1 INTRODUCTION
A range of crops has been genetically modified for tolerance to broad-spectrum herbicides. The proposed benefit of this technology is that it will reduce costs and the amount of herbicide applied, as weed management for the farmer will be easier. However, this same benefit has an important possible side effect, that weed control becomes sufficiently efficient to threaten farmland species that have already been in decline because of changes in farming practice. The purpose of this project is to assess the impacts of genetically-modified herbicide tolerant (GMHT) crops on biodiversity through the effects of the crop management. The research is also intended to contribute to an assessment of the wider question of whether the commercial use of GM crops will change the management of farming systems and the agricultural landscape. The crops under the study funded by DETR / MAFF / SE are maize, winter rape and spring rape. In addition, sugar / fodder beet will enter the study in spring 2000.
The main purpose is to be achieved by testing the null hypothesis that there are no significant differences between the biodiversity associated with the management of GM winter oilseed rape / spring oil seed rape / maize / beet crops that are tolerant to herbicides and comparable non-GM crops at the farm scale.
Here we report on progress from the period October 1999 – February 2000.
Within the project management team, we are working to the following mission:
To test the null hypotheses of no effects of GMHT maize, beet, spring rape and winter rape on biodiversity
To conduct high quality science on the biodiversity of agricultural systems, while testing the null hypotheses
To ensure that deliverables (reports to the Steering Committee, database and scientific papers) are timely and to appropriate quality
To ensure impartiality, quality and transparency of the project
To ensure the health, safety and security of the staff involved in the project
To ensure the effective presentation of the project to participants and the wider community
2 TESTING THE NULL HYPOTHESIS
The overall approach was described in section 2.1 in the November 1999 report. Put simply, indicators of soils, vegetation and invertebrates on land used for GM crops are compared with the same indicators for comparable non-GM crops. A representative range of farms is selected for each crop, from which fields are split in two, and one half allocated the GM crop and the other the non-GM crop on a random basis. Both GM and non-GM farming systems are managed according to current commercial practice, although within this constraint, management practices will be kept as similar as possible. Our methodology is the same for all crops, as far as possible. The experiment does not address gene flow, as this is being addressed under a separate contract.
2.2 Developing the methodology
Field work has continued during winter 99-00 on the three winter oil seed rape sites. Furthermore, there has been analysis of the 1999 field data to inform the decisions about the appropriate intensity of sampling particular biodiversity indicators. Finally, there has been much work on the protocols for site selection, provision of agronomic advice to farmers, and collecting data on crop management.
2.3.1 The issues surrounding site selection
The October meeting of the Scientific Steering Committee (SSC) debated the issue of selecting sites for the full sample, commencing in spring 2000, at some length, and the minutes summarise well the overall process.
It was agreed that the farms selected needed to be representative of farming in the UK. The possibility of using some form of a stratified sample was considered (farms sub-divided by type/intensity of usual management). It was discussed whether such a stratified sample would represent farm types in the proportion they occur in the UK, or be deliberately skewed in order, say, to have an equal number of each type. The committee advised that further consideration be given to this issue by the contractors, in contact with relevant committee members, with a view to confirming the approach as soon as possible.
The contractors asked to be provided with considerably more potential farms than required. They would obtain further details, select farms and a field in each farm to fulfil the selection criteria, then randomise the allocation of treatments to experimental units, in consultation with the Steering Group. Units to be planted with GM crops would then be notified to SCIMAC and growers. The committee supported this request and SCIMAC undertook to find potential sites.
The objectives of farm and field selection are:
- To optimise the power of the experiment to detect any differences between the treatments that are being compared.
- To ensure that farms are a representative sample of those within Britain that grow the crop concerned on a commercial basis
- To ensure that selection is done in a fashion that respects the different responsibilities of the organisations involved and maintains scientific probity
- To ensure that the pair of half-field-sized units within the split-field to which the treatments are allocated are as alike as possible.
2.3.2 The issue of power of the experiment
The Scientific Steering Committee has agreed that a target sample size of 25 sites per crop per year would provide adequate statistical power of the experiment (see November 1999 report, section 2.4). While any shortfall in one year could be made up in others, this situation would be unwelcome on the grounds of project management. Also, while we will be able to make up any gaps in site characteristics in future years to some extent (e.g. by proactive requests for sites with particular characteristics, as suggested by the SSC), we wish to avoid strong year on year differences between the samples of sites. The SSC revisited this issue at its February meeting, at which it was unanimously agreed that SCIMAC would need to provide sufficient farms for the consortium to select a minimum of around 12-15 farms, representing a range of intensities and geographical locations. This would also include some redundancy for operational problems. This would allow the number of farms to be increased slightly in subsequent years, and also allow any minor imbalances in the range of farms selected for year one to be redressed.
2.3.3 The issue of representativenessThe complete set of farms chosen must form a representative set of those on which the crops are grown commercially. It may be desirable but is not essential for some of the relevant factors that describe farm type to be represented in the sample in roughly the same proportion as that in which they occur in Great Britain as a whole. Whether that can be achieved or not, it is important to ensure that the extremes of all relevant factors are adequately represented in the sample. In particular, any ‘blocking’ factor, such as farming intensity, that is thought a priori to have a substantial interaction with the main treatment, should occur in the sample in sufficient numbers so as to be able to estimate the interaction reasonably accurately. For such a factor, it may well be important to depart from the criterion to be represented in the sample in roughly the same proportion as that in which they occur in Great Britain as a whole. This might occur when the proportion in the wider population of some level of a factor, such as low-intensity, high-biodiversity farms, was small; for such a factor some over-representation, say to greater than c.20% of the sample, would be wise.
This overall approach was agreed at the February meeting of the SSC.
The consortium proposed that farm intensity and geographic location were among the most important factors in selecting farms, and that statistically, the best sample of farms would be one in which the range of practices and geography was included and the extremes were somewhat over represented.
2.3.4 The issue of probity
In the farm scale evaluations of GM crops the Consortium has responsibility for the scientific conduct of the study and for the design and analysis. Responsibility for the supply of GM seed, recommendations concerning GM husbandry and the locating of potential sites rests with SCIMAC. The growers and SCIMAC have a contractual relationship; the growers agree to allow the Consortium to collect scientific data from the trial site. SCIMAC has a legal obligation to ensure the proper use of the GMHT seed and crop and the herbicide applied to it. Site selection can only take place on the basis that contracts with the growers, SCIMAC, can be established, and that the farmer is willing to undertake the experiment within the limits set by the Consortium.
The process of site selection is intended to guarantee that the study is impartial, by separating clearly the tasks of each body. First, farmers approach SCIMAC, prompted by the media articles, approaches via farmers groups etc. SCIMAC assess the potentially suitability to hold contracts for SCIMAC, and, if appropriate, pass details on to the consortium. In other words, SCIMAC have to decide on the suitability of farms in contractual terms. They should provide as many farmers as possible to the Consortium, to give maximum flexibility of choice.
We then contact the farmer to obtain key information about crop history, agronomy and measures of biodiversity value, to decide whether the farm would provide a suitable member of the desired sample. In other words, the Consortium is responsible for deciding the suitability of the farm on scientific grounds. This decision is relayed to SCIMAC for arranging a contract, or not, with the farmer.
2.3.5 The characteristics of the target populations
The target populations of sites are those farms that would be likely to grow the GM crop varieties under study in a commercial setting. This is interpreted as including the range of farms that currently grow the crop, excluding organic farms, as GM crops are not allowed within current organic standards.
These target populations have been characterised using existing data, in terms of regional distribution and agronomy. While we are confident that our characterisations are enough for the process of site selection, we wish to learn more about these target populations, notably in terms of how they would use GMHT crops within their farming systems. We have proposed a national postal farm questionnaire to this end.
2.3.6 Deciding whether or not to accept a farm
Once the farm details have been passed on by SCIMAC to the Consortium, the farmer is contacted to provide the details necessary for a choice to be made about its suitability.
Some of the information refers to the whole farm. This includes farm location and the source of agronomic advice. It also includes typical management practice, such as tillage, soil type, average yield and typical inputs on the crop under study, and a self-assessed intensity score. There follows a check list of "biodiversity potential" indicators, such as the presence of beetle banks and conservation headlands on the farm, the use of LEAF audit or other IPM protocols and the use of a FWAG advisor. Other information relates to the field or fields under offer to the Consortium. This includes field size, soil type, aspect, as well as the crop history and recent use of pesticides.
This information is used to aid site selection. It has already been stated that the key elements are geographic location and "intensity". Yield is used as the intensity score for beet, but the replies are not complete / reliable for the other crops. Therefore the self-assessment score is being used, but cross-checked against other answers. We are also keeping close track of field size and of biodiversity potential. For the latter, we are using a weighted score so that practices that will have a larger positive impact on wildlife (or show more commitment to conservation) have higher values.
Broadly speaking, the same information is required for each of the four crops. However, its interpretation may differ between crops, as may the weight given to each particular factor. For example, aspect is particularly important for beet crops. Maize tends to be grown on a greater proportion of mixed farms than the other crops; furthermore, it might be restricted to a certain group of fields on such a farm. The desirable minimum field size also differs: maize fields tend to be smaller than those used for, say, winter oilseed rape.
The data form will be re-assessed before the next round of site selection by comparing questionnaire responses with field observations.
Once the farm has been accepted by the Consortium, and a contract prepared by SCIMAC, the Consortium visits the site with the farmer to decide on the split of the field, and to agree the randomization of treatments. The field must be split so that each of the halves which form the experimental units to which the treatments are applied are as alike as possible over the range of factors that contribute to variability of wildlife within fields. Thus field and field boundary structure between the two sections should be similar as far as possible, as should soil type. A secondary criterion would be to split the field cross-ways, rather than two long, thin units. This process does not rule out unequal treatment areas, as long as the sampled areas are alike.
The randomization of the treatments is then agreed with the farmer. The scientist marks on the field map the letter A on the north / north west / west section, and B on the other. It is explained that one of these has already been randomly allocated the GM crop, and that if the farmer is unwilling to accept this allocation, then he / she cannot take further part in the experiment. This step is required to avoid any possibility of bias by the farmer about which section of the field should receive which crop. The scientist then opens a sealed envelope, prepared by the project statisticians in advance, with the allocation of the GM crop to either section A or B. This allocation is noted on the map, which is signed by both the farmer and the scientist, as a record of the allocation process.
Both GMHT and non-GMHT cropping systems are to be managed along appropriate commercial lines. The GMHT cropping systems will be managed under SCIMAC guidelines. Changes to the rotations, or to field margin management, will therefore be allowed. Where non-herbicide treatments are imposed on both GMHT and the local control, they should be applied at the same time. Any pesticide seed treatments are the same on both crops at a site.
We have already reported (November 1999, Section 2.6) on the importance of data on crop management, and of establishing the impartiality of any agronomic advice given to the farmers. The Steering Committee (October 1999) supported this point:
The committee unanimously agreed the need for openness in the advice given to farmers on pesticide use. Members agreed that SCIMAC should be advising as little as necessary for the GM crop, and should play no part in advising for the non-GM crop. It was agreed that the contractors, DETR and SCIMAC would discuss further how a greater degree of openness could be achieved.
The Consortium considers that the management of the non-GM crop should be as close as possible to the normal commercial management by the farmer. This means that SCIMAC has no involvement with advising on the non-GM crop, even if asked. Also, it means that the farmer uses whatever form of advice is usually used. Any advice is recorded in writing. SCIMAC only has an input on advising the herbicide regimes on the GM crops, and this is in order to ensure that the farmer (or more usually the advisor) understands the guidelines and the product label. All advice from SCIMAC will be aimed at ensuring cost effective weed management. Note that this may involve advising a zero herbicide regime (as may the herbicide regime on the non- GM treatment). This advice will be recorded. SCIMAC will make every effort to channel this advice through the grower’s usual agronomic consultants and to train the consultant in GMHT technology so that the consultant can give independent advice concerning this technology in the future. SCIMAC will continue to liaise with the grower to discharge its legal responsibilities concerning the treatment of the GM crop with glufosinate (or glyphosate in beet), and over site safety and the disposal of produce.
A sample of farms will be visited by an agronomist from the Consortium to ensure that advice is being given to the farmer in an appropriate and professional way. The farmer will inform the Consortium of the actual management practices. At the end of the growing season, the Consortium will ask the farmer to sign off the management diary. The farmer will also be asked to provide a more detailed field management history than given in the initial questionnaire.
Thus, the Consortium has access to records of advice, and actual management records. The former will be used purely to scrutinise for any biases in the nature of advice given to the farmers, whether by SCIMAC or others. The management data are used in two ways. One is to use in the analysis of biodiversity data, seeking relationships between field management and biodiversity indicators. The other is to establish that the management is in a real sense "typical" of the wider population of farmers. The GM herbicide regimes will be compared with those that have been developed in smaller scale experiments. The other regimes can be compared against existing farming practice (see also November 1999, section 2.6).
This can be done at several levels:
- In comparison with the same crops by the same farmers
- In comparison with the same crops by nearby farmers (this would require a new survey not currently within the project)
- In comparison with national data from farm census etc (may not be ideal, as not all data are collected in this way, and there are time lags before they are collated and disseminated).
We are proposing the first and third options, but welcome the Scientific Steering Committee’s consideration of the second option also.
Chapter 2 continued on next page
Published 13 June 2000
Farm-Scale Evaluations Index
Environmental Protection Index
Defra Home Page