| File | What's in it |
|---|---|
| County bad-air days from EPA monitors 923 counties, 120 KB | Average days a year with an AQI over 100 and over 150, 2021 to 2025, the median AQI, the main cause and how many days the monitors reported. |
| County bad-air days by year from CDC estimates 3,108 counties, 375 KB | Days each year when the worst spot in the county went over 35.4 µg/m³ of fine particles or 70 ppb of ozone. |
| Average bad-air days by month 3,108 counties, 399 KB | Average smoky days (fine particles over 35.4 µg/m³) and high-ozone days in each month, 2001 to 2022, from the CDC estimates. |
How to credit us
The files are free to reuse under the Creative Commons Attribution 4.0 license. Credit "Tested Air" and link to this page. For example: Source: Tested Air analysis of EPA and CDC data, testedair.com/data/.
What the columns mean
- fips: the county's 5-digit FIPS code.
- bad_air_days_per_year_2021_2025: average days a year with an AQI over 100 at the county's worst monitor.
- unhealthy_days_per_year_2021_2025: average days a year over 150.
- main_cause: the pollutant that set the AQI on most bad days since 2016, one of: blowing dust; blowing dust (few bad days); blowing dust (few bad days); ozone not measured; blowing dust; ozone not measured; dust and ozone; dust and ozone (few bad days); dust and particles; dust and particles; ozone not measured; dust, particles and ozone; fine particles; fine particles (few bad days); fine particles (few bad days); ozone not measured; fine particles; ozone not measured; no bad days since 2016; ozone; ozone (few bad days); ozone (few bad days); fine particles not measured; ozone and dust; ozone and dust (few bad days); fine particles not measured; ozone and particles; ozone and particles (few bad days); ozone, dust and particles; ozone, dust and particles (few bad days); ozone; fine particles not measured; particles and dust; particles and dust; ozone not measured; particles and ozone; particles and ozone (few bad days); particles, ozone and dust; summer smoke and particles; summer smoke and particles (few bad days); summer smoke and particles (few bad days); ozone not measured; summer smoke and particles; ozone not measured; winter particles; winter particles (few bad days); winter particles (few bad days); ozone not measured; winter particles; ozone not measured. Where the monitors measure only one pollutant, the value is , since bad days from the other one aren't counted.
- monitor_days_per_year: how many days the monitors reported. Under about 330 usually means an ozone-season-only monitor.
- pm25_days_YEAR and ozone_days_YEAR: days that year when the worst modeled spot in the county went over 35.4 µg/m³ of fine particles or 70 ppb of ozone.
- pm25_days_MONTH and ozone_days_MONTH: the average number of those days in that month, 2001 to 2022.
How we count explains the method, and Sources lists the original files and their limits. Updated November 8, 2026.