An integrative methodology to predict dispersal of genetically modified genotypes in oilseed rape at landscape-level—A study for the region of Schleswig-Holstein, Germany

Potential environmental as well as socio-economic effects of the cultivation of genetically modified (GM) oilseed rape (OSR) may be caused by large-scale dispersal of transgenes. We present an up-scaling approach that was based on scenario assumptions concerning the percentage of GM cultivation and took into account natural and anthropogenic variation of involved dispersal processes. The applied methods include computer modelling and spatial analysis. A simulation model (GeneTraMP) was used to calculate the spatio-temporal pattern of the spread of a neutral transgene (without any specific function) in OSR. Basic scenario calculations were carried out for different spatial configurations covering 1 km2 each and taking into account information on climate and cultivation systems of the region of the federal state of Schleswig-Holstein, Germany. For the exemplary regional study presented here, we analysed the numbers of flowering plants of GM OSR in different types of locations as predicted by the model. The results confirmed the expectation of a very high variability of GM occurrences at distinguishable intensity levels which were closely related to the proximity of areas of intended GM oilseed rape cultivation and may be described by a combination of management parameters and location type.The up-scaling method included a spatial analysis of the target region. Based on satellite images and digital maps, the structure of the region was analysed resulting in a map of Schleswig-Holstein that represents each single field, also including information on crop rotation, ownership and production systems. Applying GIS queries to this database, we identified the area of relevant location types. Both, the model results and the spatial data were used to predict the total numbers of flowering GM OSR plants for the region of Schleswig-Holstein. As an important feature, the up-scaling of modelling results to a larger scale allows for a comprehensive analysis by also enclosing regional parameters, as, for example the cropping density. The presented methods can support decision making if they are incorporated into the planning of an environmental monitoring of commercial GM crops or into life cycle assessment and cost–benefit analyses of GMO cultivation.