Environmental Engineering Reference
In-Depth Information
objectives of this system are to better characterize the emissions from the fires, to
accurately simulate the dispersion of the smoke plumes, and to increase the
resolution of regional-scale models so that the impacts of PB emissions can be
discerned from other pollution sources in the region. The approach taken is to
utilize a dynamic, solution-adaptive grid algorithm to increase grid resolution locally
around the PB plume. Turbulence parameterization is being revised in view of the
refined grid scales, and Daysmoke, a plume model specifically designed for PB
plumes, is being used as the sub-grid model.
2. Adaptive Grid Models
An adaptive grid version of MM5 was developed to predict the occurrence and
extent of optical scale turbulence [1]. Using a dynamic adaptive grid algorithm, it
is possible to resolve gravity waves, shear or optical turbulence sufficiently to
provide accurate predictions. The standard MM5 turbulence parameterization was
replaced with a physically based, four equation, LES/RANS turbulence model that
outputs directly the dissipation rates needed to calculate optical turbulence without
parameterization.
The adaptive grid air quality modeling methodology of Odman et al. [2] was
implemented in CMAQ. The grid adaptation is restricted to the horizontal plane. A
2-D weight function, consisting of the Laplacian of the ground-level PM
2.5
con-
centration field, determines where grid nodes are to be clustered for a more accurate
solution. The concentrations are then interpolated to the new grid positions.
Owing to the equivalence of interpolation to numerical advection, a higher order
advection scheme is used for interpolation. Emissions and meteorological data are
also processed to the new grid.
The grid becomes non-uniform after adaptation but, for easy computation of
the solution, a coordinate transformation reestablishes the uniform grid. The
Jacobian of the coordinate transformation from the physical (
x
,
y
,
z
) space to the
computational
(
ξ
,
,
)
space is calculated as
1
∂
z
⎛
∂
x
∂
y
∂
y
∂
x
⎞
J
=
⎜
⎝
−
⎟
⎠
.
2
m
∂
σ
∂
ξ
∂
η
∂
ξ
∂
η
The governing equation of the CMAQ model is modified as
( )
(
) (
) (
)
ξ
η
σ
∂
Jc
∂
Jv
c
∂
Jv
c
∂
Jv
c
∂
⎛
∂
c
⎞
ξξ
n
+
n
+
n
+
n
+
⎜
⎝
JK
n
⎟
⎠
∂
t
∂
ξ
∂
η
∂
σ
∂
ξ
∂
ξ
⎛
∂
∂
c
⎞
∂
⎛
∂
c
⎞
ηη
σσ
+
⎜
⎝
JK
n
⎟
⎠
+
JK
n
=
JR
+
JS
⎝
⎠
n
n
∂
η
∂
η
∂
σ
∂
σ
