This action might not be possible to undo. Are you sure you want to continue?
Gaussian Plumes from “Point” Sources
• Time averaged vs instantaneous plumes
• Simplified steadystate plume model
• “Eddy” diffusion, advection/diffusion equation
• Gaussian point source plume model
• Plume sigma values vs stability and distance
• Plume reflection
• Nongaussian plumes
• Plume Rise; plume trajectories
• Buoyancyinduced dispersion
• Stack downwash
View from below of “Coning” plume
under neutral atmospheric conditions
Source: Slade et al “Meteorology and
Atomic Energy, 1968”
instantaneous
timeaveraged
2
Instantaneous
Plume Shape
Timeaveraged
Plume Shape
Fig 43, p.44 in Martin et al
wind
Describing Plume Concentrations
3
Tim’s Simple Plume Model
h
x
1
2
3
mass/time
passing
point 1
=
mass/time
passing thru
disk area 2
mass/time
passing thru
disk area 3
C
1
> C
2
> C
3
=
Simplified SteadyState Plume Model
*Pollutant is well mixed and confined within the cone
*Pollutant is continuously swept thru the cone by the wind
Concentration vs. distance downwind depends upon cone shape
4
( )( ) 2 disk of area speed wind
rate emission Mass
2 at air of ion Concentrat ·
1
2
Simple Model #1:
( )( )
2 3
m m/sec
µg/sec
m
µg
·
Disk shape depends upon stability category
More unstable and thus more pronounced vertical spreading
5
Perspective
View of “Fanning”
Plume in Very
Stable Air
Source: Slade et al “Meteorology and
Atomic Energy, 1968”
Less vertical
motion
6
unstable
neutral
stable
A A
Timeaveraged
concentration
across AA
Gaussian (normal) distribution
occurs across AA due to changes in
wind direction over averaging time
Shape is described by
“plume sigmas”
Most probable
wind direction
source
wind
Mass is not uniformly
distributed within the
cone’s volume
σ
More Detailed Plume Model
7
( )( )
[ ] function on distributi Gaussian
2 disk of area speed wind
rate emission Mass
2 at Conc ·
1
2
Simple Model #2:
x
z
y
X is the timeaveraged
wind direction,
Y is the crosswind
direction,
Z is the vertical dimension
( )( )
[ ] − ·
2 3
m m/sec
µg/sec
m
µg
Gaussian Plume Model
1
2
In order to derive an equation describing the
distribution of mass within the plume, we must
first consider the transport of mass within a
small control volume
8
x
z
y
*Transport of mass in x direction depends on the average horizontal wind
*Transport of mass in the y and z directions depends on turbulent motions
dx
dy
dz
“Eddy” Diffusion
Pollutant molecules are
moving around in random
directions due to random
eddy motions
Consider two enclosed air volumes separated by a wall
(arrows represent eddy motions; balls represent pollutant molecules)
Random eddy motions
in pollutionfree air
Now remove the wall between the enclosures
Size of eddies >
size of pollutant
molecules
9
The rate of pollutant molecules crossing plane AA depends upon the
concentration difference between the two sides. Specifically, this rate
=K δc/δx ,where K is termed an “eddy diffusivity” with units of m
2
/sec.
The magnitude of K depends upon the magnitude of the eddy motions.
A
A
x
Eventually in this case, the rate of net rate of pollution crossing AA is
zero (an equal number of molecules cross in both directions). However
the value of K remains constant throughout the “experiment”
dx
dy
dz
Changes in x Direction:
Net rate of change of mass flow =
(Mass flow rate in)  (Mass flow rate out)
Mass Flow Rate In = C u A
yz
{ [µg/m
3
] [m/s] [m
2
]}
( )dx
x
yz yz
A u C A u C Out Rate Flow Mass
∂
∂
+ ·
( ) ( ) V u C  A u C  Change of Rate Net
yz
x
dx
x ∂
∂
·
∂
∂
·
V =Volume = dxdydz
A
xy
= dxdy
A
yz
= dydz
A
xz
= dxdz
10
dx
dy
dz
Changes in z Direction via “ Turbulent Diffusion”:
V =Volume = dxdydz
A
xy
= dxdy
A
yz
= dydz
A
xz
= dxdz
{ ¦ ] sec ][m m g ][ ][m [m C K
z
A  In Rate Flow Mass
1  2 3  1  2
z xy
µ
∂
∂
·
( ) V C K
z z
Change of Rate Net
z
¹
'
¹
¹
'
¹
∂
∂
∂
∂
·
A similar result is obtained in the y direction. Given that the net
rate of change in the volume [ = V(δc/δt)] is the change in all three
directions , we obtain an overall expression in terms of x, y and z.
¹
'
¹
¹
'
¹
∂
∂
∂
∂
+
¹
'
¹
¹
'
¹
∂
∂
∂
∂
+
∂
∂
− ·
∂
∂
z
C
z y
C
y x
C
u
C
z y
K K
t
“AdvectionDiffusion” Equation
+ other losses
due to
deposition and
chemical
reactions
= 0 for steady
state models
“Advection”, i.e.,
transport by the mean
wind, u
Effect of turbulent “diffusion”, i.e.,
exchange of polluted air parcel
with surrounding air parcels. If the
surrounding air is cleaner, δC/δz &
δC/δy are negative. K is the “eddy
diffusivity” and represents the intensity
of turbulent motions and varies with
stability
11
The Gaussian plume equation is a particular solution to
this more general equation under the following assumptions:
0
C
·
∂
∂
t
* Constant wind speed with height (u does not depend on z)
* Steady state conditions
* Constant eddy diffusivity (K does not depend on y or z)
* Mass is conserved
{ ¦ g/sec 0] [for x Q C µ > ·
∫ ∫
∞
∞ −
∞
∞ −
dydz
Define:
u
x K 2
ó
z 2
z
·
u
x K 2
ó
y 2
y
·
Gaussian “Point” Source Plume Model:
Wind speed
evaluated at
“effective”
release height
Mass
emission
rate
}
Corresponds to
disk area in
simple model
(values depend
upon downwind
distance, x)
Distribution of mass in
vertical dimension (z)
at a given downwind
distance, x (includes
the effect of surface
reflection)
Distribution of
mass in crosswind
dimension (y) at a
given downwind
distance, x
Pollutant
concentration as a
function of
downwind
position (x,y,z)
( ) ( )
( )
¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.

−
¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.

+ −
+
,
`
.

− −
·
2
y
2
2
2
2
2
z y
2ó
y
exp
2
exp
2
exp
u 2
Q
z) y, C(x,
z z
h z h z
σ σ σ σ π
“Effective” stack height,
including rise of the hot
plume near the source
12
Gaussian Plume
(Concentrations vary with x, y and z)
For a given x, the max conc. is at the
plume centerline and decreases
exponentially away from the centerline
at a rate dependent upon the sigma
values, σ
y
and σ
z
.
σ
y
and σ
z
are functions of x
Plume
Centerline
Crosswind distance from plume centerline (m)
V
e
r
t
i
c
a
l
d
i
s
t
a
n
c
e
f
r
o
m
p
l
u
m
e
c
e
n
t
e
r
l
i
n
e
(
m
)
Concentration distribution in a Gaussian plume
(σ
y
= 20 m; σ
z
= 10m; centerline concentration = 1.0)
Source: Hanna et al, 1981
Note: theoretical plume has infinite extent in all directions!
σ
y
σ
z
13
Sigmay
x
Sigmaz
14
( ) Θ · tan 11628 . 465 x
y
σ
( ) ( ) x d c ln 017453293 . 0 − · Θ
x is in kilometers
σ
y
is in meters
Θ is in radians
Crosswind distribution:
b
z
ax · σ
Vertical distribution:
x is in kilometers
σ
z
is in meters
a, b depend on x
Plume sigma formulas from EPA’s ISC Model
Pasquill
Stability
Category
x (km)
a b
A
*
<.10
0.10  0.15
0.16  0.20
0.21  0.25
0.26  0.30
0.31  0.40
0.41  0.50
0.51  3.11
>3.11
122.800
158.080
170.220
179.520
217.410
258.890
346.750
453.850
**
0.94470
1.05420
1.09320
1.12620
1.26440
1.40940
1.72830
2.11660
**
*
If the calculated value of óz exceed 5000 m, óz is set to 5000 m.
b
z
ax · σ
15
B
*
<.20
0.21  0.40
>0.40
90.673
98.483
109.300
0.93198
0.98332
1.09710
C
*
All 61.141 0.91465
D <.30
0.31  1.00
1.01  3.00
3.01  10.00
10.01  30.00
>30.00
34.459
32.093
32.093
33.504
36.650
44.053
0.86974
0.81066
0.64403
0.60486
0.56589
0.51179
*
If the calculated value of óz exceed 5000 m, óz is set to 5000 m.
**
óz is equal to 5000 m.
Pasquill
Stability
Category
x (km)
a b
b
z
ax · σ
Pasquill
Stability
Category
x (km)
a b
E <.10
0.10  0.30
0.31  1.00
1.01  2.00
2.01  4.00
4.01  10.00
10.01  20.00
20.01  40.00
>40.00
24.260
23.331
21.628
21.628
22.534
24.703
26.970
35.420
47.618
0.83660
0.81956
0.75660
0.63077
0.57154
0.50527
0.46713
0.37615
0.29592
F <.20
0.21  0.70
0.71  1.00
1.01  2.00
2.01  3.00
3.01  7.00
7.01  15.00
15.01  30.00
30.01  60.00
>60.00
15.209
14.457
13.953
13.953
14.823
16.187
17.836
22.651
27.074
34.219
0.81558
0.78407
0.68465
0.63227
0.54503
0.46490
0.41507
0.32681
0.27436
0.21716
b
z
ax · σ
16
0.36191 4.1667 F
0.54287 6.2500 E
0.72382 8.3330 D
1.0857 12.5000 C
1.8096 18.3330 B
2.5334 24.1670 A
d c Pasquill Stability
Category
( ) ( ) x d c ln 017453293 . 0 − · Θ
Plume “Reflection” off of the Ground
(pollutant cannot penetrate the ground)
Actual Source
“Virtual” Source
(below the surface)
Most of plume
above the surface
Plume begins to
“reflect” off the
surface
Reflected
material (shaded
area)
Resulting in an
asymmetric
vertical profile
Reflection is modeled by adding a “virtual” source contribution to the “real” one
h
17
x
A virtual” source can also be used to
model the effect of hot, rising plumes
Physical
stack
height,h
s
Plume
rise,
∆h
“Effective”
stack
height, h
Concentration vs. x at ground level (z=0)
(note maximum at x > 0)
Concentration in
elevated plume
at height z
Example Calculation
( ) ( ) ( )
¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.

¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.

+ −
+
,
`
.

− −
·
2
y
2
2
2
2
2
z y
2ó
y 
exp
2
exp
2
exp
u 2
Q
z) y, C(x,
z z
h z h z
σ σ σ σ π
Given:
Q = 10 grams/sec; h = 50m; x = 500 m = 0.5 km; u
100
= 6 m/s; Stability
Class “D”
Compute:
C(500, 0, 0) ,i.e., the ground level concentration at plume centerline, 500
meters downwind.
18.3m ) 32.093(0.5 ax ó
0.81066 b
z
· · ·
[ ] radians 1542 . 0 ) 0.5 0.72382ln  (8.3330 .017453293 0 · · Θ
[ ] m 1 . 36 ) 1542 . 0 tan( ) 5 . 0 ( 11628 . 65 4 ) (tan 11628 . 65 4 ó
y
· · Θ · x
( )( )( )
( )
( )
( )
( )
( )
( )
¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.

¹
¹
¹
'
¹
¹
¹
¹
'
¹
,
`
.
 + −
+
,
`
.

− −
·
2
2
2
2
2
2
36.1 2
0 
exp
3 . 18 2
50 0
exp
3 . 18 2
50 0
exp
3 . 18 1 . 36 6 2
10
C(500,0,0)
π
( )( )( )
{ ¦{ ¦
3 3 5
g/m 19.2 g/m x10 92 . 1 1 0.0479
3 . 18 1 . 36 6 2
10
C(500,0,0) µ
π
· · ·
−
18
Virtual Source Also Used to Model Reflection
Off of the Top of the Mixed Layer
(added to direct plume concentration)
Eventually
wellmixed
Plume Reflection off of the top of the mixed layer
wind
z
ii
19
Plume “Trapped” in Stable Layer Above Mixed Layer
Time of Day
Date
Latitude
Land Use
Cloud Cover
Hourly Wind Speed
Predawn dθ/dz
Stability Category
Mixing Depth
Wind Speed Profile
Atmospheric
Dispersion
Model
Minimum Necessary Information
Needed to Implement a Simple
Atmospheric Dispersion Model
Wind Direction
Source Emission Rate
Source Geometry
20
NonGaussian Plumes
water
land
source
wind
Very Narrow Stable
Plume over Water
(as viewed from above)
21
Water (stable)
Land (unstable)
source
‘Fumigation’
Plume Fumigation During Onshore Flow
wind
Plume “Trapped” in Building Wake
22
Plume “Looping” During Unstable Conditions
(largescale vertical motions)
Extreme
Departure
From
Gaussian
23
Many tall industrial stacks release hot, effluents into the air. Hot air rises
and cools. There are two well established results from the science of
plume rise: the immediate downwind trajectories of "bent over" plumes
and the maximum attained height of "vertical" plumes.
Plume Rise in Neutral or Stable Air
Vertical Plume Rise, ∆h ‘Bent over’ plume trajectory, z’
z
x
x
final
h
s
∆h
z’
h
s
h
∆h
(z’ is a function of x)
Low wind speeds in stable air Higher wind speeds in stable/neutral air
wind
24
Near the stack (x< x
final
) ,
the plume trajectory is
dominated by buoyant rise
as shown in the equation
below.
x
‘Bent over’ plume trajectory under
stable and neutral conditions
h
x
final
h
s
∆h
u
x F 1.6
'
3
2
0
· z
F
0
is the initial
buoyancy flux at the
stack exit
[m
4
s
3
]
The final plume rise, ∆h,
is a complex function of
wind speed and stability
and applies when x > x
final
There is also some rise
due to momentum
(stack gas velocity),
but this is usually
small compared with
the effect of buoyancy.
Buoyancy Flux, F
0
( ) ( )
2
0 0 0 0
R w T T
T
g
F − ·
T
0
= Stack gas exit temperature (K)
w
0
= Stack gas exit velocity (m/s)
R
0
= Stack radius at exit (m)
The larger the stack radius,
the more time it takes for
the gas to cool by mixing
with surrounding, cooler air
25
Dimensionless downwind parameter (x u
1
s
0.5
)
D
i
m
e
n
s
i
o
n
l
e
s
s
p
l
u
m
e
r
i
s
e
p
a
r
a
m
e
t
e
r
(
∆
h
[
u
s
]
0
.
3
3
3
F
0

0
.
3
3
3
)
Observed Plume Trajectories
Source: Simon and Proudfit (1967);
data from Ravenswood power plant,
New York
( )
375 . 0
25 . 0
0
4
s
F
h · ∆
,
`
.
 Θ
− ·
dz
d
T
g
s 0098 . 0
Where s is the stability
parameter, a continuous
descriptor of the strength of the
atmospheric restoring force under
stable conditions
Vertical Plume Rise Under Stable Conditions
∆h
h
s
26
x
A virtual” source can be used to approximate the
effect of plume rise on downwind concentrations
Physical
stack
height,h
s
Plume
rise,
∆h
“Effective”
stack
height, h
Buoyancy Induced Dispersion
Near the source, the rising plume entrains surrounding air as it rises.
This dilution of the plume is not accounted for in the classic plume
dispersion equations. To account for this extra dilution, an additional
term is added to the “plume sigmas” (plume spreading parameters) that
is a function of plume trajectory as follows:
( )
2
2
2
5 . 3
'
z effective z
z
σ σ +
,
`
.

·
( )
2
2
2
5 . 3
'
y
effective
y
z
σ σ +
,
`
.

·
27
StackInduced Plume Downwash
Wind flowing past a stack can create a region of lower pressure
immediately downwind of the stack. If the vertical momentum
of the stack gas is not sufficient, the plume will be drawn
downward on the downwind side of the stack, lowering the
effective stack height, h.
For w
0
< 1.5 u
,
`
.

− + · 5 . 1 2
0
0
'
u
w
D h h
s s
For w
0
>= 1.5 u h
s
’ = h
s
Where h
s
’ is the adjusted physical stack height (not including plume rise)
Instantaneous Plume Shape
Timeaveraged Plume Shape
Describing Plume Concentrations
wind
Fig 4 3, p.44 in Martin et al
2
distance downwind depends upon cone shape 3 .Tim’s Simple Plume Model h x Simplified SteadyState Plume Model *Pollutant is well mixed and confined within the cone *Pollutant is continuously swept thru the cone by the wind 3 2 1 mass/time passing point 1 = mass/time passing thru disk area 2 C1 > C 2 > C 3 = mass/time passing thru disk area 3 Concentration vs.
Simple Model #1: Concentration of air at 2 = Mass emission rate (wind speed )(area of disk 2) µg = µg/sec m 3 (m/sec)( 2 ) m 2 1 Disk shape depends upon stability category More unstable and thus more pronounced vertical spreading 4 .
Perspective View of “Fanning” Plume in Very Stable Air Less vertical motion Source: Slade et al “Meteorology and Atomic Energy. 1968” 5 .
unstable neutral stable More Detailed Plume Model Mass is not uniformly distributed within the cone’s volume wind source A A Timeaveraged concentration across AA σ Gaussian (normal) distribution occurs across AA due to changes in wind direction over averaging time Shape is described by “plume sigmas” Most probable wind direction 6 .
Simple Model #2: Conc at 2 = ( wind speed)(area of disk 2 ) Mass emission rate [Gaussian distribution function ] µg µg/sec [ ] = − 3 2 m (m/sec )( ) m 2 1 y z x X is the timeaveraged wind direction. Z is the vertical dimension Gaussian Plume Model 2 1 In order to derive an equation describing the distribution of mass within the plume. we must first consider the transport of mass within a small control volume 7 . Y is the crosswind direction.
z y x dy dz dx *Transport of mass in x direction depends on the average horizontal wind *Transport of mass in the y and z directions depends on turbulent motions “Eddy” Diffusion Consider two enclosed air volumes separated by a wall (arrows represent eddy motions. balls represent pollutant molecules) Size of eddies > size of pollutant molecules Pollutant molecules are moving around in random directions due to random eddy motions Random eddy motions in pollutionfree air Now remove the wall between the enclosures 8 .
However the value of K remains constant throughout the “experiment” dx dy dz V =Volume = dxdydz Axy = dxdy Ayz = dydz Axz = dxdz Changes in x Direction: Net rate of change of mass flow = (Mass flow rate in) .where K is termed an “eddy diffusivity” with units of m2 /sec. the rate of net rate of pollution crossing AA is zero (an equal number of molecules cross in both directions).(Mass flow rate out) Mass Flow Rate In = C u Ayz Mass Flow Rate Out = C u A yz + { [µg/m3 ] [m/s] [m2 ]} ∂ C u A yz dx ∂x ( ) Net Rate of Change =  ∂ (C u A yz )dx = . this rate =K δc/δx . Specifically.The rate of pollutant molecules crossing plane AA depends upon the concentration difference between the two sides. A A x Eventually in this case. The magnitude of K depends upon the magnitude of the eddy motions.∂ (C u ) V ∂x ∂x 9 .
δC/δz & δC/δy are negative. i. exchange of polluted air parcel with surrounding air parcels..A xy ∂ { K z C} ∂z [m 2 ][m1 ][ µg m 3 ][m 2 sec1 ] Net Rate of Change = ∂ ∂ (K z C ) V ∂ z ∂z A similar result is obtained in the y direction. y and z. Given that the net rate of change in the volume [ = V(δc/δt)] is the change in all three directions . “AdvectionDiffusion” Equation ∂C ∂C ∂ ∂C ∂ ∂C + other losses = −u + K + K due to ∂t ∂x ∂y y ∂y ∂z z ∂z deposition and chemical reactions = 0 for steadystate models “Advection”.e. transport by the mean wind. we obtain an overall expression in terms of x. If the surrounding air is cleaner.e.dx dy dz V =Volume = dxdydz Axy = dxdy Ayz = dydz Axz = dxdz Changes in z Direction via “ Turbulent Diffusion”: Mass Flow Rate In = . K is the “eddy diffusivity” and represents the intensity of turbulent motions and varies with stability 10 . i. u Effect of turbulent “diffusion”..
y.z) Gaussian “Point” Source Plume Model: Mass emission rate “Effective” stack height.y. including rise of the hot plume near the source Q C(x. x) Distribution of mass in vertical dimension (z) at a given downwind distance. x (includes the effect of surface reflection) Distribution of mass in crosswind dimension (y) at a given downwind distance. x 11 . z) = 2π u σyσz − (z − h exp 2 2σz ) 2 ( + exp − z + h 2σz2 ) 2 − (y)2 exp 2ó 2 y } Wind speed evaluated at “effective” release height Corresponds to disk area in simple model (values depend upon downwind distance.The Gaussian plume equation is a particular solution to this more general equation under the following assumptions: * Steady state conditions ∂C =0 ∂t * Constant wind speed with height (u does not depend on z) * Constant eddy diffusivity (K does not depend on y or z) Define: ó2 = z 2 Kz x u ó2= y 2 Ky x u * Mass is conserved ∞ −∞ ∫∫ ∞ −∞ Cdydz = Q [for x > 0] {µg/sec } Pollutant concentration as a function of downwind position (x.
σy and σz are functions of x Concentration distribution in a Gaussian plume (σ y = 20 m. 1981 12 . σ z = 10m. centerline concentration = 1. σy and σz. y and z) Plume Centerline For a given x. the max conc.Gaussian Plume (Concentrations vary with x.0) Note: theoretical plume has infinite extent in all directions! Vertical distance from plume centerline (m) σy σz Crosswind distance from plume centerline (m) Source: Hanna et al. is at the plume centerline and decreases exponentially away from the centerline at a rate dependent upon the sigma values.
Sigmay x Sigmaz 13 .
12620 1.520 217.40 0.11660 ** * If the calculated value of ó z exceed 5000 m. 14 .080 170.11628 x(tan Θ ) Θ = 0.21 .25 0.3.0.51 .50 0.10 0.40940 1.0.30 0.31 .11 >3.72830 2.890 346.05420 1. ó z is set to 5000 m.0.20 0. b depend on x Crosswind distribution: σ y = 465.26440 1.Plume sigma formulas from EPA’s ISC Model Vertical distribution: σ z = ax b x is in kilometers σz is in meters a.11 a 122.16 .0.10 .41 .15 0.750 453.800 158.94470 1.0.017453293(c − d ln(x )) x is in kilometers σy is in meters Θ is in radians Pasquill Stability Category A* σ z = ax b x (km) <.850 ** b 0.0.09320 1.220 179.26 .410 258.
32681 0.01 . ** Pasquill Stability Category E σ z = ax b x (km) <.483 109.650 44.651 27.40.00 7.40 a 90.29592 0.63227 0.01 .01 .628 22.10.4.00 30.219 b 0.15.37615 0.0.504 36.81066 0.141 34.00 >30.1.00 3. óz is equal to 5000 m.053 b 0.10 0.31 .21716 F <.60486 0.953 14.30.83660 0.00 >60.09710 0.70 0.00 15 .074 34.420 47.01 .618 15.40 >0.01 .673 98.00 10.093 32.1.46713 0.187 17.81558 0.823 16.00 1.30 0.00 1.3.01 .30.91465 0.64403 0.00 3.628 21.260 23.00 >40.54503 0.60.Pasquill Stability Category B * σ z = ax b x (km) <.01 .01 .0.953 13.01 .534 24.01 .2.00 4.01 .00 20.703 26.98332 1.20 0.300 61.71 .00 D * If the calculated value of óz exceed 5000 m.46490 0.10 .01 .86974 0.81956 0.51179 C * All <.836 22.01 .30 0.093 33. óz is set to 5000 m.2.31 .1.459 32.331 21.50527 0.00 a 24.21 .41507 0.93198 0.20 0.3.75660 0.0.7.00 1.00 2.01 .209 14.27436 0.20.00 2.21 .00 15.68465 0.63077 0.10.78407 0.970 35.57154 0.56589 0.00 10.457 13.
Θ = 0.5000 8.54287 0.017453293(c − d ln(x )) Pasquill Stability Category A B C D E F c 24.5334 1.0857 0.2500 4.36191 Plume “Reflection” off of the Ground (pollutant cannot penetrate the ground) Reflection is modeled by adding a “virtual” source contribution to the “real” one Most of plume above the surface Plume begins to “reflect” off the surface Resulting in an asymmetric vertical profile Actual Source h Reflected material (shaded area) “Virtual” Source (below the surface) 16 .3330 12.3330 6.72382 0.8096 1.1670 18.1667 d 2.
(y ) 2ó 2 y C(500.3330 . 500 meters downwind. Stability Class “D” Compute: C(500.5 ) 0. 0.1542 ) ] = 36 .3m Θ = 0.i.3) 2(18.81066 = 18. rising plumes “Effective” stack height. u 100 = 6 m/s. ó z = axb = 32.0. x = 500 m = 0.5 km.2µg/m3 2π (6 )(36.72382ln [0.0) = {0. ∆h Physical stack height.3) 17 .0479}{1} = 1.92 x10 −5 g/m3 = 19.11628 x(tanΘ) = 465 .1)(18. 11628 ( 0.1)(18.5]) = 0 .e.A virtual” source can also be used to model the effect of hot.. x at ground level (z=0) (note maximum at x > 0) Example Calculation Given: Q = 10 grams/sec.017453293 (8.093(0.z) = Q 2π u σ y σ z − (z − h exp 2 2σ z ) 2 + exp − (z + h 2σ 2 z ) 2 2 exp . the ground level concentration at plume centerline. 5)[tan(0.(0 ) exp 2 2 2 2(18. 0) .0) = 2 2 − (0 − 50 )2 10 + exp − (0 + 50 ) exp .3) 2π (6 )(36 .0.1m C(x.1542 radians ó y = 465 .h s Concentration in elevated plume at height z x Concentration vs.3) 2(36.0. h = 50m.1) 10 C(500. h Plume rise. y.
Virtual Source Also Used to Model Reflection Off of the Top of the Mixed Layer Eventually wellmixed (added to direct plume concentration) Plume Reflection off of the top of the mixed layer wind zi 18 .
Plume “Trapped” in Stable Layer Above Mixed Layer Minimum Necessary Information Needed to Implement a Simple Atmospheric Dispersion Model Time of Day Date Latitude Land Use Cloud Cover Hourly Wind Speed Predawn dθ/dz Stability Category Mixing Depth Wind Speed Profile Atmospheric Dispersion Model Wind Direction Source Emission Rate Source Geometry 19 .
NonGaussian Plumes Very Narrow Stable Plume over Water (as viewed from above) wind land water source 20 .
Plume Fumigation During Onshore Flow wind source Water (stable) Land (unstable) ‘Fumigation’ Plume “Trapped” in Building Wake 21 .
Plume “Looping” During Unstable Conditions (largescale vertical motions) Extreme Departure From Gaussian 22 .
Hot air rises and cools. z’ (z’ is a function of x) Vertical Plume Rise. ‘Bent over’ plume trajectory. There are two well established results from the science of plume rise: the immediate downwind trajectories of "bent over" plumes and the maximum attained height of "vertical" plumes. effluents into the air.Plume Rise in Neutral or Stable Air Many tall industrial stacks release hot. ∆h wind z’ hs z x ∆h ∆h h hs xfinal Higher wind speeds in stable/neutral air Low wind speeds in stable air 23 .
but this is usually small compared with the effect of buoyancy. is a complex function of wind speed and stability and applies when x > x final xfinal z' = 1.6 3 F0 x 2 u F0 is the initial buoyancy flux at the stack exit [m 4 s3 ] Buoyancy Flux. cooler air F0 = There is also some rise due to momentum (stack gas velocity).‘Bent over’ plume trajectory under stable and neutral conditions ∆h h hs Near the stack (x< x final) . the plume trajectory is dominated by buoyant rise as shown in the equation below. the more time it takes for the gas to cool by mixing with surrounding. F 0 The larger the stack radius. g (T0 − T ) w 0 (R 0 )2 T T0 = Stack gas exit temperature (K) w0 = Stack gas exit velocity (m/s) R0 = Stack radius at exit (m) 24 . x The final plume rise. ∆h.
25 hs Where s is the stability parameter. New York Vertical Plume Rise Under Stable Conditions 4 (F ) ∆h = 00375 s. a continuous descriptor of the strength of the atmospheric restoring force under stable conditions s= g dΘ 0.Observed Plume Trajectories Dimensionless plume rise parameter (∆h [u s]0. data from Ravenswood power plant.0098 − T dz 25 . ∆h 0.333 ) Dimensionless downwind parameter (x u 1 s0.333 F0 0.5 ) Source: Simon and Proudfit (1967).
This dilution of the plume is not accounted for in the classic plume dispersion equations.A virtual” source can be used to approximate the effect of plume rise on downwind concentrations “Effective” stack height.5 2 26 . To account for this extra dilution. the rising plume entrains surrounding air as it rises. an additional term is added to the “plume sigmas” (plume spreading parameters) that is a function of plume trajectory as follows: (σ ) 2 z effective z' 2 = +σ z 3.h s x Buoyancy Induced Dispersion Near the source.5 2 (σ ) 2 y effective z' 2 = +σy 3. h Plume rise. ∆h Physical stack height.
5 u hs’ = hs For w0 >= 1.StackInduced Plume Downwash Wind flowing past a stack can create a region of lower pressure immediately downwind of the stack. h. For w0 < 1. lowering the effective stack height. the plume will be drawn downward on the downwind side of the stack.5 u Where hs’ is the adjusted physical stack height (not including plume rise) 27 . If the vertical momentum of the stack gas is not sufficient.5 u w hs' = hs + 2 D0 0 − 1.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue listening from where you left off, or restart the preview.