AAAR 34th Annual Conference
October 12 - October 16, 2015
Hyatt Regency
Minneapolis, Minnesota, USA
Abstract View
Understanding the Character and Dynamics of Organic Aerosol in the Houston Area Using Multi-way Factor Analysis
NANCY SANCHEZ, Yu Jun Leong, Henry Wallace, Basak Karakurt Cevik, James Flynn, Barry Lefer, Robert Griffin, Rice University
Abstract Number: 105 Working Group: Urban Aerosols
Abstract Organic aerosol (OA) has been identified in several previous studies as the dominant constituent of particulate matter (PM) in different rural and urban locations. Investigation of the sources, formation pathways, transformation processes and chemical character of this fraction is crucial to better elucidate the impacts of submicron aerosol on climate and human health. Furthermore, as more stringent standards for PM concentration are promulgated by the EPA, the formulation of public policies intended to reduce/control PM levels in cities like Houston, where nonattainment zones for PM have been identified, requires an improved understanding of the evolution of this pollutant in the atmosphere. Although factor analysis techniques such as positive matrix factorization (PMF) provide information on the components of the organic fraction of submicron aerosol, the extension of PMF to multi-way analyses such as 3-D parallel factor (PARAFAC) analysis in order to incorporate information on the size distribution of the OA, offers an improved approach to aerosol apportionment. In this work, PARAFAC analysis was applied to size-resolved organic aerosol composition data sets collected with an Aerodyne high-resolution time-of-flight aerosol mass spectrometer in the Greater Houston area (GHA) during the DISCOVER-AQ 2013 campaign and the Houston Aerosol Characterization and Health Experiment, a comprehensive field study covering multiple locations in the GHA conducted from 2013 to 2015. The PARAFAC-based analysis of these data sets provided information on (i) the main fractions present in the OA in Houston and their specific size distribution, (ii) the spatial and temporal distribution of the constituents of the OA across the GHA and (iii) the variation in the OA fraction size distribution associated with aerosol processing during specific periods/events. The results obtained in this study provide further insight on the dynamics of OA in Houston and might support future policies oriented to control PM levels in the GHA