Ours is a study that quantifies the invasion speed of an invasive weedy herbicide resistant genotype. The incidence of glyphosate resistant weed species (12) has increased rapidly over the past five years as the proportion of arable land treated with glyphosate has climbed (in excess of 90% of US soybean production is now treated with glyphosate). One resistant species, Conyza canadensis, has spread rapidly in the mid-Atlantic and southeastern states. We set out to quantify the dispersal distance of the species in order to estimate its invasion speed and then use knowledge about spread rates to explore area-wide management practices that might contain the species and limit local abundance.

Results/Conclusions

In a series of large scale field studies using on-the-ground sampling and remotely piloted aircraft, we determined that dispersal is stratified with 90% of the seed settling within 15m of source plants while small numbers of seed were found at the ends of 600 m long transects. At the same time seed were trapped on the ground to quantify the two dimensional dispersal kernel, seed were detected and collected with remotely piloted aircraft at 60 and 110 m above the ground surface, well into the atmospheric boundary layer. With an exceptionally small settling velocity and many of its seed dispersed by midday, we estimate that seed in the atmospheric boundary layer are traveling tens of km in a single day. Results from the empirical studies were used to guide the development of a spatially explicit model to assess strategies for managing this invasive herbicide resistant genotype. The model projects a disturbing reality, without significant reductions of glyphosate use and without a spatially explicit and landscape scale coordination of weed and crop management practices, glyphosate resistant weeds will continue to evolve resistance to glyphosate and spread across the landscape. At the same time, the crop seed and input industry is rapidly developing crops with multiple insect and herbicide resistance traits which will only exacerbate the problem of herbicide resistance in weedy populations. Models like that reported herein provide badly needed syntheses of data and observation to guide the design of more sustainable agricultural practices. They also call into question the wisdom of largely herbicidal solutions to the glyphosate resistant problem. Such information is needed now for current and future glyphosate resistant crop regulatory decisions and cropping systems recommendations.

Keywords: Genetically modified organisms, herbicide resistance, landscape heterogeneity, invasion speed