A Numerical study on flow and dispersion in an urban area using a CFD/WRF coupled model

Abstract

This study, examined the effects of highly developed residential buildings on wind flow and pollutant dispersion in a built-up area of Seoul, South Korea. Predicting realistic flows in an urban area, it is very important to reflect variations of a mesoscale weather field. For this, a mesoscale numerical weather prediction model, the Weather Research and Forecasting (WRF) model is employed and coupled with a computational fluid dynamics (CFD) model. The CFD model is based on a Reynolds-averaged Navier-Stokes equations model with the renormalization group (RNG) turbulence closure scheme. output data sets are produced by the WRF model which can simulate mesoscale weather well are used as the initial input data sets in the CFD model. The CFD model is numerically integrated using the data sets which are properly interpolated in time and space. The WRF model is simulated for 30 hours starting from 18 UTC 6 August 2009 using NCEP final analysis data as initial and boundary conditions. During the period, a high-pressure system was dominant, clear condition is maintained, and synoptic winds are weak. Four nested computational domains are considered and grid sizes of the innermost domain are 1km in the horizontal. The coupled CFD-WRF model is integrated for 24 hours. Sea level pressures are well reproduced during the simulation period by WRF model compared with the analysis charts. The results showed the importance of the coupled CFD-WRF model in predicting the wind speed and direction in urban areas. The CFD-WRF model predicted them better than those by the WRF model only. It is because the CFD-WRF model considered adjacent buildings and more realistic topographies. The results also showed that tall buildings in the urban area affected the flow and pollutant dispersion much in the domain. The pollutant concentration inside the area bounded by the tall buildings is low due to the partial increase in the wind speed (so called, the channeling effect) even though the area is near from the pollutant sources.