GM Crop Farm-Scale Evaluations: Fact Sheets
GM Crop Farm-scale Evaluations of GM herbicide tolerant oil-seed rape and maize on farmland wildlife
Introduction
1. Over the last 50 years, steep declines in much of the wildlife that inhabits farmland has been apparent in the UK. Modern intensive farming practices, such as high pesticide use, mechanisation and changing sowing times are believed to have contributed to these declines.
2. The concern has been raised that commercial planting of genetically modified herbicide tolerant (GMHT) crops will lead to more changes in agricultural practice which could put additional pressure on farmland wildlife. The potential wider impact of the commercial use of GM crops has been discussed in the recently published paper The Commercial use of GM crops in the UK: the potential wider impact on farmland wildlife, published by the Advisory Committee of Releases to the Environment (ACRE) and available on the ACRE web-site (www.defra.gov.uk/environment/gm/wildlife/index.htm) or from the Chemicals & Biotechnology Division, DETR, Ashdown House, 123 Victoria Street, London, SW1E 6DE.
3. Government and the plant breeding industry, through the industry body SCIMAC (Supply Chain Initiative on Modified Agricultural Crops), have agreed that the first farm-scale plantings of GM herbicide tolerant crops will be the subject of a scientific investigation of the impact on farmland wildlife. The results of these 'farm-scale evaluations' will ensure that the managed development of GM crops in the UK takes place safely.
4. The farm-scale evaluations will start in spring 1999 with plantings of spring oil seed rape and fodder maize, these will be followed in autumn 1999 by winter oil seed rape. The evaluations will then run for 3-4 years depending on the particular crop.
5. The research is to be conducted by independent contractors, funded by DETR, with MAFF and The Scottish Office contributing. SCIMAC will take responsibility for providing the GM crops, and for liaising with farmers to locate sites for the evaluations. All crops will be grown in accordance with the SCIMAC Guidelines and Code of Practice.
The research contracts
6. For each of the three crops, the research will be undertaken by independent contractors. The research contracts were let by multiple competitive tender. Fifteen major research organisations were invited to tender, and four tenders were received for each of the three contracts.
7. A tender review panel including English Nature, SCIMAC, the Scottish Office, MAFF and DETR decided to let all three of the research contracts to a consortium of three organisations, led by the Institute of Terrestrial Ecology (ITE), and including the Institute of Arable Crops Research (IACR) and the Scottish Crop Research Institute (SCRI). Each institute in the consortium is taking lead responsibility for one of the crop types (ITE for maize, SCRI for spring oil seed rape, IACR for winter oil seed rape).
8. The contracts were awarded to the consortium as, among other things, they proposed the best scientific methodology and demonstrated the best logistical capability to undertake a project of this size. The successful contractors, ITE, IACR and SCRI are eminently qualified to undertake the work, and are world renowned experts in this field of research.
The steering committee
9. The progress of the research will be guided by a steering committee of independent scientific experts, independent from Government, the biotech industry and the contractors. The membership of this steering group will be chosen to reflect expertise in a range of appropriate disciplines, with the intention of ensuring that the research carried out by the contractors is of the highest scientific standards. The steering group will be made up of scientists drawn from English Nature, environmental NGOs (including RSPB) and academia. Representatives of the research contractors, the funding Departments and SCIMAC will also attend meetings; not as advising members of the steering group, but to report on progress, provide information to the members and listen to their advice.
10. The steering group will meet with the contractors three or four times per year and will monitor the progress of the work, advise on experimental methodologies and design, and review data analysis and conclusions. If any changes or alterations are required to the methods of the investigation, these will be conveyed to the contractors. The steering group will also have the important role of overseeing publication of all results (see below; 'publication of results').
11. The steering group will hold its first meeting in early June 1999 to review the experimental design and sampling methodology proposed by the ITE/IACR/SCRI consortium.
Basic project design
12. The farm-scale evaluations will investigate the null hypothesis that there are no differences in the diversity and abundance of wildlife associated with the management of GMHT crops compared with the management of equivalent non-GM crops.
13. A wide consultation was made among the scientific community and environmental groups regarding the design of the project. The basic design adopted will be to have farms on which a GM crop and a non-GM equivalent (the same crop species, of an equivalent variety, but not genetically modified) is grown. The abundance and diversity of wildlife in the two crops will then be compared. Each farm in each year will constitute one such paired comparison.
14. The management of the GM crop will follow the guidelines prepared by SCIMAC and the pesticide manufacturers guidelines regarding herbicide use. Because the exact advice given is likely to be a major influence on the outcome of the investigation, the advice provided will be under the very close scrutiny of the steering committee. The advice will be provided by the company in advance of the investigation, and will be compared with normal practice in countries where similar GM crops are already grown on a commercial scale. The management of the non-GM equivalent crop on the farm will be managed according to normal practice on that farm, this will be substantiated by reference to farm records.
15. The management of the GM crop will be compared with the management of an equivalent conventional crop. What constitutes 'conventional management' will also be carefully considered. The intention is to select farms in a variety of locations that represent the range of management practices used for growing the particular crop. For example in the case of spring oilseed rape this will include high and low intensity pesticide regimes, operating on a range of different soil and drainage types and with different background levels of weeds. It is not possible to include organic oil seed rape as this is not grown in the UK. In the case of maize, there is a small amount grown organically in Britain, but it is not intended to include this in the comparison, in part due to the separation distance required between the GM and organic crops that would prohibit a paired comparison.
16. In the first year the paired comparison between crops is proceeding in two ways: the GM and non-GM crops grown on the same farm but in separate fields (a 'paired-field design') and the GM and non-GM crops grown in a separate half of the same field (a 'split-field design'). Various scientific authorities have advocated each method and both have pros and cons: a paired field design may be more realistic, and may reduce the movement of mobile organisms between the crops that would undermine the independence of the samples, but there may be a greater background variation in the abundance of weeds and insects between the two fields, owing to their past history of management. A split field approach avoids the problem of past history to a far greater extent, but at the expense of some realism and some compromise of independence due to the movement of mobile organisms. The split-field approach may also introduce biases concerning boundaries and the borders between the two halves. In reality the two approaches are best suited to different sets of organisms. In the first year both methods will be used. The contractors and the steering committee will discuss the most appropriate method prior to the start of the second year of study.
Sample sizes/field sizes
17. In the first year of study only three or four farms will have a GM and non-GM equivalent crop sown. This number is being kept low in order to have an opportunity to refine the monitoring methods and to investigate the optimal timing of sampling with respect to different crop management strategies, and the best sample sizes for both within and between field sampling (how many samples should be taken in each field and how many fields should be sampled). It will also help the steering group and contractors decide on the best approach to use regarding split and paired fields.
18. In the second and third years it is likely that around twenty farms will have a GM and non-GM equivalent crop sown. This sample size has been selected as being appropriate following analysis of existing data on between field variation in the abundance and diversity of farmland wildlife derived from previous studies. It includes some contingency if crops fail and cannot be used. If the results of the first year indicate that the sample size is insufficient it will be reconsidered.
19. The size of fields used has been the subject of some debate. The guiding principle here is that the size of fields should reflect commercial reality. In the case of oil seed rape, large fields (around 10 ha or more) are being sought for those fields being sown with a single crop type (the paired-field design) and even larger ones for those fields being sown with half GM and half non-GM (the split-field design). In the case of maize, which is typically grown in smaller fields, the size of fields in the evaluations may be smaller than 10 ha.
20. Farms may take part in the evaluation in successive years. As crops are typically rotated around the fields of a farm as part of a rotation, it is unlikely that any field will be planted in consecutive years with a GM crop. Oil seed rape is typically rotated in a 3-5 year cycle with cereals and other crops (the possible exception being maize in which successive cropping in the same field occasionally takes place). The fields in which the GM crops are grown will be monitored in following crops regardless of whether another GM crop is sown in the field or not (see below).
Research programme
21. A comprehensive research program has been prepared by the contractors, concentrating on sampling plants and invertebrates. All sampling will take place in the crop, the headland and in the field margin. Wildlife that will be sampled includes, all vascular plants, arthropods (insects, spiders, harvestmen etc.) on the ground and on plants, caterpillars, slugs and snails, bees, butterflies, soil seed bank and earthworms.
22. The monitoring methods proposed will follow standard tried-and-tested ecological sampling practices. These will be constantly monitored by the steering group. If the range of organisms sampled, or the measures of their diversity and abundance used needs to be adjusted, the steering group will advice and the contractors will respond.
23. The list of organisms sampled is not a full list of all the organisms likely to be affected by crop management. Notably, it does not include birds and mammals. However, the existing proposed methodology should provide strong indirect evidence of the potential for an adverse affect. Including birds and mammals is logistically difficult due to the intensive nature of studies to relate aspects of their ecology to single fields as they are rarely confined to such a small area. By sampling organisms at the lower end of the food chain, knock-on effects to the larger organisms can be well predicted.
24. Sampling will take place throughout the growing season of the crops (i.e. at several intervals throughout the year). The timing of these sampling visits will be carefully controlled to coincide with particular growth stages of the crops and herbicide spray timing, in order to give an accurate picture at different times of year.
25. In the first year sampling will concentrate on refining the sampling methodology during the growing season, before the evaluations are scaled up to the full sample size. In subsequent years, vegetation and soil samples will be taken before the crops are sown to establish baseline conditions. More intensive sampling for two years before the GM crop is sown was considered unnecessary as any data collected during this time on plant or invertebrate abundance would largely reflect the crop and management practice at the time, rather than an intrinsic characteristic of the field. The paired sampling design avoids the problem by comparing the GM crop to a control rather than to what was there before.
Publication of results
26. All results of the farm-scale evaluations will be made publicly available. The steering committee will undertake the role of peer reviewers to ensure that the highest possible scientific standards are maintained. This will also prevent unnecessary delay in dissemination of results. A clear demand will exist for results as and when they are discovered. However, a balance has to be made between publishing early results to meet the demand and taking a more prudent approach to avoid misinterpretation of the data. This will be a task for the steering group, and Government will take their advice on the publication of results. It is envisaged that interim and progress reports will be made publicly available throughout the study. However it is clear that many of the final results will not be available until the end of the study, as the statistical analysis and the peer review needs to be carried out thoroughly.
Page published 13 August
1999;
Page last modified
10 August, 2002