AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Integrating Multi-source (Satellite Retrieval, Model Simulation, Ground Based Monitoring) and Low-cost Sensor Particulate Mass Data to Improve Spatio-temporal Air Quality Mapping
CARL MALINGS, Matthias Beekmann, Daniel Westervelt, Albert A. Presto, R. Subramanian, LISA
Abstract Number: 261 Working Group: Air Quality Sensors: Low-cost != Low Complexity
Abstract The availability of low-cost particulate mass sensors is creating an opportunity for community groups, citizen scientists, and governments to assess local air quality at unprecedented spatial and temporal resolutions. More traditional high-cost, high-accuracy monitoring sites, satellite data retrievals, and chemical transport models provide additional sources of air quality data, but unfortunately no source is comprehensive on its own. Traditional monitoring networks have mainly been established for developed nations. Low-cost sensors have the potential to supplement and extend these networks worldwide but require in-field calibrations to quantify their uncertainties. Satellites provide only “snapshots” of air quality. They also require “ground-truthing” against ground-based monitors, as do computationally-intensive simulation models. Together, however, these sources of information have the potential to cover each other’s weaknesses, providing a better overall spatial and temporal picture of air quality than is available from each source independently. We demonstrate such a multi-source data integration using two case studies. In Pittsburgh, Pennsylvania, a traditional monitoring network allows for ground-truthing of satellite-based retrievals and model assimilations available from NASA's Global Modeling and Assimilation Office. Combining this with a dense network of low-cost sensors further allows for spatial and temporal downscaling. In Kigali, Rwanda, and Kinshasa, DRC, traditional ground-based monitoring is lacking and must be substituted with low-cost sensor data. For example, a year of such data in Kigali suggests an average PM2.5 concentration of 53 µg/m3, while the latest population-weighted average for Rwanda based on multiple data sources (Shaddick et al. 2018, available at www.stateofglobalair.org) is 43 µg/m3. Through this case study, we assess the ability of regional-scale satellite-based retrievals and local-scale low-cost sensor measurements to correct each other and yield a more accurate overall picture of air quality. Overall, we assess the potential for a multi-source approach to improve error quantification and expand spatio-temporal air quality data coverage.