Fast network-based modelling of dispersion in city centres

S..E. Belcher1, O. Coceal1,2, E. V. Goulart3, A.C. Rudd4, A.G. Robins5

1Department of Meteorology, University of Reading
2NCAS Atmospheric Physics
3Federal University of Espirito Santo, Vitoria, Brazil
4Centre for Ecology and Hydrology, Wallingford
5Faculty of Engineering and Physical Sciences, University of Surrey

pdf Download this article as a PDF (297 KB)

logo nerclogo epsrc


Dr Omduth Coceal is an NCAS-Atmospheric Physics Research Scientist with expertise in turbulence and dispersion modelling in the atmospheric boundary layer.

What are the new findings?

Predicting dispersion of pollutants in urban areas is challenging due to the chaotic nature of the turbulent flow within a complex urban environment. We present a methodology that simplifies the problem considerably. A process-based model based on a network approach gives insight into the roles of key dominant processes in controlling dispersion around buildings in a city centre. Analytical solutions show that the concentration pattern from localised releases is determined by only three effective parameters, which depend on the building geometry and the turbulent flow properties.

Why are these findings important?

Fast modelling of accidental or terrorist releases of toxic air-borne material is critical for first responders. This requires models that can provide concentration estimates quickly and reliably, a challenging combination in view of the complexity of turbulent dispersion in urban environments. The insights and modelling approach developed here simplify the problem considerably and enable good first estimates to be made. The method could also be used in inverse modelling to find the location and strength of unknown releases in urban areas.

How did we discover this?

A conceptual framework was developed based upon empirical results obtained by the analysis of comprehensive data generated by supercomputer simulations performed previously by the authors. Then a mathematical model was formulated by treating an urban area as a street network and the governing equations derived and then solved analytically. The resulting model compared well with high-resolution numerical data and wind-tunnel experiments.


Above: Dispersion of localised releases among buildings is a complex stochastic process. But simple models give valuable insight and fast predictive capability.

Find out more:

  • Read more on the project website
  • See also Omduth's homepage
  • Email: This email address is being protected from spambots. You need JavaScript enabled to view it.
  • Take a look at the journal article here

Tell us what you think

  • How clearly was this article written?
  • How interesting or useful was it?
  • Do you have any other comments?

Please let us know:

This email address is being protected from spambots. You need JavaScript enabled to view it.

This research was funded by NCAS/NERC and EPSRC.