Analysis of short-term ozone and PM_2.5 measurements: Characteristics and relationships for air sensor messaging

ABSTRACT

Air quality sensors are becoming increasingly available to the general public, providing individuals and communities with information on fine-scale, local air quality in increments as short as 1 min. Current health studies do not support linking 1-min exposures to adverse health effects; therefore, the potential health implications of such ambient exposures are unclear. The U.S. Environmental Protection Agency (EPA) establishes the National Ambient Air Quality Standards (NAAQS) and Air Quality Index (AQI) on the best science available, which typically uses longer averaging periods (e.g., 8 hr; 24 hr). Another consideration for interpreting sensor data is the variable relationship between pollutant concentrations measured by sensors, which are short-term (1 min to 1 hr), and the longer term averages used in the NAAQS and AQI. In addition, sensors often do not meet federal performance or quality assurance requirements, which introduces uncertainty in the accuracy and interpretation of these readings. This article describes a statistical analysis of data from regulatory monitors and new real-time technology from Village Green benches to inform the interpretation and communication of short-term air sensor data. We investigate the characteristics of this novel data set and the temporal relationships of short-term concentrations to 8-hr average (ozone) and 24-hr average (PM_2.5) concentrations to examine how sensor readings may relate to the NAAQS and AQI categories, and ultimately to inform breakpoints for sensor messages. We consider the empirical distributions of the maximum 8-hr averages (ozone) and 24-hr averages (PM_2.5) given the corresponding short-term concentrations, and provide a probabilistic assessment. The result is a robust, empirical comparison that includes events of interest for air quality exceedances and public health communication. Concentration breakpoints are developed for short-term sensor readings such that, to the extent possible, the related air quality messages that are conveyed to the public are consistent with messages related to the NAAQS and AQI.

Implications: Real-time sensors have the potential to provide important information about fine-scale current air quality and local air quality events. The statistical analysis of short-term regulatory and sensor data, coupled with policy considerations and known health effects experienced over longer averaging times, supports interpretation of such short-term data and efforts to communicate local air quality.

Acknowledgment

The authors thank the entire EPA Air Sensors Team, including but not limited to those who specifically provided input on the data, analysis, and/or methods: Ron Evans, Brett Gantt, Gayle Hagler, Bryan Hubbell, Jesse McGrath, Liz Naess, Ben Wells, Karen Wesson, and Ronald Williams.

Additional information

Notes on contributors

Elizabeth Mannshardt

The authors are colleagues in EPA’s Office of Air Quality Planning and Standards in Research Triangle Park, North Carolina. Elizabeth Mannshardt is a Statistician; Kristen Benedict is a Physical Scientist; David Mintz is a Senior Statistician; and Richard Wayland is the Division Director in the Air Quality Assessment Division. Martha Keating is a Policy Advisor; Scott Jenkins is a Senior Environmental Health Scientist; and Susan Stone is a Senior Environmental Health Scientist, all with the Health and Environmental Impacts Division.

Kristen Benedict

The authors are colleagues in EPA’s Office of Air Quality Planning and Standards in Research Triangle Park, North Carolina. Elizabeth Mannshardt is a Statistician; Kristen Benedict is a Physical Scientist; David Mintz is a Senior Statistician; and Richard Wayland is the Division Director in the Air Quality Assessment Division. Martha Keating is a Policy Advisor; Scott Jenkins is a Senior Environmental Health Scientist; and Susan Stone is a Senior Environmental Health Scientist, all with the Health and Environmental Impacts Division.

Scott Jenkins

The authors are colleagues in EPA’s Office of Air Quality Planning and Standards in Research Triangle Park, North Carolina. Elizabeth Mannshardt is a Statistician; Kristen Benedict is a Physical Scientist; David Mintz is a Senior Statistician; and Richard Wayland is the Division Director in the Air Quality Assessment Division. Martha Keating is a Policy Advisor; Scott Jenkins is a Senior Environmental Health Scientist; and Susan Stone is a Senior Environmental Health Scientist, all with the Health and Environmental Impacts Division.

Martha Keating

The authors are colleagues in EPA’s Office of Air Quality Planning and Standards in Research Triangle Park, North Carolina. Elizabeth Mannshardt is a Statistician; Kristen Benedict is a Physical Scientist; David Mintz is a Senior Statistician; and Richard Wayland is the Division Director in the Air Quality Assessment Division. Martha Keating is a Policy Advisor; Scott Jenkins is a Senior Environmental Health Scientist; and Susan Stone is a Senior Environmental Health Scientist, all with the Health and Environmental Impacts Division.

David Mintz

The authors are colleagues in EPA’s Office of Air Quality Planning and Standards in Research Triangle Park, North Carolina. Elizabeth Mannshardt is a Statistician; Kristen Benedict is a Physical Scientist; David Mintz is a Senior Statistician; and Richard Wayland is the Division Director in the Air Quality Assessment Division. Martha Keating is a Policy Advisor; Scott Jenkins is a Senior Environmental Health Scientist; and Susan Stone is a Senior Environmental Health Scientist, all with the Health and Environmental Impacts Division.

Susan Stone

The authors are colleagues in EPA’s Office of Air Quality Planning and Standards in Research Triangle Park, North Carolina. Elizabeth Mannshardt is a Statistician; Kristen Benedict is a Physical Scientist; David Mintz is a Senior Statistician; and Richard Wayland is the Division Director in the Air Quality Assessment Division. Martha Keating is a Policy Advisor; Scott Jenkins is a Senior Environmental Health Scientist; and Susan Stone is a Senior Environmental Health Scientist, all with the Health and Environmental Impacts Division.

Richard Wayland

The authors are colleagues in EPA’s Office of Air Quality Planning and Standards in Research Triangle Park, North Carolina. Elizabeth Mannshardt is a Statistician; Kristen Benedict is a Physical Scientist; David Mintz is a Senior Statistician; and Richard Wayland is the Division Director in the Air Quality Assessment Division. Martha Keating is a Policy Advisor; Scott Jenkins is a Senior Environmental Health Scientist; and Susan Stone is a Senior Environmental Health Scientist, all with the Health and Environmental Impacts Division.