Development of an In-Home Air Pollutant Sensor Platform and Implications for Community Use: Framework for Selection and Quality Control
Invited conference paper presented and published in conference proceedings

CUHK Authors
Author(s) no longer affiliated with CUHK


Full Text

Other information
AbstractCharacterization of environmental exposures has been shaped by logistical, technological and financial constraints resulting in passive or integrated campaigns that do not allow fine-scale measurement of short-term events relevant to health. Recent advances allow for continuous monitoring of pollutants using comparatively small, inexpensive, real-time sensors. Such sensors have been widely used in occupational settings, but transitions to use in lower concentration environments present quality control challenges. For the Center for Research on Environmental and Social Stressors in Housing across the Life Course (CRESSH), we developed a multi-pollutant air sampling platform–environmental multi-pollutant monitoring assembly (EMMA)–for in-home measurements aiming to develop innovative methods for improved estimates in between-household variability in environmental exposures. Of one- and five-minute data collected using EMMA–carbon monoxide (CO), carbon dioxide (CO2), nitrogen oxides (NOx), particulate matter (PM2.5), temperature, relative humidity and noise–the following issues were identified and accounted for: inter-sensor variability, temperature effects and performance degradation. Gas sensors show notable initial performance (mean R2=0.99±0.005), but require constant performance scrutiny. Over 18 months, average CO sensor sensitivity decreased by 21% and NO & NO2 by 36 & 30%, respectively, coinciding with differential signal at like concentrations over time. Modeling results from lab-based controlled experiments show NO & NO2 data significantly biased at temperatures ≧93 & 90°F and below 40 & 50°F, respectively. Despite such issues, we achieved 97% data completeness. These findings may inform citizen science applications where various resources (e.g., calibration equipment, computer software) ensuring interpretable data may not be available. Our process of validating and maintaining EMMA may provide guidance for others seeking to develop and deploy such platforms.
All Author(s) ListSara Gillooly, Jose Vallarino, Yulun Zhou, MyDzung T Chu, Gary Adamkiewicz
Name of ConferenceISES-ISEE 2018 Joint Annual Meeting
Start Date of Conference26/08/2018
End Date of Conference30/08/2018
Place of ConferenceOttawa
Country/Region of ConferenceCanada
Year2018
LanguagesEnglish-United States

Last updated on 2020-06-05 at 15:56