A STUDY OF STABILITY CONDITIONS IN AN URBAN AREA Page: 4 of 10
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
near the surface, wind and concentration profiles
adjacent to the outside walls of several buildings were
also taken. Furthermore, a pseudo-tower, supported by
a 90-m crane and fitted with sonic anemometers at
eight levels, was deployed downwind at about 750 m
from downtown OKC for turbulence observations. In
one nocturnal case, balloons were deployed close to
the tracer release area. Many of the released balloons
exhibited quick ascents from ground level to the top of
buildings, implying highly convective conditions.
3. MODAL-DATA COMPARISON
In this study, airflow and dispersion simulations for
the first continuous release of IOPs 2 and 8, a daytime
and a nighttime release respectively, were performed.
In each case, SF6 was released near the ground as a
point source for 30-min, with a release rate of 5.0 g/s
for IOP 2 and 3.1 g/s for IOP 8. Shown in Fig. 1 are the
footprints of buildings in the central business district of
OKC, with the Westin release location indicated by the
red dot. The tallest building in the area is approximately
120-m high and the average building height of the area
is -30 m.
Fig. 1. Footprints of buildings in the central business
district of Oklahoma City and the Westin release
location (red dot) for IOP 2 and IOP 8.
In the numerical simulations, a domain size of
1,030 m x 3,010 m x 425 m (in lateral, longitudinal, and
vertical directions) was employed. A graded mesh
consisting of 201 x 303 x 45 grid points, with a minimal
grid spacing of -1 m near the ground surface, was
used. Most of the buildings within 500 m of the release
point were explicitly resolved and the remaining
buildings were treated as virtual buildings.
Steady logarithmic velocity profiles were used
as inflow boundary conditions. These profiles were
created, based on the 15-minute averaged wind
speeds and directions from the PNNL sodar located
approximately 2 km SSW of downtown OKC and the
hourly averaged data from the weather station on
the rooftop of St. Anthony's hospital at -1.5 km NW
of downtown OKC. The estimated wind speed is 5
m/s at z=50 m for both IOPs and the estimated wind
direction is 2150 for IOP 2 and 1550 for IOP 8,
For each simulated release, a quasi-steady
state flow field was established after -15 minutes of
simulated time prior to the start of the dispersion
simulation. The release of SF6 was modeled as a
continuous source over a small area (covered by 2 x
2 cells on the ground surface) at a constant release
rate and dispersion results indicate steady state was
reached in about 20 minutes of simulated time. For
both cases, the RANS approach with a non-linear
eddy viscosity (NEV) turbulence model (Gresho and
Chan, 1998) was used and neutral atmospheric
stability was assumed.
In the following, model predictions of flow and
concentration in the near and intermediate regions of
the release point are presented and compared with
observed data. For brevity, only major results are
presented and compared herein. Several of the
statistical performance measures recommended by
Hanna, et al. (2005) are used to indicate the
performance of our model. They are: the factor of two
or five (FAC2 or FAC5), fractional bias (FB), the
geometric mean bias (MG), and the normalized mean
square error (NMSE). For differences in angles
between predicted and measured wind vectors, the
formula of scaled angle differences (SAA) with larger
vectors carrying more weights, devised by Calhoun,
et al. (2004), is used.
3.1 IOP 2
Airflow in urban areas is extremely complex, with
features such as flow separations, local stagnation
regions, eddies of various size, and high velocity jets
in street canyons. These features were all observed in
our model simulations. Due to space limitations, the
simulated flow field is not presented here, however, a
quantitative model-data comparison of wind vectors at
a number of locations is presented. In Fig. 2, the
predicted wind vectors in the downtown area are
compared with the 30-min averaged data measured
by Dugway Proving Ground (DPG) PWIDS. In
general, the agreement between model predictions
and field observations is very good. The statistical
performance measures are: SAA=15, FAC2=0.6,
FB=-0.04, MG=0.71, and NMSE=0.41, respectively.
SoPG, SuprPWID 20,
Here’s what’s next.
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Chan, S T & Lundquist, J K. A STUDY OF STABILITY CONDITIONS IN AN URBAN AREA, article, November 1, 2005; Livermore, California. (digital.library.unt.edu/ark:/67531/metadc873199/m1/4/: accessed July 21, 2018), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.