From Concentrations to Context: Approaches for Translating Sensor Network Data into Meaningful Outputs

NAOMI ZIMMERMAN, University of British Columbia

     Abstract Number: 379
     Working Group: Aerosols Spanning Spatial Scales: Measurement Networks to Models and Satellites

Abstract
Increasingly, lower-cost air pollution sensors are being used by academics, government agencies and community organizations to understand local patterns of air quality and the impacts of sources such as traffic and wildfires. However, there is a disconnect between reported concentrations and being able to address/answer questions of interest to users, which include understanding spatiotemporal patterns, air pollution exposure, emission factors, and indoor-outdoor pollutant exchange.

In this talk, I will give an overview of different data treatment and modelling strategies to convert low-cost sensor network data into outputs that are more interpretable and actionable for user groups, with a focus on PM_2.5. This will include an introduction to data treatment methods such as isolating short-lived pollution events from persistent pollution enhancements and integrating wind data into the analysis, and then go on to discuss examples at different spatial scales: (1) building scale for indoor-outdoor exchange, (2) near-road scale for emission factors, (3) neighborhood scale for community advocacy, and (4) city scale for land use regression and exposure estimation. This will draw upon case studies from across Pittsburgh, PA, Vancouver, BC and Uttar Pradesh, India. Lastly, I will briefly touch on strategies for knowledge dissemination of maps or models built with sensor network data, since these networks are used by disparate groups who may benefit from knowledge sharing that goes beyond traditional academic approaches.