J. Wind Eng. Ind. Aerodyn.

129 (2014) 69102

Contents lists available at ScienceDirect

Journal of Wind Engineering

and Industrial Aerodynamics
journal homepage: www.elsevier.com/locate/jweia

50 years of Computational Wind Engineering: Past, present and future$

Bert Blocken a,b,n
a
Building Physics and Services, Department of the Built Environment, Eindhoven University of Technology, P.O. box 513, 5600 MB Eindhoven, The Netherlands
b
Building Physics Section, Department of Civil Engineering, Leuven University, Kasteelpark Arenberg 40 bus 2447, 3001 Heverlee, Belgium

art ic l e i nf o a b s t r a c t

Article history: In the past 50 years, Computational Wind Engineering (CWE) has undergone a successful transition from
Received 23 December 2013 an emerging eld into an increasingly established eld in wind engineering research, practice and
Received in revised form education. This paper provides a perspective on the past, present and future of CWE. It addresses three
9 March 2014
key illustrations of the success of CWE: (1) the establishment of CWE as an individual research and
Accepted 13 March 2014
application area in wind engineering with its own successful conference series under the umbrella of the
International Association of Wind Engineering (IAWE); (2) the increasing range of topics covered in
Keywords: CWE; and (3) the history of overview and review papers in CWE. The paper also outlines some of the
Review earliest achievements in CWE and the resulting development of best practice guidelines. It provides
Historical overview
some views on the complementary relationship between reduced-scale wind-tunnel testing and CFD. It
Computational Fluid Dynamics
re-iterates some important quotes made by CWE and/or CFD researchers in the past, many of which are
Urban physics
Building physics still equally valid today and which are provided without additional comments, to let the quotes speak for
Fluid mechanics themselves. Next, as application examples to the foregoing sections, the paper provides a more detailed
view on CFD simulation of pedestrian-level wind conditions around buildings, CFD simulation of natural
ventilation of buildings and CFD simulation of wind-driven rain on building facades. Finally, a non-
exhaustive perspective on the future of CWE is provided.
& 2014 Elsevier Ltd. All rights reserved.

1. Introduction At the occasion of the 6th European and African Conference on

Wind Engineering (6EACWE) in Cambridge in July 2013, the Interna-
Computational Wind Engineering (CWE) is primarily dened as tional Association of Wind Engineering (IAWE) celebrated the 50th
the use of Computational Fluid Dynamics (CFD) for wind engineering anniversary of the historical meeting in Teddington in 1963. Also the
applications, although it also includes other approaches of computer historical starting point of CWE could also be situated around 1963,
modelling and in the broadest sense also eld and wind-tunnel when Smagorinsky (1963) developed one of the rst successful
measurements supporting CWE model development and evaluation approaches to Large Eddy Simulation (LES), the SmagorinskyLilly
(Murakami, 1990a, 1990b, 1993a, 1997, 1998; Meroney, 1997; model, which is still intensively used in many areas of uid mechanics
Stathopoulos, 1997, 2002; Baker, 2000; Murakami et al., 2008; today. Which is the actual historical starting point of CWE depends on
Huber and Blocken, 2011). In the present paper however, I will restrict the spatial scale considered (see Section 2). The main research area of
the denition of CWE to the use of CFD and other computer Smagorinsky was Numerical Weather Prediction (NWP), applied at the
approaches in wind engineering. Wind engineering itself is best synoptic scale or meteorological macroscale. However, the practical
dened as the rational treatment of interactions between wind in deployment of NWP already started in the 1950s (e.g. Charney et al.,
the atmospheric boundary layer and man and his works on the surface 1950; Smagorinsky, 1953, 1958; Charney, 1955; Phillips, 1956).
of Earth. (Cermak, 1975). CWE has come a long way. 50 years ago, the In meteorology, the 1960s and later years were characterised by
present extent of achievements and applications in CWE could hardly the continued development and application of NWP (e.g. Phillips,
have been imagined. 1960; Smagorinsky, 1963, 1969; Kasahara and Washington. 1967;
Shuman and Hovermale, 1968; Kasahara, 1974). In addition, there
was the emergence of mesoscale analyses in 2D (i.e. in a vertical

This paper is an extended version of a keynote paper presented at the 6th section) of phenomena such as sea breezes with and without
European and African Conference on Wind Engineering, Cambridge, UK, prevailing (synoptic) winds (e.g. Pearce, 1955; Fisher, 1961; Estoque,
July 711, 2013. 1961, 1962; Magata, 1965), convective motions over mountain ridges
n
Correspondence address: Building Physics and Services, Department of the Built
(e.g. Fosberg, 1967, 1969) and ow over (simplied urban) heat islands
Environment, Eindhoven University of Technology, P.O. box 513, 5600 MB Eindho-
ven, The Netherlands. with and without prevailing winds (e.g. Estoque and Bhumralkar,
E-mail address: [email protected] 1969; Delage and Taylor, 1970; Meroney and Yamada, 1971, 1972).

http://dx.doi.org/10.1016/j.jweia.2014.03.008
0167-6105/& 2014 Elsevier Ltd. All rights reserved.
70 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Other numerical studies addressed the microscale ow around (2) the increasingly wide range of topics covered in CWE, ranging
explicitly modelled surface-mounted obstacles such as buildings (e.g. from pedestrian-level wind conditions over natural ventilation of
Yamada and Meroney, 1972; Hirt and Cook 1972; Frost et al., 1974) and buildings and wind loads on buildings and bridges to sports
complex terrain (e.g. Hirt and Cook 1972; Deaves, 1975; Derickson and aerodynamics; and (3) the history of review and overview papers
Meroney, 1977). Of particular importance for CWE were the pioneering in CWE. Each of these three illustrations will be addressed in this
studies by Meroney and his co-workers in which the so-called hybrid paper.
approach was pursued: systematic comparison of numerical simula- CWE is complementary to other, more traditional areas of wind
tions with dedicated wind-tunnel measurements in an atmospheric engineering, such as full-scale on-site experimentation and
boundary layer wind tunnel (e.g. Meroney and Yamada, 1971, 1972; reduced-scale wind-tunnel testing. Each approach has its specic
Yamada and Meroney, 1972; Derickson and Meroney, 1977). advantages and disadvantages. The main advantage of on-site
In Aerospace Engineering, the T3 group at the Los Alamos measurements is that they are able to capture the real complexity
National Laboratories in 1963 rst used computers to model the of the problem under study. Important disadvantages however are
2D swirling ow around an object, using the vorticity stream that they are not fully controllable due to among others the
function method, followed by the rst 3D application by Hess and inherently variable meteorological conditions, that they are not
Smith (1967) using the so-called panel method. possible in the design stage of a building or urban area and that
Driven by these early achievements, later efforts in CWE usually only point measurements are performed. The latter dis-
focused on the determination and analysis of the wind velocity advantage also holds for wind-tunnel measurements. Techniques
and pressure elds around buildings (e.g. Vasilic-Melling, 1977; such as Particle Image Velocimetry (PIV) and Laser-Induced
Hanson et al., 1986; Paterson and Apelt, 1986, 1989, 1990; Fluorescence (LIF) in principle allow planar or even full 3D data
Murakami 1990a, 1990b, 1990c, 1993a, 1993b; Murakami et al., to be obtained in wind-tunnel tests, but the cost is considerably
1987, 1990, 1992; Murakami and Mochida, 1988, 1989; Baskaran higher and application for complicated geometries can be ham-
and Stathopoulos, 1989, 1992; Stathopoulos and Baskaran, 1990; pered by laser-light shielding by the obstructions constituting the
Baetke et al., 1990; Wu et al., 1992; Mochida et al., 1993; Nicholls model, e.g. in case of an urban model consisting of many buildings.
et al., 1993). A strong impetus to CWE was provided by the Another disadvantage is the required adherence to similarity
organisation of a new symposium in 1992 by S. Murakami in criteria in reduced-scale testing, which can limit the extent and
Tokyo, Japan: the International Symposium on Computational the range of problems that can be studied in wind tunnels.
Wind Engineering (CWE). The importance of this symposium CWE/CFD has some particular advantages over experimental (full-
cannot be overemphasised (Stathopoulos, 2013): For the rst time, scale or reduced-scale) testing. It can provide detailed information on
it joined wind engineering delegates with classical aerodynami- the relevant ow variables in the whole calculation domain (whole-
cists, who were using CFD rather routinely to address and solve ow eld data), under well-controlled conditions and without
aeronautical problems. As these problems are very different from similarity constraints. However, the accuracy and reliability of CFD
wind engineering problems, the rst CWE symposium led to a simulations are of concern and solution verication and validation
very fruitful interaction between the different groups of delegates. studies are imperative. This requires high-quality full-scale or
It also marked the beginning of a period of impressive growth in reduced-scale measurements, which in turn should satisfy important
CWE developments and applications, and the beginning of a whole quality criteria. Therefore, experiments remain indispensable for
new series of symposia, demonstrating not only the growing CWE. In addition, it is widely recognised that the results of CFD
importance of CWE, but also the pioneering and visionary char- simulations can be very sensitive to the wide range of computational
acter of their founder, S. Murakami. parameters that have to be set by the user. For a typical simulation,
The difference in time between the earliest CFD developments the user has to select the target variables, the approximate form of
in the 1950s and the later application of CFD in CWE for wind the governing equations, the turbulence model, the computational
velocity and pressure elds around buildings is attributed to the domain, the computational grid, the boundary conditions, the dis-
specic difculties associated with the ow eld around bluff cretisation schemes, the convergence criteria, etc. This expresses the
bodies with sharp edges, many of which are not encountered in need for best practice guidelines for CWE. Best practice guidelines
CFD computations for simple ows such as channel ow and will also be addressed in this paper. While the foundations for these
simple shear ow (e.g. Ferziger, 1990; Leschziner, 1990, 1993; best practice guidelines were already laid in the early years of CWE
Stathopoulos, 1997; Murakami, 1998). Murakami (1998) meticu- applied to buildings and structures, only in the past 14 years have
lously outlined the main difculties in CWE: (1) the high Reynolds these efforts been compiled into extensive best practice guideline
numbers in wind engineering applications, necessitating high grid documents (e.g. Casey and Wintergerste, 2000; Franke et al. 2004,
resolutions, especially in near-wall regions as well as accurate wall 2007, 2011; Britter and Schatzmann 2007; Tominaga et al. 2008a;
functions; (2) the complex nature of the 3D ow eld with Tamura et al., 2008; Blocken et al., 2012; Blocken and Gualtieri, 2012).
impingement, separation and vortex shedding; (3) the numerical This paper provides a perspective on the past, present and future
difculties associated with ow at sharp corners and conse- of CWE. But it does not start without a major disclaimer. It will also
quences for discretisation schemes; and (4) the inow (and out- not end without a major acknowledgement section. CWE has grown
ow) boundary conditions, which are particularly challenging for to a strongly established eld in wind engineering research, practice
LES. These difculties were directly linked to limitations in and education. It is employed daily by probably thousands of
physical modelling and in computational requirements at those researchers, practitioners and teachers all over the world. In addition,
times, but many of those limitations are still to some extent the realm of CWE has now spread to so many topics that it becomes
present today. increasingly difcult to oversee them all. The perspective presented
In spite of these difculties, in the past decades and driven by in this paper is therefore inherently incomplete. And the author
the pioneering studies mentioned above, CWE has undergone a apologises to all researchers and practitioners whose valuable con-
successful transition from an emerging eld into an increasingly tributions are not included in this paper.
established eld in wind engineering research, practice and The paper starts with outlining some of the earliest CWE
education. This transition and the success of CWE are illustrated achievements, followed by an overview of the CWE best practice
by (1) the establishment of CWE as an individual research and guidelines that originated from those achievements, and some
application area in wind engineering with its own successful main aspects of these guidelines. Next, the historical background
conference series under the umbrella of the IAWE (Solari, 2007); of the CWE symposia is presented, and the increase in scope of
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 71

Fig. 1. Spatial and temporal scales of atmospheric phenomena and how these phenomena are treated in Reynolds-averaged NavierStokes (RANS) mesoscale or obstacle
resolving microscale models (right columns) (Schlnzen et al., 2011, Elsevier). The characteristic scales are based on Orlanski (1975) and Randerson (1976), the model scales
are an update of diagrams by Schlnzen (1996) and Moussiopoulos et al. (2003). Dashed areas in the right columns indicate the currently used RANS model resolutions and
the resulting possibly resolvable minimum phenomena sizes.

these symposia throughout the years is demonstrated. In addition, 2. Some early CWE achievements at macroscale, mesoscale
an overview of CWE review and overview papers is provided and and microscale
the complementary character of reduced-scale wind-tunnel test-
ing and CWE is discussed. The paper also re-iterates some As mentioned in the previous section, the historical starting
important quotes made by CWE and/or CFD researchers in the point of CWE depends on the spatial scale considered. Fig. 1,
past, many of which are still equally valid today. They are provided adopted from (Schlnzen et al., 2011), provides an overview of the
without additional comments, to let the quotes speak for them- spatial and temporal scales of atmospheric phenomena. A distinc-
selves. Next, as application examples to the preceding sections, the tion is made between the meteorological macroscale (or synoptic
paper provides a more detailed perspective on CFD simulation of scale), the mesoscale and the microscale. The American Meteor-
pedestrian-level wind conditions (PLW) around buildings, on CFD ological Society (AMS, 2014) provides the following denitions:
simulation of natural ventilation (NV) of buildings and on CFD
simulation of wind-driven rain (WDR) on building facades. A main
Macroscale or synoptic scale: the scale of atmospheric motions
reason for the choice of PLW, NV and WDR is that these are with a typical range of many hundreds of kilometres, including
increasingly recognised as topics that can be studied successfully such phenomena as cyclones and tropical cyclones.
with CFD and, as a result, some extensive case studies have been
Mesoscale: the scale of atmospheric phenomena having hor-
performed. In addition, for PLW, many early applications in CWE izontal scales ranging from a few to several hundred kilo-
focused on this topic, and the best practice guidelines by the metres, including thunderstorms, squall lines, fronts,
COST732 (Franke et al., 2004) and AIJ groups (Tominaga et al., precipitation bands in tropical and extratropical cyclones, and
2008a) were initially intended to support PLW studies. Finally, topographically generated weather systems such as mountain
some perspectives on the future of CWE are provided. waves and sea and land breezes.
The reader will notice that this paper has a larger focus on
Microscale: the scale of atmospheric motions with Lagrangian
Environmental Wind Engineering than on Structural Wind Engi- Rossby numbers greater than 200 or spatial scales of 2 km
neering. This is also reected in the choice of the applications PLW, or less.
NV and WDR that are all part of Environmental Wind Engineering.
The reason for this is threefold: (1) CWE has most extensively been In its broadest sense, CWE encompasses the entire range of spatial
applied in the eld of Environmental Wind Engineering; (2) CWE scales, from the synoptic scale down to the microscale. It even
is at present considered to be a potentially suitable approach for includes effects of microscale wind conditions on wind ow in-
more topics in Environmental Wind Engineering than in Structural side buildings (natural ventilation and inltration) and in build-
Wind Engineering (e.g. it is easier to obtain fairly accurate ing components (e.g. ventilation of wall and roof cavities, see
estimates of mean pressures for NV than similarly accurate peak Nore et al., 2010). However, in the wind engineering community,
pressures for wind loading); (3) this paper and especially the as represented by the IAWE and its ofcial journal (the Journal of
sections on PLW, NV and WDR unavoidably reect the expertise of Wind Engineering and Industrial Aerodynamics JWEIA), the
the author which is Environmental rather than Structural Wind focus is mainly but not exclusively on the meteorological
Engineering. microscale. This focus is also clearly reected in this review paper.
72 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

From the perspective of the broadest interpretation of CWE, some 2.2. Mesoscale
but denitely not all early achievements in CWE at the
different spatial scales are briey mentioned below. Early mesoscale CWE studies include 2D numerical analyses
(i.e. in a vertical section) of sea breezes with and without
2.1. Synoptic scale prevailing (synoptic) winds (e.g. Pearce, 1955; Fisher, 1961;
Estoque, 1961, 1962; Magata, 1965), convective motions over
At the synoptic scale or macroscale,1 the numerical integration mountain ridges (e.g. Fosberg, 1967, 1969) and ow over (simpli-
of the governing equations for atmospheric dynamics subject to ed urban) heat islands with and without prevailing winds (e.g.
specied initial conditions is termed Numerical Weather Predic- Estoque and Bhumralkar, 1969; Delage and Taylor, 1970). Pioneer-
tion (NWP). NWP was proposed by L.F. Richardson2 in 1922 in his ing studies that combined numerical simulations with dedicated
book Weather Prediction by Numerical Process (Richardson, wind-tunnel experiments were performed by R.N. Meroney and
1922), in which he suggested the numerical solution based on his co-workers. Meroney and Yamada (1971) provided the rst
nite differences of the governing differential equations to predict validation study of CFD simulations of the urban heat island effect
the change of atmospheric recirculation. However, at his time, the for 2D stratied airow over the island. Later, this study was
numerical computations required were unattainable and the extended to two identical heated islands in series (Meroney and
manual computations took much longer than the weather Yamada, 1972).
advances to be predicted. Although his work received generally
favourable review comments, its at that time excessively time- 2.3. Microscale
consuming and unrealisable character in combination with an
unfortunate example calculation attracted adverse criticism. At the microscale, the ow around surface-mounted obstacles
Richardson (1922) stated: such as buildings is explicitly resolved, i.e. these obstacles are
represented with their actual shape instead of parameterised as is
Perhaps some day in the dim future it will be possible to advance
typically the case in mesoscale and macroscale simulations.
the computations faster than the weather advances and at a cost
Yamada and Meroney (1972) studied 2D airow over a square
less than the saving to mankind due to the information gained. But
surface-mounted obstacle in a stratied atmosphere, both in the
that is a dream.
wind tunnel and with CFD. Hirt and Cook (1972) calculated 3D
It was only in 1950 that NWP could be applied thanks to the ow around structures and over rough terrain. Frost et al. (1974)
rst electronic computers combined with further developments in numerically analysed the 2D neutrally stratied wind ow over a
atmospheric dynamics, instrumentation and observing practice. As semi-elliptical surface obstruction, used to represent an idealised
stated by Lynch (2006), history has shown that Richardson's building. Deaves (1975) reported numerical simulations of 2D
approach was fundamentally sound and his suggested methodol- neutrally stratied wind ow over several particular hill shapes.
ogy is essentially that used in practical weather forecasting today. In addition, numerical studies of ow over hills were also
According to Chapman (1965), Charney, after his successful reported by among others Wallington and Portnall (1958),
achievements in NWP with digital computing, addressed the Royal Sawyer (1960), Taylor and Gent (1974) and Derickson and Meroney
Meteorological Society stating that: (1977). The latter authors also provided a comparison with wind-
tunnel experiments for neutrally stratied ow and further
to the extent that my work in weather prediction has been of numerical experiments with stable and unstable stratication.
value, it has been a vindication of the vision of my distinguished CFD simulation of wind ow around 3D buildings started with
predecessor, Lewis F. Richardson fundamental studies for isolated buildings, often with a cubical
Moreover, Charney sent copies of several reports to Richardson shape, to analyse the velocity and pressure elds (e.g. Vasilic-
who responded them to be an enormous scientic advance Melling, 1977; Hanson et al., 1986; Paterson and Apelt, 1986, 1989,
(Lynch, 2008). Indeed, some decades after his now very famous 1990; Murakami et al., 1987, 1990, 1992; Murakami and Mochida,
book, Richardson's dream had come true. 1988, 1989; Baskaran and Stathopoulos, 1989, 1992; Stathopoulos
NWP was also the main eld in which Smagorinsky briey and Baskaran, 1990; Murakami, 1990b, 1990c, 1993b; Baetke et al.,
mentioned in Section 1 of this paper was active. Smagorinsky 1990; Mochida et al., 1993). Together with later studies, they laid
(1963) extended early weather models to include variables such as the foundations for the current best practice guidelines, by
wind, cloud cover, precipitation, atmospheric pressure and radia- focusing on the importance of grid resolution (e.g. Murakami
tion emanating from the earth and sun. This required a method to and Mochida, 1989; Murakami, 1990b, 1990c; Baskaran and
account for atmospheric turbulence occurring on scales smaller Stathopoulos, 1992), the inuence of the boundary conditions on
than the model grid size and led to the development of one of the the numerical results (e.g. Murakami and Mochida, 1989; Paterson
rst successful approaches to Large-Eddy Simulation: the Smagor- and Apelt, 1990; Baetke et al., 1990; Stathopoulos and Baskaran,
inskyLilly model. 1990; Baskaran and Stathopoulos, 1992) and by comparing the
Early macroscale studies were reported by e.g. Charney performance of various types of turbulence models in steady RANS
et al. (1950), Smagorinsky (1953, 1958), Charney (1955), Phillips simulations (e.g. Baskaran and Stathopoulos, 1989; Murakami
(1956, 1960), Smagorinsky (1963, 1969), Kasahara and Washington et al., 1992; Murakami, 1993b; Mochida et al., 2002). Also compar-
(1967), Shuman and Hovermale (1968) and Kasahara (1974). For isons of steady RANS versus LES were performed (e.g. Murakami
historical and contemporary review and overview papers on NWP, et al., 1990, 1992; Murakami, 1990c, 1993b).
the reader is referred to (Platzman, 1979; Haltiner and Williams, In the past, especially the deciencies of the steady RANS
1980, Shuman, 1989; Kimura, 2002; Lynch, 2006, 2008). approach with the standard k model (Jones and Launder, 1972)
for wind ow around buildings were addressed. These include the
stagnation point anomaly with overestimation of turbulent kinetic
1
And increasingly also the mesoscale, as the spatial resolution of numerical energy near the frontal corner and the resulting underestimation of
weather prediction increases. the size of separation and recirculation regions on the roof and the
2
Lewis Fry Richardson (18811953), English mathematician, physicist, meteor-
ologist, psychologist, pacist. The Richardson number (dimensionless number of
side faces, and the underestimation of turbulent kinetic energy in
ratio of potential to kinetic energy) and Richardson extrapolation are named after the wake resulting in an overestimation of the size of the cavity zone
him in appreciation of his scientic achievements. and wake. Various revised linear and non-linear k models and also
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 73

second-moment closure models were developed and tested, and In 2004, Franke et al. (2004) compiled a set of specic recom-
showed improved performance for several parts of the ow-eld mendations for the use of CFD in wind engineering from a detailed
(e.g. Baskaran and Stathopoulos, 1989; Murakami et al., 1992; review of the literature, as part of the European COST6 Action C14:
Murakami, 1993b; Wright et al., 2001; Mochida et al., 2002). Impact of Wind and Storm on City Life and Built Environment. Later,
However, the main limitation of steady RANS modelling remained: this contribution was extended into an extensive Best Practice
its incapability to model the inherently transient features of the ow Guideline for the CFD simulation of ows in the urban environment
eld such as separation and recirculation downstream of windward (Franke et al., 2007, 2011), in the framework of the COST Action 732:
edges and vortex shedding in the wake. These large-scale features Quality Assurance and Improvement of Microscale Meteorological
can be explicitly resolved by LES. While unsteady RANS (URANS) has Models, managed by Schatzmann and Britter (http://www.mi.uni-
hardly been used to study wind ow around buildings, early hamburg.de/Home.484.0.html). Like the ERCOFTAC guidelines, also
applications of LES for this purpose were already made by Murakami these guidelines primarily focused on steady RANS simulations,
et al. in 1987 (Murakami et al., 1987), and later by Murakami et al. although also some limited information on URANS, LES and hybrid
(1990, 1992) and Murakami (1990c). These studies illustrated the URANS/LES was provided. When using CFD tools, whether they are
intrinsically superior performance of LES compared to RANS. Never- academic/open source or commercial codes, it is also important that
theless, as will be discussed further, LES entails specic disadvan- the code is well documented, and that basic verication tests and
tages that are not easy to overcome, including the strongly increased validation studies have been successfully performed and reported. A
computational requirements and the difculty in specifying appro- good description of how a microscale airow and dispersion model
priate time-dependent inlet and wall boundary conditions. has to be documented can be found in the Model Evaluation
The studies mentioned above are not all studies that were Guidance Document published in the COST Action 732 by Britter
performed for isolated buildings. But starting from the 1990s, and Schatzmann (2007).
supported by the previous studies and the increased availability of In Japan, working groups of the Architectural Institute of Japan
computational power and CFD codes, fundamental studies gradu- (AIJ) conducted extensive cross-comparisons between CFD simulation
ally shifted their focus to multiple-building congurations, and results and high-quality wind-tunnel measurements to support the
also application studies were increasingly performed. In addition, development of guidelines for practical CFD applications. Part of these
the sensitivity of the CFD results to the large number of computa- efforts were reported by Yoshie et al. (2007). In 2008, Tominaga et al.
tional parameters to be set by the user and the possibility of (2008a) published the AIJ guidelines for practical applications of CFD
applying CFD in practice led to the development of best practice to pedestrian wind environment around buildings, and Tamura et al.
guidelines, as discussed in the next section. (2008) wrote the AIJ guide for numerical prediction of wind loads on
buildings. The guidelines by Tominaga et al. (2008a) focus on steady
RANS simulations, while the guidelines by Tamura et al. (2008) also
consider LES, given the importance of time-dependent analysis for
3. CWE best practice guidelines wind loading of buildings and structures.
More generic best practice advice was provided by Jakeman
In CFD simulations, a large number of choices need to be made by et al. (2006) in the article Ten iterative steps in development and
the user. It is well known that these choices can have a very large evaluation of environmental models, which were later on
impact on the results. Already since the start of the application of extended to development and evaluation of process-based bio-
CFD for wind ow around bluff bodies in the late 70s and 80s, geochemical models of estuaries by Robson et al. (2008) but also to
researchers have been testing the inuence of these parameters on CFD for environmental uid mechanics (including CWE) by
the results, which has provided a lot of valuable information (e.g. Blocken and Gualtieri (2012). Blocken et al. (2012) also provided
Murakami and Mochida, 1989; Baetke et al., 1990; Stathopoulos and a general decision framework for the analysis of PLW comfort and
Baskaran, 1990; Cowan et al., 1997; Hall, 1997). In addition, safety in urban areas.
Schatzmann et al. (1997) provided an important contribution on These best practice guideline documents have been based on
validation with eld and laboratory data. However, initially this and/or reinforced by more basic guidelines and standards con-
information was dispersed over a large number of individual pub- cerning verication and validation, as outlined in e.g. Roache
lications in different journals, conference proceedings and reports. (1994, 1997); AIAA (1998),7; Oberkampf et al. (2004); Roy
In 2000, the ERCOFTAC3 Special Interest Group on Quality and (2005); Roy and Oberkampf (2010); ASME (2009),8 and others. It
Trust in Industrial CFD published an extensive set of best practice is interesting to note that the importance of numerical accuracy
guidelines for industrial CFD users (Casey and Wintergerste, control is emphasised by the Journal of Fluids Engineering Editor-
2000). These guidelines were focused on RANS simulations. ial Policy (ASME, 2011), incited by contributions by Roache et al.
Although they were not specically intended for wind engineer- (1986) and Freitas (1993), which demand at least formally second-
ing, many of these guidelines also apply for CWE. Within the EC order accurate spatial discretisation.
project ECORA,4 Menter et al. (2002) published best practice In addition to these general guidelines, also some very specic
guidelines based on the ERCOFTAC guidelines but modied and guidelines were published. These include (1) consistent modelling
extended specically for CFD code validation. Within QNET-CFD,5 of equilibrium atmospheric boundary layers in computational
the Thematic Area on Civil Construction and HVAC (Heating, domains (e.g. Richards and Hoxey, 1993; Blocken et al., 2007a,
Ventilating and Air-Conditioning) and the Thematic Area on the 2007b; Hargreaves and Wright, 2007; Franke et al., 2007; Di
Environment presented some best practice advice for CFD simula- Sabatino et al., 2007; Gorl et al., 2009; Yang et al., 2009;
tions of wind ow and dispersion (Scaperdas and Gilham, 2004; Parente et al., 2011; Richards and Norris, 2011); (2) high-quality
Bartzis et al., 2004). grid generation (e.g. Tucker and Mosquera, 2001; van Hooff and
Blocken, 2010a) and (3) validation with eld and laboratory data
(e.g. Schatzmann et al., 1997; Schatzmann and Leitl, 2011). Note
3
ERCOFTAC European Research Community on Flow, Turbulence and
Combustion.
4
ECORAEvaluation of Computational Fluid Dynamic Methods for Reactor
6
Safety Analysis. COST European Cooperation in Science and Technology.
5 7
QNET-CFD Network for Quality and Trust in the Industrial Application of AIAA American Institute of Aeronautics and Astronautics.
8
CFD. ASME American Society of Mechanical Engineers.
74 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Table 1
Overview of CWE symposia, with acronym, location, date, chair(s) and special issue.

Nr Acronym Location Date Chair(s) Special issue

1 CWE1992 Tokyo, Japan August 2123, 1992 Murakami JWEIA 1993, vol. 4647
2 CWE1996 Fort Collins, Colorado, USA August 48, 1996 Meroney, Bienkiewicz JWEIA 1997, vol. 6768
3 CWE2000 Birmingham, UK September 47, 2000 Baker WAS 2002, vol. 5
4 CWE2006 Yokohama, Japan July 1619, 2006 Murakami, Matsumoto, Tamura JWEIA 2008, vol. 96(1011)
5 CWE2010 Chapel Hill, North Carolina, USA May 2327, 2010 Huber, Blocken, Stathopoulos JWEIA 2011, vol. 99(4)
6 CWE2014 Hamburg, Germany June 813, 2014 Schlnzen, Hffer, Leitl

that most of the efforts in the rst two areas were focused on of symposia, as indicated in Table 1. In 2005, the Executive Board of
steady RANS simulations. the IAWE, under presidency of G. Solari, made the important decision
The establishment of these guidelines has been an important to manage these CWE symposia under the umbrella of the IAWE. As
step towards more accurate and reliable CFD simulations. a result, since 2010, the CWE symposia are being held in a perfect
Although several of the guideline documents mentioned above temporal sequence and distribution of the most important confer-
have been developed with focus on PLW conditions (Franke et al., ences in the calendar of wind engineering (i.e. ICWE,9 ACWE,10
2004; Tominaga et al., 2008a; Blocken et al., 2012), most of the APCWE,11 EACWE,12 BBAA13) and also in a perfect rotation of
information is also applicable to other topics in CWE. conference venues among the different regions14 (Solari, 2007).
The success of the CWE conferences can be demonstrated in
many ways, including number of presentations, number of parti-
4. CWE symposia: historical background and scope cipants, number of countries represented at the conference and
the resulting special issues in scientic journals. These numbers
An excellent and very comprehensive overview of the history, have continued to grow. There are several reasons for this.
progress and prospects of the IAWE and of the conferences in wind Certainly, the increasing availability of computational resources
engineering has been provided by Solari (2007). He recalls the rst has played a main role. But at least equally important is the fact
Symposium on Wind Effects on Buildings and Structures in Ted- that each of these successive symposia has beneted immensely
dington, 1963, organised by K. Scruton and his group, precisely 50 from the large success, strong reputation and excellent support of
years before 6EACWE. This very successful conference incited a the preceding symposia and their organisers. The latest CWE
series of conferences: the second one in Ottawa, Canada, 1967, the symposium, CWE2010, hosted by Huber, Blocken and Stathopou-
third one in Tokyo, Japan, 1971, and the fourth one in London, UK, los, had 267 presentations including 6 keynote presentations and
1975. As mentioned by Solari (2007), these conferences were it had the pleasure of welcoming 299 participants from 30
destined to become milestones of modern wind engineering. At countries. So far, each of the ve CWE symposia has also resulted
the fourth conference in London, the IAWE was founded. This in a Special Issue in either the JWEIA or the journal Wind and
incited the extension of the discipline of wind engineering beyond Structures. The growth of CWE is also demonstrated by the
the study of wind effects on buildings and structures to a wider set increasing number of research topics covered by these symposia
of topics and also a wider community (Solari, 2007). Indeed, the over time, as shown in Table 2 that was prepared for the present
fth conference in Fort Collins, Colorado in 1979, did not only focus paper. Two important comments concerning this Table are: (1) For
on wind effects on buildings and structures, but also included four CWE1992 and CWE1996, only selected papers of these symposia
innovative sections (Solari, 2007) that would later become main were compiled into the proceedings, and the classication in
CWE topics: (1) Social and economic impact of wind storms; Table 2 pertains to these proceedings; (2) The topics in Table 2
(2) Wind environment (wind erosion, natural ventilation, pedes- are not mutually exclusive. Examples are bridges and wind-
trian wind comfort); (3) Physical and mathematical modelling structure interaction; mesoscale modelling and meteorological
(stack gas dispersion and snow drifting simulations in wind phenomena; wind and thermal environment and pedestrian-
tunnels); and (4) Wind engineering applications (dispersion of level wind conditions, etc. Therefore, the comments below the
chemical vapours and natural gas, siting of wind turbines, and train Table detail the choices underlying this classication.
aerodynamics). It is important to note that this fth conference was While many factors determine the type of contributions to a
also the rst International Conference on Wind Engineering (ICWE). given symposium, Table 2 shows that the overall tendency is clearly
The later editions of the ICWE conference series saw an increasing towards a coverage of almost all traditional wind engineering topics
growth of wind engineering activity and expertise in a wide range of by CWE. In particular, the following observations are made:
topics. In the framework of CWE, it is important to mention the eight
ICWE held in London, Ontario, Canada, 1991 and chaired by A.G.
A number of topics have been present from the very beginning.
Davenport (Davenport, 1991). Not only was this conference charac- This holds in particular for the three general and fundamental
terised by a very strong increase in the number of submitted and topics: (1) ABL simulation, as an essential prerequisite for
presented papers (with 264 presentations in 48 sessions, necessitat- accurate wind-ow simulation around buildings and struc-
ing the use of parallel sessions), but it also introduced a series of new tures; (2) bluff body aerodynamics, which in this classication
specic sessions, one of which was Computational Fluid Dynamics refers to studies with a clear focus on the application of CFD to
(CFD), chaired by T. Stathopoulos and R.J. Kind. As correctly pointed explain uid mechanical/physical processes of wind ow
out by Solari (2007), the importance of this rst session devoted to around buildings and structures; (3) turbulence modelling
CFD by an ICWE cannot be overestimated.
Only one year later, another very strong impetus to CWE was
9
provided by the organisation of a new symposium in 1992 by S. ICWE International Conference on Wind Engineering.
10
Murakami in Tokyo, Japan: the International Symposium on Compu- ACWE American Conference on Wind Engineering.
11
APCWE Asia-Pacic Conference on Wind Engineering.
tational Wind Engineering (CWE). The enormous success of this 12
EACWE European-African Conference on Wind Engineering.
symposium was not only indicative of the potential of CFD for wind 13
BBAABluff Body Aerodynamics and Applications.
engineering, but it was also the very successful start of a new series 14
The American, Asia-Pacic and European-African Region.
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 75

Table 2
Overview of scope and type of contributions in CWE conference proceedings.

CWE1992a CWE1996a CWE2000 CWE2006 CWE2010

Review and overview papers x x x x x

General topics
Atmospheric boundary layer simulationc x x - x x
Bluff body aerodynamics d x x x x x
Turbulence modelling and numerical techniquese x x x x x
Environmental wind engineeringb
Wind and thermal environmentf x x x x
Pedestrian-level wind conditionsg x x x x
Air pollutant dispersionh x x x x x
Meteorological phenomena x x x x
Fire x
Flow over (complex) topographyi x x x x x
(Natural) ventilation of buildings and vehiclesj x x x
Sand, dust and snow transport x x x x
Wind-driven rain x x x x
Surface convective heat transferk x x
Acoustics x
Structural wind engineeringb
Wind loads on generic obstaclesl x x x x x
Wind loads on buildings and structuresm x x x x x
Wind loads on solar panels x
Bridges x x x x x
Vibrations / wind-structure interactions x x x x x
Wind energy
Wind energy siting assessmentn x x
Wind energy in the built environment x x
Wind turbine rotor aerodynamics x x x
Other topics
Vehicle aerodynamicso - x x x x
Sports aerodynamics x
Windborne debris x x
Computer-aided experimentsp x x x x
Experiments for CFD validation x x x x
Mesoscale modellingq x x x x x
Total number of topics 18 18 17 25 26

a
Only selected papers of these symposia were compiled into the proceedings; the indications in the table pertain to these proceedings.
b
Excluding aerodynamics of vehicles and sports.
c
Includes inow boundary conditions for RANS and LES.
d
Focus on uid mechanical and physical aspects of wind ow around buildings and structures.
e
Focus on physical and numerical modelling such as turbulence modelling, wall functions, grid generation/discretisation, assessment of physical and numerical
modelling errors, sensitivity studies and best practice guidelines.
f
Includes outdoor thermal comfort and urban heat island effect.
g
Isothermal studies of pedestrian-level wind conditions around buildings.
h
Also includes urban ventilation studies that do not explicitly model the dispersion of a scalar.
i
Excludes studies of wind ow around bluff bodies and bluff body groups, which are contained in the category Bluff body aerodynamics.
j
Only includes studies on ventilation of indoor environments, ventilation of outdoor environments is part of category Air pollutant dispersion.
k
High-resolution modelling of surface convective heat transfer including the thin viscous sublayer that represents the largest thermal resistance.
l
Circular or square cylinders, often in smooth approach ow.
m
Generic or real surface-mounted buildings and structures, in turbulent boundary layer approach ow.
n
Off-shore or on-shore, but natural terrain, not built environment.
o
Including wind loads, therefore different from the topics in Structural wind engineering.
p
Includes the use of CFD for wind-tunnel design and evaluation.
q
Mesoscale modelling alone or coupled mesoscale-microscale modelling.

and numerical techniques, which in this classication refers to thermal environment; pedestrian-level wind conditions; air
physical15 and numerical modelling (turbulence modelling, pollutant dispersion; meteorological phenomena; ow over
wall functions, grid generation, discretisation schemes, etc.), (complex) topography; sand, dust and snow transport; wind-
the related physical and numerical modelling errors and the driven rain; wind loads on generic obstacles and on buildings
resulting best practice guidelines. and structures; bridges; vibrations and/or wind-structure inter-
actions; wind turbine rotor aerodynamics; computer-aided

Also a very large number of applied topics, in spite of their experiments; and mesoscale modelling.
complexity, was already present from the beginning: wind and
New16 topics at later CWE symposia were (1) surface convec-
tive heat transfer; (2) wind and acoustics; (3) wind energy site
assessment, (4) wind-borne debris and (5) res.
15
The term physical modelling here refers to modelling of the physical

Less represented topics were (1) wind energy in the built
processes by CFD and relates to the modelling of the object geometry and ow environment; and (2) sports aerodynamics. This can partly be
physics. It is associated with the specication of the equations to be solved and the
boundary conditions for these equations. It is to be distinguished from numerical
16
modelling, which is associated with the solution procedures of the equations. It is New topics at CWE symposia does not necessarily imply that these topics
also to be distinguished from wind-tunnel testing, which is sometimes also termed were dealt with for the rst time at the CWE symposia, rather that they were not
physical modelling. presented at previous CWE symposia.
76 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Table 3
Overview of CWE review and overview papers published in archival journals

Topic Author(s) (year) of publication

General topics
Atmospheric boundary layer simulation Franke et al. (2004, 2007), Tominaga et al. (2008a), Tamura (2008), Tamura et al. (2008), Tabor and Baba-Ahmadi
(2010)
Bluff body aerodynamics, turbulence modelling Murakami (1990b, 1993b, 1997, 1998, 1999), Roache (1994, 1997), Ferziger (1990, 1993a, 1993b), Leschziner
and numerical techniques (1990,1993), Hughes and Jansen (1993), Shah and Ferziger (1997), Stathopoulos (1997, 2002), Rodi (1997), Tamura
et al. (1997, 2008), Gosman (1999), Castro and Graham (1999), Tezduyar (1999), Franke et al. (2004, 2007, 2011),
Fujii (2005), Norton and Sun (2006), Bartzis (2006), Baker (2007, 2010), Tamura (2008), Hanjalic and Kenjeres
(2008), Tominaga et al. (2008a), Squires et al. (2008), Tabor and Baba-Ahmadi (2010), Cochran and Derickson
(2011), Blocken and Gualtieri (2012), Lee et al. (2013)
IAWE Solari (2007)
Environmental wind engineering
Wind and thermal environment Murakami et al. (1999), Stathopoulos (2006), Mochida and Lun (2008), Mochida et al. (2008), Moonen et al. (2012),
Lee et al. (2013a)
Pedestrian-level wind conditions Blocken and Carmeliet (2004b), Stathopoulos (2006), Yoshie et al. (2007), Mochida and Lun (2008), Blocken et al.
(2011a, 2012), Moonen et al. (2012), Blocken and Stathopoulos (2013)
Air pollutant dispersion Lee et al. (1997), Vardoulakis et al. (2003), Meroney (2004), Canepa (2004), Li et al. (2006), Holmes and Morawska
(2006), Tominaga and Stathopoulos (2007, 2013), Fernando et al. (2010), Blocken et al. (2011a, 2013b), Gousseau
et al. (2011a), Balczo et al. (2011), Tyagi et al. (2012), Lee et al. (2013), Di Sabatino et al. (2013)
Meteorological phenomena
Fire
Flow over (complex) topography Wood (2000), Bitsuamlak et al. (2004)
(Natural) ventilation of buildings and vehicles Reichrath and Davies (2002), Norton et al. (2007), Chen (2009), Bournet and Boulard (2010), Jiru and Bitsuamlak
(2010), Ramponi and Blocken (2012a), Bjerg et al. (2013b)
Sand, dust and snow transport Livingstone et al. (2007), Tominaga et al. (2011)
Wind-driven rain Blocken and Carmeliet (2004a, 2010), Blocken et al. (2011a, 2013a)
Surface convective heat transfer Blocken et al. (2011a), Defraeye et al. (2012, 2013)
Acoustics
Structural wind engineering
Wind loads on generic obstacles
Wind loads on buildings and structures Stathopoulos (1997, 2003), Kareem (2008)
Wind loads on solar panels
Bridges Taddei and Bontempi (2003), Ge and Xiang (2008)
Vibrations and/or wind-structure interactions Tamura (1999)
Wind energy
Wind energy siting assessment Ayotte (2008), Sumner et al. (2010), Port-Agel et al. (2011), Sanderse et al. (2011), Sorensen (2011), Leung and Yang
(2012), Miller et al. (2013)
Wind energy in the built environment
Wind turbine rotor aerodynamics Vermeer et al. (2003), Snel (2003), Hansen et al. (2006), Sumner et al. (2010), Sorensen (2011), Port-Agel et al.
(2011), Roy and Saha (2013), Miller et al. (2013)
Other topics
Vehicle aerodynamics Takagi (1990), Hucho and Sovran (1993), Mueller and DeLaurier (2003), Katz (2006), Baker (2010)
Sports aerodynamics Mehta (1985)
Windborne debris
Computer-aided experiments Wu et al. (1992), Tamura and Matsui (2002), Meroney (2004)
Experiments for CFD validation Robins (2003), Schatzmann et al. (1997), Dalgliesh and Surry (2003), Meroney (2004), Schatzmann and Leitl (2011)
Mesoscale (and macroscale) modelling Platzman (1979), Shuman (1989), Pielke and Nicholls (1997), Kimura (2002), Lynch (2008), Mochida et al. (2011),
Yamada and Katsuyuki (2011), Schlnzen et al. (2011)

explained by the fact that the wind energy community has its contains the keynote and review papers published in the special
own and separate conference series, and that quite a lot of issues of the previous CWE symposia, the CWE review papers
research on wind energy and sports aerodynamics is performed published in the main wind engineering journals as well as several
under contracts with condentiality agreements, and is there- CWE review papers published in other journals. Many relevant
fore not presented at conferences and symposia and also not CFD but not CWE review papers are not included, but do
published in conference and journal papers. nevertheless constitute an important body of work on CFD.
Without wanting to detract from the excellent achievements by
many CWE researchers, special attention in this section is given to
5. An overview of CWE review papers published in scientic a few papers by Murakami. In 1990, Murakami (1990b) provided
journals what is to the best of my knowledge the rst review paper in
CWE. This review was very comprehensive, and, even though
This section is limited to mentioning only a few contributions written more than 20 years ago, it is still very relevant. It
in detail and listing many others in Table 3. This table applies the addressed physical modelling errors, their reduction by improved
same categorisation as Table 2 in Section 4 of this paper. Table 3 is turbulence modelling and wall boundary conditions, numerical
limited to peer-reviewed journal papers, with one exception.17 It errors and their assessment. It illustrated the power and potential
of LES. It also contained a practical application of CWE to
17
determine the pedestrian-level wind speed around four buildings
An exception is made for the extensive review paper by Meroney on Wind
tunnel and numerical simulation of pollution dispersion: a hybrid approach,
to be located on an urban renewal site in a city near Tokyo using
presented at the Croucher Advanced Study Institute, Hong Kong University of steady RANS with the standard k model (Fig. 2a and b). A very
Science and Technology, 6-10 December 2004. remarkable part of this study is the detailed calculation of wind
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 77

conditions at the balconies of one of these buildings, which also 6. Reduced-scale wind-tunnel testing and CWE: competing
included the evaluation of a windbreak as remedial measure for or complementary?
strong winds (Fig. 2cf). 20 years later, in the framework of
revising the Dutch Wind Nuisance Standard NEN 8100, the Dutch As already mentioned in Section 1, CWE is complementary to
Normalisation committee NEN asked the author of the present other, more traditional areas of wind engineering, such as full-
paper to investigate if any studies on wind comfort on balconies scale on-site experimentation and reduced-scale wind-tunnel
had been published in archival journals. This search did not yield testing. Each approach has its specic advantages and disadvan-
any studies that went substantially past the pioneering work by tages. CFD has some particular advantages over experimental (full-
Murakami (1990b). Recent CFD simulations for balconies were scale or reduced-scale) testing, especially the fact that it provides
published by Montazeri and Blocken (2013) and Montazeri et al. detailed information on the relevant ow variables in the whole
(2013), but these simulations although valuable are denitely calculation domain (whole-ow eld data), under well-
situated in the shadow of the achievements by Murakami (1990b), controlled conditions and without similarity constraints.
made 23 years earlier. The following topics/processes have been reported to be
In his 1997 paper Current status and future trends in compu- very difcult or even impossible to study by reduced-scale wind-
tational wind engineering, Murakami (1997) outlines in detail tunnel testing. It has been argued that they should be addressed
why CWE is more difcult compared to some other areas in CFD. by CWE:
Next, this paper provides an impressive list of application exam-
ples, all of which are still intensively studied in CWE today (Fig. 3): 1. wind ow and related processes for study areas where rela-
tively small ow features and dimensions are important, such

velocity and temperature elds around a human body (at the as natural ventilation through relatively small openings, where
real human scale, i.e. the viscous sublayer at the body surface scaling down could change the nature of the ow in these
with y values down to 3.3, which is the layer that represents openings from turbulent to transitional or even laminar;
the largest resistance to heat and moisture transfer) (Fig. 3a 2. wind ow and related processes over large areas that would
and b); require too large scaling factors, which include extensive (in

velocity and pressure elds around a bluff body (bridge deck or terms of horizontal distances) microscale studies but certainly
building structure) (Fig. 3c and d); mesoscale studies;

coupled uid-structure analysis; 3. wind ow and related processes in atmospheric boundary

pollutant dispersion around buildings; layers with stable and unstable stratication;

pedestrian wind conditions around a high-rise building; 4. buoyant ows such as buoyancy-driven natural ventilation and

analysis of outdoor climate within city blocks (including air air pollutant dispersion;
velocity, temperature, moisture and radiation); 5. multiphase ow problems such as the transport and deposition

mesoscale analysis of city and regional climates (Fig. 3e and f). of sand, dust, rain, hail and snow;
6. meteorological phenomena such as tornadoes and downbursts.
The paper continues by outlining new trends in turbulence
modelling for CWE applications, including the still very actual That CWE can overcome the important problems in reduced-
topic of inow boundary conditions for LES, but too many to scale wind-tunnel testing and should be preferred is certainly true
mention them all. for topic (1) in situations where similarity would be strongly
Table 3 provides a list of overview and review papers in CWE. It violated. This refers to similarity issues regarding the lowest range
roughly indicates which topics have been extensively investigated of spatial scales (e.g. van Hooff and Blocken, 2010a, 2010b).
and subsequently reviewed, and which topics might still be in need Concerning topic (2) there are clearly concerns related to
of a review paper. In particular, (additional) review papers are reduced-scale wind-tunnel testing (e.g. Bowen, 2003) although
welcomed on CFD simulation of the neutral, stable and unstable for specic applications also successful wind-tunnel studies with
atmospheric boundary layer, on wind and acoustics, on wind loads large scaling factors have been reported. It should be noted that
on solar panels, on wind energy in the built environment, on sports they require special preparation and care (e.g. Baker et al., 1985;
aerodynamics and on wind-borne debris. Note that a special issue Meroney, 1980, 1990; Conan et al., 2012; Sanz Rodrigo et al., 2012).
in the JWEIA on wind loads on solar panels was established by However, that CWE should be preferred over reduced-scale wind-
Kopp (2013). Furthermore, in 2013, also two Virtual Special Issues tunnel testing is not really true for topics (3), (4) and (5), which
in CWE were established, one on CFD simulation of micro-scale present an almost equally large challenge to CWE as they do to
pollutant dispersion in the built environment in the journal wind-tunnel testing and where quite a number of important and
Building and Environment (Blocken et al., 2013b) and one on CFD successful wind-tunnel studies have been reported (e.g. Iversen,
simulation of pedestrian-level wind conditions around buildings in 1981; Kothari et al., 1986; Meroney, 1982, 1987a, 1987b; Meroney
the JWEIA (Blocken and Stathopoulos, 2013). In addition, note that et al., 1984; Meroney and Neff, 1986; Shin and Meroney, 1988;
papers by Castro and Robins (1977), Robins and Castro (1977a, Meroney and Meroney, 1989; Isyumov and Mikitiuk, 1990; Avissar
1977b), Huber and Snyder (1982), Li and Meroney (1983a, 1983b), et al., 1990; Da Matha SantAnna and Taylor, 1990; Meroney and
Meroney et al. (1996), Tominaga et al. (1997), Gromke and Ruck Melbourne, 1992; Kwok et al., 1992; Smedley et al., 1993; Surry
(2007, 2008, 2009, 2012) and Gromke (2011) have provided et al., 1994; Inculet and Surry, 1994; Delpech et al., 1998; Inculet,
experimental data on pollutant dispersion that have been used 2001; Sanz Rodrigo et al., 2012). Concerning topic (6), while many
extensively for the assessment and validation of numerical past studies have been numerical, also quite some experimental
models (e.g. Li and Stathopoulos, 1997; Selvam, 1997; Tominaga studies (e.g. Chay and Letchford, 2002; Letchford and Chay, 2002;
and Stathopoulos, 2007, 2009, 2010; Milliez and Carissimo, 2007, Letchford et al., 2002; Mason et al., 2005; Xu and Hangan, 2008;
2008; Gromke et al., 2008; Blocken et al., 2008a, 2013b; Buccolieri Tari et al., 2010) or combined numerical-experimental studies
et al., 2009, 2011; Balczo et al., 2009; Santiago et al., 2010; Dejoan (Wood et al., 2001; Meroney, 2003; Sengupta and Sarkar, 2008;
et al., 2010; Salim et al., 2011a, 2011b; Di Sabatino Sengupta et al., 2008) have been published. In addition, recently
et al., 2011; Gousseau et al., 2011b, 2012; Goricsan et al., 2011; new facilities have been developed for enhanced reduced-scale
Saloranta and Hellsten, 2011; Buccolieri and Di Sabatino, 2011; Baik testing of these and other phenomena (e.g. WindEEE Dome at the
et al., 2012; Moonen et al., 2013; Tominaga et al., 2013). University of Western Ontario, Canada UWO, 2014).
78 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Fig. 2. CFD study of the airow around four buildings to be located on an urban renewal site in a city located near Tokyo. A is a 4-storey building, B is a 19-storey building, C
is a 20-storey building and D is a 7-storey building. (a) Plan view of computational grid with 120,120 cells; (b) velocity vector eld near ground level (1.5 m height) for SSW
wind; (c) detailed view of balcony at SW corner of building B; (d) velocity vector eld in vertical plane near corner of building B; (ef) velocity vector eld in horizontal plane
at 1.5 m height at 13th oor without and with windbreak by solid fence (Murakami, 1990b, Elsevier).

In the past decades, often statements have been made that CFD (Castro and Graham, 1999) and by Stathopoulos in his paper The
would replace (reduced-scale) wind-tunnel testing and that it numerical wind tunnel for industrial aerodynamics: real or virtual
would become the numerical wind tunnel. The label numerical in the new millennium? (Stathopoulos, 2002). The next section
wind tunnel was convincingly denounced by Castro and Graham provides a series of quotes from computational wind engineers on
in their paper Numerical wind engineering: the way ahead? the numerical wind tunnel, systematically denouncing this label,
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 79

Fig. 3. (a,b) Velocity and temperature elds around a human body exposed to a weak cross wind, illustrating a rising plume around the human body which is transported
downward by the cross wind; (c,d) wind ow around a real-shaped bridge deck with angle of attack (c) 41 and (d) 81; (e) velocity vector eld over the central part of Japan,
where the Japanese Alps are located; (f) corresponding measurement results (Murakami, 1997, Elsevier).

however without recognising the important complementary value dispersion simulation can be examined and boundary conditions
and potential of CWE. can be modied.
The complementary aspects of wind-tunnel testing and CWE
are multifold. Indeed, wind-tunnel testing can provide the indis-
Moonen et al. (2006) applied CFD to support the design and
pensable high-quality validation data needed for CWE, and CWE
evaluation of wind tunnels. They developed a detailed methodol-
can supplement wind-tunnel testing by providing whole-ow
ogy for numerically simulating the ow conditions in closed-
eld data on all relevant parameters. Furthermore, Leitl and
circuit wind tunnels based on the method of characteristic lines.
Meroney (1997) indicated the value of CFD to design wind-
This effort was explicitly intended as a contribution to the general
tunnel experiments:
philosophy of incorporating CFD in wind-tunnel design and
testing. They stated:
Using numerical codes can help to design and setup wind
tunnel experiments; hence reducing the time required to optimize The methodology developed in this paper and the accuracy
a physical model and expensive pre-runs in a wind tunnel. With a obtained provide perspectives for the use of this methodology
numerical simulation critical points like source design for and of CFD in general as a tool in wind tunnel design and testing
80 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

and for CFD validation studies when detailed boundary (inlet) or the need to rene the grid until converged grid-independent
conditions are not available. solutions are obtained. (Stathopoulos, 1997)
It appears that although CFD is denitely a good friend of wind
Later, Moonen et al. (2007) developed a series of new indicators
engineering, it has not yet become a true ally. (Stathopoulos,
for wind-tunnel test section ow quality and applied CFD to
1997)
illustrate the effectiveness of these indicators. The approach by
Moonen et al. (2006) was adopted by Calautit et al. (2014) for
In this paper, we have presented several applications of CFD
further development of design methodologies of closed-loop
analyses of outdoor climate ranging from human scale to urban
subsonic wind tunnels.
scale. Comparisons of CFD results with measured data are also
given when measured data are available. At this stage, the
7. Some quotes accuracy of CFD predictions is pretty good but not perfect.
However, we do think that the comprehensive assessment based
This section lists some quotes, mainly by wind engineers, on the CFD method combining various factors seems to be the only
computational wind engineers and/or uid mechanicians in the approach for clarifying such complicated phenomena. Thus,
past decades. They give the reader a avour of past discussions, further efforts are strongly required in this eld. (Murakami
concerns and trends, many of which are still equally valid today. et al., 1999)
This section is by no means intended to be complete my
incompleteness disclaimer also applies here. The quotes are Good mental health in a uid or CFD modeler is always indicated
provided without additional comments, to let them speak for by the presence of a suspicious nature, cynicism and a show me
themselves. attitude. These are not necessarily the best traits for a life mate or
a best friend, but they are essential if the integrity of the modeling
7.1. On verication and validation process is to be maintained. (Meroney, 2004)

The results of numerical simulation cannot be free from various

verication is the assessment of the accuracy of the solution to
types of numerical errors Therefore it is indispensable that the
a computational model, primarily by comparison with known
accuracy of numerical simulation be examined by comparing the
solutions. Validation is the assessment of the accuracy of a
numerical results with those from wind tunnel tests or eld
computational simulation by comparison with experimental data.
experiments Therefore the two different methods of research
In verication, the relationship of the simulation to the real world
should proceed in concert and in cooperation with each other.
is not an issue. In validation, the relationship between computa-
(Murakami, 1990c)
tion and the real world, ie, experimental data, is the issue.
(Oberkampf et al., 2004)
The frequently heard argument any solution is better than none
can be dangerous in the extreme. The greatest disaster one can
As Martin Jensen has reminded us the lack of full scale veri-
encounter in computation is not instability or lack of convergence
cation that has been tolerated is embarrassing. It is not
but results that are simultaneously good enough to be believable
characteristic of other technologies such as shipping, transporta-
but bad enough to cause trouble. (Ferziger, 1993c)
tion or aeronautics.' The quote remains relevant and there is a
continuing need for full-scale experimentation to validate loading
Which model is best for which kind of ows (none is expected to data in codes, the use of wind tunnels and CFD. (Baker, 2007,
be good for all ows) is not yet quite clear, partly due to the fact commenting on a quote by Davenport, 1999)
that in many attempts to answer this question numerical errors
played a too important role so clear conclusions were not possible
In practice the quality of model output depends not only on the
In most workshops held so far on the subject of evaluation of
accuracy of the model itself and the model input, but also on the
turbulence models, the differences between solutions produced by
qualication of the person running a model. Numerical simulation
different authors using supposedly the same model were as large if
is a knowledge-based activity. Appropriate knowledge can be
not larger than the differences between the results of the same
transferred to users by recommendations concerning the proper
author using different models. (Ferziger and Peric, 1996)
use of models. For obstacle-resolving CFD codes such recommen-
dations are not straightforward. (Schatzmann and Leitl, 2011)
The very important point, independent of the semantics, is that
use of a veried code is not enough. This point is probably well
The judicial presumption of innocence does not hold in CFD. CFD
recognized by present readers, but it is not universally so.
results are wrong, until proven otherwise (Blocken, 2014,
Especially in the commercial CFD arena, user expectations are
this paper)
often that the purchase and use of a really good code' will remove
from the user the obligation of doing his homework, that is, the
7.2. On appropriate boundary conditions in CWE
straightforward but tedious work of verication of calculations via
systematic grid-convergence studies. This unrealistic hope is In 1958 Jensen showed that in wind-tunnel testing it is just as
sometimes encouraged by advertising. (Roache, 1997) important to correctly model the wind as it is to correctly model
To simply compare model results with measured data is often the building. This lesson must surely carry over into the relatively
inappropriate since data generated in eld or laboratory experi- new eld of computational wind engineering. (Richards and
ments and those from model simulations exhibit systematic Hoxey, 1993)
differences In view of the remarks made above, it must be
concluded that such a comparison often resembles the proverbial By setting appropriate proles for wind velocity and the turbu-
comparison of apples with oranges. (Schatzmann et al., 1997) lence quantities at the inlet, it is often assumed that the boundary
layer will be maintained up to the buildings or obstructions in the
Most practitioners are more concerned with obtaining results ow. This paper shows that this is not the case, even in the
than with either the order of accuracy of their numerical schemes absence of obstructions, and that the velocity and turbulence
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 81

proles decay along the fetch under these default conditions. design for dispersion simulation can be examined and boundary
(Hargreaves and Wright, 2007) conditions can be modied. (Leitl and Meroney, 1997)

The problems typically manifest themselves as unintended It is argued that although the potential is undoubtedly great and
streamwise gradients in the vertical mean wind speed and CFD is thus increasingly being used in industry (for assessment of
turbulence proles as they travel through the computational wind loads, pollutant dispersion, etc.) there are signicant dan-
domain. These gradients can be held responsibleat least partly gers. Without a sound understanding of the uid mechanics
for the discrepancies that are sometimes found between see- appropriate to the particular problem being attacked, an aware-
mingly identical CFD simulations performed with different CFD ness of the extent to which the code being used has been validated
codes and between CFD simulations and measurements. (Blocken for similar problems and a clear understanding of the sources of
et al., 2007a) uncertainty and the accuracy levels actually needed, great caution
is required in using CFD as an integral part of the design process.
[on LES inow conditions] There is certainly scope for further (Castro and Graham, 1999)
development. For instance, the articially generated turbulence
may not be divergence free and this may have some downstream It is true, of course, that even a highly accurate solution to the
effects; this issue has rarely been addressed in the literature. A modelled equations may differ signicantly from the actual ow
more general question, perhaps, relates to just how detailed that would occur given the same boundary conditions, because of
should be the imposed statistics and, indeed, how many of the inadequacies in the turbulence modelling. But this difference is
independent quantities need to be considered. Our feeling is that often of secondary importance compared with those which arise
for general applicability it is important to model not only the because of bad choices (or even plain user mistakes) in all the
turbulence stresses but also correlation scales in all three direc- other areas. (Castro and Graham, 1999)
tions, as in the present method. (Xie and Castro, 2008)

Nevertheless, there seems to be an ever-increasing condence in

Appropriate boundary conditions for computational wind engi-
the results obtained by CFD codes and more and more papers
neering was an issue addressed by Richards and Hoxey at the rst
propagate the idea that the numerical wind tunnel does exist
Computational Wind Engineering Conference and is still a relevant
today and produces results ready to be used by practitioners.
issue today. (Richards and Norris, 2011)
In the authors opinion this is at best premature and at
worst dangerous with the exception of very limited cases.
In order to couple the LES with MMM [ Mesoscale Meteor- (Stathopoulos, 2002)
ological Models] successfully, several problems should be solved.
One of the most signicant issues to be solved from the viewpoint
In spite of some interesting and visually impressive results
of CWE applications is the treatment of the turbulent velocity
produced with Computational Wind Engineering, the numerical
uctuations imposed at the inow as the boundary conditions.
wind tunnel is still virtual rather than real. Its potential however,
(Mochida et al., 2011)
is extremely high and its progress should be monitored carefully.
Many more parallel studies - numerical and experimental - will be
The inhomogeneous ABL has a signicant effect on the prediction necessary in order to increase the present level of condence in the
of the ow and dispersion elds, depending on the percentage computational results. Practitioners should be warned about the
deviation of the incident from the inlet conditions (Ai and Mak, uncertainties of the numerical wind tunnel results and urged to
2013) exercise caution in their utilization. Finally, more effective efforts
should be made in the numerical simulation of uctuating ow
7.3. On the numerical wind tunnel eld and the numerical evaluation of peak values of variables
necessary for design. (Stathopoulos, 2002)
The large majority of ows encountered in engineering, archi-
tectural and environmental applications are considerably more
complex than those mentioned above For these ows, the Applications will become widespread in areas where wind
concept of a numerical wind tunnel generating quantitatively velocities rather than surface pressures are required, such as the
meaningful design data without careful case-related experimental assessment of pedestrian comfort. These trends may well lead to
validation may be decades away, if it is at all a sensible objective the concentration of boundary layer wind tunnel testing for
to pursue. (Leschziner, 1990) complex structures into a smaller number of institutions over
the next few decades. (Baker, 2007)

This [ computer technology] evolution has given rise to the

rather radical view - expressed predominantly among the US The biggest remaining challenge for CWE is the treatment of peak
aerodynamics fraternity - that the wind tunnel is destined to structural wind loads and peak cladding pressures on buildings.
become a convenient storage cabinet for computer output. Continued hybrid use of wind tunnels and CFD with cross
A moments contemplation leads to the conclusion that this comparison validation between wind-tunnel (or full-scale) results
view reects a rather narrow interpretation of CFD, focusing will be essential to gain condence in the methodology. (Cochran
on the particular type of ows most relevant to high-speed and Derickson, 2011)
external aerodynamics and some turbomachinery applications.
(Leschziner, 1993) As demonstrated in this paper, inevitably, high quality CFD is often
time consuming and costly. The validity of the level of expertise
Using numerical codes like Fluent can help to design and setup required and the time (cost) involved should be carefully evaluated
wind tunnel experiments; hence reducing the time required to on the basis of its purposes by comparing them with those of
optimize a physical model and expensive pre-runs in a wind other assessment methods. (Tominaga and Stathopoulos,
tunnel. With a numerical simulation critical points like source 2013)
82 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

7.4. On Large Eddy Simulation versus RANS 8. Application: CFD simulation of pedestrian-level
wind conditions around buildings
[About LES] as the model formulation increases in complexity,
the likelihood of degrading the models performance due to input High-rise buildings can introduce high wind speed at pedes-
data and model parameter uncertainty increases as well. (Hanna, trian level, which can lead to uncomfortable or even dangerous
1989) conditions. Wind discomfort and wind danger can be detrimental
to the success of new buildings. Wise (1970) reports about shops
If it turns out that LES can be done on very coarse grids, it will be that are left untenanted because of the windy environment that
one of the few times that nature has been kind to us with regard to discouraged shoppers. Lawson and Penwarden (1975) report the
turbulent ows. (Ferziger, 1990) death of two old ladies due to an unfortunate fall caused by high
wind speed at the base of a tall building. Today, many urban
authorities only grant a building permit for a new high-rise
In the event of peak wind and pressure loading having to be building after a wind comfort study has indicated that the negative
determined, a statistical framework is obviously inappropriate consequences for the pedestrian wind environment remain lim-
In this case, the only alternative route is Large Eddy Simulation. ited. Note that a wind comfort study is generally performed by a
(Leschziner, 1993) combination of three types of information/data: (1) statistical
meteorological information; (2) aerodynamic information; and
At this stage, it is clear that dynamic LES gives the best results for (3) a comfort criterion. CFD or wind-tunnel measurement data
many wind engineering applications. One disadvantage of using can be used to provide part of the aerodynamic information.
LES is that too much CPU time is required. However, rapid
evolution of CPU hardware will surely overcome this restriction, 8.1. CFD versus wind-tunnel measurements
and wide application of LES to CWE problems will certainly be
realized in the near future. (Murakami, 1997) Wind comfort studies require the knowledge of at least the
mean wind velocity vector eld at pedestrian height (z 1.75 or
2 m). This information can be obtained by wind-tunnel modelling
A major error source was found to be the stationary solution or by CFD. Wind-tunnel tests are generally point measurements
procedure that was chosen for all simulations. Since no vortex with Hot-Wire Anemometry (HWA) or Hot-Film Anemometry
shedding at the building edges is calculated less turbulent mixing (HFA) (e.g. Isyumov and Davenport, 1975; Stathopoulos and
close to the building leads to stationary high concentration areas Storms, 1986; Uematsu et al., 1992; Blocken et al., 2008b), Irwin
near the building edges. Less mixing observed for ground level probes (Irwin, 1981; Durgin 1992; Wu and Stathopoulos, 1994;
releases might also have been caused by differences in turbulent Monteiro and Viegas, 1996; Van Beeck et al., 2009; Tsang et al.,
structure close to the wall. (Leitl et al., 1997) 2012) or Laser-Doppler Anemometry (LDA) (e.g. Van Beeck et al.,
2009; Conan et al., 2012). In the past, also area techniques such as
It should be stressed however, that LES as a procedure of sand erosion (Beranek and Van Koten, 1979; Beranek, 1982, 1984;
turbulence modeling is going to be truly useful only if it reaches Livesey et al., 1990; Richards et al., 2002; Van Beeck et al., 2009;
the stage of producing peak instantaneous pressure coefcients, Conan et al., 2012) and infrared thermography (Yamada et al.,
with some reasonable accuracy. (Stathopoulos, 2002) 1996; Wu and Stathopoulos, 1997; Sasaki et al., 1997) have been
used. They are however considered less suitable to obtain accurate
quantitative information. Instead, they can be used as part of a
It is argued that RANS will further play an important role, two-step approach: rst an area technique is used to qualitatively
especially in industrial and environmental computations, and that indicate the most important problem locations, followed by
the further increase in the computing power will be used more to accurate point measurements at these most important locations
utilize advanced RANS models to shorten the design and market- (Blocken and Carmeliet, 2004b).
ing cycle rather than to yield the way to LES. (Hanjalic, 2004) One of the main advantages of CFD in pedestrian-level wind
comfort studies is avoiding this time-consuming two-step approach
The CFD techniques that will prove to be of most use will be those by providing whole-ow eld data. In spite of its deciencies,
that will faithfully model the turbulence structure within the steady RANS modelling with the k model or with other turbulence
atmospheric boundary layer, e.g. LES or DES techniques. The use of models has become the most popular CFD approach for pedestrian-
RANS based techniques will decrease over time, although their level wind studies. Two main categories of CFD studies can be
relative simplicity and economy will ensure their continued use for distinguished: (1) fundamental studies, which are typically con-
many applications. (Baker, 2007) ducted for simple, generic building congurations to obtain insight
in the ow behaviour, for parametric studies and for CFD validation,
and (2) applied studies, which provide knowledge of the wind
It should be noted that, in order to use CFD for wind load
environmental conditions in specic and often much more complex
estimation, an accurate time-dependent analysis, such as LES, is
case studies. Fundamental CFD studies beyond the case of the
denitely required, because it enables prediction of peak-type of
isolated building were performed by several authors including
quantities such as a peak pressure or maximum response of a
Baskaran and Stathopoulos (1989), Bottema (1993), To and Lam
building and a structure. Furthermore, consistency of inow
(1995), Baskaran and Kashef (1996), Yoshie et al. (2007), Blocken
turbulence characteristics for various numerical models is very
et al. (2007b, 2008b), Blocken and Carmeliet (2008), Tominaga et al.
signicant for appropriate wind load estimation. (Tamura et al.,
(2008b) and Mochida and Lun (2008). Apart from these fundamen-
2008)
tal studies, also several CFD studies of pedestrian wind conditions in
complex urban environments have been performed (Murakami,
Spatial distribution of the turbulent scalar ux inside building 1990b; Gadilhe et al., 1993; Takakura et al., 1993; Stathopoulos and
arrays has shown that inaccurate predictions of the effects of Baskaran, 1996; Baskaran and Kashef, 1996; He and Song, 1999,
intermittency are the major cause for discrepancies between RANS Ferreira et al., 2002, Richards et al., 2002, Miles and Westbury, 2002;
and experimental results (Di Sabatino et al., 2013) Westbury et al., 2002; Hirsch et al., 2002; Meroney et al., 2002;
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 83

Fig. 4. Examples of CFD studies of pedestrian-level wind conditions in urban areas: (a and b) Grid (38895 nite elements) and wind-velocity vectors based on steady RANS
simulations (Gadilhe et al., 1993 Elsevier), (c and d) Grid (total cell count unknown) and wind speed contours based on LES (He and Song, 1999, Elsevier), (e and f) Grid
(2.8 million cells) and wind speed ratio contours, based on steady RANS (Blocken and Persoon, 2009, Elsevier).

Blocken et al., 2004; Yoshie et al., 2007; Blocken and Carmeliet, can easily be compromised by the presence of a combination
2008; Blocken and Persoon, 2009; Blocken et al., 2012; Janssen et al., of numerical errors and physical modelling errors. Statements on
2013; Montazeri et al., 2013). Some of the computational grids and the accuracy of steady RANS with a certain turbulence model
some typical presentations of results of these studies are shown in should therefore be based on CFD studies that have undergone
Fig. 4. Almost all these studies focused on building groups, rather solution verication, i.e. it should be proven that numerical
than isolated buildings, which is logical because unfavourable errors are limited, so clear conclusions about the physical model-
pedestrian-level wind conditions are often the specic result of ling errors can be made. Several studies have adopted this
the interaction of the wind-ow patterns around several buildings. approach in their validation of CFD with wind-tunnel measure-
Almost all these studies were also conducted with the steady RANS ments and on-site measurements. A general observation from
approach and a version of the k- model. An exception is the study these studies is that the prediction accuracy is a pronounced
by He and Song (1999) who used LES. function of the location in the ow pattern, and therefore
of the wind direction. While several validation studies have been
8.2. Accuracy of CFD performed, at least two of those have provided conclusions
on the accuracy of steady RANS CFD that can be generalised:
Attempts to provide general statements about the accuracy of the extensive validation study by Yoshie et al. (2007) for four
steady RANS CFD for pedestrian-level wind environment studies different building and urban congurations and the validation
84 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Fig. 5. Building congurations in the validation studies by Yoshie et al. (2007), (a and b) geometry and structured grid (0.1 million cells) of isolated building, (c and d)
geometry and structured grid (1.3 million cells) of high-rise building surrounded by low-rise buildings, (e and f) geometry, immersed-boundary (0.25 million cells) and body-
tted (0.8 million cells) grids of building complex in actual urban area (Niigata), (g and h) geometry, immersed-boundary (2.95 million cells) and body-tted (1.18 million
cells) grids of building complex in actual urban area (Shinjuku, Tokyo). Courtesy of R. Yoshie and Y. Tominaga (2010).

Fig. 6. Comparison of CFD results and wind-tunnel measurements of wind speed ratio for the isolated building (see Fig. 5a) by Yoshie et al. (2007): (a) steady RANS with
standard k- model, (b) steady RANS with RNG k model. The symbols refer to: front of building; o side of building; x behind building. The different colours refer to a
variety of positions in front, beside and behind the building. Courtesy of R. Yoshie and Y. Tominaga (2010).

study by Blocken and Carmeliet (2008). These two studies are the LaunderKato k model (Kato and Launder, 1993) and the
discussed next. Renormalization Group (RNG) k model (Yakhot and Orszag,
In the framework of the development of the AIJ guideline for 1986). Comparison of the standard k model results with the
wind environment evaluation, Yoshie et al. (2007) reported wind-tunnel measurements showed that the amplication factor
validation studies for four different building and urban congura- U/U0 (which is the ratio of the local pedestrian-level wind speed U
tions (Fig. 5): (1) an isolated square prism with ratio L:W:H 1:1:2, to the wind speed U0 that would occur at the same position
(b) an idealised high-rise building surrounded by regularly spaced without buildings) is generally predicted within an accuracy of
low-rise buildings, (c) building complexes in the actual urban area 10% in the regions where U/U0 41 (see Fig. 6a). In the wake region
of Niigata, Japan, and (d) building complexes in the actual behind the building however, where U/U0 o 1, the predicted wind
Shinjuku sub-central area in Tokyo, Japan. A view of the computa- speed is generally signicantly underestimated, at some locations
tional grids is also shown in Fig. 5. In all four cases, the simulations by a factor 5 or more (Fig. 6a). The results of the other turbulence
were performed with steady RANS, combined with the standard models showed a slight improvement in the high wind-speed
k- model or with revised k models, and compared with the regions, but worse results in the wake region (Fig. 6b). The
results of wind-tunnel experiments. Note that the simulations underestimations in the wake region are attributed to the under-
included a grid-sensitivity analysis, careful application of the estimation of turbulent kinetic energy in the wake, due to the fact
boundary conditions, higher-order discretisation schemes, a com- that steady RANS is not capable of reproducing the vortex shed-
plete report of the computational settings and parameters and a ding in the wake of buildings (Yoshie et al., 2007; Tominaga
detailed comparison with the wind-tunnel measurements, all of et al., 2008b).
which are required in order to support the validity of the The simulations for the idealised high-rise building surrounded
conclusions. The simulations for the isolated building were made by low-rise buildings were made with the standard k model and
with the standard k- model and with two revised k models: the RNG k model. In the high wind-speed regions, the standard
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 85

Fig. 7. Validation study for parallel building congurations by Blocken and Carmeliet (2008): (a) sand-erosion contour plots of the amplication factor U/U0, (b) CFD results
for U/U0 (1.5 million cells); (c) sand erosion plots of U/U0, (d) CFD results for U/U0 (0.7 million cells).

k model underestimated the wind tunnel results by about 15%. regions of high U/U0 are the corner streams and the areas between
In the lower wind speed regions, differences up to a factor 4 were the buildings in which pressure short-circuiting occurs (Blocken
found. The results of the RNG k model showed improved and Carmeliet, 2008). Other results from the same study (not
performance in the high wind speed regions, but again a deterio- shown in Fig. 7) indicate that also the high U/U0 in the standing
rated performance in the lower wind speed regions. Similar vortex is predicted with good accuracy by steady RANS CFD. Note
conclusions on the different performance in high versus low wind that the standing vortex is only clearly visible for wind directions
speed regions were found in the CFD study for the actual urban that are almost perpendicular to the long building facade. Regions
area in Niigata: in high wind speed regions, the predictions are of low U/U0 do not only occur in the wake of the buildings, but
generally within 20% of the measurements, while the wind speed are also found in the low-speed stagnation zone upstream of the
in low wind speed regions is generally signicantly underesti- buildings. Similar to the results by Yoshie et al. (2007), the
mated, at some positions with a factor 5 or more. The comparisons underestimations in these regions can go up to a factor 5 or more.
for the fourth conguration, the Shinjuku sub-central area, con- Note that also these simulations were based on grid-sensitivity
rmed the ndings for the other congurations. While for all four analysis, careful application of the boundary conditions and higher
studies, large discrepancies are found in the low wind speed order discretisation schemes. It should be noted that sand-erosion
regions, it should be noted that the high wind speed regions are measurement results are generally considered to be less suitable
those of interest for pedestrian-level wind studies. In these for CFD validation, although in this study a very close agreement
regions, steady RANS was shown to provide a good to very good both qualitatively and quantitatively was found in the high U/U0
accuracy (1020%). region.
Blocken and Carmeliet (2008) performed steady RANS CFD For assessing the accuracy of CFD for pedestrian-level wind
simulations with the realisable k model (Shih et al., 1995) for studies, it is important to compare them not only with wind-
three congurations of parallel buildings and compared the results tunnel measurements where the boundary conditions are
with the sand-erosion wind-tunnel experiments by Beranek generally well-known but also with on-site measurements.
(1982). Two of these comparisons are shown in Fig. 7, yielding However, CFD pedestrian-level wind studies in complex urban
observations that are very similar to those by Yoshie et al. (2007): environments including a comparison with on-site measurements
a close to very close agreement between CFD and wind-tunnel are very scarce. To the knowledge of the author, only four such
measurements in the region of high U/U0 (about 10% accuracy) and studies have been published: the study by Yoshie et al. (2007) for
signicant underestimations in the regions of lower U/U0. The the Shinjuku Sub-central area in Tokyo (Fig. 5g and h), the study by
86 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Blocken and Persoon (2009) for the area around the multifunc- 9. Application: CFD simulation of natural ventilation
tional ArenA stadium in Amsterdam (Fig. 4e and f) and the studies of buildings
by Blocken et al. (2012) and Janssen et al. (2013) for the Eindhoven
University campus. Although these measurements were quite Natural ventilation is an important factor in the development of
limited, overall, the comparisons conrmed the conclusions made sustainable and healthy indoor environments (e.g. Finnegan et al.,
earlier, albeit that the discrepancies in the high wind speed 1984; da Graa et al., 2002; Chang, 2006; Chen et al., 2007; Chen,
regions can slightly exceed 10%. 2009; Tablada et al., 2009; Heiselberg and Perino, 2010; van Hooff
and Blocken, 2010a, 2012, 2013). It is driven by wind or buoyancy,
8.3. Practical applicability or most often by a combination of both (e.g. Linden, 1999; Hunt
and Linden, 1999; Li and Delsante, 2001; Reichrath and Davies,
In spite of the very limited number of validation studies based 2002; Tan and Glicksman, 2005; Larsen and Heiselberg, 2008). In
on on-site measurements, CFD is gaining increasing acceptance as the past decades, a lot of research efforts have contributed to the
a tool for pedestrian-level wind studies. This has been conrmed evaluation of the natural ventilation performance of buildings. A
by the publication of the Dutch Wind Nuisance Standard, NEN8100 comprehensive review on methods for ventilation performance
(Wisse and Willemsen, 2003; NEN, 2006a, 2006b; Willemsen and prediction for buildings was provided by Chen (2009). Other
Wisse, 2007) that specically allows the user to choose between reviews were provided by, among others, Etheridge and
wind-tunnel testing and CFD for obtaining the aerodynamic part Sandberg (1996), Reichrath and Davies (2002), Awbi (2003),
in the pedestrian-level wind comfort and wind safety assessment Karava et al. (2004, 2006), Norton and Sun (2006), Norton et al.
procedure. To the best of the author's knowledge, this standard is (2007), Etheridge (2011), Ramponi and Blocken (2012a) and Bjerg
the rst wind nuisance standard in the world. In addition, the et al. (2013a, 2013b).
option for the user to choose between wind-tunnel testing and
CFD can be considered as a milestone in the acceptance process of 9.1. CFD versus measurements
CFD as a tool for the evaluation of wind comfort and wind safety in
urban areas. However, it does not absolve the user from providing Ventilation performance can be assessed by on-site and
quality assurance. The decision to treat wind-tunnel experimenta- reduced-scale measurements, analytical and/or semi-empirical
tion and CFD as equals in the Dutch standard has not been made formulae, simulations with zonal and multi-zone network models
lightly and has indeed led to the specication of quality assurance and CFD (Etheridge and Sandberg, 1996; Awbi, 2003; Chen, 2009;
requirements in the standard, both for CFD and for wind-tunnel Etheridge, 2011). As opposed to on-site measurements, reduced-
testing. This reinforces the importance of best practice guidelines scale wind-tunnel measurements offer the advantage that the
and their integration in wind comfort and wind safety studies. CFD boundary conditions can be carefully controlled. But wind-tunnel
solution verication and validation and complete reporting of the measurements also exhibit some particular problems, including
followed procedure are essential components of quality assurance. incompatible similarity requirements especially when buoyancy is
In practical situations and in case of complex urban environments, involved, and the scaling of the ventilation openings. Indeed, a
when measurements are often not available, CFD model validation specic problematic aspect of reduced-scale wind-tunnel mea-
should be performed for simpler congurations, the ow features surements in natural ventilation is the scaling of the ventilation
of which show resemblance with those expected in the actual openings. It has been shown that the presence of surroundings
complex urban conguration (Oberkampf et al., 2004; Blocken et buildings is very important to determine the natural ventilation of
al., 2004, 2012; Franke et al., 2007; Yoshie et al., 2007; Blocken and a given building (e.g. van Hooff and Blocken, 2010b). This however
Carmeliet, 2008; Tominaga et al., 2008a). For these simpler cases, imposes the need to include quite a large part of the urban
wind-tunnel measurement data are generally available in the surroundings in the wind-tunnel model, which in turns requires
literature. quite large scaling factors to be able to t this urban model into the
Steady RANS is the most commonly used method, while LES is wind-tunnel test section. As a result, the ventilation openings can
still considered out of reach for practical pedestrian-level wind become so small that similarity is violated: the ow in these
studies in actual urban environments (Yoshie et al., 2007). This is openings might change from turbulent at full scale to transitional
mainly attributed to the much larger computational cost of LES but or even laminar at reduced scale. In addition, the narrow ventila-
also to the lack of best practice guidelines for LES. Indeed, as tion openings in the scaled model might become impractically
mentioned in Section 3, most best practice guidelines focus on small (and nearly impossible to manufacture). CFD therefore has
RANS. Concerning the computational cost, for pedestrian-level wind some clear advantages compared with on-site and reduced-scale
studies, simulations need to be performed for many (e.g. 12 or 16) measurements, the most important of which are that it provides
wind directions, and this needs to be repeated for congurations whole-ow eld data and that it avoids potentially incompatible
with remedial measures implemented (Yoshie et al., 2007). Never- similarity requirements because the simulations can be performed
theless, it is expected that the increase in computing power and at full scale. In addition, CFD allows full control over the boundary
speed together with the intrinsically superior potential of LES will conditions and easily and efciently allows parametric studies to
render it increasingly more attractive in the years to come. While be performed to evaluate alternative design congurations. The
LES should be preferred over RANS for several CWE applications latter is especially an advantage when the different congurations
(especially wind loads), it is not yet fully clear whether this is the are all a priori embedded within the same computational domain
case for the practical assessment of pedestrian-level wind comfort and grid (see e.g. van Hooff and Blocken, 2010a). Chen (2009)
and wind safety. There are quite some indications from recent additionally mentions that CFD models are currently most popular
studies supported by extensive solution verication and validation and particularly suited for studying indoor air quality and natural
actions that steady RANS has a fairly high accuracy in predicting the ventilation, as these are difcult to predict with other models. For
mean wind speed at least for regions with high amplication factor these reasons, many studies on evaluating and optimising the
(Yoshie et al., 2007; Blocken and Carmeliet, 2008; Blocken et al., natural ventilation potential of buildings have employed CFD.
2011a; Blocken et al., 2012; Janssen et al., 2013). It might turn out Two main categories of CFD studies can be distinguished:
that using the words by Ferziger (1990), pedestrian-level wind (1) fundamental studies, which are typically conducted for simple,
comfort is one of the few topics in CWE where nature is kind to us generic building congurations to obtain insight in the ow
concerning turbulent ows. behaviour, for parametric studies and for CFD validation, and
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 87

(2) applied studies, which provide knowledge of the natural induced cross-ventilation has applied the coupled approach
ventilation in specic and often much more complex case studies. (Ramponi and Blocken, 2012a).
In the remainder of this section, only natural cross-ventilation The main reason for the extensive use of the coupled approach
studies will be addressed. Fundamental studies on wind-induced is the knowledge that, in case of large ventilation openings, the
cross-ventilation were performed by e.g. Kato et al. (1992), Straw decoupled approach can introduce important errors. Indeed, the
et al. (2000), Kurabuchi et al. (2000), Bartzanas et al. (2002), so-called sealed-body assumption in the decoupled approach
Mistriotis and Briassoulis (2002), Shklyar and Arbel (2004), Hu implies that the pressure distribution on the building envelope is
et al. (2005, 2008), Lee et al. (2005), Evola and Popov (2006), not affected by the presence of the openings (e.g. Murakami et al.,
Chang (2006), Wright and Hargreaves (2006), Asfour and Gadi 1991; Kato et al., 1992; Karava et al., 2007). It assumes that the
(2007), Wang and Wong (2008, 2009), Kobayashi et al. (2009, turbulent kinetic energy is dissipated at the windward opening
2010), Meroney (2009), Norton et al. (2010a), Nikas et al. (2010), and that the effect of the dynamic pressure on the air ow passing
Larsen et al. (2011), Cheung and Liu (2011), Ramponi and Blocken through the opening is negligible (Murakami et al., 1991;
(2012a, 2012b), Shen et al. (2012), Stavridou and Prinos (2013), Etheridge and Sandberg, 1996; Seifert et al., 2006; Karava et al.,
Bangalee et al. (2013), Chu and Chiang (2013). CFD studies of 2007, 2011; Kobayashi et al., 2010). However, Murakami et al.
wind-induced cross-ventilation for specic buildings including (1991), Kato et al. (1992), Sandberg (2004), Karava et al. (2006,
those in complex urban environments were performed by e.g. 2011) and Kobayashi et al. (2009, 2010) correctly pointed out that
Mistriotis et al. (1997a, 1997b), Jiang and Chen (2002), da Graa in case of wind ow through large ventilation openings, the
et al. (2002), Bartzanas et al. (2004, 2007), Tan and Glicksman turbulent kinetic energy is rather preserved and the sealed-body
(2005), Mochida et al. (2005, 2006), Fatnassi et al. (2006), assumption is therefore no longer valid. A virtual stream-tube
Stavrakakis et al. (2008), Horan and Finn (2008), Teitel et al. model was introduced to explain the direct connection between
(2008), Wang and Wong (2008), Norton et al. (2009, 2010a, the inlet and outlet openings (Murakami et al. 1991; Kato et al.,
2010b), van Hooff and Blocken (2010a, 2010b, 2013), Wu et al. 1992; Sandberg, 2004; Kobayashi et al., 2009, 2010).
(2011), Wu et al. (2012), Lo et al. (2013) and Hajdukiewicz et al. Many CFD studies of natural cross-ventilation have included a
(2013). A specic type of natural ventilation studies are those comparison with either wind-tunnel measurements or on-site
concerning specially designed roof constructions to drive natural measurements, or both. Given the importance of solution verica-
ventilation of the building zones (e.g. Montazeri et al., 2010; tion including grid-convergence analysis, this paper will only
Montazeri, 2011; van Hooff et al., 2011a; Blocken et al., 2011b). highlight a very small sample of the RANS studies that included
Some of the computational grids and some typical presentations of a grid-convergence analysis, as well as a particular study per-
results from CFD studies of natural cross-ventilation are shown in formed with LES.
Fig. 8. The large majority of CFD studies of natural cross- Bartzanas et al. (2004) applied 3D steady RANS with the
ventilation focused on an isolated building rather than on building standard k model to analyse the effect of the ventilation
groups. This is to some extent surprising given the potentially conguration of a tunnel greenhouse with crop on airow and
large impact of surrounding buildings (see e.g. van Hooff and temperature patterns. The simulations were based on grid-
Blocken, 2010b). Most CFD studies of natural cross-ventilation sensitivity analysis yielding a grid with 76,800 cells. Validation
were conducted with the steady RANS approach. Some exceptions was performed with on-site measurements of local velocity and
are the studies by Kato et al. (1992), Kurabuchi et al. (2000), Jiang ventilation rate data from N2O tracer gas measurements. For wind
and Chen (2002), Hu et al. (2005, 2008), Meroney (2009) and Chu direction parallel to the ridge of the tunnel greenhouse, deviations
and Chiang (2013), who used LES, and the studies by Wright and in local wind speed varied between 10% and 35%, while deviations
Hargreaves (2006) and Meroney (2009) who used Detached Eddy in ventilation rate were only 1215%, which is considered a very
Simulation (DES). good agreement for this rather complex study.
Evola and Popov (2006) applied 3D steady RANS with the
9.2. Accuracy of CFD standard and RNG k model to analyse cross-ventilation for the
isolated cubic building model that was previously studied by Jiang
In CFD simulations of cross-ventilation involving large open- et al. (2003) (Fig. 10a). The grid-sensitivity analysis yielded a grid
ings, a major issue of concern is the accurate modelling of the of 676,000 control volumes. The wind direction was perpendicular
interaction between the outdoor wind ow around the buildings to the ventilation openings. The CFD results were compared with
and the indoor air ow inside the buildings, which interact with the wind-tunnel measurements by Jiang et al. (2003) along
each other at the ventilation openings. A distinction can be made 5 vertical lines (Fig. 10b). The results in Fig. 10c indicate that the
between a coupled and a decoupled approach. In the coupled RNG k- model outperforms the standard k- model at some
approach, there is a single computational geometry and computa- locations, while the opposite is true at other locations. Comparing
tional domain, that includes both the outside and the inside the ventilation rates obtained by both models with the LES results
environment of the building (Fig. 9a). In this approach, the by Jiang et al. (2003), the deviation by the standard k model is
ventilation openings are considered open, the outdoor wind ow 9%, while that for the RNG k model is only 3%.
and indoor air ow are solved within the same computational Meroney (2009) was the rst to perform a very extensive CFD
domain and the interaction (coupling) between the outdoor wind study of cross-ventilation for the isolated building model that was
ow and indoor air ow is resolved in detail using the appropriate experimentally studied by Karava et al. (2011) and Karava and
governing equations. Contrary to this, in the decoupled approach, Stathopoulos (2012). This study was published prior to the actual
there are two different computational geometries and two differ- publication of the experimental data. It included a detailed
ent computational domains: one for the outdoor environment and evaluation of steady RANS with the standard k model, the
one for the indoor environment of the building (Fig. 9b). In this realisable k model, the RNG k model, the standard k model
approach, the wind ow simulation is conducted for the building (Wilcox, 1998) and the Reynolds Stress Model (RSM) (Launder
as a sealed body, i.e. the openings are closed. This simulation et al., 1975), Detached-Eddy Simulation (DES) and Large-Eddy
yields the pressure coefcients at the positions of the openings Simulation (LES). A remarkable result from this study was that
and these coefcients are subsequently used as boundary condi- despite the obvious transient nature of separation and reattach-
tions for the CFD simulation of the indoor air ow. A review of the ment ows some of the RANS models performed just as well as the
literature indicates that, by far, most CFD research on wind- LES or DES models.
88 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Fig. 8. Examples of CFD studies of wind-induced cross-ventilation of buildings. (a) Isolated generic building model; (b) corresponding mean velocity vector eld in vertical
plane and (c) in horizontal plane as a result of LES simulations (Kato et al., 1992, Elsevier). (d) Computational domain; (e) horizontal distribution of velocity vectors at 4.5 m
height in Case1 (2:00 p.m.); (f) horizontal distribution of velocity vectors (zone A, at 4.5 m height) (Mochida et al., 2005, Elsevier). (g) Building geometry and dimensions
(normalised by building height); (h) velocity vector eld in vertical centreplane obtained by LES and wind-tunnel experiments; (i) same for turbulent kinetic energy (Hu
et al., 2008, Elsevier). (j) Computational grid for Amsterdam ArenA stadium (5.6 million control volumes); (k) inside view of computational grid; (l) distribution of CO2
concentration in vertical centreplane (van Hooff and Blocken 2010a, 2013).
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 89

Fig. 9. Schematic representation of (a) coupled and (b) decoupled approach for analysis of wind-induced cross-ventilation of buildings (from Ramponi and Blocken, 2012a,
Elsevier).

Ramponi and Blocken (2012a) applied 3D steady RANS for the in building groups and in complex urban environments. Such
isolated building by Karava et al. (2011) and Karava and studies however are scarce. Some exceptions are the study by Jiang
Stathopoulos (2012) (Fig. 11a). The turbulence models included and Chen (2002) that was mentioned above and the studies by
the standard k model, the realisable k model, the RNG k Mochida et al. (2005, 2006), Fatnassi et al. (2006), Teitel et al.
model, the standard k model, the shear-stress transport (SST) (2008) and van Hooff and Blocken (2010a, 2010b, 2013). However,
k model (Menter, 1994) and the Reynolds Stress Model (RSM). many more of these studies are needed to arrive at denite
The study also investigated the impact of various other computa- conclusions concerning the accuracy of CFD for natural cross-
tional parameters on the accuracy of the outdoor and indoor mean ventilation.
velocity ratios, including the size of the computational domain, the
resolution of the computational grid, the inlet turbulent kinetic 9.3. Practical applicability
energy prole of the atmospheric boundary layer, the order of the
discretisation schemes and the iterative convergence criteria. The The application of CFD for natural ventilation studies of build-
grid-sensitivity analysis yielded a grid with 575,247 control ings in practice and especially for buildings situated in complex
volumes (Fig. 11b). The best agreement with the PIV wind-tunnel urban environments has up to now remained rather limited. On
measurements by Karava et al. (2011) was obtained by the SST k the one hand, this is surprising because CFD provides some clear
model (Fig. 11c and d) followed by the RNG k model. The other practical advantages compared to wind-tunnel testing. It does not
models were insufciently capable of reproducing the magnitude suffer from potentially incompatible similarity requirements
and position of the standing vortex upstream of the building related to buoyancy effects and scaling of the ventilation openings
facade, and of the resulting direction of the jet through the because the simulations can be conducted at full scale. In addition,
ventilation opening. Apart from the area close to the ventilation the coupled approach in CFD allows resolving the interaction
openings, where the accuracy of the PIV measurements suffered between the outdoor wind ow and the indoor airow. Also,
from reections, the differences between measurements and CFD allows taking into account the natural variability in wind
simulations are generally less than 20%, and less than 10% in the directions, which was shown to be important by Jiang and Chen
central part of the building. (2002). On the other hand, it is not so surprising. For natural
Jiang and Chen (2002) compared results from LES with the ventilation studies with the coupled approach in complex urban
Smagorinsky subgrid-scale model with experimental data from environments, it is important to accurately resolve the wind ow
Katayama et al. (1992), who performed on-site measurements and in the urban area, the ow through the ventilation openings and
wind-tunnel tests for both outdoor and indoor airows on a the indoor airow. The large differences in spatial scales (from the
building site (Fig. 12a and b). In particular, Jiang and Chen (2002) extent of the urban area down to the size of the ventilation
focused on the effect of wind-direction uctuations on the wind- openings) puts quite high demands on the development of a high-
velocity patterns. Indeed, the natural wind is highly variable in resolution and high-quality computational grid, which in turn
both speed and direction, which cannot easily be generated in a translates into relatively large numbers of cells and therefore a
wind tunnel. In a conventional wind tunnel, the wind direction is high computational cost.
xed. Therefore using a tunnel-generated wind to simulate real Also for natural ventilation, solution verication and validation
wind may cause signicant errors. Jiang and Chen (2002) success- of the CFD simulations are indispensable. Although many solution
fully demonstrated that LES can simulate both cases by xing or verication and validation studies have been performed for iso-
changing the incoming wind direction. For the wind-tunnel case, lated buildings, there is a clear lack of such studies for buildings in
LES with a constant wind direction was applied. For the on-site generic or more complex (real) urban areas. As a result, there is
case, LES simulated the wind with either a normal or a uniform also a lack of specic guidelines for such simulations. This imposes
distribution depending on the magnitude of the incoming wind the need for detailed solution verication and validation for every
speed. The results in Fig. 12c and d show that the wind-tunnel data new case study. In practical situations and in case of complex
present a deep, thin and high velocity core in the north room urban environments, when on-site measurements or wind-tunnel
(upper room) of building A and that the LES results with xed measurements are often not available and/or not possible, CFD
wind direction show a similar ow pattern. The results in Fig. 12e model validation should be performed for simpler congurations,
and f show that the on-site measurements present a shallower and the ow features of which show resemblance with those expected
wider high-speed region in both rooms which is clearly repro- in the actual practical or complex urban conguration. For these
duced by LES with the varied wind direction. simpler cases, wind-tunnel measurement data should be collected
For assessing the accuracy of CFD for natural cross-ventilation to allow this validation.
studies, it is important to not only perform solution verication Steady RANS is by far the commonly used method. This is
and validation studies for isolated buildings, but also for buildings mainly attributed to the much larger computational cost of LES but
90 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Fig. 10. (a) Building geometry with dimensions in mm (modied from Evola and Popov, 2006, Elsevier); (b) sections AE along which experimental and numerical results
are compared; (c) vertical proles of ratio of horizontal velocity component to reference wind speed: dots are wind-tunnel results, solid lines are results from RNG k
model, dashed lines results from standard k model. Dimensions on vertical axis in m (from Evola and Popov, 2006, Elsevier).

also to the lack of best practice guidelines for LES. Nevertheless, it steady RANS simulations can be sufcient, and in which cases the
is expected that the increase in computing power and speed more complex LES approach should be applied. This is important
together with the intrinsically superior potential of LES will render because, as stated by Hanna (1989): as the model formulation
it increasingly more attractive in the future. Irrespective of this increases in complexity, the likelihood of degrading the models
expected development, future detailed validation studies for performance due to input data and model parameter uncertainty
practical and complex situations should indicate in which cases increases as well.
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 91

Fig. 11. (a) Building geometry and indication of vertical measurement plane; (b) computational grid (575,247 cells); (c) comparison of PIV and CFD (SST k) velocity vector
elds in vertical centreplane and horizontal plane at mid-height through the openings; (d) comparison of streamwise wind speed ratio U/Uref from PIV and CFD (SST k)
along centreline (Ramponi and Blocken, 2012a, Elsevier).

10. Application: CFD simulation of wind-driven rain suffer from large errors (Hgberg et al., 1999; van Mook, 2002;
on building facades Blocken and Carmeliet, 2005, 2006a). Recently, guidelines that
should be followed for selecting accurate and reliable WDR data
Wind-driven rain (WDR) is one of the most important moisture from experimental WDR datasets have been proposed (Blocken
sources affecting the hygrothermal performance and durability of and Carmeliet, 2005, 2006a). The strict character of these guide-
building facades. Consequences of its destructive properties can lines, however, implies that only very few rain events in a WDR
take many forms. Moisture accumulation in porous materials can dataset are accurate and reliable and hence suitable for WDR
lead to rain water penetration, frost damage, moisture induced salt studies. Other drawbacks of WDR measurements are the fact that
migration, discolouration by eforescence, structural cracking due they are time-consuming and the fact that measurements on a
to thermal and moisture gradients, to mention just a few. WDR particular building site have very limited application to other sites.
impact and runoff is also responsible for the appearance of surface These limitations drove researchers to establish semi-empirical
soiling patterns on facades that have become characteristic for so relationships between the intensity of WDR and the inuencing
many of our buildings. Assessing the intensity of WDR on building meteorological parameters wind speed, wind direction and hor-
facades is complex, because it is inuenced by a wide range of izontal rainfall intensity (i.e., the rainfall intensity through a
parameters: building geometry, environment topography, position horizontal plane, as measured by a traditional rain gauge). The
on the building facade, wind speed, wind direction, turbulence advantage of semi-empirical methods is their ease-of-use; their
intensity, rainfall intensity and raindrop-size distribution (e.g. main disadvantage is that only rough estimates of the WDR
Choi, 1993, 1994a, 1994b; Etyemezian et al., 2000; Blocken and exposure can be obtained (Blocken and Carmeliet, 2004a, 2010).
Carmeliet, 2002, 2004a; Tang and Davidson, 2004). Given the drawbacks associated with measurements and semi-
empirical methods, researchers realised that further achievements
10.1. CFD versus measurements and semi-empirical formulae were to be found by employing numerical methods. CFD can be a
valuable alternative to avoid time-consuming and expensive on-
Three categories of methods exist for the assessment of WDR site or reduced-scale experiments, and provide more detailed and
on building facades: measurements, semi-empirical methods and accurate information than (semi-)empirical formulae (Blocken
numerical methods based on Computational Fluid Dynamics et al., 2010, 2011c). To the knowledge of the author, the rst CFD
(CFD). An extensive literature review of each of these categories simulations of WDR on buildings were made by Souster (1979).
has been provided by Blocken and Carmeliet (2004a). Measure- Choi (1991, 1993, 1994a, 1994b) developed and applied a steady-
ments have always been the primary tool in WDR research, state simulation technique for WDR. It consists of solving the
although a systematic experimental approach in WDR assessment wind-ow pattern and calculating the trajectories of raindrops in
is not feasible. Different reasons are responsible for this, the most this pattern by solving their equation of motion (Lagrangian
important of which is the fact that WDR measurements can easily particle tracking). This technique allows determining the spatial
92 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Fig. 12. (a) Layout of building group with indication of building A; (b) dimensions of building A containing the two rooms under study; (cf) contours of wind speed (m/s)
obtained from (c) wind-tunnel measurements; (d) LES with xed wind direction; (e) on-site measurements; (f) LES with varied wind direction (Jiang and Chen, 2002,

Elsevier).

distribution of WDR on building facades for given (xed) values of (2009), Huang and Li (2010) and Kubilay et al. (2013). While some
the wind speed, the wind direction and the horizontal rainfall authors found signicant discrepancies between simulations and
intensity. Later, Choi's simulation technique was extended into the measurements, others indicated a fair to good agreement. Three
time domain by Blocken and Carmeliet (2002, 2007a). Choi's examples are given below.
technique (with and without the extension) has been applied by Validation studies of CFD simulations of WDR for a low-rise
many researchers to assess the WDR exposure of building facades building were rst performed by Blocken and Carmeliet in 2002
(e.g. Lakehal et al., 1995; Choi, 1997; Hangan, 1999, Etyemezian and later extended by the same authors in 2006 and 2007
et al., 2000; van Mook, 2002; Blocken et al., 2002, 2004a, 2006b, (Blocken and Carmeliet, 2002, 2006b, 2007b). WDR measurements
2007b, 2010, 2011c; Tang and Davidson, 2004). A particular were made at 9 positions on the facade of the low-rise VLIET test
application of CFD WDR simulations for buildings concerns the building during 19971999 and at 24 facade positions during 2002
wetting of stands in football stadia (Persoon et al., 2008; van Hooff (Fig. 13a). Fig. 13b illustrates contours of the catch ratio on the
et al., 2011b). The large majority of CFD WDR studies focused on south-west facade after a rain event with south-west wind
isolated buildings rather than on building groups. Most CFD WDR direction. The catch ratio is the ratio of the WDR sum at a certain
studies were based on the steady RANS approach with a k position at the facade to the horizontal rainfall sum measured by a
turbulence model to provide closure. traditional rain gauge (i.e. the unobstructed rainfall sum falling on
the ground). The three separate validation studies in (Blocken and
10.2. Accuracy of CFD Carmeliet 2007b) indicate deviations between CFD results and
measurements that are 20% on average, but that can locally go up
Although validation is an essential part of CFD WDR simula- to 50% and more. Considering the complexity of turbulent wind
tions, up to now, only a few validation attempts have been made. ow around a building and WDR deposition on building surfaces,
Hangan (1999) compared his CFD simulations with the WDR 20% is considered very good agreement. Several remarks are made
wind-tunnel tests by Inculet and Surry (1994). CFD validation here: (1) the CFD simulations were made on grids based on grid-
with on-site full-scale WDR measurements was performed by van sensitivity analysis, with second-order discretisation schemes and
Mook (2002), Blocken and Carmeliet (2002, 2004a, 2006b, 2007b), specic care was given to specication of the boundary conditions;
Tang and Davidson (2004), Abuku et al. (2009), Briggen et al. (2) the measurement data for validation were carefully selected to
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 93

Fig. 13. CFD validation studies of WDR on building facades. (a) VLIET test building in Leuven, Belgium (Blocken and Carmeliet, 2005, Elsevier); (b) catch ratio contours on
south-west facade after rain event with south-west wind direction (Blocken and Carmeliet, 2007b, Elsevier); (c) Cathedral of Learning. (d) Cathedral of learning with
surface soiling patterns (courtesy of C. Bailey 2010); (e) catch ratios on south-west facade for different reference wind speeds (Tang and Davidson, 2004, Elsevier);
(f) Hunting Lodge St. Hubertus in the Netherlands, with indication of moisture-related damage (Briggen et al., 2009, Elsevier); (g) comparison of measured (left) and
simulated (right) catch ratios at the end of a rain event with wind direction perpendicular to the facade (Briggen et al., 2009, Elsevier).

minimise measurement errors; (3) a good to very good qualitative Blocken and Carmeliet (2007b) can be attributed to the larger
agreement (wetting patterns) was obtained; (4) the error percen- geometrical complexity of the building and its high-rise character,
tages mentioned do not include the values on the west corner of as will be explained below. Fig. 13e shows the catch ratio
the building that was in reality inuenced by a row of trees that distribution for different reference wind speeds. For this type of
was not included in the CFD model. building, 25% is considered a very good agreement.
Tang et al. (2004) and Tang and Davidson (2004) performed Briggen et al. (2009) conducted WDR measurements and CFD
measurements and CFD simulations of WDR on the facades of the simulations for the south-west facade of the monumental building
Cathedral of Learning in Pittsburg, US, to explain the surface Hunting Lodge Saint Hubertus in the Netherlands, to provide the
soiling patterns on the facades (Fig. 13c and d). WDR measure- boundary conditions for numerical BE-HAM transfer models to
ments were made at 16 locations for a period of 21 months. The analyse the moisture related damage to the facades (Fig. 13f). The
CFD simulations were performed using the extended simulation grid was based on grid-sensitivity analysis, specic care was given
method by Blocken and Carmeliet (2002). The deviations were on to the boundary conditions and the measurement data for valida-
average 25%. The higher deviations compared to the study by tion were carefully selected following the guidelines by Blocken
94 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

and Carmeliet (2005). In spite of these efforts, very large discre- 11. Summary and future perspectives
pancies were found at the lower part of the south-west facade (up
to more than a factor 2), while a fair to good agreement was found In the past 50 years, Computational Wind Engineering (CWE)
at the upper part (20% on average). One set of results is shown in has undergone a successful transition from an emerging eld into
Fig. 13g. The most likely reason for these discrepancies is the role an increasingly established eld in wind engineering research,
of the turbulent dispersion of raindrops, which was neglected in practice and education. This paper addressed three key illustra-
these studies. The effect of turbulent dispersion can be very tions of the success of CWE: (1) the establishment of CWE as an
different depending on the building geometry and the position individual research and application area in wind engineering with
on the building (Briggen et al., 2009). It can be especially its own successful conference series under the umbrella of the
important for the bottom part of high-rise buildings and when International Association of Wind Engineering (IAWE); (2) the
the reference wind speed is low. The reason is that in this case, increasing range of topics covered in CWE; and (3) the history of
the raindrop trajectories (without turbulent dispersion) close to related review papers published in scientic journals.
the windward facade are almost vertical and parallel to the bottom CWE and CFD offer some particular advantages compared with
part of the windward facade, and do not always impinge on the on-site measurements and reduced-scale wind tunnel measure-
facade. Turbulent dispersion in the streamwise direction can cause ments. They can provide detailed information on the relevant ow
these raindrops to deviate from their mean trajectory and to hit variables in the whole calculation domain (whole-ow eld
the facade anyway. This means that, when including turbulent data), under well-controlled conditions and without similarity
dispersion, more rain will impinge on the lower part of the facade constraints. However, the accuracy and reliability of CFD are of
in reality than calculated with the CFD model. This statement is concern and solution verication and validation studies are
corroborated by an earlier study by Lakehal et al. (1995) who imperative. This requires high-quality full-scale or reduced-scale
found that turbulent dispersion is an important factor increasing measurements, which in turn should satisfy important quality
WDR on vertical walls in cases with weak upstream wind ow, criteria. In addition, it is widely recognised that the results of
such as in a street canyon. CFD simulations can be very sensitive to the wide range of
For assessing the accuracy of CFD for WDR studies, it is computational parameters that have to be set by the user. This
important to not only perform validation studies for isolated expresses the need for best practice guidelines for CWE.
buildings, but also for buildings in building groups and in complex CWE is complementary to other, more traditional areas of wind
urban environments. Such studies however are very scarce, but engineering, such as full-scale on-site experimentation and
they are needed to arrive at denite conclusions concerning the reduced-scale wind-tunnel testing. But it cannot replace them.
accuracy of CFD for WDR. Although several claims concerning the establishment of the
numerical wind tunnel have been made, mainly by non-wind
10.3. Practical applicability engineers, the CWE community has systematically and throughout
the past decades, up to the present day, denounced this label and
In spite of quite some research efforts, the application of CFD has continued to warn that many CWE problems are too complex
for WDR studies in practice has up to now remained very limited. to be tackled by CFD alone. Instead, exploiting the synergy
A few authors provided specic guidelines for CFD WDR simula- between experiments and CFD simulations is promoted.
tion (Choi, 1994a, 1994b; Blocken and Carmeliet, 2002, 2004a, Care for high quality and reliability of CFD simulations is
2006b; Briggen et al., 2009). It should be noted that the guidelines crucial. Poor quality in CWE can easily spread, contaminate and
mentioned in Section 4 also apply for CFD WDR studies, as damage the eld the further development of CWE requires the
accurate calculation of the wind-ow pattern is the rst step for opposite. In a personal communication with Robert N. Meroney,
successful WDR simulations. There are two main reasons for the Emeritus Professor of Civil and Environmental Engineering at
current limited practical use of CFD for WDR studies: (1) the very Colorado State University, USA, he expressed his important con-
time-consuming character of Lagrangian particle tracking of rain- cerns based on his 50 years of expertise and experience in
drops, in which the entire building facade needs to be covered by a computational and experimental wind engineering as follows:
large number of raindrops. Lagrangian particle tracking implies
solving the equation of motion of individual raindrops within the Many go directly to using commercial CFD codes without ever
wind-ow eld. Note that this wind-ow eld is generally taking a course in fundamental numerical methods. This has and
obtained with an Eulerian approach, i.e. not focusing on individual will continue to lead to inappropriate designs and decisions. A
particles but on xed positions in space. Lagrangian tracking needs similar situation occurred during the 1960s when D. Bruce Turner
to be performed for a large number of combinations of reference published his famous Workbook of Atmospheric Dispersion
wind speed, wind direction and raindrop diameter. (2) The fact Estimates, EPA, 1969. He was horried to nd that many
that steady RANS generally does not allow accurate modelling of engineering rms were handing the workbook to young hirees,
turbulence elds around buildings, and therefore also not of and telling them to apply it to stack plume calculations to real
turbulent dispersion of raindrops, which is important for calculat- situations. The result was that the methodology was applied
ing WDR intensities at the lower part of high-rise building facades. frequently outside the box of validation or intention. Often the
Accurate turbulent dispersion modelling would require transient applications were so distorted from the intended use that the
simulations with LES or hybrid URANS/LES, which would require results had no relation to reality. He told me personally that he
even more intensive Lagrangian particle tracking efforts. To wished he never wrote the document. (Meroney, 2014)
alleviate these problems, it might be necessary to abandon the
traditional Eulerian-Lagrangian framework in CFD WDR simula- Indeed, the increasing availability of (commercial) CFD codes
tions, and to resort to Eulerian-Eulerian modelling instead, in with user-friendly graphical user interfaces and the related
which not only the wind-ow pattern, but also the WDR inten- increased accessibility combined with the increasing availability
sities are computed with an Eulerian approach. This implies that of computational power has strongly increased the use of CFD in
the rain phase, like the air phase, is treated as a continuum. This Environmental and Structural Wind Engineering. This is both a
approach has been followed by Huang and Li (2010) and Kubilay blessing and a curse. A blessing because it has allowed CWE to
et al. (2013) and is expected to become the preferred approach in develop into the strong eld in wind engineering that it is today.
the future. A curse because it is precisely the increased availability and
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 95

accessibility that, far too often, has led, is leading and will continue acoustics; (3) wind-borne debris; (4) wind energy in the built
to lead to the use of CFD for CWE without the required basic environment; (5) sports aerodynamics.
knowledge. Present and future evaluation of such studies will be CWE has come a long way. While there is still a long way to go,
unforgiving. There is no substitute for thorough knowledge of uid many problems to be tackled, many research questions to be
mechanics and numerical methods. There is no substitute for addressed and many challenges to overcome, the strong progress
detailed literature study of CFD and CWE literature and for the established in the past 50 years provides a promising outlook for
systematic and strenuous application of published best practice its future.
guidelines.
Care for high quality and reliability of CFD simulations is
crucial. Therefore, building further on the many research efforts Acknowledgements
made in the past decades, present and future CWE simulations
should demonstrate their quality and reliability by: This paper is an extended version of the keynote paper by the
author at the 6th European-African Conference on Wind Engineer-

Demonstrated assessment of numerical and physical modelling ing (6EACWE) in Cambridge, UK, July 711, 2013. The author is
errors. Discretisation errors, based on at least second order very grateful to the organisers for the kind invitation to deliver a
schemes, should be assessed by spatial and temporal sensitivity keynote lecture, and for their efforts in making this conference a
analyses, with Richardson extrapolation and preferably great success. 6EACWE was organised by Dr. John Owen, Prof. Dr.
reported using the convergence index by Roache (1997). After Mark Sterling, Dr. David Hargreaves, Prof. Dr. Chris Baker, Dr.
assessment of the numerical modelling errors, physical model- Hassan Hemida and Dr. Andrew Quinn.
ling errors should be estimated by comparison with high- The author is deeply indebted to his mentor in (computational)
quality wind-tunnel or on-site measurement data. If such data wind engineering, Prof. Dr. Ted Stathopoulos, whose group at Con-
are not available, the validation should be performed for cordia University, Montreal, he joined in 20042005, thanks to a
simpler congurations, the ow features of which show resem- fellowship from the Research Fund of Flanders (FWO-Vlaanderen) in
blance with those expected in the case under study. Flanders, Belgium. He is also very grateful to his mentors in building

Demonstrated adherence to general best practice guidelines in physics, Prof. Dr. Jan Carmeliet, previously full professor at Leuven
CWE, such as those by Franke et al. (2007, 2011), Britter and University in Belgium and now Chair of Building Physics at ETH
Schatzmann (2007), Tominaga et al. (2008a), Tamura et al. Zurich, Switzerland, and Prof. Dr. Hugo Hens, emeritus full professor
(2008) and Blocken and Gualtieri (2012) and to specic best at Leuven University.
practice guidelines such as those by Richards and Hoxey (1993), The author is a member of the Dutch-Flemish Wind Engineer-
Blocken et al. (2007a), Hargreaves and Wright (2007), Di ing Association, which is part of the IAWE, and as such he
Sabatino et al. (2007) and others. Note however that most best expresses his gratitude to his colleagues in this Association, in
practice guidelines for CFD focus on steady RANS rather than particular Prof. Jacob Wisse (founding member of the Association),
on LES and that the development of specic guidelines for LES Prof. Dr. Chris Geurts, Carine van Bentum, Eddy Willemsen and
in CWE is an important task for the future. Prof. Dr. Jeroen van Beeck.
In 2010, the author had the opportunity to contribute to the
organisation of CWE2010 in Chapel Hill, North Carolina, as chair of
It has been shown that the steady RANS approach in CWE has the scientic committee under the excellent chairmanship of Dr.
several important limitations, although it appears to be rather Alan Huber and in close collaboration with Prof. Dr. Ted Statho-
successful for pedestrian-level wind conditions, at least when poulos and Prof. Dr. Robert Meroney. In 2011, the author assisted in
high-quality and high-resolution grids are applied and when the organisation of ICWE13 in Amsterdam, as chair of the scientic
focusing on areas with high amplication factors, which are often committee, under the excellent chairmanship of Prof. Dr. Chris
the areas of primary interest in those studies. In this respect, it Geurts. Many thanks are due to both conference organising
might turn out that using the words by Ferziger (1990), committees for the very enjoyable and successful collaboration
pedestrian-level wind conditions is one of the few topics in CWE in support of the activities of the IAWE.
where nature is kind to us concerning turbulent ows. The author expresses his great appreciation to the Executive
In addition, steady RANS has also been used successfully for Board of the IAWE, in particular to Prof. Dr. Giovanni Solari,
studies of natural cross-ventilation of buildings and wind-driven previous IAWE president, and to Prof. Dr. Yukio Tamura, present
rain on building facades, although by far most of these studies only IAWE president, for their extensive and very successful efforts in
considered isolated buildings. Future studies should analyse the building the future of wind engineering in the framework of
capabilities of steady RANS when applied to buildings in both the IAWE.
generic and realistic urban environments. On the other hand, The author thanks his national and international colleagues,
steady RANS is clearly decient when mean and especially peak many of which are cited in this paper. A very special acknowl-
values of pollutant concentration or surface pressure are needed. edgement is given to Prof. Dr. Robert N. Meroney for his many
LES is intrinsically capable of providing this information, on invaluable comments to this paper based on his 50 years of
condition of sufcient computational resources, accurate initial expertise and experience in computational and experimental wind
and boundary conditions and sufciently high spatial and tem- engineering. Special thanks also go both Prof. Dr. Robert N.
poral resolution. Particular challenges for the future are exploring Meroney and Prof. Dr. Shuzo Murakami for their pioneering
the boundaries of coarse-grid LES and LES with efcient time- achievements in CWE that have shaped the past of CWE and will
stepping and the establishment of best practice guidelines for LES continue to shape its future. Thanks also go to Prof. Dr. Y. Tominaga
in CWE. Important achievements here would open up a wide and Prof. Dr. R. Yoshie, for the kind permission to reproduce some
range of additional opportunities and applications for CWE. of their gures in this paper.
Past research efforts have shown the application of CFD to Finally, last but certainly not least, a warm and sincere
many areas of wind engineering. Nevertheless, some areas appear acknowledgement is given to the postdocs and Ph.D. students in
still fairly unexploited or at least unpublished and might my group, whose youthful motivation, hard work and inspiration
represent fertile ground for new CWE developments and related keep us going every day: Postdocs Dr. Twan van Hooff and Dr.
publications: (1) surface convective heat transfer; (2) wind and Christof Gromke, and Ph.D. students Dr. Rubina Ramponi (who
96 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

received her Ph.D. between submission and revision of this paper), Beranek, W.J., Van Koten, H., 1979. Beperken van windhinder om gebouwen, deel 1,
Hamid Montazeri, Wendy Janssen, Yasin Toparlar, Adelya Khayr- Stichting Bouwresearch no. 65, Kluwer Technische Boeken BV, Deventer (in
Dutch).
ullina, Okke Bronkhorst, Jorge Isaac Peren Montero, Katarina Bitsuamlak, G., Stathopoulos, T., Bdard, C., 2004. Numerical evaluation of wind
Kosutova, Raffaele Vasaturo, Samy Iousef and Alessandro Pizzo- ow over complex terrain: review. J. Aerosp. Eng. 17 (4), 135145.
ferrato. The same warm acknowledgement is given to the previous Bjerg, B., Liberati, P., Marucci, A., Zhang, G., Banhazi, T., Bartzanas, T., Cascone, G., Lee, I.B.,
Norton, T., 2013a. Modelling of ammonia emissions from naturally ventilated
Ph.D. students that I supervised: Dr. Thijs Defraeye and Dr. Pierre
livestock buildings. Part 2: air change modelling. Biosyst. Eng. 116, 246258.
Gousseau. And to the current visiting and/or external Ph.D. Bjerg, B., Cascone, G., Lee, I.B., Bartzanas, T., Norton, T., Hong, S.W., Seo, I.H., Banhazi,
students in the group: Thijs van den Brande, Aytac Kubilay, T., Liberati, P., Marucci, A., Zhang, G., 2013b. Modelling of ammonia emissions
Antonio Castillo Torres and Alessio Ricci. from naturally ventilated livestock buildings. Part 3: CFD modelling. Biosyst.
Eng. 116, 259275.
Blocken, B., Carmeliet, J., 2002. Spatial and temporal distribution of driving rain on
a low-rise building. Wind Struct. 5 (5), 441462.
Blocken, B., Carmeliet, J., 2004a. A review of wind-driven rain research in building
References
science. J. Wind Eng. Ind. Aerodyn. 92 (13), 10791130.
Blocken, B., Carmeliet, J., 2004b. Pedestrian wind environment around buildings:
Abuku, M., Blocken, B., Nore, K., Thue, J.V., Carmeliet, J., Roels, S., 2009. On the literature review and practical examples. J. Therm. Envel. Build. Sci. 28 (2),
validity of numerical wind-driven rain simulation on a rectangular low-rise 107159.
building under various oblique winds. Build. Environ. 44 (3), 621632. Blocken, B., Roels, S., Carmeliet, J., 2004. Modication of pedestrian wind comfort in
Ai, Z.T., Mak, C.M., 2013. CFD simulation of ow and dispersion around an isolated the Silvertop Tower passages by an automatic control system. J. Wind Eng. Ind.
building: effect of inhomogeneous ABL and near-wall treatment. Atmos. Aerodyn. 92 (10), 849873.
Environ. 77, 568578. Blocken, B., Carmeliet, J., 2005. High-resolution wind-driven-rain measurements on
AIAA, 1998. Guide for the Verication and Validation of Computational Fluid a low-rise building experimental data for model development and model
Dynamics Simulations. American Institute of Aeronautics and Astronautics, validation. J. Wind Eng. Ind. Aerodyn. 93 (12), 905928.
AIAA, Reston, VA (AIAA-G-077-1998) Blocken, B., Carmeliet, J., 2006a. On the accuracy of wind-driven rain measurements
AMS, 2014. Meteorological Glossary. American Meteorological Society. http:// on buildings. Build. Environ. 41 (12), 17981810.
glossary.ametsoc.org/wiki/Mesoscale (Retrieved 04.03.14.). Blocken, B., Carmeliet, J., 2006b. The inuence of the wind-blocking effect by a building
Asfour, O.S., Gadi, M.B., 2007. A comparison between CFD and Network models for on its wind-driven rain exposure. J. Wind Eng. Ind. Aerodyn. 94 (2), 101127.
predicting wind-driven ventilation in buildings. Build. Environ. 42, 40794085. Blocken, B., Carmeliet, J., 2007a. On the errors associated with the use of hourly
ASME, 2009. Standard for Verication and Validation in Computational Fluid data in wind-driven rain calculations on building facades. Atmos. Environ. 41
Dynamics and Heat Transfer. ASME VandV 20-2009, The American Society of (11), 23352343.
Mechanical Engineers. Blocken, B., Carmeliet, J., 2007b. Validation of CFD simulations of wind-driven rain
ASME, 2011. http://journaltool.asme.org/Templates/JFENumAccuracy.pdf (Retrieved on a low-rise building. Build. Environ. 42 (7), 25302548.
30.06.11). Blocken, B., Carmeliet, J., 2008. Pedestrian wind conditions at outdoor platforms in
Avissar, R., Moran, M.D., Wu, G., Meroney, R.N., Pielke, R.A., 1990. Operating rages of a high-rise apartment building: generic sub-conguration validation, wind
mesoscale numerical models and meteorological wind tunnels for the simula- comfort assessment and uncertainty issues. Wind Struct. 11 (1), 5170.
tion of sea and land breezes. Bound.-Layer Meteorol. 50, 227275. Blocken, B., Stathopoulos, T., Carmeliet, J, 2007a. CFD simulation of the atmospheric
Awbi, H.B., 2003. Ventilation of Buildings. Spon Press, London: New York boundary layer: wall function problems. Atmos. Environ. 41 (2), 238252.
Ayotte, K., 2008. Computational modelling for wind energy assessment. J. Wind Blocken, B., Carmeliet, J., Stathopoulos, T., 2007b. CFD evaluation of wind speed
Eng. Ind. Aerodyn. 96 (1011), 15711590. conditions in passages between parallel buildingseffect of wall-function
Baetke, F., Werner, H., Wengle, H., 1990. Numerical simulation of turbulent ow roughness modications for the atmospheric boundary layer ow. J. Wind
over surface-mounted obstacles with sharp edges and corners. J. Wind Eng. Ind. Eng. Ind. Aerodyn. 95 (911), 941962.
Aerodyn. 35 (13), 129147. Blocken, B, Stathopoulos, T, Saathoff, P, Wang, X., 2008a. Numerical evaluation of
Baik, J.-J., Kwak, K.-H., Park, S.-B., Ryu, Y.-H., 2012. Effects of building roof greening pollutant dispersion in the built environment: comparisons between models
on air quality in street canyons. Atmos. Environ. 61, 4855. and experiments. J. Wind Eng. Ind. Aerodyn. 96 (1011), 18171831.
Baker, C.J., 2000. Foreword. In: Proceedings of the 3rd International Symposium on Blocken, B., Stathopoulos, T., Carmeliet, J., 2008b. Wind environmental conditions
Computational Wind Engineering. Birmingham, UK, 47 September, 2000.
in passages between two long narrow perpendicular buildings. J. Aerosp. Eng.
Baker, C.J., 2007. Wind engineering past, present and future. J. Wind Eng. Ind.
ASCE 21 (4), 280287.
Aerodyn. 95 (911), 843870.
Blocken, B., Persoon, J., 2009. Pedestrian wind comfort around a large football
Baker, C.J., 2010. The ow around high speed trains. J. Wind Eng. Ind. Aerodyn. 98
stadium in an urban environment: CFD simulation, validation and application
(67), 277298.
of the new Dutch wind nuisance standard. J. Wind Eng. Ind. Aerodyn. 97 (56),
Baker, C.J., Wood, C.J., Gawthorpe, R.G., 1985. Strong winds in complicated hilly
255270.
terrain eld measurements and wind-tunnel study. J. Wind Eng. Ind.
Blocken, B., Carmeliet, J., 2010. Overview of three state-of-the-art wind-driven rain
Aerodyn. 18 (1), 126.
assessment models and comparison based on model theory. Build. Environ. 45
Balczo, M., Gromke, C., Ruck, B., 2009. Numerical modelling of ow and pollutant
(3), 691703.
dispersion in street canyons with tree planting. Meteorol. Z. 18 (2), 197206.
Blocken, B., Desz, G., van Beeck, J., Carmeliet, J., 2010. Comparison of calculation
Balczo, M., Balogh, M., Goricsan, I., Nagel, T., Suda, J.M., Lajos, T., 2011. Air quality
methods for wind-driven rain deposition on building facades. Atmos. Environ.
around motorway tunnels in complex terrain computational uid dynamics
44 (14), 17141725.
modeling and comparison to wind tunnel data. IDOJARAS 115 (3), 179204.
Blocken, B., Stathopoulos, T., Carmeliet, J., Hensen, J.L.M., 2011a. Application of CFD
Bangalee, M.Z.I., Miau, J.J., Lin, S.Y., Yang, J.H., 2013. Flow visualization, PIV
measurement and CFD calculation for uid-driven natural cross-ventilation in in building performance simulation for the outdoor environment: an overview.
a scale model. Energy Build. 66, 306314. J. Build. Perform. Simul. 4 (2), 157184.
Bartzanas, T., Boulard, T., Kittas, C., 2002. Numerical simulation of the airow and Blocken, B., van Hooff, T., Aanen, L., Bronsema, B., 2011b. Computational analysis of
temperature distribution in a tunnel greenhouse equipped with insect-proof the performance of a venturi-shaped roof for natural ventilation: venturi-effect
screen in the openings. Comput. Electron. Agric. 34, 207221. versus wind-blocking effect. Comput. Fluids 48 (1), 202213.
Bartzanas, T., Boulard, T., Kittas, C., 2004. Effect of vent arrangement on windward Blocken, B., Abuku, M., Nore, K., Briggen, P.M., Schellen, H.L., Thue, J.V., Roels, S.,
ventilation of a tunnel greenhouse. Biosyst. Eng. 88, 479490. Carmeliet, J., 2011c. Intercomparison of wind-driven rain deposition models
Bartzanas, T., Kittas, C., Sapounas, A.A., Nikita-Martzopoulou, C., 2007. Analysis of based on two case studies with full-scale measurements. J. Wind Eng. Ind.
airow through experimental rural buildings: Sensitivity to turbulence models. Aerodyn. 99 (4), 448459.
Biosyst. Eng. 97, 229239. Blocken, B., Gualtieri, C., 2012. Ten iterative steps for model development and
Bartzis, J.G., 2006. Turbulence modelling in the atmospheric boundary layer: a review evaluation applied to computational uid dynamics for environmental uid
and some recent developments. WIT Trans. Ecol. Environ.: Air Pollut. XIV 86, 312. mechanics. Environ. Model. Softw. 33, 122.
Bartzis, J.G., Vlachogiannis, D., Sfetsos, A., 2004. Thematic area 5: best practice Blocken, B., Janssen, W.D., van Hooff, T., 2012. CFD simulation for pedestrian wind
advice for environmental ows. QNET-CFD Netw. Newsl. 2 (4), 3439. comfort and wind safety in urban areas: general decision framework and case
Baskaran, A., Kashef, A., 1996. Investigation of air ow around buildings using study for the Eindhoven University campus. Environ. Model. Softw. 30, 1534.
computational uid dynamics techniques. Eng. Struct. 18 (11), 861873. Blocken, B., Derome, D., Carmeliet, J., 2013a. Rainwater runoff from building
Baskaran, A., Stathopoulos, T., 1989. Computational evaluation of wind effects on facades: a review. Build. Environ. 60, 339361.
buildings. Build. Environ. 24 (4), 325333. Blocken, B., Tominaga, Y., Stathopoulos, T., 2013b. Editorial to virtual special issue:
Baskaran, A., Stathopoulos, T., 1992. Inuence of computational parameters on the CFD simulation of micro-scale pollutant dispersion in the built environment.
evaluation of wind effects on the building envelope. Build. Environ. 27 (1), Build. Environ. 64, 225230.
3949. Blocken, B., Stathopoulos, T., 2013. Editorial to virtual special issue: CFD simulation
Beranek, W.J., 1982. Beperken van windhinder om gebouwen, deel 2, Stichting of pedestrian-level wind conditions around buildings: past achievements and
Bouwresearch no. 90, Kluwer Technische Boeken BV, Deventer (in Dutch). prospects. J. Wind Eng. Ind. Aerodyn. 121, 138145.
Beranek, W.J., 1984. Wind environment around single buildings of rectangular Bottema, M., 1993. Wind Climate and Urban Geometry (Ph.D. thesis). Eindhoven
shape. Heron 29 (1), 431. University of Technology (212 p)
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 97

Bournet, P.E., Boulard, T., 2010. Effect of ventilator conguration on the distributed Defraeye, T., Blocken, B., Derome, D., Nicolai, B., Carmeliet, J., 2012. Convective heat
climate of greenhouses: a review of experimental and CFD studies. Comput. and mass transfer modelling at airporous material interfaces: overview of
Electron. Agric. 74 (2), 195217. existing methods and relevance. Chem. Eng. Sci. 74, 4958.
Bowen, A.J., 2003. Modelling of strong wind ows over complex terrain at small Defraeye, T., Verboven, P., Ho, Q.T., Nicolai, B., 2013. Convective heat and mass
geometric scales. J. Wind Eng. Ind. Aerodyn. 91 (1215), 18591871. exchange predictions at leaf surfaces: applications, methods and perspectives.
Briggen, P.M., Blocken, B., Schellen, H.L., 2009. Wind-driven rain on the facade of a Comput. Electron. Agric. 96, 180201.
monumental tower: numerical simulation, full-scale validation and sensitivity Dejoan, A., Santiago, J.L., Martilli, A., Martin, F., Pinelli, A., 2010. Comparison between
analysis. Build. Environ. 44 (8), 16751690. Large-Eddy Simulation and Reynolds-Averaged Navier-Stokes computations for the
Britter, R., Schatzmann, M., 2007 (Eds.). Model Evaluation Guidance and Protocol MUST eld experiment. Part II: effects of incident wind angle deviation on the
Document COST Action 732. COST Ofce Brussels, 3-00-018312-4. mean ow and plume dispersion. Bound.-Layer Meteorol. 135 (1), 133150.
Buccolieri, R., Di Sabatino, S., 2011. MUST experiment simulations using CFD and Delage, Y., Taylor, P.A., 1970. Numerical studies of heat island circulations. Bound.-
integral models. Int. J. Environ. Pollut. 44 (14), 376384. Layer Meteorol. 1, 201226.
Buccolieri, R., Gromke, C., Di Sabatino, S., Ruck, B., 2009. Aerodynamic effects of Delpech, Ph., Palier, P., Gandemer, J., 1998. Snowdrifting simulation around Antartic
trees on pollutant concentration in street canyons. Sci. Total Environ. 407 (19), buildings. J. Wind Eng. Ind. Aerodyn. 7476, 567576.
52475256. Derickson, R.G., Meroney, R.N., 1977. A simplied physics airow model for evaluating
Buccolieri, R., Salim, S.M., Leo, L.S., Di Sabatino, S., Chan, A.D., Ielpo, P., de Gennaro, wind power sites in complex terrain. In: Proceedings of the Summer Computer
Simulation Conference, July 1820, 1977, Hyatt Regency, Chicago, Illinois.
G., Gromke, C., 2011. Analysis of local scale tree-atmosphere interaction on
Di Sabatino, S., Buccolieri, R., Pulvirtenti, B., Britter, R., 2007. Simulations of
pollutant concentration in idealized street canyons and application to a real
pollutant dispersion within idealised urban-type geometries with CFD and
urban junction. Atmos. Environ. 45 (9), 17021713.
integral models. Atmos. Environ. 41, 83168329.
Calautit, J.K., Chaudry, H.N., Hughes, B.R., Sim, L.F., 2014. A validated design
Di Sabatino, S., Buccolieri, R., Olesen, H.R., Ketzel, M., Berkowicz, R., Franke, J.,
methodology for a closed-loop subsonic wind tunnel. J. Wind Eng. Ind.
Schatzmann, M., Schlnzen, H., Leitl, B., Britter, R., Borrego, C., Costa, A.M., Trini
Aerodyn. 125, 180194.
Castelli, S., Reisin, T., Hellsten, A., Saloranta, J., Moussiopoulos, N., Barmpas, F.,
Canepa, E., 2004. An overview about the study of downwash effects on dispersion
Brzozowski, K., Goricsan, I., Balcz, M., Bartzis, J., Efthimiou, G., Santiago, J.L.,
of airborne pollutants. Environ. Model. Softw. 19 (12), 10771087. Martilli, A., Piringer, M., Hirtl, M., Baklanov, A., Nuterman, R., Starchenko, A.,
Casey, M., Wintergerste, T., 2000. Best Practice Guidelines, ERCOFTAC Special 2011. COST 732 in practice: the MUST model evaluation exercise. Int. J. Environ.
Interest Group on Quality and Trust in Industrial CFD, ERCOFTAC, Brussels. Pollut. 44, 403418.
Castro, I.P., Graham, J.M.R., 1999. Numerical wind engineering: the way ahead? Di Sabatino, S., Buccolieri, R., Salizzoni, P., 2013. Recent advancements in numerical
Proc. Inst. Civil Eng. Struct. Build. 134 (3), 275277. modelling of ow and dispersion in urban areas: a short review. Int. J. Environ.
Castro, I.P., Robins, A.G., 1977. Flow around a surface-mounted cube in uniform and Pollut. 52 (34), 172191.
turbulent streams. J. Fluid Mech. 79, 307. Durgin, F.H., 1992. Pedestrian level wind studies at the Wright brothers facility. J.
Cermak, J.E., 1975. Applications of uid mechanics to wind engineering - A Freeman Wind Eng. Ind. Aerodyn. 44 (13), 22532264.
Scholar Lecture. J. Fluids Engng., ASME, 938 Estoque, M.A., 1961. A theoretical investigation of the sea breeze. Quart. J. R.
Chang, W.R., 2006. Effect of porous hedge on cross ventilation of a residential Metorol. Soc. 87, 136146.
building. Build. Environ. 41, 549556. Estoque, M.A., 1962. The sea breeze as a function of the prevailing synoptic
Chapman S., 1965. Introduction to Dover Edition of Weather Prediction by situation. J. Atmos. Sci. 19, 244250.
Numerical Process. Estoque, M.A., Bhumralkar, C.M., 1969. Flow over a localized heat source. Mon.
Charney, J.G., 1955. The use of the primitive equations of motion in numerical Weather Rev. 97, 850859.
prediction. Tellus 7, 2226. Etheridge, D.W., Sandberg, M., 1996. Building Ventilation: Theory and Measure-
Charney, J.G., Fjortoft, R., von Neumann, J., 1950. Numerical integration of the ment. John Wiley & Sons, Chichester, New York
barotropic vorticity equation. Tellus 2, 237254. Etheridge, D.W., 2011. Natural ventilation of buildings. Theory, measurement and
Chay, M.T., Letchford, C.W., 2002. Pressure distributions on a cube in a simulated design. Wiley p. 454
thunderstorm downburst Part A: stationary downburst observations. J. Wind Etyemezian, V., Davidson, C.I., Zufall, M., Dai, W., Finger, S., Striegel, M., 2000.
Eng. Ind. Aerodyn. 90 (7), 711732. Impingement of rain drops on a tall building. Atmos. Environ. 34 (15),
Chen, Q., Glicksman, L.R., Lin, J., Scott, A., 2007. Sustainable urban housing in China. 23992412.
J. Harbin Inst. Tech. (New Ser.) 14, 69. Evola, G., Popov, V., 2006. Computational analysis of wind driven natural ventila-
Chen, Q., 2009. Ventilation performance prediction for buildings: a method over- tion in buildings. Energy Build. 38, 491501.
view and recent applications. Build. Environ. 44 (4), 848858. Fatnassi, H., Boulard, T., Poncet, C., Chave, M., 2006. Optimisation of greenhouse
Cheung, J.O.P., Liu, C.H., 2011. CFD simulations of natural ventilation behaviour in insect screening with computational uid dynamics. Biosyst. Eng. 93, 301312.
high-rise buildings in regular and staggered arrangements at various spacings. Fernando, H.J.S., Zajic, D., Di Sabatino, S., Dimitrova, R., Hedquist, B., Dallman, A.,
Energy Build. 43, 11491158. 2010. Flow, turbulence and pollutant dispersion in urban atmospheres. Phys.
Choi, E.C.C., 1991. Numerical simulation of wind-driven rain falling onto a 2-D Fluids 22 (5), 051301.
building. in: Proceedings of the Asia Pacic Conference on Computational Ferreira, A.D., Sousa, A.C.M., Viegas, D.X., 2002. Prediction of building interference
Mechanics, Hong Kong, pp. 17211728. effects on pedestrian level comfort. J. Wind Eng. Ind. Aerodyn. 90 (45),
Choi, E.C.C., 1993. Simulation of wind-driven rain around a building. J. Wind Eng. 305319.
Ind. Aerodyn. 46&47, 721729. Ferziger, J.H., 1990. Approaches to turbulent ow computation: applications to ow
Choi, E.C.C., 1994a. Determination of wind-driven rain intensity on building faces. J. over obstacles. J. Wind Eng. Ind. Aerodyn. 35, 119.
Wind Eng. Ind. Aerodyn. 51, 5569. Ferziger, J.H., 1993a. A computational uid dynamicist's view of CWE. J. Wind Eng.
Choi, E.C.C., 1994b. Parameters affecting the intensity of wind-driven rain on the Ind. Aerodyn. 4647, 879880.
Ferziger, J.H., 1993b. Simulation of complex turbulent ows: recent advances and
front face of a building. J. Wind Eng. Ind. Aerodyn. 53, 117.
prospects in wind engineering. J. Wind Eng. Ind. Aerodyn. 4647, 195212.
Choi, E.C.C., 1997. Numerical modelling of gust effect on wind-driven rain. J. Wind
Ferziger, J.H., 1993c. Estimation and reduction of numerical error. FED vol. 158, In:
Eng. Ind. Aerodyn. 72, 107116.
Proceedings of the Symposium on Quantication of Uncertainty in Computa-
Chu, C.R., Chiang, B.F., 2013. Wind-driven cross ventilation with internal obstacles.
tional Fluid Dynamics, ASME Fluid Engineering Division, Summer Meeting,
Energy Build. 67, 201209.
Washington DC, June 2024, pp. 18.
Cochran, L., Derickson, R., 2011. A physical modeler's view of computational wind
Ferziger, J.H., Peric, M., 1996. Computational Methods for Fluid Dynamics. Springer,
engineering. J. Wind Eng. Ind. Aerodyn. 99 (4), 139153.
Berlin p. 356
Conan, B., van Beeck, J., Aubrun, S., 2012. Sand erosion technique applied to wind
Finnegan, M.J., Pickering, C.A., Burge, P.S., 1984. The sick building syndrome:
resource assessment. J. Wind Eng. Ind. Aerodyn. 104106, 322329.
prevalence studies. Br. Med. J. (Clin. Res. Ed.) 289, 15731575.
Cowan, I.R., Castro, I.P., Robins, A.G., 1997. Numerical considerations for simulations
Fisher, E.L., 1961. A theoretical study of the sea breeze. J. Meteorol. 18, 216233.
of ow and dispersion around buildings. J. Wind Eng. Ind. Aerodyn. 67 and 68, Fosberg, M.A., 1967. Numerical analysis of convective motions over a mountain
535545. ridge. J. Appl. Meteorol. 6, 889904.
da Graa, G., Chen, Q., Glicksman, L.R., Norford, L.K., 2002. Simulation of wind- Fosberg, M.A., 1969. Airow over a heated coastal mountain. J. Appl. Meteorol. 8,
driven ventilative cooling systems for an apartment building in Beijing and 436442.
Shanghai. Energy Build. 34, 111. Franke, J., Hirsch, C., Jensen, A.G., Krs, H.W., Schatzmann, M., Westbury, P.S., Miles,
Da Matha SantAnna, F., Taylor, D.A., 1990. Snow drifts on at roofs: wind tunnel S.D., Wisse, J.A., Wright, N.G., 2004. Recommendations on the use of CFD in
tests and eld measurements. J. Wind Eng. Ind. Aerodyn. 34 (3), 223250. wind engineering. In: van Beeck, J.P.A.J. (Ed.), Proceedings of the International
Dalgliesh, W.A., Surry, D., 2003. BLWT, CFD and HAM modelling vs. the real world: Conference on Urban Wind Engineering and Building Aerodynamics. COST
bridging the gaps with full-scale measurements. J. Wind Eng. Ind. Aerodyn. 91, Action C14, Impact of Wind and Storm on City Life Built Environment. Von
16511669. Karman Institute, Sint-Genesius-Rode, Belgium, 57 May 2004.
Davenport, A.G., 1991. Preface to the Proceedings of the 8th International Con- Franke, J., Hellsten, A., Schlnzen, H., Carissimo, B., 2011. The COST 732 best practice
ference on Wind Engineering, London, Ontario. guideline for CFD simulation of ows in the urban environment a summary.
Davenport, A.G., 1999. The missing links. In: Proceedings of the 10th International Int. J. Environ. Pollut. 44 (14), 419427.
Conference on Wind Engineering, Copenhagen, pp. 315. Franke, J., Hellsten, A., Schlnzen, H., Carissimo, B. (Eds.). 2007. Best practice
Deaves, D.M., 1975. Wind over hills: a numerical approach. J. Wind Eng. Ind. guideline for the CFD simulation of ows in the urban environment. COST
Aerodyn. 1, 371391. Ofce Brussels, 3-00-018312-4.
98 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Freitas, C.J., 1993. Journal of uids engineering editorial policy statement on the Hgberg, A.B., Kragh, M.K., van Mook, F.J.R., 1999. A comparison of driving rain
control of numerical accuracy. J. Fluids Eng. 115, 339340. measurements with different gauges. In: Proceedings of the Fifth Symposium of
Frost, W., Maus, J.R., Fichtl, G.H., 1974. A boundary-layer analysis of atmospheric Building Physics in the Nordic Countries, Gothenburg. pp. 3618.
motion over a semi-elliptical surface obstruction. Bound.-Layer Meteorol. 7, Holmes, N.S., Morawska, L., 2006. A review of dispersion modelling and its
165184. application to the dispersion of particles: an overview of different dispersion
Fujii, K., 2005. Progress and future prospects of CFD in aerospacewind tunnel and models available. Atmos. Environ. 40 (30), 59025928.
beyond. Prog. Aerosp. Sci. 41 (6), 455470. Horan, J.M., Finn, D.P., 2008. Sensitivity of air change rates in a naturally ventilated
Gadilhe, A., Janvier, L., Barnaud, G., 1993. Numerical and experimental modelling of atrium space subject to variations in external wind speed and direction. Energy
the three-dimensional turbulent wind ow through an urban square. J. Wind Build. 40, 15771585.
Eng. Ind. Aerodyn. 4647, 755763. Hu, C.H., Kurabuchi, T., Ohba, M., 2005. Numerical study of cross-ventilation using
Ge, Y.J., Xiang, H.F., 2008. Recent development of bridge aerodynamics in China. J. two-equation RANS turbulence models. Int. J. Vent. 4, 123132.
Wind Eng. Ind. Aerodyn. 96 (67), 736768. Hu, C.H., Ohba, M., Yoshie, R., 2008. CFD modelling of unsteady cross ventilation
Goricsan, I., Balczo, M., Balogh, M., Czader, K., Rakai, A., Tonko, C., 2011. Simulation ows using LES. J. Wind Eng. Ind. Aerodyn. 96, 16921706.
of ow in an idealised city using various CFD codes. Int. J. Environ. Pollut. Huang, S.H., Li, Q.S., 2010. Numerical simulations of wind-driven rain on building
44 (14), 359367. envelopes based on Eulerian multiphase model. J. Wind Eng. Ind. Aerodyn. 98 (12),
Gorl, C., van Beeck, J., Rambaud, P., Van Tendeloo, G., 2009. CFD modelling of small 843857.
particle dispersion: the inuence of the turbulence kinetic energy in the Huber, A., Blocken, B., 2011. Preface to the special issue of the 5th symposium on
atmospheric boundary layer. Atmos. Environ. 43 (3), 673681. computational wind engineering. J. Wind Eng. Ind. Aerodyn. 99 (4) (vii-v).
Gosman, A.D., 1999. Developments in CFD for industrial and environmental Huber, A.H., Snyder, W.H., 1982. Wind tunnel investigation of the effects of a
applications in wind engineering. J. Wind Eng. Ind. Aerodyn. 81 (13), 2139. rectangular-shaped building on dispersion of efuents from short adjacent
Gousseau, P., Blocken, B., Stathopoulos, T., van Heijst, G.J.F., 2011a. CFD simulation of stacks. Atmos. Environ. 16 (12), 28372848.
near-eld pollutant dispersion on a high-resolution grid: a case study by LES and Hucho, W.H., Sovran, G., 1993. Aerodynamics of road vehicles. Annu. Rev. Fluid
RANS for a building group in downtown Montreal. Atmos. Environ. 45 (2), Mech. 25, 485537.
428438. Hughes, T.J.R, Jansen, K., 1993. Finite element methods in wind engineering. J. Wind
Gousseau, P., Blocken, B., van Heijst, G.J.F., 2011b. CFD simulation of pollutant Eng. Ind. Aerodyn. 4647, 297313.
dispersion around isolated buildings: on the role of convective and turbulent Hunt, G.R., Linden, P.F., 1999. The uid mechanics of natural ventilation
mass uxes in the prediction accuracy. J. Hazard. Mater. 194, 422434. displacement ventilation by buoyancy-driven ows assisted by wind. Build.
Gousseau, P., Blocken, B., van Heijst, G.J.F., 2012. Large-Eddy Simulation of pollutant Environ. 34, 707720.
dispersion around a cubical building: analysis of the turbulent mass transport Inculet, D., Surry, D. 1994. Simulation of Wind-driven Rain and Wetting Patterns on
mechanism by unsteady concentration and velocity statistics. Environ. Pollut. 167, Buildings, BLWTLSS30-1994, Final Report.
4757. Inculet, D.R. 2001. The design of Cladding Against Wind-driven Rain (Ph.D.) thesis,
Gromke, C., 2011. A vegetation modeling concept for building and environmental The University of Western Ontario, London, Canada, 297 p.
aerodynamics wind tunnel tests and its application in pollutant dispersion Irwin, H.P.A.H., 1981. A simple omnidirectional sensor for wind-tunnel studies of
studies. Environ. Pollut. 159 (89), 20942099.
pedestrian-level winds. J. Wind Eng. Ind. Aerodyn. 7 (3), 219239.
Gromke, C., Ruck, B., 2007. Inuence of trees on the dispersion of pollutants in an
Isyumov, N., Davenport, AG., 1975. Comparison of full-scale and wind tunnel wind
urban street canyon experimental investigation of the ow and concentration
speed measurements in the commerce court plaza. J. Wind Eng. Ind. Aerodyn.
eld. Atmos. Environ. 41 (16), 32873302.
1, 201212.
Gromke, C., Ruck, B., 2008. Aerodynamic modelling of trees for small-scale wind
Isyumov, N., Mikitiuk, M., 1990. Wind tunnel model tests of snow drifting on a two-
tunnel studies. Forestry 81 (3), 243258.
level at roof. J. Wind Eng. Ind. Aerodyn. 36 (2), 893904.
Gromke, C., Ruck, B., 2009. On the impact of trees on dispersion processes of trafc
Iversen, J.D., 1981. Comparison of wind-tunnel model and full-scale snow fence
emissions in street canyons. Bound.-Layer Meteorol. 131 (1), 1934.
drifts. J. Wind Eng. Ind. Aerodyn. 8 (3), 231249.
Gromke, C., Ruck, B., 2012. Pollutant concentrations in street canyons of different
Jakeman, A.J., Letcher, R.A., Norton, J.P., 2006. Ten iterative steps in development
aspect ration with avenues of trees for various wind directions. Bound.-Layer
and evaluation of environmental models. Environ. Model. Softw. 21 (5),
Meteorol. 144 (1), 4164.
602614.
Gromke, C., Buccolieri, R., Di Sabatino, S., Ruck, B., 2008. Dispersion study in a street
Janssen, W.D., Blocken, B., van Hooff, T., 2013. Pedestrian wind comfort around
canyon with tree planting by means of wind tunnel and numerical investigations
buildings: comparison of wind comfort criteria based on whole-ow eld data
evaluation of CFD data with experimental data. Atmos. Environ. 42 (37), 86408650.
for a complex case study. Build. Environ. 59, 547562.
Hajdukiewicz, M., Geron, M., Keane, M.M., 2013. Calibrated CFD simulation to
Jiang, Y., Chen, Q., 2002. Effect of uctuating wind direction on cross natural
evaluate thermal comfort in a highly-glazed naturally ventilated room. Build.
ventilation in buildings from large eddy simulation. Build. Environ. 37,
Environ. 70, 7389.
Hall, R.C., (Ed.), 1997. Evaluation of Modelling Uncertainty. CFD Modelling of Near- 379386.
eld Atmospheric Dispersion. Project EMU Final Report, European Commission Jiang, Y., Alexander, D., Jenkins, H., Arthur, R., Chen, Q., 2003. Natural ventilation in
DirectorateGeneral XII Science, Research and Development Contract EV5V- buildings: measurement in a wind tunnel and numerical simulation with large-
CT94- 0531. Surrey: WS Atkins Consultants Ltd. eddy simulation. J. Wind Eng. Ind. Aerodyn. 91, 331353.
Haltiner, G.J., Williams, R.T., 1980. Numerical prediction and dynamic meteorology. Jiru, T.E., Bitsuamlak, G.T., 2010. Application of CFD in modelling wind-induced
Wiley, 477 p. natural ventilation of buildings a review. Int. J. Vent. 9 (2), 131147.
Hangan, H., 1999. Wind-driven rain studies. A C-FD-E approach. J. Wind Eng. Ind. Jones, W., Launder, B., 1972. The prediction of laminarization with a two-equation
Aerodyn. 81, 323331. model of turbulence. Int. J. Heat Mass Transf. 15, 301314.
Hanjalic, K., 2004. Will RANS survive LES? A view of perspectives. J. Fluids Eng. Karava, P., Stathopoulos, T., Athienitis, A.K., 2004. Wind driven ow through
Trans. ASME 127 (5), 831839. openingsa review of discharge coefcients. Int. J. Vent. 3, 255266.
Hanjalic, K., Kenjeres, S., 2008. Some developments in turbulence modeling for Karava, P., Stathopoulos, T., Athienitis, A.K., 2006. Impact of internal pressure
wind and environmental engineering. J. Wind Eng. Ind. Aerodyn. 96 (1011), coefcients on wind-driven ventilation analysis. Int. J. Vent. 5, 5366.
15371570. Karava, P., Stathopoulos, T., Athienitis, A.K., 2007. Wind-induced natural ventilation
Hanna, S.R., 1989. Plume dispersion and concentration uctuations in the atmo- analysis. Sol. Energy 81, 2030.
sphere. Encyclopedia of environmental control technology. Air Pollution Con- Karava, P., Stathopoulos, T., Athienitis, A.K., 2011. Airow assessment in cross-
trol, vol. 2. Gulf Publishing Company, Houston, TX, pp. 547582. ventilated buildings with operable facade elements. Build. Environ. 46,
Hansen, M.O.L., Srensen, J.N., Voutsinas, S., Srensen, N., Madsen, H.A., 2006. State 266279.
of the art in wind turbine aerodynamics and aeroelasticity. Prog. Aerosp. Sci. 42 Karava, P., Stathopoulos, T., 2012. Wind-induced internal pressures in buildings
(4), 285330. with large faade openings. J. Eng. Mech. 138 (4), 358370.
Hanson, T., Summers, D.M., Wilson, C.B., 1986. Validation of a computer simulation Kareem, A., 2008. Numerical simulation of wind effects: a probabilistic perspective.
of wind ow over a building model. Build. Environ. 21, 97111. J. Wind Eng. Ind. Aerodyn. 96 (1011), 14721497.
Hargreaves, D.M., Wright, N.G., 2007. On the use of the k model in commercial Kasahara, A., Washington, W.M., 1967. NCAR global general circulation model of the
CFD software to model the neutral atmospheric boundary layer. J. Wind Eng. atmosphere. Mon. Weather Rev. 95, 389402.
Ind. Aerodyn. 95 (5), 355369. Kasahara, A., 1974. Various vertical coordinate systems used for numerical weather
He, J., Song, C.C.S., 1999. Evaluation of pedestrian winds in urban area by numerical prediction. Mon. Weather Rev. 102 (7), 509522.
approach. J. Wind Eng. Ind. Aerodyn. 81, 295309. Katayama, T., Tsutsumi, J., Ishii, A., 1992. Full-scale measurements and wind tunnel
Heiselberg, P., Perino, M., 2010. Short-term airing by natural ventilationimplica- tests on cross-ventilation. J. Wind Eng. Ind. Aerodyn. 4144, 25532562.
tion on IAQ and thermal comfort. Indoor Air 20, 126140. Kato, M., Launder, B.E., 1993. The modelling of turbulent ow around stationary and
Hess, J.L., Smith, A.M.O., 1967. Calculation of potential ow about arbitrary bodies. vibrating square cylinders. In: Proceedings of the Ninth Symposium on
Prog. Aeronaut. Sci. 8, 1138. Turbulent Shear Flows. pp. 1014.
Hirsch, C., Boufoux, V., Wilquem, F., 2002. CFD simulation of the impact of new Kato, S., Murakami, S., Mochida, A., Akabayashi, S., Tominaga, Y., 1992. Velocity-
buildings on wind comfort in an urban area. Workshop Proceedings, Cost pressure eld of cross ventilation with open windows analyzed by wind tunnel
Action C14, Impact of Wind and Storm on City Life and Built Environment, and numerical simulation. J. Wind Eng. Ind. Aerodyn. 44, 25752586.
Nantes, France. Katz, J., 2006. Aerodynamics of race cars. Annu. Rev. Fluid Mech. 38, 2763.
Hirt, C.W., Cook, J.L., 1972. Calculating three-dimensional ows around structures Kimura, R., 2002. Numerical weather prediction. J. Wind Eng. Ind. Aerodyn. 90,
and over rough terrain. Journal of Computational Physics 10, 324340. 14031414.
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 99

Kobayashi, T., Sagara, K., Yamanaka, T., Kotani, H., Takeda, S., Sandberg, M., 2009. Magata, M., 1965. A study of the sea breeze by numerical experimentation. Papers
Stream tube based analysis of problems in prediction of cross-ventilation rate. Meteor. Geophys. Tokyo, vol. 16. pp. 2336.
Int. J. Vent. 7, 321334. Mason, M.S., Letchford, C.W., James, D.L., 2005. Pulsed wall jet simulation of a
Kobayashi, T., Sandberg, M., Kotani, H., Claesson, L., 2010. Experimental investiga- stationary thunderstorm downburst: part A: physical structure and ow eld
tion and cfd analysis of cross-ventilated ow through single room detached characterization. J. Wind Eng. Ind. Aerodyn. 93 (7), 557580.
house model. Build. Environ. 45, 27232734. Mehta, R.D., 1985. Aerodynamics of sports balls. Annu. Rev. Fluid Mech. 17,
Kopp, G.A., 2013. Guest editorial for SI: solar array and wind loads. J. Wind Eng. Ind. 151189.
Aerodyn. 123, 191. Menter, F., Hemstrom, B., Henrikkson, M., Karlsson, R., Latrobe, A., Martin, A.,
Kothari, K.M, Peterka, J.A., Meroney, R.N., 1986. Perturbation analysis and measure- Muhlbauer, P., Scheuerer, M., Smith, B., Takacs, T., Willemsen, S., 2002. CFD Best
ments of building wakes in a stably stratied turbulent boundary layer. J. Wind Practice Guidelines for CFD Code Validation for Reactor-Safety Applications,
Eng. Ind. Aerodyn. 25 (1), 4974. Report EVOLECORA-D01, Contract no. FIKS-CT-2001-00154, 2002.
Kubilay, A., Derome, D., Blocken, B., Carmeliet, J., 2013. CFD simulation and Menter, F.R., 1994. Two-equation eddy-viscosity turbulence models for engineering
validation of wind-driven rain on a building facade with an Eulerian multi- applications. AIAA J. 32, 15981605.
phase model. Build. Environ. 61, 6981. Meroney, B.N., Meroney, R.N., 1989. Snow control with vortex and blower fences. In:
Kurabuchi, T., Ohba, M., Arashiguchi, A., Iwabuchi, T., 2000. Numerical study of Proceedings of the Conference on a Multidisciplinary Approach to Snow
airow structure of a cross ventilated model building. In: Air Distribution in Engineering, Santa Barbara, California, 1015 July 1988.
Rooms: Ventilation for Health and Sustainable Environment, pp. 313-8. Meroney, R.N., Yamada, T., 1971. Wind tunnel and numerical experiments of two-
Kwok, K.C.S., Kim, D.H., Smedley, D.J., Rohde, H.F., 1992. Snowdrift around buildings dimensional stratied airow over a heated island. Winter Annual Meeting of
for antartic environment. J. Wind Eng. Ind. Aerodyn. 44 (13), 27972808. ASME, Nov. 28-Dec. 2, Washington DC.
Lakehal, D., Mestayer, P.G., Edson, J.B., Anquetin, S., Sini, J.F., 1995. Eulero- Meroney, R.N., Yamada, T., 1972. Numerical and physical simulation of a stratied
Lagrangian simulation of raindrop trajectories and impacts within the urban airow over a series of heated islands. In: Proceedings of the Summer
canopy. Atmos. Environ. 29 (23), 35013517. Simulation Conference, June 1316, 1972, San Diego, California.
Larsen, T.S., Heiselberg, P., 2008. Single-sided natural ventilation driven by wind Meroney, R.N., 1980. Wind-tunnel simulation of the ow over hills and complex
pressure and temperature difference. Energy Build. 40, 10311040. terrain. J. Ind. Aerodyn. 5, 297321.
Larsen, T.S., Nikolopoulos, N., Nikolopoulos, A., Strotos, G., Nikas, K.S., 2011. Meroney, R.N., 1982. Wind-tunnel experiments on dense gas dispersion. J. Hazard.
Characterization and prediction of the volume ow rate aerating a cross Mater. 6 (12), 85106.
ventilated building by means of experimental techniques and numerical Meroney, R.N., Neff, D.E., Heskestad, G., 1984. Wind-tunnel simulation of eld
approaches. Energy Build. 43, 13711381. dispersion tests (by the U.K. Health and Safety Executive) of water-spray
Launder, B.E., Reece, G.J., Rodi, W., 1975. Progress in the development of a Reynolds- curtains. Bound.-Layer Meteorol. 28, 107119.
stress turbulence closure. J. Fluid Mech. 68, 537566. Meroney, R.N., 1987a. Validation of uid modelling techniques for assessing
Lawson, T.V., Penwarden, A.D., 1975. The effects of wind on people in the vicinity of hazards of dense gas cloud dispersion. J. Hazard. Mater. 15 (3), 377417.
buildings. In: Proceedings 4th International Conference on Wind Effects on Meroney, R.N., 1987b. Guidelines for uid modelling of dense gas cloud dispersion.
Buildings and Structures, Cambridge University Press, Heathrow, pp. 605622. J. Hazard. Mater. 17 (1), 2346.
Lee, I., Lee, S., Kim, G., Sung, J., Sung, S., Yoon, Y., 2005. PIV verication of Meroney, R.N., Neff, D.E., 1986. Heat transfer effects during cold dense gas
greenhouse ventilation air ows to evaluate CFD accuracy. Trans. ASAE (Am. dispersion: wind-tunnel simulation of cold gas spills. J. Heat Transf. 108, 916.
Soc. Agric. Eng.) 48, 22772288. Meroney, R.N., 1990. Fluid dynamics of ow over hills and mountains: insights
Lee, I.-B., Bitog, J.P.P., Hong, S.-W., Seo, I.-H., Kwon, K.-S., Bartzanas, T., Kacira, M., obtained through physical modelling. Chapter 7 of AMS Monograph on Current
2013. The past, present and future of CFD for agro-environmental applications. Directions in Atmospheric Processes over Complex Terrain, AMS Monograph
Comput. Electron. Agric. 93, 168183. vol. 23, Nr. 45, June 1990, pp. 145172.
Lee, R.L., Albritton, J.R., Ermak, D.L., Kim, J., 1997. Computational uid dynamics Meroney, R.N., Melbourne, W.H., 1992. Operating ranges of meteorological wind
modeling for emergency preparedness and response. Environ. Model. Softw. 12 tunnels for the simulation of convective boundary layer (CBL) phenomena.
(1), 4350. Bound.-Layer Meteorol. 61, 145174.
Leitl, B.M., Meroney, R.N., 1997. Car exhaust dispersion in a street canyon. Meroney, R.N., Pavageau, M., Rafailidis, S., Schatzmann, M., 1996. Study of line
Numerical critique of a wind tunnel experiments. J. Wind Eng. Ind. Aerodyn. source characteristics for 2-D physical modelling of pollutant dispersion in
67&68, 293304. street canyons. J. Wind Eng. Ind. Aerodyn. 62 (1), 3756.
Leitl, B.M., Kastner-Klein, P., Rau, M., Meroney, R.N., 1997. Concentration and ow Meroney, R.N., 1997. Preface to special issue of second international symposium on
distributions in the vicinity of U-shaped buildings: wind-tunnel and computa- Computational Wind Engineering (CWE-96). J. Wind Eng. Ind. Aerodyn. 6768,
tional data. J. Wind Eng. Ind. Aerodyn. 67&68, 745755. viiviii.
Leschziner, M.A., 1990. Modelling engineering ows with Reynolds stress turbu- Meroney, R.N., Neff, D.E., Chang, C.H., 2002. Diagnosis of a sick building by the wind
lence closure. J. Wind Eng. Ind. Aerodyn. 35, 2147. doctors. In: Proceedings of 2nd International Symposium on Advances in Wind
Leschziner, M.A., 1993. Computational modelling of complex turbulent ow and Structures (AWAS '02), 2123 August 2002, Pusan Convention Center,
expectations, reality and prospects. J. Wind Eng. Ind. Aerodyn. 4647, 3751. Pusan, Korea.
Letchford, C.W., Chay, M.T., 2002. Pressure distributions on a cube in a simulated Meroney, R.N., 2003. Fire whirls, re tornadoes and restorms: physical and
thunderstorm downburst. Part B: moving downburst observations. J. Wind Eng. numerical modelling. In: Proceedings of PHYSMOD2003: International Work-
Ind. Aerodyn. 90 (7), 733753. shop on Physical Modelling of Flow and Dispersion Phenomena, 35 September
Letchford, C.W., Mans, C., Chay, M.T., 2002. Thunderstorms their importance in 2003, Prato, Italy.
wind engineering (a case for the next generation wind tunnel). J. Wind Eng. Ind. Meroney, R.N., 2004. Wind tunnel and numerical simulation of pollution dispersion: a
Aerodyn. 90 (1215), 14151433. hybrid approach. Paper for Invited Lecture at the Croucher Advanced Study
Leung, D.YC., Yang, Y., 2012. Wind energy development and its environmental Institute, Hong Kong University of Science and Technology, 610 December 2004.
impact: a review. Renew. Sustain. Energy Rev. 16 (1), 10311039. Meroney, R.N. 2009. CFD prediction of airow in buildings for natural ventilation.
Li, W.W., Meroney, R.N., 1983a. Gas dispersion near a cubical model building. Part II. In: Proceedings 11th Americas Conference on Wind Engineering, San Juan,
Concentration uctuation measurements. J. Wind Eng. Ind. Aerodyn. 12 (1), Puerto Rico, pp. 111.
3547. Meroney, R.N., 2014. Personal Commucation, March 7, 2014.
Li, W.W., Meroney, R.N., 1983b. Gas dispersion near a cubical model building. Miles, S.D., Westbury, P.S., 2002. Assessing CFD as a tool for practical wind
Part I. Mean concentration measurements. J. Wind Eng. Ind. Aerodyn. 12 (1), engineering applications. In: Proceedings Fifth UK Conference on Wind
1533. Engineering, September.
Li, Y., Stathopoulos, T., 1997. Numerical evaluation of wind-induced dispersion of Miller, A., Chang, B., Issa, R., Chen, G., 2013. Review of computer-aided numerical
pollutants around a building. J. Wind Eng. Ind. Aerodyn. 67&68, 757766. simulation in wind energy. Renew. Sustain. Energy Rev. 25, 122134.
Li, X.X., Liu, C.H., Leung, D.Y.C., Lam, K.M., 2006. Recent progress in CFD modelling Milliez, M., Carissimo, B., 2007. Numerical simulations of pollutant dispersion in an
of wind eld and pollutant transport in street canyons. Atmos. Environ. 40 (29), idealized urban area, for different meteorological conditions. Bound.-Layer
56405658. Meteorol. 122 (2), 321342.
Li, Y.G., Delsante, A., 2001. Natural ventilation induced by combined wind and Milliez, M., Carissimo, B., 2008. Computational uid dynamical modelling of
thermal forces. Build. Environ. 36, 5971. concentration uctuations in an idealized urban area. Bound.-Layer Meteorol.
Linden, P.F., 1999. The uid mechanics of natural ventilation. Annu. Rev. Fluid Mech. 127 (2), 241259.
31, 201238. Mistriotis, A., Arcidiacono, C., Picuno, P., Bot, G.P.A., Scarascia-Mugnozza, G., 1997a.
Livesey, F., Inculet, D., Isyumov, N., Davenport, A.G., 1990. A scour technique for Computational analysis of ventilation in greenhouses at zero-and low-wind-
evaluation of pedestrian winds. J. Wind Eng. Ind. Aerodyn. 36, 779789. speeds. Agric. For. Meteorol. 88, 121135.
Livingstone, I., Wiggs, G.F.S., Weaver, C.M., 2007. Geomorphology of desert sand Mistriotis, A., Bot, G., Picuno, P., Scarascia-Mugnozza, G., 1997b. Analysis of the
dunes: a review of recent progress. Earth-Sci. Rev. 80 (34), 239257. efciency of greenhouse ventilation using computational uid dynamics. Agric.
Lo, L.J., Banks, D., Novoselac, A., 2013. Combined wind tunnel and CFD analysis for For. Meteorol. 85, 217228.
indoor airow prediction of wind-driven cross-ventilation. Build. Environ. 60, Mistriotis, A., Briassoulis, D., 2002. Numerical estimation of the internal and
1223. external aerodynamic coefcients of a tunnel greenhouse structure with
Lynch, P., 2006. The Emergence of Numerical Weather Prediction: Richardson's openings. Comput. Electron. Agric. 34, 191205.
Dream. Cambridge University Press, Cambridge, UK, pp. 127 Mochida, A., Iizuka, S., Tominaga, Y., Lun, I.Y.F., 2011. Up-scaling CWE models to
Lynch, P., 2008. The origins of computer weather prediction and climate modelling. include mesoscale meteorological inuences. J. Wind Eng. Ind. Aerodyn. 99 (4),
J. Comput. Phys. 227, 34313444. 187198.
100 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Mochida, A., Lun, I.Y.F., 2008. Prediction of wind environment and thermal comfort Murakami, S., Ooka, R., Mochida, A., Yoshida, S., Kim, S., 1999. CFD analysis of wind
at pedestrian level in urban area. J. Wind Eng. Ind. Aerodyn. 96 (1011), climate from human scale to urban scale. J. Wind Eng. Ind. Aerodyn. 81 (13),
14981527. 5781.
Mochida, A., Murakami, S., Shoji, M., Ishida, Y., 1993. Numerical simulation of ow NEN, 2006a. Wind Comfort and Wind Danger in the Built Environment, NEN 8100
eld around Texas Tech Building by Large Eddy Simulation. J. Wind Eng. Ind. (in Dutch) Dutch Standard.
Aerodyn. 4647, 455460. NEN, 2006b. Application of Mean Hourly Wind Speed Statistics for the Netherlands,
Mochida, A., Tabata, Y., Iwata, T., Yoshino, H., 2008. Examining tree canopy models NPR 6097:2006 (in Dutch). Dutch Practice Guideline.
for CFD prediction of wind environment at pedestrian level. J. Wind Eng. Ind. Nicholls, M., Pielke, R., Meroney, R., 1993. Large eddy simulation of microburst
Aerodyn. 96 (1011), 16671677. winds owing around a building. J. Wind Eng. Ind. Aerodyn. 46-47, 229237.
Mochida, A., Tominaga, Y., Murakami, S., Yoshie, R., Ishihara, T., Ooka, R., 2002. Nikas, K.S., Nikolopoulos, N., Nikolopoulos, A., 2010. Numerical study of a naturally
Comparison of various k models and DSM applied to ow around a high-rise cross-ventilated building. Energy Build. 42, 422434.
buildingreport on AIJ cooperative project for CFD prediction of wind envir- Nore, K., Blocken, B., Thue, J.V., 2010. On CFD simulation of wind-induced airow in
onment. Wind Struct. 5 (24), 227244. narrow ventilated facade cavities: coupled and decoupled simulations and
Mochida, A., Yoshino, H., Takeda, T., Kakegawa, T., Miyauchi, S., 2005. Methods for modelling limitations. Building and Environment 45 (8), 18341846.
controlling airow in and around a building under cross-ventilation to improve Norton, T., Sun, D.W., 2006. Computational uid dynamics (CFD) an effective and
indoor thermal comfort. J. Wind Eng. Ind. Aerodyn. 93, 437449. efcient design and analysis tool for the food industry: a review. Trends Food
Mochida, A., Yoshino, H., Miyauchi, S., Mitamura, T., 2006. Total analysis of cooling Sci. Technol. 17, 600620.
effects of cross-ventilation affected by microclimate around a building. Sol. Norton, T., Sun, D.W., Grant, J., Fallon, R., Dodd, V., 2007. Applications of computa-
Energy 80, 371382. tional uid dynamics (CFD) in the modelling and design of ventilation systems
Montazeri, H., Montazeri, F., Azizian, R., Mostafavi, S., 2010. Two-sided wind catcher in the agricultural industry: a review. Bioresour. Technol. 98 (12), 23862414.
performance evaluation using experimental, numerical and analytical model- Norton, T., Grant, J., Fallon, R., Sun, D.W., 2009. Assessing the ventilation effective-
ing. Renew. Energy 35 (7), 14241435. ness of naturally ventilated livestock buildings under wind dominated condi-
Montazeri, H., 2011. Experimental and numerical study on natural ventilation perfor- tions using computational uid dynamics. Biosyst. Eng. 103, 7899.
mance of various multi-opening wind catchers. Build. Environ. 46 (2), 370378. Norton, T., Grant, J., Fallon, R., Sun, D.W., 2010a. Optimising the ventilation
Montazeri, H, Blocken, B., 2013. CFD simulation of wind-induced pressure coef- conguration of naturally ventilated livestock buildings for improved indoor
cients on buildings with and without balconies: validation and sensitivity environmental homogeneity. Build. Environ. 45, 983995.
analysis. Build. Environ. 60, 137149. Norton, T., Grant, J., Fallon, R., Sun, D.W., 2010b. Improving the representation of
Montazeri, H, Blocken, B, Janssen, WD, van Hooff, T., 2013. CFD evaluation of new thermal boundary conditions of livestock during CFD modelling of the indoor
second-skin facade concept for wind comfort on building balconies: case-study environment. Comput. Electron. Agric. 73, 1736.
for the Park Tower in Antwerp. Build. Environ. 68, 179192. Oberkampf, W.L., Trucano, T.G., Hirsch, C., 2004. Verication, validation, and
Monteiro, J.P., Viegas, D.X., 1996. On the use of Irwin and Preston wall shear stress predictive capability in computational engineering and physics. Appl. Mech.
probes in turbulent incompressible ows with pressure gradients. J. Wind Eng. Rev. 57 (5), 345384.
Ind. Aerodyn. 64 (1), 1529. Orlanski, I., 1975. A rational subdivision of scales for atmospheric processes. Bull.
Moonen, P., Blocken, B., Roels, S., Carmeliet, J., 2006. Numerical modeling of the Am. Meteorol. Soc. 56, 527530.
ow conditions in a low-speed closed-circuit wind tunnel. J. Wind Eng. Ind. Parente, A., Gorl, C., van Beeck, J., Benocci, C., 2011. Improved k model and wall
Aerodyn. 94 (10), 699723. function formulation for the RANS simulation of ABL ows. J. Wind Eng. Ind.
Moonen, P., Blocken, B., Carmeliet, J., 2007. Indicators for the evaluation of wind Aerodyn. 99 (4), 267278.
tunnel test section ow quality and application to a numerical closed-circuit Paterson, D.A., Apelt, C.J., 1986. Computation of wind ows over three-dimensional
wind tunnel. J. Wind Eng. Ind. Aerodyn. 95 (911), 12891314. buildings. J. Wind Eng. Ind. Aerodyn. 24, 192213.
Moonen, P., Defraeye, T., Dorer, V., Blocken, B., Carmeliet, J., 2012. Urban Physics: Paterson, D.A., Apelt, C.J., 1989. Simulation of wind ow around three-dimensional
Effect of the micro-climate on comfort, health and energy demand. Front. buildings. Build. Environ. 24, 3950.
Archit. Res. 1 (3), 197228. Paterson, D.A., Apelt, C.J., 1990. Simulation of ow past a cube in a turbulent
Moonen, P., Gromke, C., Dorer, V., 2013. Performance assessment of Large Eddy boundary layer. J. Wind Eng. Ind. Aerodyn. 35, 149176.
Simulation (LES) for modeling dispersion in an urban street canyon with tree Pearce, R.P., 1955. The calculation of the sea breeze circulation in terms of the
planting. Atmos. Environ. 75, 6676. differential heating across the coast line. Quart. J. R. Meteorol. Soc. 81, 351381.
Moussiopoulos, N., Schlnzen, K.H., Louka, P., 2003. Modelling urban air pollution Persoon, J., van Hooff, T., Blocken, B., Carmeliet, J., de Wit, M.H., 2008. Impact of roof
(Chap. 6). In: Moussiopoulos, N. (Ed.), Air Quality in Cities. SpringerVerlag, Berlin, geometry on rain shelter in football stadia. J. Wind Eng. Ind. Aerodyn. 96 (89),
pp. 121154 12741293.
Mueller, T.J., DeLaurier, J.D., 2003. Aerodynamics of small vehicles. Annu. Rev. Fluid Phillips, N.A., 1956. The general circulation of the atmosphere: a numerical
Mech. 35, 89111. experiment. Q. J. R. Meteorol. Soc. 82, 123164.
Murakami, S., Mochida, A., Hayashi, Y., 1990. Examining the k- model by means of Phillips, N.A., 1960. On the problem of initial data for the primitive equations. Tellus
a wind tunnel test and large-eddy simulation of the turbulence structure 12, 121126.
around a cube. J. Wind Eng. Ind. Aerodyn. 35, 87100. Pielke, R.A., Nicholls, M.E., 1997. Use of meteorological models in computational
Murakami, S., 1990a. Preface to special issue on computational wind engineering. wind engineering. J. Wind Eng. Ind. Aerodyn. 6768, 363372.
J. Wind Eng. Ind. Aerodyn. 35 (1), ixxi. Platzman, G.W., 1979. The ENIAC computations of 1950 gateway to numerical
Murakami, S., 1990b. Computational wind engineering. J. Wind Eng. Ind. Aerodyn. weather prediction. Bull. Am. Meteorol. Soc. 60, 302312.
36 (1), 517538. Port-Agel, F., Wu, Y.T., Lu, H., Conzemius, R.J., 2011. Large-eddy simulation of
Murakami, S., 1990c. Numerical simulation of turbulent oweld around cubic atmospheric boundary layer ow through wind turbines and wind farms. J.
model: current status and applications of ke model and LES. J. Wind Eng. Ind. Wind Eng. Ind. Aerodyn. 99 (4), 154168.
Aerodyn. 33 (12), 139152. Ramponi, R., Blocken, B., 2012a. CFD simulation of cross-ventilation for a generic
Murakami, S., Kato, S., Akabayashi, S., Mizutani, K., Kim, Y., 1991. Wind tunnel test isolated building: impact of computational parameters. Build. Environ. 53,
on velocity-pressure eld of cross-ventilation with open windows. ASHRAE 3448.
Trans. 97, 525538. Ramponi, R, Blocken, B., 2012b. CFD simulation of cross-ventilation ow for
Murakami, S., 1993a. Preface to the rst international symposium on Computa- different isolated building congurations: validation with wind tunnel mea-
tional Wind Engineering (CWE92). J. Wind Eng. Ind. Aerodyn. 4647, 12. surements and analysis of physical and numerical diffusion effects. J. Wind Eng.
Murakami, S., 1993b. Comparison of various turbulence models applied to a bluff Ind. Aerodyn. 104106, 408418.
body. J. Wind Eng. Ind. Aerodyn. 46 and 47, 2136. Randerson, D., 1976. An overview of regional-scale numerical models. Bull. Am.
Murakami, S., 1997. Current status and future trends in computational wind Meteorol. Soc. 57, 797804.
engineering. J. Wind Eng. Ind. Aerodyn. 67 and 68, 334. Reichrath, S., Davies, T.W., 2002. Using CFD to model the internal climate of
Murakami, S., 1998. Overview of turbulence models applied in CWE-1997. J. Wind greenhouses: past, present and future. Agronomie 22, 319.
Eng. Ind. Aerodyn. 74-76, 124. Richards, P.J., Hoxey, R.P., 1993. Appropriate boundary conditions for computational
Murakami, S., Matsumoto, M., Tamura, Y., 2008. Preface to the special issue of the wind engineering models using the k turbulence model. J. Wind Eng. Ind.
fth symposium on Computational Wind Engineering. J. Wind Eng. Ind. Aerodyn. 46&47, 145153.
Aerodyn. 96 (1011), 14491450. Richards, P.J., Mallison, G.D., McMillan, D., Li, Y.F., 2002. Pedestrian level wind
Murakami, S., Mochida, A., 1988. 3-D numerical simulation of airow around a cubic speeds in downtown Auckland. Wind Struct. 5 (24), 151164.
model by means of the k model. J. Wind Eng. Ind. Aerodyn. 31 (23), 283303. Richards, P.J., Norris, S.E., 2011. Appropriate boundary conditions for computational
Murakami, S., Mochida, A., Hibi, K., 1987. Three-dimensional numerical simulation wind engineering models revisited. J. Wind Eng. Ind. Aerodyn. 99 (4), 257266.
of airow around a cubic model by means of large eddy simulation. J. Wind Richardson, L.F., 1922. Weather Prediction by Numerical Process. The University
Eng. Ind. Aerodyn. 25, 291305. Press, Cambridge
Murakami, S., Mochida, A., 1989. Three-dimensional numerical simulation of Roache, P.J., 1994. Perpective a method for uniform reporting of grid renement
turbulent ow around buildings using the k  turbulence model. Build. studies. J. Fluids Eng. Trans. ASME 116 (3), 405413.
Environ. 24 (1), 5164. Roache, P.J., 1997. Quantication of uncertainty in computational uid dynamics.
Murakami, S., Mochida, A., Hayashi, Y., Sakamoto, S., 1992. Numerical study on Annu. Rev. Fluid Mech. 29, 123160.
velocity-pressure eld and wind forces for bluff bodies by k, ASM and LES. J. Roache, P.J., Chia, K.N., White, F., 1986. Editorial policy statement on the control of
Wind Eng. Ind. Aerodyn. 44 (13), 28412852. numerical accuracy. J. Fluids Eng. 108, 2.
 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102 101

Robins, A., 2003. Wind tunnel dispersion modelling some recent and not so recent Shuman, F.G., 1989. History of numerical weather prediction at the National
achievements. J. Wind Eng. Ind. Aerodyn. 91 (1215), 17771790. Meteorological Center. Weather Forecast 4, 286296.
Robins, A.G., Castro, I.P., 1977a. Wind-tunnel investigation of plume dispersion in Shuman, F.G., Hovermale, J., 1968. An operational six-layer primitive equation
vicinity of a surface mounted cube. 1. Flow eld. Atmos. Environ. 11 (4), model. J. Appl. Meteorol. 7, 525547.
291297. Smagorinsky, J., 1953. The dynamical inuence of large-scale heat sources and sinks
Robins, A.G., Castro, I.P., 1977b. Wind-tunnel investigation of plume dispersion in on the quasi-stationary mean motions of the atmosphere. Q. J. R. Meteorol. Soc.
vicinity of a surface mounted cube. 2. Concentration eld. Atmos. Environ. 11 79, 342366.
(4), 299311. Smagorinsky, J., 1958. On the numerical integration of the primitive equations of
Robson, B.J., Hamilton, D.P., Webster, I.T., Chan, T., 2008. Ten steps applied to motion for baroclinic ow in a closed region. Mon. Weather Rev. 86 (12), 457466.
development and evaluation of process-based biogeochemical models of Smagorinsky, J., 1963. General circulation experiments with the primitive equations
estuaries. Environ. Model. Softw. 23 (4), 369384. I. The basic experiment. Mon. Weather Rev. 91 (3), 99164.
Rodi, W., 1997. Comparison of LES and RANS calculations of the ow around bluff Smagorinsky, J., 1969. Problems and promises of deterministic extended range
bodies. J. Wind Eng. Ind. Aerodyn. 6971, 5575. forecasting. Bull. Am. Meteorol. Soc. 50 (5), 286311.
Roy, C.J., 2005. Review of code and solution verication procedures for computa- Smedley, D.J., Kwok, K.C.S., Kim, D.H., 1993. Snowdrifting simulation around Davis
tional simulation. J. Comput. Phys. 205 (1), 131156. Station workshop, Antartica. J. Wind Eng. Ind. Aerodyn. 50, 153162.
Roy, C.J, Oberkampf, W.L., 2010. A complete framework for verication, validation, Snel, H., 2003. Review of aerodynamics for wind turbines. Wind Energy 6 (3),
and uncertainty quantication in scientic computing. 48th AIAA Aerospace 203211.
Sciences Meeting Including the New Horizons Forum and Aerospace Exposition Solari, G., 2007. The International Association for Wind Engineering (IAWE):
47 January 2010, Orlando, Florida.
progress and prospects. J. Wind Eng. Ind. Aerodyn. 95, 813842.
Roy, S., Saha, U.K., 2013. Review on the numerical investigations into the design and
Sorensen, J.N., 2011. Aerodynamic aspects of wind energy conversion. Annu. Rev.
development of Savonius wind rotors. Renew. Sustain. Energy Rev. 24, 7383.
Fluid Mech. 43, 427448.
Salim, S.M., Cheah, S.C., Chan, A., 2011a. Numerical simulation of dispersion in
Souster, C., 1979. A Theoretical Approach to Predicting Snow Loads and Driving
urban street canyons with avenue-like tree plantings: comparison between
Rain Deposition on Buildings (Ph.D. thesis). University of Shefeld, UK
RANS and LES. Build. Environ. 46 (9), 17351746.
Squires, K.D., Krishnan, V., Forsythe, J.R., 2008. Prediction of the ow over a circular
Salim, S.M., Buccolieri, R., Chan, A., Di Sabatino, S., 2011b. Numerical simulation of
cylinder at high Reynolds number using detached-eddy simulation. J. Wind
atmospheric pollutant dispersion in an urban street canyon: comparison
Eng. Ind. Aerodyn. 96 (1011), 15281536.
between RANS and LES. J. Wind Eng. Ind. Aerodyn. 99, 103113.
Stathopoulos, T., 1997. Computational Wind Engineering: past achievements and
Saloranta, J., Hellsten, A., 2011. Evaluation of a general CFD-solver for a micro-scale
urban ow. Int. J. Environ. Pollut. 44 (14), 368375. future challenges. J. Wind Eng. Ind. Aerodyn. 6768, 509532.
Sandberg, M., 2004. An alternative view on the theory of cross-ventilation. Int. J. Stathopoulos, T., 2002. The numerical wind tunnel for industrial aerodynamics: real
Vent. 4, 409418. or virtual in the new millennium? Wind Struct. 5 (24), 193208.
Sanderse, B., van der Pijl, S.P., Koren, B., 2011. Review of computational uid Stathopoulos, T., 2003. Wind loads on low buildings: in the wake of Alan
dynamics for wind turbine wake aerodynamics. Wind Energy 14 (7), 799819. Davenport's contributions. J. Wind Eng. Ind. Aerodyn. 91 (1215), 15651585.
Santiago, J..L., Dejoan, A., Martilli, A., Martin, F., Pinelli, A., 2010. Comparison Stathopoulos, T., 2006. Pedestrian level winds and outdoor human comfort. J. Wind
between Large-Eddy Simulation and Reynolds-Averaged Navier-Stokes compu- Eng. Ind. Aerodyn. 94 (11), 769780.
tations for the MUST eld experiment. Part I: study of the ow for an incident Stathopoulos, T., Storms, R., 1986. Wind environmental conditions in passages
wind directed perpendicularly to the front array of containers. Bound.-Layer between buildings. J. Wind Eng. Ind. Aerodyn. 24 (1), 1931.
Meteorol. 135 (1), 109132. Stathopoulos, T., Baskaran, A., 1990. Boundary treatment for the computation of 3D
Sanz Rodrigo, J., van Beeck, J., Buchlin, J.M., 2012. Wind engineering in the turbulent conditions around buildings. J. Wind Eng. Ind. Aerodyn. 35, 177200.
integrated design of princess Elisabeth Antarctic base. Build. Environ. 52, 118. Stathopoulos, T., Baskaran, A., 1996. Computer simulation of wind environmental
Sasaki, R., Uematsu, Y., Yamada, M., Saeki, H., 1997. Application of infrared conditions around buildings. Eng. Struct. 18 (11), 876885.
thermography and a knowledge-based system to the evaluation of the Stathopoulos, T., 2013. Personal Communication, 9 May 2013.
pedestrian-level wind environment around buildings. J. Wind Eng. Ind. Aero- Stavridou, A.D., Prinos, P.E., 2013. Natural ventilation of buildings due to buoyancy
dyn. 6768, 873883. assisted by wind: investigating cross ventilation with computational and
Sawyer, J.S., 1960. Numerical calculation of the displacements of a stratied laboratory simulation. Build. Environ. 66, 104119.
airstream crossing a ridge. Quart. J. Roy. Meteorol. Soc. 86, 326345. Stavrakakis, G.M., Koukou, M.K., Vrachopoulos, M.G., Markatos, N.C., 2008. Natural
Scaperdas, A., Gilham, S., 2004. Thematic Area 4: best practice advice for civil cross-ventilation in buildings: building-scale experiments, numerical simula-
construction and HVAC. QNET-CFD Netw. Newsl. 2 (4), 2833. tion and thermal comfort evaluation. Energy Build. 40, 16661681.
Schatzmann, M., Rafailidis, S., Pavageau, M., 1997. Some remarks on the validation Straw, M.P., Baker, C.J., Robertson, A.P., 2000. Experimental measurements and
of small-scale dispersion models with eld and laboratory data. J. Wind Eng. computations of the wind-induced ventilation of a cubic structure. J. Wind Eng.
Ind. Aerodyn. 67 and 68, 885893. Ind. Aerodyn. 88, 213230.
Schatzmann, M., Leitl, B., 2011. Issues with validation of urban ow and dispersion Sumner, J., Watters, C.S., Masson, C., 2010. CFD in wind energy: the virtual,
CFD models. J. Wind Eng. Ind. Aerodyn. 99, 169186. multiscale wind tunnel. Energies 3, 9891013.
Schlnzen, K.H., 1996. Validierung hochausender Regionalmodelle. Ber. aus dem Surry, D., Inculet, D.R., Skerlj, P.F., Lin, J.-X., Davenport, A.G., 1994. Wind, rain and
Zentrum f. Meeres- und Klimaforschung, Meteorologisches Institut, Universitt the building envelope: a status report of ongoing research at the University of
Hamburg, A23, 184. http://www.bis.zmaw.de/leadmin/Bib/Voll texte/ZMK-A23. Western Ontario. J. Wind Eng. Ind. Aerodyn. 53, 1936.
pdf. Tablada, A., De Troyer, F., Blocken, B., Carmeliet, J., Verschure, H., 2009. On natural
Schlnzen, K.H., Grawe, D., Bohnenstengel, S.I., Schlter, I., Koppmann, R., 2011. ventilation and thermal comfort in compact urban environmentsthe Old
Joint modelling of obstacle induced and mesoscale changescurrent limits and Havana case. Build. Environ. 44, 19431958.
challenges. J. Wind Eng. Ind. Aerodyn. 99 (4), 217225. Tabor, G.R., Baba-Ahmadi, M.H., 2010. Inlet conditions for large eddy simulation: a
Seifert, J., Li, Y., Axley, J., Rsler, M., 2006. Calculation of wind-driven cross
review. Comput. Fluids 39 (4), 553567.
ventilation in buildings with large openings. J. Wind Eng. Ind. Aerodyn. 94,
Taddei, D., Bontempi, F., 2003. A reference framework for the aerodynamic and
925947.
aeroelastic analysis of long span bridges with computational uid dynamic.
Selvam, R.P., 1997. Numerical simulation of pollutant dispersion around a building
Syst.-Based Vis. Strateg. Creat. Des. 1-3, 25112516.
using FEM. J. Wind Eng. Ind. Aerodyn. 67&68, 805814.
Takagi, M., 1990. Application of computers to automobile aerodynamics. J. Wind
Sengupta, A., Sarkar, P.P., 2008. Experimental measurement and numerical simula-
Eng. Ind. Aerodyn. 33 (12), 419428.
tion of an impinging jet with application to thunderstorm microburst winds.
Takakura, S., Suyama, Y., Aoyama, M., 1993. Numerical simulation of
J. Wind Eng. Ind. Aerodyn. 96 (3), 345365.
oweld around buildings in an urban area. J. Wind Eng. Ind. Aerodyn. 46
Sengupta, A., Haan, F.L., Sarkar, P.P., Balaramudu, V., 2008. Transient loads
on buildings in microburst and tornado winds. J. Wind Eng. Ind. Aerodyn. 96 47, 765771.
(1011), 21372187. Tamura, T., 1999. Reliability on CFD estimation for wind-structure interaction
Shah, K.B., Ferziger, J.H., 1997. A uid mechanicians view of wind engineering: large problems. J. Wind Eng. Ind. Aerodyn. 81, 117143.
eddy simulation of ow past a cubic obstacle. J. Wind Eng. Ind. Aerodyn. 6768, Tamura, T., 2008. Towards practical use of LES in wind engineering. J. Wind Eng.
211224. Ind. Aerodyn. 96 (1011), 14511471.
Shen, X., Zhang, G., Bjerg, B., 2012. Comparison of different methods for estimating Tamura, T., Kawai, H., Kawamoto, S., Nozawa, K., Sakamoto, S., Ohkuma, T., 1997.
ventilation rates through wind driven ventilated buildings. Energy Build. 54, Numerical prediction of wind loading on buildings and structures activities of
297306. AIJ cooperative project on CFD. J. Wind Eng. Ind. Aerodyn. 67-68, 671685.
Shih, T.H., Liou, W.W., Shabbir, A., Yang, Z., Zhu, J., 1995. A new k eddy viscosity Tamura, T., Nozawa, K., Kondo, K., 2008. AIJ guide for numerical prediction of wind
model for high Reynolds number turbulent ows. Comput. Fluids 24, 227238. loads on buildings. J. Wind Eng. Ind. Aerodyn. 96 (1011), 19741984.
Shin, S.H., Meroney, R.N., 1988. Surface pattern comparability of wind-tunnel simula- Tamura, Y., Matsui, M., 2002. Recent topics in wind engineering focusing on
tions of the Thorney Island dense gas dispersion trials. 17th NATO/COMS Interna- monitoring techniques. Adv. Build. Technol. (I), 6574
tional Technical Meeting on Air Pollution Modelling and its Application. Cambridge, Tan, G., Glicksman, L.R., 2005. Application of integrating multi-zone model
UK, 1922 September 1988, Downing College, Cambridge University. with CFD simulation to natural ventilation prediction. Energy Build. 37,
Shklyar, A., Arbel, A., 2004. Numerical model of the three-dimensional isothermal 10491057.
ow patterns and mass uxes in a pitched-roof greenhouse. J. Wind Eng. Ind. Tang, W., Davidson, C.I., 2004. Erosion of limestone building surfaces caused by
Aerodyn. 92, 10391059. wind-driven rain. 2. Numerical modelling. Atmos. Environ. 38 (33), 56015609.
102 B. Blocken / J. Wind Eng. Ind. Aerodyn. 129 (2014) 69102

Tang, W., Davidson, C.I., Finger, S., Vance, K., 2004. Erosion of limestone building van Hooff, T., Blocken, B., 2013. CFD evaluation of natural ventilation of indoor
surfaces caused by wind-driven rain: 1. Field measurements. Atmos. Environ. environments by the concentration decay method: CO2 gas dispersion from a
38 (33), 55895599. semi-enclosed stadium. Build. Environ. 61, 117.
Tari, P.H., Gurka, R., Hangan, H., 2010. Experimental investigation of tornado-like van Mook, F.J.R., 2002. Driving rain on building envelopes (Ph.D. thesis). Building
vortex dynamics with swirl ratio: the mean and turbulent ow elds. J. Wind Physics and Systems, Technische Universiteit Eindhoven, Eindhoven University
Eng. Ind. Aerodyn. 98 (12), 936944. Press, Eindhoven, The Netherlands (198 p.)
Taylor, P.A., Gent, P.R., 1974. A model of atmospheric boundary-layer ow above an Vardoulakis, S., Fisher, B.E.A., Pericleous, K., Gonzalez-Flesca, N., 2003. Modelling air
isolated two-dimensional hill; an example of ow above gentle topography. quality in street canyons: a review. Atmos. Environ. 37 (2), 155182.
Bound.-Layer Meteorol. 7, 349362. Vasilic-Melling, D., 1977. Three dimensional turbulent ow past rectangular bluff
Teitel, M., Ziskind, G., Liran, O., Dubovsky, V., Letan, R., 2008. Effect of wind bodies (Ph.D. thesis). Imperial College of Science and Technology, London
direction on greenhouse ventilation rate, airow patterns and temperature Vermeer, L.J., Srensen, J.N., Crespo, A., 2003. Wind turbine wake aerodynamics.
distributions. Biosyst. Eng. 101, 351369. Prog. Aerosp. Sci. 39 (67), 467510.
Tezduyar, T.E., 1999. CFD methods for three-dimensional computation of complex Wang, L., Wong, N.H., 2008. Coupled simulations for naturally ventilated residential
ow problems. J. Wind Eng. Ind. Aerodyn. 81 (13), 97116. buildings. Autom. Constr. 17, 386398.
To, A.P., Lam, K.M., 1995. Evaluation of pedestrian-level wind environment around a Wallington, C.E., Portnall, J., 1958. A numerical study of the wavelength and
row of tall buildings using a quartile-level wind speed descripter. J. Wind Eng. amplitude of lee waves, Quart. J. Roy. Meteorol. Soc. 84, 3845.
Ind. Aerodyn. 5455, 527541. Wang, L., Wong, N.H., 2009. Coupled simulations for naturally ventilated rooms
Tominaga, Y., Murakami, S., Mochida, A., 1997. CFD prediction of gaseous diffusion
between building simulation (BS) and computational uid dynamics (CFD) for
around a cubic model using a dynamics mixed SGS model based on composite
better prediction of indoor thermal environment. Build. Environ. 44, 95112.
grid technique. J. Wind Eng. Ind. Aerodyn. 67-68, 827841.
Westbury, P.S., Miles, S.D., Stathopoulos, T., 2002. CFD Application on the Evaluation
Tominaga, Y., Stathopoulos, T., 2007. Turbulent Schmidt numbers for CFD analysis
of Pedestrian-level Winds. Workshop on Impact of Wind and Storm on City Life
with various types of oweld. Atmos. Environ. 41 (37), 80918099.
and Built Environment, Cost Action C14, CSTB, June 34, Nantes, France.
Tominaga, Y., Stathopoulos, T., 2009. Numerical simulation of dispersion around an
Wilcox, D.C., 1998. Turbulence modeling for CFD. La Canada. DCW Industries, Inc.,
isolated cubic building: comparison of various types of k models. Atmos.
California
Environ. 43 (20), 32003210.
Willemsen, E., Wisse, J.A., 2007. Design for wind comfort in The Netherlands:
Tominaga, Y., Stathopoulos, T., 2010. Numerical simulation of dispersion around an
isolated cubic buildings: model evaluation of RANS and LES. Build. Environ. 45 procedures, criteria and open research issues. J. Wind Eng. Ind. Aerodyn.
(10), 22312239. 95 (911), 15411550.
Tominaga, Y., Stathopoulos, T., 2013. CFD simulation of near-eld pollutant disper- Wise, A.F.E., 1970. Wind effects due to groups of buildings. In: Proceedings of the
sion in the urban environment: a review of current modelling techniques. Royal Society Symposium Architectural Aerodynamics, London.
Atmos. Environ. 79, 716730. Wisse, J.A., Willemsen, E., 2003. Standardization of wind comfort evaluation in the
Tominaga, Y., Mochida, A., Yoshie, R., Kataoka, H., Nozu, T., Yoshikawa, M., Netherlands. In: Proceedings of the 11th International Conference on Wind
Shirasawa, T., 2008a. AIJ guidelines for practical applications of CFD to Engineering (11ICWE), Lubbock, Texas.
pedestrian wind environment around buildings. J. Wind Eng. Ind. Aerodyn. Wood, G.D., Kwok, K.C.S., Motteram, N.A., Fletcher, D.F., 2001. Physical and
96 (1011), 17491761. numerical modelling of thunderstorm downbursts. J. Wind Eng. Ind. Aerodyn.
Tominaga, Y., Mochida, A., Murakami, S., Sawaki, S., 2008b. Comparison of various 89 (6), 535552.
revised k models and LES applied to ow around a high-rise building model Wood, N., 2000. Wind ow over complex terrain: a historical perspective and the
with 1:1:2 shape placed within the surface boundary layer. J. Wind Eng. Ind. prospect for large-eddy modelling. Bound.-Layer Meteorol. 96, 1132.
Aerodyn. 96 (4), 389411. Wright, N.G., Hargreaves, D.M., 2006. Unsteady CFD Simulations for natural
Tominaga, Y., Okaze, T., Mochida, A., 2011. CFD modeling of snowdrift around a ventilation. Int. J. Vent. 5, 1320.
building: an overview of models and evaluation of a new approach. Build. Wright, N.G., Easom, G.J., Hoxey, R.J., 2001. Development and validation of a non-
Environ. 46 (4), 899910. linear k-epsilon: model for ow over a full-scale building. Wind Struct. 4 (3),
Tominaga, Y., Iizuka, S., Imano, M., Kataoka, H., Mochida, A., Nozu, T., Ono, Y., 177196.
Shirasawa, T., Tsuchiya, N., Yoshie, R., 2013. Cross comparisons of CFD results of Wu, H., Stathopoulos, T., 1994. Further experiments on Irwin's surface wind sensor.
wind and dispersion elds for MUST experiment: evaluation exercises by AIJ. J. J. Wind Eng. Ind. Aerodyn. 53 (3), 441452.
Asian Archit. Building Eng. 12 (1), 117124. Wu, H.Q., Stathopoulos, T., 1997. Application of infrared thermography for pedes-
Tsang, C.W., Kwok, K.C.S., Hitchcock, P.A., 2012. Wind tunnel study of pedestrian trian wind evaluation. J. Eng. Mech. ASCE, 123; , pp. 978985.
level wind environment around tall buildings: effects of building dimensions, Wu, G., Higuchi, K., Meroney, R.N., 1992. Applications of digital image processing in
separation and podium. Build. Environ. 49, 167181. wind engineering. J. Wind Eng. Ind. Aerodyn. 42 (13), 9991010.
Tucker, P.G., Mosquera, A., 2001. NAFEMS Introduction to Grid and Mesh Generation Wu, Y.C., Yang, A.S., Tseng, L.Y., Liu, C.L., 2011. Myth of ecological architecture
for CFD. NAFEMS CFD Working Group, R0079, 56 pp. designs: comparison between design concept and computational analysis
Tyagi, S.K., Pandey, A.K., Pant, P.C., Tyagi, V.V., 2012. Formation, potential and results of natural-ventilation for tjibaou cultural center in new caledonia.
abatement of plume from wet cooling towers: a review. Renew. Sustain. Energy Energy Build. 43, 27882797.
Rev. 16 (5), 34093429. Wu, W., Zhai, J., Zhang, G., Nielsen, P.V., 2012. Evaluation of methods for determining air
UWO, 2014. http://www.eng.uwo.ca/windeee/facilities.html. University of Wes- exchange rate in a naturally ventilated dairy cattle building with large openings
tern Ontario. (Retrieved March 04.03.14.). using computational uid dynamics (CFD). Atmos. Environ. 63, 179188.
Uematsu, Y., Yamada, M., Higashiyama, H., Orimo, T., 1992. Effects of the corner
Xie, Z.-T., Castro, I.P., 2008. Efcient generation of inow conditions for large eddy
shape of high-rise buildings on the pedestrian-level wind environment with
simulation of street-scale ows. Flow, Turbul. Combust. 81, 449470.
consideration for mean and uctuating wind speeds. J. Wind Eng. Ind. Aerodyn.
Xu, Z., Hangan, H., 2008. Scale, boundary and inlet condition effects on impinging
44 (13), 22892300.
jets. J. Wind Eng. Ind. Aerodyn. 96 (12), 23832402.
Van Beeck, J.P.A.J., Dezs, G., Planquart, P., 2009. Microclimate Assessment by Sand
Yakhot, V., Orszag, S.A., 1986. Renormalization group analysis of turbulence. J. Sci.
Erosion and Irwin Probes for Atmospheric Boundary Layer Wind Tunnels.
Comput. 1 (1), 351.
PHYSMOD 2009, Von Karman Institute (VKI), Sint-Genesius-Rode, Belgium.
Yamada, M., Uematsu, Y., Sasaki, R., 1996. A visual technique for the evaluation of
van Hooff, T, Blocken, B., 2010a. Coupled urban wind ow and indoor natural
the pedestrian-level wind environment around buildings by using infrared
ventilation modelling on a high-resolution grid: a case study for the Amster-
dam ArenA stadium. Environ. Model. Softw. 25 (1), 5165. thermography. J. Wind Eng. Ind. Aerodyn. 65 (13), 261271.
van Hooff, T., Blocken, B., 2010b. On the effect of wind direction and urban Yamada, T., Katsuyuki, K., 2011. Downscaling mesoscale meteorological models for
surroundings on natural ventilation of a large semi-enclosed stadium. Comput. computational wind engineering applications. J. Wind Eng. Ind. Aerodyn. 99
Fluids 39, 11461155. (4), 199216.
van Hooff, T., Blocken, B., Aanen, L., Bronsema, B., 2011a. A venturi-shaped roof for Yamada, T., Meroney, R.N. 1972. Numerical and wind tunnel simulation of airow
wind-induced natural ventilation of buildings: wind tunnel and CFD evaluation over an obstacle. National Conference on Atmospheric Waves, American
of different design congurations. Build. Environ. 46 (9), 17971807. Meteorological Society, Salt Lake City, October 12-15, 1971.
van Hooff, T., Blocken, B., van Harten, M., 2011b. 3D CFD simulations of wind ow Yang, Y., Gu, M., Chen, S., Jin, X., 2009. New inow boundary conditions for
and wind-driven rain shelter in sports stadia: inuence of stadium geometry. modelling the neutral equilibrium atmospheric boundary layer in computa-
Build. Environ. 46 (1), 2237. tional wind engineering. J. Wind Eng. Ind. Aerodyn. 97 (2), 8895.
van Hooff, T., Blocken, B., 2012. Full-scale measurements of indoor environmental Yoshie, R., Mochida, A., Tominaga, Y., Kataoka, H., Harimoto, K., Nozu, T., Shirasawa,
conditions and natural ventilation in a large semi-enclosed stadium: possibi- T., 2007. Cooperative project for CFD prediction of pedestrian wind
lities and limitations for CFD validation. J. Wind Eng. Ind. Aerodyn. 104106, environment in the Architectural Institute of Japan. J. Wind Eng. Ind. Aerodyn.
330341. 95 (911), 15511578.