Modelling nitrogen dioxide dispersion in urban street canyons through sensor-based emission assessment.

This study assesses the performance of the ADMS-Urban dispersion model in estimating 1-h mean nitrogen dioxide (NO) concentrations within the street canyons of Prague. While traditional air quality modeling that relies on sparse data from localized monitoring stations, this approach pioneers the integration of traffic, background, and rooftop sensor network, to archive a more granular validation of model outputs. The results demonstrate robust model performance, with FAC2 values ranging from 0.307 to 0.552, NMSE consistently below 1.8, and fractional bias (FB) largely within ±0.4. Notably, the rooftop sensors exhibited a higher FB of 0.775, suggesting reduced accuracy in zones with lower ambient concentrations and complex vertical mixing. Spatial mapping revealed that NO concentrations exceeded 50 μg/m at key traffic intersections, driven by vehicle idling, congestion, and restricted dispersion within enclosed canyon geometries. Street morphology played a critical role in pollutant retention, with narrow, high-walled corridors accumulating higher NO loads than broader, vegetated streets that allowed greater airflow and dilution. Temporal analysis indicated a marked seasonal trend, with elevated concentrations during the colder months (October to February), reflecting increased vehicular activity and meteorological influences such as lower wind speeds and temperature inversions. Overall, the findings confirm the ADMS-Urban model's capacity to accurately reflect both spatial and temporal variability in urban NO distribution. The results underscore the interplay between urban form, traffic dynamics, and pollutant dispersion, offering valuable insights for air quality planning in similarly dense urban environments.