Modelling airborne concentration and deposition rate of maize pollen

The introduction of genetically modified (GM) crops has reinforced the need to quantify gene flow from crop to crop. This requires predictive tools which take into account meteorological conditions, canopy structure as well as pollen aerodynamic characteristics. A Lagrangian Stochastic (LS) model, called SMOP-2D (Stochastic Mechanistic model for Pollen dispersion and deposition in 2 Dimensions), is presented. It simulates wind dispersion of pollen by calculating individual pollen trajectories from their emission to their deposition. SMOP-2D was validated using two field experiments where airborne concentration and deposition rate of pollen were measured within and downwind from different sized maize (Zea mays) plots together with micrometeorological measurements. SMOP-2D correctly simulated the shapes of the concentration profiles but generally underestimated the deposition rates in the first 10 m downwind from the source. Potential explanations of this discrepancy are discussed. Incorrect parameterisation of turbulence in the transition from the crop to the surroundings is probably the most likely reason. This demonstrates that LS models for particle transfer need to be coupled with air-flow models under complex terrain conditions.