Multiscale Analysis of Air Quality in New York City

Dense urban areas experience poor air quality due to increased anthropogenic emissions and complex urban morphology that restricts ventilation. While our general understanding of urban air pollution has improved considerably through advances in numerical modeling and sensor platforms, our insights into neighborhood-to-building-scale pollution remain insufficient. Most current forecasting models do not account for urban climate processes, and traditional in-situ observations do not capture street-level variability in primary pollutants. Here, we investigate the street- and neighborhood-scale dynamics of air pollution in New York City, using both mobile and in situ observations. We used a backpack fitted with research-grade instruments to monitor particulate matter (PM2.5) and ozone. In-situ observations from multiple public air quality networks were also included in our analysis. Our results show a high degree of uniformity in street-level ozone concentrations in NYC, whereas particulate matter concentrations varied significantly. On days impacted by synoptic disturbances, both ozone and particulate matter concentrations were nearly uniform throughout the city. The fixed ground stations adequately captured the median PM2.5 concentration. However, they missed the extremes, which were, in some cases, two to five times the median value. The observations were also used to validate an urbanized WRF-Chem model and satellite-derived measurements. The numerical simulations conducted at 4km X 4km resolution performed better than the current forecast model in predicting both PM2.5 and ozone concentration. The model accounted for the impacts of the urban heat island effect and local sea breeze flows on air pollutants. The model particularly captured the ozone dynamics accurately.