AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Design and Evaluation of a Portable PM Monitor Featuring a Low-Cost Light Scattering Sensor in Line with an Active Filter Sampler
JESSICA TRYNER, Casey Quinn, Bret Windom, John Volckens, Colorado State University
Abstract Number: 615 Working Group: Air Quality Sensors: Low-cost != Low Complexity
Abstract Fine particulate matter (PM2.5) concentrations can vary by orders of magnitude over time and across microenvironments. Light scattering sensors can report real-time PM2.5 measurements from many locations at low cost; however, these sensors have limited accuracy. Gravimetric filter samples provide more accurate, albeit time-integrated, measurements that can be used to correct light scattering sensor data. Correction of light scattering sensor data is key; for example, gravimetric correction factors for nephelometer-derived PM2.5 concentrations varied by a factor of three in two recent studies: one involving personal sampling of adults commuting two and from work in Colorado and one involving stationary sampling in Honduran homes with biomass-fueled cooking stoves.
This work describes a new portable PM2.5 monitor that features a low-cost light scattering sensor in-line with an active filter sampler. Laboratory tests were used to determine (1) the accuracy and precision of PM2.5 concentrations derived from the filter sample and (2) gravimetric correction factors for the response of the low-cost sensor to ammonium sulfate, Arizona road dust, urban PM, and match smoke. Filter samples collected at 0.25 L·min-1 (n = 12; 4 tests with 3 monitors) had a mean bias of -10% (relative to a TEOM). The relative standard deviation of three concurrent samples ranged from 7% to 17%. Correction factors varied from 1.2 to 6.3 depending on the test aerosol and the individual monitor. Gravimetric correction improved the accuracy and precision of 1-hour average concentrations reported by the light scattering sensor. The monitor was also deployed in a week-long field experiment to study sources and concentrations of residential air pollution. Field data were used to identify: (1) pollution events resulting from occupant cooking and heating activities and (2) variations in the number of air changes per hour inside the residence.