BioRisk 7: 73-97 (20 | 2) ae ige oneiie iccess journa I
doi: 10.3897/biorisk.7.1969 RESEARCH ARTICLE &% B | O R IS

www.pensoftonline.net/biorisk

A framework for a European network for a systematic
environmental impact assessment of genetically
modified organisms (GMO)

Frieder Graef"””, Jorg Rombke’, Rosa Binimelis?, Anne Ingeborg Myhr’,
Angelika Hilbeck®, Broder Breckling®, Tommy Dalgaard’, Ulrich Stachow',
Georgina Catacora-Vargas*, Thomas Bohn**, David Quist*, Béla Darvas*”’,

Gert Dudel’, Bernadette Oehen'®, Hartmut Meyer'', Klaus Henle'?,
Brian Wynne'?, Marc J. Metzger'*, Silvio Knabe'®, Josef Settele'®,
Andras Székacs*”’, Angelika Wurbs', Jeannette Bernard",

Donal Murphy-Bokern'®, Marcello Buiatti'?, Manuela Giovannetti'’,
Marko Debeljak”®, Erling Andersen*', Andreas Paetz!'’, Saso Dzeroski”®,
Beatrix Tappeser”, Cornelis A.M. van Gestel”, Werner Wosniok”,
Gilles-Eric Séralini*®, Iulie Aslaksen”**, Roland Pesch®, Stanislav Maly?’,
Armin Werner!

| ZALE Leibniz Centre for Agricultural Landscape Research, Institute of Land Use Systems, Eberswalder Str.

84, 15374 Miincheberg, Germany 2. ECT Ockotoxikologie GmbH; Boéttgerstr. 2-14; 65439 Florsheim a.M.,

Germany 3 Center for Agro-food Economy and Development-CREDA-UPC-IRTA; Parc Mediterrani de la

Tecnologia- ESAB Building; C/ Esteve Terrades 8; 08860 Castelldefels (Barcelona), Spain 4 Gen@Ok; Centre
for Biosafety, Science Park, 9294 Tromso; Norway 5 ETH; Swiss Federal Institute of Technology, Institute of
Integrative Biology, Universitaetstr. 16; 8092 Ziirich; Switzerland 6 University of Vechta; Chair of Landscape
Ecology; Driverstr. 22; 49377 Vechta; Germany 7 Aarhus University; Department of Agroecology; Blichers Allé
20; 8830 Tjele; Denmark 8 Plant Protection Institute of the Hungarian Academy of Sciences; Department of
Ecotoxicology and Environmental Analysis; Herman Otto ut 15; 1022 Budapest; Hungary 9 TUD; Technische
Universitat Dresden; Faculty of Geo-, Forest- and Hydroscience; Helmholtzstr. 10; 01069 Dresden; Germany
10 FiBL; Forschungsinstitut fiir Biologischen Landbau; Ackerstr. 1; 5070 Frick; Switzerland \\_ ENSSER,

Postfach 1102, 15832 Rangsdorf, Germany 12 UFZ; Helmholtz Centre for Environmental Research; De-

partment of Conservation Biology; Permoserstr. 15; 04318 Leipzig; Germany 13 ESRC Cesagen, Lancaster
University; Sociology; Bailrigg; LAI 4YD Lancaster; UK \4 The University of Edinburgh; School of GeoScien-

ces; Drummond Street; Edinburgh EH8 9XP; UK 15 EAS; Eurofins Agroscience Services GmbH; Eutinger
Strasse 24; 75223 Niefern-Oschelbronn; Germany \6 UFZ; Helmholtz Centre for Environmental Research;

Department of Community Ecology; Theodor-Lieser-Str. 4; 06120 Halle; Germany \1 DIN; Deutsches In-

stitut fiir Normung; Burggrafenstr. 6; 10787 Berlin; Germany \8 Donal Murphy-Bokern; Lindenweg 12;

49393 Kroge-Ehrendorf; Germany \9 UDP; University of Pisa; Department of Crop Plant Biology; Via del
Borghetto 80, 56124 Pisa; Italy 20 JST; Josef Stefan Institute; Department of Knowledge Technologies; Jamova

39; 1000 Ljubljana; Slovenia 2\ University of Copenhagen; Faculty of Life Sciences; Rolighedsvej 23; 1958
Frederiksberg C; Denmark 22. BEN; Federal Agency for Nature Conservation; Division GMO-Regulation, Bio-

Copyright Frieder Graef et al. This is an open access article distributed under the terms of the Creative Commons Attribution License 3.0
(CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

74 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

safety; Konstantinstr. 110; 53179 Bonn; Germany 23 VU University, Amsterdam; Faculty of Earth and Life
Sciences; De Boelelaan 1085; 1081 HV Amsterdam; The Netherlands 24 University of Bremen, Department
of Mathematics and Computer Science; Bibliothekstr. 1; 28359 Bremen; Germany 25 CRIIGEN; University
of Caen; IBFA Laboratory of Biochemistry; Esplanade de la Paix ; 14032 Caen; France 26 Statistics Norway,

0033 Oslo; Norway 27 UKZUZ; Central Institute for Supervising and Testing in Agriculture; Foreign Rela-

tions and EU Department; Hroznovd 2; 65606 Brno; Czech Republik 28 Institute of Pharmacy, Faculty of
Health Sciences, University of Tromso; Norway 29 Central Food Science Research Institute; Herman Otto ut
15; 1022 Budapest; Hungary

Corresponding author: Frieder Graef (fgraef@zalf.de)

Academic editor: K. L. Heong | Received 26 August 2011| Accepted 16 March 2012 | Published 17 October 2012

Citation: Graef F Rombke J, Binimelis R, Myhr AI, Hilbeck A, Breckling B, Dalgaard T, Stachow U, Catacora-Vargas G,
Bohn T, Quist D, Darvas B, Dudel G, Oehen B, Meyer H, Henle K, Wynne B, Metzger MJ, Kniabe S, Settele J, Székacs
A, Wurbs A, Bernard J, Murphy-Bokern D, Buiatti M, Giovannetti M, Debeljak M, Andersen E, Paetz A, Dzeroski S,
Tappeser B, van Gestel CAM, Wosniok W, Séralini G-E, Aslaksen I, Pesch R, Maly S, Werner A (2012) A framework for a
European network for a systematic environmental impact assessment of genetically modified organisms (GMO). BioRisk

7: 73-97. doi: 10.3897/biorisk.7.1969

Abstract

The assessment of the impacts of growing genetically modified (GM) crops remains a major political
and scientific challenge in Europe. Concerns have been raised by the evidence of adverse and unexpected
environmental effects and differing opinions on the outcomes of environmental risk assessments (ERA).
The current regulatory system is hampered by insufficiently developed methods for GM crop safety testing
and introduction studies. Improvement to the regulatory system needs to address the lack of well designed
GM crop monitoring frameworks, professional and financial conflicts of interest within the ERA research
and testing community, weaknesses in consideration of stakeholder interests and specific regional condi-
tions, and the lack of comprehensive assessments that address the environmental and socio-economic risk
assessment interface. To address these challenges, we propose a European Network for systematic GMO
impact assessment (ENSyGMO) with the aim directly to enhance ERA and post-market environmental
monitoring (PMEM) of GM crops, to harmonize and ultimately secure the long-term socio-political
impact of the ERA process and the PMEM in the EU. These goals would be achieved with a multi-dimen-
sional and multi-sector approach to GM crop impact assessment, targeting the variability and complexity
of the EU agro-environment and the relationship with relevant socio-economic factors. Specifically, we
propose to develop and apply methodologies for both indicator and field site selection for GM crop ERA
and PMEM, embedded in an EU-wide typology of agro-environments. These methodologies should be
applied in a pan-European field testing network using GM crops. The design of the field experiments and
the sampling methodology at these field sites should follow specific hypotheses on GM crop effects and
use state-of-the art sampling, statistics and modelling approaches. To address public concerns and cre-
ate confidence in the ENSyGMO results, actors with relevant specialist knowledge from various sectors

should be involved.

Keywords
GM crops, field testing network, environmental risk assessment, post-market environmental monitoring,

typology of EU agro-environment, stakeholder involvement, socio-economic impact assessment

A framework for a European network for a systematic environmental impact assessment... 75

Introduction

Cultivation of genetically modified (GM) crops in European agriculture is, compared to
other developed countries, limited due to the significant public opposition and scientific
research on their potential adverse effects (Lemaire et al. 2010; Myhr 2010; FOE 2011).
The concerns centre on the potential risks of GM crop cultivation, evidenced by adverse
direct or indirect environmental and health effects (Heard et al. 2003; Giovannetti et al.
2005; Relyea 2005, Benachour and Séralini 2009; Graef 2009; Lang and Otto 2010;
Séralini et al. 2011). In relation to potential environmental effects in soil, a number of
unexpected research results have been reported, for instance on the transfer of engi-
neered genes from transgenic plants to soil bacteria (Gebhard and Smalla 1998; Nielsen
et al. 2000), and the release of insecticidal and fungicidal toxins by the roots of trans-
genic plants into the surrounding environment (Saxena et al. 1999; Turrini et al. 2004).

The resulting societal attention to risk demands a robust and independent regula-
tory system. The regulatory system that has evolved is subject to criticism, particularly
with regard to inadequately designed GM crop testing and introduction studies (Hil-
beck et al. 2011), and differing conclusions of the ERAs, for instance with respect to
health risks or nutritional assessment studies due to financial or professional conflicts
of interest (Diels et al. 2011). There has been insufficient attention given to full envi-
ronmental problem-formulation, protection and developmental goals, and other soci-
etal concerns (Nelson et al. 2009). This has fed scepticism. Other factors underpinning
uncertainties in the environmental safety of GM crops that have engendered public
distrust in regulatory practices around GM crops include a) conflicting or negative
results of GM crop effects on non-target organisms (NTO) (Castaldini et al. 2005;
Lovei and Arpaia 2005; Rosi-Marshall et al. 2007; Bohn et al. 2008, 2010), b) lacking
environmental baseline data from prospective GM crop cultivation areas as required
by Directive 2001/18/EC (European Commission 2001), c) poor monitoring designs
(De Jong 2010), d) missing studies and/or data relevant to the approval process (Graef
et al. 2010), e) undesirable impacts on organic farming (Binimelis 2008; Henle et
al. 2008), and f) the differing interpretations of Directive 2001/18/EC among EU
Member State authorities (B£N et al. 2011). Doubts have been raised about whether
the EU regulations, and especially their implementation, appropriately protect pub-
lic interest and goods, and are instead biased towards supporting unsustainable high
input agriculture. The insufficient involvement of local stakeholders and insufficient
transparency in regulatory processes feed the scepticism about GM crop import and
cultivation, and have led to polarized discussions and strong reactions from the public,
for instance destruction of field trials (Lemaire et al. 2010).

These shortcomings are partly related to lack of independent biosafety research and
to prevailing simplistic and sometimes misleading research approaches, which generally
undervalue the complex network of interactions governing ecosystem functions. The
selection of field sites, indicators, detection methods, assessment schemes and other com-
ponents among the ERA for GMOs often contain a significant degree of arbitrariness
(Hilbeck et al. 2011). In particular, monitoring of single trait transgenic proteins can be

76 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

burdened with substantial systematic errors (Székacs et al. 2010) rendering corresponding
literature data hardly comparable to each other. Such research and monitoring methods
require better standardisation among laboratories (Székacs et al. 2011) and should also
respect particular characteristics of the different receiving environments. Also, the insuf-
ficient consideration of regional particularities (Schermer and Hoppichler 2004; Graef
et al. 2005) and of the socio-economic context of European farming systems (Ohl et al.
2007; Binimelis et al. 2009) in many cases has contributed to questionable relevance of
field studies submitted in dossiers seeking approval from European authorities for field
trials, cultivation or import. Previous EU research in this area (European Commission
2010, 2011; Biota 2011) has placed little emphasis on these issues, despite the critical
importance of these aspects for achieving the desired outcomes from the EU Directives.

Requirements and challenges for a European-wide network for system-
atic GMO impact assessment (ENSyGMO)

According to the EU Directive 2001/18/EC, GM crops considered for placing on
the market must be subjected to satisfactory field testing at research and development
stages in all those ecosystems which could be affected by their use (European Com-
mission 2001). Furthermore, GM crop introduction into the environment must be
carried out following a precautionary approach by using the “step-by-step” principle,
gradually increasing the scale of release if data obtained at previous steps does not
provide evidence for biosafety concerns (Hilbeck et al. 2011). GM crop introductions
in the EU must follow this regulatory framework that requires a systematic environ-
mental risk assessment (ERA) and mandatory post-market environmental monitoring
(PMEM) after approval. While the ERA is primarily based on short-term and small-
scale introduction of the GM crops, PMEM is intended to handle uncertainties about
remaining potential adverse environmental effects after the ERA, comprising immedi-
ate, direct, indirect, delayed, long term as well as combinatorial and cumulative effects
(European Commission 2001). The approval process of GM crops in the EU must
consider both the sustainability of agricultural systems and environmental protection
goals. However, both intentions need long-term interdisciplinary perspectives and sys-
temic assessments, including social and economic ones, for generalising possible GM
crop impacts across the variable European agricultural and environmental conditions
(Ohl et al. 2007; Graef et al. 2010; BEN et al. 2011).

Taking a long-term view and allowing for systematic pre-release and continued
assessments of GM crops introduced into differing receiving agricultural and natural
environments will require that representative indicators are identified, developed, vali-
dated and harmonized with regard to the different ecological and socio-economic con-
texts within Europe. Also, detection methods and a process of selecting representative
field test sites across the biogeographic and agro-ecological regions and socio-economic
contexts of the EU need to be established in a transparent and scientifically sound
manner, taking into consideration specific regional protection goals.

A framework for a European network for a systematic environmental impact assessment... 77

To achieve these ends, we propose the establishment of a Europe-wide network
for systematic GMO impact assessment (ENSyGMO) that simultaneously targets the
following core issues:

¢ Harmonized and whenever possible standardized key indicators and sampling
methods to quantify possible impacts (EFSA 2010a). This leads to reliable and
comparable data within representative testing sites across Europe and can be
used as a scientific basis for a realistically differentiated EU-wide ERA.

e A representation of the variability of agro-ecosystems and its biological and so-
cio-economic components into which GM crops are proposed for introduction.

e Design of statistically robust representative field tests on the European scale,
and protocols for data analysis as a basis for the ERA and PMEM studies.
The challenge here is not so much to ensure the detection of adverse effects in
agricultural systems, but to discriminate measured effects with regard to cause-
effect relationships, for instance potential impact of GM cropping on other
agricultural practices, taking into account also the dynamics of agricultural
and environmental changes.

e Stakeholder involvement for i) communication of field test regions and sites;
ii) feedback from the relevant local actors, such as the farming communities
and bee keepers among others, on the design of comparisons (including iden-
tification of salient indicators) between GM cropping and non-GM cropping
systems; iii) a sound basis for socio-economic assessments and monitoring of
conflicts (Henle et al. 2008); and iv) effective dissemination of methods, pro-
cedures and approaches to the administration and decision makers, and other
stakeholders and users.

e Public and scientific validation on development, application and improvement
of ERA procedures and protocols through enhanced stakeholder involvement
and transparency.

We suggest establishing the ENSyGMO framework for the ERA (and in part for
the PMEM) using as the first cases the GM crops authorised for cultivation and com-
parative assessment with near isogenic lines or other conventional counterparts in re-
gionally differing agricultural systems with specific crop rotations. However, since con-
ventional non-GM agriculture may also create adverse effects, the assessment of these
effects should not be restricted to comparative approaches only, but include additional
sustainability criteria for agriculture and its environment. This will require modifica-
tions to existing frameworks. For example, the PMEM design may be inadequate to
cover such effects and will require a more advanced monitoring approach (BfN et al.
2011). ENSyGMO must aim primarily to create trust in its scientific independence,
robustness and societal utility. Accordingly, the participation of relevant stakeholders
from the public sector, researchers, and the private sector is central in the ENSyGMO
approach. Where appropriate this includes attunement of prevailing scientific indica-
tors and parameters to relevant stakeholder (e.g. farmer) knowledge and concerns.

78 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

Objectives of the ENSyGMO framework

The overall goal of the ENSyGMO framework is to design and apply harmonized pro-
cedures for detecting and analysing GM crop effects across the variability of European
agricultural environments and socio-economic contexts. A further key goal is to make
the EU regulatory framework as well as the appraisal of GM crop introduction propos-
als more scientifically, socially and technically robust. These overarching goals can be
broken down into the following objectives:

1) The development of a harmonized catalogue of evaluated indicator organisms,
from both a pan-European and regional view, based on defined criteria for identifying
indicators (e.g. functional groups, traits, communities, red list species, etc.) that cap-
ture possible impacts on biodiversity and other national or regional protection goals.

2) The development and validation of a harmonized catalogue of standardized
sampling, analyses and evaluation methods, as a basis for ERA and for possible long-
term PMEM studies.

3) The creation of a database of current agro-environmental (baseline) character-
istics of the main biogeographic regions in Europe, consisting of a) a biogeographical
inventory of indicator organisms and their variability across European agricultural ar-
eas and b) the typologies of agricultural systems and surrounding environments with
respect to potential GM crop introductions.

4) The design and establishment of a pan-European network of representative sites
tested and verified for ERA, and for long-term PMEM studies and representative for
EU biogeographic regions and farming systems.

5) The analysis of the socio-economic impacts of GM crop cultivation and its
management (e.g. including co-technology such as herbicides used) in relation to the
eco-social context of introduction, non-GM crop cultivation contexts, and regionally
differing agricultural practices.

6) The participation of a wide community of stakeholders representing diverse
social and ecological values and criteria of performance and the communication of the
ENSyGMO framework design, activities and results with all relevant stakeholders and
the public, beyond those already involved.

These objectives should be designed for regulatory authorities of relevant re-
gional to European levels, field assessors, farmer representatives, and scientists in the
relevant fields (inter alia agronomy, ecology, and socio-economy). The ENSyGMO
framework also includes an analysis of its potential to expand the network structures
and protocols, for instance the methods to derive appropriate indicators for specific
GM crops. A general task of the ENSyGMO framework is the development, testing
and application of the harmonized ENSyGMO outcomes to serve as a model and
basis for future ENSyGMO refinement, and for the development of other network
systems that assess and/or monitor technology and innovation impacts in agricul-
ture, environment and socio-economy, which are still missing in the EU (Henle et

A framework for a European network for a systematic environmental impact assessment... 79

al. 2008; The Royal Society 2009). The results gained by such a network could be
also used for other stressors in agricultural landscapes. For instance, the pesticide
registration procedure in EU requires distinguishing between bio-geographic re-
gions in Europe (European Commission 2009), yet its implementation is seriously
hampered by the lack of basic data on the composition of organism communities

(EFSA 2010b).

Design of the ENSyGMO framework

The ENSyGMO framework must encompass all relevant dimensions for a comprehen-
sive and regionally specific GM crop assessment scheme, including adaptability to fu-
ture scenarios and challenges. For designing and implementing the ENSyGMO frame-
work, we suggest six interlinked Thematic Clusters (TC) reflecting the main aforemen-
tioned objectives and directly leading to core products (Figure 1). The core products
are: a harmonized catalogue of evaluated indicator organisms and sampling methods
to quantify possible GM crop impacts (TC1); a database of baseline variability of EU
agro-ecosystems (TC2); an EU network scheme for statistically-based representative
field tests (I'C3); a socio-economic impact assessment framework (TC4); public and
stakeholder participation and dissemination, thus improved public legitimacy (T'C5);

core products

catalog:
harmonized indicators
catalog:
harmonized methods
database:
EU- baseline conditions

design, methodology:
pan-European network

analysis:
socio-economy of GM crops

meetings, website:
communication / dissem.

Figure |. Interrelationship of thematic clusters (TC) and core products in the ENSyGMO framework:
Indicator and sampling methods are selected (TC1) and baseline data and typology generated (TC2),
which are then integrated and validated in the field testing network under real agricultural field condi-
tions (C3) and the socio-economic context (C4). Based on the GM crop field testing results the field
network is successively adapted to represent the EU typology of European agro-ecosystems and biogeo-
graphic regions (TC2). Field trial results and supplementing GM cropping data, as well as the stakeholder
analysis (TC5), need to be included with the socio-economic impact assessment of the GM crops (TC4).
Between the TCs 1-4 there are feedback loops to iteratively and mutually adapt/improve the ENSyGMO
framework. All themes are integrated, evaluated and synthesised (TC6) to ensure the applicability of EN-
SyGMO products to the ERA, PMEM and SEIA regulatory frameworks.

80 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

and an integration and synthesis of the different scientific improvements across these
ENSyGMO domains, (TC6) particularly taking care of the need to transfer the EN-
SyGMO products to the regulatory framework.

Indicators and sampling methods (TC1)

The identification of indicators and sampling methods for detecting potential GM
crop impacts is a crucial step in the ERA methodology. A proposed ERA concept (Hil-
beck et al. 2008a, 2011) that was partially included in the EFSA (2010a) guidelines
places the whole GM organism at the centre of the assessment. The concept includes
potential adverse effects arising from direct and indirect exposure to the GM crop and
also secondary stressors such as inherent management practices to the specific GM
crops (e.g. the application of broad spectrum herbicides) (Andow and Hilbeck 2004).

To achieve a comprehensive and solid foundation for indicator identification, hy-
potheses and evidences about ecological impacts of GM crop cultivation in various
regions and environmental conditions have to be compiled. Both direct and indirect
effects must be covered. Direct effects include, for instance toxic effects on non-target
fauna, mainly invertebrates but also mammals and microbes (Relyea 2005; Giovan-
netti et al. 2005; Benachour and Séralini 2009). Indirect effects refer for instance to
altered rotation and other production schemes, pesticide applications rates and timing,
and tillage system (Graef 2009). Also combinatorial or cumulative effects, for instance
alterations in biodiversity or food webs, and pest-resistance development should be
covered (Heard et al. 2003). Furthermore, the relevant environmental compartments
(terrestrial both below- and above-ground, and aquatic systems) and land-use forms
(agricultural sites and other potential receiving environments) should be represented
(EFSA 2010a; BfN et al. 2011).

Indicators must be then selected in a step-wise process, which begins with iden-
tifying the most important ecological functions and protection goals relevant to sus-
tainability in agriculture and continues with the identification of possible exposure
pathways, relevant species in the local ecosystem, their suitability for testing, their
sampling methods, and their practical testing (Hilbeck et al. 2008b). Such indicators
are a) organisms at the species and/or community level including functional groups
such as earthworms (Bouché 1977), trophic groups such as nematodes (Yeates 2003)
or trait groups such as aquatic invertebrates (Liess and Beketov 2011); b) direct func-
tional endpoints such as litter decomposition, biogeochemical cycles completion; c)
indirect functional endpoints such as ecological functions provided by single species
or communities (Hilbeck 2008a, b; Schmitt-Jansen et al. 2008) and ecosystem ser-
vices such as biodiversity and habitat provision and/or pollinators securing food crop
production (Faber and van Wensem 2012; Mace et al. 2012); and d) landscape-scale
related indicators such as land use diversity which may be affected by altered rotation
and other production schemes (Graef 2009). They must represent not only arable ar-
eas, but neighbouring receiving environments including wild habitats, where the GM

A framework for a European network for a systematic environmental impact assessment... 81

crops may have a potential impact or could occur. There should be a representation
of at least a) the main environmental compartments (terrestrial below- and above-
ground, aquatic); b) functional groups such as predators, herbivores, saprophages,
and symbionts; and c) different physiological, taxonomical groups, for instance,
mainly arthropods but also oligochaetes, microbes and/or fungi (Hilbeck et al. 2008a;
Rombke et al. 2009).

Detecting possible impacts on indicators requires appropriate laboratory testing
methods suitable for the ERA, as well as field testing and monitoring methods (Hil-
beck et al. 2006, 2008b; EFSA 2010a). These methods should be preferentially stand-
ardised, for instance, by OECD, ISO, VDI or IOBC methods (Fink et al. 2006; VDI
2010). Depending on the selected indicator species, these methods may require modi-
fications or new methods must be developed. Sub-lethal endpoints, such as reproduc-
tion, should be included as criteria since they can also give indications of possible
long-term effects and are more sensitive than acute (lethal) harm (Rémbke et al. 2009).
The methods identified have to be examined in practice, preferably in inter-laboratory
comparison tests, and developed into a comprehensive testing protocol.

The hypotheses on GM crop effects need to be practically tested using the selected
indicators and laboratory methods. Current ERA procedures are expected to undergo
considerable improvement (EFSA 2010a). Lab tests should be performed with the GM
plant material as well as with mixtures of GM plant and conventional counterpart ma-
terial, and compared to a non-GM conventional counterpart. Test data-sets must be
statistically evaluated to control the test method performance under routine conditions
and to help focus subsequent field testing (Rémbke et al. 2009). Field tests are also
essential in higher tier evaluation, and must be performed, even if the proposed mode
of action is well understood and laboratory tests indicate no observed effect on a given
species (Romeis et al. 2011).

Finally, assessing socio-economic impacts of GM crops in European agro-ecosys-
tems and regions require specific indicators as part of the ENSyGMO framework.
These indicators need to combine the relevant socio-economic impact assessment
(SEIA) factors, GM crop environmental monitoring data, and the associated agricul-
tural management practices (Henle et al. 2008). Using regional rules derived inter alia
from the research we propose and/or scenarios for identification and measurement of
socio-economic indicators is particularly useful, for instance, relating to management
or co-existence inputs of GM crops compared to non-GM crops, in the specific eco-
social context of the GM crop receiving environment (Binimelis 2008).

Baseline conditions of European agro-ecosystems (TC2)

In 2008, the European Commission mandated the EFSA to develop methodologies
and recommendations for establishing relevant GM crop baseline information. The
guidance (EFSA 2010a), however, does not yet provide substantial improvement in
this regard. Europe covers a wide range of agro-environmental conditions, which are

82 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

reflected in a wide variety of agro-ecosystems with specific biodiversity, climate, land
use and management systems and agricultural productivity. Spatial classification of
information and geographical data is essential for their analysis and communication
(Metzger et al. 2005). Increased availability of spatial environmental data and ad-
vances in spatial data processing has led to a range of new European classifications
and typologies of biophysical regions (Hazeu et al. 2011; EEA 2011). However, few
attempts have been made to develop useful classifications and/or typologies focus-
sing on environmental impacts of agricultural innovations. This requires the linkage
of information on farming systems and information on the biophysical endowment
(Kempen et al. 2010).

For the ENSyGMO framework, we suggest establishing a comprehensive spatial
agro-ecosystems typology and regional baselines of EU agro-landscapes and the wider
potential receiving environments. This should build on or be co-developed with ex-
isting stratifications, typologies and classifications (Andersen et al. 2007; Petit et al.
2008; Hazeu et al. 2011; EEA 2011) with biophysical data relevant for discriminating
potential environmental GM crop effects on the previously identified indicators (TC1)
from the continuously changing agro-environment. Ecological information on habi-
tats, species, sites with local biological diversity importance, and protected areas should
also be integrated. Scale-related omissions in geographic regions represented, habitats,
ecosystems and taxa must be identified throughout the ENSyGMO framework in
order to collate additional data if possible (Dalgaard et al. 2003).

Baseline information primarily for the aforementioned environmental and socio-
economic indicators must be compiled at the European level to efficiently assess the
sensitivity of European regions and agro-ecosystems, particularly in relation to poten-
tial adverse effects of GM crops within differing protection, developmental and socio-
economic goals (Dziock et al. 2006). Other baseline information and indicators are
essential for explaining GM crop effects. These may relate, for instance, to characteris-
tics of farming systems, biophysical and ecological conditions for agro-ecosystems, and
protected wildlife and habitats, and ecosystem functions (Settele and Ktihn 2009; Bi-
ota 2011; Jansch et al. 2011), and should refer to EEA and OECD standards (OECD
2008; EEA 2010). Established environmental monitoring programmes (EMP) may
also provide baseline information needed for targeting field sites and for field testing.
EMPs are established and integrated, for instance, under the Water Framework Di-
rective, the Habitats Directive (Graef et al. 2008) and for the Long-Term Ecosystem
Research Network sites (LTER 2011) but exist also at the national level (Schmeller
and Henle 2008; EuMon 2011).

The ENSyGMO baseline information on European agro-ecosystems must be
managed and analysed with a geo-database including onthology, its access, main-
tenance and meta-data information. This geo-database should include the spatial
agro-ecosystems typology and the indicators determined (Andersen et al. 2007), and
should be accessible through standard database browsers and (Web-)GIS-programmes
(Kleppin et al. 2011).

A framework for a European network for a systematic environmental impact assessment... 83

EU wide network for GM crop field testing and monitoring (TC3)

Practical field testing and PMEM in the EU and worldwide are lacking a scientifically
robust and spatially representative design (BfN et al. 2011). GM crop introduction
trials in general are concentrated on one or a few locations only and are restricted to
short-term studies (Lévei and Arpaia 2005). These shortcomings, together with the
fact that GM plant approval dossiers sometimes crucially rely on tests done in non-EU
overseas regions are major reasons for public concerns and for the often contradictory
comments of EU Member State experts during GM crop application and decision-
making processes (Graef et al. 2010). The only larger scale experiments conducted this
far in the EU are the Farm Scale Evaluations in the UK (Firbank et al. 2003, Heard et
al. 2003). ‘Thus, the step-by-step principle of gradual spatial increase of the GM crop
introduction as required by Directive 2001/18/EC (European Commission, 2001) is
not implemented in practice. Methodologies for designing regionally representative
field tests are scarce (e.g. Stein and Ettema 2003; Graef et al. 2005); and networks for
carrying out these studies do not exist yet. Both ERA and PMEM require a representa-
tion of the variable European agro-environment (EFSA 2010a). Therefore, we consider
that the implementation of a comprehensive network approach such as the ENSyG-
MO in the near future is critical to address these deficiencies in current practice.

Given the inherent agro-biodiversity in the EU a fully functioning representative
network cannot be implemented right from the start. Rather, initially this has to be
a prototype requiring incremental adaptations and refinements, based on field test-
ing results and multi-/trans-disciplinary experience gained (Lindemayer and Likens
2009). Hence, the proposed ENSyGMO must be based on a statistically verified ex-
perimental field study design comprising a test site network. Being based on the agro-
environmental baselines and typologies developed, this design should have sufficient
power to explain the EU-wide variability of different indicators. The network should
not only cover present GM cultivation regions (FOE 2011) but include environments
where potential future GM crop cultivation could take place. Existing field test sites
of agricultural companies and/or authorities and of agricultural research institutions
could provide the core of such a network (Figure 2). The initial design, depending on
the typology outcome, may include 8-10 sites including sufficient replications (Figure
2). To achieve public acceptance of the GM crop cultivation and the assessment pro-
cess the involvement of local or other stakeholders into design and conducting the field
experiments is crucial (Lemaire et al. 2010).

Practical field testing using established indicators and sampling methods should be
done under controlled conditions in parallel at all sites. ENSyGMO sites will also serve as
facilities for socio-economic impact assessments (SEIA). To assess laboratory and field lev-
el indicators and methods for their suitability and extrapolation, field testing in additional
GM crop cultivation regions outside Europe, for instance those with significant levels of
GM crop cultivation, (e.g. Brazil, India and China) should be undertaken. It is expected
that at least the same taxonomic or functional groups can be used in the different areas.

84 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

Environmental Zone

[ALN - Alpine North
BOR - Boreal
[J NEM - Nemeral
[i ATN - Atlantic North
[> ALS - Alpine South
(ER CON - Continental

ATC - Atlantic Central
ll PAN - Pannonian
[__] LUS - Lusitanian
{| ANA - Anotolian
a MOM - Mediteranean Mountains
[9 MDN - Mediteranean North
{| MDS - Mediteranean South

HB network institutions
A field stations

Figure 2. Potential distribution of research institutions and available field research stations in an EN-
SyGMO representing the Environmental Stratification of Europe (Metzger et al. 2005). Field station

sites for practical testing of GM crops should be located in all agro-ecosystems relevant for GM cropping.

Managing the ENSyGMO field testing data requires a well-designed database, one
that can be used to analyse the ENSyGMO data in present and future ERA and PMEM
assessments (Reuter et al. 2010) and also store additional data collected from various
sources. [he data may also be used for feeding decision support systems with appropriate
inputs and for creating predictive models from the field testing data (Bohanec et al. 2008).

To achieve sustainability of an ENSyGMO framework, structural, financial, and
organisational requirements need to be identified. This analysis should also include
the potential for other institutions with wider environmental, ERA and monitoring
expertise to co-operate and give further methodological input. For long-term establish-
ment of the ENSyGMO sites, local or other relevant stakeholders are expected to be
involved (Lemaire et al. 2010).

Socio-economic impact assessment (TC4)

Analysis of the current research shows that a significant amount of work on SEIA is nar-
row in scope and contested in terms of assumptions, applied methodologies and find-

ings (Desquilbet and Bullock 2009; Demont et al. 2010; Glover 2010). We find that:

A framework for a European network for a systematic environmental impact assessment... 85

¢ most socio-economic impact research focuses on ex-ante and purely economic
parameters of a limited number of GM crops cultivated in only a few regions,
which creates a knowledge gap on the actual impacts after GM crops are intro-
duced into the environment (Smale et al. 2009);

¢ comprehensive comparative analysis between GM and non-GM crop pro-
duction systems (e.g. conventional, GMO-free, organic) along the produc-
tion chain and implications of co-existence (Coléno 2008) are missing. This
approach includes the analysis of entangled socio-economic and ecological
relationships. For instance, potential undesired processes, such as gene flow
resulting in transfer of GM pollen to honey, may have implications at the
commercial (e.g. honey with GM pollen would require a specific authorisa-
tion for its consumption) (European Court of Justice 2011), managerial (e.g.
beekeepers moving their bee colonies to other areas to avoid contamination)
(Lezaun 2011), and the ecological level (e.g. displaced bee colonies may sig-
nificantly reduce pollination of plants in agricultural and natural ecosystems);

¢ socio-economic impacts that might go beyond the monetary assessment analy-
sis are rarely considered (Binimelis 2008);

e usually adequate costs and benefits analyses are unfeasible, due to restricted
knowledge on both potential adverse effects and benefits of GM crops in the
medium and long term (Messéan et al. 2009);

¢ integrated socio-economic analysis with respect to other impacts, including
unintended environmental effects, usually is ignored (Pavone et al. 2011);

e only asmall community of researchers is involved in SEIA, restricting the vari-
ety of research perspectives and narrowing the range of methodologies mainly
to agro-economic aspects such as yield, prices, cost of production, profits and
consumer acceptance (Smale et al. 2009).

These shortcomings are tackled in the ENSyGMO framework in light of the
recognition of the multifunctionality of agriculture (The Royal Society 2009), by a)
including the intertwined relationship between the ecological and socio-economic
context where GM crops are introduced, and b) facilitating the participation of
relevant stakeholders as central in the ENSyGMO methodological approach for a
comprehensive SEIA.

Accordingly, the SEIA of the ENSyGMO framework includes, in a first step, de-
fining potential or observed adverse socio-economic impacts of GM crops. ‘This is
based on a) an analysis of the existing cases of GM crop introductions, b) existing
information and knowledge provided by integrated SEIA methods and experiences,
c) analysis of the socio-cultural and institutional context of GM crop introductions,
taking into account the private sector (farms, traders, supply chains) and the public
sector (national, regional governments and communities), d) identification of protec-
tion goals in relation to socio-economic welfare and sustainable development, and e)
identification of relevant knowledge gaps.

86 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

The SEIA framework of GM crops must be developed from the baseline informa-
tion obtained prior to or simultaneous with their introduction. This includes consulta-
tion with relevant private and public sector stakeholders. In the ENSyGMO frame-
work, relevant stakeholders refer to the different actors along the GM crop value chain
that are affected in monetary and non-monetary socio-economic terms and include,
inter alia, GM and non-GM farmers, the agribusiness sector (e.g. importers of inputs
for GM crop production and traders), local communities and families surrounding
the GM crop cultivation, consumers, policy and decision makers, and practitioners
of SEIA. The SEIA consultation process includes the identification of a) relevant en-
vironmental, cultural, institutional and political factors with socio-economic impacts;
b) socio-economic and development protection goals, c) monetary and non-monetary
implications at farm and community level (T'C6); and d) a decision support system for
a comprehensive and comparative assessment of socio-economic impacts of GM crops.

Finally, the SEIA of the GM crops tested within the ENSyGMO framework need
to be validated. This requires the implementation of the SEIA on the value chain of
the GM crops under the EU relevant scenarios, taking into account the regional spe-
cificities, environmental and socio-economic protection goals and data availability in
different test site regions. Moreover, complementary studies in GM crop commodity-
exporting countries such as Argentina or Canada should be included in order to attain
a more comprehensive knowledge base on the interrelated socio-economic pathways.
Adaptations of the SEIA framework based on the experience gained from implementa-
tion and feedback from actors and stakeholders involved, and integration in terms of
approach, findings, lessons learned and policy recommendations, would be the final
steps in this framework.

Communication and dissemination in the ENSyGMO framework (TC5)

The benefits and potential adverse effects of GM crops are highly contested in the sci-
entific, public and policy spheres. In these debates environmental and socio-economic
harm has typically been viewed as a purely scientific matter. In the framing of harm,
benefits and risks the analysis of social conditions (Myhr 2010) and human values
(Wynne 2001) also are considered. ‘These are relevant for assessing impacts (Felt et al.
2007) as it corresponds to the Problem-Framing within the Problem Formulation and
Options Assessment (PFOA) framework for ERA (Nelson et al. 2009). The ENSyGMO
framework recognizes the importance of the social, cultural and ethical factors relevant
to different actors and stakeholders. As a result the framework includes a two-way com-
munication approach as an essential component of good scientific research practice, as
well as for social and ethical considerations in policy and decision making (Dalgaard et
al. 2003). This allows a multi-sector and multi-disciplinary dialogue among the major
groups of stakeholders (i.e. private sector / policy-makers / researchers / civil society).
Communication within the ENSyGMO framework takes place in various forms,
including information provided through scientific conferences and other meetings,

A framework for a European network for a systematic environmental impact assessment... 87

scientific journals, policy reports and technical reports, also more proactive exchange
of knowledge (e.g. workshops). Moreover, the ENSyGMO methodology and results
— being cross-cutting issues — should be actively communicated to the full range of
GM crop ERA and SEIA practitioners, for instance, the European Commission and
national Competent Authority scientists, environmental agencies, land managers, and
policy-makers.

Communication in the ENSyGMO framework must also include pre-assessment
communication. This should refer to the dissemination of project aims, particularly
field trial objectives, design, requirements, and envisaged uses in conjunction with
the other ENSyGMO framework actors through a) an early-established multi-lingual
interactive website that is regularly updated and informs scientists, policy-makers, au-
thorities, NGOs and the interested public and also elicits public and other stakeholder
responses, and b) stakeholder-dialogue meetings in selected Member States (Myhr
2010). Long-term commitments of local and other stakeholders need to be established
for conducting the field trials and the agro-environmental and socio-economic moni-
toring.

Stakeholder knowledge and concerns should be included in the assessment plan-
ning itself (Wynne 2001). Both public and private stakeholder knowledge and con-
cerns about GM crops and site-specific or more widespread potential hazards, which
could contribute to the scientific ERA and the SEIA, need to be retrieved. This requires
analysis of the ENSyGMO interactive website, analysis of responses from the stake-
holder dialogue meetings, the combined evaluation of the ENSyGMO lab, field test
and SEIA results, and application of the PFOA framework (Nelson et al. 2009). It also
requires input from GM crop cultivation scenarios (TC6), and from EFSA stakeholder
and public consultation processes (Koutalakis et al. 2007).

The ENSyGMO framework should include training and capacity building through
approaches adapted to the different audiences, targeting a) the broad range of actors
and stakeholders; b) practitioners including social scientists, especially at postgraduate
levels; and c) EU, Member States and developing country policymakers. The purpose
is to provide understanding of the nature and conduct of such field trials and assess-
ments necessary to satisfy GM crop regulatory requirements.

Integration, evaluation and synthesis in the regulatory context (TC6)

The interpretation and implementation of the ERA and PMEM principles, as laid
down in Directive 2001/18 and Annex III of the Cartagena Protocol, is an ongoing
process defining and refining what is needed and how it can be achieved (Hilbeck et
al. 2008a,b, Myhr 2010, BEfN et al. 2011). Scientific, legislative and regulatory re-
quirements, as well as societal or political perceptions, frame the ERA, PMEM and
the SEIA. The ENSyGMO framework aims at providing sound data with the field
studies for the ERA and PMEM, providing appropriately harmonised and, if possible,
standardised methods and indicators for detection and analysis (VDI 2010). It aims at

88 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

transferring the ERA mainly based on short-term observations to an ecosystem-based
integrated assessment of the GM crop impact on the farming systems, environment
and socio-economic context specifics. The ENSyGMO framework also requires con-
tinuous review and adaptation including and targeting new and/or unforeseen devel-
opments, new knowledge, and change in cultivation practice and field sites in the EU.
Accordingly, the ENSyGMO framework is an iterative process for constantly review-
ing and improving research hypothesis and methods in the ERA, PMEM and SEIA.

Hence, an integrating platform is required to create maximum impact and usabil-
ity of core ENSyGMO products (Figure 1) for the existing ERA, PMEM and SEIA
frameworks. ‘This platform requires the permanent involvement and inputs of all EN-
SyGMO partners and vice versa. The core objective of this platform is the development
and synthesis of a comprehensive, interdisciplinary scenario framework for GM crops
adoption and associated changes in EU agriculture. This is based on inputs by the EN-
SyGMO actors and themes (indicator organisms, sampling methods, baseline require-
ments, overall network design, socio-economic impact, stakeholder dialogue) and on
previous or ongoing EU scenario oriented projects, for instance, ALARM, SEAMLESS
and SENSOR (Rounsevell and Metzger 2010). The scenario framework should be
applied to check the usability and predictive power of the ENSyGMO results and for
deriving suggestions for the ERA, PMEM and SEIA frameworks. For instance, GM
crop cultivation scenarios could serve to extrapolate the results on field testing, regional
protection goals, and regional farming systems to possible future situations.

The ENSyGMO lab and field studies require synthesis and comparison to other
GM crops, and other studies (including peer-reviewed literature) to attain an overall
ERA, PMEM and SEIA of GM and non-GM farming systems in the EU. ‘This entails
the following steps, a) process-related findings using tested assessment methodologies
and protocols from lab and field trials. The pros and cons, as well as required technical,
financial and person-power input of each methodology should be analysed and recom-
mendations for use formulated; b) synthesis of lab and field trial data and results using
state-of-the-art statistics. [he predictive power of lab studies indicating environmental
impact need to be evaluated by modelling and comparing the ENSyGMO and other
lab data to field findings; c) evaluating remaining uncertainties and their impact on the
accuracy of the ERA; and d) the evaluation of lab and field trial and other data using
the SEIA framework.

The transferability of the ENSyGMO framework results to the existing ERA,
PMEM and SEIA framework in the EU and the Cartagena Protocol on Biosafety has
to be ensured. ‘Therefore, the TCs should be monitored and supported from the begin-
ning to provide appropriate baseline information, methodologies and input to exist-
ing protocols. The ENSyGMO outputs require synthesizing and fulfilment of their
objectives for use by decision makers and relevant stakeholders on regional, Member
States, European and global scales. The final recommendations require a consensus-
building process within the ENSyGMO framework. Representing different interests
of diverse groups in society and regulatory science requires transparency, accountabili-

ty and participation of stakeholders along the ENSyGMO implementation. A series of

A framework for a European network for a systematic environmental impact assessment... 89

stakeholder meetings are required for applying and enhancing the PFOA methodology
(Nelson et al. 2009). Finally, the PFOA needs to be tested as a tool for accompanying
field introduction trials and PMEM by validating the outcomes of the ERA against the
initial ERA assumptions.

Outlook

The ENSyGMO framework endeavours to address the many concerns about GM
cropping systems. These concerns centre, for instance, on inadequately designed GM
crop testing studies and PMEM, the lack of environmental baseline data and repre-
sentativeness, non-consideration of regional environmental and socio-economic spe-
cifics, the conflicting interpretation and under-implementation of EU regulations, and
the poor involvement of local and other stakeholders. It is, however, neither a “cure-
all” for addressing conflicts, nor can it provide answers to all uncertainties connected
to GM agriculture. However, the ENSyGMO can provide a long-term scientifically
sound basis for the ERA of GM crops and for long-term monitoring studies in the EU.
For the proper PMEM as defined by the EU Directive 2001/18/EC additional sites
in real GM cropping regions and farming systems are required, generally with a more
prolonged timeline. The ENSyGMO framework as proposed in detail here requires
field implementation and validation to effectively contribute to broadening the scope
of requirements and potentials linked to the ERA, PMEM, SEIA and the regulatory
framework of GM crops. ENSyGMO is a flexible framework that will be improved
based on the experience gained, the changing contexts and the development of novel
GM crops. As a result and taking into account that the framework operates on a case-
by-case and step-by-step basis, an additional outcome of ENSyGMO is the potential
for organizing permanent or ad hoc expert working teams or sub-networks, depending
on the GM crop (e.g. Bt-Maize working teams) or the potential impacts (e.g. biodi-
versity or socio-economic impacts working team). The ENSyGMO framework is not
only applicable to GM crop impact assessment, but to assessing and monitoring the
implementation, impact and sustainability of EU policies and/or impacts of other ag-
ricultural technologies and innovations, (e.g. synthetic fertilisers, harvesting systems,
plant protection products and the production of non-food crops). In particular, the
ERA of plant protection products, currently under review, would clearly benefit from

the activities in the ENSyGMO framework.

Acronyms
EEA European Environment Agency
EFSA European Food Safety Authority

ENSyGMO European Network for Systematic GMO impact assessment

ERA environmental risk assessment

90 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

GMO genetically modified organism

IOBC Int. Organisation for Biological Control of Noxious Animals and Plants
ISO International Organization for Standardization

OECD Organisation for Economic Co-operation and Development

PFOA Problem formulation and options assessment

PMEM post-market environmental monitoring

SEIA socio-economic impact assessment

TC thematic cluster

VDI Verein Deutscher Ingenieure

Acknowledgments

Funding from the Federal Ministry of Food, Agriculture and Consumer Protection
(BMELV) and the Ministry of Infrastructure and Agriculture (MIL) Brandenburg has
supported this work. This article is the outcome of a consortium of authors sum-
marizing their jointly developed concept for a pan-European framework for the sys-
tematic assessment of GMO impacts (ENSyGMO) submitted to the 7° Framework
Programme of the European Union. While another project was selected for funding,
the consortium members wished to put the outcome of their joint effort forward for
further discussion and possible uptake or inspiration to a wider community of sci-
entists, regulators and interested stakeholders. We maintain that such a network is
urgently needed not only for GMO impact assessment but also for other agricultural
policies that require science-based EU-wide oversight. We also would like to express
our gratitude for the critical and constructive comments received by three anonymous
reviewers.

References

Andersen E, Elbersen B, Godeschalk F, Verhoog D (2007) Farm management indicators and
farm typologies as a basis for assessments in a changing policy environment. Journal of
Environmental Management 82: 353-362. doi: 10.1016/j.jenvman.2006.04.021

Andow DA, Hilbeck A (2004) Science-Based Risk Assessment for Nontarget Effects of Trans-
genic Crops. BioScience 54(7): 637-649. doi: 10.1641/0006-3568(2004)054[0637:SRA
FNE]2.0.CO;2

Benachour N, Séralini GE (2009) Glyphosate formulations induce apoptosis and necrosis in hu-
man umbilical, embryonic, and placental cells. Chemical Research in Toxicology 22: 97—
105. doi: 10.1021/tx800218n

BfN, FOEN, EEA (2011) Monitoring of genetically modified organisms: A policy paper represent-
ing the view of the National Environment Agencies in Austria and Switzerland and the Fed-
eral Agency for Nature Conservation in Germany. Umweltbundesamt-Reports 305 (Vienna):
1-55.

A framework for a European network for a systematic environmental impact assessment... 91

Binimelis R (2008) Coexistence of plants and coexistence of farmers: is an individual choice
possible? Journal of Agricultural and Environmental Ethics 21: 437-457. doi: 10.1007/
s10806-008-9099-4

Binimelis R, Monterroso I, Rodriguez-Labajos B (2009) Catalan agriculture and genetically
modified organisms (GMOs) — An application of DPSIR model. Ecological Economics
69(1): 55-62. doi: 10.1016/j.ecolecon.2009.02.003

Biota (2011) Biota list of FP biodiversity projects: http://www.edinburgh.ceh.ac.uk/biota/

Bohanec M, Messean A, Scatasta S, Angevin F Griffiths B, Krogh PH, Znidar’i¢ M, Dzeroski
S (2008) A qualitative multi-attribute model for economic and ecological assessment of
genetically modified crops. Ecological Modelling 215(1—3): 247-261. doi: 10.1016/j.ecol-
model.2008.02.016

Bohn T, Primicerio R, Hessen DO, Traavik T (2008) Reduced fitness of Daphnia magna fed
a Bt-transgenic maize variety. Archives of Environmental Contamination and Toxicology
55: 584-592. doi: 10.1007/s00244-008-9150-5

Bohn T, Traavik T, Primicerio R (2010) Demographic responses of Daphnia magna fed trans-
genic Bt-maize. Ecotoxicology 19: 419-430. doi: 10.1007/s10646-009-0427-x

Bouché MB (1977) Stratéegies lombriciennes. In: Lohm U, Persson T (Ed) Soil organisms as
components of ecosystems. Ecological Bulletins NFR 25: 122-132.

Castaldini M, Turrini A, Sbrana C, Benedetti A, Marchionni M, Mocali S, Fabiani A, Landi
S, Santomassimo F, Pietrangeli B, Nuti MP, Miclaus N, Giovannetti M (2005) Impact of
Bt corn on rhizospheric and soil eubacterial communities and on beneficial symbiosis in
experimental microcosms. Applied and Environmental Microbiology 71: 6719-6729. doi:
10.1128/AEM.71.11.6719-6729.2005

Coléno FC (2008) Simulation and evaluation of GM and non-GM segregation management
strategies among European grain merchants. Journal of Food Engineering 88(3): 306-314.
doi: 10.1016/j.jfoodeng.2008.02.013

Dalgaard T, Hutchings NJ, Porter JR (2003) Agroecology, scaling and interdisciplinarity. Agri-
culture Ecosystems and Environment 100: 39-51. doi: 10.1016/S0167-8809(03)00152-x

De Jong T (2010) General surveillance of genetically modified plants in the EC3 and the need
for controls. Journal ftir Verbraucherschutz und Lebensmittelsicherheit 5: 181-183. doi:
10.1007/s00003-009-0547-5

Demont M, Dillen K, Daems W, Sausse C, Tollens E, Mathijs E (2010) On the Proportiona-
lity of EU Spatial ex ante Coexistence Regulations: Reply. Food Policy 35(2): 183-184.
doi: 10.1016/j.foodpol.2010.03.001

Desquilbet M, Bullock DS (2009) Who Pays the Cost of GMO Segregation and Identity Pres-
ervation? American Journal of Agricultural Economics 91: 656-672. doi: 10.1111/j.1467-
8276.2009.01262.x

Diels J, Cunha M, Manaia C, Sabugosa-Madeira B, Silva M (2011) Association of financial
or professional conflict of interest to research outcomes on health risks or nutritional
assessment studies of genetically modified products. Food Policy 36(2): 197-203. doi:
10.1016/j.foodpol.2010.11.016

92 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

Dziock F, Henle K, Foeckler F, Follner K, Scholz M (2006) Biological indicator systems
in floodplains — a review. International Review of Hydrobiology 91(4): 271-291. doi:
10.1002/iroh.200510885

EFSA (European Food Safety Authority) (2010a) Guidance on the environmental risk assess-
ment of genetically modified plants, EFSA Journal 8(11):1879.

EFSA (European Food Safety Authority) (2010b) Scientific Opinion on the development of a
Soil Ecoregions concept. EFSA Journal 20(8): 1820.

European Commission (2001) Directive 2001/18/EC of the European Parliament and of the
Council of 12 March 2001 on the deliberate release into the environment of genetically
modified organisms and repealing Council Directive 90/220/EEC (Luxembourg, Publica-
tions Office) Official Journal of the European Communities 44(L106): 1-39.

European Commission (2009) Regulation (EC) no 1107/2009 of the European Parliament
and of the Council of 21 October 2009 concerning the placing of plant protection prod-
ucts on the market and repealing Council Directives 79/117/EEC and 91/414/EEC. Of-
ficial Journal of the European Union L309: 1-50.

European Commission (2010) A decade of EU-funded GMO research (2001 — 2010). Luxem-
bourg, Publications Office of the European Union, 1-264.

European Commission (2011) Environment — including Climate change. http://cordis.europa.
eu/fp7/environment/home_en.html

EEA (European Environment Agency) (2010): Biodiversity — 10 messages for 2010. http://
www.eea.europa.eu/publications/10-messages-for-2010

EEA (European Environment Agency) (2011) Biogeographical regions in Europe. http://www.
eea.europa.eu/data-and-maps/figures/main-threats-to-biodiversity-by-biogeographic-region

EuMon (2011) EU-wide monitoring methods and systems of surveillance for species and habi-
tats of Community interest. http://eumon.ckff:si/index1.php

European Court of Justice (2011) Judgement of the Court (Grand Chamber) of 6 September
2011. Karl Heinz Bablok and others v Freistaat Bayern. Case C-442/09. http://curia.eu-
ropa.eu/juris/liste.jsfelanguagesen&num=C-442/09

Faber JH, van Wensem J (2012) Elaborations on the use of the ecosystem services concept for
application in ecological risk assessment for soils. The Science of the Total Environment
415: 3-8. doi: 10.1016/j.scitotenv.2011.05.059

Felt U, Wynne B, Callon M, Gongalves ME, Jasanoff S, Jepsen M, Joly PB, Konopasek Z,
May S, Neubauer C, Rip A, Siune K, Stirling A, Tallacchini M (2007) Taking European
Knowledge Society Seriously. Report of the Expert Group on Science and Governance, to
the Science, Economy and Society Directorate, Directorate-General for Research, Euro-
pean Commission. European Commission, Brussels, Belgium, 1—95.

Fink M, Seitz H, Beismann H (2006) Concepts for General Surveillance: VDI Proposals.
Standardisation and harmonisation in the field of GMO-Monitoring. Journal for Con-
sumer Protection and Food Safety 1, Suppl. 1: 11-14.

Firbank LG, Heard MS, Woiwod IP, Hawes C, Haughton AJ, Champion GT, Scott RJ, Hill
MO, Dewar AM, Squire GR, May MJ, Brooks DR, Bohan DA, Daniels RE, Osborne
JL, Roy DB, Black HIJ, Rothery P, Perry JN (2003) An introduction to the Farm-Scale

A framework for a European network for a systematic environmental impact assessment... 93

Evaluations of genetically modified herbicide-tolerant crops, Journal of Applied Ecology
40: 2-16. doi: 10.1046/j.1365-2664.2003.00787.x

FOE (2011) GM crops continue to fail in Europe. www.foeeurope.org/GMOs/download/
FoEE_Who_benefits_fact_sheet.pdf

Gebhard F, Smalla K (1998) Transformation of Acinetobacter sp. Strain BD413 by transgenic
sugar beet DNA. Applied and Environmental Microbiology 64: 1550-1554.

Giovannetti M, Sbrana C, Turrini A (2005) The impact of genetically modified crops on soil
microbial communities. Biology Forum 98: 393-418.

Glover D (2010) Is Bt Cotton a Pro-Poor Technology? A Review and Critique of the Em-
pirical Record. Journal of Agrarian Change 10(4): 482-509. doi: 10.1111/j.1471-
0366.2010.00283.x

Graef F (2009) Review: Potential environmental effects of altering cultivation practice with ge-
netically modified herbicide-tolerant oilseed rape and implications for monitoring. Agron-
omy for Sustainable Development 29: 31-42. doi: 10.1051/agro:2007055

Graef F, De Schrijver A, Murray B (2008) GMO monitoring data coordination and harmo-
nisation at the EU level — Outcomes of the European Commission Working Group on
Guidance Notes supplementing Annex VII of Directive 2001/18/EC. Journal for Con-
sumer Protection and Food Safety 3, Supplement 2: 17—20.

Graef F, Zighart W, Hommel B, Heinrich U, Stachow U, Werner A (2005) Methodological
scheme for designing the monitoring of genetically modified crops at the regional scale. En-
vironmental Monitoring and Assessment 111(1—3): 1-26. doi: 10.1007/s10661-005-8044-5

Graef F, Schiitte G, Winkel B, Teichmann H, Mertens M (2010) Scale implications for en-
vironmental risk assessment and monitoring of the cultivation of genetically modified
herbicide-resistant sugar beet: A review. Living Rev. Landscape Res. 4, 3. http://www.
livingreviews.org/Irlr-2010-3

Hazeu GW, Metzger MJ, Miicher CA, Perez-Soba M, Renetzeder Ch, Andersen E (2011) Eu-
ropean environmental stratifications and typologies: An overview. Agriculture, Ecosystems
and Environment 142: 29-39. doi: 10.1016/j.agee.2010.01.009

Heard MS, Hawes C, Champion GT, Clark SJ, Firbank LG, Haughton AJ, Parish AM, Perry
JN, Rothery P, Roy BA, Scott RJ, Skellern MP, Squire GR, Hill MO (2003) Weeds
in fields with contrasting conventional and genetically modified herbicide-tolerant crops.
I. Effects on individual species, Philosophical Transactions of the Royal Society B 358:
1833-1846. doi: 10.1098/rstb.2003.1401

Henle K, Alard D, Clitherow J, Cobb P, Firbank L, Kull T, McCracken D, Moritz RFA,
Niemela J, Rebane M, Wascher D, Watt A, Young J (2008) Identifying and managing the
conflicts between agriculture and biodiversity conservation in Europe — a review. Agricul-
ture Ecosystems and Environment 124: 60—71. doi: 10.1016/j.agee.2007.09.005

Hilbeck A, Andow DA, Arpaia S, Birch ANE, Fontes EMG, Lévei GL, Sujii E, Wheatley
RE, Underwood E (2006) Methodology to support non-target and biodiversity risk as-
sessment. In: Hilbeck A, E Fontes, Andow DA (Eds) Environmental Risk Assessment of
Transgenic Organisms, Volume 2: Methodologies for assessing Bt cotton in Brazil. CABI
Publishing, Wallingford, UK, 108-132.

94 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

Hilbeck A, Jansch S, Meier M, Rombke J (2008a) Analysis and validation of present ecotoxi-
cological test methods and strategies for the risk assessment of genetically modified plants.
BEN Skript 236. Federal Agency for Nature Conservation, Bonn — Bad Godesberg, 1-138.

Hilbeck A, Meier M, Benzler A (2008b) Identifying indicator species for post-release monitor-
ing of genetically modified, herbicide resistant crops. Euphytica 164(3): 903-912. doi:
10.1007/s10681-008-9666-9

Hilbeck A, Meier M, Rombke J, Jansch S, Teichmann H, Tappeser B (2011) Environmental
Risk Assessment of Genetically Modified Plants —- Concepts and Controversies. Environ-
mental Sciences Europe 23(13). doi: 10.1186/2190-47 15-23-13

Jansch S, Rombke J, Hilbeck A, WeifS G, Teichmann H, Tappeser B (2010) Classification
of the receiving environment in the context of environmental risk assessment (ERA) of
genetically modified plants (GMP) in Europe. BioRisk 6: 19-40.

Kempen M, Elbersen BS, Staritsky I, Andersen E, Heckelei T (2010) Spatial allocation of farm-
ing systems and farming indicators in Europe. Agriculture, Ecosystems & Environment
142: 51-62. doi: 10.1016/j.agee.2010.08.001

Kleppin L, Schmidt G, Schréder W (2011) Cultivation of GMO in Germany: support of
monitoring and coexistence issues by WebGIS technology. Environmental Sciences Eu-
rope 23(4): doi: 10.1186/2190-4715-23-4

Koutalakis C, Wendler F, Borras S (2007) European Agencies and Input Legitimacy: EFSA,
EMEA and EPO in the Post-Delegation Phase. Journal of European Integration 29 (5):
583-600. doi: 10.1080/07036330701694899

Lang A, Otto M (2010) A synthesis of laboratory and field studies on the effects of transgenic
Bacillus thuringiensis (Bt) maize on non-target Lepidoptera. Entomologia Experimentalis
et Applicata 135: 121-134. doi: 10.1111/j.1570-7458.2010.0098 1.x

Lemaire O, Moneyron A, Masson JE, The Local Monitoring Committee (2010) “Interactive
Technology Assessment” and Beyond: the Field Trial of Genetically Modified Grapevines
at INRA-Colmar. PLoS Biol 8(11): €1000551. doi: 10.1371/journal.pbio. 1000551

Lezaun J (2011) Bees, beekeepers, and bureaucrats: parasitism and the politics of transgenic
life. Environment and Planning D: Society and Space 29: 738-756. doi: 10.1068/d0510

Liess M, Beketov M (2011) Traits and stress-keys to identify community effects at low toxicant
level. Ecotoxicology 20: 1328-1340. doi: 10.1007/s10646-01 1-0689-y

Lindemayer DB, Likens GE (2009) Adaptive monitoring: A new paradigm for long-term re-
search and monitoring. Trends in Ecology and Evolution 24(9): 482-486. doi: 10.1016/j.
tree.2009.03.005

Lévei GL, Arpaia S (2005) The impact of transgenic plants on natural enemies: a critical
review of laboratory studies. Entomologia Experimentalis et Applicata 114: 1-14. doi:
10.1111/j.0013-8703.2005.00235.x

LTER (2011) European Long-Term Ecosystem Research Network. http://www.lter-europe.net

Mace GM, Norris K, Fitter AH (2012) Biodiversity and ecosystem services: a multilayered
relationship. Trends Ecol Evol 27(1): 19-26. doi: 10.1016/j.tree.2011.08.006

Messéan A, Squire GR, Perry JN, Angevin F, Gomez-Barbero M, Townend D, Sausse C,
Breckling B, Langrell S, Dzeroski S$, Sweet JB (2009) Sustainable introduction of GM

A framework for a European network for a systematic environmental impact assessment... 95

crops into european agriculture: a summary report of the FP6 SIGMEA research project.
Oléagineux, Corps Gras, Lipides. 16 (1): 37-51.

Metzger M, Bunce B, Jongman R, Miicher S, Watkins JW (2005) A climatic stratification
of the environment in Europe. Global Ecology and Biogeography 14: 549-563. doi:
10.1111/j.1466-822X.2005.00190.x

Myhr AI (2010) A Precautionary Approach to Genetically Modified Organisms: Challenges
and Implications for Policy and Science. Journal of Agricultural and Environmental Ethics
23: 501-525. doi: 10.1007/s10806-010-9234-x

Nelson K, Andow DA, Banker MJ (2009) Problem Formulation and Option Assessment
(PFOA) Linking Governance and Environmental Risk Assessment for Technologies: A
Methodology for Problem Analysis of Nanotechnologies and Genetically Engineered Or-
ganisms. The Journal of Law, Medicine & Ethics 37(4): 732-748. doi: 10.1111/j.1748-
720X.2009.00444.x

Nielsen KM, van Elsas JD, Smalla K (2000) Transformation of Acinetobacter sp. Strain BD413
(pFG4nptll) with Transgenic plant DNA in Soil Microcosms and Effects of Kanamycin
on Selection of Transformants. Applied and Environmental Microbiology 66: 1237-1242.
doi: 10.1128/AEM.66.3.1237-1242.2000

OECD (2008) Paying for Biodiversity. Enhancing the cost-effectiveness of payments for eco-
system services: Organization for Economic Co-operation and Development, Paris. www.
oecd.org/env/biodiversity/pes

Ohl C, Krauzeb K, Griinbithel C (2007) Towards an understanding of long-term ecosystem
dynamics by merging socio-economic and environmental research — Criteria for long-
term socio-ecological research sites selection. Ecological Economics 63: 383-391. doi:
10.1016/j.ecolecon.2007.03.014

Pavone V, Goven J, Guarino R (2011) From risk assessment to in-context trajectory evalua-
tion — GMOs and their social implications. Environmental Sciences Europe 23(3): 1-13.

Petit S, Vinther FP, Verkerk PJ, Firbank LG, Halberg N, Dalgaard T, Kjeldsen C, Kohleb N,
Lindner M, Zudin S (2008) Indicators for environmental impacts of land use changes. In
Helming K, Pérez-Soba M, Tabbush P (Eds) Sustainability Impact Assessment of Land
Use Changes. Springer Verlag (Berlin): 305-324. doi: 10.1007/978-3-540-78648-1_16

Relyea R (2005) The impact of insecticides and herbicides on the biodiversity and productiv-
ity of aquatic communities. Ecological Applications 15: 618-627. doi: 10.1890/03-5342

Reuter H, Middelhoff U, Graef F, Verhoeven R, Batz T, Weiss M, Schmidt G, Schréder W,
Breckling B (2010) Information system for monitoring environmental impacts of geneti-
cally modified organisms. Environmental Science and Pollution Research 17: 1479-1490.
doi: 10.1007/s11356-010-0334-y

Romeis J, Hellmich RL, Candolfi MP, Carstens K, De Schrijver A, Gatehouse AM, Herman RA,
Huesing JE, McLean MA, Raybould A, Shelton AM, Waggoner A (2011) Recommendations
for the design of laboratory studies on non-target arthropods for risk assessment of genetically
engineered plants. Transgenic Research 20: 1-22. doi: 10.1007/s11248-010-9446-x

Rémbke J, Jansch S, Meier M, Hilbeck A, Teichmann H, Tappeser B (2009) General Recom-

mendations for Soil Ecotoxicological Tests Suitable for The Environmental Risk Assess-

96 Frieder Graef et al. / BioRisk 7: 73-97 (2012)

ment of Genetically Modified Plants. Integrated Environmental Assessment and Manage-
ment 6(2): 287-300.

Rosi-Marshall EJ, Tank JL, Royer TV, Whiles MR, Evans-White M, Chambers C, Griffiths
NA, Pokelsek J, Stephen ML (2007) Toxins in transgenic crop by products may affect
headwater stream ecosystems. PNAS 104: 16204-16208. doi: 10.1073/pnas.0707177104

Rounsevell MDA, Metzger MJ (2010) Developing qualitative scenario storylines for environ-
mental change assessment. Wiley Interdisciplinary Reviews: Climate Change 1: 606-619.
doi: 10.1002/wec.63

Saxena D, Flores S, Stotzky G (1999) Insecticidal toxin in root exudates from Bt corn. Nature
402: 480.

Schermer M, Hoppichler J (2004) GMO and sustainable development in less favoured re-
gions—the need for alternative paths of development. Journal of Cleaner Production
12(5): 479-489. doi: 10.1016/S0959-6526(03)00110-0

Schmeller DS, Henle K (2008) Cultivation of genetically modified organisms: resource needs
for monitoring adverse effects on biodiversity. Biodivers. Conserv. 17: 3551-3558. doi:
10.1007/s10531-008-9404-6

Schmitt-Jansen M, Veit U, Dudel G, Altenburger R (2008) An ecological perspective in aquat-
ic ecotoxicology: Approaches and challenges. Basic and Applied Ecology 9: 337-345. doi:
10.1016/j.baae.2007.08.008

Séralini G-E, Mesnage R, Clair E, Gress S, Spiroux J, Cellier D (2011) Genetically modified
crops safety assessments: present limits and possible improvements. Environmental Sci-
ences Europe 2011, 23:10. http://www.enveurope.com/content/23/1/10

Settele J, Kiihn E (2009) Insect Conservation. Science 325: 41-42. doi: 10.1126/sci-
ence. 1176892

Smale M, Zambrano P, Gruére G, Falck-Zepeda J, Matuschke I, Horna D, Nagarajan L, Yer-
ramareddy I, Jones H (2009) Measuring the Economic Impacts of Transgenic Crops in
Developing Agriculture during the First Decade. IFPRI Food Policy Review 10. http://
www. ifpri.org/publication/measuring-economic-impacts-transgenic-crops-developing-
agriculture-during-first-decade

Stein A, Ettema C (2003) An overview of spatial sampling procedures and experimental design of
spatial studies for ecosystem comparisons. Agriculture Ecosystems and Environment 94: 31-47.
doi: 10.1016/S0167-8809(02)00013-0

Székacs A, Lauber E, Takdcs E, Darvas B (2010) Detection of CrylAb toxin in the leaves of
MON 810 transgenic maize. Analytical and Bioanalytical Chemistry 396(6): 2203-2211.
doi: 10.1007/s00216-009-3384-6

Székacs A, Weiss G, Quist D, Takacs E, Darvas B, Meier M, Swain T, Hilbeck A (2011) Inter-
laboratory comparison of Cry1Ab toxin quantification in MON 810 maize by enzyme-im-
munoassay. Food and Agricultural Immunology 22. doi 10.1080/09540105.2011.604773

The Royal Society (2009) Reaping the benefits — Science and the sustainable intensification of
global agriculture. RS Policy document 11/09. http://royalsociety.org/Reapingthebenefits

Turrini A, Sbrana C, Pitto L, Ruffini Castiglione M, Giorgetti L, Briganti R, Bracci T, Evange-
lista M, Nuti MP, Giovannetti M (2004) The antifungal Dm-AMP1 protein from Dahlia

merckii expressed in Solanum melongena is released in root exudates and differentially

A framework for a European network for a systematic environmental impact assessment... 97

affects pathogenic fungi and mycorrhizal symbiosis. New Phytologist 163: 393-403. doi:
10.1111/j.1469-8137.2004.01107.x

VDI (2010) VDI-Fachbereich Gentechnik: http://www.vdi.de/42479.0.html

Wynne B (2001) Creating public alienation: Expert cultures of risk and ethics on GMOs. Sci-
ence as Culture 10(4): 445-482. doi: 10.1080/09505430120093586

Yeates GW (2003) Nematodes as soil indicators: functional and biodiversity aspects. Biol. Fer-
til. Soils 37: 199-210.