Building Drag Parameterization Including Anisotropy and Its Coupling with the Weather Research and Forecasting Model

Buildings increase the urban surface roughness and reduce near-surface wind speeds due to the drag effect, which depends on the flow direction. In this study, a building drag parameterization scheme including the building anisotropy for all flow directions was developed through approximating buildings with elliptical columns to represent anisotropic frontal area index. The new scheme was coupled with the Weather Research and Forecasting (WRF) model to improve urban simulations in those including near-surface wind speeds. The conducted offline sensitivity tests with the developed scheme, using horizontal wind along different directions, show continuous transitions of drag coefficient and other variables depending on flow direction. The maximum difference of drag coefficient between the new and the original scheme reached 10%–20% of that from the original one. These monthly simulations of the WRF model with the new building drag scheme for Chengdu were conducted to validate the updated model against station observation and reanalysis data. Compared to the original scheme, the updated scheme reduces overestimation of 10-m wind speed by 0.1–0.2 m s⁻¹ (5%–15% of the original bias), overestimation of 2-m temperature by 0.1°C–0.4°C (20%–60%), and underestimation of 2-m relative humidity by 1%–3% (20%–60%). This is achieved by increasing the drag coefficient through an enhanced frontal area index and reducing wind speed. The diminished wind speed reduces sensible heat flux, enhances latent heat flux, and suppresses vertical motions, resulting in humidity accumulation and cooling in the lower atmosphere. These suggest that reasonable representation of the building anisotropy is important in researching urban climate.