29th Conference on Agricultural and Forest Meteorology

2B.2

Modeling pollen dispersion in a Swiss valley using a Lagrangian dispersion model

PAPER WITHDRAWN

Michael J. Klocke-Sullivan, Iowa State University, Ames, IA; and R. W. Arritt

The increased use of genetically modified crops has created the need for accurate modeling of pollen dispersion over various topographical conditions. We have used the WRF model, a high resolution atmospheric model, along with a Lagrangian dispersion model to predict pollen dispersion over highly complex terrain in a narrow valley in Urner Reusstal Switzerland. The valley is approximately 2.5 km wide at an altitude of 440 m. The valley is surrounded by a lake to the north and mountains on either side rising to approximately 1800-2600 m altitude.

Bannert et al. (2006) quantified cross pollination using different colors of maize varieties in this valley. Observations showed that cross-pollination rates were lower than other international studies but the investigators were unable to give a specific reason for this occurrence. Modeling pollen dispersion in this narrow and isolated valley allowed us to run the WRF model at a higher resolution than previous studies.

We modeled pollen dispersion for this observation period and compared the out-crossing quantified in Bannert et al. (2006) with that predicted in our dispersion model. The topography for this case produced complex flow fields that showed strong variability throughout the day. This variability is attributed to interactions of terrain with the diurnal cycle of heating and cooling. These temporally and spatially varying flow fields in turn produced variations in the transport and deposition of simulated maize pollen. Winds at the valley floor were generally weak, likely explaining the low rate of cross-pollination found in the observations.

Session 2B, Pollen Transport in the Atmospheric Boundary Layer
Monday, 2 August 2010, 3:30 PM-4:15 PM, Crestone Peak III & IV

Previous paper  Next paper

Browse or search entire meeting

AMS Home Page