Understanding the variability of air pollutants outdoors, as well as identifying their sources, is integral to our understanding of how humans are exposed to these compounds.  Since the concentrations of air pollutants present in outdoor air undoubtedly influence the potential for human exposure, identifying the sources and understanding the behavior of these compounds can be used to produce effective mitigation strategies.  To that end, the U.S. EPA is conducting the Detroit Exposure and Aerosol Research Study (DEARS) to measure concentrations of acrolein and volatile organic compounds (VOCs) inside and outside of participants' homes, from personal exposures and at a central community site.  Residences were chosen in five neighborhoods based on proximity to a major roadway or point source region in the Detroit area.  Outdoor acrolein concentrations from summer 2004 data indicated no significant differences between the neighborhoods when analyzed using the least squares means function at a significance level of p<0.05.  This suggests that the compound was predominantly formed in the atmosphere as a secondary pollutant, rather than from direct emissions.  Outdoor acrolein concentrations at each site, however, showed high temporal variability with standard deviations exceeding the means.  For example, the mean outdoor acrolein concentration measured in one Detroit neighborhood was 1.8 ± 2.4 µg/m³ (mean ± std. dev.).  Inspection of the impact of a residence's location downwind versus upwind of a potential roadway source indicated no significant differences with acrolein.  Analysis of select outdoor VOC concentrations (benzene, toluene, 1,3-butadiene) showed significant spatial variability, where concentrations at the urban sites (1.4 ± 1.5 ppb for benzene) were generally twice as high as the rural monitoring site (0.8 ± 1.7 ppb for benzene).  This indicates that these VOCs are likely products of urban/industrial sources.  Substantial temporal variability was evident with benzene (1.4 ± 1.4 ppb in one Detroit neighborhood) and 1,3-butadiene (0.3 ± 1.6 ppb in this same neighborhood) concentrations.  While acrolein did not show evidence of primary emission from a local source, results suggest that concentrations of benzene, toluene, and 1,3-butadiene were emitted primarily from urban/industrial sources.  Between the various urban sites, levels of these VOCs did not vary significantly.  Disclaimer:  Although this work was reviewed by EPA and approved for publication, it may not necessarily reflect official Agency policy.