Coverage of endangered species in environmental risk assessments at EFSA


Scientific Committee
EFSA Journal
EFSA Journal 2016; 14(2):4312 [124 pp.].
Panel members at the time of adoption
Simon More, Alicja Mortensen, Antonia Ricci, Vittorio Silano, Katrine Helle Knutsen, Guido Rychen, Hanspeter Naegeli, Dominique Turck, Michael John Jeger, Colin Ockleford, Diane Benford, Thorhallur Halldorsson, Anthony Hardy, Hubert Noteborn, Josef R. Schlatter, Roland Solecki

The Scientific Committee wishes to thank the following members, chairs* and scientific secretariat# of the working group on overarching elements of environmental risk assessment (coverage of endangered species) for drafting this scientific opinion: Michael Bonsall, Theo Brock*, Gianni Gilioli, Christer Hogstrand, Jonathan Jeschke, Mira Kattwinkel, Robert Luttik*, Ad Ragas, Paulo Sousa and Claus Svendsen, as well as the following EFSA staff members for their support: Fernando Alvarez, Yann Devos, Jean-Lou Dorne, Angelo Maggiore, Agnes Rortais, Reinhilde Schoonjans#, FranzStreissl, Jose Tarazona, Sara Tramontini and Maria Vittoria Vettori

Opinion of the Scientific Committee/Scientific Panel
On request from
Question Number
11 November 2015
3 February 2016
European Food Safety Authority (EFSA), Parma, Italy
Download Article (7.79 MB)

The EFSA performs environmental risk assessment (ERA) for single potential stressors such as plant protection products, genetically modified organisms and feed additives, and for invasive alien species that are harmful to plant health. This ERA focusses primarily on the use or spread of such potential stressors in an agricultural context, but also considers the impact on the wider environment. It is important to realise that the above potential stressors in most cases contribute a minor proportion of the total integrated pressure that ecosystems experience. The World Wildlife Fund listed the relative attribution of threats contributing to the declines in animal populations as follows: 37% from exploitation (fishing, hunting, etc.), 31% habitat degradation and change, 13% from habitat loss, 7% from climate change, and only 5% from invasive species, 4% from pollution and 2% from disease. In this scientific opinion, the Scientific Committee gathered scientific knowledge on the extent of coverage of endangered species in current ERA schemes that fall under the remit of EFSA. The legal basis and the relevant ecological and biological features used to classify a species as endangered are investigated. The characteristics that determine vulnerability of endangered species are reviewed. Whether endangered species are more at risk from exposure to potential stressors than other nontarget species is discussed, but specific protection goals for endangered species are not given. Due to a lack of effect and exposure data for the vast majority of endangered species, the reliability of using data from other species is a key issue for their ERA. This issue and other uncertainties are discussed when reviewing the coverage of endangered species in current ERA schemes. Potential tools, such as population and landscape modelling and trait-based approaches, for extending the coverage of endangered species in current ERA schemes, are explored and reported.


Plant protection products (PPPs), feed additives (FAs) and genetically modified organisms (GMOs) are subject to a risk analysis and regulatory approval before being placed on the market, released into the environment, spread or used in agriculture. In this process, the role of the European Food Safety Authority (EFSA) is to independently assess and provide scientific advice to risk managers on possible risks that PPPs, GMOs and FAs may pose to the environment. EFSA also assesses the environmental risks related to the entry and spread of invasive alien species (IAS) that are harmful to plant health and the effects of their management. PPP, GMOs, FAs and IAS are herein commonly denominated as potential stressors.

It is important to realise that the above ‘potential stressors’ in most cases contribute only a minor proportion of the total integrated pressure that ecosystems experience. The World Wildlife Fund (WWF) Living Planet Report (WWF, 2014) listed the relative attribution of threats contributing to the declines in animal populations as follows: 37% from exploitation (fishing, hunting etc.), 31% habitat degradation and change, 13% from habitat loss, 7% from climate change, and only 5% from invasive species, 4% from pollution and 2% from disease.

It has become clear over the past few years (e.g. at the EFSA 10th Anniversary Conference (EFSA, 2012)) that EFSA’s environmental risk assessment (ERA) schemes have evolved independently in the different areas within its remit (see EFSA, 2011), and that further harmonisation is desirable and possible on specific topics. EFSA therefore mandated (under mandate M-2013-0098) the Scientific Committee (SC) to harmonise EFSA’s ERA schemes with regard to: (1) accounting for biodiversity and ecosystem services when defining protection goals (PGs) for ERA; (2) coverage of endangered species as non-target organisms (NTOs) in single-stressor ERA; and (3) assessing temporal and spatial recovery of NTOs from effects of potential stressors in ERA. The SC therefore prepared three separate scientific documents to address the abovementioned issues.

The current opinion deals with the coverage of endangered non-target species in ERA and considered common approaches across EFSA’s areas of responsibility. The focus of this scientific opinion is on single-stressor ERA schemes under the remit of EFSA. The specific questions to be addressed include the following Terms of Reference (ToR):

ToR1: Are endangered species more vulnerable than other species

  • based on their toxicological sensitivity, probability/possibility of exposure, specific potential for recovery, low genetic diversity, or
  • because of other stressors, e.g. limited, marginal or fragmented habitat?

ToR2: Do the current ERA schemes appropriately cover endangered species?

ToR3: If not, what risk mitigation measures can be envisaged to prevent endangered species being put at risk from potential stressors?

ToR4: Is monitoring needed to check the efficacy of risk mitigation measures for the occurrence of endangered species?

Scientifically, there is no generally accepted definition for endangered species as endangerment is related to spatio-temporal scales. In this opinion, an endangered species is defined as a species that is either:

  1. listed in one or more ‘red lists’ as threatened (i.e. vulnerable, endangered, or critically endangered, or variants thereof), where the considered red lists are: (1) the European Red List, (2) the global International Union for Conservation of Nature (IUCN) Red List of Threatened Species, and (3) national and other regional red lists within Europe that follow the IUCN or another suitable classification scheme;
  2. rare based on the classification of Rabinowitz’s seven classes of rarity (including ‘endemics’, ‘classic rarity’, ‘habitat specialists’ and ‘truly sparse’ species).

Regarding question (ToR 1) Are endangered species more vulnerable than other species?, it was found that their endangerment might be due to particular characteristics that relate to vulnerability, namely (1) exposure, (2) recovery and/or (3) sensitivity to the potential stressor, directly or via indirect effects. No convincing scientific evidence was found indicating that endangered species have in general a higher exposure than other species, with the exception of top predators due to biomagnification. It appears that not the potential stressor or the endangered species per se may be decisive for ecological recovery from impact, but their interaction with (the properties of) the environments/landscapes impacted by stressors, in which endangered species (temporarily) dwell.

However, it seems that endangered species more often exhibit traits that are related to a decreased ability for recovery (e.g. they often have a slow life history). With respect to sensitivity against toxicological stressors, there is no evidence that endangered species are per se more sensitive towards regulated chemicals. However, as many of the endangered species are highly specialised, e.g. in their food or habitats, they may only have been exposed to a restricted range of natural chemicals, therefore resulting in the phylogenetic loss of certain detoxifying pathways relevant for assessed chemicals. The available data in the scientific literature do not allow concluding that endangered species generally suffer more from indirect effects than other related non-target species from potential stressors that fall under the remit of EFSA. However, the conditions described in this opinion suggest that endangered species that highly depend on obligate relationships with other species and/or are being part of complex ecological networks may suffer pronounced indirect effects. This may warrant a precautionary approach until more information becomes available for the specific endangered species under evaluation. In conclusion, question (1) cannot be answered in general, but anecdotal examples illustrate why, where and when endangered species can be more vulnerable than the species or the vulnerable taxa currently considered in ERAs. It is therefore important to identify these more vulnerable endangered species and to explicitly consider them in ERAs.

Regarding question (ToR2) Do the current ERA schemes appropriately cover endangered species?, the SC notes that many non-regulated factors and factors not subject to ERA by EFSA constitute subsets of threats that endangered species face. These factors include climate change, water contamination, soil erosion, nutrient stress in aquatic habitats, habitat destruction or fragmentation, or predator pressure in areas with decreased predator control. While it is clear that these cannot be regulated via the assessment of products which are deliberately placed on the market, the possible role of prospective ERA procedures in protecting endangered species should be considered by EFSA in addition to (or irrespective of) protection offered by measures in line with legislation like the Habitats Directive and Birds Directive. There are four types of potential stressors undergoing ERA within EFSA’s remit and (mainly) in an agricultural context: PPPs, GMOs, IAS and FAs. For GMO and IAS, the protection of endangered species is explicitly considered during the problem formulation phase in the respective ERA schemes. These ERA schemes allow a tailor-made assessment and the selection of one or more relevant endangered species. For PPPs, the EFSA Panel on PPPs and their Residues (PPR) adopted an approach to species selection for prospective risk assessment of an individually assessed pesticide using (or considering the option for) the concept of vulnerable species. Many endangered species are probably covered by the vulnerable species approach, although anecdotal observations suggest that some endangered species may be more vulnerable than those normally considered in ERA. While the vulnerable species concept takes account of exposure, sensitivity and recovery, it does not usually consider that the conservational state of a species can already be unfavourable. For FAs, the ERA does not tolerate population effects on any species in the environment and, thus, endangered species are implicitly included by the assumption that no FA is allowed on the market should a species be at risk. Thus, it currently varies among EFSA ERA schemes to which degree (implicit or explicit) endangered species are covered and how they are covered. The level of protection afforded by these four ERA schemes seems to vary for endangered species. However, the limited data availability does not allow the drawing of a firm conclusion and also does not allow an assessment of the level of protection achieved (regardless whether endangered species are implicitly or explicitly covered). Hence, current risk assessments are primarily conducted via selected standard and/or surrogate test species and it is assumed that the assessment factors applied offer a sufficient extrapolation to endangered species (bottom-up approach). Whether the assumption above is correct needs to be verified, e.g. by conducting landscape-level assessments (per potential stressor or for multiple stressors) that may include integrating all relevant experimental and monitoring data in spatial-explicit population models (top-down approach). Such an approach would need to account for the interaction of endangered species, stressors and the environmental properties on an appropriate spatio-temporal scale.

Regarding question (ToR3) What risk mitigation measures can be envisaged to prevent endangered species being put at risk from stressors resulting from the application of a regulated product?, the mitigation for and monitoring of endangered species can often be best addressed in a site-specific manner (e.g. by specific conservation areas for weeds or hamsters in specific crops; financial compensation of farmers to implement specific land-use requirements that favour endangered species) rather than generically. Two objectives of mitigation can be distinguished: (1) to better achieve a safe use of the product under assessment; (2) specific risk mitigation measures that can be proposed as a result of observations from monitoring schemes. Very often, farmers will be the in-field, and often the Endangered species in ERA 4 EFSA Journal 2016;14(2):4312 edge-of-field, risk managers; hence their education and training should be supported. The importance of risk mitigation measures should be emphasised and well communicated to farmers.

Finally, regarding question (ToR4) Is monitoring needed to check the efficacy of risk mitigation measures for the occurrence of endangered species?, the Scientific Committee (SC) considers it important to monitor the level of protection achieved by all management measures or mitigation measurements taken to protect endangered species (either compliance or supplementary monitoring). At present, only the GMO Panel is actively involved in regulated monitoring of potential stressors. For invasive species, surveillance and monitoring is advisable in any case. For PPPs and FAs, EFSA is currently not involved in monitoring. At the Member State (MS) level, information on chemical and biological monitoring, for instance conducted within the context of the Water Framework Directive (WFD), may inform the reregistration of PPPs.

During the scientific analysis for ToR2, scientific knowledge was collected on diverse approaches that are available to risk assessors. Without judging their necessity or implementation, potential approaches for extending the coverage of endangered species in ERA schemes include the following: Explicit inclusion of endangered species in ERA schemes requires a detailed specification of the PGs for endangered species, particularly in terms of what species (groups) should be protected, where, when and to what level. The establishment of these specific protection goals (SPGs) for endangered species requires a joint coordinated effort involving risk managers, risk assessors, scientists and other stakeholders.

Different approaches can be followed to cover endangered species in ERA schemes in EFSA’s remit. There is, however, not one single approach that suits all EFSA sectors (i.e. PPR Panel, GMO Panel, Panel on Plant Health (PLH) and Panel on Additives and Products or substances used in Animal Feed (FEEDAP)). For example, currently the surrogate species concept is frequently applied to assess GMOs, whereas a generic protection level in combination with a species-specific trait-based assessment (the vulnerable species concept) is more often used to assess PPPs.

Trait-based approaches, in which species traits are being used as indicators of potential (increased) risk, provide promising opportunities for including endangered species in ERA schemes. Further exploration and elaboration of the potential of this type of approaches is recommended, i.e.:

  • Identification and validation of species traits that, in combination with ecological conditions, drive the ecological vulnerability of endangered species for different types of stressors, i.e. traits related to exposure, stressor sensitivity, recovery and susceptibility to indirect effects.
  • Development of a systematic procedure in which species traits are being used to obtain a qualitative and/or quantitative estimate of the environmental risk of stressors for endangered species.
  • Construction of a species trait database that can be used as a basis to assess the contextdependent ecological vulnerability of endangered species for different types of stressors.

The rapid advancements in ‘omics’ and in silico techniques are resulting in large amounts of data that provide information about the molecular mechanisms and species traits driving the sensitivity of organisms to stressors.

Current practical and ethical limitations involved in testing endangered species in the field or the laboratory can be overcome if this type of information can be applied in a predictive way, i.e. to predict the sensitivity of species based on molecular traits regarding the phylogenetic relationships between endangered and non-endangered species of the same group (next generation RNA sequencing and whole genome sequencing). However, these novel techniques need to be further developed in order to check their potential use in ERA.

Mathematical models linking individual species traits and behaviours to populations, communities and landscapes provide a promising tool that can aid the risk assessment of potential stressors for endangered (if information on the actual impairment of the population is available at the ecologically relevant spatial scale) and other species.

Because the coverage of endangered species in ERA schemes cannot a priori be limited to one particular spatial scale, risk assessment might require ERAs to be conducted at different spatial scales. This also depends on the overlap between the sphere of influence of the potential stressor and the occurrence of the endangered species.

At the end of this opinion, further tools are recommended for covering endangered species in ERA. The SC supports to convene a stakeholder workshop for setting SPGs for endangered species. As the development of trait-based approaches for ERA is supported and current information for endangered species is fragmented, the SC recommends the establishment of an integrated database that can support the identification of relevant traits (e.g. through a systematic study to identify species traits that drive the vulnerability of endangered species), the centralisation of information and making it accessible and available to the public.

Additionally, it is suggested to develop methods and predictive tools that allow the assessment of potential effects of multiple stressors at different spatial scales for the improvement of ERA in general and also to support for the interpretation of chemical and biological monitoring data (to facilitate the feed-back mechanisms between prospective and retrospective approaches).

endangered species, environmental risk assessment, food production, plant protection products, genetically modified organisms, feed additives, invasive alien species
Print on demand
Number of Pages