Improved parameterisation for the numerical modelling of air pollution within an urban street canyon

Efisio Solazzo, Xiaoming Cai, Sotiris Vardoulakis

Research output: Contribution to journalArticlepeer-review

46 Citations (Scopus)

Abstract

Numerical modelling for application to wind flow and dispersion in urban environments has noticeably progressed in recent years, to currently represent a widely used tool for simulating mechanical processes governing air pollution in complex geometries. In particular, Computational Fluid Dynamic (CFD) techniques based on RANS (Reynolds-Averaged Navier-Stokes equations) models, are extensively used to produce detailed simulations of the wind flow and turbulence in the urban canopy. However, several studies have indicated that RANS models, and in particular the widely used standard k-ε turbulence model, are sensitive to the particular form of inlet profiles for turbulence and velocity. In the present study, simulations of the wind flow and dispersion within an idealised street canyon were carried out using the standard k-ε turbulence model provided by the commercial software FLUENT. The aim of this study was to improve the standard k-ε model performance by modifying the model parameters according to the chosen form of inlet profiles for velocity and turbulence. Capability of the model to reproduce real wind flow fields, turbulence and concentration patterns was evaluated by comparing the model results against recently published wind tunnel data. Results for turbulent kinetic energy and concentration showed that the redefinition of the default dispersive parameters can significantly enhance the model performance. The newly proposed parameterisations of the standard k-ε turbulence model can be readily implemented within commercial CFD software packages, offering a reliable modelling tool for application to urban air pollution and other environmental studies.

Original languageEnglish
Pages (from-to)381-388
Number of pages8
JournalEnvironmental Modelling and Software
Volume24
Issue number3
DOIs
Publication statusPublished - Mar 2009
Externally publishedYes

Fingerprint

Dive into the research topics of 'Improved parameterisation for the numerical modelling of air pollution within an urban street canyon'. Together they form a unique fingerprint.

Cite this