In the mid-1970s it was widely recognized that we needed a uniform and easily understood system for monitoring air quality across the country to replace the ad hoc measures that had been used in various localities to alert the public to air pollution health hazards. An index subsequently was developed, primarily to warn people who are most likely to be susceptible to respiratory ailments but also to provide an ongoing record of changes in air quality across the United States. Today, as a result, residents of metropolitan areas are accustomed to daily reports on local air quality expressed by a common index (the Pollutant Standard Index, or PSI). At a level of 100, the air is considered unhealthful; at 200, it is very unhealthful; and at 300 and above, it is hazardous.
The creation of this nationwide system (which is supposed to cover the entire country by the end of 1983) is a notable administrative achievement, but it has many limitations, even assuming that all of the monitoring stations are functioning properly, following procedures perfectly, and are located so as to assure representative sampling. The limitations reflect a number of difficult decisions that go into the construction of such an index—decisions that often have to be made on the basis of inadequate information.
What pollutants should be included? PSI incorporates five—carbon monoxide, sulfur dioxide, nitrogen dioxide, ozone, and total suspended particulates. But these may not be the most important.
At what concentration level does each pollutant have a slight, moderate, or serious impact on comfort? (And in the long run, what is the cumulative effect on health of such exposures?) The PSI adopts the National Ambient Air Quality Standards previously established for these pollutants.
What relevant importance should be given to each pollutant or to the synergistic effect of a combination of pollutants in combining the data into a single index? The PSI finesses this problem by basing the index on whichever pollutant has the highest reading in relation to its ambient standard. Thus, an index of 150 may reflect a high concentration of carbon monoxide in one area and a high ozone level in another—a simplification that is perhaps acceptable for the public Information purposes for which the index was designed but that is dubious for broader scientific and policy analysis.
Whether the PSI serves its purpose well enough may be debated. Perhaps, over time, the choice of pollutants may have to be changed in response to new knowledge, ambient standards may be revised, or the procedure for weighing the importance of the different pollutants may be altered in calculating the index. Similarly, one may question whether a satisfactory general-purpose database, one that will be useful in answering a number of questions, is being developed as a result of the monitoring program being carried out in conjunction with the PSI. The program, in short, illustrates both a reasonable response to a need and the difficulty of achieving a purpose that has been articulated.
A version of this article appeared in print in the May 1981 issue of Resources magazine.
