Characterizing Intra-Urban Air Quality Gradients with a Spatially-Distributed Network

City-wide air pollution measurements have typically relied on regulatory or research monitoring sites with low spatial density to assess population-scale exposure. However, air pollutant concentrations 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 at 1 km resolution, a network of 12 real-time air quality monitoring stations was deployed beginning July 2016 in Pittsburgh, PA. The stations were deployed at sites along an urban-rural transect and in urban locations with a range of traffic, restaurant, and tall building densities to examine the impact of various modifiable factors. Measurements from the stationary monitoring stations were further supported by mobile monitoring, which provided higher spatial resolution pollutant measurements on nearby roadways and enabled routine calibration checks. The stationary monitoring measurements comprise ultrafine particle number (Aerosol Dynamics "MAGIC" CPC), PM2.5 (Met One Neighborhood PM Monitor), black carbon (Met One BC 1050), and a new low-cost air quality monitor, the Real-time Affordable Multi-Pollutant (RAMP) sensor package for measuring CO, NO2, SO2, O3, CO2, temperature and relative humidity. High time-resolution (sub-minute) measurements across the distributed monitoring network enable insight into dynamic pollutant behaviour. Our preliminary findings show that our instruments are sensitive to PM2.5 gradients exceeding 2 micro-grams per cubic meter and ultrafine particle gradients exceeding 1000 particles per cubic centimeter. Additionally, we have developed rigorous calibration protocols to characterize the RAMP sensor response and drift, as well as multiple linear regression models to convert sensor response into pollutant concentrations that are comparable to reference instrumentation.