AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Evaluation of Low-Cost PurpleAir Monitors and In-Field Correction Using Co-Located Portable Filter Samplers
JESSICA TRYNER, Christian L'Orange, John Mehaffy, Dan Miller-Lionberg, Josephine Hofstetter, John Volckens, Colorado State University
Abstract Number: 353 Working Group: Air Quality Sensors: Low-cost != Low Complexity
Abstract Low-cost aerosol monitors have the potential to provide more spatially- and temporally-resolved data on ambient fine particulate matter (PM2.5) concentrations than are typically available from regulatory monitoring networks; however, low-cost monitors—which do not measure PM2.5 mass directly and tend to be sensitive to variations in particle size and refractive index—sometimes produce inaccurate estimates of PM2.5 concentrations. In this study, we investigated laboratory- and field-based approaches for calibrating low-cost monitors against gravimetric filter samples. First, we investigated the PurpleAir response to NIST Urban PM and derived a laboratory-based gravimetric correction factor. Then, we co-located PurpleAir monitors with compact, portable filter samplers at 15 outdoor sites spanning a 3×3-km area in Fort Collins, Colorado. We evaluated whether PM2.5 correction factors derived from periodic co-locations with portable filter samplers improved the accuracy of the PurpleAir monitors (relative to reference filter samplers operated at 16.7 L·min-1). The atmospheric (ATM) PM2.5 concentrations reported by the PurpleAir monitors appeared to increase linearly with the concentration of NIST Urban PM measured by a TEOM, but neither the factory ATM calibration nor the laboratory-based correction factor translated to the field. Correction factors derived in the field from monthly, weekly, semi-weekly, and concurrent co-locations with portable filter samplers increased the fraction of 72-hour average PurpleAir PM2.5 concentrations that were within 20% of the concentration measured using the reference filter sampler from 15% (for uncorrected measurements) to 45%, 59%, 56%, and 70%, respectively. In addition, 72-hour average PM2.5 concentrations measured using the portable and reference filter samplers agreed (bias ≤ 20% for 71% of samples). These results demonstrate that periodic co-location with portable filter samplers can improve the accuracy of 72-hour average PM2.5 concentrations reported by PurpleAir monitors.