COMPUTATIONAL MODELING OF ENVIRONMENTAL AND ATMOSPHERIC FLOWS: AIR POLLUTION DISPERSION, URBAN WIND DYNAMICS, AND CLIMATE APPLICATIONS

Abstract

Computational modeling plays a critical role in understanding environmental and atmospheric flows in complex urban environments, where interactions between wind dynamics, turbulence, pollutant dispersion, and climate-driven stability effects govern air quality and human exposure. This study presents a comprehensive numerical framework for modeling urban atmospheric flows by integrating geometry-adaptive turbulence closure, urban canopy drags, and stability-aware pollutant diffusion within a Reynolds-Averaged Navier–Stokes (RANS) formulation. The model explicitly accounts for urban morphological characteristics and atmospheric stratification, enabling improved representation of momentum exchange, turbulence production, and scalar transport. Numerical simulations are conducted for idealized urban configurations under unstable, neutral, and stable atmospheric conditions. The results demonstrate enhanced prediction of vertical wind profiles, turbulent kinetic energy, eddy viscosity, and pollutant concentration compared to conventional RANS models. Ground-level pollutant concentrations are shown to increase by up to 30–40% under stable stratification, while unstable conditions promote enhanced vertical mixing and ventilation. Validation against representative benchmarks indicates a reduction in prediction error of approximately 30–35% relative to traditional approaches. The proposed framework provides a scalable and climate-aware tool for urban air quality assessment and offers a pathway for integration into mesoscale and climate modeling systems.