Agron. Sustain. Dev. 28 (2008) 195-206
DOI: 10.1051/agro:2007039

Feasibility of isolation perimeters for genetically modified maize

Yann Devos¹, Olivier Thas², Mathias Cougnon¹, Eva M. De Clercq³, Karl Cordemans⁴ and Dirk Reheul<sup>1

1</sup>  Department of Plant Production, Faculty of Bioscience Engineering, University of Ghent, Coupure Links 653, 9000 Ghent, Belgium
²  Department of Applied Mathematics, Biometrics and Process Control, Faculty of Bioscience Engineering, University of Ghent, Coupure Links 653, 9000 Ghent, Belgium
³  Department of Forest and Water Management, Faculty of Bioscience Engineering, University of Ghent, Coupure Links 653, 9000 Ghent, Belgium
⁴  Department of Rural Development, Flemish Land Agency, Gulden Vlieslaan 72, 1060 Brussels, Belgium

Accepted 9 August 2007; published online 20 October 2007

Abstract - Using geographic information system datasets and Monte Carlo simulations, this study investigates to what extent the intra-regional variability in maize share and field distribution affects the feasibility of implementing isolation perimeters between genetically modified (GM) and non-GM maize fields. More specifically, five scenarios differing in shares and spatial distributions of GM maize were tested for various isolation perimeters in six agricultural areas in Flanders (Belgium). Proportions of non-GM maize fields and farmers having at least one non-GM maize field occurring within isolation perimeters were calculated to assess how spatial co-existence measures would affect the freedom of choice of neighbouring farmers to grow non-GM maize on their fields. Irrespective of the scenario tested, our results demonstrated that the proportions of non-GM maize fields and their corresponding farmers falling within the isolation perimeters are approximately two to eight times higher in areas with the highest maize share than in areas with the lowest maize share. The higher the share of GM maize and the wider the imposed isolation perimeter, the lower the intra-regional differences became. Hence, those findings confirm that farmers will not be equally affected by isolation perimeters, indicating the importance of considering intra-regional differences in the choice of appropriate spatial co-existence measures. Since uniform and wide isolation perimeters tend to be difficult to implement in practice and hardly reflect the diversity of the agricultural landscape, relying on flexible or combining various co-existence measures is worthwhile considering. As an alternative to single co-existence measures for limiting the GM input from cross-fertilisations between neighbouring maize fields, the appropriateness of other on-farm co-existence measures is discussed for Flemish agricultural conditions. Proposing the most appropriate co-existence measures on a case-by-case basis may be one step forward in reaching proportionate, fair and consistent co-existence at the regional and landscape level.

Key words: adventitious mixing / co-existence / cross-fertilisation / genetically modified crops / geographic information system / isolation perimeters  / pollen flow / maize / regional variation / simulations

Corresponding author: Yann.Devos@UGent.be

© INRA, EDP Sciences 2008