AAAR 34th Annual Conference
October 12 - October 16, 2015
Hyatt Regency
Minneapolis, Minnesota, USA
Abstract View
Development of Fine Particulate Matter Source Profiles Using a Nonlinear Optimization Approach
Cesunica Ivey, Nabil Abdurehman, Xinxin Zhai, Yongtao Hu, James Mulholland, ARMISTEAD G. RUSSELL, Georgia Institute of Technology
Abstract Number: 265 Working Group: Source Apportionment
Abstract In previous studies, a nonlinear optimization approach was used to estimate daily fine particulate matter (PM2.5) source impacts over continental U.S (36-km). This hybrid approach takes into account observed speciated PM2.5 concentrations, chemical transport model estimates of PM2.5, as well as uncertainty of both data sources, in order to minimize the discrepancy between the two. Optimized source impacts capture regional and seasonal trends in PM2.5 concentrations. In this work, a similar nonlinear optimization approach is used to estimate PM2.5 source profiles for primary emissions from twenty sources of interest, including mobile sources (both diesel and gasoline), open biomass burning, aircraft, and non-road sources. Optimized source impacts are used as inputs in the calculation of new source profiles. The profiles estimate relative contributions of elemental and organic carbon, major ions, and 17 trace metals. Newly optimized source profiles are evaluated by implementation in CMB at various monitoring locations and comparing results to observed concentrations. Optimizing source profiles using chemical transport model results captures ambient processes that are not often considered in traditional source profile estimation methods. The optimization approach can be applied regionally to capture differences in source composition over continental U.S.