Comparisons of simple and complex methods for quantifying exposure to individual point source air pollution emissions

Area of Science:

• Atmospheric Chemistry
• Environmental Modeling
• Public Health

Background:

• Chemical transport models (CTMs) provide detailed atmospheric simulations but are computationally intensive for large-scale source impact analyses.
• Epidemiology and environmental justice studies increasingly require evaluating impacts from numerous sources, necessitating efficient modeling approaches.

Purpose of the Study:

• To compare the accuracy of reduced-complexity models against a sophisticated CTM for assessing population-weighted PM2.5 impacts from U.S. coal power plants.
• To evaluate the performance of a wind field-based Lagrangian model (HyADS) and an inverse distance method against CTM adjoint sensitivities.

Main Methods:

• Modeled population-weighted PM2.5 impacts from over 1100 U.S. coal power plants for 2006 and 2011.
• Utilized three approaches: GEOS-Chem CTM adjoint sensitivities, the HyADS model, and an inverse distance calculation.
• Quantified errors using normalized mean error and root mean square error.

Main Results:

• HyADS and the inverse distance approach showed normalized mean errors of 20–28% and RMSEs of 0.0003–0.0005 µg m⁻³ compared to CTM adjoint sensitivities.
• Reduced complexity models performed best near and downwind of sources.
• Model accuracy decreased farther from sources and upwind, especially when wind fields were not considered.

Conclusions:

• Simplified modeling approaches, like HyADS and inverse distance, offer viable, computationally efficient alternatives to CTMs for assessing PM2.5 impacts from large numbers of sources.
• The accuracy of these simplified models is dependent on proximity to the source and consideration of wind fields.
• These findings support the expanded use of reduced-complexity models in environmental justice and epidemiological research.