The maize trials in the UK's farm scale evaluations (FSEs) have come under fire for being "misleading", "worthless" and "a complete waste of time". Robert Vint and Lim Li Ching investigate.
The FSEs compared the impact of managing GM herbicide-tolerant crops on farmland biodiversity, with that of their conventional counterparts. Three spring-sown crops were examined - beet, oilseed rape and maize. For beet and oilseed rape, clear negative impacts on farmland biodiversity were found (see "GM crops harm wildlife"). GM herbicide-tolerant maize, however, was said to have positive effects, a claim widely highlighted in the media.
But the maize trials have been called into question.
Analysis of the methodology reveals systemic bias - underestimating the environmental impact of the GM crops whilst overestimating the likely environmental impact of future non-GM cultivation. The failure to measure the yield of the GM crop makes it impossible to confirm that the cultivation method was viable. In addition, published yield figures for the GM crop are derived from cultivation using a different herbicide, adding to the deception.
The GM maize used in the FSEs is Chardon LL (Liberty Link) developed by Aventis (now Bayer CropScience), and engineered to be tolerant to its Liberty herbicide (glufosinate ammonium). The GM maize herbicide management regime in the FSEs thus used Liberty, a herbicide less powerful than that used in the non-GM halves of the fields (see later).
However, research and farmers' experience have shown that the GM maize cannot be grown viably unless Liberty is mixed with other more aggressive herbicides. A Texas Agricultural Extension Service report, Weed Control in Liberty Link Corn 1996 to 1999 by Brent Bean and Matt Rowland, concludes that a single Liberty application should not be relied upon for season-long weed control and that control was greatly improved with the addition of atrazine. Similarly, a 1998 paper by Berzsenyi et al. concluded that in Hungary, "the results of field experiments showed that a weed management strategy with glufosinate must include multiple applications, residual herbicides or mechanical control".
Of US farmers growing Liberty Link GM maize, 75%-90% now need to use Liberty ATZ (a more powerful and environmentally harmful tank mix of Liberty and atrazine) rather than plain Liberty to obtain adequate weed control and maintain yields. Aventis/Bayer has marketed Liberty ATZ in the US for use on Liberty Link maize at least since 12 March 2001, as indicated on their product data sheet.
According to Pesticide Action Network UK, maize farmers in the UK have been using increasing amounts of atrazine in recent years. It seems highly likely that if UK farmers grow GM maize, they would want the same mixed formulations as US farmers - if not mixed with atrazine then with other powerful herbicides.
Furthermore, the spread of glufosinate-resistant weeds is a potential problem likely to make the use of Liberty ATZ almost essential in areas where GM maize has been grown for several years*. In the absence of any UK research on Liberty-resistance in weeds, this must be assumed to be a likely problem to emerge in the UK.
If Liberty ATZ or any other Liberty-based herbicide mix was ever licensed for use in the UK, it would have a much more dramatic effect on biodiversity than the FSEs suggest.
The decision to use Liberty alone on the FSEs' GM maize, rather than Liberty ATZ or a mixture of Liberty and another herbicide, ensures that there will be more weeds and wildlife in the GM fields than would be likely under commercial cultivation and makes it unlikely that a commercially viable yield could be obtained. It also means that the GM maize plots were subjected to a herbicide management regime that is likely to quickly become obsolete.
This flaw was highlighted as early as 25 June 2002 in a BBC Newsnight programme 'Weeds fight back', and subsequently in The Times and Farmers Guardian, but no action seems to have been taken by the Scientific Steering Committee (SSC) overseeing the FSEs to correct this or even to discuss the matter. Furthermore, Aventis/Bayer must have known that Liberty on its own was ineffective, as it was already recommending in other countries that its Liberty ATZ be used in conjunction with its GM maize.
Subsequently, Brian Johnson, biotechnology adviser to the Government's advisory body English Nature, commented, "If I were being cynical I would say that Aventis told the government that only GA [glufosinate ammonium] would be used on these crops in the hope that more weeds would survive in the LL [Liberty Link] crops in the FSEs. If so, and I have no idea that this is right, then they could argue that the GM crops were better for the environment! They might then gain marketing consent for LL crops, only for the company then to change the pesticide recommendations to ATZ-type tank mixes."
The non-GM control crops in the FSEs were cultivated commercially by the farmers for sale or for feeding to their own dairy cows. In the overwhelming majority of cases, atrazine - a particularly toxic and persistent herbicide - was used on the conventional maize plots.
However, atrazine is now to be banned by the EU, a decision expected for several years because of its environmental impact. It was already banned in Austria, Denmark, Finland, Germany, Italy, the Netherlands and Sweden. This destroys the validity of the maize trials, as they no longer reflect the real conditions under which non-GM crops will be grown. Atrazine's replacement is likely to be less harmful to the environment.
The use of atrazine on the non-GM crop thereby misleadingly gives the impression that the GM crop is relatively benevolent.
Michael Meacher, who as Environment Minister commissioned the trials, said "The ban on atrazine means that the trials are no longer valid because they no longer make a true comparison between the herbicides that would be used on GM and conventional maize… I do not see how the Government can now responsibly license GM crops."
The suitability or otherwise of the herbicide regime used on the GM crop cannot be assessed because the crop yield was not measured. The FSEs were supposedly designed to mimic expected future UK commercial farming practice with GM crops, but FARM, the Independent Farmers' Union, argues that because no attention was paid to yield the maize trials cannot be shown to reflect normal commercial practice. Furthermore, there is no way of knowing whether commercial farmers would have been satisfied with this level of weed control or with the starch or dry matter yield of the resultant crop.
The measurement of biodiversity, which the FSEs studied, is a complex and time-consuming task. But the measurement of yield - which could be as simple and quick as weighing the crop or the cobs - was not even attempted in these £5.5 million trials. The farmers hosting the trials were merely asked to 'estimate' the success of the crop without providing any evidence!
Independent observers of the FSEs have reported low yields and fields full of weeds in the GM maize plots, raising suspicion that the GM crops were managed to limit adverse effects on wildlife, and not to maximise commercial yields. The results are thus irrelevant to farmers, who would not accept such yield penalties. The absence of yield measurements further increases suspicion that a deliberate attempt was made to conceal the commercial unviability of the herbicide regime selected.
The principal measurements of yield and dry matter reported for Chardon LL are derived from the National Seed List trials, which, in common with non-GM varieties, were grown using atrazine. However, as Chardon LL was engineered for use with Liberty, these figures are irrelevant and almost certainly misleadingly high. Most of the GM maize trials were treated with only one spray of Liberty at rates averaging just 3.5 litres of glufosinate per hectare (FSE report, p. 1815), allowing weeds to flourish, whereas a maximum total dose of 8 litres of glufosinate per hectare was permitted in the efficacy trials to efficiently kill weeds (PSD Notice 1123).
John Sherrell, FARM founding member and South West dairy farmer, said: "These trials are completely useless for working farmers. Not only have they been invalidated by the use of the now banned herbicide atrazine, but they also provide no evidence of how these crops would perform under practical commercial conditions. It is amazing how the Government are trying to force farmers to grow these crops without providing the information farmers need."
GM Free Cymru has accused the SSC, which oversaw the FSEs, the Department for Environment, Food and Rural Affairs (DEFRA), and its scientific advisor, the Advisory Committee on Releases to the Environment (ACRE), of scientific fraud in the GM maize trials. In their view, the SSC should have recommended the cancellation of the maize trials as soon as it discovered that they were not replicating commercial management regimes.
Needless to say, the maize trials did not assess other important questions such as the threat posed to organic and other non-GM maize crops via pollen contamination, or the rate of emergence of Liberty-resistant weeds.
These flaws, in combination, render the FSEs of GM maize misleading and worthless. Ian Panton of GM Free Cymru said, "It would be an act of gross irresponsibility and negligence should the Government seek to authorise the commercialisation of GM maize on the basis of this cynical and dishonest science."
Article first published 29/12/03
*Erratum: In an earlier version of this article, we had stated that US researchers have documented the emergence since 1996 of heritable glufosinate-resistance in ryegrass, goosegrass, horsetail and waterhemp in areas of high glufosinate (Liberty) use. We haven't been able to further verify this. So far, to our knowledge, glufosinate-resistant weeds have not emerged, although the potential remains. But resistance to glyphosate (Roundup), the other major herbicide used with herbicide tolerant GM crops, has been documented since 1996 in rigid ryegrass, Italian ryegrass, goosegrass and marestail (horseweed).
Got something to say about this page? Comment