Long-Term Performance of Five Air Sensor Models Across Seven U.S. Sites
Presented by: Karoline Johnson Barkjohn, US EPA
Summary: For any single project, air sensor performance is typically evaluated at a single site or within a single region over a short time period (weeks to months). However, sensor performance often depends on environmental conditions, pollutant concentrations, and particle properties. In addition, performance may change due to seasonal changes in pollutant mixtures or environmental conditions and degradation of the sensor performance. This makes it difficult for potential users to decipher expected performance for their location and conditions. With this in mind, in August 2019 five types of air sensors were set up for a yearlong collocation at seven air monitoring stations across the United States. Sensors were located at stations across climate regions of the U.S. including North Carolina, Georgia, Delaware, Arizona, Colorado, Oklahoma, and Wisconsin. The five sensor models were the Clarity Node from Clarity Movement, PA-II-SD from PurpleAir, AQY1 from Aeroqual, Maxima from Applied Particle Technologies, and the RAMP from SENSIT. The selected sensors measure either particulate matter (PM), or a combination of PM and some gas pollutants. The performance of the PM2.5 and ozone (O3) measurements will be discussed. Sensors were evaluated for accuracy, precision, and the influences of environmental conditions including temperature and relative humidity. Accuracy was evaluated monthly at each site so that comparisons could be made over time and across sites. The results from the first six months of the project provide valuable insights into the performance of air sensors in different climates, including changing seasons. In addition, broader lessons learned about in-field sensor management, data completeness, and data management will be discussed.
Although this abstract was reviewed by EPA and approved for publication, it may not necessarily reflect official Agency policy. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.