### Air Quality Models

```Air Quality Models
Web Sites
• USEPA, Technology Transfer Network (TTN),
Support Center for Regulatory Atmospheric
Modeling (SCRAM),
http://www.epa.gov/ttn/scram/
• 台灣空氣品質模式支援中心
http://aqmc.epa.gov.tw/

• 高斯模式(Gaussian Model)
– Gaussian Plume Model
– Gaussian Puff Model
• 拉格蘭吉軌跡模式(Lagrangian trajectory
Model)
• 尤拉網格(Eulerian Grid Model)

USEPA 將模式分為:
• Dispersion Modeling : These models are typically used in the
permitting process to estimate the concentration of pollutants at
specified ground-level receptors surrounding an emissions source.
– Preferred/Recommended Models: AERMOD, CALPUFF, …
– Alternative Models
– Screening Models
• Photochemical Modeling : These photochemical models are largescale air quality models that simulate the changes of pollutant
concentrations in the atmosphere using a set of mathematical
equations characterizing the chemical and physical processes in the
atmosphere. These models are applied at multiple spatial scales
from local, regional, national, and global.
• Receptor Modeling : Receptor models are mathematical or
statistical procedures for identifying and quantifying the sources of
air pollutants at a receptor location.

• 座標系統
– 尤拉系統(Eulerian
system)
– 拉格藍奇系統
(Lagrangian system)
• 所考慮的污染物
–
–
–
–

• 依所考慮的空間尺度，可分為：
• 近場(near field)：距污染源數公里範圍內，主要由煙

• 局部(local)或短距離：距污染源數十公里範圍左右，

• 都市(urban)尺度：數十至數百公里範圍，包含各種類

• 區域(regional)尺度：數百公里以上，必須考慮沈降及

• 全球(global)尺度。

(Gaussian Plume Models)
•The earliest air models
• Use many simplifying assumptions to obtain
closed-form analytical solutions
–
–
–
–
–
Spatially uniform (homogeneous) dispersion
Plume coherency
Inert or first-order decay
Slender plume assumption
• Gaussian plume models are not capable of
treating nonlinear problems (such as
photochemistry).
Computer Gaussian Plume Models
• ISCST: ISC3 is a steady-state Gaussian plume model which can be
used to assess pollutant concentrations from a wide variety of
sources associated with an industrial complex. This model can
account for the following: settling and dry deposition of particles;
downwash; point, area, line, and volume sources; plume rise as a
function of downwind distance; separation of point sources; and
• AERMOD: A steady-state plume model that incorporates air
dispersion based on planetary boundary layer turbulence structure
and scaling concepts, including treatment of both surface and
elevated sources, and both simple and complex terrain.
• CALINE3: A steady-state Gaussian dispersion model that is
designed to determine air pollution concentrations at receptor
locations downwind of highways located in relatively
uncomplicated terrain.
Control
Source
Receptor
Meteorology
Output
ISCST輸入檔樣版
Gaussian Puff Models
–Fewer simplifying assumptions
–Accounting for spatial variability of meteorological and dispersion
conditions,
–causality effects,
–low wind speed dispersion,
–temporal variability in emission rates, etc.
– employ analytical solutions for each puff but computers are required to
track the large number of puffs
– some models retain the plume coherency assumption, they are not
suitable for diffusion in complex terrain, sea breeze, mountain-valley
wind, …
– a few have been developed for individual reactive plumes, but they are
difficult to consider the pollutants outside the plume
– numerical solution methods are needed to solve chemical kinetics
equations
•Examples: CALPUFF, RPM-IV, SCIPUFF, SCICHEM
CALPUFF
• CALPUFF is a multi-layer, multi-species non-steadystate puff dispersion model that simulates the effects
of time- and space-varying meteorological conditions
on pollution transport, transformation and
removal. CALPUFF can be applied on scales of tens to
hundreds of kilometers. It includes algorithms for
subgrid scale effects (such as terrain impingement), as
well as, longer range effects (such as pollutant removal
due to wet scavenging and dry deposition, chemical
transformation, and visibility effects of particulate
matter concentrations).
• Complicated 3-D meteorological input is required.
Lagrangian Models
• approximate emissions as particles of mass and follow
the particles as they are transported and diffused by
atmospheric flow and turbulence.
• can estimate the dispersion very close to source or in
complex topography, e,g., re-circulation, sea breeze,
valley flows, etc.
• Lagrangian particle models are:
– State-of-the-science approach, especially for simulation of
inhomogenous (convective) turbulence
– Computationally demanding
– More difficult to deal with wet and dry deposition,
chemistry

（Lagrangian particle diffusion modeling）

x i 1  x i  u ( x , y , z )  t  u  ( x , y , z , t )  t
y i 1  y i  v ( x , y , z )  t  v  ( x , y , z , t )  t
z i 1  z i  w ( x , y , z )  t  w  ( x , y , z , t )  t
xi ,y i ,zi
: 質點在i時間的位置
x i  1 , y i  1 , z i  1 : 質點在i+1時間的位置
u ,v ,w
: 平均風速
u  ,v  ,w  : 瞬時風速擾動量
Grid Models
Governing equation for Atmospheric Diffusion
  uc i   vc i    wc i  

  


t
y
z 
 x
 ci
Change in
=
Concentration

 
 ci 
 ci 
 
 
 ci 
 K H
 
KH

 KV

x 
x  y 
y  z 
z 
Turbulent Diffusion
+
Ri
Chemical
Reaction
+
Si
Emissions
+
Li
Surface
Removal/Deposition
Grid Models
• 網格模式(Grid model)為採用尤拉系統的數值模式，此

• 此種方法之優點為：
– 能考慮 各項影響因素（包括傳送、擴散、排放、反應、沈

– 能考慮非線性的化學反應。
• 缺點：
– 須大量的計算機儲存位置和時間。
– 須輸入大量的資料。
– 數值延散(numerical dispersion)和擴散(numerical
diffusion)的困擾。
– Subgrid解析的問題: 網格較小結果，較準確，但須越多計

• 排放源
– 地面排放源 (移動源, 面源, 低點源, 生物源, 逸

– 點源 (大排放源的高煙囪)
• 擴散 (Diffusion)
• 化學轉換
– VOC 、 NOx 、自由基的氣相化學反應和循環
– 氣膠的熱動力學和液相的化學反應
• 沉降(Deposition)
– 乾沉降(Dry deposition)
– 濕沉降(Wet deposition，包括rain out and wash out)
Required Input Data (1)
• Emissions Inputs
– Low-level anthropogenic
emissions
•
•
•
•
Point sources
Area sources
– Elevated point source
emissions
– Biogenic emission
estimates
• Meteorological Inputs
– Three-dimensional winds
– Three-dimensional
temperatures
– Three-dimensional watervapor concentration
– Surface pressure
– Three-dimensional vertical
diffusivity (effective mixing
height)
– Two-dimensional cover
– Rainfall rate
Required Input Data (2)
• Air quality related input files
– Initial conditions (all grids, initial hour)
– Boundary conditions (outermost grid, all hours)
• Chemistry input files
– Chemical reaction rates
– Photolysis rates
• Geographic/other input files
– Land-use, Land-cover
– Albedo, turbidity, and ozone column
Modules in a Grid Model
• Emissions Modeling System
– EPS2x, SMOKE, BEIS, MEGAN, …..
• Meteorological Modeling System
– MM5 , RAMS , WRF, …..
• Preprocessors for Other Inputs
– TUV (photolysis Rates)
– Initial Concentrations and Boundary Conditions
• Air Quality Model
– CAMx, Models-3/CMAQ , UAM-V, MAQSIP, …..
• Post-Processors and Visualization
– Model Performance Evaluation
– PAVE
– NCL, …..

Model Perfromance Evaluation

• 所有模式都有無法避免的限制、不確定性

• 這項差異也可能是輸入資料不準確或缺乏

• 要將這兩種影響模式表現的因素分離非常

Model Evaluation (Validation)
• Operational Validation(作業性驗證): 將模式

• Scientific Validation(科學性驗證): 比較全面

• Diagnostic Evaluation(診斷性評估): 評估模

• 時間演變比較圖：對於O3影響，需作模擬值與監

• 地面等濃度圖：網格模式需選擇適當時間繪出地

• 散佈圖：繪製模擬值與監測值比較之散佈圖，以

• 非配對峰值之常化偏差(MB)：計算同一天O3最大監測小時濃

• 配對值之常化偏差(OB)：針對O3之模擬計算同一小時O3、
NOx/NO2、NMHC，針對PM10之模擬計算同一日PM10、SO2、
NOx/NO2模擬與監測平均濃度之常化偏差，瞭解模式是低估

• 配對值之絕對誤差(GE)：針對O3之模擬計算同一小時O3、
NOx/NO2、NMHC，針對PM10之模擬計算同一日PM10、SO2、
NOx/NO2所有模擬與監測濃度之平均常化絕對誤差量。O3濃

O3
NO2
NMHC

0.05
0.16
0.26
0.64
0.89
-0.01
0.41

--0.16
0.22
0.33
0.01
0.5

--0.06
0.23
0.31
0.16
0.5

0.03
0.08
0.21
0.33
0.45
-0.16
0.37

--0
0.04
0.18
0.35
0.55

0.02
0.09
0.26
0.38
0.49
0.06
0.38

-0.02
0.01
0.19
0.16
0.37
-0.16
0.38

--0.04
-0.01
0.18
0.32
0.45

-0.08
-0.05
0.2
0.93
0.97
-0.27
0.5

0.04
0.11
0.24
-0.01
0.33
-0.18
0.34

-0.03
-0.01
0.25
-0.07
0.44
-0.28
0.4

--0.05
0.08
0.25
-0.06
0.44

0.19
0.25
0.37
-0.04
0.42
-0.59
0.59

--0.1
0.2
0.29
0.11
0.36

--0.27
0.42
0.43
-0.33
0.39

--0.12
0.15
0.27
0.19
0.53

0.06
0.12
0.26
0.19
0.5
-0.2
0.42

±10%
±15%
<35%
-40%~+50%
<80%
-40%~+50%
<80%
PM事件日

PM2.5
PM10

-0.16
-0.11
-0.2
-0.18
-0.18
-0.02
-0.07
-0.03
-0.13
-0.21
-0.31
-0.31
-0.32
-0.2
-0.18
-0.34
-0.18
-50%
~+80%
NO2

0.32
0.41
0.29
0.24
0.39
0.34
0.31
0.39
0.33
0.29
0.33
0.36
0.32
0.22
0.3
0.34
0.32

0.34
0.28
0.2
0.27
0.02
0.24
0.14
0.14
0.09
-0.07
0.09
-0.16
0.21
-0.06
0.02
-0.11
0.1

0.5
0.5
0.31
0.4
0.31
0.32
0.35
0.4
0.31
0.28
0.33
0.22
0.36
0.21
0.33
0.32
0.34
150%

-0.1
-0.17
-0.11
0.05
0.14
0.1
0.01
0.08
0.54
0.03
0.03
0.14
-0.02
-0.03
-0.44
0.05
0.02
-40%
~+50%
SO2

0.26
0.3
0.17
0.09
0.18
0.1
0.08
0.09
0.54
0.1
0.22
0.24
0.06
0.12
0.44
0.06
0.19
80%

0.1
0.06
0.23
0.23
0.38
0.29
0.25
0.27
0.49
0.01
0.01
--0.02
--0.94
0.25
-40%
~+50%

0.17
0.12
0.23
0.32
0.38
0.29
0.26
0.34
0.49
0.03
0.11
-0.02
--0.94
0.28
80%

```