Building and Environment 96 (2016) 12e21

Contents lists available at ScienceDirect

Building and Environment

journal homepage: www.elsevier.com/locate/buildenv

10 Questions

Ten questions concerning hybrid computational/physical model

simulation of wind flow in the built environment
Robert N. Meroney
Colorado State University, Fort Collins, CO, USA

a r t i c l e i n f o a b s t r a c t

Article history: For over fifty years the common presumption has been that computational fluid dynamics (CFD) and
Received 17 September 2015 experimental fluid dynamics (EFD) were mutually exclusive and competitive. Often the question was
Received in revised form posed: When can we get rid of our physical modeling facilities? This question does not recognize the
3 November 2015
tremendous synergistic leverage of combining the best qualities of both CFD and EFD as a research and
Accepted 4 November 2015
Available online 21 November 2015
design methodology. Coordinating the application of both CFD and EFD in a hybrid management
approach can expedite results, improve understanding of flow phenomena, and often reduce research
costs and time. This paper considers some of the common questions that arise as one considers hybrid
Keywords:
Hybrid research
research or design methods as it is applied to wind engineering and the built environment.
Computational fluid dynamics (CFD) © 2015 Elsevier Ltd. All rights reserved.
Physical modeling
Experimental flow dynamics (EFD)
Built environment
Similitude

1. Introduction 1.1. Advantages of physical modeling

Wind Engineering was first identified and defined as a specific
subtopic in engineering in the 1975 Freeman lecture prepared by
Jack E. Cermak [1]. The application of experimental fluid dynamics
(EFD)1 in fluid modeling facilities (wind or water tunnels) to wind
engineering applications began as early as 1891 when Professor la
Cour of Denmark established a windmill experimental station at
Askov, Denmark. Even earlier in 1759 John Smeaton did experi-
mental investigations of wind effects on solid objects. However,
serious application of fluid similarity and physical modeling did not
really begin until the 1940e1960 period, when attempts were
made to estimate loads on the proposed World Trade Center in New
York, the selection of stack heights for power stations, the esti-
mation of flow and dispersion in forests, and the effects of winds
and turbulence on pedestrian comfort near buildings and in cities
[1,2]. The application of computational fluid dynamics (CFD) to
wind engineering occurred somewhat later, since significant soft-
ware and computational capacity were developed only after about
1960 [3e5].
Wind or water tunnels are, in effect, analog computers that have
the advantage of “near-infinitesimal” resolution and “near-infinite
memory.” A fluid modeling study employs “real fluids” not models
of fluids; hence, the fluid model is implicitly non-hydrostatic, tur-
bulent, includes variable fluid properties, non-slip boundary con-
ditions, and dissipation. Real fluids permit flow separation and
recirculation. All conservation equations are automatically included
in their correct form without truncation or differencing errors, and
there are no missing terms or approximations. The basic equations
of motion and transport are solved by simulating the flow at a
reduced scale, and then the desired quantity is measured. Finally,
the fluid model bridges the gap between the fluid-mechanician's
analytic or numeric models of flow, turbulence and dispersion and
their application in the field. Fluid modeling may be used to plan
field experiments, provide conservative estimates of plume trans-
port, wind flows, wind forces, and validate modules of numeric
code [6e8]. Some limitations will, of course, exist such as inability
to model geophysical scales larger than test facilities.

1.2. Advantages of numerical modeling

1
In this paper experimental fluid dynamics (EFD) will be limited to physical Numerical modeling, despite its many limitations associated
modeling and not full-scale field experiments. with grid resolution, choice of turbulence model, or assignment of

http://dx.doi.org/10.1016/j.buildenv.2015.11.005
0360-1323/© 2015 Elsevier Ltd. All rights reserved.
 R.N. Meroney / Building and Environment 96 (2016) 12e21 13

boundary conditions is not intrinsically limited by similitude or advantages of an old tool, fluid (or physical or scale) modeling
scale constraints. Thus, in principle, it should be possible to (EFD), with the speed and convenience of a new technology,
numerically simulate all aspects of fluid motions, thermal stratifi- computational fluid dynamics (CFD). The traditional view is that
cation, induced forces (such as Coriolis effects), plume transport, the scientific method has two foundations, experimental and
dispersion, and/or drift. In addition it should be possible to examine theoretical (Fig. 1). While the traditional scientific method does not
all interactions of these properties individually, sequentially and acknowledge the role for computing and simulation, a new para-
combined to evaluate nonlinear effects [9,10]. Recently, modelers digm establishes a foundation for the extension of the traditional
have even managed to perform down-scaling from meso-scale processes to include verification and scientific software develop-
synoptic programs and up-scaling from surface layer CWE calcu- ment that results in the notional framework known as Sargent's
lations [11,12]. It is this tremendous potential that has led wind Framework. This framework elucidates the relationships between
engineering practitioners to more frequently present results of the processes of scientific model development, computational
such numerical studies in professional and trade journals and model verification, and simulation validation (Fig. 2). “The outer
promotional materials. circle together with data validity are the technical processes that
must be addressed to show that a model is credible … .Assessment
1.3. The CFD/EFD dilemma activities are spawned from each of these technical processes.” (See
Knepell and Arangno, Chapter 2 [35]) For example, fluid modeling
Initially the relationship between proponents of CFD and EFD for can initially provide data from which CFD turbulence models are
wind engineering applications, however, was very uneasy. CFD created, CFD calculations can use such turbulence models to quickly
proponents were enthusiastic about the future of their craft, and survey alternate solution strategies using simplified domain sce-
frequently endorsed numerical methods as the wave of the future, narios, then physical modeling can examine in greater depth design
open to immediate use on almost any application, and predicted consequences, and finally, CFD can extend initial conclusions to a
the eventual demise of “old fashioned” and “surely expensive” and broader set of similar cases. Combining experiments with numer-
cumbersome fluid modeling facilities. Some predicted that CFD ical simulations also provides new educational opportunities for
would be dominant by 1985 [13]. Perhaps this is not surprising the next generation of engineers and scientists [36].
considering that computer costs were declining by factors of 10 Thus, both CFD and EFD are in many respects mature tools to
every 5 years. EFD specialists were not convinced. Peter Bradshaw apply to wind engineering. Recently, there has been a courtship
(1975) [14] pointed out that whereas computational number between the two methodologies as one identifies how they can
crunching capacity had increased exponentially, our fundamental complement one another. Indeed, today there are opportunities for
understanding of turbulence had only grown slowly; hence a “fact a marriage of experimental and numerical technologies. This new
gap” which made proposed turbulent models approximate and discipline has been called “hybrid” wind engineering.
most numerical results questionable unless carefully validated for
the given application. The tendency for many CFD users to believe
2. Ten questions (and answers) concerning hybrid simulation
implicitly in the realism of the beautiful graphical displays that
of the wind environment
their software produces is implied when one says that CFD is really
an acronym for “Colorful Fluid Dynamics.” Harsher critics say that
Many readers may still feel uncertain as to exactly what this new
“Cheats, Frauds and Deceivers” would be more appropriate [10].
“hybrid” design dogma entails; hence in the next sections we will
Since 1960 a number of researchers have reviewed CFD and
consider ten questions about “hybrid” wind engineering research.
EFD methods to judge how well they represent the “full scale” or
original wind phenomena. Several committees have been formed
to systematically examine CFD and EFD reliability and propose 2.1. What is hybrid modelling for wind engineering?
minimum domain, boundary conditions, initial conditions, and
appropriate turbulence models [15,16]. Other researchers have Answer: It is the complementary use of CFD and EFD modeling
compiled experimental data sets suitable for validation of CFD to simulate wind engineering (WE) problems.
models (NPARC Alliance, 2005; ERCOFTAC, 2015; CEDVAL; CSU/
TTU Cooperative Program in Wind Engineering, 1987e2002;
Architectural Institute of Japan Working Group, 1992e1994)
[17e22]. Periodically papers appear that critique CFD methods,
consider software verification, and propose validation scenarios
[10,21,23e33]. Recently Kraft (2010) [34] even asked the ques-
tion “After 40 years why hasn't the computer replaced the wind
tunnel?”
Nonetheless, over the last fifty years the application of CFD to
wind engineering (or Computational Wind Engineering, CWE) has
matured substantially. The increase in CWE skill parallels the
similar growth in fluid-physical or EFD modeling. Cochran and
Derickson [33] summarized the sixty year struggle for physical
modeling to develop facilities, instrumentation, and model re-
quirements. Today there is general consensus about what one can
do with EFD with confidence, what remains uncertain, and what
should not be attempted at all. Indeed, EFD for wind engineering
went through decades of validation, and so it is not surprising that
CWE has followed a similar path. Blocken [3] and Meroney and
Derickson [5] have recently reviewed the birth, growing pains,
teenage status, and recent maturity of CWE.
A reliable new hybrid methodology has arisen that combines the Fig. 1. The “old” scientific method [82].
14 R.N. Meroney / Building and Environment 96 (2016) 12e21
Fig. 2. The “new” scientific method [82,83].
There is a renewed interest in modelling structure and envi-
ronment interactions to plan for consequences of wind related
hazards, atmospheric transport and dispersion, pedestrian comfort,
weather related transportation effects, and wind effects on alter-
native energy solar or wind energy projects. The interaction of
these needs are included in the taxonomy of the discipline of wind
engineering by Derickson and Cochran [37]. Such configurations
are often too complicated to predict reliably by the host of analytic
or code models developed for stationary effects in limited or
idealized situations. A new hybrid methodology has arisen that
combines the advantages of an old tool, fluid (or physical or scale)
EFD modeling, with the speed and convenience of a new technol-
ogy, computational fluid dynamics, CFD [4,38].
2.2. Why perform joint hybrid modeling versus CFD/Physical
modelling independently?
Answer: Each method can provide insight, accuracy levels and
reliability that are jointly synergistic.
The very use of the word “hybrid” implies a combination of
initially unrelated concepts of heterogeneous origin that results in
an offspring that has hopefully the best qualities of both parents. In-
deed, such crossbreeding often results in ‘hybrid vigor’ that has
even better and stronger characteristics than either parent. For
example.

Both methods permit the individual control of many variables,
which allows the researcher to isolate driving or dominant
transport mechanisms,

Fluid modeling can provide data in situations where the actual
physical mixing mechanisms are vague, confused or obscured by
other phenomena,

Computational modeling can provide data at greater spatial
resolution than is practicable during fluid modeling due to
instrumentation, time or cost limitations,

Flow in fluid modeling situations incorporate mixing at all time-
scales, whereas time dependent solutions in CFD often require
the dedication of massive computational resources, and

Computational modeling can be employed to explore further
the parametric space of a problem while keeping specific pa-
rameters constant.
2.3. What problems are particularly suitable for hybrid modeling?
Answer: Wind-engineering situations where a large number of
alternatives need to be considered, but validation data are limited;
situations where several physical time or space scales are involved
simultaneously or sequentially; and situations where critical sub-
modules or mechanisms need to be identified, defined, or mathe-
matically modeled before either CFD or physical modeling can
proceed.
Successful hybrid modeling usually involves maximum “cross-
talk” between the numerical and fluid modelers. Each needs to
know the other discipline to the extent that intelligent questions
are asked before and during modeling. The process can be
considered as one where there is frequent leapfrogging in questions
and answers, and a “bootstrap” process can occur where EFD sug-
gests boundary conditions, and subsequently CFD suggests model
modifications, which are validated by further EFD.
2.4. What type of problems may be unsuitable for hybrid modeling?
Answer: Problems which depend primarily on dimensionless
scales where the phenomena are not correctly reproduced during
physical simulation, e.g. surface heat transfer, combustion, radia-
tion, and problems during which important phenomena are
disconnected due to a mismatch in time or space scales.
Although some flow characteristics are known to be scale in-
dependent, there are others which are not Reynolds or Peclet
number independent. In particular flow phenomenon which are
limited by microscopic fluid properties such as heat or mass
transfer in the laminar sublayer, chemical reaction rates which have
different time scales than the surrounding flow motions, stratified
flows that scale as Froude number (L1/2) while mean or mixing
flows scale by Reynolds number (L), or problems in which con-
duction, convection and radiation produce similar order driving
forces, yet they occur over different time scales.
2.5. Will performing both CFD and physical simulations during
hybrid modeling be more expensive?
Answer: Often simplified or approximate CFD modeling will
permit the elimination of many expensive physical modeling sim-
ulations; thus, permitting a focus on critical options, or alterna-
tively, early physical modeling validation may identify what
linearization or simplifying assumptions can be incorporated into a
CFD model.
Modeling costs depend on facility operational costs and labor
costs. Larger physical modeling facilities can cost millions of dollars
to construct and operate, similarly large scale computing requires
huge investments for maintenance, cooling, and software support.
But although significant reductions have been made in the unit cost
of computer hardware and software, physical laboratory facilities
costs have remained constant or even risen. Thus, any project
management scheme that can leverage on the high cost physical
model research to reduce or extend the value of their measure-
ments is a savings in time and money. Without EFD verification, the
CFD results often remain questionable, but, on the other hand, once
verified CFD systems can also quickly and more cheaply extend
basic EFD measurements (e.g. CFD and EFD teamwork used during
solar panel analysis, Meroney and Neff [39] as described in Section
2.9).
2.6. What size CFD/Physical modeling facilities are required to
perform realistic simulations?
Answer: CFD calculation capacity in terms of grid size and speed
 R.N. Meroney / Building and Environment 96 (2016) 12e21
must be large enough to capture relevant flow scales over times
sufficient to reach flow equilibrium, and physical modeling facil-
ities must be large enough to reproduce relevant scaled turbulence
length and time scales of both the atmospheric boundary layer and
those generated by building models.
Different wind engineering phenomenon tend to be dominated
by specific turbulence scales. For example although a dispersing
plume is influenced by all scales of atmospheric motion from the
micro scales to the macro scales, most of the mixing of a plume (as
opposed to its displacement) is caused by eddies close to the same
size as the local plume. Thus, initially at a plume source large eddies
displace the entire plume back and forth, but only the smaller size
eddies entrain nearby air; however, at larger sizes the smallest
eddies merely “vibrate” the edges of the plume, the largest eddies
still cause plume “meander:, and the eddies of the same order of
magnitude as the plume entrain nearby fluid. Similarly, phenom-
enon such as cladding loading, snow drifting, pedestrian comfort,
etc. require consideration of the scales of motion that dominate
their behavior. Cermak et al. [2] and Cermak [1,40] proposed facility
limitations associated with atmospheric boundary layer simulation,
structural loading, canopy flows, and atmospheric transport,
Snyder [6] identified significant facility requirements for dispersion
and transport from stack plumes, Meroney [7,41] extended the
dispersion discussion to include transport and mixing of dense gas
clouds, and Meroney and Melbourne [8] identified operating ranges
to simulate the convective boundary layer in different size wind-
tunnel facilities.
2.7. Will it be necessary for a researcher to be an authority in both
CFD and physical modeling to achieve reliable results?
Answer: Researchers should share a foundation in meteoro-
logical, fluid mechanics, and structural mechanic fundamentals, but
the details of metrology and numerical simulation need not be
duplicated in every member of a research team.
One of the fascinating features of wind engineering research has
been the successful participation of researchers with many back-
grounds. At Colorado State University (CSU) from 1950 through
2010 the Fluid Mechanics and Wind Engineering Program involved
staff trained in Civil Engineering, Mechanical Engineering, Chemi-
cal Engineering, Atmospheric Science, Engineering Mechanics,
Economics, Physics, Computer Science, and Electrical Engineering.
The synergism of this wide variety of education and training
resulted in over 50 years of productive research and engineering
design. It is however very important that there is some cross-
disciplinary training, especially with respect to a base under-
standing of micro-meteorology and atmospheric surface layer
phenomena. Both clients and engineers need to talk the same
language and have the same expectations about the contributions
of wind engineering [42].
2.8. What role does full field scale research play during hybrid
modelling?
Answer: Both CFD and physical simulation need “truth” testing
against full size environments if possible; however, it is sometimes
impossible to perform reliable or repeatable full scale tests due to
inherent meteorological variabilities, safety, and cost constraints.
Engineering decisions frequently entail risk; hence, there will
always be some doubt and reluctance to make design decisions
based completely on “models”, “simulations” or “virtual” reality,
especially as the consequences of a wrong decision may entail
extreme cost or lives. Thus, as we extend our predictions into
formerly unexplored areas, it will always be wise to take the time to
confirm our estimates when possible by comparison with the actual
15
full-scale flow situations. When the situations are very complex,
enormously large, or involve extended times, then engineers will
attempt to examine all the possible submodules of their design in
the place of testing everything simultaneously.
2.9. What are some examples where hybrid wind-engineering has
achieved credibility?
Answers:

Building structures: bridge and tower vibrations; exterior
cladding loading; pedestrian comfort: street canyon flows;
vegetation effects [32,39,42e45,39,46,47].

Solar panel loads: isolated performance; installation on and
near structures [39,48,49].

Building ventilation: infiltration; window/door movement;
combined internal and external flows; transport in stairwells or
between floor [32,50,51].

Dispersion and diffusion of stack gases: modification and
entrainment of stack plumes by buildings; cooling tower drift;
dense gas dispersion [4,7,32,41,52e55].

Fire and smoke transport: cases involving combined heat
conduction, radiation, convection and even chemical reaction;
evacuation of personnel; control of sprinklers or automated
vents [5,32,56e60].

Snow, sand, dust or rain drifting/loading: complex interaction
of particle propagation and interaction with structures; thermal
environment variations; rain penetration or impact on struc-
tural cladding [5,32,61,62].

Wind energy production and siting: turbine performance;
wake and terrain interaction; built environment [63e65].

Flow over complex terrain: terrain enhanced flows; landesea
interfaces [63,64,66e68].

Vehicle aerodynamics: exhaust behavior, overturning mo-
ments, loads and stability of rail and road vehicles; rocket
launch site characteristics [69e72].

Meteorological phenomena: flow physics of tornadoes;
downbursts; hurricanes; interaction with built environment;
etc. [8,11,12,73e76].

Facility design: wind tunnel specifications [8,77].
There are so many application examples it is difficult to list them
all, so here are a couple cases to provide insight into the advantages
of hybrid analysis.
Solar Panels: Computational fluid dynamics (CFD) can provide a
flexible and cost-effective tool to select promising configurations
for solar panels from alternative design strategies, which can be
subsequently tested in detail either in the laboratory or the field.
Two-dimensional CFD models were performed for single, double,
and multiple collector arrangements before more time-consuming
and detailed three-dimensional calculations were performed for
selected cases. EFD studies were performed first at near full scale to
provide a validation base for CFD trials (See Figs. 3 and 4). Only after
a protocol was found that reproduced the initial EFD measure-
ments, were other alternatives considered. Then based on these
results the “best” solar arrangements were studied again by EFD
[39,49,50].
Cladding and Pedestrian Effects: A wind-sick building at the
Denver Federal Center, Colorado, USA was examined by the CSU
Wind Doctors using MRI (Measurement Research Investigation)
and CATscan (CFD Aerodynamic Testing) diagnostic techniques to
determine a proper course of treatment to alleviate severe skin and
breathing problems. The evaluation determined that the building's
sickness was due to severe extreme wind exposure (See Figs. 5 and
6). A prescription was written to provide protection through
16 R.N. Meroney / Building and Environment 96 (2016) 12e21

Fig. 3. Static pressure contours for CFD 3-d model (Left e 0 and right 180 wind orientations) from Meroney and Neff [39].

Fig. 4. Overall load comparisons between CFD 3-d and wind-tunnel models (Left for 0 and right for 180 wind orientations) from Meroney and Neff [39].

Fig. 5. CFD computed average pressure coefficients for NNW winds, N face in stag-
nation, from Meroney et al. [44].
Fig. 6. MRI and CATscan pressure coefficient comparison, West Face, NNW winds, from
Meroney et al. [44].

constructive plastic surgery on the face of the building and the

 R.N. Meroney / Building and Environment 96 (2016) 12e21
insertion of supportive landscaping orthotics around the foot of the
structure. The prognosis is for significantly improved quality of life,
but problems will persist even if various landscaping improve-
ments mature [45].
a. By looking at both CFD and EFD results one could determine
how building façade (mullions) affected variations in pressure.
b. It was also possible to determine if the mullions present during
the EFD tests modified the general behavior of pressure distri-
butions by looking at CFD results. The CFD also filled in infor-
mation in regions that were not instrumented.
c. Both methods produced results similar enough to make equiv-
alent design decisions.
Cooling Tower Drift: EFD was used to determine general flow
patterns in an urban environment with cooling towers, but simu-
lating drift as small scales was not considered feasible. Hence, the
EFD measurements were used to confirm CFD simulations, and
then Lagrangian particle predictions were produced in the resultant
wind fields using CFD (Figs. 7 and 8). Tests were performed with
and without structures present to determine multiplicative factors
to correct seasonal or annual predictions using approved codes like
ISCST3 and SACTI [55].
Fire Propagation: Fire can propagate through vegetative can-
opies, around buildings and within buildings. Flow and turbulence
measurements within model vegetative canopies have been
extensively studied by EFD [78] but since fires involve scales
ranging from the micro to the macro scales, it is not feasible to
realistically simulate them in the laboratory. Thus, EFD has been
used to validate CFD models absent fires; then, fire and smoke as-
pects are added into the CFD simulations to complete the evalua-
tion (Fig. 9) [57e59].
Facility Evaluation: Given that different wind or water test fa-
cilities can have different sizes and configurations (open vs closed,
various combinations of flow straighteners and screens) it is not
surprising that they may have different wind, turbulence, and
secondary flow fields. Thus, even when EFD tests are repeated by
different research organizations in what was assumed to be
equivalent initial and boundary conditions, significant variations in
results can occur. It has been difficult to identify how often and how
large such deviations affect measurements. Recently Moonen et al.
[77] have shown by using CFD that numerical methods can repro-
duce these facility idiosyncrasies by including all facility turns,
contractions, screens, flow straighteners and boundaries (See
Figs. 10 and 11). This provides an important means to critique EFD
Fig. 7. Discrete Particle Model predicted particle tracks (scale ¼ sec traveled) from all
Cooling Tower units for wind approach angle 240 and Rosin-Rammler source distri-
bution (dmean ¼ 0.1 mm, n ¼ 1) from Meroney [55].
17
Fig. 8. Discrete Particle Model predicted nodal contours of deposition from all Cooling
Tower units (kg/m2 s) for wind approach angle 2400 from Meroney [55].
results and to design more configuration independent EFD
facilities.
2.10. What does the future hold for hybrid modeling in wind
engineering?
Answer: The maturity of CFD and improved computers suggests
that IT will eventually permeate EFD metrology and analysis, and in
some situations CFD design techniques will dominate with EFD
being used to validate solutions or extend numerical submodules.
One can even imagine future situations where EFD measure-
ments may drive a CFD interpretation which controls subsequent
EFD measurements in a feedback manner; thus improving
predictions.
3. Summary and conclusions
Good mental health in a fluid or numerical modeler is always
indicated by the presence of a suspicious nature, cynicism and a
“show me” 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 process is to be maintained.
3.1. Limitations of similitude
It is important to remember that models whether CFD or EFD
based are “virtual” reality and only as accurate or realistic as our
own imagination. When we insist on modeling at full or reduced
length scale ratios then simulation criteria often require metrology
decisions that may enhance one flow characteristic while degrad-
ing another. We will never fully be able to answer the question “Just
how reliable are the results?” Simulation must be limited by un-
certainties in our understanding of the physical phenomena, un-
certainties about the initial or boundary conditions, uncertainties
about our measuring or calculation equipment, uncertainties about
our prototype observations, and uncertainty about what we really
want to know.
We must also take care that our search for agreement and cor-
relation does not itself lead to “spurious” errors and self-deception
[79]. Data presentation suggested by scaling variables and simu-
lation criteria might itself misrepresent the results. Flawed analysis
either intentional or through misunderstanding of commonly
accepted data analysis methods can lead to erroneous results and
presumption of correlation of cause and effect, when, in fact, there
is little or none. Classical dimensional analysis combined with
statistical regression of such scaled data may produce apparent
18 R.N. Meroney / Building and Environment 96 (2016) 12e21


Fig. 9. Large eddy calculations of 3-d thermal plume behavior at 160 s after ignition, Centerline section, T ( K) for flow over clear ground (left) and flow within a porous canopy
(right) as calculated by Meroney [59].

Fig. 10. The Jules Verne thermal wind tunnel a) Plan view, b) and c) are longitudinal and lateral cross-sectional views, and d) is a photograph of the nozzle, screen and guide vanes
as noted in Moonen et al. [77].

correlation of information resulting in “virtual” or “spurious” cor-
relation. Such inadvertent correlation errors can result in inap-
propriate conclusions and self-deception concerning the actual
relationships between scaling variables and the associated reduc-
tion in variance found in tables and graphs. A number of dimen-
sionless expressions used in meteorology and wind engineering
induce large magnitudes of spurious correlation when plotted
against other commonly accepted parameters. For example, when
drag coefficients, CD, pressure coefficients, Cp, dimensionless shear,
S*, or dimensionless concentrations, K, are regressed against Rey-
nolds numbers, Re, dimensionless height, z/LMO, or stratified Jensen
numbers, zo/LMO, inherent virtual correlations can exist with values
from 50 to 95% even when random numbers are used to generate
the component parts of the dimensionless groups!
Despite the limitations noted above, careful CFD/EFD modeling
are often still the best and only reliable predictive tool available!
3.2. Specific modeling limitations yet unresolved
Sometimes CFD, EFD, and hybrid model capacity are constrained
by our own limited understanding of the fluid phenomena. For
example, better simulations await the ability to provide:

Consistent inlet flow during CFD simulation of scales and char-
acteristics of the atmospheric boundary layer (reproduction
simultaneously of velocity, turbulence and integral length scale
profiles which are in equilibrium); (e.g. adjustment of numerical
wall functions [80].

Simultaneous resolution of small and large scales (e.g. fire in-
volves scales from microns to km);

Simultaneous similitude when conflicting similarity re-
quirements exist (e.g. Fr and Re number, one requires U pro-
portional to L and the other L1/2);

Turbulence models that correctly model the balance of turbu-
lent production and dissipation (e.g. standard k-epsilon model
presumes equilibrium in production and destruction of turbu-
lence energy in situations where significant convection is
actually present, but large eddy simulation models are known to
incorrectly model jet growth rates if incorrect constants are
chosen) [32]; and

Chemical reaction models that do not impose reaction time
scales which are incompatible with other fluid modeling
similitude requirements.

Studies which define model sensitivities [e.g. Beychok (2005)
[81] notes that uncertainties in input information can propagate
 R.N. Meroney / Building and Environment 96 (2016) 12e21
Fig. 11. Contours of dimensionless velocity magnitude a 1.80 m height (left) and in the vertical center plane (right) for different calculated situations a) no vanes, b) only horizontal
vanes, c) only vertical vanes, and d) both. Wind speeds higher than 2.1 are colored white as noted in Moonen et al. [77].
through models for plume rise, dispersion coefficients and
concentration averaging to produce potential errors by factors of
six to eighty].
3.3. Hybrid modeling in the 21st and 22nd Centuries
The primary role of EFD modeling of building aerodynamics
during the next century will not always be the direct measurement
of data to be used during engineering design of specific facilities.
Physical modeling is often not fast or flexible enough to perform the
sensitivity studies commonly required making engineering de-
cisions about very complex systems. Instead EFD and CFD may be
combined:

To explore atmospheric interactions not yet fully understood,

To tune and justify turbulence models incorporated into CFD
models,

To complement (or replace) numerical measurements when the
veracity of CFD modeling is constrained by computational ca-
pacity, understanding, or economics,

To devise new analytic models suitable for inclusion in larger
numerical systems, and

To validate computational modules as they are incorporated into
computer design codes.

To assist in the “education” of a new generation in fluid dy-
namics and wind engineering.
The proliferation of conference titles and sessions focusing on
EFD/CFD and wind engineering suggests that this refocus is already
underway.
19
Acknowledgments
This paper could not have been written without the contribu-
tions of the cohort of international engineers and scientists who
have contributed to the growth of wind engineering using physical
and numerical modeling. The author acknowledges their creativity,
genius and hard work. The author would also admit that not all
relevant references could be included in this review. Papers chosen
were based on relevance and familiarity to the author, and it is
highly likely additional and perhaps better examples are available.
References2
[1] J.E. Cermak, Applications of Fluid Mechanics to Wind EngineeringeA
Freeman Scholar Lecture, J. Fluids Eng. 97 (1) (1975) 9e38.
[2] J.E. Cermak, V.A. Sandborn, E.J. Plate, G.H. Binder, H. Chuang, R.N. Meroney,
S. Ito, Simulation of Atmospheric Motion by Wind-tunnel Flows, Civil Engi-
neering Report CER66JEC-VAS-EJP-GJB-HC-RNM-SI17, Colorado State Uni-
versity Fort Collins, CO, 1966, p. 111, http://dx.doi.org/10.13140/
RG.2.1.2298.5687, http://hdl.handle.net/10217/169834 (Valid 10/2015).
[3] B. Blocken, 50 Years of Computational Wind Engineering: Past, Present and
future, J.Wind Eng. Ind. Aerodyn. 129 (June 2014) 69e102.
[4]* R.N. Meroney, Wind Tunnel and Numerical Simulation of Pollution Disper-
sion: a Hybrid Approach, Proceedings of Croucher Advanced Study Institute,
Hong Kong University of Science and Technology, 2004, p. 66, 6e10
December 2004, and again 7e8 December 2005.
[5] R.N. Meroney, R.E. Derickson, Virtual reality in wind engineering: The windy
world within the computer, Indian J. Wind Eng. Indian Soc. Wind Eng. 11 (2)
(July 2014) 11e26.
[6] W.H. Snyder, Guideline for Fluid Modeling of Atmospheric Diffusion, US
Environmental Protection Agency, 1981, p. 199. Report EPA-600/8-81-009,
http://nepis.epa.gov/Adobe/PDF/2000BDW8.PDF (Valid 10/2015).
[7] R.N. Meroney, Guideline for Fluid Modeling of Liquefied Natural Gas Cloud
2
*- These references are available to download at http://www.engr.colostate.edu/
~meroney/ under the subject category html link contributions of Robert N. Mer-
oney. (Valid 10/2015).
20 R.N. Meroney / Building and Environment 96 (2016) 12e21
Dispersion, Report No. GRI-86/0102.1, Chicago, Il, in: Instruction Guide, Vol. 1,
Gas Research Institute, 1986, p. 62. Also available as US NTIS PB87-120051.
[8] R.N. Meroney, W.H. Melbourne, Operating ranges of meteorological wind
tunnels for the simulation of convective boundary layer phenomena,
J. Bound. Layer Meteorol. 61 (1992) 145e174.
[9] S. Murakami, Past, present and future of CWE: the view from 1999, in:
A. Larson, G.L. Larose, F.M. Livesy (Eds.), Wind Engineering into the 21st
Century, Balkema Press, 1999, pp. 91e104.
[10] T. Stathopoulos, Computational wind engineering: Is it mature for civil en-
gineering applications? (Editorial), J. Aerosp. Eng. 12 (1999) 111e112.
[11] T. Yamada, K. Koike, Downscaling mesoscale meteorological models for
computational wind engineering applications, J. Wind Eng. Ind. Aerodyn. 99
(2011) 199e216.
[12] A. Mochida, S. Iizuka, Y. Tominaga, I.Y. Lun, Up-scaling CWE models to include
mesoscale meteorological influences, J. Wind Eng. Ind. Aerodyn. 99 (4) (2011)
187e198.
[13] D.R. Chapman, H. Mark, M.W. Pirtle, Computers Vs. Wind Tunnels for Aero-
dynamic Flow Simulations, Aeronaut. Astronaut. J. (April 1975) 22e30, 35.
[14] P. Bradshaw, Letter to Editor on “Computers vs Wind Tunnels”, Aeronaut.
Astronaut. J. (September 1975) 6.
[15] AIAA, Guide for the Verification and Validation of Computational Fluid Dy-
namic Simulation, American Institute of Aeronautics and Astronautics,
Reston, VA, 1998, p. 19. AIAA G-077-1998.
[16] QNET-CFD, Repository of Structured Knowledge and Advice for Industrial and
Academic Application of CFD, 2004. http://www.ercoftac.org/products_and_
services/wiki/. http://uriah.dedi.melbourne.co.uk/w/index.php/Main_Page
(Valid 9/1/15).
[17] NPARC Alliance, CFD and Validation, NPARC Alliance CFD Verification and
Validation Web Site, 2005. http://www.grc.nasa.gov/WWW/wind/valid/
homepage.html. http://www.grc.nasa.gov/WWW/wind/valid/archive.html
(Valid 9/1/15).
[18] ERCOFTAC, European Research Community on Flow, Turbulence and
Combustion-Best Practice Guidelines-Publications, 2015. http://www.
ercoftac.org/publications/. http://www.ercoftac.org/publications/ercoftac_
best_practice_guidelines/ (Valid 9/1/15).
[19] B. Leitl, Compilation of Experimental Data for Validation of Microscale
Dispersion Models, CEDVAL, Website maintained by Environmental Wind
Tunnel Laboratory (EWTL), Meteorological Institute, University of Hamburg,
Germany, 2006 (Valid 9/1/15), http://www.mi.zmaw.de/index.php?id¼433.
[20] B. Bienkiewicz, Animation and Data from CSU/TTU Cooperative Program in
Wind Engineering (1987-2000), 2006 (Valid 9/1/15), http://www.windlab.
colostate.edu/research_and_service.htm.
[21] A.H.P. Smith, Data from CSU/TTU Cooperative Program in Wind Engineering
(1987-2000), Wind Science and Engineering Research Center, Texas Tech
University, Lubbock, TX, 2006. http://www.wind.ttu.edu/index.php (Valid 9/
1/15).
[22] T. Tamura, Numerical Prediction of Wind Loading on Buildings and
Structures-Activities of AIJ Cooperative Project on CFD by Working Group of
Architectural Institute of Japan, in: IWEF Workshop on CWE/CFD’96, Fort
Collins, CO, 9 August 1996, p. 72. See also (1997) J. Wind Eng. Ind. Aerodyn.,
Vol. 67-68, pp. 671e685.
[23] P. Bearman, The Tortoise and the Hare: the Physical Experiment versus the
Computer Model Has Been a Scientific Tortoise and Hare Race over the Last
30 Years, BMT Focus, BMT (British Maritime Technology) Group Ltd, Ted-
dington, U.K., 1996, p. 12. Also see link : http://www.zoominfo.com/
CachedPage/?archive_id¼0&page_id¼2069416270&page_url¼//www.
bmtfm.com/?/377/694/487&;page_last_updated¼2012-10-11T11:57:
39&firstName¼Peter&lastName¼Bearman (Valid 10/2015).
[24] A.J. Baker, R.M. Kelso, E.B. Gordon, S. Ray, E.G. Schaub, Computational Fluid
dynamics: A Two-Edged Sword, ASHRAE J. 29 (8) (August, 1997) 51e58.
[25] T. Stathopoulos, Computational wind engineering: Past achievements and
future challenges, J. Wind Eng. Ind. Aerodyn. 67&68 (1997) 509e532.
[26] T. Stathopoulos, A State-of-the art Report on Computational Wind Engi-
neering, in: C.K. Choi, C.B. Yun, H.G. Kwak (Eds.), 7th Int. Conf. On Computing
in Civil and Building Engineering, 1997, pp. 2081e2086. Seoul, Korea.
[27] T. Stathopoulos, The numerical wind tunnel for industrial aerodynamics: Real
or virtual in the new millennium? Wind Struct. 5 (2e4) (2002) 193e208.
[28] Y. Li, T. Stathopoulos, Computational evaluation of pollutant dispersion
around buildings: Estimation of numerical errors, J. Wind Eng. Ind. Aerodyn.
77&78 (1998) 619e630.
[29] I.P. Castro, J.M.R. Graham, Numerical wind engineering: the way ahead?,
Proc. Institute of Civil Engineers, Struct. Build. 134 (1999) 275e277.
[30]* D. Banks, R.N. Meroney, R.L. Petersen, J.J. Carter, Evaluation of Fluent for
Predicting Concentrations on Buildings, Paper No. 70223, in: Proceedings of
the Annual Meeting of the AWMA (Air and Waste Management Association),
San Diego, CA, June 2003, p. 11.
[31] T. Scott, D. Banks, Virtual Versus Physical: Examining the Capabilities of CWE/
CFD Simulations through Comparisons to Wind Tunnel Observations, in: 4th
Int. Proceedings of Symposium on CWE (CWE2006), Yokohama, Japan, 2006,
pp. 581e584, in: http://www.iawe.org/Proceedings/CWE2006/TB2-06.pdf
(Valid 10/15).
[32] B. Blocken, T. Stathopoulos, J. Carmeliet, J. Hensen, Application of CFD in
building performance simulation for the outdoor environment: an overview,
J. Build. Perf. Simul. 4 (2) (2011) 157e184.
[33] L. Cochran, R. Derickson, A physical modeler's view of Computational Wind
Engineering, J. Wind Eng. Ind. Aerodyn. 99 (2011) 139e153.
[34] E.M. Kraft, After 40 Years Why Hasn't the Computer Replaced the Wind
Tunnel? ITEA J. 31 (2010) 329e346.
[35] P.L. Knepell, D.C. Arangno, in: Simulation Validation: a Confidence Assess-
ment Methodology, Wiley-IEEE Computer Society Press, Los Alamitos, CA,
1993, p. 168.
[36] L. Guessous, R. Bozinoski, R. Kouba, D. Woodward, Combining Experiments
with Numerical Simulations in the Teaching of Computational Fluid Dy-
namics, Comput. Educ. J. ASEE XIV (3) (2004) 55e60.
[37] R.G. Derickson, L. Cochran, Making the Wind Engineer's Role More Effective
in Urban Projects: A Proposed Three-Facet Framework for Enhanced Success,
in: 12th Americans Conference on Wind Engineering (12ACWE), Proceedings
of, Vol. 2, Seattle, Washington, USA, June 16-20, 2013, pp. 1237e1250.
[38] H.M. Hangan, Wind-driven rain studies. A C-FD-D approach, J. Wind Eng. Ind.
Aerodyn. 81 (1999) 323e331.
[39]* R.N. Meroney, D.E. Neff, Wind effects on roof-mounted solar photovoltaic
arrays: CFD and wind-tunnel evaluation, in: Proceedings of 5th Int. Sym-
posium on Computational Wind Engineering (CWE2010), 23e27 May, Chapel
Hill, NC, 2010, p. 8.
[40] J.E. Cermak, Wind-tunnel design for physical modeling of atmospheric
boundary layers, Special Volume on Advances in Fluid Mechanics, ASCE,
J. Eng. Mech. 107 (EM3) (1981) 623e642.
[41] R.N. Meroney, Wind-tunnel experiments on dense gas dispersion, Special
Issue, J. Hazard. Mater. 6 (1982) 85e106. also in book Dense Gas Dispersion,
R. E. Britter and R. F. Griffiths, eds. Chemical Engineering Monograph No. 16,
Elsevier Scientific Pub. Co., 1982.
[42]* L. Cochran, R. Derickson, R.N. Meroney, H. Sharp, On what new building
project managers need to know about wind engineering, in: 17th Austral-
asian Wind Engineering Society Workshop, Wellington, New Zealand, 11e13
February, 2015, p. 4.
[43] R.N. Meroney, C.W. Letchford, P.P. Sarkar, Comparison of numerical and wind-
tunnel simulation of wind loads on smooth and rough domes immersed in a
boundary layer, in: 3rd Int. Symposium on Computational Wind Engineering,
Univ. of Birmingham, UK, 4e7 September 2000, 2002, p. 12. Also Journal of
Wind and Structures (2002), Vol. 5, No. 2-4, 347-358.
[44]* R.N. Meroney, D.E. Neff, C.H. Chang, Diagnosis of a Sick Building by the Wind
Doctors, in: C.K. Choi, et al. (Eds.), Proc.. of the Second International Sym-
posium on Advances in Wind and Structures (AWAS'02), August 21e23,
2002, pp. 43e52. Busan, Korea.
[45]* R.N. Meroney, D.E. Neff, C.H. Chang, R. Predoto, Computational Fluid Dy-
namics and Physical Model Comparisons of Wind Loads and Pedestrian
Comfort Around a High Rise Building, Session on Computational Evaluation
of Wind Effects on Buildings, in: Proceedings of ASCE 2002 Structures
Congress, Denver, CO, April 4e6, 2002, p. 2.
[46] Y. Tominaga, A. Mochida, R. Yoshie, H. Kataoka, T. Nozu, M. Yoshikawa,
T. Shirasawa, AIJ guidelines for practical applications of CFD to pedestrian
wind environment around buildings, J. Wind Eng. Ind. Aerodyn. 96 (10e11)
(2008) 1749e1761.
[47] M.H. Salim, K.H. Schlunzen, D. Grawe, Including trees in the numerical
simulations of the wind flow in urban areas: Should we care? J. Wind Eng.
Ind. Aerodyn. 144 (September 2015) 84e95.
[48]* D.E. Neff, R.N. Meroney, Wind Performance of Photovoltaic Arrays, Final
Report, Wind Engineering and Fluids Laboratory, Colorado State University,
2003, p. 181, http://dx.doi.org/10.13140/RG.2.1.1798.8563.
[49] D. Banks, Panels in the Wind: How Wind Load Studies Will Impact the Solar
Industry, World Renewable Energy Forum, Denver, CO, 2012, p. 7.
[50] P. Karava, Airflow Prediction in Buildings for Natural Ventilation Design:
Wind-Tunnel Measurements and Simulation, Ph.D. Thesis, Dept. of Building,
Civil and Environmental Engineering, Concordia University, Montreal,
Quebec, 2008, p. 242.
[51]* R.N. Meroney, CFD Prediction of Airflow in Buildings for Natural Ventilation,
in: Proceedings 11th Americas Conference on Wind Engineering, San Juan,
Puerto Rico, June 22-26, 2009, p. 11.
[52] C.H. Chang, R.N. Meroney, Concentration and flow distributions in urban
street canyons: Wind tunnel and computational data, J. Wind Eng. Ind.
Aerodyn. 91 (2003) 1141e1154.
[53] B. Leitl, P. Klein, M. Rau, R.N. Meroney, Concentration and flow distributions
in the vicinity of U-shaped buildings: Wind-tunnel and computational data,
J. Wind Eng. Ind. Aerodyn. 67&68 (1997) 745e755.
[54] R.N. Meroney, B.M. Leitl, S. Rafailidis, M. Schatzmann, M. Pavegeau, Wind-
tunnel and numerical modeling of flow and dispersion about several building
shapers, in: Int. Workshop on ACFD for Wind Climate in Cities, Hayama,
Japan, August 24-26, 1998, p. 10. Special issue of J. of Wind Engineering and
Industrial Aerodynamics, Vol 81(1999), pp. 333e346.
[55] R.N. Meroney, CFD Prediction of cooling tower drift in an urban environment,
in: Proceedings of 4th International Symposium on Computational Wind
Engineering (CWE2006), 16-19 July 2006, Yokohama, Japan, 2006, p. 4. Also
Journal of Wind Engineering, Vol. 108, pp. 809e812.
[56] H.R. Baum, D.B. McGrattan, R.G. Rehm, Three dimensional simulations of fire
plume dynamics, in: Fie Safety Science Proceedings of the Fifth International
Symposium, International Association of Fire Science, March 3e7, 1997,
Melbourne, AU, 1997, pp. 511e522 (also J. Heat Transfer Society of Japan, Vol.
35, pp. 45e52.).
[57]* R.N. Meroney, Fire whirls, fire tornadoes, and fire storms: Physical and nu-
merical modeling, in: Proceedings of PHYSMOD'03: International Workshop
 R.N. Meroney / Building and Environment 96 (2016) 12e21
on Physical Modelling of Flow and Dispersion Phenomena, 3e5 September
2003, p. 9. Prato, Italy.
[58]* R.N. Meroney, Numerical prediction of fire propagation in idealized wildland
and urban canopies, in: Proceedings of 12th International Conference on
Wind Engineering (ICWE12), 1e6 July 2007, p. 8. Cairns, North Queensland,
Australia.
[59] R.N. Meroney, Fires in porous media: natural and urban canopies (Chapter 8),
in: Y.A. Gayev, J.C.R. Hunt (Eds.), Flow and Transport Processes with Complex
Obstructions, NATO Science Series, Springer, Berlin, 2007.
[60] R.N. Meroney, Protocol for CFD prediction of cooling tower drift in an urban
environment, in: 4th Int. Symp on Computational Wind Engineering
(CWE2006), 2008, p. 12. Final publication in J. Wind Engr. Ind. Aerod, Vol. 96,
(2008), pp. 1789e1804.
[61] B. Bang, A. Nielsen, P.A. Sundsbo, T. Wilk, Computer simulation of wind
speed, wind pressure and snow accumulation around buildings (SNOW-SIM),
Energy Build. 21 (1994) 235e243.
[62] B. Blocken, J. Carmeliet, Spatial and temporal distribution of driving rain on a
low-rise building, Wind Struct. 5 (5) (2002) 441e462.
[63] D.E. Neff, R.N. Meroney, (1997), Wind-tunnel modelling of hill and vegetation
influence on wind power availability, in: Proceedings of Wind Energy and
Landscape (WEL), Satellite International Workshop of the 2nd European &
African Conference on Wind Engineering (2 EACWE), Genova, Italy, June 26-
27, 1997, p. 10. Also published as reviewed paper in J. of Wind Engineering
and Industrial Aerodynamics, Vol. 74-76, 1998, pp. 335e343.
[64] R.G. Derickson, J.A. Peterka, Development of a powerful hybrid tool for
evaluating wind power in complex terrain: Atmospheric Numerical Models
and Wind Tunnels, in: 42nd AIAA Aerospace Sciences Meeting and Exhibit,
5e8 January 2004, Reno, Nevada, American Institute of Aeronautics and
Astronautics, 2004, p. 25, http://dx.doi.org/10.2514/6.2004-1005. Paper No.
AIAA-2004-1005.
[65] C.J. Desmond, S.J. Watson, S. Aubrun, S. Avila, P. Hancock, A. Sayer, A study on
the inclusion of forest canopy morphology data in numerical simulations for
the purpose of wind resource assessment, J. Wind Eng. Ind. Aerodyn. 126
(March 2014) 24e37.
[66]* R.G. Derickson, R.N. Meroney, A simplified physics airflow model for evalu-
ating wind power sites in complex terrain, in: Proceedings of Summer
Computer Simulation Conference, July 18-20, 1977, p. 14. Chicago, Illinois.
[67] R.N. Meroney, CFD Modeling of dense gas cloud dispersion over irregular
terrain, in: 13th Int. Conf. on Wind Engineering, July 10-15, 2011, Amsterdam,
The Netherlands, 2011, p. 8. Journal of Wind Engineering and Industrial
Aerodynamics (2012), Vol. 104-106, pp. 500e508.
[68]* R.N. Meroney, , Wind-tunnel Modelling of Hill and Vegetation Influence on
Wind Power Availability: Task 1: Literature Review, Prepared for Meteoro-
logical Services, U.S. Windpower, Inc, Livermore, CA, 1992, p. 123, http://
dx.doi.org/10.13140/RG.2.1.2560.7124. CSU Contract No. 298610, November
1992.
[69] B. Leitl, R.N. Meroney, Car exhaust dispersion in a street canyon e Numerical
critique of a wind tunnel experiment, J. Wind Eng. Ind. Aerodyn. 67e78
(1997) 293e304.
[70] M. Desai, S.A. Channiwala, H.J. Nagarsheth, Computational validation of
experimental aerodynamic predictions of a car, in: FMA’08: Proceedings of
6th IASME/WSEAS Int. Conf. On Fluid Mechanics and Aerodynamics, Rhodes,
Greece, August 20-22, 2008, p. 3. New Aspects of Fluid Mechanics and
Aerodynamics, Series WSEAS Mechanical Engineering, pp. 125e128.
[71] Y. Sakuma, M.P. Paidoussis, S.J. Price, Effect of boundary layer development
on the dynamics of trains and train-like articulated systems travelling in
confined fluid, J. Wind Eng. Ind. Aerodyn. 126 (March 2014) 38e47.
[72] B. Blocken, T. Defraeye, E. Koninckx, J. Carmeliet, P. Hespel, CFD simulations of
the aerodynamic drag of two drafting cyclists, Comput. Fluids 71 (2013)
435e445.
21
[73] R.N. Meroney, Wind-tunnel simulation of convective boundary layer phe-
nomena: Simulation Criteria and Operating Ranges of Laboratory Facilities,
in: E.J. Plate, E.E. Fedorovich, D.X. Viegas, J.C. Wyngaard (Eds.), (NATO
Advanced Study Institute on Buoyant Convection in Geophysical Flows, 17-27
March 1997, Pforzheim, Germany.) Buoyant Convection in Geophysical Flows,
Kluwer Press, August 1998, pp. 313e326.
[74] R. Avissar, R.N. Meroney, M.D. Moran, G. Wu, R.A. Pielke, Operating range of
numerical and physical models for the simulation of coastal marine flows,
Bound. Layer. Meteorol. 50 (1990) 227e275.
[75] A. Sengupta, F.L. Haan, P.P. Sarkar, V. Balaramudu, Transient loads on build-
ings in microburst and tornado winds, J. Wind Eng. Ind. Aerodyn. 96 (2008)
2173e2187.
[76] C.W. Letchford, M.T. Chay, Pressure distributions on a cube in a simulated
thunderstorm downburst. Part A: stationary downburst observations, Part B:
moving downburst observations, J. Wind Eng. Ind. Aerodyn. 90 (7) (2002),
711-732 and 733e753.
[77] P. Moonen, B. Blocken, J. Carmeliet, Indicators for the evaluation of wind
tunnel test section flow quality and application to a numerical closed-circuit
wind tunnel, J. Wind Eng. Industrial Aerodynamics 95 (2007) 1289e1314.
[78] R.N. Meroney, Characteristics of wind and turbulence in and above model
forests, J. Appl. Meteorology 4 (5) (October 1968) 780e788.
[79] R.N. Meroney, Spurious or Virtual Correlation Errors Commonly Encountered
in Reduction of Scientific Data, J. Wind Eng. Industrial Aerodyn. 77e78
(1998a) 543e553.
[80] B. Blocken, T. Stathopoulos, J. Carmeliet, CFD Simulation of the Atmospheric
Boundary Layerdwall Function Problems, Atmospheric Environment, 2000.
[81] M.R. Beychok, Fundamentals of Stack Gas Dispersion, fourth ed., 2005, p. 201
(Published by author), Newport Beach, CA.
[82] C.P. Rahaim, The AIAA Committee on Standard for CFD e Verification Project
Status and Plans, 3rd QNET-CFD Workshop Program, Prague, CZ, 29-30 May
2003, Slides 13 & 14 of 39 slides. Or see: Rahaim, C.P., Oberkampf, W., Cosner,
R.R. and Dominik, D.F. (2003), in: AIAA Committee on Standards for
Computational Fluid Dynamicsestatus and Plans, AIAA Aerospace Sciences
Meeting, Reno, NV, 2003, p. 21, http://dx.doi.org/10.2514/6.2003-844. Paper
No. 2003-0844, 6e9 January.
[83] R.G. Sargent, Proceedings of, in: B. Johansson, S. Jain, J. Montoya-Torres,
J. Hugan, E. Yucesan (Eds.), Verification and Validation of Simulation Models,
2010 Winter Simulation Conference, 5-8 December, 2010, pp. 166e183. Bal-
timore, MD.
Dr. Meroney has over 50 years of experience in both numerical methods and
metrology applied to wind-engineering problems. He is the author of more than 350
papers and reports. He is an Emeritus Professor of Wind Engineering in the Civil and
Environmental Engineering Department at Colorado State University (CSU) at which
he was Director of the Fluid Dynamics and Diffusion Laboratory (FDDL) and Leader
of the Fluid Mechanics Program for over 15 years. For over ten years he co-directed
the joint CSU/TTU Cooperative Program in Wind Engineering with Texas Tech Univer-
sity which studied extreme wind effects about low-rise buildings, diffusion, tornadoes,
and wind turbines. Between 1965 and 2000 he supervised $31 million or $1 million/
year in 31 years (discounting sabbatical years) in 2010 adjusted CPI dollars. Retired
in 2004 he has continued as a Senior Research Scientist/Scholar at CSU to the present.
Dr. Meroney also consults and works as an expert witness on fluid mechanics problems
in the areas of water and waste treatment, building ventilation, smoke and fires, forest
and urban dispersion. He has taught several workshops at national meetings on
Computational Wind Engineering (2006, 2007, and 2010). He regularly contributes pa-
pers at national and international wind engineering conferences on CFD applications
to wind flows about buildings, vegetation, solar systems, complex terrain, wind tur-
bines, and drifting snow.