PNNL

Abstract

Allergenic pollen has been an increasing concern under the changing climate because its synergism with air pollutants to cause allergic airway disease. A modeling system incorporating pollen emission and transport has been developed to simulate the spatiotemporal distributions of allergenic pollen of representative trees, weeds and grasses. The emission model was described in details in an accompanying paper. The model performance was evaluated using hit and false rates and fractional bias based on the observed pollen count at monitor stations across the contiguous United States in 2004. For oak and ragweed pollen, the hit rates for three ambient pollen levels were generally between 70% and 100% at the majority of studied stations; the false 9 rates were generally between 0% and 10%; the fractional bias were between -0.4 and 0.4. The dry deposition, emission and vertical eddy diffusion were the dominant processes determining ambient pollen concentrations. The boundary condition exerted less influence on mean pollen concentrations, but remarkable influences on maximum pollen concentrations. The modeling system was also tested as a tool to study the co-occurrences of pollen and air pollutants, and to investigate climate change effects on allergenic pollen season.