Hindawi Publishing Corporation
Te Scientifc World Journal
Volume 2013, Article ID 781748, 11 pages
http://dx.doi.org/10.1155/2013/781748
Research Article
Microscale Obstacle Resolving Air Quality Model Evaluation
with the Michelstadt Case
Anikó Rakai and Gergely Kristóf
Department of Fluid Mechanics, Budapest University of Technology and Economics, Budapest 1111, Hungary
Correspondence should be addressed to Anik´ o Rakai; rakai@ara.bme.hu
Received 30 April 2013; Accepted 27 June 2013
Academic Editors: J. Corte-Real, M. A. Gal´ an, and J. Koch
Copyright © 2013 A. Rakai and G. Krist´ of. Tis is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
Modelling pollutant dispersion in cities is challenging for air quality models as the urban obstacles have an important efect on the
fow feld and thus the dispersion. Computational Fluid Dynamics (CFD) models with an additional scalar dispersion transport
equation are a possible way to resolve the fowfeld in the urban canopy and model dispersion taking into consideration the efect
of the buildings explicitly. Tese models need detailed evaluation with the method of verifcation and validation to gain confdence
in their reliability and use them as a regulatory purpose tool in complex urban geometries. Tis paper shows the performance of
an open source general purpose CFD code, OpenFOAM for a complex urban geometry, Michelstadt, which has both fow feld
and dispersion measurement data. Continuous release dispersion results are discussed to show the strengths and weaknesses of the
modelling approach, focusing on the value of the turbulent Schmidt number, which was found to give best statistical metric results
with a value of 0.7.
1. Introduction
Prognostic microscale obstacle resolving meteorological
models and Computational Wind Engineering (CWE) mod-
els deal with the common felds of wind and pollutant
dispersion modelling inside the urban canopy. Baklanov
and Nuterman [1] show that these models with increasing
computational capacity can be the fnal scale in a nested
multiscale meteorological and dispersion model. In this
paper the intersection of the two described disciplines is
investigated with the assumptions of quasistationary fow
and neutral meteorological conditions and the pollutant is
considered a passive scalar.
Stull [2] defnes microscale in meteorology as a few kilo-
meters or less where the typical phenomena include mechan-
ical turbulence caused by the buildings. Britter and Hanna
[3] suggest the following lengthscales: regional (up to 100 or
200 km), city scale (up to 10 or 20 km), neighborhood scale
(up to 1 or 2 km), and street scale (less than 100 to 200 m).
Te last two correspond to the microscale defnition of Stull
and are used here.
Te general approach in CWE for air pollution dispersion
modelling is adding a passive scalar transport equation
decoupled from the solution of the fowfeld or modelling
Lagrangian particle dispersion in the computed fowfeld.
Although this is essentially less efort than computing the
fowfeld itself, if there are errors already in the mean velocity
feld, for example, reattachment length overestimation, or in
the turbulent felds, for example, stagnation point anomaly,
those errors are propagated in the pollutant transport mod-
elling. Especially the turbulent felds which are not always
vital for the fowfeld are equally important in the dispersion
model as they are responsible for the turbulent difusion.
A comparison of the Eularian additional transport equa-
tion approach and the Lagrangian approach in a single
building confguration is given in Gorl´ e et al.’s [4]. Tey
fnd no great diference between the two approaches, but the
additional transport equation has slightly better statistical
results. However, the shape of the plume is signifcantly
diferent between simulations and experiments. It must be
added that the test case they are using has a pollutant
source in the wake of the building, where the fow feld
modelling is already problematic, and they focus on the
diferent modelling approaches of the fow feld. Gorl´ e et al.
[5] have investigated the efect of turbulent kinetic energy
inlet boundary conditions on the dispersion results and