Field trials could be underestimating the potential for cross-pollination between GM and conventional crops, according to new research by the University of Exeter. The research team recommends a new method for predicting the potential for cross-pollination, which takes account of wind speed and direction.

The research, published in the journal Ecological Applications, used records of wind speed and direction from weather stations across Europe to predict the movement of pollen in the air. The findings show huge variation in the amount of cross-pollination between GM and non-GM crops of maize, oilseed rape, rice and sugar beet. Levels vary according to whether the GM field is upwind or downwind of the non-GM field given the direction of the prevailing wind over the flowering period of the crop.

Field trials are regularly carried out to measure the potential for cross-pollination between GM and conventional crops. Current guidelines for minimum field-to-field distances are based on the results from these trials. However, if the GM field in a trial is downwind of the non-GM field, the trial will underestimate the potential for cross-pollination.

'We were struck by the strong influence of wind direction on the amount of cross-pollination', said Martin Hoyle of the University of Exeter. 'Wind speed and direction are important factors outside of our control that have not previously been used to inform guidelines on minimum field-to-field distances. Recommended minimum distances between GM and conventional crops may need to be increased based on our findings.'

Field trials are time-consuming and expensive, so measuring the potential for cross-pollination across the full range of weather conditions is not feasible. This research resulted in the development of a theoretical computer model to analyse the effects of wind on pollen travel. The model, together with measurements of cross-pollination and wind speed and direction from field trials, can be used to predict cross-pollination at other times and sites.

'If the production of GM crops becomes widespread in Europe, it is essential that measures are taken to minimise cross-pollination from GM to conventional non-GM crops,' said Hoyle. 'The recommended minimum distances between GM and conventional crops should be informed by weather data, which is possible using our model of pollen dispersal in the wind.'

Funded by The Natural Environment Research Council, a group funding research on environmental issues such as climate change, biodiversity and natural hazards.