AAAR 35th Annual Conference October 17 - October 21, 2016 Oregon Convention Center Portland, Oregon, USA
Abstract View
Measurement of Emissions and Air Quality Near a Major Refinery
Henry Wallace, Nancy Sanchez, Courtney L. Herring, Timothy M. VanReken, James Flynn, Matthew H. Erickson, Barry Lefer, ROBERT GRIFFIN, Rice University
Abstract Number: 566 Working Group: Urban Aerosols
Abstract During the Benzene and other Toxics Exposure (BEETEX) Study during February 2015, we made stationary measurements of trace gases and particulate matter using a mobile air quality laboratory (MAQL). The site was located 50 m south of a large refinery along the notoriously polluted Houston Ship Channel. Being proximate to the largest petrochemical refining complex and the second busiest port in the United States, neighborhoods in the area are impacted by a variety of emissions. The location and time of the campaign were ideal for observations of primary emissions. Observations of non-refractory submicron particulate matter were made with a High-Resolution Time-of-Flight Aerosol Mass Spectrometer, of particle-bound polycyclic aromatic hydrocarbons (PAH) with a photoelectric aerosol sensor (PAS), and of volatile organic compounds with a Proton Transfer Reaction Mass Spectrometer. In addition, the MAQL has the capability to measure photochemically relevant trace gases and meteorological parameters.
This work presents our results and analyses of this data set. First, we compare PAH measurements estimated using analysis of ion fragments from the HR-ToF-AMS to those made with the PAS. Second, we perform principle component analysis of particulate organics, nitrate, sulfate, ammonium, and PAH and gas-phase compounds including VOCs, nitrogen oxides, total reactive nitrogen, sulfur dioxide, carbon monoxide, and ozone. We present a five-factor solution that explains 69.4% of the variance and identifies sources influencing the site. Factors affecting the composition of the site are vehicle emissions, tank and oil line emissions, primary emissions of particulate sulfate, nitrate and ammonium, diel variability and industrial solvent use. Finally, we present a three-factor solution (hydrocarbon-like, less oxidized, and more oxidized organic aerosol for positive matrix factorization modeling for the source apportionment of particulate matter, which indicates that hydrocarbon-like OA is the dominant factor at this site.