AAAR 36th Annual Conference October 16 - October 20, 2017 Raleigh Convention Center Raleigh, North Carolina, USA
Abstract View
Characterizing Intra-Urban Air Pollution Gradients with a Spatially-Distributed Network of Lower Cost Sensors
NAOMI ZIMMERMAN, Zhongju Li, Ellis Shipley Robinson, Aja Ellis, R. Subramanian, Allen Robinson, Joshua Apte, Albert A. Presto, Carnegie Mellon University
Abstract Number: 338 Working Group: Urban Aerosols
Abstract Urban air pollution measurements have typically relied on a small number of widely separated regulatory monitoring sites to assess population-scale exposure. However, air pollutant concentrations may exhibit significant spatial variability depending on local sources and features of the built environment, which may not be well captured by the existing monitoring regime. To better understand urban spatial and temporal pollution gradients on the <1 km scale, a network of 12 air quality monitoring stations was deployed for one year of monitoring beginning July 2016 in Pittsburgh, PA. The stations were deployed both at sites along the urban-rural transect and in downtown urban locations with a range of traffic, restaurant and tall building densities. The stationary monitoring measurements comprise ultrafine particle number (UFP, Aerosol Dynamics “MAGIC” CPC), PM2.5 (Met One Neighborhood PM Monitor), black carbon (Met One BC 1050), and a low-cost air quality monitor, the Real-time Affordable Multi-Pollutant (RAMP) sensor package for measuring CO, NO2, O3, and CO2. Measurements from the monitoring stations were reported at high time resolution (2 min or faster), enabling insight into dynamic pollutant behaviour. RAMP data was calibrated using machine learning-based models to convert sensor response into pollutant concentrations at sensitivities comparable to reference instrumentation. Intra-urban pollutant differences were largely associated with traffic and restaurant density. For example, a site within 30 m of a restaurant plume had on average 3x higher UFP concentrations compared to an urban background site, and 50% higher UFP compared to a high traffic site less than 250 m away. Three sites each within 15 m of a busy roadway also showed elevated NO2 and suppressed O3 concentrations compared to the urban background. Findings from this study combined with land use and population metrics will enable better air pollutant exposure estimates and aid in environmental policy priorities towards improved air quality.