Characterization of National Spatial Variation

Spatial variability is an important quality of air pollutants for many areas of policy within the U.S. Environmental Protection Agency (EPA). Obviously, monitoring regulations depend heavily on knowledge of spatial variability. In addition, control strategies depend on this knowledge, which helps determine whether a local or regional program would be more effective. Action day programs and public information programs also benefit from this knowledge. Traditionally, spatial variation has been depicted by isopleth maps, concentration maps, and box plots of various sites. Does this really give us useful knowledge about spatial variation? This paper explores a new way to examine spatial variability on a national scale and also presents an extension of this method in an attempt to characterize spatial variability in a useful way. The new methodology is presented along with its application using PM 2.5 and ozone data.