Comments on Monitoring of LMOs released into the environment
Version of 22 February 2012 (3rd AHTEG, CBD)
(All the remarks below are directed to post release monitoring, under the light of the Brazilian experience with GMOs and comparatively to the Brazilian normative for GMO post release monitoring - Paulo Paes de Andrade, D. Sc., Federal University of Pernambuco – firstname.lastname@example.org)
Please visit http://genpeace.blogspot.com/2011/12/brazils-new-post-release-monitoring.html for explanations and comments on the Brazilian post release monitoring system.
Please visit http://genpeace.blogspot.com/2011/12/brazils-new-post-release-monitoring.html for explanations and comments on the Brazilian post release monitoring system.
This document complements and builds on the Roadmap for Risk Assessment of Living Modified Organisms.
From the very beginning: monitoring is risk management, not risk assessment. Therefore, the document should not be presented as an addendum to the Roadmap, or a form of building up on it, except for controlled release experiments, before commercial release.
In the context of this guidance, monitoring of LMOs refers to the systematic observation, collection, and analysis of data undertaken based on the risk assessment and following the release of an LMO into the environment, and in accordance with the objective of the Protocol.
As far as the Brazilian experience goes, all risk assessments of commercially approved plants pointed to negligible risks. In such a scenario, the monitoring plan has no support from the risk evaluation results, except this: there are no risks...If the idea is to monitor the very small risks that were considered negligible during the risk evaluation steps, it is just a loss of time and money. Usually, the risk assessments ended up with a large degree of certainty about the risk levels. Therefore, the next paragraph is just void. Moreover, the idea that monitoring could bring new data for risk assessment is kind of a dream, as monitoring conditions are far from the ideal, controlled conditions needed for solid scientific conclusions . It could, nevertheless, disclose some unexpected harm. Starting from this disclosure, the risk analyst would have to return to controlled field releases or lab experiments.
In the context of paragraph 8(f) of Annex III, which states that “where there is uncertainty regarding the level of risk, it may be addressed by requesting further information on the specific issues of concern or by implementing appropriate risk management strategies and/or monitoring the living modified organism in the receiving environment”. As such, monitoring is one of the possibilities to reduce uncertainty related to the level of risk of an LMO. In accordance with the terms of reference for the AHTEG, this document provides guidance on “monitoring of the long-term effects of living modified organisms released in the environment”. In addition, recognizing the importance of in situ conservation, Parties to the Protocol may consider monitoring within the broader context of the provisions of article 7, “Identification and Monitoring”, of the Convention of Biological Diversity (CBD) (e.g. monitoring of protected areas or keystone species). Article 16 of the Protocol and, in particular, paragraphs 2 and 4 may be relevant with respect to the implementation of monitoring.
Monitoring may help detect changes related to adverse effects, in a timely manner, before the consequences are realized, and inform the need for appropriate response measures (e.g. changes to risk management strategies, emergency response measures, a new risk assessment, or re-evaluation of prior decisions).
The Brazilian National Technical Biosafety Committee, in its new normative on monitoring, crystallized the idea that damage (adverse effects) is the sole variable that can be observed in the real world of agribusiness. Moreover, adverse effects, damage and consequences are just synonyms, in the context of risk analysis. What can be extracted from the above paragraph is: if we follow some baselines, we could maybe see changes that would signal for a late damage. This is, however, not practical in Brazil, where 30 mi. hectares of GM crops were harvested last year. Baselines are both scientifically very hard to establish in non-controlled, commercial fields and very expensive. Monitoring could, therefore, just detect some damage, not necessarily due to the GMOs around the monitored areas. Inevitably, any suspicion would have to be followed by a long, scientifically driven process of investigation.
OBJECTIVE AND SCOPE
The present document aims at providing conceptual, science-based and practical guidance for monitoring changes that could be related to adverse effects of LMOs released into the environment and that could affect the conservation and sustainable use of biological diversity, taking into account risks to human health. This guidance may be applicable to all classes of LMOs, and scales of release into the environment (e.g. small- and large-scale releases).
Monitoring of potential adverse effects to human health in the context of environmental risk assessment is considered under this guidance (e.g. inhalation of pollen from LM plants).
Again, monitoring is risk management, not a part of the risk assessment. A far as these critics are concerned, we deal here only with post release monitoring
Issues related to the decision as to whether or not monitoring should be implemented, or who bears the responsibility for its implementation and associated costs, are not addressed in this document.
MONITORING AND ITS PURPOSES
Monitoring can be done in a case-specific manner to address questions and uncertainties related to level of risk identified in a risk assessment. When recommended in step 5 of the Roadmap, the case-specific monitoring reflects the considerations in the earlier steps of the risk assessment and the considerations on uncertainty with regard to the overall risk of the LMO.
Please, see comments above on usual risk levels and uncertainty levels in real life GMO risk assessments
The implementation of case-specific monitoring in conjunction with an approved release may provide observational data about specific effects of the LMO on relevant components of the ecosystem.
Although this may be true for pre-market, controlled field releases, its efficacy in producing data from commercial field observations is far from granted...
Case-specific monitoring of the environmental release may be done for different purposes, depending on the type (e.g. experimental or commercial), duration (e.g. short- or long-term) and scale (e.g. small- and large-scale) of release, as well as on uncertainties regarding the level of risk or its management:
• Monitoring during experimental, short-term and/or small-scale environmental releases
Monitoring can generate data during experimental, short-term and small-scale releases in order to provide supporting data for future risks assessments that may involve a larger scale of release of the same LMO. When environmental releases of an LMO are conducted in a step-wise manner, monitoring at smaller scales may increase the scientific strength or certainty of risk assessments for subsequent larger scale releases.
• Monitoring during long-term and/or large-scale environmental releases
During long-term and large-scale releases of an LMO (e.g. for commercial purposes), monitoring may be conducted in order to address remaining uncertainties identified in the risk assessment, or to confirm that conclusions of the risk assessment are accurate once the environmental release has taken place.
Again, the uncertainties remaining in real life risk assessments are usually very near to zero. The huge costs and inevitable flaws in any monitoring plan are not justified from the strict pre-market risk assessment results. Nevertheless, a general surveillance directed towards the identification of damages to certain protection goals may be justified, as far as the fulfillment of a political wish is concerned...
• Monitoring to evaluate the efficacy of specific risk management strategies
In cases where risk management strategies are implemented along with an environmental release, monitoring may be used to evaluate the effectiveness of these risk management strategies.
This is absolutely correct, but only happens when non negligible risks are detected during risk assessment, what is seldom the case.
Another type of a case-specific monitoring may be undertaken to detect changes related to potential adverse effects that were identified but not addressed in the risk assessment (e.g. effects such as long-term, tri-trophic, cumulative, as well as changes to management practices and effects to human health).
The idea is nice, but since post release monitoring is dependent on commercial areas, where a number of different parameters are fully out of the control of the monitoring personnel, it is highly improbable that sound scientific results would come out of these field monitoring. Even if a clear question could be addressed, the many different field practices in the area, including crop rotation, pesticide use, pollution, bird migration, etc, etc would possible preclude any useful results.
Broader environmental monitoring for unanticipated adverse effects that were not identified in the risk assessment may be conducted to address more general questions related to the conservation and sustainable use of biological diversity, taking into account risks to human health. Monitoring for unanticipated effects starts with general observations of changes in indicators and parameters, which are often defined within national protection goals, and that could be related to adverse effects. Should monitoring for unanticipated adverse effects that were not identified in the risk assessment detect changes that could be related to an adverse effect, a more specific hypothesis may be formulated to establish a causal relationship between the LMO(s) and the adverse effect, and be followed up by case-specific monitoring or further research. When monitoring for unanticipated adverse effects that were not identified in the risk assessment, programmes already established for the surveillance of broader protection goals may be used in order for the monitoring to be more cost-effective.
Why should a broader environmental monitoring be conducted to address questions related to conservation and sustainable use of biological diversity? This is by no means the scope of GMO monitoring.
The European conception of general surveillance based on baselines for a couple of protection goals is, as stated before, too expensive and generally impossible to be conducted in commercial fields. On the other hand, the transition from general surveillance to case-specific monitoring is correct: the identification of a possible lin between an adverse effect detected in the monitored area and a GMO should be further evaluated by a case-specific monitoring approach, However, even before this specific field observation begins, new lab experiments should signal the pathways to damage.
Annex 1 provides a diagram outlining the purposes of monitoring in the risk assessment process under the Protocol.
DEVELOPMENT OF A MONITORING PLAN
A monitoring plan is developed when the recommendation of a risk assessment and/or the national biosafety policy calls for monitoring activities to be carried out in conjunction with the environmental release of the LMO. In such cases, the competent authority(ies) or the entity responsible for the risk assessment may outline the requirements of the monitoring strategy (including the reporting of monitoring data). The monitoring plan should be transparent, of scientific quality and presented in sufficient detail so that the relevance of the data can be appraised.
If the monitoring plan is to be developed by the notifier, it may be evaluated by the competent national authority and may be subject to modification before a decision for release is granted. It is important to consider that the proposed monitoring activities should be commensurate with the uncertainty regarding the level of risk posed by the LMO under consideration.
Absolutely! However, it is important to keep in mind that in the real world of GMO risk evaluation risks are until now considered negligible and the uncertainty of this decision approaches zero!
Information relevant for developing the monitoring plan may be available from the risk assessment and, if applicable, from previous monitoring activities, including those from other countries. For example, the choice of protection goals, as well as of indicators and parameters may often be derived from the context and scoping phase of the risk assessment (See Roadmap, “Setting the context and scope”). The scientific and technical details of the specific LMO, including detection methods, would be available from the information required for conducting the risk assessment as outlined in Annex III.
This guidance focuses on the development of a monitoring plan to address uncertainty regarding the level of risk of an LMO in the context of (i) the results and recommendations of the risk assessment, including adverse effects that were identified but not addressed in the risk assessment and (ii) unanticipated adverse effects that were not identified in the risk assessment. When both types are to be undertaken, separate plans may be developed. When developing (or evaluating) a monitoring plan, the following may be considered:
Again, the level of uncertainty is usually very low. Moreover, if adverse effects were identified during risk evaluation, but not considered, than something weird happened and RA should be restarted again. By no means monitoring should substitute such a flaw.
Monitoring could be useful to detect unanticipated adverse effects (using the broader approach of general surveillance), but a realistic way of doing it is not to follow baseline fluctuations, but to concentrate on damage reports.
1. Description of how monitoring data would address the uncertainty regarding the level of risk of an LMO (“why to monitor?”);
Again, as stated many times before: the uncertainty on the level of risk is usually nil in real life RA. Unanticipated risk, however, may exist. Therefore, the main reason to monitor is to try to detect those unanticipated damages.
2. Choice of indicators and parameters for monitoring (“what to monitor?”);
3. Monitoring methods, including the establishment of baselines and the duration of monitoring (“how to monitor?”);
4. Monitoring sites and regions (“where to monitor?”);
5. Reporting of monitoring results (“how to communicate?”).
The sections below address these issues in terms of rationales and points to consider.
1. Description of how monitoring data would address the uncertainty regarding the level of risk of an LMO (“why to monitor?”)
The monitoring plan may differ according to the uncertainties regarding the level of risk of an LMO, including (i) risks that were identified but either not addressed or resolved in the risk assessment, as well as monitoring of the efficacy of risk management measures, and (ii) risks that were not identified in the risk assessment and, therefore, related to unanticipated adverse effects. The monitoring plan should be described in such a way that it will contribute to achieving its expected outcomes.
Clearly, although the text is confusing, two strategies are delineated: a case-specific monitoring for those risks which are not negligible and a general surveillance to depict any unanticipated negative effect. However, as stated before, non negligible risks have not been reported in all 34 GMOs seeded and harvested in Brazil. Moreover, the uncertainty level of risk measurement is effectively zero. Therefore, case specific monitoring is useless.
Points to consider:
a. Uncertainties regarding the level of risk of the LMO; effectively, zero.
b. Identified causal pathways from the LMO to potential adverse effects, if applicable, in relation to the risk hypothesis; What is the real meaning of this elliptic phrasing? Causal links are part of the RA, they may guide the design of the monitoring plan. Is that the idea under item b?
c. Uncertainties related to the duration and scale of the release; Crop rotation, new technologies, market variations, etc., all these questions transform any attempt to follow baselines or transform commercial plantations in lab experiments in a real nightmare. Only theoreticians can imagine that a certain GMO will stay in the same area for many years, without any disturbance from the neighbourhood...
d. Uncertainties related to the effectiveness of the implementation of risk management measures. Only if previously known, non-negligible risks exist...
2. Choice of indicators and parameters for monitoring (“what to monitor?”)
The selection of indicators and parameters to be monitored will vary from case to case, depending on the LMO, characteristics of the receiving environment, specific risk scenarios established during the risk assessment (see the Roadmap), and on the protection goals and biosafety legislation or policies of each country.
Considering again the real life scenario, previously evaluated risks will be negligible. Therefore, the risk scenario will be again that of the unanticipated damages. Protection goals may be followed to identify any damage.
The indicators (e.g. species, populations, groups of species, environmental processes, etc.) and parameters (i.e. a component to be measured in the observation of an indicator) chosen are ideally those that can reliably signal potential adverse effects and address uncertainties in the level of risks.
Here baselines are at the heart of the question. We just do not believe on any possible correlation between baseline fluctuations and GMOs, due to the rather uncontrolled commercial field situation.
Annex 2 provides examples of indicators and parameters that may be part of a monitoring plan.
Points to consider:
- The potential of the indicators and parameters to signal potential adverse effects, in particular, before the consequences are realized;
Again, the phrasing is confusing. Adverse effects, damage and consequences are just synonyms. We can, however, admit that some discrete value fluctuation of certain parameters could signal for a small damage, before some serious damage could happen. However, such a reliable parameter in commercial field conditions is simply impossible to design.
- Characteristics of the indicators, as well as the distribution and abundance of those indicators that are species and, if applicable, their level of exposure to the LMO;
- Variability of the parameters to be measured;
- The usefulness of the chosen indicators and parameters to establish relevant baselines, including reference points;
- The importance of the indicators and parameters to relevant key ecological processes and functions or to the identified protection goals;
- Whether sampling and analysis would be easy or difficult and how these would affect the choice of indicators and parameter.
All items above are useless if reliable baselines are not available. Costs will be immense to establish many baselines in different crop areas and even larger costs will hinder the baseline fluctuation follow up. The predictive power of these baselines, due to the unavoidable time and place variation of commercial fields, will be very low.
3. Monitoring methods, baselines and duration of monitoring (“how to monitor?”)
a) Selecting monitoring methods
Monitoring methods are largely dependent on the indicators and parameters chosen in the preceding step and their ability to address uncertainty regarding the level of risk and to signal adverse effects. The selection of monitoring methods should also take into account their level of sensitivity and specificity needed to detect changes in the indicators and parameters.
As stated above, monitoring sounds like a very controlled scientific experiment, running in controlled conditions. This is definitely not the case. It does not matter which indicator, it will be under the influence of a myriad of factors out of the observer´s control and, specially, without a similar indicator in non GM areas. This is the case of many large producers, as Brazil, where adoption rates for GM technology are very high, different GM varieties are seeded every year and crop rotation is the rule. It is just a dream to imagine it would be possible to establish solid baselines for convenient protection goals.
The description of the monitoring methodology includes the means for sampling and observing indicators and parameters, and analyzing the resulting data. Appropriate methods, observations, descriptive studies, or questionnaires may be useful in the collection of data for monitoring, including questionnaires addressed to those who are exposed to the LMO. For ecological issues, or effects occurring outside of the receiving environment, additional knowledge and tools may be required to gather relevant data.
Harmonization of methods, data formats, and analytical approaches facilitates the comparison of results from monitoring. When the use of existing monitoring networks is to be considered, the monitoring plan should specify the criteria for their selection and utilisation.
In Brazil, post release monitoring is based on damage detection, not on baseline fluctuations. A network of informants is an important element, but far more important are questionnaires that are developed and applied by specialists.
Points to consider:
a. Relevance of the monitoring methodology to generate information to address uncertainty related to the level of risk;
b. The nature of the effect to be monitored (e.g. whether short- or long-term, delayed or indirect, cumulative, etc.);
Most of these effects will not be detected by base line fluctuations, especially those long-term, delayed or cumulative. 5 year baselines are just impossible in such a rapidly changing environment as the Brazilian agriculture.
- Relevance, suitability and adaptability of existing broader monitoring schemes, as well as the accessibility to those data, in the context of broader environmental monitoring for unanticipated adverse effects that were not identified in the risk assessment;
d. The specification of the ranges or degrees of changes in a parameter or indicator to signal an adverse effect;
e. The scientific quality of the sampling, analytical and statistical methods to be employed;
f. The availability of relevant standardized methods, and whether and how these could be taken into account;
g. Whether methods are adequate to meet the objectives of the proposed monitoring plan;
h. The use of descriptive studies or questionnaires, taking into account their replicability and verifiability;
i. Findings of the ongoing and/or other monitoring activities, if relevant;
j. Relevant local, regional and international monitoring practices.
All these questions are relevant and point towards the VERY DIFFICULT task of establishing such a monitoring system, based on baseline values.
b) Establishing baselines, including reference points
The establishment of relevant baselines, including reference points is necessary for observing and analysing changes during monitoring. In practice, the baseline is a measurement of the relevant indicators and parameters in the likely potential receiving environment, or in a comparable environment. Therefore, the baseline should be described in the monitoring methodology in order to verify that it accurately represents the environment where the LMO will be released. Natural and human induced variation that may occur in baseline data should be taken into account when analysing monitoring data.
Absolutely! The central problem here is exactly this variation, which is not related to the GMO (or, more specifically, to any detrimental effect of the GMO when compared to the non GM counterpart). The effects of human activities, natural, seasonal variations and other sources of baseline fluctuations are far stronger than the unanticipated effect a GMO may have...
Points of consider:
- The scientific quality of methods used for generating baseline data;
- The appropriate spatial scale over which to establish the baseline;
- Effects of temporal and spatial variation (i.e. human induced or natural variation);
- The scale of potential spread of the LMO.
c) Establishing the duration of monitoring
The duration of the monitoring, including the frequency of observations necessary, is chosen on a case-by-case basis and will depend on the type of adverse effects that are to be monitored (e.g. immediate or delayed, short- or long-term), type of LMO (e.g. short or long life cycles, transgenic traits introduced), or duration of proposed environmental release. The duration of monitoring may be changed, if appropriate, on the basis of the results of on-going monitoring activities.
Points to consider:
a. The duration necessary for changes in a parameter related to the adverse effects to likely become apparent;
b. Life-cycle and generation time of species to be used as indicators;
c. Life-cycle and generation time of the LMO as being used in the environment;
d. Whether variability in the monitored parameters over time could affect the results of the monitoring;
e. Potential for environmental changes.
Nice points, but again just theoretical considerations. In real life there will be a restricted time window determined by crop rotation, adoption of new technologies, periodical agricultural pests, market issues and many other circumstances that will destroy any attempt to pursue a monitoring goal for a long period.
4. Choice of monitoring sites (“where to monitor?”)
4. Choice of monitoring sites (“where to monitor?”)
Monitoring sites are selected on a case-by-case basis depending on the parameters and indicators that will be used in the monitoring and the likely potential receiving environment, as well as the intended use of the LMO, and taking into account the associated management practices. The likely potential receiving environment may include areas that extend beyond the intended receiving environment where the LMO may be introduced.
Relevant information regarding the sites to be monitored include, for example, specific locations, their size and relevant environmental characteristics.
Points to consider:
a. Dissemination and establishment of the LMO in the likely potential receiving environment;
b. The type of LMO as well as indicators and parameters to be monitored and, in case of indicators species, their biological or ecological characteristics and life cycles;
c. Appraisal of suitable, relevant reference sites where the LMO is not present for a comparison over the duration of the monitoring, if applicable;
d. Pathways through which the environment is likely to be exposed to the LMO(s);
e. The distribution patterns, including seasonal distribution (e.g. migration), of the selected indicator species in the receiving environment for consistent detection and observation;
f. Appraisal of protected areas and centres of origin and genetic diversity or ecologically sensitive regions, particularly in the context of monitoring the presence of LMOs;
g. The appropriate number of monitoring sites sufficient to support meaningful statistical analysis;
h. The continued availability of the monitoring sites throughout the duration of monitoring;
i. Current management practices and possible changes to those practices over the duration of monitoring.
Again, the points are relevant, but just emphasize what we said before: in real life post release monitoring should be done in commercial fields, under normal management, in a sufficiently large area and under these circumstances baseline fluctuations will be very hard to be established, even harder to be followed up, making the identification of a causal link between the GMO (its gene products, gene flow, etc) and the damage a very difficult exercise.
5. Reporting of monitoring results (“how to communicate?”)
Reporting of monitoring results serves four main objectives: i) to inform competent authorities of any changes that could be related to adverse effects, ii) to provide feedback as to whether the monitoring activities have been carried out in a manner that meets the intended objectives set out in the monitoring plan, iii) to indicate, if appropriate, the need for changes to the monitoring strategy and/or other risk management strategies (or for follow-up studies or risk assessments), and iv) to recommend, if appropriate, the re-evaluation of a decision and the necessity of any emergency measures.
The reporting of monitoring activities may be communicated in different forms depending on the target audience. Since monitoring is both a scientific and regulatory activity, the report should clearly describe how the scientific results relate to the original regulatory need for monitoring. From the report, the regulatory authority should be able to interpret the results and decide whether or not a specific action is required.
Points to consider:
a. Reporting requirements set out by the competent authority(ies) or in national biosafety regulations, if available;
b. The completeness of the report, including transparency in presentation of methods, data and analytical tools used to draw conclusions;
c. Accessibility to raw data accrued during the monitoring activities, taking into account information that may be confidential.
CHALLENGES IN THE IMPLEMENTATION OF A MONITORING STRATEGY
In the development (or evaluation) of a monitoring plan, it may become apparent that resource limitations or technical and scientific challenges may affect its effective implementation. Therefore, an analysis of the capacities and resources, human and financial, helps to ensure the maintenance and completion of the proposed monitoring strategy. Amendments to the strategy may be required in some cases to ensure the monitoring strategy is efficient and cost-effective in relation to monitoring needs and expected outcomes.
Because changes or effects observed through monitoring may be a consequence of complex interactions of various biological and non-biological factors within the environment, it is essential that the monitoring activities are designed in a way to give meaningful information towards determining whether the observed effects and an LMO have a causal link (which may require further monitoring information or data).
Examples of challenges that may be encountered during the implementation of monitoring may include i) lack of capacity for the establishment of robust detection or identification methodologies, ii) determination of cause-effect relationships (causalities) between the LMO(s) and observed changes in the indicator(s) or parameter(s); and iii) the interpretation of monitoring results and relating them to further specific actions.
Yes, we do agree: three very limiting challenges!!!
[Add graphic representation of the revised text]
The graphic was missing in the original CBD document
Examples of monitoring in relation to protection goals/objectives
Example(s) of monitoring
Reduction of levels of significant uncertainty of potential effects identified in the RA
We did not see any significant level of uncertainty in risk levels identified in our RA until now.
Target organisms, Non-target organisms, environmental parameters, etc.
• Confirming host-range effects of target transgenic proteins, resistance development,
• Confirming exposure routes or levels, if not maximized in the considerations of the risk assessment (worst case approach)
Impact on assessment endpoints or related indicators identified and evaluated in the RA
Target organisms, non-target organisms, environmental parameters, etc.
• Presence and population levels of key selected NTOs
• Food web and predator/prey interactions of key selected NTOs at different trophic levels
These levels are subject to intense variations, both seasonal and geographical, due to many different reasons, We strongly doubt that any baseline values could be established and, even more difficult, that any meaningful correlation fo baseline values changes and GMO could be established.
Confirmation of in vivo exposure levels
Non-target organisms, etc.
• Direct or indirect uptake/exposure of NTOs to transgenic pesticidal proteins
• Existence of weed species in herbicide tolerant (HT) fields
• Accumulation of transgenic products in the soil
This is a scientific experiment, not a monitoring activity. Moreover, some questions extrapolate the monitoring scope (as far as biological diversity is concerned)
Impact on production systems in relation to sustainability
Functional organisms, key environmental services, etc.
• Pollination impacts
• Pest control efficacy
Monitoring for scale-dependent effects
Wild and weedy relatives, HGT candidates
• Persistence of DNA or transgenic products in the soil
• Frequency of gene transfer potential
Again, this is a scientific research, not a monitoring activity. Some of these questions have been already approached and the papers are published to help the risk manager.
Efficacy of risk management strategies
Weed populations, resistance development
• Efficacy of refugia strategies to delay resistance development of pesticide-producing crops by testing susceptibility of target pests
• Recording weed populations in HT crop fields or adjacent areas
This is technology follow-up, something that companies do very well. It is definitely not environmental monitoring.
Conservation of biodiversity (including genetic diversity) and ecosystems
The examples listed in the last column are curious: why should anyone monitor dispersal, establishment and persistence of a GM maize or a GM soybean compared to its non GM counterpart? GM plants are by no means more invasive than the non modified recipient organism. In the specific case of maize, it is just nonsense to ask this question. It is a useless question for most normal commercial crops, even for beans. Many other questions grossly exaggerate the “power” of GMOs (specifically emanated from their GM phenotype) to change our world (e.g., landscape alterations).
Primary producers (e.g. plants) and vertebrates (mammals, birds, fish, etc.), invertebrates (arthropods, fungi) with a focus on beneficial/functional organisms, important sources of genetic diversity or protected species
• Abundance and population changes
• Resistance development, changes in pest prevalence or pathology
• Effects of agrochemical usage associated with the LMO in indicator species
• Developmental and fitness changes (direct and indirect) in indicator species
• Host range or key behavioral changes in indicator species
• Changes in dispersal, establishment and persistence in the LMO compared to the non-modified recipient organism
• Landscape alterations
• Outcrossing/hybridization with wild or weedy relatives
Soil quality and functional processes
Here technology impact and GMO impact are mixed...
Soil microbes and invertebrates (e.g. bacteria, fungi, and arthropods) particularly those providing key soil ecological services (nutrient cycling and decomposition)
• Population changes
• Gene transfer frequencies
• Organic compound changes
• Effects of agrochemical usage associated with the LMO
• Soil fertility changes
• Changes to degradation processes
• Soil erosion and compaction changes
Water quality and water pollution prevention
Physical and chemical pollutants in water, etc.
• Nutrient levels
• Pollutants: pesticides, herbicides, etc.
• Emission of transgenic product to water
Plant diseases, pests and weeds, etc.
• Incidence of disease, pests and weeds
• Pesticide usage
Human health (e.g. LMO handlers)
Handlers of LMOs or their products (e.g. farmers, research technicians, mill workers, etc.)
• Exposure analysis
• Screens for toxic or immunogenic effects
• Epidemiological surveys
Floral and faunal indicators of functionality (pollinator populations, beneficial plant communities)
• Foraging behaviors and pollination levels
• Soil indicators
Food and Agriculture Organization of the United Nations. (2011). Biosafety resource book. Rome: FAO, Module B: Ecological Aspects and Module D: Test and Post-Release Monitoring of GMOs.
VDI-Guideline 4330 Part 1: Monitoring the ecological effects of genetically modified organisms, Genetically modified plants, Basic principles and strategies, 2006.
EFSA Panel on GMO; Scientific Opinion on guidance on the Post-Market Environmental Monitoring (PMEM) of genetically modified plants. EFSA Journal 2011;9(8):2316. [40 pp.]
Conclusions: the proposal lists important points related to GMO environmental monitoring, but faithfully reflects a point-of-view that is focused on the generation of numerical data for baseline establishment and follow-up. It will be very difficult and very expensive to establish these baselines which, in real life circumstances, will be of little or no help. Moreover, the text is confusing, listing aims not related to monitoring as part of it; conceptual mistakes abound and, specially, wrong assumptions on the real life risk assessment results. This is an inborn mistake that wrongly directs monitoring to such irrelevant targets as invasiveness, for example. The text is of little help to risk managers, especially those with limited experience on the subject.
 See Article 1 of the Protocol.
 COP-MOP decision BS-IV/11 (http://bch.cbd.int/protocol/decisions/decision.shtml?decisionID=11690).
 See CBD article 7(a) to (d).
 See Roadmap “Overarching issues”, “Quality and relevance of information”.
 See Roadmap “Overarching issues”, “ Identification and consideration of uncertainty”.
 See Annex III pagraph 9 (a thru h)
 See also considerations on “Quality and relevance of information” in the Roadmap.
 See article 16(4) of the Protocol.
 See article 21 of the Protocol.
 This table includes a non-exhaustive list of examples that may be taken into account on a case-by-case basis, as appropriate, when developing a monitoring strategy.