MEAN FLOW AND TURBULENCE MEASUREMENTS
AROUND A 2-D ARRAY OF BUILDINGS IN A WIND TUNNEL
Michael J, Brown1
Los Alamos National Laboratory, Los Alamos, New Mexico
Robert E. Lawson, Jr.2
Atmospheric Sciences Modeling Division, ARL, NOAA, Research Triangle Park, North Carolina
David S, DeCroix
Los Alamos National Laboratory, Los Alamos, New Mexico
Rdbert L. Lee
Lawrence Livermore National Laboratory, Livermore, California
1. INTRODUCTION
In order to predict the dispersion of harmful
materials released in or near an urban environment, it is
important to first understand the complex flow patterns
which result from the interaction of the wind with
buildings and, more commonly, clusters of buildings.
Recent advances in the application of computational
fluid dynamics (CFD) models to such problems have
shown great promise, but there is a need for high-quality
data with which to evaluate CFD models. This study
was performed to fill that need for a limited range of
conditions.
High-resolution measurements of the three
components of the mean and turbulent velocity statistics
were obtained around a 2-d array of model buildings in
the USEPA meteorological wind tunnel. In this paper,
we briefly review prior field and laboratory experiments
on building flows, describe our experimental set-up and
measurement apparatus, present the flow measure-
ments, and discuss their significance in relation to
current understanding.
2. BACKGROUND
A large number of flow and tracer experiments have
been performed around single buildings (e.g., see
reviews by Meroney (1982), Hosker (1984), Peterka et
al. (1985) and more recent studies by Lee et al. (1991),
Snyder and Lawson (1994), Kastner-KIein et al. (1997),
and Cowan (1997)). There have been relatively fewer
measurement campaigns around groups of buildings.
For two buildings, much of the basic understanding of
the flow in the urban street canyon has been obtained
through tracer dispersion and smoke visualization
experiments. Field experiments by Johnson et al.
(1973), Dabberdt et al. (1973), DePaul and Sheih
(1985), Yamartino and Wiegand (1986), and
Kitabayashi (1992) all confirmed the presence of a large
vortex circulation within the urban canyon, although
there was some disagreement under what conditions
' Corresponding author address: Michael J. Brown, Los
Alamos National Laboratory, Group TSA-4, MS F604, Los
Alamos, NM 87545. E-mail: mbrown@lanl.gov.
2 On assignment to the National Exposure Research
Laboratory, U.S. Environmental Protection Agency.
the vortex would form and what the controlling factors
were for the vortex strength. These differences can
probably be attributed to differences in the building
configurations and meteorological conditions, and to
uncertainties in the measurements.
Even fewer field experiments have been performed
for obtaining flow fields around groups of buildings. A
very nice study by Depaul and Sheih (1986) measured
the vortex circulation using tracer balloons and rapid
sequence photography for a street canyon in Chicago.
A few wind measurements were also obtained with a
hot-wire anemometer. In a street canyon in Tsukuba,
Japan, Kitabayashi (1992) measured the horizontal
winds at 4 positions along each canyon wall at two
heights. Turbulence statisitics were also obtained at 2
sites near the wall using sonic anemometers. Qin and
Kot (1993) took horizontal wind speed and wind
direction measurements at several positions within
urban canyons at 3 sites in Guangzhou City, China. At
one site a U-V-W propellor anemometer was used in the
center of the street canyon. A quality, long-term dataset
was obtained by Rotach (1995) in Zurich that included
two vertical profiles in the street canyon and one profile
at rooftop. Eighteen months of data were analyzed from
six 3-axis sonic anemometers, 12 cup anemometers,
two wind vanes, and four temperature sensors. A
unique experiment was performed by Nakamura and
Oke (1988) in Kyoto where 63 temperature sensors
were used to measure the temperature field within a
street canyon. Only two sonic anemometers were used,
one at the canyon floor and one at rooftop. Data were
collected over two 2-day periods. Although the experi-
ments described above contain valuable information, it
is difficult to use this data for rigorous CFD model
validation due to sparsity of measurements and lack of
knowledge of the upstream boundary conditions.
More measurements can, in general, be obtained in
wind-tunnel experiments and the upstream boundary
conditions can be accurately defined. Flow measure-
ments in urban canyons have been performed by Britter
and Hunt (1979), Hussain and Lee (1980), and Lawson
and Ohba (1993), for example. In general, the nature of
the flow between two buildings of equal height is
determined by the ratio of the width between buildings
(W) to the building height (H) (Hussain and Lee, 1980).
There is also a weak dependence on the cross-sectional
-------
length of the buildings. As summarized by Oke (1987),
a single vortex develops between buildings for skimming
flow (W/H < 1), two counter-rotating vortices may
develop tor wake interference flow (W/H - 1,5), and for
isolated roughness flow (W/H > 3) the flow field looks
similar to the single building case,
Hosker (1987) reported that several studies have
shown that a helical vortex will form between two
buildings if the wind is within 60 degrees of
perpendicular to the building face, otherwise no vortex
forms. Several wind-tunnel studies (Wedding et al.,
1977; Hoydysh and Dabberdt, 1988; and Theurer et a!.,
1992) demonstrated that building height differences caW-j
significantly change the urban canyon flow field. In
addition, peaked roofs and non-rectilinear buildings can
alter urban canyon circulation (e.g., Rafailidis and
Shatzmann, 1995), Based on smoke visualization
studies, Meroney et al. (1996) found that rooftop
recirculation zones do not form on a series of buildings
of equal height, except for the one furthest upstream.
Wind-tunnei flow measurements for rnulti-building
arrays have been carried out by a few researchers. For
example, Roth and Ueda (1998) measured the
longitudinal and vertical components of the mean wind
and turbulent intensity for staggered and unstaggered
cubical arrays using a laser doppler anemometer.
Lateral traverses were made at 5 heights at one
downwind location within the cubical array and vertical
profiles were taken at two locations: directly behind a
cube and in the street channel. Similarly, Davidson et al.
(1996) obtained multiple vertical and lateral wind and
turbulent intensity profiles using a pulsed wire
anemometer for staggered and unstaggered cubical
arrays. Both Roth and Ueda (1998) and Davidson et al.
(1996) found significant wind speed reduction within the
obstacle arrays. However, in both cases, the
measurements were not sufficiently dense to fully
capture the canyon vortex and rooftop recirculations.
Components of the wind-tunnel experiments per-
formed by Raifailidis and Shatzmann (1995), Meroney
et al. (1996), and Theurer et al. (1992) and the small-
scale field experiments by McDonald et al. (1998) have
similar set-ups as described here, i.e., an array of wide
buildings. However, the first two experiments and the
last one focused primarily on concentration measure-
ments, whiie the third took mean wind and turbulence
Intensity measurements above the building tops only. It
is our intent to supplement these experiments with high
density mean wind and turbulent kinetic energy
measurements in and around a 2-d building array.
3. EXPERIMENTAL SET-UP
The experiments were carried out in the wind tunnel
of the U.S. Environmental Protection Agency's Fluid
Modeling Facility (Snyder, 1979). The wind-tunnel is of
the open-return type with a test section 3,7 m wide, 2.1
m high and 18.3 m long. Airspeed in the test section
can be varied from about 0.3 to 8 m/s. The ceiling of the
test section is adjustable in height to compensate for
blockage effects due to large models or to compensate
for the growth of a thick floor boundary layer by allowing
for a non-acceterating freestream flow. An automated
Figure 1. The two-dimensional building array mounted
in the wind tunnel.
instrument carriage system provides the capability for
positioning a probe anywhere in the test section,
acquiring data, then moving to the next measurement
location and repeating the process without intervention.
The general setting for the building models was
assumed to be an urban environment typica! of the
downtown areas of modern cities. Ideally, the building
height, shape and separation between the buildings
should all be varied over the full range of typical values;
however, the total number of combinations would
quickly become excessive. Based on flow-visualization
studies in which many of these parameters were varied,
we elected to restrict detailed quantitative measure-
ments to two building configurations, a two-dimensionaf
array as reported herein and a three-dimensional array
that will be reported later.
The two-dimensional building array examined in this
study consisted of rectangular blocks with equal height
and length (H = L = 150mm) and extending from wall-to-
wall in the spanwise direction (Fig. 1). The blocks were
spaced 1H apart in the alongwind direction. The
building models were immersed in a simulated neutral
atmospheric boundary layer which was created in the
wind tunnel using spires (Irwin, 1981) and floor rough-
ness elements. This combination produced a simulated
boundary layer with depth of 1,8m, a roughness length
of 1 mm, and a power law exponent of 0.16, The array
was located 10.9m from the leading edge of the spires
to allow sufficient upstream fetch for the boundary layer
to grow to equilibrium. While no specific scale ratio was
chosen, a representative value would be 250:1, hence
the building models would correspond to full-scale
buildings on the order of 30 to 40m in height.
Similarity criteria for modeling flow around a
building immersed in a neutral atmospheric boundary
layer in a wind tunnel require that the Rossby,
Reynolds, Peclet or Reynolds-Schmidt numbers, plus a
set of non-dimensional boundary conditions, be
matched in both model and prototype. Referring to
Snyder (1981), the Rossby number can be neglected
when modeling prototype flows with a length scale less
than about 5 km. Also, provided the model Reynolds
number is sufficiently large, it is not necessary to match
the Reynolds number, Peclet number or Reynolds-
Schmidt product between model and prototype. The
reference velocity in this study (3m/s at z=H) was
-------
£
downwind distance, mm
3.0 m/s
Figure 2, Wind vector and turbulent kinetic energy fields measured along centerline around the 2-d building array in
the USEPA meteorological wind tunnel. The first three of seven buildings are shown here.
chosen such that the building Reynolds number was
greater than that regarded as the critical value for
Reynolds number independence,
A hot-wire anemometer with an X-array sensor
was used to measure the mean velocity and turbulence
intensity profiles of the approach flow in the absence of
any buildings. The hot-wire anemometer is useful when
the turbulence intensities are relatively low (e.g. 20% or
so), or where the instantaneous velocity vector remains
within a cone with a total angle of about 30°. Significant
errors can occur, however, when the hot-wire anemo-
meter is used in high-intensity or reversing flows such
as that found near buildings. As a result, measure-
ments within and near the building array were made
with a pulsed wire anemometer (PWA, Bradbury and
Castro, 1971), The basic principle of operation of the
PWA is measurement of the transit time of a heat pulse
from a central wire to either of two sensor wires, one
located upstream and the other downstream. The
central wire is pulsed with a high current for a few
microseconds, raising the temperature of the wire to
several hundred degrees Celsius and releasing a tracer
of heated air which is convected away at the
instantaneous flow velocity. The sensor wires are
operated as resistance thermometers and are used to
measure the time-of-arrival of the heated air parcel. The
use of two sensor wires, one on either side of the pulsed
wire, ensures that the flow direction is unambiguously
determined. While the PWA probe can sense only one
velocity component at a time, it can be oriented to
measure velocity components in all three coordinate
directions, PWA calibrations were performed against a
Pilot-static tube mounted in the free-stream of the wind
tunnel in the absence of the spires. All PWA measure-
ments were obtained using a pulsing rate of "lOHzand
an averaging time of 120 seconds at each measurement
location. For each point, the digital time-of-flight signals
were first converted to velocity using the appropriate
calibration curve for each sensor, then the mean,
standard deviation, skewness and kurtosis were calcu-
lated for the sample. These statistics were computed for
each of 1016 coordinate locations extending from 3.5H
upstream of the first building to 7.5H downstream of the
last building and to 3H in the vertical.
The velocity measurements were supplemented
with measurements of pressure coefficient on the top
and on both faces of each building in the array using a
capacitance manometer, but are not reported here.
4. RESULTS AND DISCUSSION
Mean flow and turbulence measurements were
made down the centerline of the 2-d array of buildings in
the direction of the mean wind. Figure 2 shows the
mean wind vectors and the turbulent kinetic energy {tke)
around and upstream of the first three buildings. Here
we see the rotor that forms on the upstream face of the
first building and the single large vortex that forms
between the buildings within each canyon. The
stagnation point on the upstream face of the first
building appears to occur at about 1/2 building height in
close agreement with the findings of Snyder and
Lawson (1994) for a single laterally-wide building. Of
particular interest is the separation zone and reverse
flow that forms on the first building rooftop, but not on
subsequent rooftops. This is in agreement with the
smoke visualization studies of Meroney et al. (1996) that
showed reverse flow occurred on only the first building
for buildings of equal height with W/H = 1. The
streamlines above the leading edge of the second
building are descending slightly, resulting in stronger
downward motion in the first canyon vortex circulation.
Figure 2 also shows a large tke maximum at the
leading edge of the first building and an elevated "tail" of
large tke values extending above the first rooftop, where
the flow jets and shear is strong. A tke "plume" extends
downstream above rooftop level and damps out with
downwind distance. An interesting feature is the
relatively large tke values extending about 1.5 building
heights upstream of the first building. A local maximum
-------
450
0.4 0.6
TKE (mV)
Figure 3. Profiles of tke in the second street canyon.
The downwind position x is measured relative to the
edge of the upwind building.
-1.B -1.S -1.2 -0.9 -0.6 -0.3 0.0 0.3 0.6 0.9 1,2 1.5 1.8
w (ms1)
Figure 4. Mean vertical velocity measured at five
positions in each canyon. Filled symbols - canyon 1,
open symbols - canyon 2, lines - canyons 3-6.
occurs on the upstream face of the first building. (It
remains to be seen how the modified k-E models (e.g.,
Kate and Launder, 1993; Selvam, 1996) will reproduce
this feature, as the traditional k-e models have been
modified to reduce the computed tke upstream of sharp-
edged obstacles.) Relatively speaking, the first canyon
exhibits the largest values of tke. This may result from
advection of a stronger source of tke and/or from
stronger shear due to more impingement on the
downstream building face. Between the buildings,
larger tke values are found on the downstream side of
the canyon. Figure 3 illustrates this more clearly, where
the tke profile closest to the downstream building wall
has nearly twice the tke as compared to the other in-
canyon profiles. Above the rooftop, tke values are
relatively large with the biggest values upstream. Even
over the short distance illustrated here, the above-
rooftop tke decays significantly with downwind distance.
Figure 4 shows vertical profiles of the mean
vertical velocity in several street canyons. In all cases,
the magnitude of the downward motion on the
downstream side of the canyon is stronger than the
magnitude of the upward motion on the upstream side.
From continuity, this suggests that the area of
downward motion should be smaller than the area of
upward motion. The vertical motions are stronger in the
first canyon relative to the other canyons probably due
to the slight downward curvature of the mean flow
streamlines above the leading edge of the second
building (see Rg. 2). Of particular note is that the
vertical velocity appears to reach equilibrium relatively
quickly: in the 3rd or 4th street canyon. This agrees with
the experiments of McDonald et al. (1997) and Hussain
and Lee (1980) who found that equilibrium was reached
after 4 to 5 building rows based on mean velocity
measurements made near rooftop and drag measure-
ments, respectively.
Figure 5 depicts vertical profiles of tke at the center
of each canyon. Above the rooftop, the tke is continual-
ly decreasing in magnitude with increasing distance
from the upstream edge of the first building, the location
of most of the tke production. Hence, it appears that it
takes more than seven building rows to reach equilibri-
um above building rooftop, whereas, below rooftop, the
tke reaches equilibrium relatively quickly by the 3rd or 4*
canyon.
Vertical profiles of the mean horizontal velocity
downstream of the buildings show that the building array
impacts the flow at distances greater than x = 7.5H
downstream of the last building row {Fig. 6). The
reattachment point is between x = 3.5H and 5.5H,
presumably closer to the former. Within a half-building
height downstream of the last building the velocities are
near zero up to about z = 0.8H. Strong reverse flow is
found near the ground between x = 1 .OH and 2.5H.
The velocity profiles upstream of the building array
reveal that the flow at the furthest upstream position
(x/H = -3,3) is already being influenced by the building
-------
J1J
500
400 •
250
Figure 5, Turbulent kinetic energy profiles measured
at the center of each of the canyons in the 2-d building
array. Downwind position x in mm.
array (not shown). The small rotor just upstream of the
first building identified in Fig, 1 extends approximately
0.5H upstream of the array.
5. CONCLUSIONS
Detailed flow measurements have been obtained in
a wind tunnel around a 2-d building array. Measure-
ments include the U, V, and W mean velocity compo-
nents, the ou, ov, and 0W turbulence components, and
pressure on building surfaces. Major findings include:
1. separation only occurred on the first rooftop in
agreement with the smoke visualization studies of
Meroneyetal. (1996);
2. above rooftop tke damped out rapidly with
downwind distance, but still had not reached
equilibrium at building no. 7. The tke and mean
flow within the street canyon, however, reached
equilibrium by the 3rd or 4* canyon;
3. elevated levels of tke extended about 1.5H
upstream of the first building;
4. tke and mean vertical velocity magnitude were
found to be higher on the downstream side of each
canyon; and
5. the reattachment point was found to be close to x =
3.5H downstream of the last building.
Work continues on analysis of the wind and pressure
fields around the 2-d building array. Similar
i
Velocity proiles downstream of 2-D array
compared to approach profile
(x is distance downstream of last building)
— upstream approach flow
* x=0.1H
•» x=0.25H
-O X=0.5H
•m- x=i.oH
B- x=1.5H
-•• x=2.5H
-•- X=3,5H
-I- x=5,5H
-X- X--7.5H
Figure 6. Mean wind velocity profiles measured
downstream of the building array.
experiments have just begun for an 11x7 array of cubes
in order to test the 3-d behavior of CFD codes.
Acknowledgements. We are grateful to Steve Perry,
Jon Reisner, Scott Smith and Roger Thompson for
assistance in planning and preparation for this study.
This work was supported under the DOE CBNP
program managed by Page Stoutland.
6. REFERENCES
Bradbury, L. & Castro, I., 1971: A pulsed-wire technique
for velocity measurements in highly turbulent flows,
J. Fluid Mech., 49, pp. 657-91.
Britter, R. & Hunt, J., 1979: Velocity measurements and
order of magnitude estimates of the flow between
two buildings in a simulated atmospheric boundary
layer. J. Indus. Aerodyn., 4, pp. 165-82.
Cowan, I., 1997: A comparison of wind-tunnel experi-
ments and computational simulations of dispersion
in the environs of buildings, Int. J. Env. & Poll., 8, pp.
699-707.
Dabberdt, W., F. Ludwig, and W. Johnson, 1973:
Validation and applications of an urban diffusion
model for vehicular pollutants, At. Env., 7, pp. 603-
618.
Davidson, M., Snyder, W., Lawson, R. & Hunt, J., 1996:
Wind tunnei simulations of plume dispersion through
groups of obstacles. Atm, Env., 30, pp. 3715-31.
-------
DePaul, F. and C. Sheih, 1985: A tracer study of
dispersion in an urban street canyon, At Env., 19,
pp. 555-559.
DePaut, F. and C. Sheih, 1986: Measurements of wind
velocities in a street canyon, At, Env., 20, pp. 455-
459.
Hosker, R.P. Jr. 1984 Flow and Diffusion Near
Obstacles. Atmospheric Science & Power
Production, DOE/T1C-27601, D. Randerson, Ed., Ch.
7, 241-326. U.S. Dept. of Energy, Wash., DC.
Hosker, R., 1987: The effects of buildings on local
dispersion, Modeling the Urban Boundary Layer,
Am. Met. Soc., Boston.
Hoydysh, W. & W. Dabberdt, 1988: Kinematics and
dispersion characteristics of flows in asymmetric
street canyons. Atm. Env., 22, pp. 2677-89.
Hussain and Lee, 1980: An investigation of wind forces
on three dimensional roughness elements in a
simulated atmospheric boundary layer flow, Report
BS 56, Dept. B!dg. Set, Univ. Sheffield.
Irwin, H., 1981: The design of spires for wind simulation,
J. WindEng, Indust. Aero., 7, pp. 361-366.
Johnson, W., F. Ludwig, W. Dabberdt, & R. Allen, 1973:
An urban diffusion simulation model for carbon
monoxide. J. Air Poll. Cont Assoc., 23, pp, 490-8
Kastner-Klein, P., E. Plate, and E, Fedorovich, 1997:
Gaseous pollutant dispersion around urban-canopy
elements: wind-tunnel case studies, Int. J. Env. &
Poll., 8, pp. 727-737.
Kato, M. and B. Launder, 1993: The modeling of
turbulent flow around stationary and vibrating square
cylinders, Sfh Symp. on Turb. Shear Flows, Kyoto,
Japan.
Kitabayashi, K., 1992: A field study of airflow and tracer
gas diffusion in a model street canyon, 3 World
Clean Air Congr., Montreal, Canada.
Lawson, R. and Ohba, M., 1993: Physical modeling of
the flow field around twin high-rise buildings, fr
AMS J, Cont on Appl. of Air Poll. Met. with AWMA,
Nashville, TN.
Lee, J., D. Call, R, Lawson, W. Clements, and D. Hoard,
1991: A video image analysis system for
concentration measurements and flow visualization
in building wakes, At. Env., 25A, pp, 1211-1225.
McDonald, R., R. Griffiths, & S. Cheah, 1997: Field
experiments of dispersion through regular arrays of
cubic structures, At. Env., 31, pp. 783-795.
McDonald, R. R. Griffiths, D. Hall, 1998: A comparison
of results from scaled field and wind tunnel modeling
of dispersion in arrays of obstacles, At. Env,, 32, pp.
3845-3862.
Meroney, R.N., 1982: Turbulent Diffusion near
Buildings. In Engineering Meteorology (E.J. Plate,
ed.), Ch. 11, 481-525. Elsevier, Amsterdam.
Meroney, R., S. Rafailldis, and M. Paageau, 1996:
Dispersion in idealized urban street canyons, in Air
Pollution Modeling and Its Application XI, 21, NATO.
Nakamura, and T. Oke, 1988: Wind, temperature, and
stability conditions in an east-west oriented urban
canyon, At. Env., 22, pp. 2691-2700.
Oke,T., 1987: Boundary-Layer Climates, Routledge,
London.
Peterka, J., R. Meroney, K. Kothari, 1985, Wind flow
patterns about buildings, J. Wind Eng, & Ind. Aero.,
21, pp. 21-38.
Qin, Y. and S. Kot, 1993: Dispersion of vehicular
emission in street canyons, Guangzhou City, South
China (P.R.C.), At. Env., 27B, pp. 283-291.
' Rafailidis, S. and M. Schatzmann, 1995: Concentration
measurements with different roof patterns in street
canyons with aspect ratios B/H=1/2 and B/H=1,
Universitat Hamburg, Meteorologisches Institut.
Rotach, M., 1995: Profiles of turbulence statistics in and
above an urban street canyon. Atm. Env., 29, pp.
1473-86.
Roth, M. and H. Ueda, 1998: A wind tunnel study of
turbulent flow over a rough surface, 2nd AMS Urb.
Env. Conf, Albuquerque, NM.
Selvam, P., 1996: Computation of flow around Texas
Tech building using k-e and Kato-Launder k-E
turbulence model, Eng. Struct, 18, pp. 856-860,
Snyder, W., 1979: The EPA Meteorological Wind
Tunnel: Its Design, Construction, and Operating
Characteristics, Rpt No. EPA-600/4-79-051, Envir.
Prot, Agcy., Res, Tri. Pk., NC, 78p.
Snyder, W., 1981: Guideline for Fluid Modeling of
Atmospheric Diffusion. Rpt. No. EPA-600/8-81-009,
Envir. Prot. Agcy., Res. Tri. Pk., NC, 200p.
Snyder, W, & R. Lawson, Jr., 1994: Wind-tunnel
measurements of flow fields in the vicinity of
buildings, fl* AMS J. Conf. on Appl. of Air Poll. Met.
with AWMA, Nashville, TN.
Theurer, W., W. Baechlin",' and E. Plate, 1992: Model
study of the development of boundary layers above
urban areas, J. Wind Eng. & Ind. Aerodyn., 41-44,
pp, 437-448.
Wedding, J,, D. Lombard!, and J. Cermak, 1977: A
wind-tunnel study of gaseous pollutants in city street
canyons, J. Air Poll. Control Assoc., 27, pp. 557-566.
Yamartino and Wiegand, 1986: Development and
evaluation of simple models for the flow, turbulence,
and pollutant concentration fields within an urban
street canyon. At. Env., 20, pp. 2137-2156.
DISCLAIMER: This paper has been reviewed in
accordance with the U.S. Environmental Protection
Agency's peer and administrative review policies and
approved for presentation and publication. Mention of
trade names or commercial products does not constitute
endorsement or recommendation for use.
-------
' TflEILL-RTP-AMP-00-01. 1
TECHNICAL REPORT DATA
1. REPORT NO.
EPA/600/A-00/011
2.
3.RECIPIENTS ACCESSION NO.
4. TITLE AND SUBTITLE
Mean Flow and Turbulence Measurements Around a 2-D Array of Buildings
in a Wind Tunnel
5.REPORT DATE
6.PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
1 Michael J. Brown, 2Robert E. Lawson, Jr., 'David S. Decroix, and 'Robert E,
Lee
8.PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
'Los Alamos National Laboroatory
Los Alamos, New Mexico
2Same as Block 12
3Lawrence Livermore National Laboroatory
Livermore, California
10.PROGRAM ELEMENT NO.
II. CONTRACT/GRANT NO,
12. SPONSORING AGENCY NAME AND ADDRESS
National Exposure Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13.TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/9
15, SUPPLEMENTARY NOTES
16. ABSTRACT
In order to predict the dispersion of harmful materials released in or near an urban environment, it is important to first
understand the complex flow patterns which result from the interaction of the wind with buildings and, more commonly,
clusters of buildings. Recent advances in the application of computational fluid dyanmics (CFD) models to such problems
have shown great promise, but there is a need for high-quality data with which to evaluate CFD models. This study was
performed to fill that need for a limited range of conditions.
High-resolution measurements of the three coponenets of the mean and turbulent velocity statistics were obtained around a
2-D array of model buildings in the USEPA meteorological wind tunnel. In this paper, we briefly review prior field and
laboroatory experiemnts on building flows, describe our experimental set-up and measurement apparatus, present the flow
measurements, and discuss their significance in relation to current understanding.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/ OPEN ENDED
TERMS
c.COSATI
18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
7
20. SECURITY CLASS (This Page)
22. PRICE
------- |