AAAR 32nd Annual Conference
September 30 - October 4, 2013
Oregon Convention Center
Portland, Oregon, USA
Abstract View
A Novel Method for Reliable Long-term Assessment of Exposure to Traffic-related Air Pollution Mixtures
NATALIA MYKHAYLOVA, Kelly Sabaliauskas, Jon M Wang, Ezzat Jaroudi, Cheol-Heon Jeong, Jeff Brook, Greg J. Evans, SOCAAR, University of Toronto
Abstract Number: 409 Working Group: Aerosol Exposure
Abstract Exposure to air pollutant mixtures at high temporal and spatial resolution is an important predictor for many respiratory and cardiovascular conditions and its accurate assessment remains the holy grail of air quality monitoring. However, attaining high temporal resolution data of multiple pollutants, while maintaining low cost and reliable long-term performance, remains a challenge.
To address this issue, we conducted a thorough investigation of 30 different low-cost commercial gas and particulate matter sensors from 5 manufacturers. The types compared included n-type MOS, p-type MOS, optical, and electrochemical sensors. We identified the best-performing sensors through controlled lab calibrations and long-term ambient inter-comparisons between sensors and standard real-time measuring techniques. Best-performing sensors for detection of nitrogen dioxide, carbon monoxide, ozone, volatile organic compounds, fine particulate matter and coarse particulate matter, were selected, further optimized and integrated into a portable sensor array system. Temperature and humidity sensors were included for data normalization. We evaluated the performance of sensors in field during two 2-week deployments in winter and summer, using a network of high time-resolution sensor array devices deployed at 5 different locations within a 30km radius of downtown Toronto. These deployments allowed long-term sensor performance to be evaluated under different meteorological conditions as well as different ranges of pollutant concentrations.
The portable sensor array system was shown to be sensitive enough to measure ambient levels of pollutants even in Toronto’s relatively clean air. Specifically the system could detect concentrations as low as 5ppm level for total volatile organic compounds, 20 micrograms per meter cubed for fine particulate matter, 5ppb for ozone, 15ppb for nitrogen dioxide and 400ppb for carbon monoxide. The pollutant levels showed large variation both temporally and spatially, indicating the importance of high resolution monitoring. Methods for evaluating measurement accuracy and precision as well as reducing measurement biases will be discussed.