Changes in criteria air pollution levels in the US before, during, and after Covid-19 stay-at-home orders: Evidence from regulatory monitors

The widespread and rapid social and economic changes from Covid-19 response might be expected to dramatically improve air quality. However, national monitoring data from the US Environmental Protection Agency for criteria pollutants (PM(2.5), ozone, NO(2), CO, PM(10)) provide inconsistent support for that expectation. Specifically, during stay-at-home orders, average PM(2.5) levels were slightly higher (~10% of its multi-year interquartile range [IQR]) than expected; average ozone, NO(2), CO, and PM(10) levels were slightly lower (~30%, ~20%, ~27%, and ~1% of their IQR, respectively) than expected. The timing of peak anomaly, relative to the stay-at-home orders, varied by pollutant (ozone: 2 weeks before; NO(2), CO: 3 weeks after; PM(10): 2 weeks after); but, by 5–6 weeks after stay-at-home orders, the concentration anomalies appear to have ended. For PM(2.5), ozone, CO, and PM(10), no US state had lower-than-expected pollution levels for all weeks during stay-at-home-orders; for NO(2), only Arizona had lower-than-expected levels for all weeks during stay-at-home orders. Our findings show that the enormous changes from the Covid-19 response have not lowered PM(2.5) levels across the US beyond their normal range of variability; for ozone, NO(2), CO, and PM(10) concentrations were lowered but the reduction was modest and transient.