Evaluating the Meteorological Effects on the Urban Form-Air Quality Relationship Using Mobile Monitoring

Predictive models based on mobile measurementshave been increasingly used to understand the spatiotemporalvariations of intraurban air quality. However, the effects ofmeteorological factors, which significantly affect the dispersion ofair pollution, on the urban-form-air-quality relationship have notbeen understood on a granular level. We attempt tofill this gap bydeveloping predictive models of particulate matter (PM) in theBronx (New York City) using meteorological and urban formparameters. The granular PM data was collected by mobile low-cost sensors as the ground truth. To evaluate the effects ofmeteorological factors, we compared the performance of modelsusing the urban form withinfixed and wind-sensitive buffers,respectively. Wefind better predictive power in the wind-sensitive group (R= 0.85) for NC10(number concentration for particleswith diameters of 1 mu m-10 mu m) than the control group (R= 0.01), and modest improvements for PM2.5(R= 0.84 for the windsensitive group,R= 0.77 for the control group), indicating that incorporating meteorological factors improved the predictive powerof our models. We also found that urban form factors account for 62.95% of feature importance for NC10and 14.90% for PM2.5(9.99% and 4.91% for 3-D and 2-D urban form factors, respectively) in our Random Forest models. It suggests the importance ofincorporating urban form factors, especially for the uncommonly used 3-D characteristics, in estimating intraurban PM. Our methodcan be applied in other cities to better capture the influence of urban context on PM levels