Laboratory and field evaluation of real-time and near real-time PM2.5 smoke monitors

ABSTRACT Increases in large wildfire frequency and intensity and a longer fire season in the western United States are resulting in a significant increase in air pollution, including concentrations of PM2.5 (particulate matter <2.5 µm in aerodynamic diameter) that pose significant health risks to nearby communities. During wildfires, government agencies monitor PM2.5 mass concentrations providing information and actions needed to protect affected communities; this requires continuously measuring instruments. This study assessed the performance of seven candidate instruments: (1) Met One Environmental beta attenuation monitor (EBAM), (2) Met One ES model 642 (ES642), (3) Grimm Environmental Dust Monitor 164 (EDM), (4) Thermo ADR 1500 (ADR), (5) TSI DRX model 8543 (DRX), (6) Dylos 1700 (Dylos), and (7) Purple Air II (PA-II) in comparison with a BAM 1020 (BAM) reference instrument. With the exception of the EBAM, all candidates use light scattering to determine PM2.5 mass concentrations. Our comparison study included environmental chamber and field components, with two of each candidate instrument operating next to the reference instrument. The chamber component involved 6 days of comparisons for biomass combustion emissions. The field component involved operating all instruments in an air monitoring station for 39.5 days with hourly average relative humidity (RH) ranging from 19% to 98%. Goals were to assess instrument precision and accuracy and effects of RH, elemental carbon (EC), and organic carbon (OC) concentrations. All replicate candidate instruments showed high hourly correlations (R2 ≥ 0.80) and higher daily average correlations (R2 ≥ 0.90), where all instruments correlated well (R2 ≥ 0.80) with the reference. The DRX and Purple Air overestimated PM2.5 mass concentrations by a factor of ~two. Differences between candidates and reference were more pronounced at higher PM2.5 concentrations. All optical instruments were affected by high RH and by the EC/OC ratio. Equations to convert candidate instruments data to FEM BAM type data are provided to enhance the usability of data from candidate instruments. Implications: This study tested the performance of seven candidate PM2.5 mass concentration measuring instruments in two settings - environmental chamber and field. The instruments were tested to determine their suitability for use during biomass combustion events and the effects of RH, PM mass concentrations, and concentrations of EC and OC on their performance. The accuracy and precision of each monitor and effect of RH, PM concentration, EC and OC concentrations are varied. The data show that most of these candidate instruments are suitable for measuring PM2.5 concentration during biomass combustions with a proper correction factor for each instrument type.