Real-time traffic emission inventory development using GeoVideo data and its application to urban street-level air quality modelling

Urban air pollution poses significant risks to public health, necessitating high-resolution monitoring and modelling approaches. This study developed a GeoVideo-based intelligent sensing system for real-time monitoring of vehicle activity levels on urban streets, capturing traffic parameters such as flow, speed, and vehicle type. Applied in Kaifeng, China, the system provided essential input data for street-scale air pollutant modelling. Using the collected data, this study created a high spatiotemporal resolution traffic emission inventory for Kaifeng’s street network in 2018, estimating annual emissions of approximately 27,839 tons of CO, 2,845 tons of HC, 1,273 tons of NOx, 23.56 tons of PM2.5, and 25.18 tons of PM10. This emission inventory was integrated with regional (WRF-Chem) and street-level (MUNICH) air quality models to conduct fine-scale spatiotemporal simulations of NO2 and O3 concentrations. The simulation results showed strong agreement with observed data (R = 0.98 for O3 and R = 0.89 for NO2), demonstrating the effectiveness of our approach in capturing fine-scale variations of street-level air pollutants. The simulations revealed significant spatiotemporal variations in pollutant concentrations, with NO2 pollution concentrated on peripheral streets peaking around 10:00 AM, and O3 pollution on secondary and minor roads peaking between 3:00 PM and 6:00 PM. This study presents a novel methodology for fine-scale urban air quality modelling, providing valuable insights for intelligent urban management and real-time regulation, and contributing to effective air quality management strategies.