Evaluating Fine-Scale Air-Quality Heterogeneity Using a Low-Cost Multipollutant Sensor Network in Twin Cities, Minnesota

Abstract

There is growing interest and effort in deploying low-cost sensors (LCSs) to understand fine-scale spatiotemporal heterogeneity of ambient air pollutants in urban areas. The Assessing Urban Air Quality project conducted by the Minnesota Pollution Control Agency deployed 45 low-cost air-quality monitors (AQMesh, Environmental Instruments Ltd.) in the Minneapolis/St. Paul metropolitan area network. Hourly averaged pollutant concentrations or mixing ratios for carbon monoxide (CO), ozone (O3), nitrogen dioxide (NO2), nitric oxide (NO), and sulfur dioxide (SO2), as well as particulate matter (PM), including PM1, PM2.5, and PM10, were obtained. The network data from June 2019 to June 2021 were analyzed for temporal air-quality variations related to traffic patterns and atmospheric chemistry. Compared to the PM concentrations, gaseous pollutant (CO, NO2, O3, and SO2) mixing ratios among the different sites showed higher correlations (higher pairwise Pearson r2 values). However, pairwise coefficients of divergence (COD) show higher spatial heterogeneity for O3 and NO2. The conditional bivariate probability function coupling the sensor and meteorological data can partly explain the fine-scale heterogeneity. There was little to no correlation between the pollutant concentrations/mixing ratios and census-based income or site groups in different redlined areas. This study demonstrates the unitality of a multiyear, dense network of LCS to provide high-resolution spatiotemporal data with desirable performance.