Copyright © 2007 Elsevier Ltd All rights reserved.
Sources of ambient fine particulate matter at two community sites in Detroit, Michigan
Davyda M. Hammonda,
, 
, J. Timothy Dvoncha,
Gerald J. Keelera,
b,
Edith A. Parkerc,
Ali S. Kamala,
James A. Barresa,
Fuyuen Y. Yipa,
d
and Wilma Brakefield-Caldwelle
aDepartment of Environmental Health
Sciences, University of Michigan School of Public Health, Ann Arbor, MI,
USA
bDepartment of Atmospheric, Oceanic, and
Space Sciences, College of Engineering, University of Michigan, Ann Arbor, MI,
USA
cDepartment of Health Behavior and Health
Education, University of Michigan School of Public Health, Ann Arbor, MI,
USA
dNational Center for Environmental Health,
Centers for Disease Control and Prevention, Atlanta, GA, USA
eCommunity Action Against Asthma, Detroit, MI, USA
Received 28 June 2007; revised 26 September 2007; accepted 28
September 2007. Available online 5 October 2007.
Abstract
Detroit, Michigan is a non-attainment area of the annual PM2.5
(particles 
2.5 μm in diameter) National Ambient Air Quality Standard (NAAQS),
and contains a host of local pollution contributors including high diesel
traffic from a nearby international border crossing. A source apportionment
analysis was conducted using PM2.5 data collected from 1999 to 2002
by the Community Action Against Asthma (CAAA) project in Detroit, Michigan. CAAA
used a community-based participatory research approach to identify and address
the environmental triggers for asthma among children residing in southwest and
east Detroit. The data used for the study included 24-h measurements of
PM2.5 mass, elemental and organic carbon, and a suite of trace
element species, along with hourly measurements of PM2.5 mass and
black carbon. Positive matrix factorization (PMF2) was used to quantitatively
apportion the sources of ambient PM2.5 at each of two Detroit
community sites. Results showed that southwest Detroit PM2.5 levels
can be apportioned to seven source categories: secondary sulfate/coal
combustion, gasoline vehicles, diesel vehicles, refinery/oil combustion,
iron–steel manufacturing/waste incineration, automotive electroplating, and
sewage sludge incineration that includes crustal material from runoff. The PMF2
model apportioned the east Detroit PM2.5 data into five source
categories: secondary sulfate/coal combustion, motor vehicles/combustion,
refinery/oil combustion, iron–steel manufacturing/waste incineration, and
automotive electroplating. For both locations, approximately over 60% of the
PM2.5 mass was attributed to secondary sulfate/coal combustion
sources, approximately 30% to vehicular sources, and 1–5% to local industrial
sources. The unexplained mass accounted for <2% of the measured
PM2.5 mass. This study illustrates that regional secondary
sulfate/coal combustion and local motor vehicle emissions alone are enough for
this mid-western US city to be in non-attainment for the annual PM2.5
NAAQS.
Keywords: Receptor modeling; Positive matrix factorization; PM2.5; Trace elements