Pollen transport and viability evaluated using combined large-eddy simulation and Lagrangian stochastic models

We have coupled a large-eddy simulation model, a Lagrangian particle dispersion model, and a pollen viability model to assess the effect of boundary-layer dynamics and thermodynamics on the dispersion of maize pollen. The large settling velocity of maize pollen implies that it can be transported long distances only if there are turbulent motions sufficient to counteract the gravitational settling of maize pollen grains and lift them to substantial heights. Recent studies using both piloted an remote-controlled aircraft have in fact shown the presence of viable maize pollen through the entire depth of the convective boundary layer, suggesting a mechanism for long-range transport of viable maize pollen. Our goal here is to develop an interdisciplinary modeling framework to explore this mechanism.

The large-eddy simulation model is a version of the Advanced Regional Prediction System (ARPS) numerical model. We used ARPS to simulate boundary-layer evolution from sunrise to sunset and archived the predicted velocity components, thermodynamic variables, and turbulent kinetic energy every 60 seconds of simulated time. These results provided the flow and turbulence fields for a Lagrangian-stochastic model of particle dispersion. We tracked the motions of approximately 100,000,000 particles that were released near the surface and transported using the velocity fields and turbulent kinetic energy simulated by ARPS. The tracer particles were interpreted as a statistical sample of the actual pollen cloud. The Lagrangian formalism is especially useful for the present application because it permits diagnosis of environmental temperature and moisture experienced by the pollen grain as it travels. Accordingly, we diagnosed pollen moisture content using a formulation that accounts for the drying of pollen grains as a function of atmospheric vapor pressure deficit and time. Viability was then found a function of pollen moisture content; i.e., as pollen grains dried, they gradually lost viability. The viability of each sample particle was evaluated as a function of time according to the environment through which it traveled until it was deposited at the surface.

Results from the combined models show that small amounts of pollen were transported 5 km or more from the source. About 30% of pollen was predicted to retain its viability at this distance. This can be attributed in part to the thermal and moisture structure that typifies the daytime atmospheric boundary layer. Temperature decreases approximately as the dry adiabatic lapse rate while absolute humidity remains approximately constant, so pollen that is lofted in the boundary layer enters an environment with low vapor pressure deficit which helps preserve its viability. The amount of pollen deposited at this distance was very low, so that it may be difficult for such small quantities of fugitive pollen to out-compete locally generated pollen. We recommend that field experiments be carried to determine the existence of low levels of outcross at long distances. Our study illustrates the multi-disciplinary nature of this problem, namely the relation of pollen transport and viability to atmospheric dynamics and thermodynamics.