Downscaled simulation approach for near-field nuclear facility pollutant atmospheric transport and dispersion using a...

Nuclear facilities can release radioactive pollutants into the atmosphere during daily operations and accidents. These pollutants are transported and dispersed through atmospheric processes, eventually depositing on the surface and posing potential risks to ecosystems and public health. Accurately describing their atmospheric dispersion is therefore essential for scientific assessments of radioactive impact. In this study, a one-way downscaling model framework of atmospheric dispersion of pollutants was developed, which coupled the mesoscale Weather Research and Forecasting (WRF) model with a microscale computational fluid dynamics (CFD) model. To enhance the microscale model and make it inherit the flow field characteristics of the mesoscale model, closure constants of the microscale turbulence model were adjusted according to the planetary boundary layer scheme in the mesoscale model. The simulation performance of the WRF-CFD downscaling model framework and its adjusted model was evaluated by meteorological observations and an SF₆ tracer experiment conducted in Fuqing Nuclear Power Plant. The results demonstrated that WRF reasonably reproduced the mesoscale wind field and provided reliable inflow conditions for the CFD simulations. The coupled WRF-CFD downscaling model had high accuracy and reliability in reproducing wind profiles and concentration field distributions, while the adjustment strategy for turbulence model closure constants could effectively enhance the simulation performance of the downscaling model. Although wind shear-induced plume swing introduced spatial deviations between the simulation results based on steady RANS and the measured values in far-field regions, the model still captured peak concentration magnitudes comparable to measurements. Overall, the proposed downscaling strategy enabled reliable high-resolution simulations of pollutant transport and dispersion in the near-field atmosphere of nuclear facilities. The turbulence constant adjustment improved model performance, while limitations remain in steady-state RANS for resolving plume swing. Future work is needed to explore transient approaches to better capture shear-induced plume variability. This study provided a new approach for the detailed investigation of pollutant transport and dispersion in the real atmospheric environment, as well as a practical and scientifically robust tool for environmental impact assessment and emergency preparedness in nuclear facility management.