Use of Air Quality Dispersion Models to Reconstruct Site Exposure Profiles

Air quality dispersion modeling can be used to estimate both past and future localized exposures to occupational pollutants based on emission sources and meteorological conditions. There are several air quality dispersion models available, each with capabilities that are specialized for particular air quality situations. To demonstrate which models are appropriate for different applications, a suite of air quality models and geographical data analysis software was reviewed, including AERMOD, CALPUFF, HYSPLIT, QUIC, and ArcGIS. A similar simulation was run with all models except CALPUFF, which was not run due to weather data availability limitations. All models produced similar air pollutant dispersion patterns when given similar emission sources and meteorological input. However, QUIC produced substantially larger concentrations near emission sources than the other models reviewed. This is most likely due to a combination of QUIC overpredicting concentrations and the other models underpredicting concentrations. Comparison of each models unique features and capabilities indicates that QUIC is the most relevant for short-term exposure applications where the exposure map for time steps less than 1 hour is desired, while AERMOD is more appropriate for estimating long-term exposures where the exposure map for time steps greater than 1 hour up to several years is desired. HYSPLIT is appropriate for a preliminary assessment of a current air quality situation e.g., smoke from a forest fire. Uses of ArcGIS for the creation of surface geometry and for analysis of observed air quality were also discussed. Based on the information reviewed in this report, air quality dispersion modeling has been shown to be a useful tool to estimate past exposures and predict future exposures to support Air Force infrastructure decisions, epidemiological investigations, and the Air Force Total Exposure Health Initiative.