Modeling Roadside NO₂ Concentrations Using CMAQ-CFD Model with Dynamic Spatio-Temporal Boundary Conditions

 Pollutant concentrations near roadways exhibit significant variability due to vehicle emissions and street canyon structures. Computational Fluid Dynamics (CFD) models are effective in simulating microscale pollutant dispersion. However, numerical models are prone to errors caused by uncertainties. Some recent studies have improved CFD model performances by using larger-scale model data, such as Community Multi-scale Air Quality (CMAQ), as initial and boundary conditions and calibrating model results based on observational data. This study aims to evaluate urban air quality modeling performance by applying mesoscale model results to generate initial and boundary conditions that reflect spatiotemporal variability and by applying them to a CFD model. The study period was from January 7 to 8, 2021, focusing on a 9 km × 9 km area centered in Seocho-gu, Seoul, Republic of Korea. The data used includes NO₂ concentrations from Seoul's roadside air quality monitoring stations, as well as air temperature, wind direction, and wind speed data. The models used in this study were the CMAQ-CFD models, and the initial and boundary conditions were established using 9-km grid data obtained from the Weather Research and Forecasting (WRF) model and CMAQ model. This study corrected temporal deviations between simulation and measurement values for every hour and incorporated spatiotemporal variability in the initial and boundary conditions by considering land-use types at corresponding grids. Across all stations, the method that accounted for spatiotemporal variability in both initial and boundary conditions outperformed CMAQ in terms of R, RMSE, and IOA. In conclusion, we suggest that applying CMAQ model results to CFD models as initial and boundary conditions can improve roadside air quality simulations. This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (RS-2024-00356913).