Embedded large eddy simulation of typhoon wind field and its effects in a large-scale complex urban area with field validations

Accurate simulation of wind fields in complex urban areas with dense buildings and surrounding topography necessitates sufficiently large computational domains; however, conventional Large Eddy Simulation (LES) remains prohibitively expensive. To address this challenge, this study proposes a numerical simulation framework based on the Embedded LES (ELES) model, which optimizes mesh allocation to significantly reduce computational costs while maintaining high accuracy. Compared to previous ELES-based urban wind simulations, two key innovations are introduced. First, the ELES model is applied, for the first time, to a large-scale real-world urban area with complex terrain. The simulation employs a 14 000 × 14 000 m2 computational domain, where buildings and terrain in the cover region are explicitly represented using body-fitted meshes, and source terms are incorporated into the momentum equations to account for velocity loss and disturbance caused by airflow passing through tree-covered areas. The simulation remained stable throughout, demonstrating the robustness of the proposed ELES framework for a more realistic representation of the urban wind environment. Second, the framework's performance is quantitatively validated against both field measurements and conventional LES results for surface pressure on a 599 m high skyscraper within the urban area under real typhoon conditions. The comparison confirms the model's ability to accurately reproduce time-averaged values, root mean square values, and probability density distributions of the wind pressures on the building cladding. These findings underscore the framework's potential for accurately simulating the urban wind field and its effects under extreme wind conditions, offering significant promise for urban wind environment assessment and wind-induced disaster mitigation.