Investigating the impacts of driving restriction on NO_2 concentration by integrating citywide scale cellular data and traffic simulation

摘要

Urban traffic is one of the main sources of NO2, and driving restriction (DR) has been widely used to alleviate air pollution in China, which means that private vehicles are not allowed to enter or pass a specific area according to their last digit of license plate numbers. This study investigates the impacts of DR on NO2 concentration at the traffic analysis zone (TAZ) level, which is particularly defined for traffic-related spatial data analysis. Measured NO2 data, citywide scale cellular data, and traffic-related data were collected for analysis. Inverse distance weighting (IDW) model was used to obtain NO2 at each TAZ. The traffic model of Nanjing was built in TranStar, a mesoscopic traffic simulation platform. Based on TAZ level NO2, traffic demand, and traffic status data, a geographically weighted regression (GWR) model was developed. DR with different proportions and spatial scales were simulated in TranStar and evaluated based on the predictions of the GWR model. Results suggest that morning rush hour is the only time in the day when urban traffic is the main cause of NO2's rise. Larger traffic volume and severer congestion with lower speed lead to the higher rise of NO2. The main factors influencing the rise of NO2 could be different at distinct locations. DR strategies can decrease the rise of NO2 significantly for most TAZs. Restriction proportion has significantly higher impacts than spatial scale, and the effects of scale enlarge with the rise of proportion. A side effect of DR is that the rise of NO2 would be higher in certain regions due to travelers' mode shift and detour, which is a key point for policymakers to weigh the pros and cons.