Air quality dataset details

Interpreting the data

What these data tell us:

These data can be used to identify yearly measures of ozone levels within counties where measurements have been taken.
These data represent two metrics for ozone.
These data also include four metrics for fine particulate matter (PM) smaller than 2.5 microns in diameter.

What these data do not tell us:

These data do not give us measures of actual exposure to ozone or PM 2.5 for individual or communities.
These data can only reflect air quality in specific locations as values are taken from fixed points. However, air quality can vary within a county.
Comparisons of air quality data to health measures are done at an aggregate level. Areas with poorer air quality do not necessarily mean there will be higher rates for certain health effects. Different factors, such as lifestyle, can contribute to the health of an individual.

Limitations of the data

It is not representative of overall air quality. The data are only a small snapshot of air quality characteristics in an area.

When there is more than one air monitor within a county, the highest reading on any day is used. Estimates for larger counties with more than one air monitor may be biased higher than counties with one air monitor.

These data do not give us any information about air quality in counties where there are not air monitors. Air monitors tend to be located in urban areas, where the most people live. However, air quality is often worse than in rural areas.

About these measurements

The ozone indicator is comprised of two measures. The first measure is the number of days above the National Ambient Air Quality Standards (NAAQS). The second measure is the number of person-days above the NAAQS. These measures are provided only for counties where air quality monitors are maintained and measurements are taken. Not all counties have monitors or take measurements. Data is available for the years 2001 – 2016. Data may not be available for each monitor each year.

The NAAQS for ozone data collected for an 8-hour average:

Prior to 2008 - 0.080 parts per million (ppm).
From 2008 until October 2015 - 0.075 ppm.
From October 2015 to present - 0.070 pp.

These data also include four metrics for fine PM smaller than 2.5 microns in diameter:

Annual average PM 2.5 concentration.
Number of person-days over the air quality standard.
Percent of days over the air quality standard.
Percent of the population living in areas where the air quality standard for PM 2.5 is met. This metric is compared to areas where standards are not met, or there is no monitoring or data available.

The air quality standard (NAAQS) for PM 2.5 is measured multiple ways. Two primary standards were used in the development of these metrics. The first standard is the annual average concentration averaged over 3 years, 12 µg/L. The second standard is the 24 hour 98th percentile averaged over 3 years, 35 µg/L.

The data are obtained only from monitors that are designated as Federal Reference Methods or equivalent. The data include values associated with exceptional events such as high winds, fires, or construction.

Calculation methods

EPA’s DataMart was used to access datasets and supplemental data for all monitoring sites across the United States. These calculations only include data from monitors that meet the minimum data completeness criteria set by the EPA.

The ozone measure is of the 8-hour average of hourly measurements taken. A day is considered to be above the standard for ozone if the maximum 8-hour average exceeds the NAAQS.

Calculating person-days is done by multiplying the number of days that monitored above in a county by it's population.

The PM 2.5 measure is measured in two ways:

Annual average, averaged over 3 years.
- An annual average for an area is considered to be above the standard if the average is 12 µg/L.
24-hour standard.
- This standard is taken at the 98th percentile for a 3-year average. Then, it is applied to the percent of days and person-days measure.
- The annual standard is applied to the annual concentration and percent of population measures.