An empirically derived model of field-scale gene flow in winter wheat

The potential introduction of wheat (Triticum aestivum L.) cultivars with transgenic traits has generated increased interest in pollen-mediated gene flow (PMGF). The objectives of this study were to estimate wheat PMGF between commercial fields across multiple years and locations, and to compare estimates from large fields to those from smaller experimental plots. The study was conducted in a total of 56 commercial field locations in eastern Colorado in 2003, 2004, and 2005. We measured PMGF by tracking the movement of an imidazolinone herbicide resistance gene from resistant to susceptible cultivars, sampled at distances of 0.23 to 61 m. At least one sample from all 56 fields and from all 18 evaluated cultivars had detectable PMGF. The highest observed PMGF was 5.3% at 0.23 m. The farthest distance at which PMGF was detected was 61 m and the highest PMGF at that distance was 0.25%. Higher levels and greater distances of PMGF were detected in commercial fields than in experimental plots. Based on estimates from a generalized linear mixed model with a random location effect, the distance required to ensure 95% confidence that 95% of locations have PMGF less than 0.9% is 41.1 m for cultivars heading earlier than the pollen source and 0.7 m for cultivars heading later than the pollen source. These confidence limits should represent the highest levels of PMGF expected to occur in winter wheat in the west-central Great Plains and will be useful for wheat biotechnology regulation.