated. If it is determined that airborne drift contributes significantly to NTTP exposure, then a dedicated exposure model will be developed and parameterisation based on new data will be sought. In this case, the Ecotoxicologically Relevant Exposure Quantity (EREQ) – for spray application will be derived as the sum of both sedimentary and airborne spray drift components. To support this evaluation, data generated under the PERA FPA EUBA/EFSA/PREV/2023/01 will be used where available, including both systematic literature review outcomes and experimental field study data. These activities are expected to provide information on spray drift deposition patterns in complex 3‐D vegetative structures and will help to inform whether and how airborne drift should be incorporated into the revised guidance. Conversely, if airborne spray drift is judged to be of limited relevance in the overall exposure assessment, then the pathway will not be considered in the updated Guidance Document.

The plan to address this issue is illustrated in Figure 15.

• 3Root uptake from soil in 2‐D field margin scenario and (4) Root uptake from soil in 3‐D field margin scenario

The root uptake of PPPs from pore water of the soil is not considered currently, but it is agreed that this pathway will be considered in the revised guidance document.

The pore water concentration resulting from deposition onto the soil is characterised in EFSA (2017) and PECpw are estimated using the model PERSAM, PEARL and PELMO. PECs are estimated at different soil depths. See Section 6.3.1.5 for the proposal on how to estimate soil concentrations in the off‐field area.

Two pathways are considered relevant for the off‐field exposure, the deposition from the air and the soil contamination via surface run‐off. However, the likelihood and relevance of the contribution of run‐off into the PECpw still needs to be investigated.

The Scientific Opinion recommends that quantification of exposure via run‐off be based on the modelling approaches developed under the FOCUS surface water guidance (FOCUS, 2001), particularly using PRZM at Step 3, which simulates run‐off and erosion losses from different crops as a function of chemical properties (e.g. Koc) and environmental conditions. Although originally developed for aquatic exposure, the models and default values can be used to estimate the fraction of PPP lost from the field via run‐off and its potential deposition onto terrestrial off‐field compartments such as vegetated buffer strips or other field margin habitats. See Section 6.3.1.5 for the approach on how to further explore the relevance of surface run‐off/erosion.

The method for quantification of exposure from the air will follow the same approach as described in points (1) and (2) above. However, a consideration of plant interception should be made by the WG, since assuming that all PPP available in the air will be deposited to the soil would be an unrealistic assumption for this route of exposure. The proportion of interception should be dependent on the season, the region and the off‐field habitat type, which should be taken into consideration. The interception values available for crops (e.g. EFSA, 2023a) will be first considered for this task. Consequently, if the WG deems this is insufficient for off‐field vegetation then additional data will be sought through a narrative literature review.

The proposed work strategy is illustrated in Figure 16.

FIGURE 16.

Envisaged strategy for developing exposure estimation method for root uptake of non‐target terrestrial plants.

No exposure model was developed in this protocol. However, the WG discussed the key parameters to be considered for the exposure model and the key aspects of the envisaged risk assessment for NTTPs. It was agreed that the PEQ for root uptake should be linked to the PECpw. The WG considered that the exposure of individual plants in laboratory tests could be considered as worst‐case compared to field situations. This is because the individuals in the tests (both OECD 227 and 208) have more space to grow, potentially more spray liquid can reach their above ground parts and more residues can be taken up by the roots from the soil. A typical field margin habitat may be very dense where the individual plants are continuously competing for light, water and the other resources. This results in a potentially lower exposure at individual level via both routes of exposure (foliar uptake and root uptake) in‐field situations. Therefore, the application rate was agreed to be considered as a good predictor for the amount (mass) received or taken up by each plant. However, in order to maintain the application rate as the predictor, a conversion between PECpw and the mass of residues taken up by the plants in the seedling emergence (OECD 208) would be needed. It was agreed that considering the size of the pots used in the test, this conversion could be done (i.e. the mass of residues potentially available for the test plants in the pot could be calculated). It was also discussed that the PECpw is not routinely available from the laboratory tests. Nevertheless, it could be measured or predicted. For the prediction, certain soil property data should be available and adequately reported (see also Section 6.3.3.1).

7.3.1.6. Overview of methods for exposure sub‐questions

Table 16 summarises the main outcomes of SQ1 to SQ4, highlighting gaps, relative priorities and methods for addressing those gaps. SQ5 is not reported in the table since it considers all the aspects addressed in the previous questions. It should be noted that all the considerations made apply to both active substances and degradation products.

TABLE 16.

Overview of gaps, relative priority and methods for addressing sub‐questions (SQs) related to exposure for non‐target terrestrial plants. In bold activities that should be conducted to gather additional data. Please note that degradation products are not specifically mentioned, since relevant aspects listed in the table are applicable for both active substances/PPP and degradation products.

Sub‐question		Priority	Method
SQ1 – pathways	Airborne spray drift	Potentially high	Preliminary investigation of relevance.
	Sedimentary spray drift/dust drift	High	Systematic literature review and Experimental work under contract FPA EUBA/EFSA/PREV/2023/01
	Surface run‐off/erosion	Potentially high	Preliminary investigation for its relevance via modelling (for details see Section 6.3.1.5).
	Volatilisation	Low	No data gathering needed
SQ2 – Environmental matrix	Air	High	Exposure model proposed but no data gathering is necessary
	Soil	High	No data gathering needed as already covered by existing models (PERSAM; PEARL, PELMO)
SQ3 – Exposure scenario	2‐D field margin scenario	High	Once data are gathered as outlined for SQ1, an exposure model will be developed.
	3‐D field margin scenario	High	Once data are gathered as outlined for SQ1, an exposure model will be developed.
SQ4 – Exposure route	Foliar uptake	High	Already addressed but data gathered under SQ1 will be considered
	Root uptake	High	Systematic literature review and experimental work under PERA FPA EUBA/EFSA/PREV/2023/01 will further elucidate its relevance. Established modelling methods are available and will be implemented for quantification.

7.3.2. Hazard identification and characterisation sub‐questions

According to the currently in‐use SANCO Guidance (European Commission, 2002), the Tier 1 risk assessment is based on screening data from efficacy trials and/or other assays. The WG will consider if this method should continue to be recommended; if so, it will be considered as a screening step. This screening step is not further discussed in this protocol. The risk assessment method that is called Tier 2 in the SANCO Guidance (European Commission, 2002), is considered the basis for the development of a Tier 1 risk assessment in the updated guidance document for NTTPs. Therefore, the considerations below will refer to that tier as Tier 1, although it is called ‘Tier 2’ in the SANCO GD.

The current risk assessment schemes only consider testing PPPs (or active substances). If testing of metabolites becomes necessary, the same principles apply as discussed below.

7.3.2.1. Species and life stage sensitivity (sub‐question 2.1)

To answer the question, the current knowledge and practice with the prioritisation exercise (see Section 7.2 above) was taken into consideration.

In the current practice of Tier 1 risk assessments (called Tier 2 in SANCO, 2000), two types of laboratory tests are conducted and 6–10 species are selected in those tests (usually crop species). The species are selected to cover a number of taxonomic groups (some dicotyledon and some monocotyledon species are always represented). This testing strategy is suitable to investigate the interspecies sensitivity distribution and for constructing SSD curves, which may optionally be considered in the risk assessment.

The current practice has been criticised because only crop species are selected for testing and wild plant species are not represented, while they may have different sensitivity (EFSA PPR Panel, 2014). As discussed in the Scientific Opinion, the sensitivity differences between crop and wild species are not evident when annual/biannual species at juvenile stages are compared. However, when other crop stages (e.g. reproductive stages) or some other plant groups, like woody plant species or ferns, are compared with the typical test species, some differences in sensitivity are more apparent.

These concerns were investigated through targeted literature reviews by Christl et al. (2017a, 2017b). Both literature reviews collected and analysed relevant published and unpublished laboratory and field studies on plants. The authors concluded that there were no consistent differences in sensitivity between wild plant species and crop species, and there were no consistent differences in sensitivity between vegetative and reproductive ecotoxicological endpoints. However, some uncertainties and some exceptions were identified (e.g. low reliability of detecting small differences was noted, endpoints not fitting with the general trend seemed to be substance‐specific).

Therefore, the WG agreed that further investigation will be carried out. Results from the ongoing EFSA‐funded projects will be evaluated first. Sensitivity differences between plant species will be investigated by a systematic literature review in the context of the PERA FPA EUBA/EFSA/PREV/2023/01. Following the public consultation, any proposed change in the protocol will be considered. Final results should be available in the second quarter of 2027.

If additional data are required to adequately address all the aspects considered here, targeted literature searches will be conducted to complement existing findings, e.g. by extracting pertinent data from regulatory dossier studies.

It is expected that the database obtained by these data collections could also be used to answer some additional issues (linked to other SQs) as needed for the guidance development. For example, questions like ‘could particularly sensitive endpoints be identified?’, ‘what could be a realistic worst‐case factor to be applied for extrapolation between certain endpoints?’ or ‘what could be a realistic worst‐case factor to be applied for extrapolation between the least and most sensitive plant groups?’ may be able to be answered by further analysis.

Since sensitivity differences between plant groups should be answered by this SQ, the WG will establish the groups and traits of EU wild plants to be considered. Certain characteristics like herbaceous or woody, the height and morphology of the species will be considered for these groupings.

7.3.2.2. Relevant exposure duration for a proper hazard identification and characterisation (sub‐question 2.2)

After PPP application in phytotoxicity tests, dissipation/degradation processes take place (the test plants are not transferred to clean media or a clean environment in these tests). It may be considered that the rate of dissipation/degradation processes in laboratory conditions differ from field conditions, since the environmental conditions in‐field situations are more variable than in the laboratory, where many parameters (e.g. light regime, temperature, relative humidity, certain soil characteristic) are kept within certain ranges. Therefore, the exposure duration in the laboratory could be shorter than in certain field situations. However, the study duration in the standard laboratory studies is typically 3–4 weeks, which represents a substantial proportion of the lifespan of the test plants. During this period considerable growth and development take place, and the main effect endpoints are related to growth. On the other hand, in general, the half‐life of pesticides in/on plants is usually shorter than 10 days (EFSA, 2023a, 2023b). Therefore, considering all the aspects above, the WG considered that the exposure duration as it results from the setup of the relevant phytotoxicity tests (OECD 208 ²⁵ and 227 ²⁶ ), seems to be sufficient. Nevertheless, the WG concluded that the actual exposure in the study must be controlled through analytical methods during testing. Accordingly, relevant criteria will be developed for inclusion in the updated Guidance Document.

Furthermore, for multiple applications, the relevance of potential effects of the exposure time between repeated applications will be investigated (see also 7.3.3).

7.3.2.3. Relevant ecotoxicological type of effects to be considered at tier 1 and their characterisation (sub‐question 2.3)

In the currently used Tier 1 tests (OECD 208 and OECD 227) the following observations are performed:

• Emergence (just in OECD 208)

• Visual observations for phytotoxicity

• Mortality

• Dry or fresh shoot weight

• Shoot height (optional)

In addition, reproductive structure measurements are requested by OECD 227, if such structures are present; however, such endpoints are typically not available.

The endpoints from the above observations should be expressed as % deviation from the control. For studies with dose–response design, if possible, regression analysis should be performed, and the endpoints should be expressed as ECx or ERx. In addition, NOEC/LOEC values should be derived if possible.

According to the Scientific Opinion, in order to satisfy the proposed SPG in the Scientific Opinion, the recommended endpoints to be considered in the future risk assessment schemes are (note that in the Opinion HC5 are mentioned in place of HR5):

• HR5 ERrepro10: 5%‐ile hazard concentration derived from SSD curve made up from 10% reproductive effect rates of the individual assays (single test species)

• HR5 ERveg10: 5%‐ile hazard concentration derived from SSD curve made up from 10% vegetative effect rates of the individual assays (single test species)

• HR5 ERvisual50: 5%‐ile hazard concentration derived from SSD curve made up from 50% visual damage effect rates of the individual assays (single test species)

It is noted that the proposed endpoints may need to be reconsidered once agreed SPGs are set. Nevertheless, substantial deviations from the above proposed protection goals are not expected. The WG considered that the recommendations in the Scientific Opinion, as presented above, are reasonable and should be the basis for future risk assessment schemes. Therefore, the relevant ecotoxicological effects are visually observable including phytotoxicity, effects on vegetative development and effects on reproduction.

As regards the characterisation of those effects, again the above recommendations from the Scientific Opinion are to be considered. Using the hazard concentration (or hazard rate) approach results in the discontinuation of the deterministic risk assessment option at Tier 1 (as is recommended in SANCO, 2002 and is currently the common practice). Instead, probabilistic risk assessment will be conducted.

As regards the three types of effects, the following considerations and actions will be made:

Visual phytotoxic effects (i.e. visual damage) are covered by the current test guidelines. Either qualitative assessments or quantitative ratings are recommended. However, qualitative assessments may prevent any ERx calculations, therefore the endpoint to be used in the risk assessment cannot be expressed in the preferred way (HC5 ERvisual50). Therefore, the WG will develop recommendations for collecting and reporting those observations in order to guarantee that the preferred endpoint could be established. It is considered that those recommendations will only be amendments to the guidelines and will not overrule the relevant aspects of the test guideline.

Vegetative effects are measured in different ways in the test guidelines: emergence, mortality, dry or fresh shoot weight, shoot height. The WG considers that those observations sufficiently cover the need to investigate potential effects of PPPs on vegetative development. Nevertheless, the need for measurements for shoot height in the presence of measurements of shoot weight will be discussed, and clear recommendations proposed. Also, whether dry or fresh shoot weights are equally appropriate, or there is a preference for one or the other will also be discussed. In general, the outcomes of the tasks in 7.3.2.1 will be used to investigate whether some types of observations can be omitted and further simplifications applied. Again, it is considered that those recommendations will be amendments only and will not overrule the relevant aspects of the existing test guideline.

Reproductive effects are not covered by the current testing protocols. However, the already available assessments, namely the relevant information in the Scientific Opinion and publications by Christl et al., suggest that reproductive effects may be predictable from the vegetative effects by applying appropriate assessment factor(s). Therefore, the WG considered that first it will be investigated whether such assessment factor(s) could be established for the Tier 1 risk assessments. The method of these investigations is detailed in 7.3.2.1, above. This would also mean that no amendments to the current test methods would be necessary. Recommendations for amendments to the test methods (e.g. tests performed with plants in reproductive stages or longer test duration to cover reproductive stages, type of reproductive endpoint to be considered) will be developed for higher tiers of the risk assessment or if reliable and applicable assessment factor(s) cannot be established.

The preferred type of endpoint to be derived from each assay (test on individual plant species) and each type of effect is the ERx. If no ERx can be derived, a NOER value should be established. Since the exposure in the test is linked to and expressed by the application rate (g/ha), the units of the endpoints discussed above will be g/ha. Exceptionally, in studies according to OECD 208, the exposure may be set by mixing the PPP into the soil, therefore the exposure is expressed as soil concentration (mg/kg). In such cases, the exposure should be converted to g/ha by considering the concentration, the dimensions of the pot and the weight (or density) of the soil per pot.

7.3.2.4. Information to be derived from Tier 1 studies (sub‐question 2.4)

Based on the above considerations (especially 7.3.2.3), for NTTPs, the WG considers that a single‐point estimate, such as ERx, should be derived from each assay (test on individual plant species). For this, regression analysis should be performed, and a reasonable dose–response curve derived for studies with dose–response design. For tests with limited response even at high application rate(s) (e.g. limit tests), the point estimate will be represented by the NOER value. Where possible, HR5 values should be calculated. Where no HR5 values can be calculated (e.g. low number of suitable ERx values are available, poor fit of the regression curve of ERx values), the WG will develop alternative recommendations (e.g. lowest reliable point estimate). As regards regression analysis, as general guidance, EFSA (2019) ²⁷ will be considered.

7.3.2.5. Overview of methods for hazard sub‐questions

The Table 17 summarises the main outcomes of sub‐questions 1–4 with the proposed tasks and methods for addressing these questions. As reported in 7.3.1.6, also in the case of hazard, all the considerations apply to both active substances and their degradation products.

TABLE 17.

Overview of gaps, relative priority and methods for addressing sub‐questions (SQs) related to hazard characterisation of in‐soil organisms. In bold activities that should be conducted to gather additional data.

Sub‐question	Priority	Method
SQ1 – Species and life stage and sensitivity	High	Systematic literature review under contract systematic literature review in the context of the PERA project FPA.
SQ2 – Exposure duration in ecotoxicological studies	–	No need for data collection or particular action is currently identified. Pending on the outcomes of SQ1, additional activities may be identified.
SQ3 – Ecotoxicological type of effect	–	No need for data collection is identified. Potential actions depend on outcomes on tasks on SQ1 (as in 7.3.2.1, above) and on the agreed SPG.
SQ4 – Type of information form Tier 1 ecotoxicological studies		No need for data collection is identified. Potential actions depend on SQ3.

7.3.3. Risk characterisation sub‐questions

7.3.3.1. Linking the predicted exposure with the effect as characterised for each scenario and risk case

According to the current plans, different exposure pathways and exposure routes will be considered. Nevertheless, the exposure characterisation will result in a single PEQ value for both scenarios with the unit of g/ha. The exposure for the two scenarios will not be combined since the two types of habitats are considered standalone entities with no overlap. Even if one of the two scenarios systematically indicates higher PEQs than the other; both PEQs have to be calculated, since the hazard endpoint may be different for the two scenarios.

The hazard characterisation will result in three different risk cases: effects on visual damage, growth (vegetative development) and reproduction. It is proposed that separate endpoints are derived for the following two scenarios: For the 2‐D scenario, the SSD curves should be made only from endpoints of bioassays on annual and biannual test species. The SSD curves made for the 3‐D scenario must include a consideration of the sensitivity of annual ‘tall’ plants. Theoretically, this could be done by adding phytotoxicity endpoints from bioassays using such plants. However, currently such data are hardly available (and not routinely generated since standardised test methods are not available). Considering the outcome of the tasks outlined in 7.3.2.1, the WG will discuss how annual ‘tall’ plants should be considered (e.g. using assessment factors). Regardless, the unit of the hazard endpoints will be g/ha. The distinction between these three risk cases stems from the specific protection goals, as currently proposed in the Scientific Opinion. Once agreed SPGs are set, the list of risk cases (and related endpoints) maybe revised.

Since both the exposure and the effect values are expressed as mass per area (g/ha), they can directly be linked and compared. Six toxicity‐exposure ratios have to be calculated in order to cover all scenarios and risk cases, as presented in Table 18.

TABLE 18.

Risk cases for Non‐Target Terrestrial Plants.

Exposure scenario	Risk case
2‐D field margin	Visual damage
	Effects on growth and development
	Effects on reproduction
3‐D field margin	Visual damage
	Effects on growth
	Effects on reproduction

The WG will investigate whether simplification could be possible, considering the outcomes of the tasks in SQ 2.1 and the agreed SPGs.

As reported above for NTAs, it is noted that some aspects of the exposure estimation are specific for the active substance (e.g. proportion of the active substance absorbed or taken up by the plant), while for hazard characterisation the formulated product is used. Linking these two aspects may not be straightforward for PPPs with more than one active substance. Therefore, the WG will develop a specific approach to address this issue.

In addition to the aspects mentioned above, the WG will further consider how to link the exposure and effects in case of multiple applications (i.e. potential effects of multiple pulses and the exposure time between repeated applications).

7.3.3.2. Methods for assessing the outcome of the risk assessment in relation to the agreed SPGs

The proposed endpoints (HR5 based on ERx values) are aligned with the currently proposed specific protection goals (SPGs) outlined in the Scientific Opinion. These endpoints adequately address key aspects such as interspecies sensitivity and sensitivity across different life stages, and the laboratory exposure conditions are considered sufficiently worst‐case compared to field situations.

As a result, applying these endpoints in the risk assessment is considered to provide a sufficient level of protection without requiring additional considerations, such as trigger values including extra assessment factors. Consequently, if the predicted environmental concentration (PEQ) is lower than the hazard threshold (HR5), the SPG can be considered respected. Conversely, if the exposure exceeds the endpoint, the Tier 1 risk assessment does not demonstrate compliance with the SPG.

Nevertheless, the WG will review remaining uncertainties to determine whether additional assessment factors are needed. Moreover, once agreed SPGs are set, the above considerations may be reviewed.

7.3.3.3. Overview of methods for risk assessment sub‐questions

Table 19 summarises the main outcomes of sub‐questions 1 and 2. As reported in 7.3.1.6 and 7.3.2.5, also in the case of the risk assessment, all the considerations apply to both active substances and their degradation products.

TABLE 19.

Overview of gaps, relative priority and methods for addressing sub‐questions (SQs) related to risk assessment of in‐soil organisms.

Sub‐question	Priority	Method
SQ1 – Linking exposure and effect	–	No need for data collection is identified. Potential actions depend on outcomes on tasks identified in 7.3.2.1 and on the agreed SPG.
SQ2 – Methods for assessing the outcome of the risk assessment in relation to the agreed SPGs	–	No need for data collection is identified. Potential actions depend on the agreed SPG.

7.3.4. Metabolites/degradation and transformation products

The Scientific Opinion highlighted two relevant compartments where toxicologically relevant metabolites may be formed: soil and plant. However, the WG considered that metabolites formed in or on the plants are covered by the Tier 1 testing, as it is expected that they would be formed in the test according to OECD 227. It is acknowledged that some differences in the environmental conditions between the laboratory and field conditions exist. Nevertheless, as assessed in 7.3.2.2 above, those differences are not considered substantial. Therefore, no risk assessment methods will be developed for degradation and transformation products that maybe formed in or on plants.

Similarly, degradation and transformation products that maybe formed early in the degradation processes in soil are covered by the Tier 1 testing.

However, soil metabolites that are formed exclusively in later phase of the degradation processes may not be formed in the laboratory tests and are therefore not covered in Tier 1 testing. The WG considered that risk assessment methods will be developed for such soil metabolites.

As regards the determination of the relevant metabolites, information from route and rate of degradation in soil and, if available, information from rotational crop studies will be considered.

As regards the exposure, the only relevant route is the root uptake from soil. As for the PPPs (active substances), the PEQ for the metabolites will be derived considering the PECpw (see Section 7.3.1.5).

As regards the hazard characterisation for Tier 1 risk assessment, the same aspects as for the PPP apply (see Section 7.3.2) with the exception that testing only according to OECD 208 TG is recommended. Alternative methods for the hazard characterisation (in order to avoid unnecessary testing) and a screening step for the risk assessment will be developed for the updated Guidance Document.

GLOSSARY AND ABBREVIATIONS

AENEAS

Advancing the Environmental Risk Assessment of Non‐Target Arthropods for Plant Protection Products by Accounting for the Impact on Ecosystem Services and Ecological Function

AMF

arbuscular mycorrhizae

AR

application rate

AW

absorption factor from water

Ctgb

Dutch Board for the Authorisation of Plant Protection Products and Biocides

DDF

deposition distribution factor

DegT50

time taken for 50% of substance to disappear from a compartment as a result of degradation processes alone

DRAW

Drift Risk Assessment Workshop

DT50

generic term to describe the time required for disappearance of 50% of the residue. Ideally, which loss processes the disappearance time relates to should be clarified, e.g. DegT50 within the soil matrix degradation, DisT50 for dissipation processes. If the calculation of the DT50 is performed using single first‐order (SFO) kinetics, the DT50 can also be referred to as a ‘half‐life’

EEASE

EU Environmental Scenarios for ERA of Non‐target Organisms

EF

exposure factor

EREQ/C

Environmental Relevant Exposure Quantity/Concentration

EUPAF

European Precision Application Task Force

GBIF

Global Biodiversity Information Facility

GMS

generic model species

KFoc

Freundlich soil‐water partition coefficient

Koc

partitioning coefficient between (soil) organic carbon and water

LAI

leaf area index

LERAP

Local Environment Risk Assessment for Pesticides

MAF

multiple application factor

NOEC

no observed effect concentration

NOER

no observed effect rate

NTA

non‐target arthropod

NTTP

non‐target terrestrial plants

PEARL

pesticide emission at regional and local scales

PEC/Q

predicted environmental concentration/quantity

PECpw

predicted environmental concentration in pore water

PELMO

pesticide leaching model

PERA

Advancing the ERA of Plant Protection Products towards a system‐based approach

PERSAM

persistence in soil analytical model

PPP

Plant Protection Product

PRZM

Pesticide Root Zone Model for calculating fate and behaviour of substances in the unsaturated zone of the soil

SPG

specific protection goals

SQ

sub‐question

TK‐TD

toxicokinetic‐toxicodynamic

ToR

Term of Reference

TRT

time‐reinforced toxicity

VDF

vegetation distribution factor

WU

water uptake

REQUESTOR

European Commission

QUESTION NUMBER

EFSA‐Q‐2025‐00658

COPYRIGHT FOR NON‐EFSA CONTENT

EFSA may include images or other content for which it does not hold copyright. In such cases, EFSA indicates the copyright holder and users should seek permission to reproduce the content from the original source.

MAP DISCLAIMER

The designations employed and the presentation of material on any maps included in this scientific output do not imply the expression of any opinion whatsoever on the part of the European Food Safety Authority concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.

Supporting information

Annex A: Public consultation on the draft protocol

LIST OF ANNEXES A.  

Annex A: Public consultation on the draft protocol.

EFSA (European Food Safety Authority) , Adriaanse, P. , Arts, G. , Lassen, S. B. , Brown, M. , Chaton, P.‐F. , Fürst, M. , Ingels, B. , Pieper, S. , Rundlöf, M. , Sousa, J. P. , van Eekelen, R. , Alvarez, F. , Arena, M. , Fait, G. , Ippolito, A. , Linguadoca, A. , Padovani, L. , Szentes, C. , & Auteri, D. (2026). Protocols for the development of the guidance on the risk assessment of PPP to non‐target arthropods, in‐soil organisms and non‐target terrestrial plants. EFSA Journal, 24(7), e10228. 10.2903/j.efsa.2026.10228