A novel modeling system for estimating pollen emissions: Application to Northeast US
Christos Efstathiou1, Leonard Bielory2, and Panos Georgopoulos1
1UMDNJ - RW Johnson Medical School; 2UMDNJ - NJ Medical School
Rationale: Allergic airway diseases represent a complex health problem which can be exacerbated by the synergism of pollen and air pollutants. Understanding exposures to pollen requires accurate estimates of the spatial distribution of pollen emissions. Currently there are no established methods for estimating pollen emissions across the United States.
Methods: A prototype algorithm for estimating emissions of pollen particles from major allergenic tree and plant families in the United States was developed, extending the approach for modeling biogenic gas emissions in the Biogenic Emission Inventory System (BEIS). Pollen emissions are influenced by meteorological factors such as temperature, humidity, wind speed, photosynthetically active radiation, and total growing degree days. A spatio-temporal vegetation map was constructed from different remote sensing sources and local surveys, and was coupled with the Mesoscale Meteorological model (MM5) to develop pollen emissions rates. Ground level measurements of pollen levels (from trees, grasses, and weeds) in the NE US were used to evaluate predictions with respect to the start of pollen release for various species.
Results: A novel system for modeling pollen emissions was developed and applied. Allergenic species-specific emission rates exceeding 3000 grains/m2 were estimated in some locations, and strong correlations were observed between predicted pollen emission rates and measured pollen counts.
Conclusion: The approach presented here allows prediction of pollen emissions, and provides an important component in assessing exposures of populations to pollen. This system overcomes limitations posed by the lack of temporally resolved dynamic vegetation mapping in traditional pollen emission estimation methods.