AAAR 34th Annual Conference
October 12 - October 16, 2015
Hyatt Regency
Minneapolis, Minnesota, USA
Abstract View
Assessing the Quantitative Potential of Distributed Low-cost Air Quality Sensor Networks
EBEN CROSS, David Hagan, David Ogutu, Jonathan Franklin, Gary Adamkiewicz, Ann Backus, Jose Vallarino, Douglas Worsnop, John Jayne, Colette Heald, Jesse Kroll, MIT
Abstract Number: 487 Working Group: Urban Aerosols
Abstract Because of the large expense and expertise required to set up and maintain air quality (AQ) monitoring stations, our understanding of communities’ exposures to air pollutants is generally based upon an extremely limited number of measurements. Such measurements (typically no more than 1-4 monitoring stations per urban area, reporting concentrations on hourly or daily basis) do not capture the enormous temporal and spatial variability of air pollutants across densely populated areas. This greatly limits our ability to estimate spatiotemporal variation in air quality in community-based health studies and inform the public about local sources and levels of air pollution. In this presentation, we will describe results from two low-cost air quality sensor networks: [1] CLAIRITY, a 25-node network designed, built, and deployed across the MIT campus and [2] the Dorchester Air Quality Sensor System (DAQSS) a 5-node network distributed in the south Boston neighborhood of Dorchester. Each network provides an opportunity to assess the quantitative potential of electrochemical sensors (Alphasense model B4) for measurement of criteria gas pollutants (carbon monoxide, nitric oxide, nitrogen dioxide, and ozone), as well as low-cost (~$200-$400/unit) Optical Particle Counters (Dylos DC1100; Alphasense OPC-N2) for measurement of particulate matter. Presented work will highlight the need for laboratory calibration and/or in-field co-location with research grade instrumentation to ensure robust, quantitative, low-cost air pollution measurement outputs. Characterizing sensor response across a range of ambient temperature, relative humidity, and interfering gas concentrations is especially important. Ultimately, low-cost AQ sensor networks with improved spatiotemporal resolution can complement existing regulatory monitoring networks - enabling community members to exercise data-driven decisions that minimize their exposure to harmful air pollution, and researchers to better understand pollutant emissions, transport, and chemistry.