Skip to main content
European Commission logo print header

Assessing and Monitoring the Impacts of Genetically modified plants on Agro-ecosystems

Periodic Report Summary 3 - AMIGA (Assessing and Monitoring the Impacts of Genetically modified plants on Agro-ecosystems)

Project Context and Objectives:
The AMIGA project operates in a hotly debated area for the EU where, unlikely other geographic areas in the world, the dispute on GM organisms is still polarized between “for and against” and the positions also differ among EU Member States. The biggest effort made by the members of the Consortium was to shed light on contentious issues (e.g. receiving environments, non-target organisms, the role of statistical significance vs. biological relevance) with a scientifically sound approach and to help risk assessment and risk management activities in Europe for the years to come. The environmental risk assessment (ERA) framework has recently been reinforced by the European Food Safety Authority (EFSA), whose guidance document (GD) requires harmonized and efficient methodologies. In this scenario, AMIGA project aimed to improve ERA process in Europe to reduce uncertainty in decision-making for the cultivation of GM plants by developing and verifying robust ERA methods, to increase confidence in the practicability of the EFSA ERA GD and to contribute to the development of more effective post-market environmental monitoring designs and risk mitigation procedures.

The project has based its work on two case studies tested in different biogeographical zones: GM maize resistant to Lepidoptera and cisgenic potato resistant to late blight. The diversity of selected zones in terms of environmental conditions, protection goals and ecosystem services providing taxa were investigated. The analysis of the most abundant non-target organisms (NTO) in the two agro-ecosystems in the five biogeographical regions detected several differences between regions.

Rhizosphere samples were collected from experimental field sites to characterize microbial biodiversity and the responsiveness of different functional guilds, microorganism, nematodes and earthworms, to GM maize and potato. The diversity of microorganisms present was clearly distinct between the two crops, demonstrating that plants have a strong effect on selecting specific communities of soil organisms. Differences in soil biodiversity across the European field sites were found leading to the conclusion that each rizosphere microbial community of maize and potato consisted of a local and geographically independent assemblages. Feeding studies with identified nematode focal species indicated for all selected endpoints no significant difference between GM and non-GM field plots.

AMIGA has developed protocols for evaluating the effect of GM-crops on NTO, herbivores, natural enemies and detritivores, in different European receiving environments. Experimental data enabled to verify the cost effectiveness of different methods and recommend commonly shared protocols for laboratory, as well as, field experiments. Protocols for in planta testing were implemented and evaluated by different laboratories during tritrophic experiments with maize and potato. Surveys in commercial fields triggered the recommendation that region-specific NTO-indicator species should be included in ERA protocols.

The design and validation of a cost-effective Lepidoptera monitoring protocol of farmland butterflies has demonstrated that its use for detecting adverse effects on butterflies is realistic and feasible.

To assess the potential risks due to GM crops for important pollinator species and their related ecosystem service, standardized sampling and testing methods suitable for ERA and long term monitoring in pollinator representative species, were established and evaluated. Digestibility and larval developmental time were used as additional measurement endpoints to evaluate cry toxin effects via pollen; this may represent an important addiction to the commonly used ERA protocols, thus closing an important knowledge gap in ERA. AMIGA contributed to the development and implementation of harmonized monitoring methods to gather quantitative baseline data for pollinator diversity in important crops before commercialisation of GM varieties. These data are essential for conducting PMEM over time. Pollen use and foraging distances of honey bee colonies were investigated to evaluate the potential exposure to GM crops in different agricultural landscapes. Results show that semi-natural habitats and mass-flowering crops can reduce foraging distances of honey bees at different scales and seasons with possible benefits for the performance of their colonies. Results could be used to setup mitigation measures, shortening foraging distances and to reduce the exposure risks to GM crops and/or additional environmental stressors, e.g. systemic chemical plant protection products.

Simulation models were built to anticipate possible scenarios at landscape level as consequence of locally adverse effects considering the possible introduction of insect resistant and herbicide tolerant GM crops. For example, preliminary results on prediction of the impact of Bt-maize cultivation on non-target lepidopteran populations, confirmed that pollen dispersion is the main driver of possible negative effects, but landscape patterns and crop management also affect the overall risk and could be used to refine mitigation measures. The study highlighted the sensitivity of the risk outcome to the eco-toxicological component of the model and namely to the actual deposition rate on host plants. Additional biological data from field experiments are therefore needed to reduce the level of uncertainty related to these components.
Similarly, risk management measures should be adapted to the specific cropping systems into which herbicide tolerant maize is to be introduced and monitoring needs to be adapted to actual management systems. AMIGA modelling approach helps to identify those endpoints that should primarily be monitored.

The project has provided a detailed guidance in the form of protocols for design and analysis of field experiments. The AMIGA Power Analysis software, that was developed as one of the main outcomes of the project, enables to complement the traditional testing for the statistical significance of differences with equivalence tests based on limits of concern. The software can calculate the power of difference and equivalence tests for the comparison of GM events and comparator varieties in field trials. As taxonomical endpoints with sufficient abundance may be very different between environments, it is advised to define a relevant set of endpoints for each environment. Results falling outside the equivalence region are to be interpreted as a trigger for further research of potential biological relevance. The proposed statistical analysis have been applied to data from AMIGA field work on maize and on potato.

Principles and methods were established and tested for investigating long term effects. Pre-existing trends were quantified, a major suite of indictors assembled and concepts established for setting safe ecological limits. Case studies, mainly of GM herbicide tolerant oilseed rape and blight resistant potato, showed that impacts of some innovations can be tracked for 10 years or more. However, the long term effects of existing trends and dynamics in cropping systems were likely to be far greater than effects of introducing the GM crops that have been tested to date in Europe.

A Guidance toolkit to support the selection of ERA and PMEM sites according to the criteria indicated in the EFSA ERA GD, was finally developed and is proposed to the attention of the stakeholders. AMIGA has implemented a prototype central database incorporating all the tools to aid ERA and PMEM which could have many applications beyond AMIGA.

Several tools and resources have also been developed to aid socio economic evaluations that can be potentially of use to policy-makers, technologies providers and to farmers. These include the implementation of a literature database related to GM crops, a bio-economic model that enables primarily the ex-ante investigation of the economic consequences of growing a range of GM crops in different regions and at individual farm level. A risk benefit analysis was conducted considering a thorough list of environmental, social and economic factors that might be evaluated prior the introduction of a GM crop in European.

A specific communication plan has supported the spread of information from AMIGA consortium to the media, the stakeholders and the public at large. AMIGA scientists have been involved for the duration of the project in many public events to disseminate the approach of AMIGA and to highlight the value of experimental data in supporting the biosafety evaluation of innovations in agriculture.

Project Results:
To prepare a list of relevant protection goals, a survey was conducted with experts from EU Member States. Protected species and areas were indicated as most relevant for environmental risk assessment. A literature surveys to assess the current status of knowledge on the trophic webs in maize and potato agro-ecosystems, including the EFSA arthropod data base. Data obtained in specific field surveys in different regions were obtained, also from areas not included in the existing literature data base, which are considered very useful to complement the existing information. Laboratory studies have being conducted using focal arthropod species selected according to the criteria indicated in the EFSA Guidance Document on the risk assessment for Genetically modified plants (GMPs). Specific protocols were developed and the effectiveness of in planta studies during environmental risk assessment has been demonstrated. Data collected in laboratory experiments and in the experimental fields allowed the preparation of a list of potential indicator species tailored to specific geographic regions.
A list of indicators for environmental changes due to agricultural practices was prepared based on existing data from 5 European regions. These indicators were then tested in other AMIGA work packages to select suitable indicators for long term effects. A multi-attribute decision model (DEXiES), a Process-based model of C and N changes in crop growth and a spatial upscaling model to derive regional possible impacts were produced and can provide useful support to risk managers.
Field experiments with GMPs were conducted with maize (5 countries) and potato (2 countries). Rhizosphere samples were collected from experimental field sites to characterize the responsiveness of different functional guilds of arthropods and microorganisms to GM crops. These activities have allowed the production of a wealth of biological data, which will enable to improve our knowledge of the two agro-ecosystems studied. Selection of above- and below-ground focal species of NTOs for laboratory, greenhouse and field experiments was finalized. Bioassays including herbivores and their natural enemies were performed, and measurement endpoints that are suitable predictor for immediate and delayed effects were selected.
The experimental designs and protocols for field and laboratory studies with non-target organisms were designed according to the requirement of the EFSA Guidance Document and validated during project’s activities. The statistical model previously prepared in the specific work package was used to support experimental design for difference and equivalence testing approach.
An inventory of existing butterfly monitoring schemes was completed and butterfly sampling was conducted for three following seasons in 3 countries. A field monitoring on commercial GMHT crops was conducted in Argentina on GM Sorghum, to characterize existing populations of pests that showed resistance to Cry toxins and herbicides. Area-wide samplings and modelling approaches (i.e. for estimating effects on non-target Lepidoptera and on weed abundance and diversity) were developed to indicate possible effective strategies to be used in post market environmental monitoring of GMPs.
Field experiments have been conducted to assess the feasibility of using Integrated Pest Management (IPM) programs, including GM maize or potato. Locally-appropriate practices have been included in the protocols for IPM to reduce pesticide use, and enhance ecosystem services. A IPM strategy was implement for management of potato late blight; the strategy builds on host resistance as the primary defence measure against the pathogen. Fungicides are not used as long as virulence is not found. When virulence to the R gene(s) is found, a low input fungicide spray strategy is used to mitigate the effects. Overall, the IPM control strategy developed reduced the average fungicide input, in a total of five field trials in two different countries by 80 - 90%.
An economic model has been developed to explore a range of GM adoption scenarios in Europe. Further development of case studies will successively demonstrate the financial impact of growing GMPs at farm level in representative EU regions. A risk-benefit analysis including economic, social and environmental factors has been conducted considering the two case studies (GM maize and potato) explored during AMIGA activities.
A constant dialogue with interested stakeholders and the wider public was conducted using different media: the project's web site, the production of newsletters, press releases and interviews. Specific actions were undertaken to communicate with targeted interest groups (e.g. the stakeholder consultation platform, the organization of summer schools).
Potential Impact:
The possible environmental impacts of GM crops are still a controversial issue in Europe, limiting the commercial adoption of the technology compared to other world areas. AMIGA has contributed to facilitate the implementation of the new principles included in the “Guidance on the environmental risk assessment of genetically modified plants” of the European Food Safety Authority (Deliverables 1.14 9.6) and to harmonize approaches, methods and tools among applicants (summarised in Deliverable 1.13).
The project has delivered a set of evaluated, standardised and harmonised cost effective sampling and testing methods on key aspects of environmental risk assessment such as: comparative assessment of potential impacts on non-target organisms (NTO), soil biology, and post-market monitoring.
In addition, the biogeographical variation across Europe was specifically addressed thanks to the concerted efforts of the project team to conduct experimental work in a diversity of geographic areas in the EU (Boreal, Atlantic, Central, Mediterranean and Balkans), including areas where the GM crops had not previously been grown. This enabled to consider the regional variability in terms of climatic, agricultural and socio-economic conditions for maize and potato agro-ecosystems , as well as refining receiving environments also in terms of arthropod biodiversity. The project with its activity of literature review and sampling in commercial potato and maize areas (including some for which no data were available) has established current baseline conditions of different bio-geographic regions in Europe with regard to ecological aspects important for Environmental Risk Assessment and PMEM. This effort will support tailoring regional approaches to Environmental Risk Assessment and PMEM, taking into account the differences in the ecology of the agroecosystems. AMIGA has also provided modelling tools to improve knowledge on potential long-term effects of GM crops
The methodologies developed during the project and the results obtained will provide scientific support to applicants, risk assessors, risk managers and regulators both at European and national level with benefits to the quality of safety assessment of future possible introductions of GM crops in Europe.
The tools developed and the results obtained were presented and discussed with the relevant stakeholders (European Commission, EFSA, EU Member States, EuropaBio, FAO, NGOs) during the AMIGA final event and they will become available to applicants, risk assessors and risk managers in Europe.

List of Websites: