Heterogeneous impacts of local traffic congestion on local air pollution within a city: Utilizing taxi trajectory data

This study estimates the causal effects of traffic congestion on air quality in local areas within a city, highlighting the spatial and temporal variation in local traffic, local air pollution, and health risks to residents due to local pollutant exposure. We employ a large-scale taxi driving trajectory data to construct measures of traffic congestion in local areas surrounding air quality monitoring stations in Beijing, on an hourly basis. The station-by-hour panel data enables the estimation of both contemporaneous and lagged marginal effects of local traffic on local air pollution. We find that local traffic congestion increases concentrations of NO2 and CO contemporaneously, but concentrations of PM2.5 and PM10 in a 2-h lagged manner. The magnitude and significance of the estimates vary with local traffic and pollution conditions, implying heterogeneous marginal effects on local air pollution of local traffic across locations, hours of the day, months and seasons. These results stress the importance of designing traffic policies that are both spatially and temporally fine-grained to attain efficiency and equity.