AAAR 33rd Annual Conference
October 20 - October 24, 2014
Rosen Shingle Creek
Orlando, Florida, USA
Abstract View
Temporal and Regional Analysis of Spatially-Resolved PM2.5 Source Apportionment Results over Continental US
CESUNICA IVEY, Heather Holmes, Yongtao Hu, James Mulholland, Armistead Russell, Georgia Institute of Technology
Abstract Number: 190 Working Group: Source Apportionment
Abstract In recent investigations, human exposure to particulate matter with an aerodynamic diameter of 2.5 microns or less (PM2.5) was found to be correlated with adverse health impacts such as low birth weight, increased mortality rate, and exacerbation of existing cardiovascular and respiratory illness. Consequently, it is of interest to determine the sources of PM2.5 and how ambient concentrations attributed to those sources correlate with negative health impacts. Determining these associations is challenging due to spatially and temporally limited ground measurements of PM2.5 and PM species concentrations, uncertainty in PM2.5 emissions inventories, and source impact modeling limitations. Epidemiological studies that investigate these relationships are limited by spatially and temporally sparse source impact data.
In this work, a novel hybrid source apportionment method is employed that addresses these challenges. First, the Community Multi-scale Air Quality Model (CMAQ), equipped with the decoupled direct method for analysis of PM2.5 sensitivities to emissions, is used to produce daily, 36-km resolution source impacts over continental U.S. Then the hybrid method uses nonlinear optimization to minimize the discrepancies between ground observations of speciated PM2.5 and source impact results from CMAQ. Observations are available every third day from the Chemical Speciation Network. Finally, spatiotemporal kriging is applied to spatially interpolate adjustments, and the adjustments are implemented to compute daily, spatial source impact results for the entire year 2006. Source impacts are estimated for 20 source categories including, fossil fuel combustion, biogenics, livestock operations, aircraft, diesel and gasoline vehicles, metals processing, sea salt, and dust. Spatial trends are analyzed for 10 regions, which are chosen based on the Environmental Protection Agency region classification. Seasonal trends are determined by performing time series analyses. The results of this work provide spatially and temporally complete source impact data for use in studies seeking to evaluate the relationship of PM2.5 sources to adverse human health impacts.