ABSTRACT
Although spatial variations in ambient air quality are important to human health and efforts to improve air quality, a historical lack of data has hampered spatial analysis of air quality. The mapping of recent ambient atmospheric particulate monitoring data in ten urban areas reveals certain similarities, including concentric patterns and localized maxima and minima. More detailed analyses utilizing a stratified random sampling of the concentration isopleths reveal two important conclusions. First, existing monitoring data are not spatially representative, but rather tend to overestimate both average and extreme areawide concentrations. Second, particulate concentrations are approximately lognormally distributed in space. This latter result can be explained by distribution of areal land uses rather than by physical factors.
Notes
1I wish to thank Mr. Frank Record of GCA Technology Division for providing me with a complete copy of the National Urban Particulates Study and the data that greatly facilitated this research. I am also indebted to D. Michael Kirchoff for expert cartographic assistance, Professors David de Laubenfels and Mark Monmonier, Department of Geography, Syracuse University, and an anonymous reviewer for helpful comments on an earlier draft of this paper.