Predicting the spread of herbicide resistance in Australian canola fields

Abstract

A common concern expressed about the commercial release of transgenic canola into cropping systems is the risks of unwanted gene flow between varieties. Experimental data is emerging that answers some of the theoretical questions that have been posed when considering gene flow on a landscape scale. This study developed models that utilise some of this published data in an attempt to quantify the spread of transgenes in a commercial farming system. The models, which included bootstrapping the empirical data and three mathematical simulations, were compared with each other and the published data. One of the mathematical models estimated average resistance frequency by imposing a Poisson distribution around the published mean value for a single transgenic field surrounded by conventional canola fields and the other two were derived from the theory that pollen flow decreased with distance in the form of a log decay curve. The predictions of all models suggested that the average frequency of resistance occurring from pollen flow in neighbouring canola fields, even when multiple transgenic fields are adjacent to the conventional fields, are likely to be below the current internationally accepted thresholds for contamination.