Maize Pollen Dispersal under Convective Conditions

Abstract

The widespread adoption of genetically modified (GM) crops has led to a need to better understand the atmospheric transport of pollen because of concerns over potential cross-pollination between GM and non-GM crops. Maize pollen concentrations were modeled by a modified Lagrangian stochastic (LS) model of the convective boundary layer (CBL) and were compared with concentrations measured by airborne remotely piloted vehicles (RPVs) flown from directly above to 2 km from source fields. The turbulence parameterization in an existing CBL LS model was modified to incorporate the effects of shear-driven turbulence, which has an especially large impact near the surface, where maize pollen is released. The modified model was used to calculate concentrations corresponding to the RPV flight tracks. For the most convective cases, when at least 95% of the pollen came from sources near the RPV flight track for which source strength measurements are available and the results are less sensitive to uncertainty in wind direction since most of the pollen came from directly beneath the flight track, the geometric mean of the ratio between the modeled and measured concentrations was 0.94. When cases with larger contributions from more distant fields were included, the overall geometric mean decreased to 0.43. The scatter of the measured concentrations about the modeled values followed a lognormal distribution. These results indicate that the modified model presented herein can substantially improve the description of the near-source dispersion of heavy particles released near the surface during convective conditions.